Ozonolysis of bicyclic 1,2-dioxines: Initial scope and mechanistic insights

Article abstract of DOI:10.1021/jo3001518

The ozonolysis of bicyclic 1,2-dioxines was investigated using a variety of 1,4-disubstituted 1,2-dioxines along with a 1,3-dialkyl and steroidal example, with yields ranging from moderate to excellent. Two different pathways were observed upon reaction of the 1,4-disubstituted 1,2-dioxines with ozone; one pathway saw the "expected" results, that is, cleavage of the olefinic moiety with generation of 1,4-dicarbonyl 1,2-dioxines, while the other pathway revealed a previously unobserved rearrangement involving cleavage of the peroxide linkage along with loss of either CO or CO₂. Several unsymmetrical ozonolyses were also performed to further investigate the origins of this rearrangement, and initial mechanistic insights into the fragmentation pathways are discussed.

Full text of DOI:10.1021/jo3001518

Article
pubs.acs.org/joc
Ozonolysis of Bicyclic 1,2-Dioxines: Initial Scope and Mechanistic
Insights
Nicole M. Cain,† Edward R. T. Tiekink,‡ and Dennis K. Taylor*,†
†School of Agriculture, Food and Wine, The University of Adelaide, South Australia 5005, Australia
‡Department of Chemistry, University of Malaya, Kuala Lumpur, 50603, Malaysia
S
* Supporting Information
ABSTRACT: The ozonolysis of bicyclic 1,2-dioxines was
investigated using a variety of 1,4-disubstituted 1,2-dioxines
along with a 1,3-dialkyl and steroidal example, with yields
ranging from moderate to excellent. Two different pathways
were observed upon reaction of the 1,4-disubstituted 1,2-
dioxines with ozone; one pathway saw the “expected” results, that is, cleavage of the olefinic moiety with generation of 1,4-
dicarbonyl 1,2-dioxines, while the other pathway revealed a previously unobserved rearrangement involving cleavage of the
peroxide linkage along with loss of either CO or CO2. Several unsymmetrical ozonolyses were also performed to further
investigate the origins of this rearrangement, and initial mechanistic insights into the fragmentation pathways are discussed.
synthesized in 15 steps from levopimaric acid in an overall yield
of 2.7%.34 This synthesis involved the ozonolysis of bicyclic 1,2-
dioxine 1 to furnish keto-aldehyde 2 (Scheme 1), in a near
quantitative crude yield, although this paper’s discussion placed
little emphasis on either the reaction or details of the stability or
characterization of the resulting dicarbonyl product.
INTRODUCTION
■
Cyclic peroxides are an important class of organic compounds
classified by their weak O−O linkage. They are abundant in
nature with many natural products being isolated that exhibit a
wide spectrum of biological activities including antimalarial,1
antifungal2,3 and cytotoxic activities against cancer cells.4 1,2-
Dioxines, also known as endoperoxides, are a specific type of
cyclic peroxide, characterized by an unsaturated six-membered
peroxide ring. Synthetically, they have proven to be extremely
versatile starting materials with extensive research highlighting
numerous examples whereby the peroxide bond is either
cleaved5−12 or maintained12−18 during the course of further
reaction. Continuing our studies into the chemistry of 1,2-
dioxines, we have now turned our attention to the products that
result, and the possible synthetic utility generated, from the
ozonolysis of 1,2-dioxines.
The ozonolysis of alkenes, first reported in 1840, remains
one of the most important and classical methods for oxidative
cleavage of alkenes.19 Since the basic mechanism was
formulated by Criegee in the mid-1950s,20,21 a large amount
of work has been done on the mechanism, although not all
details are fully understood.22−25 While the Criegee mechanism
has become generally accepted, a number of exceptions to this
mechanism have appeared, prompting numerous “modified-
Criegee” mechanisms to be reported.26−28 Decomposition of
the ozonides that result from this reaction can be carried out
using a variety of different reagents, to furnish a number of
functional products, with triphenylphosphine (PPh3) being a
widespread reagent for their reduction resulting in the
formation of dicarbonyl products.29,30 Despite all this attention,
little precedent exists for the ozonolysis of bicyclic or other
polycyclic alkenes,31−33 and to the best of our knowledge, only
two examples can be found within the literature involving the
ozonolysis of alkene systems incorporating the 1,2-dioxine
functionality. The natural product (−)-Warburganal (3) was
The other example of interest involved work done some 30
years ago to investigate the ozonolysis of 7-dehydrocholesterol
acetate endoperoxide (4).35,36 Ozonolysis was performed on
the named steroid in a bid to understand the reaction
mechanism and pathways undertaken to the final products,
Scheme 2. The resulting ozonide was not reduced but rather
allowed to decompose over time with various decomposition
products being isolated and characterized along the way. A full
transformation into the final products, 5 and 6, reportedly took
approximately eight weeks at ambient temperature, with
dialdehyde 7 being isolated as an unstable intermediate.
Unfortunately, due to the reported vague reaction conditions
along with the insufficient characterization of the products, the
only, but important, conclusion that can be drawn from their
findings is that ozonolysis of bicyclic 1,2-dioxines appears to
proceed with no disruption of the peroxide linkage and
dicarbonyl products are formed in the first instance.
Given this clear dearth in the literature concerning the
reactivity of bicyclic 1,2-dioxines toward ozone we decided to
explore this reaction on a variety of substrates in order to
investigate both the synthetic scope and probe the mechanism
further. Thus, given our knowledge on the reactivity of 1,2-
dioxines coupled with the two examples above, we propose
herein that reaction of 1,2-dioxines of the type 8 would
consume one mole equivalent of ozone to afford primary
ozonide (molozonide) 9. This would then collapse to ozonide
Received: January 19, 2012
Published: March 16, 2012
© 2012 American Chemical Society
3808
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
Scheme 1
Scheme 2
Scheme 4
10 with reductive workup affording dialdehyde 11, but
importantly keeping the peroxide linkage intact, Scheme 3. A
Scheme 3
a
O2, DCM, hv, rose bengal.
mixture of O2/O3. Reactions were continued until the solution
turned pale blue, indicating saturation with ozone and cessation
of reaction. After completion, the mixture was raised to room
temperature, and reduced with PPh3 (typically 1.1 equiv),
unless otherwise stated. For ease of discussion, the results for
these ozonolyses have been divided into two categories: first
the substrates that behaved as ‘expected’ toward ozone, and
second the substrates that behaved in an “unexpected” manner
toward ozone.
1,2-Dioxines that Behaved as “Expected” Toward
Ozone. Ozonolysis was initially performed on steroid 4 in
order to expand upon the work previously published35,36 and
apply a more rigorous scientific approach toward this reaction
by immediately reducing the ozonide upon formation in order
to validate that the expected dialdehyde was indeed formed,
and to confirm that the peroxide bond was not broken during
the process. We found that ozonolysis of dioxine 4 followed by
immediate PPh3 reduction furnished dialdehyde 7 in excellent
yield (92%), although product instability was an issue, Scheme
5.
It is known that the peroxide bond of 1,2-dioxines can be
readily cleaved upon treatment with an excess of LiAlH4, to
furnish the respective diol,38,39 although in contrast to this,
there are also reports detailing the successful use of LiAlH4 to
reduce carbonyl functionalities in the presence of peroxide
bonds without their rupture.40 Thus, upon using an excess of
LiAlH4 (3 equiv) at 0 °C, dialdehyde 7 was successfully reduced
to triol 16, thus further confirming the presence of the aldehyde
moieties within 7, along with the peroxide bond remaining
intact. Previous NMR and computational studies, along with a
range of simple 1,4-disubstituted bicyclic 1,2-dioxines were
chosen for this study, along with a 1,3-dialkyl example
containing a substituent directly on the C−C double bond,
and a full re-examination of steroid 4 to probe the scope and
possible mechanistic outcomes resulting from the ozonolysis of
bicyclic 1,2-dioxines.
RESULTS AND DISCUSSION
■
Synthesis of Bicyclic 1,2-Dioxines (4, 8a−e and 13). All
1,2-dioxines were synthesized via the rose bengal bis
(triethylammonium) salt sensitized [4π +2π] cycloaddition of
their respective cyclic 1,3-hexadienes with singlet oxygen,
Scheme 4.8,37 The 1,4-disubstituted bicyclic 1,2-dioxines 8a−e,
the 1,3-dialkyl substituted analogue 13 and steroid 4 would
allow for exploration into the effects of differing substitution
patterns, steric and electronic factors on the proposed
ozonolysis reaction highlighted within Scheme 3. The synthesis
of all requisite bicyclic 1,2-dioxines proceeded smoothly with
purified yields ranging from average to excellent, Scheme 4.
Ozonolysis of Bicyclic 1,2-Dioxines (4, 8a−e and 13).
All standard ozonolysis reactions were carried out in dichloro-
methane at −78 °C under an atmosphere of argon. Ozone was
generated from oxygen using corona discharge via an ozone
generator, and bubbled through the solution as a gaseous
3809
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
Scheme 5
a
b
1. O3, DCM, −78 °C, 2. PPh3 (1.1 equiv). LiAlH4 (3 equiv), THF, 0 °C.
Scheme 6
a
b
1. O3, DCM, −78 °C, 2. PPh3 (1.1 equiv). Ph3PCHCO2Et (2 equiv), DCM.
crystal structure, have confirmed that the peroxide bond of 4 is
facing in a fixed downward position.41,42 Upon cleavage of the
alkene bond by ozone, the resulting dialdehyde must therefore
assume the stereochemistry as shown in Scheme 5, and this
relative and absolute stereochemistry must also be that for the
triol 16. The successful formation of 16 clearly confirms that
the ozonolysis of bicyclic 1,2-dioxine 4 proceeds via
molozonide 9 and ozonide 10 as summarized in Scheme 3
and highlights the potential synthetic utility of this process.
The ozonolysis and subsequent PPh3 reduction of the 1,3-
dimethyl susbsituted 1,2-dioxine 13 also proceeded smoothly,
resulting in the desired keto-aldehyde 17 in average yield
(45%), Scheme 5. The product was stable upon purification
the reaction mixture for the ozonolysis of 1,2-dioxine 8a.
Product instability did not allow for full characterization,
although careful column chromatography provided enough
pure sample to characterize dialdehydes 11a and 11b by 1H and
13C NMR. Attempts to purify diester dialdehyde 11d were
unsuccessful, with the product appearing to decompose upon
exposure to silica, although the crude 1H NMR spectrum
showed the presence of two aldehyde peaks, indicating that
ozonolysis had been successful. The crude dialdehyde products
for 11a, 11b and 11d were therefore subjected to a Wittig
reaction using an ethyl ester stabilized phosphorus ylide to form
the more stable alkene products (18a, 18b and 18d) to allow
for the full characterization of the sensitive dialdehydes.
1
with the aldehyde peak at δ = 9.75 ppm in the H NMR
Interestingly, the ylide only added to one aldehyde moiety of
11a and 11b, determined by 2D NMR to be the least hindered
side of each molecule, yielding 18a and 18b. Even though this
left one aldehyde unprotected, both compounds were
sufficiently stable to allow for full characterization, confirming
peroxides (18a and 18b), which in turn confirmed the presence
of the preceding dialdehydes (11a and 11b). The ylide added
to both aldehydes of the diester dialdehyde 11d, giving the
tetra-ester 18d. Standard proton and carbon NMR spectra of
the tetra-ester 18d acquired at ambient temperature showed
broadened peaks, which sharpened upon raising the temper-
ature to +50 °C. Table 1 summarizes the reaction yields over
the three steps from dioxines 8a, 8b and 8d to alkenes 18a, 18b
spectrum showing a small long-range coupling constant of J =
1.8 Hz. The ROESY spectrum showed a correlation between
the aldehyde proton and its peri-methyl protons, which
accounts for this long-range coupling. Again, “attack” of
ozone on the alkene moiety of the precursor bicyclic 1,2-
dioxine 13 dictates that the newly formed ketone and aldehyde
functionalities attain the cis stereochemistry as shown in
Scheme 5.
Ozonolysis and subsequent PPh3 reduction of dioxines 8a,
8b and 8d also furnished the expected dialdehydes 11a, 11b
and 11d in good to average yields with the peroxide bond
remaining intact during the course of reaction, Scheme 6. It is
also interesting to note the formation of diketone 19 (vide
infra), appeared in small quantities (ca. 15%) as a byproduct in
1
and 18d respectively. The H NMR spectrum of 11b showed
splitting of J = 1.5 Hz of one of the aldehyde peaks, with similar
3810
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
sion rate. Increasing the temperature to +50 °C saw a single set
of sharp peaks, allowing 21 to be fully characterized.
Table 1. Yields for Ozonolysis and Wittig Reactions on 1,2-
Dioxines 8a,b and d
1,2-Dioxines that Behaved “Unexpectedly” Toward
Ozone. The ozonolysis of 1,2-dioxines 8c and 8e did not
follow the same pathway as the other dioxines mentioned (i.e.,
to yield the “expected” dialdehydes 11c and 11e), and the
outcomes of these two reactions are individually discussed
below.
a
dioxine
dialdehyde
yield (%)
alkene
yield (%)
b
8a
8b
8d
11a
11b
11d
77
18a
18b
18d
44
10
21
c
33
b
64
a
b
Over three-steps from dioxine to alkene. Estimated, based on the
amount of crude reaction material remaining after column
chromatography in order to remove the PPh3O byproduct. Isolated
Ozonolysis of the diphenyl substituted dioxine 8c resulted in
the formation (vide infra) of keto-aldehyde 22 (26%) as a
crystalline solid (Scheme 8), with single crystal X-ray analysis
unambiguously confirming the structure and stereochemistry.
Along with the keto-aldehyde 22, two other products were
isolated, namely terphenyl (23) (21%) and the diphenyl
diketone 24 (7%). Previously, we saw that the ozonolysis of 8a
resulted in the formation of a similar diketone byproduct, 19a
and their formation will be discussed later. There is no
precedent for the formation of terphenyl (23) under these
circumstances, which raises the question of how and why does
23 form. One possibility is that ozone on interaction with the
alkene moiety of 8c lowers the transition state for cyclo-
reversion aiding the loss of singlet oxygen (to yield diene 12c)
followed by aerial oxidation.
It was considered that dialdehyde 11c may not have formed
due to the presence of the two bulky phenyl groups α to the
alkene moiety inhibiting PPh3 from being able to reduce the
ozonide, thereby allowing it to decompose directly into keto-
aldehyde 22. This theory was tested by reducing the ozonolysis
products with Me2S (1.1 equiv) to see whether the same
products were formed upon using a smaller reducing agent.
Indeed, the same products did form in similar yields and ratios,
suggesting the possibility that the combination of the bulkiness
of the PPh3 and phenyl groups of the 1,2-dioxines is not the
cause this unusual fragmentation.
Purification of the products formed via the ozonolysis and
subsequent reduction of diester dioxine 8e proved difficult,
although the crude 1H NMR spectrum of the products showed
two distinct aldehyde protons as singlets at δ = 9.60 and 9.58
ppm, integrating in a 1:1 ratio. Treatment of this crude mixture
with 2 equiv of ethyl ester ylide, resulted in the formation of
three products, determined to be 25, 26 and 27, totaling 53%
yield over the three steps. These three Wittig products appear
to have come from “unsymmetrical” keto-aldehyde 28 and
“symmetrical” dialdehyde 29 (Scheme 9) in a ratio of
approximately 6.5:1 respectively. Symmetrical dialdehyde 29
and its subsequently trapped tetra-ester 27 appear to result
from direct cleavage of the peroxide bond during the course of
reaction. Structural isomers 25 and 26 occur in an
approximately 1:2 ratio, with 25 being the “expected” product
upon aldehyde protection of the “unsymmetrical” keto-
aldehyde 28 with the ylide, while 26 appears to have undergone
alkene migration. Upon subjecting both 25 and 26 to heat and
acidic conditions in a bid to facilitate isomerization between the
two, no interconversion was observed. Given that five of the
bicyclic 1,2-dioxines (4, 13, 8a, 8b and 8d) behaved as
expected, affording dicarbonyl products upon ozonolysis, while
two dioxines (8c and 8e) deviated from the expected
mechanistic outcome, we felt it appropriate the examine the
unsymmetrical ozonolysis of bicyclic 1,2-dioxines 8a and 8e to
probe these mechanistic outcomes further.
c
pure sample.
1
splitting seen again in the H NMR spectra for the protected
adduct 18b (doublet at δ = 9.71 ppm, J = 2.1 Hz), indicative of
long-range coupling between the aldehyde and methylene
protons.
Although the yields obtained for the formation of alkenes
18a, 18b and 18d appear low to moderate, one must bear in
mind that this is the total yield for a three-step process
(ozonolysis, reduction and Wittig). It is also worth bearing in
mind the complexity of this procedure and the highly reactive
intermediates formed upon reacting 1,2-dioxines with ozone.
The highly oxygenated primary ozonides (molozonides) and
final ozonides are packed with a high density of oxygen atoms
with numerous unfavorable lone-pair repulsions possible. These
sensitive intermediates would have ample opportunity for
cleavage of the weak O−O linkages followed by downstream
decomposition, thereby contributing to a lowered overall yield.
With these possible constraints in hand, we therefore consider
the overall yields obtained for the substrates to be reasonable.
It was decided to also attempt a hydride reduction as an
alternative “trapping” technique for the unstable 1,2-dioxine
aldehyde 11a, as this proved successful for the steroid
dialdehyde 7. Ozonolysis was performed on dioxine 8a to
afford the crude dialdehyde 11a, which was triturated with
hexane to remove most of the PPh3O and subsequently treated
with 3 equiv of LiAlH4. The crude mixture containing diol 20
proved difficult to purify thus acetylation was employed to
furnish diacetate 21 in 32% yield (over four steps from the
1
dioxine), Scheme 7. Broadened peaks were observed in the H
and 13C NMR spectra of the diacetate 21 upon acquisition at
ambient temperature. Upon lowering the temperature to −50
°C all the signals separated out into multiple peaks, likely due
to “freezing” out the conformers by lowering the interconver-
Scheme 7
Unsymmetrical Ozonolysis of Bicyclic 1,2-Dioxines 8a
and 8e. One way to test the mechanistic question of whether
these latter two unusual rearrangements were occurring via
a
b
1. O3, DCM, −78 °C, 2. PPh3 (1.1 equiv). LiAlH4 (3 equiv) 0 °C.
cAc2O, DMAP, pyridine.
3811
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
Scheme 8
Scheme 9
a
b
1. O3, DCM, −78 °C, 2. PPh3 (1.1 equiv). Ph3PCHCO2Et (2 equiv), DCM.
direct decomposition of the ozonide (and therefore not actually
forming the dialdehydes at all) or via another sequence, and
also to address the possibility that the rearrangements may be
occurring due to the nature of the bridgehead substitution was
to investigate these ozonolysis reactions under a different set of
conditions, commonly known as “unsymmetrical ozonolysis”.
Consequently we decided to subject one dioxine that
underwent rearrangement upon standard ozonolysis, 8e, and
one dioxine that gave the expected results, 8a, to these
alternative conditions.
Unsymmetrical ozonolysis was first reported by Schreiber et
al. in 1982,43 and as its name suggests results in two different
functional groups being produced from the oxidative cleavage
of an alkene, in this case, an aldehyde and an ester. The primary
ozonide (molozonide) undergoes cleavage to produce the
carbonyl/carbonyl oxide intermediate 30. The alcohol then
traps the carbonyl oxide to produce a hydroperoxy acetal, which
is subsequently dehydrated to form an ester.43,44 A key aspect
of this reaction worth noting is that no ozonide is formed in
contrast to “standard” ozonolysis.
To the best of our knowledge, the unsymmetrical ozonolysis
protocol has not been previously reported on a 1,2-dioxine
(either monocyclic or bicylic), however one would expect the
product to be a dioxine incorporating aldehyde-ester
functionalities, 32 (Scheme 10). If the two substituents (R
and R1) are different, then two regioisomers will be possible.
Treatment of 8a to the unsymmetrical ozonolysis conditions,
8a yielded the “expected” ester-aldehydes 33a and 33b in poor
yield, with little selectivity seen between the two regioisomers,
Scheme 11. The two isomers were easily separable via flash
chromatography, and HMBC correlations were used to aid in
the identification of the two isomers. In addition, 33a was
crystalline allowing for single crystal X-ray analysis, unambig-
uously confirming its structure and stereochemistry. Diketone
19, which was seen as a byproduct in the standard ozonolysis of
8a was also observed under these unsymmetrical conditions, in
3812
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
standard ozonolysis of 8c. The respective 1,4-dicarbonyls were
not observed in the ozonolysis of the other 1,2-dioxines (8c, d
and e, 4 or 13) employed within this study. 1,4-Dicarbonyls of
this nature have previously been observed as byproduct of 1,2-
dioxine chemistry. A recent example reported by our group
found 19 and 24 were formed as the result of thermal
decomposition of peroxide-diols involving a radical-induced
homolytic cleavage of the peroxide followed by subsequent
double β-scission.14 It is unlikely that the formation of 19 and
24 in this instance resulted from this mechanistic process, due
to the absence of either heat or light, thereby limiting the
likelihood of a radical process. The ozonolysis of endoperoxide
37 in the presence of tetracyanoethylene was previously
reported to furnish ketone 38, which was unstable and readily
decarbonylated at −10 °C to give succinaldehyde (39), Scheme
13.45 Interestingly, formation of products 19 and 24, 22, 28 and
Scheme 10
Scheme 11
Scheme 13
a
b
O3, DCM/MeOH (5:1), −78 °C. NEt3, Ac2O.
21% yield. 1,2-Dioxane 33b was an oil which showed early signs
of decomposition thereby making purification difficult. The
reason for this may be due to the iso-propyl flanking the
aldehyde in 33a, thereby creating a shielding effect aiding in
protection against decomposition, whereas 33b’s aldehyde
moiety is adjacent to the smaller methyl group.
36 within this study also require the loss of CO or CO2 at some
stage.
Reaction conditions common to both the symmetrical and
unsymmetrical ozonolyses in our work was the use of weak
base, namely triphenylphosphine or triethylamine. Although it
is difficult to postulate how byproducts 19 and 24 were formed
in this instance, there are several potential mechanistic
pathways possible, including breakdown of either dioxine-
aldehyde 11a, 11c or 33, or alternatively and probably more
likely, via decomposition of the molozonide intermediates 9,
Scheme 14. Their formation is beyond the scope of this study,
Only one major product, furan 34 was isolated upon
1
subjecting 8e to the unsymmetrical ozonolysis conditions. H
and 13C NMR data showed that three ester groupings were
present, but it was immediately evident through the lack of an
aldehyde peak in the NMR spectra that the theoretically
expected triester aldehyde 35 had not formed. It is likely that
34 exists in equilibrium with ketol 36, resulting from a ring-
opening rearrangement similar to that seen for the standard
ozonolysis of 8e, Scheme 12.
Scheme 14
Formation of Diketone Products 19 and 24. The
formation of two different diketone products were observed
within this body of work. 19 was seen in both the standard and
unsymmetrical ozonolysis of 8a, being formed in 15 and 21%
yield, respectively, while 24 was formed in 7% yield during the
Scheme 12
however, is of mechanistic interest and will be the subject of
more detailed investigations in the future.
Overall Mechanistic Scope and Insights for the
Ozonolysis of Bicyclic 1,2-Dioxines. Overall we have seen
steroid 4 reacted as “expected” toward ozone to furnish
dialdehyde 7 in excellent (92%) yield, along with the dimethyl
1,2-dioxine 13 yielding the expected dicarbonyl 17 indicating
that substituents on the double bond appear to be tolerated
during these ozonolysis reactions. Of the bridgehead 1,2-
dioxines investigated within this study, 8a, 8b and 8d all
reacteing as “expected” toward ozone to afford their respective
dialdehydes (11a, 11b and 11d) in poor to good yields,
a
b
O3, DCM/MeOH (5:1), −78 °C. NEt3, Ac2O.
3813
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
whereas ozonolysis of 8c and 8e resulted in a range of
fragmentation products (22 and 25, 26, 27, respectively)
showing a loss of either CO or CO2 (depending on the
proposed mechanism). Finally, unsymmetrical ozonolysis was
utilized to further probe the reaction mechanism, with 8a
furnishing the expected aldehyde esters (33a,b), while 8e once
again resulted in formation of fragmentation product, 34. A
general summary of all outcomes is depicted in Scheme 15.
Several of the reactions afforded products that showed the
loss of CO or CO2 (depending on the proposed mechanism),
along with cleavage of the peroxide bond, with a similar
mechanism appearing to be at play under both symmetrical and
unsymmetrical ozonolysis conditions. In attempting to explain
how these rearrangements occur, several questions are
paramount. Do these rearrangements proceed via ozonide 10
decomposing into the expected dialdehydes 11 followed by
facile rearrangement into 22, 28 or 36, Scheme 16, Pathway A,
or, alternatively does ozonide 10 directly decompose into 22,
28 or 36, Scheme 16, Pathway B? Moreover, does the addition
of PPh3 assist in these rearrangements?
Scheme 15
Since the resultant ozonide from the phenyl, phenyl
disubstituted 1,2-dioxine 8c rearranged upon treatment with
both PPh3 and Me2S, it is unlikely that the bulky PPh3 was
solely responsible for the rearrangement seen under the
standard ozonolysis conditions. As the unsymmetrical ozonol-
ysis mechanism does not involve the formation of an ozonide,
and since the same rearrangement was observed under
unsymmetrical conditions, it is likely that the rearranged
products do not arise directly from ozonide 10 under standard
conditions, thereby ruling out Pathway B in Scheme 16.
Pathway A is further supported by the two aldehyde peaks, with
1
1:1 integrations, that were seen in the crude H NMR spectra
from the ozonolysis of diester dioxine 8e.
If dialdehyde 11, or aldehyde esters 32, 33 and 35 for
unsymmetrical ozonolysis, are indeed forming (cannot be ruled
out due to the low product yields) and spontaneously
decomposing to furnish the rearranged products, there are
several possibilities via how this could occur. Scheme 16
outlines a couple of potential pathways involving both
concerted and stepwise mechanisms for Pathway A, both
involving decarbonylation (−CO) of the aldehyde moiety.
Decarbonylation of aldehydes is usually seen in the presence of
transition metal catalysts,49−51 or radical initiators.52,53 There is
no precedent for the loss of a carbonyl group α to a peroxide
bond, making this type of rearrangement unique. At this stage,
either a concerted or stepwise pathway is plausible.
Another mechanistic possibility is that either the aldehyde-
carbonyl-oxide species 30 or the molozonide 9 may directly
rearrange into products 22, 28 or 36, and may be induced by
the electron withdrawing nature of the substituents (vide supra)
Scheme 16 Pathways C and D respectively. These inter-
mediates are common to both the symmetrical and unsym-
metrical ozonolysis reactions, and involve a total loss of CO2. It
appears reasonable that decarboxylation via loss of the carbonyl
oxide group from 30 is plausible, Scheme 16. Breakdown of the
molozonide 9 into 22, 28 or 36 could also occur via a concerted
mechanism (Scheme 16, Pathway D), or through a series of
stepwise reactions involving homolysis of the molozonide, 1,5-
hydrogen atom abstraction with ring-opening, followed by loss
of CO2 to furnish compounds of type 22 or 36.54 While the
formation of fragmentation products of type 22 was not the
primary objective of this study, it is clear that there are several
underlying rearrangements that may occur when 1,2-dioxines
are treated with ozone, with Scheme 16 highlighting the most
likely.
It is apparent that the nature of the substituents plays a part
in influencing the reaction outcome, as the alkyl substituted 1,2-
dioxine 8a did not rearrange under either standard or
unsymmetrical ozonolysis, whereas the diester dioxine 8e
rearranged in both cases. It seems reasonable from our initial
studies to postulate that electronic factors play an important
part, with electron donating groups appearing to produce the
expected dialdehydes, whereas upon removing electron density
from the peroxide via electron withdrawing groups, rearrange-
ments become dominant. It also appears plausible at this stage
to propose that one bridgehead phenyl grouping (dioxine 8b)
has little effect on altering the reaction outcome, with normal
ozonolysis proceeding, whereas two bridgehead phenyl
moieties (dioxine 8c) remove enough electron density to
weaken the peroxide bonds and enable rearrangements to
readily occur. Previous research has shown that reactions of
alkenes with ozone are sensitive to electronic effects, with
electron deficient alkenes having much lower rate constants
than those alkenes with greater electron density.46−48 This is
believed to be due to the electrophilic nature of ozone, which
adds to nucleophilic π-bonds of alkenes that are enriched by
lone pair donation by electron donating groups, thereby
assisting in the lowering of the activation barrier, and thus
enhancing the reaction rate.46,48 When the rate is increased,
there is less chance for rearrangements to occur, making it
reasonable to suggest that electron withdrawing groups
attached to the bridgehead positions of these bicyclic 1,2-
dioxines may assist in slowing the rate of reaction and thereby
raising the likelihood of rearrangements.
CONCLUSIONS
■
The ozonolysis of bicyclic 1,2-dioxines is an unusual and unique
reaction and this study has clearly exemplified that the overall
process of oxidative cleavage of the olefinic unit within 1,2-
dioxines by ozone to furnish 1,4-dicarbonyl cyclic peroxides of
3814
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
Scheme 16
silica gel 60 F254 from Merck, and visualized under 254 nm light or
developed in either vanillin or potassium permanganate dip. THF was
distilled over sodium wire with benzophenone as indicator and freshly
distilled prior to use. Ozone was generated from oxygen via corona
discharge and all solutions purged with argon. Flow rates and ozone
concentration levels were not recorded, as reactions proceeded until
solutions were saturated with ozone. Compounds 4,55 8a,56 8b,14 8c,57
8d14 19,58 2359 and 2460 showed physical and chemical properties
identical to those previously reported in the literature.
General Procedure for the Preparation of 1,2-Dioxines (4,
8a−e and 13). To a solution of the requisite 1,3-cyclohexadiene (14,
12a-e or 15) in dichloromethane (30 mL/g of 1,3-diene), in a custom-
made pyrex flask fitted with a cooling jacket, was added rose bengal,
bis(triethylammonium) salt (100 mg). Ice water was pumped
throughout the cooling jacket to maintain a temperature of ca 5−10
°C within the reaction mixture at all times. Oxygen was bubbled
through the solution, and the contents irradiated with 3 × 500 W
tungsten halogen lamps until complete via TLC (1−8 h). The mixture
was then concentrated in vacuo and the residue purified by flash
chromatography. Any unreacted diene was also recovered at this time.
( )-Dimethyl (1R,4S)-2,3-dioxabicyclo[2.2.2]oct-5-ene-1,4-
dicarboxylate (8e). Colorless needles. Yield: 46%. Mp 66−68 °C.
Rf 0.37 (3:7 ethyl acetate/hexane) 1H NMR (300 MHz, CDCl3): 6.88
(s, 2H), 3.86 (s, 6H), 2.56−2.49 (m, 2H), 1.83−1.77 (m, 2H); 13C
NMR (75 MHz, CDCl3): 168.6, 132.5, 77.9, 53.1, 26.3; IR (nujol)
type 11 is of general synthetic utility. Five of the seven bicyclic
1,2-dioxines furnished the desired 1,4-dialdehydes in moderate
to excellent yields with a range of substitution and electronic
patterns being explored. This study also elucidated for the first
time several new background fragmentations and/or rearrange-
ments that can occur upon treatment of 1,2-dioxines with
ozone. While a full study into the genesis of these outcomes
was not the primary purpose of this study, initial mechanistic
studies have highlighted several plausible breakdown pathways
as detailed within Scheme 16. Finally, we have also reported for
the first time on the outcomes of several unsymmetrical
ozonolyses of bicyclic 1,2-dioxines, which behaved in a similar
manner to those of standard ozonolysis and as such further
increases the synthetic scope and utility of these unique
reactions.
EXPERIMENTAL SECTION
■
General Experimental Methods. 1H and 13C NMR spectra were
recorded in CDCl3 on either a 300 or 600 MHz instrument. TMS
(0.00 ppm) and CDCl3 (77.00 ppm) were used as internal standards.
Melting Points are uncorrected. Flash chromatography was performed
using Merck silica gel 60 (230−400 mesh ASTM). Thin-layer
chromatography (TLC) was performed using aluminum sheets of
3815
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
2927, 1745, 1306, 1119, 989, 702 cm−1; HRMS calcd for (M)+
C10H12O6: 228.0634; found 228.0630.
A small amount was able to be isolated for NMR analysis before
decomposition. H NMR (300 MHz, CDCl3): 9.69 (s, 1H), 9.42 (s,
1
(
)-(1R,4S)-1,5-Dimethyl-2,3-dioxabicyclo[2.2.2]oct-5-ene
1H), 2.12−2.03 (m, 2H), 1.85−1.78 (m, 2H), 1.59 (m, 1H), 1.44 (s,
3H), 0.98 (d, 3H, J = 7.2 Hz), 0.97 (d, 3H, J = 7.2 Hz); 13C NMR (75
MHz, CDCl3): 202.7, 199.1, 89.1, 84.4, 31.8, 24.4, 19.5, 17.3, 16.3,
15.9.
1
(13). Colorless oil. Yield 76%. Rf 0.27 (1:9 ethyl acetate/hexane). H
NMR (300 MHz, CDCl3): 6.03 (s, 1H), 4.42 (m, 1H), 2.29 (m, 1H),
2.02 (m, 1H), 1.95 (s, 3H), 1.56−1.40 (m, 2H), 1.35 (s, 3H); 13C
NMR (75 MHz, CDCl3): 141.5, 128.9, 75.4, 75.2, 29.1, 23.1, 21.8,
18.5; IR (neat) 2932, 2360, 1660, 1444, 1374, 1224, 1158, 886, 764
cm−1; LRP (+LSIMS) m/z (%) 190 (4), 173 (2), 158 (36), 155 (13),
141 (95), 125 (M+, 13), 123 (100), 113 (9); HRMS calcd. for (M +
H)+ C8H13O2: 141.0916; found 141.0910.
( )-(3R,6R)-3-Phenyl-1,2-dioxane-3,6-dicarbaldehyde (11b).
Colorless oil. Decomposes readily. Typically the crude sample was
used immediately for Wittig protection and subsequent full character-
ization. Yield 33%. Rf 0.38 (1:1 ethyl acetate/hexane). A small amount
1
was able to be isolated for NMR analysis before decomposition. H
General Procedure for Ozonolysis of 1,2-Dioxines. A solution
of 1,2-dioxine (4, 8a−e or 13) (3 mmol) in dichloromethane (50 mL)
was cooled to −78 °C under a continuous atmosphere of argon. A
stream of ozone was bubbled through the mixture until the solution
turned pale blue. The mixture was bought back to ambient
temperature, followed by the addition of triphenylphosphine (3.3
mmol). The resulting mixture was stirred at room temperature
overnight and then concentrated in vacuo and the products were
purified by flash chromatography.
(1R,3aR,3bS,5aR,7S,9aR,9bR,11aR)-3b,5a-diformyl-9a,11a-
Dimethyl-1-(6-methylheptan-2-yl)tetradecahydro-1H-benzo-
[c]indeno[5,4-e][1,2]dioxin-7-yl Acetate (7). Colorless oil. Yield
92%. Rf 0.60 (3:7 ethyl acetate/hexane). 1H NMR (300 MHz,
CDCl3): 10.01 (s, 1H), 9.62 (s, 1H), 5.28 (m, 1H), 2.83 (dd, 2H, J =
15.6, 3.6 Hz), 2.13 (m, 1H), 2.02 (s, 3H), 1.92−1.11 (m, 22H), 1.02
(s, 3H), 0.89 (d, 3H, J = 6.3 Hz), 0.87 (d, 3H, J = 1.2 Hz), 0.85 (d, 3H,
J = 1.2 Hz), 0.74 (s, 3H); 13C NMR (75 MHz, CDCl3): 202.5, 196.8,
169.8, 88.5, 87.7, 69.2, 56.5, 55.8, 45.8, 45.5, 40.5, 39.4, 35.7, 35.4,
35.2, 31.0, 28.8, 28.01, 27.96, 24.0, 23.6, 22.7, 22.5, 21.2, 19.8, 19.3,
18.3, 18.1, 13.2. Dialdehyde 7 decomposed readily over time thus was
further characterized as triol 16.
NMR (300 MHz, CDCl3): 9.69 (d, 1H, J = 1.5 Hz), 9.61 (s, 1H),
7.52−7.30 (m, 5H), 4.69 (dd, 1H, J = 11.1, 3.0 Hz), 2.84 (m, 1H),
2.18−1.79 (m, 3H); 13C NMR (75 MHz, CDCl3): 197.7, 197.0, 129.2,
129.1, 129.0, 128.6, 125.4, 125.2, 89.0, 84.7, 26.6, 21.3.
(
)-Dimethyl-2,2′-[(3R,6S)-3,6-diformyl-1,2-dioxane-3,6-
diyl]diacetate (11d). Purification and characterization proved
difficult, as the sample decomposed upon attempted column
chromatography. Typically the crude sample was used immediately
for Wittig protection and subsequent full characterization. Rf 0.61 (7:3
ethyl acetate/hexane).
General Procedure for Wittig Reaction of Ozonolysis
Products. To a solution of dialdehyde (11a,b, or d) (3 mmol) in
dichloromethane (50 mL) under an atmosphere of nitrogen was added
ethyl 2-(triphenylphosphanylidene) acetate (6 mmol). The mixture
1
was stirred at ambient temperature until complete by H NMR or
TLC. The mixture was then concentrated in vacuo and the products
purified by flash chromatography.
( )-Ethyl-(2E)-3-[(3R,6R)-6-formyl-3-methyl-6-(propan-2-yl)-
1,2-dioxan-3-yl]prop-2-enoate (18a). Colorless oil. Yield 44%
(over three steps: ozonolysis, reduction and Wittig). Rf 0.55 (1:4 ethyl
1
acetate/hexane). H NMR (600 MHz, CDCl3): 9.74 (br s, 1H), 6.72
( 1 R , 3 a R , 3 b S , 5 a R , 7 S , 9 a R , 9 b R , 1 1 a R ) - 3 b , 5 a - B i s -
(hydroxymethyl)-9a,11a-dimethyl-1-(6-methylheptan-2-yl)-
tetradecahydro-1H-benzo[c]indeno[5,4-e][1,2]dioxin-7-ol (16).
To a solution of dialdehyde (7) (648 mg, 1.32 mmol) in anhydrous
THF (5 mL) at 0 °C under an atmosphere of nitrogen was added
LiAlH4 (215 mg, 5.67 mmol) in portions, with stirring. The mixture
was kept at 0 °C until TLC showed the disappearance of the starting
material (60 min). The mixture was quenched with saturated NH4Cl
(5 mL) and diluted with diethyl ether (10 mL). The mixture was then
acidified to pH 1 with the addition of HCl (2 M, few drops). The
layers were separated and the aqueous layer extracted with diethyl
ether (3 × 25 mL). The combined organic layers were washed with
saturated aqueous NaHCO3 (10 mL), brine (10 mL), dried (MgSO4)
and concentrated in vacuo. The crude mixture was then purified by
flash chromatography to furnish a white solid (106 mg, 55%) (16). Mp
(d, 1H, J = 16.2 Hz), 5.84 (d, 1H, J = 16.2 Hz), 4.19 (q, 2H, J = 7.2
Hz), 2.09 (ddd, 1H, J = 13.8, 5.4, 2.4 Hz), 1.93 (sept 1H, J = 7.2 Hz),
1.81 (m, 1H), 1.69 (apt dt, 1H, J = 13.2, 4.8 Hz), 1.63 (ddd, 1H, J =
13.8, 5.4, 3.0 Hz), 1.52 (s, 3H), 1.28 (t, 3H, J = 7.2 Hz), 0.97 (d, 3H, J
= 7.2 Hz), 0.93 (d, 3H, J = 7.2 Hz); 13C NMR (150 MHz, CDCl3):
204.0, 166.1, 148.4, 120.3, 89.1, 79.0, 60.6, 32.3, 29.4, 20.8, 19.8, 16.3,
15.7, 14.1; IR (neat) 2976, 1723, 1655, 1369, 1309, 1184, 1036, 749
cm−1; LRP (+LSIMS) m/z (%) 271 (M+, 8), 258 (10), 242 (12), 225
(100), 195 (12), 179 (50); HRMS calcd. for (M + H)+ C14H23O5:
271.1545; found 271.1535.
( )-Ethyl-(2E)-3-[(3R,6R)-6-formyl-6-phenyl-1,2-dioxan-3-yl]-
prop-2-enoate (18b). Colorless oil. Yield 10% (over three steps:
ozonolysis, reduction and Wittig). Rf 0.61 (dichloromethane). 1H
NMR (300 MHz, CDCl3): 9.71 (d, 1H, J = 2.1 Hz), 7.43−7.30 (m,
5H), 6.74 (dd, 1H, J = 16.2, 5.1 Hz,), 5.98 (dd, 1H, J = 16.2, 1.6 Hz,),
4.88 (m, 1H), 4.21 (q, 2H, J = 7.2), 2.87 (m, 1H), 2.06−1.74 (m, 3H),
1.30 (t, 3H, J = 7.2 Hz); 13C NMR (75 MHz, CDCl3): 198.8, 165.7,
142.0, 134.9, 129.0, 125.1, 123.2, 88.6, 79.7, 60.8, 27.8, 26.5, 14.2 (3
aromatic C masked); IR (neat) 2987, 1724, 1658, 1452, 1309, 1275,
1193, 1036, 700 cm−1; FTMS (+ESI) m/z (%): 291 (M+, 4) 279 (5),
277 (12), 262 (16), 261 (100), 259 (2); HRMS calcd. for (M + H)+
C16H19O5: 291.1232; found 291.1229.
20
174−175 °C. Rf 0.27 (6:4 ethyl acetate/hexane). [α]D = +111.76 (c
0.34, THF). 1H NMR (300 MHz, CDCl3): 4.61 (d, 1H, J = 12.3 Hz),
4.33 (s, 1H), 3.90 (d, 2H, J = 12.9 Hz), 3.48 (d, 1H, J = 12.3 Hz), 2.67
(dd, 1H, J = 15, 3 Hz), 2.41−1.06 (m, 27H), 1.01 (s, 3H), 0.93 (d, 3H,
J = 10.8 Hz), 0.90 (s, 3H), 0.87 (d, 3H, J = 1.2 Hz), 0.85 (d, 3H, J =
1.2 Hz); 13C NMR (75 MHz, CDCl3): 85.8, 85.2, 67.3, 64.6, 61.7,
57.8, 56.8, 45.5, 45.3, 41.6, 39.4, 35.8, 35.38, 35.35, 31.2, 29.0, 28.2,
28.0, 27.5, 23.7, 22.8, 22.5, 21.2, 18.9, 18.8, 18.4, 13.9; IR (nujol) 3313,
2927, 1464, 1379, 1059, 1034 cm−1; HRMS calcd. for (M + NH4)+
C27H51NO5: 470.3840; found 470.3835.
(
)-Diethyl-(2E,2′E)-3,3′-[(3R,6S)-3,6-bis(2-methoxy-2-ox-
oethyl)-1,2-dioxane-3,6-diyl]bisprop-2-enoate (18d). Colorless
oil. Yield 21% (over three steps: ozonolysis, reduction and Wittig). Rf
0.51 (1:1 ethyl acetate/hexane). 1H NMR (50 °C, 300 MHz, CDCl3):
6.93 (d, 2H, J = 16.2 Hz), 6.00 (d, 2H, J = 16.2 Hz), 4.20 (q, 4H, J =
7.2 Hz), 3.67 (s, 6H), 2.92−2.63 (m, 4H), 2.23−2.15 (m, 2H), 1.95−
1.88 (m, 2H), 1.29 (t, 6H, J = 7.2 Hz); 13C NMR (50 °C, 75 MHz,
CDCl3): 169.1, 165.8, 146.4, 122.2, 80.3, 60.6, 51.7, 41.1, 27.1, 14.1;
IR (neat) 2955, 1720, 1656, 1438, 1311, 1179, 1034, 865, 731 cm−1;
LRP (+LSIMS) m/z (%): 857 (10), 429 (M+, 100), 412 (19), 337 (8),
235 (50); HRMS calcd. for (M+ + H)+ C20H29O10: 429.1761; found
429.1752.
( )-(3R,6S)-6-Acetyl-3-methyl-1,2-dioxane-3-carbaldehyde
(17). Pale-yellow oil. Yield 45%. Rf 0.46 (3:7 ethyl acetate/hexane). 1H
NMR (300 MHz, CDCl3): 9.75 (d, 1H, J = 1.8 Hz), 4.55 (dd, 1H, J =
11.7, 3.0 Hz), 2.39 (m, 1H), 2.14 (s, 3H), 1.95 (m, 1H), 1.75−1.51
(m, 2H), 1.19 (s, 3H); 13C NMR (75 MHz, CDCl3): 204.0 202.2,
85.8, 85.3, 27.9, 26.4, 22.8, 19.8; IR (neat) 2934, 2360, 1724, 1446,
1363, 1115, 935, 749 cm−1; LRP (+LSIMS) m/z (%) 295 (20), 279
(33), 255 (9), 215 (12), 211 (97), 195 (100), 181 (22), 173 (M+, 2),
139 (36), 123 (19) HRMS calcd. for (M + H)+ C8H13O4: 173.0814;
found 173.0808.
( )-[(3R,6R)-3-Methyl-6-(propan-2-yl)-1,2-dioxane-3,6-diyl]-
dimethanediyl Diacetate (21). Ozonolysis was performed on
ascaridole (8a) (260 mg, 1.55 mmol) and reduced with triphenyl-
phosphine (407 mg, 1.55 mmol) via the general method to afford a
( )-(3R,6R)-3-Methyl-6-(propan-2-yl)-1,2-dioxane-3,6-dicar-
baldehyde (11a). Colorless oil. Decomposes readily. Typically the
crude sample was used immediately for Wittig protection and
subsequent full characterization. Rf 0.52 (3:7 ethyl acetate/hexane).
3816
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
crude oily solid. This was triturated with hexane to give a crude yellow
oil (141 mg). To a stirred solution of this crude dialdehyde (141 mg,
0.70 mmol (assuming 100% product)) in anhydrous THF (5 mL) at 0
°C under an atmosphere of nitrogen was added LiAlH4 (80 mg, 2.11
mmol) in portions, with stirring. The resulting mixture was stirred at 0
°C until TLC showed the disappearance of the starting material (30
min). The solution was quenched with saturated NH4Cl (5 mL) and
diluted with diethyl ether (10 mL). The mixture was acidified to pH 1
with the addition of conc. HCl (2 M, 2 drops). The layers were
separated and the aqueous layer extracted with diethyl ether (3 × 10
mL). The combined organic layers were washed with saturated
aqueous NaHCO3 (10 mL), water (10 mL), dried (MgSO4) and
concentrated in vacuo. To this crude mixture of diol 20 was added
pyridine (1 mL), acetic anhydride (0.64 g, 6.34 mmol) and DMAP (17
mg, 0.14 mmol), and the resulting mixture stirred overnight. The
mixture was diluted with dichloromethane (5 mL) and washed with
10% HCl (5 mL). The organic layer was then removed and the
aqueous layer extracted with dichloromethane (3 × 5 mL). The
organic layers were combined, washed with water (5 mL) and brine (5
mL), dried (MgSO4) and concentrated in vacuo. The residue was
purified via flash chromatography to afford a colorless oil (65 mg,
(1:1 ethyl acetate/hexane) 1H NMR (600 MHz, CDCl3): 6.94 (d, 1H,
J = 15.6 Hz), 6.25 (d, 1H, J = 15.6 Hz), 5.83 (t, 1H, J = 1.5 Hz), 4.21
(q, 2H, J = 7.2 Hz), 4.18 (q, 2H, J = 7.2 Hz), 3.84 (s, 3H), 3.82 (s,
3H) 3.51 (s, 1OH), 2.46 (m, 1H), 2.31 (m, 1H), 2.05 (m, 1H), 1.97
(m, 1H), 1.30 (t, 3H, J = 7.2 Hz), 1.27 (t, 3H, J = 7.2 Hz); 13C NMR
(150 MHz, CDCl3): 173.8, 168.7, 165.9, 164.7, 148.0, 146.1, 122.5,
120.9, 76.4, 60.8, 60.7, 53.8, 52.4, 35.9, 28.2, 14.2, 14.1; IR (neat)
3484, 2957, 1717, 1437, 1259, 1173, 1030, 983, 735 cm−1; LRP
(+LSIMS) m/z (%) 373 (M+, 100), 341 (17), 327 (32), 309 (5), 295
(44), 281 (26), 263 (12), 249 (18), 235 (12), 221 (13), 207 (14);
HRMS calcd. for (M + H)+ C17H25O9: 373.1499; found 373.1502.
( )-1,6-Dimethyl-3,4-dideoxy-2,5-bis-C-[(1E)-3-ethoxy-3-ox-
oprop-1-en-1-yl]-D-erythro-hexarate (27). Colorless oil. Yield: 7%
(over three steps: ozonolysis, reduction and Wittig). Rf 0.48 (1:1 ethyl
1
acetate/hexane). H NMR (600 MHz, CDCl3): 6.93 (d, 2H, J = 15.3
Hz), 6.23 (d, 2H, J = 15.3 Hz), 4.21 (q, 4H, J = 7.2 Hz), 3.83 (s, 6H),
3.52 (s, 2H), 1.95 (dd, 2H, J = 13.8, 4.8 Hz), 1.72 (dd, 2H, J = 13.8,
4.8 Hz), 1.30 (t, 6H, J = 7.2 Hz); 13C NMR (150 MHz, CDCl3):
174.0, 166.0, 146.4, 122.3, 76.5, 60.7, 53.7, 32.4, 14.2; IR (neat) 3458,
2958, 1710, 1437, 1231, 1105, 996 cm−1; LRP (+LSIMS) m/z (%) 403
(M+, 100), 369 (18), 327 (32), 311 (21), 281 (28), 251 (15), 235
(20), 219 (27), 191 (23); HRMS calcd. for (M + H)+ C18H27O10:
403.1604; found 403.1602.
1
32%) (21). Rf 0.57 (3:7 ethyl acetate/hexane). H NMR (50 °C, 300
MHz, CDCl3): 4.28 (d, 1H, J = 12.3 Hz), 4.16 (d, 1H, J = 12.3 Hz),
4.10 (br s, 2H), 2.23 (m, 1H), 2.08 (s, 3H), 2.07 (s, 3H), 1.82−1.54
(m, 4H), 1.27 (s, 3H), 0.97 (d, 6H, J = 6.9 Hz); 13C NMR (50 °C, 75
MHz, CDCl3): 170.6 (×2 overlapping), 82.4, 78.8, 67.3, 63.1, 31.9,
26.5, 22.0, 20.8, 20.7, 19.9, 16.7, 16.6; IR (neat): 2974, 2364, 1747,
1458, 1377, 1246, 1049 cm−1; LRP (+LSIMS) m/z (%): 530 (67), 475
(9), 338 (26), 289 (M+, 35), 243 (24) (219 (100). HRMS calcd. for
(M + H)+ C14H25O6: 289.1651; found 289.1639.
Attempted Isomerization of 25 and 26.
Method A: Heat. An NMR tube containing 25 (10 mg) or 26 (27
mg) in CDCl3 was heated to 60 °C, and the temperature was
maintained for 7 h. No change was detected for either sample via TLC
1
or H NMR.
Method B: Heat and Acidic Conditions. An NMR tube containing
25 (10 mg) or 26 (27 mg) in CDCl3 with a catalytic amount (10 mol
%) of PTSA was heated to 60 °C, and the temperature was maintained
( )-(2S)-2-Hydroxy-5-oxo-2,5-diphenylpentanal (22).
Method A: Reduction with PPh3. Followed general procedure
outline for the ozonolysis of 1,2-dioxines.
1
for 6 h. No change was detected for either sample via TLC or H
NMR.
Method B: Reduction with Me2S. A solution of 1,2-dioxine (8c) (3
mmol) in dichloromethane (50 mL) was cooled to −78 °C under a
continuous atmosphere of argon. A stream of ozone was bubbled
through the mixture until the solution turned pale blue. The mixture
was bought back to room temperature and placed under an
atmosphere of nitrogen. Dimethyl disulfide (3.3 mmol) was slowly
added, and the resulting mixture was stirred at room temperature
overnight. The mixture was then concentrated in vacuo and the
products purified by flash chromatography.
General Procedure for Unsymmetrical Ozonolysis of 1,2-
Dioxines. An ozone stream was bubbled through a stirred suspension
of 1,2-dioxine 8a or 8e (1 mmol) and NaHCO3 (0.5 mmol) in a 5: 1
mixture of CH2Cl2:MeOH (10 mL) at −78 °C under an atmosphere
of argon, until the solution turned pale blue. The solution was bought
back to room temperature. The NaHCO3 was filtered off and the
mother liqueur concentrated in vacuo. The residue was taken up in
CH2Cl2 (10 mL) and cooled to 0 °C. Triethylamine (1.5 mmol) and
acetic anhydride (3 mmol) were then added, and the solution brought
to room temperature and stirred for 1 h. This mixture was treated with
methanol (1 mL), stirred for 15 min and then diluted with diethyl
ether (5 mL) and washed with 5% H2SO4 (3 × 10 mL), saturated
NaHCO3 (10 mL), brine (10 mL), dried (MgSO4) and concentrated
in vacuo. The crude mixture was purified by flash chromatography.
( )-Methyl-[(3R,6R)-6-formyl-3-methyl-6-(propan-2-yl)-1,2-
dioxan-3-yl]acetate (33a). Colorless crystals. Yield 13%. Rf 0.70
(3:7 ethyl acetate/hexane). Mp 64−66 °C. 1H NMR (600 MHz,
CDCl3): 9.71 (br d, 1H, J = 1.5 Hz), 3.71 (s, 3H), 2.10−1.74 (m, 5H),
1.58 (s, 3H), 0.94 (d, 3H, J = 6.9 Hz), 0.92 (d, 3H, J = 6.9 Hz); 13C
NMR (150 MHz, CDCl3): 203.5, 171.7, 89.2, 81.2, 52.5, 32.1, 27.3,
19.9, 19.8, 16.3, 15.8; IR (neat) 2971, 1738, 1456, 1374, 1288, 1248,
1123, 748 cm−1; LRP (+LSIMS) m/z (%): 483 (22), 304 (30), 269
(16), 248 (100), 231 (5), 186 (23); HRMS calcd. for (M + H)+
C11H19O5: 231.1232; found 231.1227.
Colorless needles. Yield 26%. Mp 82−84 °C. Rf 0.29 (1:4 ethyl
1
acetate/hexane) H NMR (300 MHz, CDCl3): 9.64 (d, 1H, J = 1.2
Hz), 7.93−7.89 (m, 2H), 7.58−7.30 (m, 8H), 4.34 (d, OH, J = 1.2
Hz), 3.14 (dt, 1H, J = 18.2, 7.5 Hz), 3.02 (dt, 1H, J = 18.2, 6.6 Hz),
2.52 (dd, 2H, J = 7.5, 6.6 Hz); 13C NMR (75 MHz, CDCl3): 200.1,
199.7, 138.0, 136.4, 133.4, 128.9, 128.6, 128.1, 128.1, 125.8, 81.0, 32.2,
30.7; IR (nujol) 3448, 2927, 2859, 1722, 1676, 741 cm−1; LRP
(+LSIMS) m/z (%) 269 (M+, 1), 239 (100), 221 (24), 193 (9), 178
(7), 161 (43), 133 (18), 115 (23), 105 (87), 77 (48); HRMS calcd. for
(M)+ C17H16O3: 268.1099; found 268.1105.
(
)-1,7-Diethyl-3-methyl (1E,3S)-3-hydroxy-6-propanoyl-
hepta-1,6-diene-1,3,7-tricarboxylate (25). Colorless oil. Yield
15% (over three steps: ozonolysis, reduction and Wittig). Rf 0.69
(1:1 ethyl acetate/hexane) 1H NMR (600 MHz, CDCl3): 6.99 (d, 1H
J = 15.3 Hz), 6.77 (s, 1H), 6.27 (d, 1H J = 15.3 Hz), 4.24 (q, 2H, J =
7.2 Hz), 4.19 (q, 2H, J = 7.2 Hz), 3.82 (s, 3H), 3.81 (s, 3H), 2.89−
2.80 (m, 2H), 2.10 (ddd, 1H, J = 15.9, 9.9, 6.6 Hz), 1.96 (ddd, 1H, J =
15.9, 9.9, 6.6 Hz), 1.31 (t, 3H, J = 7.2 Hz), 1.29 (t, 3H, J = 7.2 Hz)
(hydroxl proton not detected); 13C NMR (150 MHz, CDCl3): 174.0,
166.8, 166.1, 165.5, 146.7, 146.1, 127.7, 121.8, 76.8, 60.9, 60.5, 53.3,
52.5, 37.4, 22.3, 14.1, 14.0; IR (neat) 3497, 2956, 1711, 1647, 1438,
1263, 1176, 1028, 983, 732 cm−1; LRP (+LSIMS) m/z (%) 373 (M+,
100), 355 (19), 327 (30), 309 (34), 295 (32), 281 (74), 267 (36), 249
(28), 235 (25), 221 (60), 207 (68); HRMS calcd. for (M + H)+
C17H25O9: 373.1499; found 373.1499.
( )-Methyl-[(3S,6R)-6-formyl-6-methyl-3-(propan-2-yl)-1,2-
dioxan-3-yl]acetate (33b). Colorless oil. Yield 11%. Rf 0.53 (3:7
1
ethyl acetate/hexane). H NMR (600 MHz, CDCl3): 9.58 (s, 1H),
3.78 (s, 3H), 2.41−2.32 (m, 1H), 2.14−1.98 (m, 4H), 1.32 (s, 3H),
0.99 (d, 3H, J = 7.2) 0.97 (d, 3H, J = 7.2 Hz); 13C NMR (150 MHz,
CDCl3): 200.9, 170.8, 87.4, 84.6, 52.1, 31.9, 24.9, 22.7, 18.2, 16.9, 16.8;
IR (neat) 2972, 1737, 1444, 1372, 1243, 1043, 754 cm−1; LRP
(+LSIMS) m/z (%): 231 (M+, 10), 201 (5), 185 (21), 171 (8), 155
(100), 143 (19); HRMS calcd. for (M + H)+ C11H19O5: 231.1232;
found 231.1240.
( )-1,7-Diethyl-3-methyl (1E,3S,5E)-3-hydroxy-6-propanoyl-
hepta-1,5-diene-1,3,7-tricarboxylate (26). Colorless oil. Yield
31% (over three steps: ozonolysis, reduction and Wittig). Rf 0.57
( )-Trimethyl-(5S)-5-hydroxydihydrofuran-2,2,5(3H)-tricar-
boxylate (34). Colorless waxy oil. Yield: 25%. Rf 0.52 (3:7 ethyl
acetate/hexane). 1H NMR (600 MHz, CDCl3): 4.26 (s, OH), 3.84 (s,
3817
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
3H), 3.82 (s, 3H), 3.80 (s, 3H), 2.74−2.68 (m, 2H), 2.54 (dt, 1H, J =
13.2, 9 Hz), 2.17 (m, 1H); 13C NMR (150 MHz, CDCl3): 169.4,
169.2, 168.4, 104.1, 87.3, 53.4, 53.3, 53.2, 34.4, 32.2; IR (neat) 3471,
2964, 2364, 1749, 1441, 1288, 1209, 1076, 1016, 669 cm−1; LRP
(+LSIMS) m/z (%): 295 (18), 267 (30), 245 ((M − OH)−, 33), 221
(100), 204 (33), 193 (33), 185 (34); HRMS calcd. for (M − OH)−
C10H13O7: 245.0661; found 245.0666.
(17) Bascetta, E.; Gunstone, F. D.; Scrimgeour, C. M. J. Chem. Soc.,
Perkin Trans. 1 1984, 2199−2205.
(18) Foster, C. H.; Berchtold, G. A. J. Org. Chem. 1975, 40, 3743−
3746.
(19) Schwartz, C.; Raible, J.; Mott, K.; Dussault, P. H. Tetrahedron
2006, 62, 10747−10752.
(20) Criegee, R. Ann. 1953, 583, 1−2.
(21) Criegee, R. Angew. Chem. 1975, 87, 765−771.
(22) Jung, I. C. Eur. J. Org. Chem. 2001, 1899−1901.
(23) Geletneky, C.; Berger, S. Eur. J. Org. Chem. 1998, 1625−1627.
(24) Griesbaum, K.; Volpp, W.; Greinert, R.; Greunig, H. J.; Schmid,
J.; Henke, H. J. Org. Chem. 1989, 54, 383−389.
(25) Smith, M.; March, J. March’s Advanced Organic Chemistry, 5th
ed.; John Wiley & Sons, Inc.: New York, 2001.
(26) Schank, K. Helv. Chim. Acta 2004, 87, 2074−2084.
(27) Bailey, P. S.; Ferrell, T. M. J. Am. Chem. Soc. 1978, 100, 899−
905.
ASSOCIATED CONTENT
■
S
* Supporting Information
Crystallographic data for compounds 22 and 33a; H and 13C
1
NMR data for compounds 8e, 13, 7, 16, 17, 11a, 18a, 18b, 22,
25, 26, 27, 33a and 34; VT H and 13C NMR data for
1
compounds 18d and 21. This material is available free of charge
via the Internet at http://pubs.acs.org.
(28) Ponec, R.; Yuzhakov, G.; Haas, Y.; Samuni, U. J. Org. Chem.
AUTHOR INFORMATION
■
1997, 62, 2757−2762.
(29) Higley, D. P.; Murray, R. W. J. Am. Chem. Soc. 1976, 98, 4526−
4533.
Corresponding Author
*dennis.taylor@adelaide.edu.au
(30) Hon, Y.-S.; Wong, Y.-C. Tetrahedron Lett. 2005, 46, 1365−1368.
(31) Hansen, M. M.; Bertsch, C. F.; Harkness, A. R.; Huff, B. E.;
Hutchison, D. R.; Khau, V. V.; LeTourneau, M. E.; Martinelli, M. J.;
Misner, J. W.; Peterson, B. C.; Rieck, J. A.; Sullivan, K. A.; Wright, I. G.
J. Org. Chem. 1998, 63, 775−785.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Thomas D. Avery and Dr. Gordon M. Elsey for
their initial assistance with this work. N.M.C. thanks the Faculty
of Science, The University of Adelaide, for a postgraduate
scholarship. This project was supported by the School of
Agriculture, Food and Wine, University of Adelaide, as well as
by Australia’s grapegrowers and winemakers through their
investment body, the Grape and Wine Research and Develop-
ment Corporation, with matching funds from the Australian
government. The Ministry of Higher Education (Malaysia) is
thanked for funding structural studies through the High-Impact
Research scheme (UM.C/HIR/MOHE/SC/12).
(32) Shiao, H.-Y.; Hsieh, H.-P.; Liao, C.-C. Org. Lett. 2008, 10, 449−
452.
(33) Mehta, G.; Vidya, R. J. Org. Chem. 2000, 65, 3497−3502.
(34) Ayer, W. A.; Talamas, F. X. Can. J. Chem. 1988, 66, 1675−1685.
(35) Gumulka, J.; Szczepek, W. J.; Wielogorski, Z. A. Pol. J. Chem.
1983, 57, 403−411.
(36) Gumulka, J.; Szczepek, W. J.; Wielogorski, Z. Tetrahedron Lett.
1979, 4847−4850.
(37) Matsumoto, M.; Dobashi, S.; Kuroda, K.; Kondo, K. