Generation of singlet oxygen from fragmentation of monoactivated 1,1-dihydroperoxides

Article abstract of DOI:10.1021/jo202265j

The first singlet excited state of molecular oxygen (¹O ₂) is an important oxidant in chemistry, biology, and medicine. ¹O₂ is most often generated through photosensitized excitation of ground-state oxygen. ¹O₂ can also be generated chemically through the decomposition of hydrogen peroxide and other peroxides. However, most of these "dark oxygenations" require water-rich media associated with short ¹O₂ lifetimes, and there is a need for oxygenations able to be conducted in organic solvents. We now report that monoactivated derivatives of 1,1-dihydroperoxides undergo a previously unobserved fragmentation to generate high yields of singlet molecular oxygen (¹O₂). The fragmentations, which can be conducted in a variety of organic solvents, require a geminal relationship between a peroxyanion and a peroxide activated toward heterolytic cleavage. The reaction is general for a range of skeletal frameworks and activating groups and, via in situ activation, can be applied directly to 1,1-dihydroperoxides. Our investigation suggests the fragmentation involves rate-limiting formation of a peroxyanion that decomposes via a Grob-like process.

Full text of DOI:10.1021/jo202265j

Featured Article
pubs.acs.org/joc
Generation of Singlet Oxygen from Fragmentation of Monoactivated
1,1-Dihydroperoxides
Jiliang Hang,† Prasanta Ghorai,†,§ Solaire A. Finkenstaedt-Quinn,‡ Ilhan Findik,‡ Emily Sliz,‡
Keith T. Kuwata,‡ and Patrick H. Dussault*,†
†Department of Chemistry, University of Nebraska−Lincoln, Lincoln, Nebraska 68588-0304, United States
‡Department of Chemistry, Macalester College, Saint Paul, Minnesota 55105-1899, United States
S
* Supporting Information
ABSTRACT: The first singlet excited state of molecular oxygen (1O2)
is an important oxidant in chemistry, biology, and medicine. 1O2 is most
often generated through photosensitized excitation of ground-state
1
oxygen. O2 can also be generated chemically through the decom-
position of hydrogen peroxide and other peroxides. However, most of
these “dark oxygenations” require water-rich media associated with short
1O2 lifetimes, and there is a need for oxygenations able to be conducted
in organic solvents. We now report that monoactivated derivatives of 1,1-
dihydroperoxides undergo a previously unobserved fragmentation to gener-
ate high yields of singlet molecular oxygen (1O2). The fragmentations, which
can be conducted in a variety of organic solvents, require a geminal relationship between a peroxyanion and a peroxide activated toward
heterolytic cleavage. The reaction is general for a range of skeletal frameworks and activating groups and, via in situ activation, can be
applied directly to 1,1-dihydroperoxides. Our investigation suggests the fragmentation involves rate-limiting formation of a peroxyanion
that decomposes via a Grob-like process.
However, preparative oxidations have been achieved for relatively
reactive substrates using biphasic or emulsion conditions.19
Generation of 1O2 in organic solvents has been achieved through
thermal decomposition of calcium peroxide (suspension),16
arene endoperoxides,20 phosphite ozonides,21 silyl hydrotri-
oxides,14 and alkyl hydrotrioxides.15 However, the latter class of
reagents are often thermally unstable and must be prepared
immediately prior to reaction. Major classes of reagents for
INTRODUCTION
■
Singlet molecular oxygen (1O2) is an important oxidant in
chemistry, biology, and medicine and the source of energy for
chemical oxygen−iodine (COIL) lasers.1a−c Several typical
reactions are illustrated in Scheme 1.2−4
1
Scheme 1. Examples of O2 Reactivity
1
chemical generation of O2 are overviewed below.
1O2 is most commonly generated through dye-sensitized exci-
tation of ground-state dioxygen (3O2).1b,5 However, preparative
application can be limited by the need for high-volume photo-
reactors, as well as by safety issues associated with reactions
under oxygen.6 Following the discovery of the chemical genera-
tion of 1O2 from the reaction of hydrogen peroxide and bleach,7
a number of methods for “dark oxygenation” have been reported
based upon decomposition of ozone,8 hydrogen peroxide,1a,9−13
hydrotrioxides,14,15 and several inorganic peroxides.16,17 The
majority of these methods require the use of water-enriched
Received: November 23, 2011
Published: January 11, 2012
18
1
media, conditions associated with a very short O2 lifetime.
© 2012 American Chemical Society
1233
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
Our interest in this area arose during investigations of a
“reductive ozonolysis” of alkenes.22 This transformation was
hypothesized to involve nucleophilic addition of amine N-
oxides to short-lived carbonyl oxides to generate peroxyanion/
oxyammonium acetals. These previously unknown intermedi-
ates were surmised to undergo fragmentation to generate the
Thermal Stability. Dihydroperoxide 1a, which melts without
decomposition at 78−80 °C and is not detonated by a sharp ham-
mer blow, undergoes a highly exothermic decomposition when
heated beyond 100 °C. Monoacetate 1b is stable for prolonged
periods at freezer temperature and for days at room temperature
but undergoes an exothermic decomposition beginning at approxi-
mately 76 °C (Figure 1). Monocarbonate 1d undergoes partial
1
observed “reduction” products, along with a molecule of O2,
and a molecule of amine (eq 1). Reasoning that a similar
fragmentation should be possible for any species featuring a
geminal relationship between a peroxyanion and a heterolytically
activated O−X bond, we explored the reactivity of monoper-
esters of readily available 1,1-dihydroperoxides. Deprotonation of
these species resulted in rapid fragmentation accompanied by
Figure 1. Thermal stability of dihydroperoxide monoester 1b.
23
1
highly efficient liberation of O2 (eq 2). We now describe our
investigations into the scope and mechanism of this process.
decomposition within days at room temperature to form a
substituted caprolactone and undergoes exothermic decomposi-
tion when heated above 70 °C.
Assay for 1O2. The release of 1O2 was quantified by chemical
reaction with terpinene (T) or diphenylisobenzofuran (DPBF, D).
RESULTS
■
Range of Acetal/Ketal Scaffolds. 1,1-Bishydroperoxides
1a−8a were prepared through Re(VII)-promoted acetalization
of the corresponding ketones (Scheme 2).24 Monoacetylation
1
Terpinene (T) undergoes addition of O2 to form the stable
endoperoxide ascaridole (T-O2, eq 3).27 Furan D undergoes an
Scheme 2. Synthesis of 1,1-Dihydroperoxides and Derived
Monoperesters
even faster addition to generate diketone D-O2 via an unstable
endoperoxide (eq 4). T-O2 and D-O2 are easily quantified through
isolation or by NMR or GC/MS analysis in the presence of an
internal standard. Details are provided in Supporting Information.
Variation of the Bisperoxyacetal Skeleton. We next
investigated the generality of the fragmentation across a range
of peroxyacetal skeletons. Solutions of monoesters 1b−4b and
6b−8b were decomposed with TBAF in the presence of
0.5 equiv of terpinene (Table 1); the relative stoichiometry was
chosen to facilitate product separation. Addition of base was
accompanied by the disappearance (TLC) of the monoesters
and the appearance of the corresponding ketone and T-O2. The
yield of trapped 1O2 varied little with peroxyacetal structure. By
comparison, a 1,1-dihydroperoxide (1a) did not react under the
reaction conditions.
with acetic anhydride and pyridine furnished isolable mono-
peresters (1b−4b, 6b−8b);25 the exception was adamantanone
dihydroperoxide (5a), which instead underwent ring expansion
to a lactone. 1,1-Dihydroperoxides derived from aldehydes (not
shown) reacted to form highly polar byproducts, presumably
reflecting E1cb fragmentation of peresters bearing an adjacent
C−H.26
Influence of Peroxyacetal Backbone on Relative
Reactivity. The decomposition of mixtures of dihydroperoxide
monoesters in the presence of a limiting amount of TBAF revealed
little difference in reactivity between acyclic and cyclic substrates or
between cyclic substrates possessing different amounts of ring
strain (Scheme 3).
1234
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
reactive bisperesters (vida infra). However, a reinvestigation of the
phenomenon using peracetate 1b and TBAF revealed insignificant
differences (Table 3).
