Revising the Role of a Dioxirane as an Intermediate in the Uncatalyzed Hydroperoxidation of Cyclohexanone in Water

Article abstract of DOI:10.1021/acs.joc.5b00861

The mechanism of the oxidation of cyclohexanone with an aqueous solution of hydrogen peroxide has been investigated. Experiments revealed the preliminary formation of an intermediate, identified as cyclohexylidene dioxirane, in equilibrium with the ketone, followed by formation of 1-hydroperoxycyclohexanol (Criegee adduct). Computational analysis with explicit inclusion of up to two water molecules rationalized the formation of the dioxirane intermediate via addition of the hydroperoxide anion to the ketone and revealed that this species is not involved in the formation of the Criegee adduct. The direct addition of hydrogen peroxide to the ketone is predicted to be favored over hydrolysis of the dioxirane, the latter in competition with ring opening to carbonyl oxide followed by hydration. However, dioxirane may account for the formation of the bis-hydroperoxide derivative.

Full text of DOI:10.1021/acs.joc.5b00861

Article
pubs.acs.org/joc
Revising the Role of a Dioxirane as an Intermediate in the
Uncatalyzed Hydroperoxidation of Cyclohexanone in Water
Elena Rozhko,† Stefania Solmi,† Fabrizio Cavani,† Angelo Albini,‡ Paolo Righi,*,† and Davide Ravelli*,‡
†Department of Industrial Chemistry “Toso Montanari” Alma Mater Studiorum, University of Bologna, Viale del Risorgimento 4,
40136 Bologna, Italy
‡PhotoGreen Lab, Department of Chemistry, University of Pavia, Viale Taramelli 12, 27100 Pavia, Italy
S
* Supporting Information
ABSTRACT: The mechanism of the oxidation of cyclohexanone with an aqueous solution of
hydrogen peroxide has been investigated. Experiments revealed the preliminary formation of an
intermediate, identified as cyclohexylidene dioxirane, in equilibrium with the ketone, followed by
formation of 1-hydroperoxycyclohexanol (Criegee adduct). Computational analysis with explicit
inclusion of up to two water molecules rationalized the formation of the dioxirane intermediate via
addition of the hydroperoxide anion to the ketone and revealed that this species is not involved in the
formation of the Criegee adduct. The direct addition of hydrogen peroxide to the ketone is predicted
to be favored over hydrolysis of the dioxirane, the latter in competition with ring opening to carbonyl
oxide followed by hydration. However, dioxirane may account for the formation of the bis-
hydroperoxide derivative.
hydrogen peroxide and tert-butyl hydroperoxide have also been
used. In comparison with peroxyacids, which are often
explosive and form carboxylic acid as co-product at any rate,
the cheaper and safer hydrogen peroxide in the presence of a
catalyst is obviously advantageous (higher content of active
oxygen, water as the only co-product).
For the mechanism, Baeyer and Villiger proposed a dioxirane
(III in Scheme 1b) as an intermediate in the original paper,2
while Criegee suggested the involvement of perester IVa.4
Later, von Doering and Dorfman excluded the formation of a
dioxirane in the BV oxidation with peroxybenzoic acid,5 while
Curci and co-workers claimed its involvement when the
bis(trimethylsilyl) derivative of peroxymonosulfuric acid
(H2SO5, Caro’s acid) was used as the oxidant.6 However,
NMR analysis showed that dioxiranes survived only a few
minutes at low temperature (from −20 to −10 °C, depending
on the actual structure) in chloroform.7
The convenience of the hydrogen peroxide choice fostered
several recent investigations, with particular regard to the role
of intermediates and the mechanism. Thus, cyclohexanone was
reported to react with H2O2 and give the Criegee adduct 1-
hydroperoxycyclohexanol (IVb; R = R′ = −(CH2)5−) at room
temperature8a as well as ε-caprolactone at 90 °C in the
presence of a heterogeneous catalyst (Sn-zeolite beta) in
dioxane.8b
INTRODUCTION
■
Intermediates in oxygen transfer reactions have been the
subject of intensive research and some controversy, in
particular, for the role of dioxiranes.1 Apart from that case,
such species have not often been considered. An important
exception is the versatile Baeyer−Villiger (BV) oxidation,
largely used in organic synthesis, particularly for the preparation
of pharmaceuticals, such as antibiotics, steroids, and pher-
omones, as well as of fine chemicals and intermediates for the
chemical industry.2,3 The reaction consists of the oxidation of a
carbonyl group (I) to give an ester (II) via insertion of oxygen
from a peroxyl derivative (X−OOH; Scheme 1a). m-Chloro-
peroxybenzoic acid, peroxytrifluoroacetic acid, peroxybenzoic
acid, and peroxyacetic acid are typically used as the reagent, but
Scheme 1. (a) Baeyer−Villiger Oxidation of Ketones to
Esters and (b) Proposed Intermediates Involved in the
Process
Further products formed from ketones and hydrogen
peroxide included 1,1-dihydroperoxycyclohexane (from cyclo-
hexanone; V) as well as other cyclic bis-hydroperoxides.9 The
same derivatives were also prepared by ozonolysis of vinyl
10
ethers in the presence of H2O2 or by reaction of ketals with
Received: April 17, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b00861
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
hydrogen peroxide in the presence of tungstic acid11 or in
anhydrous ether under BF3 catalysis.12 Condensation of bis-
hydroperoxides leading to dihydroperoxydicycloalkyl peroxides
and tetraoxadispiroalkanes (VI) or ozonides was also
reported.13−15
The formation of the Criegee adduct (IV) from ketone (I)
and the ensuing rearrangement of IV to the final ester (II) has
been analyzed computationally,16−19 while, to the best of our
knowledge, the involvement of dioxiranes (III) has never been
considered.
In view of the general interest for mild oxidations, it appeared
worthwhile to establish the precise role, if any, of the dioxirane
and the relation among the intermediates involved, as we report
in the following through a combined experimental/computa-
tional investigation for the paradigmatic case of the non-
catalyzed reaction between cyclohexanone (1) and aqueous
hydrogen peroxide (Scheme 2).
the ΔG of solvation, calculated as recommended by the
Gaussian 09 software, viz. by using the SMD model23 by
Truhlar and co-workers at the B3LYP/6-31G(d,p) level of
theory.
The computed enthalpy, entropy, and Gibbs free energy
were converted from the 1 atm standard state into the standard
state of molar concentration (ideal mixture at 1 mol L−1 and 1
atm) in order to allow a direct comparison with the
experimental results in water solution. Thus, the contribution
RT ln R′T, where R′ is the value of R in L atm K−1 mol−1 had to
be added to the Gibbs free energy term.24 This contribution
always cancels out unless a process where a molecularity change
(Δn) between reagents and products occurs. Accordingly, this
contribution should be written as ΔnRT ln R′T. As an example,
in the reaction A + B → C, Δn = −1 and the contribution will
be −RT ln R′T (−1.90 kcal mol−1 at 298.150 K).
Importantly, the DFT level of theory has only been adopted
for preliminary and screening purposes, while all the Gibbs free
energy data reported in the text do refer to the CBS-QB3 level
of theory, also including the solvent effect (named “SMD-CBS-
QB3”).
Scheme 2. Process Investigated in the Present Work
RESULTS AND DISCUSSION
■
Experimental Studies. Cyclohexanone 1 (98 mg, 1 mmol;
1.85 M) was dissolved in water (100 μL) and then was treated
with 3 molar equiv of H2O2 in aqueous solution (30%; 340 μL,
3 mmol). Monitoring by ESI-MS revealed that the Criegee
adduct 2 (peak + Na+, m/z = 155) was formed immediately
after mixing the reagents and increased with time. A further
peak compatible with cyclohexylidene dioxirane (hereafter
termed 4) structure (m/z = 115) was formed, but there were
no oligomeric compounds. After 2 h, at the end of the reaction,
the aqueous reaction mixture was extracted with 0.8 mL of
CDCl3. The organic phase was then separated with the aid of a
membrane phase separator, and its analysis by NMR showed
the formation of the Criegee adduct 2 only (Figure S1 in
Supporting Information).8a
2,2,6,6-Tetradeuterocyclohexanone (1-d4) gave the corre-
sponding ESI-MS peaks for 2-d4 and 4-d4 and followed a
practically identical kinetics (see Figure S2), thus supporting
that α-H atoms were not involved in the rate-determining step
of the reaction. The experimental activation energy was
measured to be 14 2 kcal mol−1 in both cases.
A second set of experiments was run directly in a NMR tube
on the 1-13C-cyclohexanone (1-13C; see the corresponding C
NMR spectrum in Figure S3; peak at 220 ppm)/H2O2 system
by recording 13C NMR spectra at regular intervals at 50 °C in
D2O (with ≈0.1 M 1-13C starting concentration and all other
parameters unchanged with respect to batch experiments; no
effect due to the concentration change, except for a slower
reaction). A new compound was formed within the 5 min
required for recording the first spectrum (see inset in Figure 1
and Figure S4 for details). This showed signals conserving the
cyclohexanone symmetry and very close to those reported for
cyclohexylidene dioxirane at low temperature.7 NMR spectra
were run at regular intervals over 24 h (see Figure S5) and
showed that, after “immediate” initial fast formation of the
dioxirane (4-13C; peak at 103 ppm), the Criegee adduct (2-13C;
peak at 110 ppm) was formed in several hours. Separate
integration of the 13C-enriched peaks at 220, 110, and 103 ppm
evidenced (Figure 1) that starting cyclohexanone and dioxirane
were initially in a ratio of about 1:1 and that this ratio was
COMPUTATIONAL DETAILS
■
All of the simulations have been carried out with the Gaussian
09 software.20 Reactants, intermediates, transition state
structures, and products have been initially optimized having
recourse to density functional theory (DFT), viz. adopting the
B3LYP functional21 and the standard 6-31G(d,p) basis set in
the gas phase. To confirm the nature of the stationary points,
vibrational frequencies (in the harmonic approximation) have
been calculated for all of the optimized structures at the same
level of theory as geometry optimizations, and it was verified
that local minima had only real frequencies, while transition
states (TSs) were identified by the presence of a single
imaginary frequency corresponding to the expected motion
along the reaction coordinate. Unscaled results from the
frequency calculations have been used to compute zero-point
energies and their thermal corrections, enabling the calculation
of the Gibbs free energy of the system considered.
For each reacting situation, a systematic investigation on all
the possible conformations has been carried out by running
several constrained optimizations. In most cases, the con-
formers differed for the arrangement of the O−H groups not
involved in the reaction. However, since the difference in
energy of these conformers was small (in most cases, a few kcal
mol−1), only the most stable conformation has been considered
for further work.
In each case, the most stable conformation has been used as
the starting point for a refinement at a higher level of theory,
viz. using the composite method CBS-QB322 in the gas phase.
The CBS-QB3 approach is part of the so-called “complete basis
set” (CBS) methods of Petersson and co-workers for
computing very accurate energies.
With both levels of theory (DFT and CBS-QB3), Gibbs free
energies (G) obtained in vacuo have been corrected by adding
B
DOI: 10.1021/acs.joc.5b00861
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
The first situation examined was indicated as “N-0″, where N
stands for neutral (only noncharged species involved) and “0”
identifies the number of water molecules included explicitly
into the optimization (see Figures 2 and 3a for details). In this
model, the proximal oxygen (OP; red color in Scheme 3)
became linked to the carbonyl group, giving the dioxirane,
while the distal oxygen (OD; blue) and HP (red) gave the co-
product of the reaction, viz. a water molecule. This process
occurred with an activation Gibbs free energy (ΔG*) of more
than 50 kcal mol−1, independently from the trajectory adopted.
The inclusion of an explicit water molecule (Figure 2;
pathway “N-1”) resulted in the formation of dioxirane through
a five-membered cyclic transition state. However, ΔG*
remained >50 kcal mol−1 in this case, and the same held for
“N-2” pathway (Figure 2), with an activation Gibbs free energy
around 45 kcal mol−1. As apparent from Figure 2, the geometric
parameters of these three TSs vary little from each other, with
the bonds involved in dioxirane formation, viz. the two C−O
bonds and the O−O bond, comprised in a narrow range (<0.1
Å difference). The situation including two molecules of water
was then examined by considering different arrangements of
the reacting cluster, with a water molecule acting in the role of
proton relay (as in pathway N-1) and the other coordinated via
hydrogen bonding to either the distal (N-2D pathway) or the
carbonyl oxygen (N-2C pathway). Activation Gibbs free
energies consistently around 50 kcal mol−1 were again obtained.
Hydrogen peroxide is slightly more acidic than water
(reported pKa value at 25 °C of 11.6)25 and is partially
dissociated in the reaction mixture, so that a non-negligible
amount of the hydroperoxide anion is present in solution.
Thus, the reaction of cyclohexanone with HOO− to give the
dioxirane, hereafter termed as “anionic pathway” and indicated
as “A”, was likewise considered. This process was slightly more
endergonic with respect to the neutral case (+ 7.73 kcal mol−1).
First, we examined whether this reaction occurred either via a
two-step mechanism, involving the initial nucleophilic addition
of the anion to the carbonyl to give a tetrahedral intermediate
(the 1-hydroperoxycyclohexanolate anion) later evolving into
the product, or in a single step. However, any effort to locate
such a structure was unsuccessful, and only a sort of loose
complex was found, the energy of which was close to the sum
of the energies of cyclohexanone and hydroperoxide anion
alone (data not shown).
Figure 1. Relative amounts of starting cyclohexanone (1-13C), Criegee
adduct (2-13C), and dioxirane (4-13C), as determined from integration
of the corresponding 13C signals. Inset: Partial spectrum registered
immediately after adding hydrogen peroxide (only 13C-enriched
carbon portion shown).
maintained throughout, while the Criegee adduct increased
steadily.
That 4 was an intermediate on the way to 2 contrasted,
however, with the purported instability of the dioxirane,
reported not to survive longer than a few minutes in CDCl3
at room temperature.7 We confirmed this finding, as outlined in
the Supporting Information (see section 1.1), and showed that
in CDCl3 the dioxirane either reverted back to starting 1 or was
destroyed in a few minutes. We found, however, that it was
stabilized by the water environment, and NMR monitoring
showed that it remained unchanged for as long as 2 days.
Computational Study. These findings demanded a
computational study on the role of 4 and its relation to 2
and 3, as summarized below (values at the SMD-CBS-QB3
level of theory are reported throughout).
Dioxirane 4 Formation. Theoretical simulations predicted
that dioxirane formation from cyclohexanone was a slightly
endergonic process, with an associated Gibbs free energy
change of +1.64 kcal mol−1. In this and other cases (see below),
attack to non-equivalent faces of the carbonyl group in the
chairlike configuration of 1, viz. axial (“AX”) and equatorial
(“EQ”; see Scheme 3), was considered. A realistic scenario was
obtained by incorporating explicitly a network of water
molecules (in the role of proton relays) in the optimization
of the reacting system18 and then including the solvent effect
(as an implicit model) via single-point calculations.
Adopting the same approach as above, we devised several
scenarios and found that this process occurs with a markedly
lower activation Gibbs free energy with respect to the neutral
paths described above. Thus, path “A-0” (see Figures 2 and 3b
for details) showed activation Gibbs free energies of ca. 19 and
25 kcal mol−1 for the axial and the equatorial trajectories,
respectively. Introducing a molecule of water did not lead to a
cyclic TS, but we checked the effect of coordination of the
distal oxygen, where the most important structural changes
were occurring (pathway “A-1”). This caused a further decrease
of ΔG*, with both AX ad EQ pathways dropping to ca. 14 kcal
mol−1. For the geometric aspect, it is worth noting that the TSs
belonging to the anionic pathway more resemble the situation
of the final products (late TS), where this feature is slightly
more pronounced in the A-0 path than in the A-1 one (see
Figure 2).
Scheme 3. Different Modes of Approach to the
Cyclohexanone Molecule Considered in the Present Study
and the Labeling Adopted for the Hydrogen Peroxide
Molecule
As expected when large complexes are considered (e.g., the
TSs pertaining to the “2” classes) as well as with charged
species, the energy of both complexed reagents and products
was much higher than the sum of isolated compounds (see
C
DOI: 10.1021/acs.joc.5b00861
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 2. Optimized transition state structures for selected reacting situations describing the formation of the dioxirane intermediate. The bonds
depicted with a blue dotted line and a green dashed line represent, respectively, those bonds formed and broken during the process. The
corresponding bond lengths are reported (in Å), adopting the same color coding; the length of the C−O bond of the original carbonyl group is
likewise reported in gray color.
Figure 3. Energy profile at the SMD-CBS-QB3 level for the pathways: (a) “N” and (b) “A”. In each case, the energy of complexed reagents has been
taken as the zero-point of the energy scale. Further notice that pathways N-2D and N-2C have been omitted for clarity (see Table 1).
Table 1), but these contributions canceled out in the
calculation of ΔG* values.
have involved the same (already excluded) anionic intermediate
1-hydroperoxycyclohexanolate (see above for details).
For the conversion of dioxirane 4 to the Criegee adduct 2
(pathway “D”), this involved the hydrolysis of the three-
membered ring that could occur via cleavage of either the O−O
or one of the two (non-equivalent) C−O bonds of the
dioxirane moiety. As in the case of cyclohexanone, the usual
AX/EQ dichotomy has been confronted also with this
intermediate (see Scheme 3). For the O−O cleavage pathway,
we found several TSs (the corresponding pathways have been
named as “DO”), but none of them pertained to the direct
process (no TS DO-0 structure found). On the other hand, the
corresponding structures including one or two molecules of
water acting as proton relays ranked activation Gibbs free
energies around 48 and 34 kcal mol−1, respectively. This
supported the role of water in facilitating this rearrangement
(see Figure 4b).
Criegee Adduct 2 Formation. Next, the formation of the
Criegee adduct was considered. This may be formed either
from the starting ketone 1 or from the dioxirane intermediate 4.
Both processes are slightly exergonic, −0.94 and −2.57 kcal
mol−1, respectively. Formation of the Criegee adduct 2 from
cyclohexanone 1 (pathway “K”) can be described as the formal
addition of a molecule of hydrogen peroxide to the carbonyl
moiety.
The uncatalyzed pathway (“K-0” path; already reported in
the literature)19c occurs with an activation Gibbs free energy
around 40 kcal mol−1, with the axial trajectory favored by ca. 2
kcal mol−1 over the equatorial one. By contrast, the inclusion of
one (K-1) or two (K-2) molecules of water resulted in the
location of the sought for cyclic transition states, with activation
Gibbs free energies around 29 and 23 kcal mol−1, respectively,
when considering the most favorable trajectories (see Figure
4a). The corresponding anionic pathway involving the
hydroperoxide anion was not investigated because this would
In contrast, all the attempts to identify a similar process
involving the cleavage of either of the C−O bonds in the
dioxirane moiety failed because the system evolved toward the
formation of a carbonyl oxide intermediate (named 5; this
D
DOI: 10.1021/acs.joc.5b00861
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
pathway is hereafter termed as “DC”). Again, both the axial and
the equatorial pathways describing 5 formation occurred with
activation Gibbs free energies around 37 kcal mol−1. The
carbonyl oxide, a valence isomer of the dioxirane, underwent
hydration to give the Criegee adduct 2 with very low activation
Gibbs free energies (around 10 kcal mol−1 in the most favorable
case; see the reacting situations labeled with “DC2” in Table 1).
Bis-hydroperoxide 3 Formation. Finally, the formation of
the α,α-bis-hydroperoxide (3) has been considered. Kinetic
considerations support that 3 does not arise from the ketone
since this would involve addition of two H2O2 molecules at the
same time, a kinetically unlikely process, but rather from the
dioxirane (pathway “DB”; exergonic with a ΔG = −8.54 kcal
mol−1). However, any attempt to depict either the O−O or the
dioxirane C−O bond cleavage in the presence of hydrogen
peroxide and variable amounts of water molecules (from 0 to
2) failed. It appeared likely that the course of this reaction was
similar to that observed for the Criegee adduct; that is, it
involved the carbonyl oxide intermediate 5, which further
underwent addition of a hydrogen peroxide molecule to give
the final bis-hydroperoxide (however, this step, named DC3, was
not investigated in detail).
Summarizing the Mechanism. A plausible reaction
mechanism supported by both theoretical simulations and
experimental data is reported in Scheme 4. For dioxirane 4
formation, two different paths can take place, viz. addition of
neutral H2O2 or its anionic form HOO− to cyclohexanone 1.
Under the present reaction conditions, the equilibrium between
the hydrogen peroxide molecule and the hydroperoxide anion
(concentration estimated ca. 4 × 10−6 M under the present
conditions) can be safely considered fast, due to the aqueous
environment facilitating proton exchange. Accordingly, the
observed reactivity can be rationalized via the Curtin−
Hammett principle,26 that is, by considering the difference in
energy of the representative transition states. Thus, dioxirane
results from a kind of nucleophilic addition onto the carbonyl
by the hydroperoxide anion (Scheme 4, path A). The huge
activation Gibbs free energy observed for the corresponding
addition of neutral hydrogen peroxide (path N) militates
against the direct intervention of such species in the reaction
(kA ≫ kN). Calculations likewise excluded the formation of the
1-hydroperoxycyclohexanolate anion. The coincidence between
the experimental activation energy for cyclohexanone con-
sumption (ca. 14 kcal mol−1) and the calculated activation
Table 1. Selected Calculated Parameters for the Reacting
Situations Described in the Text
a
Gibbs free energy (G), [kcal mol−1
]
b
reacting
situation
complexed
reagents
complexed
products
ΔG*
TS
[kcal mol−1
]
N-0AX
N-0EQ
N-1AX
N-1EQ
N-2AX
N-2EQ
N-2DAX
N-2DEQ
N-2CAX
N-2CEQ
A-0AX
3.02
1.90
56.02
54.73
62.73
61.29
60.85
58.77
68.32
67.89
66.25
64.72
34.32
35.21
33.32
29.82
42.52
43.49
36.87
36.96
36.37
35.61
59.94
59.55
52.91
50.44
36.87
36.62
18.16
18.38
19.38
18.99
26.76
24.86
6.84
6.76
53.00
52.83
55.31
52.25
46.96
44.63
55.06
53.72
52.49
49.97
19.33
25.48
14.02
13.34
39.79
41.60
29.43
28.69
23.44
23.49
47.88
48.49
35.23
33.48
36.87
36.62
14.14
14.31
10.63
10.07
13.14
10.68
7.42
12.47
12.37
18.89
17.67
19.12
20.73
18.14
18.03
34.70
33.40
31.29
29.90
9.04
13.89
14.14
13.26
14.17
13.76
14.75
14.99
9.73
A-0EQ
A-1AX
19.30
16.39
2.73
A-1EQ
K-0AX
K-0EQ
K-1AX
1.89
7.44
5.70
5.82
K-1EQ
K-2AX
8.27
12.93
12.12
12.06
11.06
17.68
16.96
10.55
10.14
4.29
K-2EQ
DO-1AX
DO-1EQ
DO-2AX
DO-2EQ
4.34
14.20
9.93
DC
DC
16.66
16.66
−19.06
−19.16
−13.55
−13.75
−10.22
−10.22
AX
EQ
DC2-0AX
DC2-0EQ
DC2-1AX
DC2-1EQ
DC2-2AX
DC2-2EQ
4.02
4.07
8.75
8.92
13.62
14.18
a
A more comprehensive list is available in Supporting Information
Table S1. Gibbs free energies at the SMD-CBS-QB3 level of theory
have been reported. In each case, the sum of the Gibbs free energies of
the involved reagents has been taken as the zero-point of the energy
b
scale. ΔG* values have been calculated by taking the difference
between the energy of TS and the corresponding complexed reagents.
Figure 4. Energy profile at the SMD-CBS-QB3 level for the pathways: (a) “K” and (b) “D”. In each case, the energy of complexed reagents has been
taken as the zero-point of the energy scale (see Table 1).
E
DOI: 10.1021/acs.joc.5b00861
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
peroxidation of cyclohexanone with hydrogen peroxide in
water. This is the first formed intermediate that is in fast
equilibrium with the starting ketone. The computational
analysis fully supports this hypothesis and provides a scenario
in excellent agreement with experimental data. The demon-
stration that the hydroperoxide anion (not hydrogen peroxide)
is involved in dioxirane formation suggests that this mechanism
is operative only in neutral and/or basic conditions. Another
important aspect is the reaction medium because water is
apparently required to stabilize the dioxirane.
Scheme 4. Proposed Reaction Mechanism
This mechanism can be easily adapted to rationalize the
formation of all of the intermediates reported in Scheme 1b
because dioxirane could act as a sort of reservoir of “oxidized”
ketone and may be involved in the formation of oligomeric
species, formally arising via addition of several hydrogen
peroxide molecules to the starting ketone and ensuing
condensation (e.g., species V, VI, and higher homologues).
By contrast, since the formation of Criegee adduct 2 occurs
directly from 1, the parallel pathway through dioxirane 4
actually slows the formation of 2. The scenario may be
different, however, under catalyzed conditions, where the
present mechanistic evidence may help in the search for a
catalyst, which should be tailored in order to intercept the
dioxirane intermediate.
a
The energy changes (ΔG; expressed in kcal mol−1) for selected
pathways at the SMD-CBS-QB3 level of theory have been likewise
reported.
Gibbs free energy for addition of the hydroperoxide anion to
cyclohexanone further supports the present proposal.
EXPERIMENTAL SECTION
Experiments suggest that dioxirane 4 and starting ketone 1
are in fast equilibrium with each other (1/4 ratio slightly >1),
and both of them may be responsible for the formation of the
Criegee adduct. Indeed, theoretical calculations support that
addition of hydrogen peroxide to the ketone (path K) is more
favored than hydrolysis of the dioxirane intermediate. If
operating, dioxirane hydrolysis may occur directly through
O−O bond cleavage (path DO) or involve a further
intermediate, viz. the carbonyl oxide 5, through C−O bond
cleavage (path DC) and ensuing (facile) addition of water (path
DC2). Thus, considering that the ΔG* values for paths DO and
DC (the rate-determining step on the way from 1 to 2 through
5) are similar (in the 33 to 37 kcal mol−1 range), the conversion
of 4 to 2 may occur through either of them, with the first
slightly favored over the latter. By contrast, the ΔG* value for
path K is approximately 10 kcal mol−1 lower (around 23 kcal
mol−1), indicating that the formation of 2 occurs preferentially
■
Typical Conditions for Batch Reactions. In a 10 mL Pyrex
reactor equipped with screw stopper, a mixture of 100 μL of H2O, 3
mmol 30% H2O2 (340 μL), and 1 mmol cyclohexanone (98 mg) was
dropped in. The resulting mixture was then stirred at 50 °C for 2 h.
Experimental Details. Cyclohexanone, 2,2,6,6-tetradeuterocyclo-
hexanone (98 atom % D), and 1-13C-cyclohexanone (99 atom % 13C)
were commercially available and used as received.
1H and 13C NMR spectra were recorded at 300 and 75 MHz or at
600 and 150 MHz, respectively, at 25 °C, except where otherwise
noted. Chemical shifts (δ) for 1H and 13C are given in parts per million
relative to residual signals of the solvents. CDCl3 was passed over a
short pad of alumina before use.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedure for attempted dioxirane isolation and
full details about computational data. The Supporting
Information is available free of charge on the ACS Publications
website at DOI: 10.1021/acs.joc.5b00861.
O
C
from 1 and not from 4, that is, kK ≫ kD > kD .
Formation of the bis-hydroperoxide presumably involves
again the dioxirane, likewise via the carbonyl oxide, then
undergoing addition of hydrogen peroxide (via paths DC, DC3),
while no evidence was found for the direct reaction of dioxirane
with H2O2 (path DB). The involvement of a carbonyl oxide
(also called “Criegee intermediate” in the literature, as opposed
to “Criegee adduct” 2)27 in the present reacting system also
finds support from the literature since ozonolysis of vinyl ethers
in the presence of H2O2,10 where the involvement of a carbonyl
oxide intermediate is generally accepted, leads to the same
geminal bis-hydroperoxides proposed here.
AUTHOR INFORMATION
Corresponding Authors
*Tel: +39 051 2093640. E-mail: paolo.righi@unibo.it.
*Fax: +39 0382 987323. Tel: +39 0382 987198. E-mail: davide.
ravelli@unipv.it.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The experimental data presented above also allowed us to
exclude mechanisms different from that depicted in Scheme 4,
This work was funded by the CINECA Supercomputer Center,
with computer time granted by ISCRA project HETFRA
(code: HP10CSWWDQ). D.R. thanks prof. P. Caramella and
R. Gandolfi (University of Pavia) for fruitful discussions.
28
such as radical reactions via HO• or HOO• since deuterium
atoms in α do not affect the kinetics29 and are conserved in the
Criegee adduct formed from 1-d4.
REFERENCES
■
CONCLUSIONS
In conclusion, the present work strongly supports the
involvement of cyclohexylidene dioxirane in the hydro-
■
(1) Murray, R. W. Chem. Rev. 1989, 89, 1187−1201.
(2) Baeyer, A.; Villiger, V. Ber. Dtsch. Chem. Ges. 1899, 32, 3625−
3633.
F
DOI: 10.1021/acs.joc.5b00861
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(3) For reviews, see: (a) Renz, M.; Meunier, B. Eur. J. Org. Chem.
1999, 737−750. (b) ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R.
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
́ ́
A. Chem. Rev. 2004, 104, 4105−4124. (c) Jimenez-Sanchidrian, C.;
Ruiz, J. R. Tetrahedron 2008, 64, 2011−2026. (d) Michelin, R. A.;
Sgarbossa, P.; Scarso, A.; Strukul, G. Coord. Chem. Rev. 2010, 254,
646−660.
(4) Criegee, R.; Kaspar, R. Liebigs Ann. Chem. 1948, 560, 127−135.
(5) von Doering, W.; Dorfman, E. J. Am. Chem. Soc. 1953, 75, 5595−
5598.
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision D.01; Gaussian, Inc.: Wallingford, CT, 2009.
(21) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(b) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623−11627.
(6) Camporeale, M.; Fiorani, T.; Troisi, L.; Adam, W.; Curci, R.;
Edwards, J. J. Org. Chem. 1990, 55, 93−98.
(22) Montgomery, J. A., Jr.; Frisch, M. J.; Ochterski, J. W.; Petersson,
G. A. J. Chem. Phys. 1999, 110, 2822−2827.
(7) Murray, R. W.; Singh, M.; Jeyaraman, R. J. Am. Chem. Soc. 1992,
114, 1346−1351.
(23) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B
2009, 113, 16378−6396.
(8) (a) Prechter, A.; Heinrich, M. R. Synthesis 2011, 10, 1515−1525.
(b) Corma, A.; Nemeth, L. T.; Renz, M.; Valencia, S. Nature 2001,
412, 423−425.
(24) Benson, S. Thermochemical Kinetics; Wiley: New York, 1968; p
8.
(9) (a) Ledaal, T.; Solbjor, T. Acta Chem. Scand. 1967, 21, 1658−
̈
(25) Douglas, D. D. Ionisation Constants of Inorganic Acids and Bases,
2nd ed.; Pergamon Press: New York, 1982.
1659. (b) Terent’ev, A. O.; Platonov, M. M.; Kashin, A. S.; Nikishin,
G. I. Tetrahedron 2008, 64, 7944−7948.
(26) Seeman, J. I. Chem. Rev. 1983, 83, 83−134.
(27) (a) Bailey, P. S. Ozonolysis in Organic Chemistry; Academic
Press: New York, 1978. (b) Donahue, N. M.; Drozd, G. T.; Epstein, S.
A.; Presto, A. A.; Kroll, J. K. Phys. Chem. Chem. Phys. 2011, 13, 10848−
10857.
(28) For a previous work on this topic, see: Cavani, F.; Raabova, K.;
Bigi, F.; Quarantelli, C. Chem.Eur. J. 2010, 16, 12962−12969.
(29) Mendes, J.; Zhou, C.-W.; Curran, H. J. J. Phys. Chem. A 2013,
117, 4515−4525.
(10) (a) Kim, H.-S.; Tsuchiya, K.; Shibata, Y.; Wataya, Y.; Ushigoe,
Y.; Masuyama, A.; Nojima, M.; McCullough, K. J. J. Chem. Soc., Perkin
Trans. 1 1999, 1867−1870. (b) Tsuchiya, K.; Hamada, Y.; Masuyama,
A.; Nojima, M.; McCullough, K. J.; Kim, H.-S.; Shibata, Y.; Wataya, Y.
Tetrahedron Lett. 1999, 40, 4077−4080.
(11) Jefford, C. W. Synth. Commun. 1990, 20, 2589−2596.
(12) (a) Terent’ev, A. O.; Kutkin, A. V.; Platonov, M. M.; Ogibin, Y.
N.; Nikishin, G. I. Tetrahedron Lett. 2003, 44, 7359−7363.
́
(b) Terentev, A. O.; Kutkin, A. V.; Platonov, M. M.; Vorontsov, I.
I.; Antipin, M. Y.; Ogibin, Y. N.; Nikishin, G. I. Russ. Chem. Bull. 2004,
53, 681−687.
(13) (a) Jefford, C. W. In Advances in Drug Research; Testa, B.,
Meyer, U. A., Eds.; Academic Press: San Diego, CA, 1997; Vol. 29, pp
271−325. (b) Dong, Y.; Matile, H.; Chollet, J.; Kaminsky, R.; Wood, J.
K.; Vennerstrom, J. L. J. Med. Chem. 1999, 42, 1477−1480.
(14) (a) Schenck, G. O.; Becker, H. D. Angew. Chem. 1958, 70, 504.
(b) Rieche, A.; Meister, K. Ber. Dtsch. Chem. Ges. 1939, 72, 1938−
1940.
(15) (a) Busch, P.; Story, P. R. Synthesis 1970, 181−183. (b) Story,
R.; Busch, P. In Advances in Organic Chemistry: Methods and Results;
́
Taylor, E. C., Ed.; Wiley: New York, 1972; Vol. 8, p 67. (c) Terentev,
A. O.; Kutkin, A. V.; Platonov, M. M.; Starikova, Z. A.; Ogibin, Y. N.;
Nikishin, G. I. Russ. Chem. Bull. 2005, 54, 1214−1218.
(16) (a) Stoute, V. A.; Winnik, M. A.; Csizmadia, I. G. J. Am. Chem.
Soc. 1974, 96, 6388−6393. (b) Hannachi, H.; Anoune, N.; Arnaud, C.;
́
Lanteri, P.; Longeray, R.; Chermette, H. J. Mol. Struct. (THEOCHEM)
1998, 434, 183−191.
(17) (a) Alvarez-Idaboy, J. R.; Reyes, L.; Cruz, J. Org. Lett. 2006, 8,
1763−1765. (b) Alvarez-Idaboy, J. R.; Reyes, L.; Mora-Diez, N. Org.
Biomol. Chem. 2007, 5, 3682−3689. (c) Reyes, L.; Alvarez-Idaboy, J.
R.; Mora-Diez, N. J. Phys. Org. Chem. 2009, 22, 643−649. (d) Mora-
Diez, N.; Keller, S.; Alvarez-Idaboy, J. R. Org. Biomol. Chem. 2009, 7,
3682−3690. (e) Itoh, Y.; Yamanaka, M.; Mikami, K. J. Org. Chem.
2013, 78, 146−153.
(18) Yamabe, S.; Yamazaki, S. J. Org. Chem. 2007, 72, 3031−3041.
(19) (a) Carlqvist, P.; Eklund, R.; Brinck, T. J. Org. Chem. 2001, 66,
1193−1199. (b) Sever, R. R.; Root, T. W. J. Phys. Chem. B 2003, 107,
10848−10862. (c) Boronat, M.; Corma, A.; Renz, M.; Sastre, G.;
Viruela, P. M. Chem.Eur. J. 2005, 11, 6905−6915. (d) Long, Q.; Ji,
H.; Lu, S. J. Mol. Struct. (THEOCHEM) 2009, 901, 117−127. (e) Xu,
S.; Wang, Z.; Li, Y.; Zhang, X.; Wang, H.; Ding, K. Chem.Eur. J.
2010, 16, 3021−3035.
(20) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
G
DOI: 10.1021/acs.joc.5b00861
J. Org. Chem. XXXX, XXX, XXX−XXX