Table 1. Influence of Peroxyacetal Backbone
a
monoperester or bisOOH
yield of T-O2 (%)
Table 3. Influence of Order of Addition of Reagents
1b
2b
3b
4b
5b
7b
8b
1a
26
26
23
31
a
conditions
yield of T-O2 (%)
27
add n-Bu4NF to solution of 1b and T
add 1b to solution of n-Bu4NF and T
35
38
24
23
a
Based upon 1b.
NR
a
Relative to monoperester
Range of Activating Groups. Base-promoted decom-
position of a perbenzoate (1c) and a percarbonate (1d)
1
a
proceeded rapidly to furnish yields of trapped O2 similar to
Scheme 3. Influence of Scaffold on Reactivity
those observed for the peracetate (Table 4). Attempts to
Table 4. Investigation of Other Peracyl Activating Groups
a
reagent
yield of 1c or 1d (%)
yield of T-O2 (%)
a
Products analyzed by GC/MS.
PhCOCl
EtOC(O)Cl
PhNCO
TsCl
1c (37)
1d (76)
dec
24
20
Influence of Solvent. Decomposition of monoester 1b in a
number of solvent systems (Table 2) revealed the yield of
dec
a
Based upon T-O2 vs stoichiometry of 1c or 1d
1
Table 2. Influence of Solvent on Trapped O2
prepare the analogous percarbamate or persulfonate led only to
ring-expansion.
Influence of Base/Counterion. As shown in Table 5, the
decomposition of 1b in the presence of LiN(TMS)2 resulted in
a
solvent
base
yield of T-O2 (%)
CH3CN
CHCl3
CDCl3
n-Bu4NF
n-Bu4NF
n-Bu4NF
n-Bu4NF
n-Bu4NF
n-Bu4NF
KOtBu
26
17
35
33
30
Table 5. Influence of Base
acetone-d6
C6D6
b
C6F6
7
a
b
base
yield of T-O2 (%)
C6F6
1
KOtBu
30
4.1
35
C6F6/ CH3CN (9:1)
C6F6/ CH3CN (1:1)
C6F14/ CH3CN (1:1)
n-Bu4NF
n-Bu4NF
n-Bu4NF
n-Bu4NF
24
36
28
39
LiN(SiMe3)2 (in THF)
nBu4NF (in THF)
c
a
Freon-113/CH3CN (1:1)
Based upon 1b.
a
b
c
Based upon 1b. Limited reagent solubility. Biphasic.
1
much lower yields of O2 trapping compared to reactions
involving the comparably basic KOtBu or TBAF.
dioxygen transfer (as T-O2) to be increased in solvents
18
1
associated with greater O2 lifetime. The exception was for
perfluorinated media, in which the reaction appeared limited by
reagent solubility. However, improved yields were possible in
fluorocarbon/CH3CN mixtures.
Order of Addition. In early experiments involving decom-
position of monoperester 1b,23 we had observed the yield of 1O2
to vary with the relative order of addition of KOtBu and perester,
suggesting the possibility of disproportionation to form poorly
Effect of Temperature and Rate of Addition. The base-
promoted decomposition of the 1,1-dihydroperoxide monoesters
is rapid; for example, rapid (1−2 s) injection of a 1 M THF solu-
tion of TBAF into a rapidly stirring solution of monoester 1b
(1 M in THF) results in immediate effervescence. Although the
decomposition of monoperester 1b proceeds significantly faster
1
at higher temperatures, the yield of trapped O2 is improved at
lower temperatures (Table 6, entries 1−4). Slowing the rate of
1235
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
not encountered with either the monoperesters (e.g., 1b) or a
bisperester (16, vida infra).
Table 6. Influence of Temperature and Rate of Addition
Reaction of the silylated monoesters with TBAF in the
presence of a trapping agent resulted in yields of transferred
1O2 equal to or exceeding those observed through deprotona-
tion of the dihydroperoxide monoesters (Table 7).
a
entry
T (°C)
solvent
addition mode
yield of T-O2 (%)
1
2
3
4
5
6
rt
0
THF
b
b
b
b
b
10
13
19
24
24
42
Table 7. Fragmentation of Silylated Dihydroperoxide
Monoesters
THF
−40
−78
rt
THF
THF
CH3CN
rt
CH3CN
b
c
a
subs
SiR3
time (min)
trap
yield (%)
a
c
Relative to perester. Syringe (≤10−3 h). Syringe pump (0.25 h).
9a
9b
9b
9b
10
11
12
13
14
SiEt3
TBS
TBS
TBS
SiEt3
SiEt3
SiEt3
SiEt3
SiEt3
5
5
D
D
T
33
29
25
25
34
37
41
38
28
reagent addition resulted in a significant increase in the yield of
trapped O2 (entries 5 and 6).
Attempted Gas-Phase Transfer. The inverted relationship
between the yield of oxygen transfer and the rate of peroxide
decomposition (Table 6, entries 5 and 6) led us to investigate
1
5
10
30
20
10
16
5
D
D
D
D
D
D
1
whether O2 could be escaping into the gas phase. The head-
space of a septum-covered flask containing an acetonitrile solu-
tion of perester 1b was exhausted via a short Teflon tube into an
acetonitrile solution of terpinene (T). Rapid (≤10 s) addition of
TBAF/THF into the solution of monoester led to vigorous
effervescence and net efflux of a high volume of gas through the
receiving solution. However, no significant amount of ascaridole
(T-O2) was detected.
a
Based on silylated perester.
Reactivity of Bissilyl and Bisacyl Derivatives. Prepara-
tions of bissilyl and bisacyl derivatives of dihydroperoxide 1a
are illustrated in Scheme 5. The bistriethylsilyl derivative (15)
Generation of the Peroxyanion via Nucleophilic
Desilylation. We investigated an alternative route to the pre-
sumed peroxyanion intermediate through nucleophilic desilyla-
tion. Monotrialkylsilylated substrates were readily available via
silylation of the monoesters (Scheme 4). Curiously, the perester
Scheme 5. Reactivity of Bissilyl and Bisacyl Derivatives of
1,1-Dihydroperoxides
Scheme 4. Synthesis of Silylated Peresters
was an oily compound that decomposed at ∼140 °C. The
bisacyl derivative (16) was a solid that melted without
decomposition at 54−56 °C.29 Saponification of 16 with
cesium hydroxide in CH3CN resulted in reformation of the
1
parent ketone and generation of O2. Bissilyl ether 15 was
decomposed by n-Bu4NF without apparent transfer of oxygen.
Tandem Activation and Fragmentation. Intrigued by
the possibility of generating and decomposing the monoactiva-
ted bisperoxide intermediate in situ, we investigated the reac-
tivity of dihydroperoxide 1a toward four classes of activating
agents: an electron-poor nitrile, a heteroaromatic known to under-
go SNAr reaction with tertiary hydroperoxides,30 a sulfonyl chloride,
and an acid anhydride. Our initial assay qualitatively monitored the
formation of ketone upon addition of base to solutions containing
the dihydroperoxide and the individual additive (Table 8). In each
case, the addition of stoichometric electrophile and excess KOtBu
resulted in rapid conversion of 1a to 4-tert-butylcyclohexanone with-
out any evidence of intermediates. Although these reactions should
in theory require only two equivalents of base, we found that four
carbonyl remained invisible by 13C NMR despite use of pulse
delays (≥5 s) or the addition of Ni(acac)2 as a relaxation agent.28
The perester carbonyl could be observed as a single sharp signal
at 167.8 ppm if spectra were acquired at ≤40 °C (9a, 9b, 11).
We are unsure as to the basis of this phenomenon, which was
1236
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
Table 8. Investigation of Electrophiles for Activation/
Fragmentation
Reaction Order and Apparent Rate Constant. The
base-promoted decomposition of monoperbenzoate 1c was
monitored at 272 nm (Figure 2). In the first experiment,
a
activator
KOtBu (equiv)
time (h)
conversion
CCl3CN
CCl3CN
triazene
triazene
p-TsCl
p-TsCl
Ac2O
2.0
4.0
2.0
4.0
2.0
4.0
2.0
4.0
0.75
≤0.1
>12
incomplete
complete
incomplete
complete
≤0.1
0.3
incomplete
complete
≤0.1
0.5
incomplete
complete
Ac2O
0.25
a
TLC and/or NMR
equivalents were invariably required to achieve complete consump-
tion of the dihydroperoxide.
We next reinvestigated the most promising conditions in the
presence of a trapping agent. The yields of oxygen transfer, which
were as good as those obtained from the dihydroperoxide mono-
esters, were similar regardless of activating agent (Table 9).
Figure 2. Monitoring rate of decompositions.
rapidly stirred THF solutions (0.5 mM) of dihydroperoxide
monoester 1c were mixed with THF solutions of TBAF ranging
in concentration from 0.50 to 2.00 mM. The reaction solutions
were then placed in the UV spectrophotometer, and the absor-
bance at 272 nm monitored for 1.0 min. Plotting the log-
[TBAF]initial against the −log(Δabs) revealed an apparent reac-
tion order of 0.94 in TBAF and a calculated rate constant of 102 L/
mol-s; details are provided in the Supporting Information. Repeating
the experiment with a constant concentration of TBAF and varying
concentrations of 1c indicated an order of 1.2 in peroxide and a
calculated rate constant of 102 L/mol-s. Overall, the results support
a mechanism that is first-order in both peroxide and base.
Table 9. Oxygen Transfer via in Situ Activation
a
activator
time (min)
yield of D-O2 (%)
CCl3CN
triazene
p-TsCl
Ac2O
5
5
5
5
36
39.5
39
31
a
Relative to dihydroperoxide 1a.
Theoretical Investigations of the Putative Fragmen-
tation Step. The base-promoted fragmentation of the
monoactivated 1,1-dihydroperoxides could conceivably involve
either a concerted Grob-like fragmentation of a zwitterion or the
formation and decomposition of an intermediate trioxetane.31a,b
The latter pathway cannot be simply disregarded. Dioxetanes are
readily formed by intramolecular nucleophilic displacements of
2-halohydroperoxide,32 while dioxiranes are generated with good
efficiency from intramolecular attack of a hemiacetal oxygen on
Preparative Reactions. Our experiments until this point
had focused on the yield of liberated oxygen. We next inve-
stigated the amount of reagent necessary to completely con-
sume a trapping reagent using either the fragmentation of a
preformed perester (1b, 8b) or the tandem activation/fragmen-
tation of dihydroperoxide 1a. As illustrated in Table 10, com-
plete consumption of terpinene was generally possible with as
little as 5 equiv of reagent.
Table 10. Investigation of Preparative Oxygenations
a
subs (equiv)
trap
reagents (equiv)
TBAF (5)
time (min)
product, yield (%)
1b (5)
1b (6)
1b (2.5)
1a (5)
1a (7)
8b (6)
T
T
D
T
T
T
10
10
5
T-O2 (83)
T-O2 (100)
D-O2 (100)
T-O2 (69)
T-O2 (100)
T-O2 (100)
TBAF (6)
TBAF (2.5)
triazine (5), KOtBu (20)
triazine (7), KOtBu (28)
TBAF (6)
15
15
10
a
Yield of T-O2 or D-O2.
1237
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
an electrophilically activated peroxide.33 We therefore turned to
density functional theory (specifically, the B3LYP hybrid
functional and the 6-31+G(d,p) basis set) to investigate each
transition state (TS) and intermediate (MIN) involved in the
decomposition of the monoactivated dihydroperoxides. We
focused our efforts on heterolytic processes as fragmentations
involving radical intermediates would be expected to generate
1
3O2 and not O2.
Decomposition of the Monoperester. Comparison of
the predicted zero-point corrected relative energies for
transition states and intermediates in the decomposition of a
peroxyanion found a concerted fragmentation to offer the
lowest energy pathway for generation of O2 and a carbonyl
(Scheme 6). The alternative pathway is disfavored by the strain
Figure 3. (a) Optimized geometry of the ground-state conformer,
MIN-1a. (b) Optimized geometry of conformer MIN-1c. Bond
lengths are in angstroms, and bond angles and dihedral angles are in
degrees.
1
Several lines of evidence suggest the presence of a stabilizing
intramolecular hydrogen bond between the anionic oxygen
(O1) and the C3−H1 bond. The O1---H1 distance is less than
1.9 Å, and the C3−H1 bond is 0.03 Å longer than the other two
methyl C−H bonds. The C3−H1 bond also vibrates at a much
lower frequency (2693 cm−1) compared with the other
hydrogen atoms on C3 (∼3100 cm−1). These data support a
normal hydrogen bond in which electron density from the
peroxyanion is delocalized into the sigma antibonding orbital of
the C−H bond. There is precedence for normal C−H
hydrogen bonds in the literature.35 Conformer MIN-1b (not
shown) is structurally very similar to MIN-1a. In particular,
MIN-1b also possesses an intramolecular hydrogen bond, with
an O1---H1 distance of 1.851 Å. In contrast, the O1−O2 and
C1−O3 bonds in conformer MIN-1c are antiperiplanar instead
of synclinal (Figure 3b). This difference deprives MIN-1c of
the intramolecular hydrogen bond and removes one of the
hyperconjugative interactions, raising the energy of MIN-1c by
7−8 kcal/mol.34
Conformational Analysis of the Grob-Like Transition
States. The lowest energy conformer of the transition state for
Grob fragmentation, TS-2 (Figure 4a), is predicted to lie only
6.7 kcal/mol higher in energy than the ground-state reactant
conformer; TS-2 benefits from antiperiplanar O2−C1 and O3−
O4 bonds. As described by Grob,31a,b this arrangement
facilitates fragmentation since electron density in the O2−C1
bond can be donated into the sigma antibonding orbital of the
O3−O4 bond. Intrinsic reaction coordinate calculations reveal
that TS-2 does not correlate directly with any of the reactant
conformers (see the Supporting Information), but rather the
unnumbered van der Waals complex shown in Figure 4b. This
complex lies only 0.4 kcal/mol below the transition state in
energy, and the excess −1 charge in the complex is evenly
distributed between the OOCH2O and OC(O)CH3 moieties.
Another fragmentation transition state, TS-2* (Figure 4c),
which is only 0.7 kcal higher in energy than TS-2, is derived
from the most stable peroxyanion conformer, MIN-1a (Figure 3a).
TS-2* lacks antiperiplanar O2−C1 and O3−O4 bonds but is
stabilized by a weak intramolecular hydrogen bond between O1 and
H1: the C3−H1 bond is 0.01 Å longer than the other two methyl
C−H bonds, and the C3−H1 stretch is ∼200 cm−1 to the red of the
other methyl C−H stretches.
Scheme 6. Available Pathways and Predicted Relative
Energies (kcal/mol) for Fragmentation of a Deprotonated
Monoester
a
a
Energetics are based on the most stable conformer of each structure.
in the trioxetane intermediate, presumably reflecting electron
1
pair repulsions, and by the high barrier for loss of O2. (The
combined energy of the trioxetane and acetate is 6.5 kcal/mol
higher than the trioxetane formation transition state. This
reflects the existence of an exit channel complex of the
trioxetane and acetate that is bound by 16.3 kcal/mol.) For
simplicity, calculations were performed on an unsubstituted
acetal framework. Repeating the computations on a more
substituted acetal framework (not shown) lowered the energies
of most intermediates and transition states on the order of
5 kcal/mol but did not, in general, alter the relative energies of
the transition states. The exception was the transition state for
trioxetane formation, which increased significantly in energy for
the more substituted scaffold. Optimized coordinates for cal-
culated structures are provided in the Supporting Information.
Conformational Analysis of the Peroxyanion. Figure 3
shows the optimized geometries of two of the 12 possible
conformers of the peroxyanion in the unsubstituted acetate
system. There is a striking difference in energy between
conformers MIN-1a (and the structurally and energetically
similar MIN-1b, not shown) versus the remaining 10
conformers, exemplified by MIN-1c, all of which lie at least
8 kcal/mol higher in energy.34
The most stable conformer (MIN-1a, Figure 3a) possesses
two pairs of synclinal O−O and C−O bonds, allowing for
hyperconjugation of oxygen lone pairs from O2 and O3 into
neighboring C−O antibonding orbitals (C1−O3; C1−O2) The
antiperiplanar orientation of the C2−O5 and O3−O4 bonds
minimizes the repulsion between the lone pairs on O3 and O5.
For a more substituted system (methine rather than
methylene acetal), the two lowest energy transition states are
structurally analogous to those shown in Figure 4. However, the
methylated transition state lacking antiperiplanar O2−C1 and
O3−O4 bonds, but possessing an intramolecular hydrogen bond,
1238
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
Figure 4. (a) Optimized geometry of the most stable conformer of TS-2. (b) Optimized geometry of the reactant complex leading to the most stable
conformer of TS-2. (c) Optimized geometry of the second most stable transition-state conformer, TS-2*.
is slightly (0.8 kcal/mol) lower in energy than the one possessing
antiperiplanar O2−C1 and O3−O4 bonds.
Pathways for Fragmentation of the Monopercarbon-
ate. We were interested in whether the pathways for
fragmentation varied with the nature of the electrophilic activa-
ting group. Scheme 7 shows calculated zero-point corrected
Scheme 7. Calculated Relative Energies (kcal/mol) for
Selected Intermediates (MIN) and Transition States (TS) in
Fragmentation of Percarbonates
Figure 5. Optimized geometries of the most stable conformers of (a)
TS-4 and (b) TS-5. Bond lengths are in angstroms, and bond angles
and dihedral angles are in degrees.
the dihedral angle between the O2−C1 and O3−O4 bonds is
significantly less than 180°.
The peroxyanion derived from the dihydroperoxide mono-
carbonate can also undergo decomposition via a six-membered
ring intermediate (Scheme 7). The barrier for cyclization to the
1,2,4,5-tetroxan-3-oxide is predicted to be nearly identical to
that for fragmentation of the acyclic peroxyanion. The pre-
relative energies for selected intermediates and transition states
in the decomposition of a peroxyanion derived from a dihydro-
peroxide monopercarbonate, as modeled for the unsubstituted
(formaldehyde acetal) backbone.34 The methyl carbonate nucl-
eofuge in the percarbonate stabilizes negative charge more
effectively than the acetate leaving group in the perester, as
indicated by a calculated ΔE of decomposition ∼6 kcal/mol
more exothermic than in the acetate series (compare Schemes 6
ference for the cyclic isomer is substantial: the ground state con-
former of the tetroxane oxide (MIN-3a) is over 9 kcal/mol lower
in energy that the most stable acyclic conformation (MIN-3d),34
reflecting the significant number of hyperconjugative interactions
between C−O sigma antibonding orbitals and oxygen lone pairs in
the chairlike conformation of the cyclic isomer. Figure 5b shows the
transition state, TS-5, for the fragmentation of the cyclic peroxide to
generate a carbonyl, 1O2, and a methyl carbonate anion. The activa-
tion barrier for fragmenting the cyclic structure, while greater than
that for fragmenting the acyclic peroxyanion (9.6 vs 4.3 kcal/mol),
is still lower than that for ring-opening to the acyclic form. As was
the case in the monoester series, fragmentation of a more sub-
stituted acetal is predicted to be more favorable by ∼5 kcal/mol;
however, there is predicted to be a minimal impact, typically less
than 1 kcal/mol, on the relative energies of individual pathways.
Comparison with Intermediates in “Reductive”
Ozonolysis. We also modeled analogous intermediates and
transition states postulated to be involved in the fragmentation
of zwitterionic peroxy/ammonium acetals during reductive ozo-
nolyses (Scheme 8).22 As a result of a stabilizing electrostatic
interaction, the zwitterionic acetal is predicted to strongly
favor a ground-state conformation with a gauche O−C−O−O.
The low-energy transition state for fragmentation lies less
than 5 kcal/mol above the ground state; the fragmentation itself
is predicted to be significantly downhill. (These calculations
are based upon a zwitterionic acetal incorporating a fully
substituted nitrogen; repeating these calculations on a species
1
and 7). The two lowest energy pathways for liberation of O2
from the deprotonated monocarbonate are illustrated in
Scheme 7. A third pathway involving an intermediate trioxe-
tane, analogous to one illustrated in Scheme 6 for the mono-
perester, was predicted to involve a very high energy transition
state and is not described here.
Concerted fragmentation of the most stable conformer of the
acyclic peroxyanion of the dihydroperoxide monocarbonate is
calculated to occur with a significantly lower activation barrier
(4.3 vs 6.7 kcal/mol) compared with the corresponding trans-
formation of the perester (Scheme 6). We note that bond
breaking in TS-4 (see Figure 5) is significantly less advanced
than in the most stable transition state for the acetate system
(TS-2, Figure 4a) and is comparable to the bond breaking that
exists in the second most stable acetate conformer (TS-2*,
Figure 4c). As is the case for the most stable peracetate transi-
tion state conformer, TS-4 (Figure 5) benefits from
antiperiplanar O2−C1 and O3−O4 bonds. However, overlap
of the σ(O2−C1) and σ*(O3−O4) orbitals is not optimal, since
1239
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
1
effervescence, we could detect no evidence for transfer of O2
Scheme 8. Calculated Energies for Decompositions of
Zwitterionic Peroxyacetals
into the gas phase. We therefore suspect that yields of oxygen trans-
fer under conditions associated with higher 1O2 flux are limited by
self-quenching.40
Relationship to Known Fragmentations. A number of
heterolytic fragmentations involve formation of a new C−O
bond through migration of a C−C σ bond with cleavage of an
activated O−O bond.41a,b,c,d These include the Criegee rear-
rangement of peresters,41a the Hock rearrangement of protona-
ted or Lewis acid-complexed peroxides,41b the fragmentation of
primary ozonides derived from allylic alcohols,41c and the
fragmentation of 2-hydroperoxycarboxylates.41d Overall, the
new fragmentation bears a close topological resemblance to the
Grob fragmentation, an extended elimination reaction in which
formation of CO and CC pi systems drives the cleavage of
C−C and C−X σ bonds.31a,b The fragmentations described
above are likely to require a similar antiperiplanar alignment of
the breaking peroxyanion C−O with the activated O−O bond.
containing one or more N−H bonds predicted proton transfer
would take place in preference to fragmentation).
DISCUSSION
■
The reactions described here constitute a new class of fragmen-
tations involving a peroxyanion geminally linked to an elec-
trophilically activated peroxide (Scheme 9).36 The peroxyanion
Scheme 9. Mechanistic Overview
can be generated via deprotonation, desilylation, or deacylation.
The higher yields observed in the presence of tetraalkylammo-
nium or potassium counterions vs lithium point to the potential
importance of anion dissociation, but must be interpreted with
caution given the potential for different levels of 1O2 quenching
by the different bases and their protonated byproducts. The
data from competition reactions suggests that deprotonation
may be rate limiting.
The ability to achieve tandem activation and fragmentation
of readily available 1,1-dihydroperoxides is noteworthy. Hydro-
peroxyimidates, while not isolable, have been postulated as
short-lived intermediates in epoxidations.37 Alkylperoxytria-
zenes, which have been prepared from tertiary hydroperoxides
via SNAr reactions, have mainly been of interest as radical
initiators.30
CONCLUSIONS
■
The decomposition of monoactivated derivatives of 1,1-
dihydroperoxides is established as a new peroxide fragmenta-
tion that provides an efficient means for preparative generation
1
of O2 in organic media.
EXPERIMENTAL SECTION
■
Standard Procedure for Synthesis of 1,1-Dihydroperoxides.
1,1-Dihydroperoxides were prepared by a reported procedure.24
1,1-Dihydroperoxy-4-tert-butylcyclohexane (1a): 84%, white
solid; mp 81−83 °C; Rf = 0.33 (30% EA/Hex); other physical data
were identical to literature reports.24
1,1-Dihydroperoxycycloheptane (2a): 83%, white solid; mp 59−61 °C;
Rf = 0.33 (30% EA/Hex); other physical data were identical to
literature reports.24
1,1-Dihydroperoxycyclooctane (3a): 58%, colorless oil; Rf = 0.33
(30% EA/Hex); other physical data were identical to literature
reports.24
4-Phenyl-2,2-dihydroperoxybutane (4a): 72%, white solid; mp
61−63 °C; Rf = 0.36 (40% EA/Hex); other physical data were
identical to literature reports.24
1,1-Dihydroperoxyadamantane (5a): 88%, white solid; mp 148−
150 °C; Rf = 0.30 (30% EA/Hex); other physical data were identical to
literature reports.24
5,5-Dihydroperoxynonane (6a): 75%, colorless oil; Rf = 0.25 (20%
EA/Hex); other physical data were identical to literature reports.24
1,1-Dihydroperoxy-3-tert-butylcyclopentane (7a): 84%, colorless
oil; Rf 1= 0.30 (30% EA/Hex); IR 3420, 2956, 2870, 1365, 1104, 963
cm−1; H NMRδ 9.83 (s, 2H), 2.05 (m, 2H), 1.9 (m, 2H), 1.7 (m,
2H), 1.47 (m, 1H), 0.85 (s, 9H); 13C NMR δ 121.7, 49.2, 34.5, 32.6,
Relationship between Reaction Conditions and
1
Efficiency of O2 Transfer. The observation of more efficient
oxygen transfer in deuterated or halogenated solvents is in keeping
with the enhanced lifetime of 1O2 in solvents lacking C−H, O−H,
or N−H bonds.18 The failure of the results to scale linearly to 1O2
lifetime results from the presence of the trapping substrate, by
definition a potent quenching agent. The observation of improved
trapping yields at reduced reaction temperatures is not unexpected.
1
While the lifetime of O2 in organic solvents varies little with
temperature,38 reaction of 1O2 with electron-rich alkenes has a near-
zero enthalpy of activation and a strongly negative entropy of
activation;39 photosensitized oxygenations are often quite rapid even
at low temperatures.5 However, the observation of increased yields
from a slower rate of fragmentation is not easily explained. Although
rapid injection of reactants was associated with significant
1240
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
31.5, 27.4, 25.6; HRMS (ESI, MeOH/H2O, NaOAc) calcd for
C9H18NaO4 (M + Na)+ 213.1103, found 213.1110.
1,1-Dihydroperoxycyclododecane (8a): 93%, colorless oil; Rf =
0.33 (30% EA/Hex); other physical data were identical to literature
reports.24
(2.46 g, 10 mmol) in CH2Cl2 (150 mL) were added DMAP (0.24 g,
2.0 mmol) and Et3N (2.7 mL, 20 mmol). The reaction mixture was
cooled to 0 °C, and a solution of TESCl (1.5 g, 10 mmol) in CH2Cl2
(50 mL) was added dropwise over 10 min. The reaction was stirred for
30 min at 0 °C, and then the solvent was removed in vacuo. The
residue obtained was purified by silica flash chromatography (5% EA/
Hex) to give a colorless, oily product (2.09 g, 58% yield).
1-Acetyldioxy-1-triethylsilyldioxy-4-tert-butylcyclohexane (9a):
58%, colorless oil; Rf = 0.65 (20% EA/Hex); IR 2954, 1777, 1366,
1180, 865 cm−1; 1H NMR δ 2.33 (d, J = 14.48 Hz, 2H), 2.15 (s, 3H),
1.70 (d, J = 13.2 Hz, 2H), 1.46 (dt, J = 3.78, 13.8 Hz, 2H), 1.30 (m,
2H), 1.05 (m, 1H), 0.98 (t, J = 7.69 Hz, 9H), 0.87 (s, 9H), 0.69 (dd,
J = 7.55, 16.27 Hz, 6H); 13C NMR (233K) δ 167.9, 110.6, 47.2, 32.5,
30.4, 27.7, 23.2, 18.3, 7.0, 3.4; HRMS (ESI, MeOH/H2O, NaOAc)
calcd for C18H36NaO5Si (M + Na)+ 383.2230, found 383.2224.
1-Acetyldioxy-1-tert-butyldimethylsilyldioxy-4-tert-butylcyclo-
hexane (9b). TESCl was replaced by TBSCl without any other
change: 65%, colorless oil; Rf = 0.65 (20% EA/Hex); IR 2954, 2859,
1777, 1365, 1178, 873, 830, 784 cm−1; 1H NMR δ 2.33 (m, 2H), 2.14
(s, 3H), 1.68 (m, 2H), 1.46 (dt, J = 3.85, 13.63 Hz, 2H), 1.29 (m, 2H),
1.05 (m, 1H), 0.93 (s, 9H), 0.87 (s, 9H), 0.14 (s, 6H); 13C NMR
(233K) δ 167.9, 110.7, 47.2, 32.5, 30.4, 27.7, 26.3, 23.2, 18.6, 18.3 ;
HRMS (ESI, MeOH/H2O, NaOAc) calcd for C18H36NaO5Si (M +
Na)+ 383.2230, found 383.2235.
1-Acetyldioxy-1-triethylsilyldioxycycloheptane (10): 81%, color-
less oil; Rf = 0.65 (20% EA/Hex); IR 2969, 1776, 1409, 1170, 1056,
796 cm−1; 1H NMR δ 2.08 (s, 3H), 1.95 (m, 4H), 1.55 (br, 8H), 0.97
(t, J = 7.86 Hz, 9H), 0.68 (q, 6H); 13C NMR δ 115.9, 33.0, 30.2, 22.8,
17.8, 6.6, 3.7; HRMS (ESI, MeOH/H2O, NaOAc) calcd for
C15H30NaO5Si (M + Na)+ 341.1760, found 341.1759.
1-Acetyldioxy-1-triethylsilyldioxycyclooctane (11): 70%, colorless
oil; Rf = 0.65 (20% EA/Hex); IR 2956, 1776, 1363, 1184, 1066, 833,
729 cm−1; 1H NMR δ 2.10 (s, 3H), 1.97 (m, 4H), 1.56 (br, 10H), 0.97
(t, J = 7.63 Hz, 9H), 0.68 (q, 6H); 13C NMR (233K) δ 168.1, 115.0,
27.8, 27.2, 24.7, 22.0, 18.2, 7.0, 3.7; HRMS (ESI, MeOH/H2O,
NaOAc) calcd for C16H32NaO5Si (M + Na)+ 355.1917, found
355.1906.
Standard Procedure for Preparation of 1,1-Dihydroperoxide
Monoacetates (Illustrated for 1b). To a solution of 4-tert-
butylcyclohexyl-1,1-dihydroperoxide 1a (1.48 g, 7.2 mmol) in
CH2Cl2 (15 mL) were added DMAP (0.09 g, 0.75 mmol) and
pyridine (0.57 g, 7.2 mmol). The reaction mixture was cooled to 0 °C,
and a solution of Ac2O (0.74 g, 7.2 mmol) in CH2Cl2 (10 mL) was
added dropwise over 10 min. Upon completion of addition, the
reaction was stirred for 30 min at 0 °C and then diluted with CH2Cl2
(100 mL). The solution was washed with saturated NaHCO3 (20 mL),
water (20 mL), and brine (20 mL) and then dried over anhydrous
Na2SO4. The residue obtained upon removal of the solvent in vacuo
was purified by silica flash chromatography (5% EA/Hex) to give a
white solid (1.47 g, 84% yield). The acid anhydride could be replaced
by acid chloride (peresters) or ethoxycarbonyl chloride (percarbon-
ates) without any other change.
1-Acetyldioxy-1-hydroperoxy-4-tert-butylcyclohexane (1b): 84%,
white solid; mp 35−37 °C; Rf = 0.33 (10% EA/Hex); other physical
data were identical to literature reports.23
1-Acetyldioxy-1-hydroperoxycycloheptane (2b): 76%, colorless
oil; Rf = 0.33 (10% EA/Hex); IR 3345, 2930, 2859, 1756, 1425,
1171, 1006, 899 cm−1; 1H NMRδ 10.13 (s, 1H), 2.09 (2, 3H), 1.92 (m,
4H), 1.54 (br, 8H); 13C NMR δ 171.7, 117.4, 32.5, 29.6, 22.6, 17.5;
HRMS (ESI, MeOH/H2O, NaOAc) calcd for C9H16NaO5 (M + Na)+
227.0895, found 227.0898.
1-Acetyldioxy-1-hydroperoxycyclooctane (3b): 73%, colorless oil;
Rf = 0.33 (10% EA/Hex); IR 3340, 2924, 1756, 1409, 1200, 1078,
1
898 cm−1; H NMRδ 10.13 (s, 1H), 2.15 (s, 3H), 2.00 (m, 2H), 1.89
(m, 2H), 1.63 (br, 4H), 1.57 (br, 6H); 13C NMR δ 171.7, 116.8, 27.8,
27.4, 24.9, 21.8, 17.6; HRMS (ESI, MeOH/H2O, NaOAc), calcd for
C10H18NaO5 (M + Na)+ 241.1052, found 241.1046.
4-Phenyl-1-acetyldioxy-1-hydroperoxybutane (4b): 86%, colorless
oil; Rf = 0.33 (10% EA/Hex); IR 3340, 2988, 2900, 1758, 1378, 1066,
899, 748, 698 cm−1; 1H NMR δ 10.39 (s,1H), 7.35 (m, 2H), 7.28 (m,
3H), 2.86 (m, 2H), 2.24 (dd, J = 5.46, 12.9 Hz, 1H), 2.20 (s, 3H), 2.09
(dd, J = 5.46, 12,2 Hz, 1H), 1.61 (s, 3H); 13C NMR δ 171.7, 141.0,
128.6, 128.4, 126.3, 113.7, 34.9, 30.2, 18.1, 17.6; HRMS (ESI, MeOH/
H2O, NaOAc) calcd for C12H16NaO5 (M + Na)+ 263.0895, found
263.0899.
5-Acetyldioxy-5-hydroperoxynonane (6b): 76%, colorless oil; Rf =
0.33 (10% EA/Hex); IR 3350, 2960, 2873, 1760, 1198, 1056,
899 cm−1; 1H NMR δ 10.13 (s, 1H), 2.12 (s, 3H), 1.73 (m, 2H), 1.56
(m, 2H), 1.34 (m, 8H), 0.90 (t, J = 7.15 Hz, 3H); 13C NMR δ 171.6,
116.0, 28.9, 25.6, 22.7, 17.4, 13.8; HRMS (ESI, MeOH/H2O, NaOAc)
calcd for C11H22NaO5 (M + Na)+ 257.1365, found 257.1361.
1-Acetyldioxy-1-hydroperoxy-3-tert-butylcyclopentane (7b): 84%,
colorless oil; Rf = 0.33 (10% EA/Hex); IR 2959, 2869, 1758, 1365,
1184, 1070, 962; 1H NMR δ 10.516/10.501 (two s, totaling 1H), 2.17
(s, 3H), 2.0 (m, 4H), 1.75 (m, 2H), 1.54 (m, 1H), 0.88 (s, 9H); 13C
NMR δ 171.6, 122.9, 122.8, 49.3, 49.2, 34.6, 34.5, 32.7, 32.6, 31.6,
27.38, 27.36, 25.6, 17.6; HRMS (ESI, MeOH/H2O, NaOAc) calcd for
C11H20NaO5 (M + Na)+ 255.1208, found 255.1199.
4-Phenyl-1-acetyldioxy-1-triethylsilyldioxybutane (12): 79%, col-
orless oil; Rf = 0.65 (20% EA/Hex); IR 2957, 1780, 1182, 1103, 825,
1
698 cm−1; H NMR δ 7.31 (m, 2H), 7.23 (m, 2H), 2.78 (t, J = 8.95
Hz, 2H), 2.15 (m, 5H), 1.57 (s, 3H), 1.04 (t, J = 7.95 Hz, 9H), 0.75
(q, 6H); 13C NMR δ 141.3, 128.5, 128.3, 126.1, 111.9, 35.5, 30.4, 18.5,
17.8, 6.7, 3.7; HRMS (ESI, MeOH/H2O, NaOAc) calcd for
C18H30NaO5Si (M + Na)+ 377.1760, found 377.1756.
1-Acetyldioxy-1-triethylsilyldioxynonane (13): 73%, colorless oil;
Rf = 0.65 (20% EA/Hex); IR 2959, 1781, 1409, 1184, 1066, 802, 729
1
cm−1; H NMR δ 2.04 (s, 2H), 1.69 (m, 2H), 1.30 (br, 8H), 0.93 (t,
J = 7.85 Hz, 9H), 0.85 (t, J = 6.35 Hz, 6H), 0.63 (q, 6H); 13C NMR δ
114.1, 29.6, 25.5, 22.7, 17.7, 13.7, 6.5, 3.6; HRMS (ESI, MeOH/H2O,
NaOAc) calcd for C18H30NaO5Si (M + Na)+ 377.1760, found
377.1768.
1-Acetyldioxy-1-triethylsilyldioxycyclododecane (14): 80%, color-
less oil; Rf = 0.65 (20% EA/Hex); IR 2929, 1779, 1470, 1186, 1003,
851, 728 cm−1; 1H NMR δ 2.08 (s, 3H), 1.73 (m, 4H), 1.37 (br, 18H),
0.96(t, J = 7.96 Hz, 9H), 0.66 (q, 6H); 13C NMR δ 114.9, 26.7, 26.1,
26.0, 22.3, 21.9, 19.4, 6.7, 3.7; HRMS (ESI, MeOH/H2O, NaOAc)
calcd for C20H40NaO5Si (M + Na)+ 411.2543, found 411.2536.
1,1-Triethylsilyldioxy-4-tert-butylcyclohexane (15). To a solution
of 4-tert-butylcyclohexyl-1,1-dihydroperoxide 1a (0.50 g, 2.5 mmol) in
DMF (30 mL) was added Et3N (0.76 mL, 5.4 mmol). The reaction
mixture was cooled to −40 °C, and TESCl (0.91 mL, 5.4 mmol) was
added. The reaction was stirred for 1 h at −40 °C and then quenched
with brine (100 mL). The solution was extracted with hexane (50 mL)
three times, and the combined organic layer was dried over anhydrous
Na2SO4. The residue obtained upon removal of the solvent in vacuo
was purified by silica flash chromatography (5% EA/Hex) to give a
colorless oil (0.57 g, 53%), Rf = 0.80 (20% EA/Hex); other physical
data were identical to literature reports.42
1-Acetyldioxy-1-hydroperoxycyclododecane (8b): 81%, colorless
oil; Rf = 0.33 (10% EA/Hex); IR 3307, 2946, 2851, 1748, 1426, 1190,
1
997, 850 cm−1; H NMRδ 10.22 (s, 1H), 2.17 (s, 3H), 1.76 (m, 2H),
1.59(m, 6H), 1.39 (br, 14H); 13C NMR δ 171.7, 116.8, 26.1, 25.9,
25.8, 22.1, 21.8, 19.3, 17.7; HRMS (ESI, MeOH/H2O, NaOAc), calcd
for C14H26NaO5 (M + Na)+ 297.1678, found 297.1679.
1-Benzoyldioxy-1-hydroperoxy-4-tert-butylcyclohexane (1c):
37%, white solid; mp 76−78 °C; Rf = 0.43 (10% EA/hex); other
physical data were identical to literature reports.23
4-tert-Butyl-1-hydroperoxycyclohexyl ethyl carbonoperoxoate
(1d): 76%, colorless oil; Rf = 0.22 (10% EA/Hex); other physical
data were identical to literature reports.23
1,1-Diacetyldioxy-4-tert-butylcyclohexane (16). To a solution of
4-tert-butylcyclohexyl-1,1-dihydroperoxide 1a (1.48 g, 7.2 mmol) in
Standard Procedure for Preparation of Monosilylated
Peresters (Illustrated for 9a). To a solution of monoacetate 1b
1241
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
CH2Cl2 (15 mL) were added DMAP (0.09 g, 0.75 mmol), and
pyridine (0.57 g, 7.2 mmol). The reaction mixture was cooled to 0 °C,
and a solution of Ac2O (1.48 g, 14.4 mmol) in CH2Cl2 (10 mL) was
added dropwise over 10 min. Upon completion of addition, the
reaction was stirred for 30 min at 0 °C and then diluted with CH2Cl2
(100 mL). The solution was washed with saturated NaHCO3 (20 mL),
water (20 mL), and brine (20 mL) and then dried over anhydrous
Na2SO4. The residue obtained upon removal of the solvent in vacuo
was purified by silica flash chromatography (10% EA/Hex) to give a
white solid (1.55 g, 75% yield): mp 54−56 °C; Rf = 0.30 (20% EA/
approximate treatment is to relax the restriction that α and β orbitals
be identical. While this allows B3LYP to provide a qualitatively correct
description of the electronic structure, this treatment also introduces
triplet character into the wave function. We corrected for this spin
contamination using the approach of Houk and co-workers.43 This
approach predicts a singlet−triplet gap for molecular oxygen of 20.8
kcal/mol, in fair agreement with the experimental value of 22.5 kcal/
mol.1a Individual conformers within the monoperacetate and
monopercarbonate systems, along with optimized coordinates for
each minimized structure or transition state, are described in the
Supporting Information.
1
Hex); IR 2960, 2869, 1785, 1361, 1187, 1057, 896 cm−1; H NMR δ
2.34 (d, J = 12.81 Hz, 2H), 2.08 (s, 3H), 2.07 (s, 3H), 1.75 (d, J =
14.12 Hz, 2H), 1.62 (m, 2H), 1.31 (m, 2H), 1.09 (m, 1H), 0.86 (s,
9H); 13C NMR δ 167.5, 167.4, 111.5, 47.1, 32.2, 30.1, 27.5, 23.2, 17.5,
17.4; HRMS (ESI, MeOH/H2O, NaOAc) calcd for C14H24NaO6 (M +
Na)+ 311.1471, found 311.1467.
NMR Method for Quantification of Oxygen Transfer
(Illustrated for Dioxygenation of Terpenine). To a solution of
1b (50 mg, 0.2 mmol) and α-terpinene (T, 90%, 15 mg, 0.1 mmol) in
CH3CN (2 mL) was injected TBAF (1 M in THF, 0.24 mL,
0.24 mmol) within 5 s and the reaction mixture stirred at room
temperature for 5 min. Solvent was removed by vacuum. Internal
standard 1,2-dichloroethane was added, and T-O2 was measured by
NMR. The alkene peak at 6.3 ppm (dd, J = 8.56, 25.93 Hz) of T-O2
will be used for calculation via following equations:
ASSOCIATED CONTENT
■
S
* Supporting Information
Details regarding experimental procedures, spectral listings for
1
new molecules, H and 13C NMR data for new compounds,
thermal analysis for 1a and 1c, conformational analysis of
model monoperoxyacetate and monopercarbonate structures,
and optimized coordinates of calculated structures described in
the paper. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
actual mole
= mole
·2·area
/area
standard
Corresponding Author
*E-mail: pdussault1@unl.edu.
T‐O
standard
T‐O
2
2
Therefore, the calculated % yield =100*actual moleT‑O /theoretical moleT‑O
2
2
Present Address
Alternatively, quantification of T-O2 can be conducted by GC/MS
using a reported procedure5
§Department of Chemistry, Indian Institute of Science and
Education Research, Bhopal, India.
GC/MS Quantification of Relative Reactivity of Dihydroper-
oxide Monoesters (Illustrated for 1b and 4b). Prepare a standard
solution of 4-tert-butylcyclohexanone (1 mmol) and benzylacetone
(1 mmol) in CH3CN (2 mL). The standard ratio α =
area4‑tert‑butylcyclohexanone standard/areabenzylacetone standard. The mole ratio of
the ketone products from the competitive reaction is
area4‑tert‑butylcyclohexanone actual/(α·areabenzylacetone actual).
Sample Procedure for Oxygen Transfer (Illustrated for
Decomposition of 1b and Dioxygenation of Terpenine). To a
solution of 1b (50 mg, 0.20 mmol) and α-terpinene (T, 90%, 15 mg,
0.10 mmol) in CH3CN (2 mL) was added via syringe a solution of
TBAF (nominally 1 M in THF, 0.24 mL, 0.24 mmol) over 5 s. The
reaction mixture was stirred at room temperature for 5 min and then
the solvent was removed under vacuum. 1,2-Dichloroethane was
added as an internal standard and T-O2 determined by NMR, GC/
MS, or isolation, as described in the Supporting Information.
Standard Procedure for Fluoride-Mediated Decomposition
of Silylated Monoesters. To a solution of 9a (77 mg, 0.2 mmol)
and 1,3-diphenylisobenzofuran (DPBF, D, 27 mg, 0.10 mmol) in
CH3CN (2 mL) was added via syringe a solution of TBAF (nominally
1 M in THF, 0.24 mL, 0.24 mmol) and the reaction mixture stirred at
room temperature for 5 min. Then solvent was removed by vacuum,
and the residue was purified by column chromatography on silica gel
to give the pure product D-O2 (19 mg, 66%).
Standard Procedure for Tandem Activation and Decom-
position of 1,1-Dihydroperoxides (Illustrated for Trichloro-
acetonitrile). To a solution of 1a (41 mg, 0.2 mmol), CCl3CN
(29 mg, 0.2 mmol), and 1,3-diphenylisobenzofuran (DPBF, D, 27 mg,
0.1 mmol) in CH3CN (2 mL) was added KOtBu (90 mg, 0.8 mmol).
The reaction was monitored by TLC and stirred at room temperature
for 5 min. Then solvent was removed by vacuum, and the residue was
purified by column chromatography on silica gel to give the pure
product D-O2 (21 mg, 72% yield based upon D or 36% based upon
consumed 1,1-dihydroperoxide).
Calculation Methods. The B3LYP/6-31+G(d,p) model chemistry
was used in all calculations. Differences in electronic energies were
corrected by inclusion of zero-point vibrational energies scaled by
0.9806. Because B3LYP is based on a single Slater determinant, it
cannot rigorously describe either singlet molecular oxygen or
transition states in which states O2 is a distinct moiety. The standard
ACKNOWLEDGMENTS
■
Research was conducted with funding from NSF (CHE-
0749916, −1057982) and the Camille and Henry Dreyfus
Foundation in facilities remodeled with NIH support
(RR016544). Calculations were performed at Macalester
College, the National Center for Supercomputing Applications
facility at the University of Illinois (CHE090078), and the
Midwest Undergraduate Computational Chemistry Consorti-
um facility at Hope College (NSF-CHE-0520704). We thank
Dr. Joe Dumais and Prof. Charles Kingsbury for useful
discussions.
REFERENCES
■
(1) (a) Foote, C. S.; Clennan, E. L. In Active Oxygen in Chemistry;
Foote, C. S., Valentine, J. S., Greenberg, A., Liebman, J. F., Eds.;
Blackie Academic and Professional: London, 1995; p 105.
(b) Clennan, E.; Pace, A. Tetrahedron 2005, 61, 6665. (c) Endo,
M. Russ. J. Phys. Chem. A 2007, 81, 1497.
(2) Gollnick, K.; Kuhn, H. J. In Singlet Oxygen; Wasserman, H. H.,
Murray, R. W., Eds.; Academic Press: New York, 1979; pp 287−427.
(3) Clennan, E. L.; L’Esperance, R. P. Tetrahedron Lett. 1983, 24,
4291.
(4) Sheu, C.; Kang, P.; Khan, S.; Foote, C. S. J. Am. Chem. Soc. 2002,
124, 3905.
(5) Schmidt, R. Photochem. Photobiol. 2006, 82, 1161.
(6) Wootton, R. C. R.; Fortt, R.; de Mello, A. J. Org. Process Res. Dev.
2002, 6, 187.
(7) Foote, C. S.; Wexler, S.; Ando, W.; Higgins, R. J. Am. Chem. Soc.
1968, 90, 975. Greer, A. Acc. Chem. Res. 2006, 39, 797.
(8) Jary, W. G.; Ganglberger, T.; Pochlauer, P.; Falk, H. Monatsh.
̈
Chem. 2005, 136, 537. Ganglberger, T.; Jary, W. G.; Pochlauer, P.;
̈
Aubry, J.-M.; Nardello, V.; Falk, H. Monatsh. Chem. 2004, 135, 501.
(9) Adam, W.; Kazakov, D. V.; Kazakov, V. P. Chem. Rev. 2005, 105,
3371. Wahlen, J.; De Vos, D. E.; Jacobs, P. A.; Alsters, P. L. Adv. Synth.
Catal. 2004, 346, 152.
1242
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243

The Journal of Organic Chemistry
Featured Article
(10) Catir, M.; Kilic, H.; Nardello-Rataj, V.; Aubry, J.-M.; Kazaz, C.
J. Org. Chem. 2009, 74, 4560.
(11) Sels, B. F.; De Vos, D. E.; Jacobs, P. A. J. Am. Chem. Soc. 2007,
(30) Davies, A. G.; Sutcliffe, R. J. J. Chem. Soc., Perkins Trans. 2 1981,
1512.
(31) (a) Grob, C. A.; Bolleter, M.; Kunz, W. Angew. Chem., Int. Ed.
Engl. 1980, 19, 708. (b) Grob, C. A. Angew. Chem., Int. Ed. Engl. 1969,
8, 535.
(32) Baader, W. J. In Science of Synthesis; Berkessel, A., Ed.; Georg
Thieme Verlag: Stuttgart, 2009; Vol. 38 (Four-Membered Ring Cyclic
Peroxides), pp 323−344.
(33) Adam, W.; Curci, R.; Edwards, J. O. Acc. Chem. Res. 1989, 22,
205. Murray, R. W. Chem. Rev. 1989, 89, 1187.
(34) All identified conformers are described within the Supporting
Information.
129, 6916.
(12) Nardello, V.; Barbillat, J.; Marko, J.; Witte, P. T.; Alsters, P. L;
Aubry, J.-M. Chem.Eur. J. 2003, 9, 435.
(13) Renirie, R.; Pierlot, C.; Aubry, J.-M.; Hartog, A. F.; Schoemaker,
H. E.; Alsters, P. L.; Wever, R. Adv. Synth. Catal. 2003, 345, 849.
(14) Corey, E. J.; Mehrotra, M. M.; Khan, A. U. J. Am. Chem. Soc.
1986, 108, 2472. Posner, G. H.; Weitzberg, M.; Nelson, W. M.; Murr,
B. L.; Seliger, H. H. J. Am. Chem. Soc. 1987, 109, 278. Cerkovnik, J.;
Tuttle, T.; Kraka, E.; Lendero, N.; Plesnicar, B.; Cremer, D. J. Am.
Chem. Soc. 2006, 128, 4090.
̂
(35) Tame Parreira, R. L.; Galembeck, S. E.; Hobza, P. Chem. Phys.
Chem. 2007, 8, 87.
́
(15) Deslongchamps, P.; Atlani, P.; Frehel, D.; Malaval, A.; Moreau,
(36) The successful alkylation of 1,1-dihydroperoxides using CsOH
also demonstrates the stability of 1,1-dihydroperoxides and
1-alkyldioxy-1-hydroperoxyalkanes toward strongly basic conditions,
although it is interesting to note that ketones are obtained as minor
byproducts in these reactions: Hamada, Y.; Tokuhara, H.; Masuyama,
A.; Nojima, M.; Kim, H.-S.; Ono, K.; Ogura, N.; Wataya, Y. J. Med.
Chem. 2002, 45, 1374.
(37) Payne, G. B. Tetrahedron 1962, 18, 763. Payne, G. B.; Deming,
P. H.; Williams, P. H. J. Org. Chem. 1961, 26, 659.
(38) Long, C. A.; Kearns, D. R. J. Am. Chem. Soc. 1975, 97, 2018.
(39) Frimer, A. A. Chem. Rev. 1979, 79, 359. Gorman, A. A.; Gould,
I. R.; Hamblet, I. J. Am. Chem. Soc. 1982, 104, 7098.
C. Can. J. Chem. 1974, 52, 3651. Abdrakhmanova, A. R.; Khalitova,
L. R.; Spirikhin, L. V.; Dokichev, V. A.; Grabovskiy, S. A.; Kabal’nova,
N. N. Russ. Chem. Bull. Int. Ed. 2007, 56, 271.
(16) Pierlot, C.; Nardello, V.; Schrive, J.; Mabille, C.; Barbillat, J.;
Sombret, B.; Aubry, J.-M. J. Org. Chem. 2002, 67, 2418.
(17) Carreno, M. C.; Gonzal
́
ez-Lopez, M.; Urbano, A. Angew. Chem.,
̃
Int. Ed. 2006, 45, 2737.
1
(18) The half-life of O2 ranges from 1−2 μs in H2O to 60 μs in
CH3CN to nearly milliseconds in perhalogenated solvents. See ref 1a
and Wilkinson, F.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data
1995, 24, 663.
(19) See, for example: Aubry, J.-M.; Adam, W.; Alsters, P. L.; Borde,
C.; Queste, S; Marko, J.; Nardello, V. Tetrahedron 2006, 62, 10753.
Caron, L.; Nardello, V.; Alsters, P. L.; Aubry, J.-M. J. Mol. Catal. A:
Chem. 2006, 251, 194.
(20) Aubry, J.-M.; Pierlot, C.; Rigaudy, J.; Schmidt, R. Acc. Chem. Res.
2003, 36, 668.
(21) Thompson, Q. E. J. Am. Chem. Soc. 1961, 83, 845. Mendenhall,
G. D. In Advances in Oxygenated Processes; Baumstark, A. L., Ed.; JAI:
Greenwich, 1990; Vol. 2, p 203.
(22) Schwartz, C.; Raible, J.; Mott, K.; Dussault, P. H. Org. Lett.
2006, 8, 3199; Tetrahedron 2006, 62, 10747.
(23) Ghorai, P.; Dussault, P. H. Org. Lett. 2009, 11, 4572.
(24) Ghorai, P.; Dussault, P. H. Org. Lett. 2008, 10, 4577.
(25) Cudden, R. C. P.; Hewlett, C. J. Chem. Soc. C 1968, 2983.
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.
(40) Matheson, I. B. C.; Lee, J.; Yamanashi, B. S.; Wolbarsht, M. L.
Chem. Phys. Lett. 1974, 27, 355.
(41) (a) Criegee: Goodman, R. M.; Kishi, Y. J. Am. Chem. Soc. 1998,
120, 9392. Criegee, R. Lieb. Ann. 1948, 560, 127. Kropf, H. In
Methoden der Organischen Chemie, 4th ed.; Kropf, H., Ed.; Thieme:
Stuttgart, 1988; Vol. E13; pp 1095−1099; (b) Hock: Hock, H.; Kropf,
H. Angew. Chem. 1957, 69, 313. Dussault, P. H.; Lee, H. J.; Liu, X.
Perkin 2000, 1, 3006. (c) Anomalous ozonolysis: Jung, M. E.; Davidov,
P. Org. Lett. 2001, 3, 627. (d) Fragmentation of 2-hydroperoxy acids:
Richardson, W. H; Smith, R. S. J. Am. Chem. Soc. 1967, 89, 2230.
(42) Ramirez, A.; Woerpel, K. A. Org. Lett. 2005, 7, 4617.
(43) Goldstein, E.; Beno, B.; Houk, K. N. J. Am. Chem. Soc. 1996,
118, 6036.
(26) Dussault, P. H. In Active Oxygen in Chemistry; Foote, C. S.,
Valentine, J. S., Greenberg, A., Liebman, J. F., Eds.; Blackie A&P:
London, 1995; pp 141−203.
1
(27) The rate constants for reaction of O2 with terpinene and 1,3-
diphenylisobenzofuran (DPBF, D) are on the order of 108 and 109
L/mol-s, respectively; see ref 18. Caution: DPBF is both an efficient
1O2 trap and a visible photosensitizer. In control reactions, we found
that stirring a solution of 1b with DPBF in an unlit fume hood gave
little oxidized product (D-O2) over the typical period (≤0.5 h) of
these reactions. In contrast, care had to be taken to prevent extensive
formation of D-O2 during chromatographic purification of reaction
mixtures under normal laboratory lighting. See: Nowakowska, M.
J. Chem. Soc., Faraday Trans. 1 1984, 80, 2119; Howard, J. A.;
Mendenhall, G. D. Can. J. Chem. 1975, 53, 2199. No precautions were
required with terpinene (T).
(28) Miller, J.; Abernathy, S.; Sharp, R. J. Phys. Chem. A 2000, 104,
4839.
(29) Bis perester derivatives of 1,1-dihydroperoxides have been
prepared: Criegee, R.; Schnorrenberg, W.; Becke, J. Annalen 1949,
565, 7. Warnant, J.; Joly, R.; Mathieu, J.; Velluz, L. Bull. Soc. Chim. Fr.
1957, 331. Milas, N. A.; Golubovic, A. J. Am. Chem. Soc. 1959, 81,
6461. They have also been investigated as polymerization initiators:
Van de Bovenkamp-Bouwman, A. G.; Van Gendt,J. W. J.; Meijer, J.;
Hogt, Andreas H.; Van Swieten, A. P. (Akzo Nobel N.V., Neth.) PCT
Int. Appl. 1999, 42 pp. CODEN: PIXXD2 WO 9932442 A1 19990701.
1243
dx.doi.org/10.1021/jo202265j | J. Org. Chem. 2012, 77, 1233−1243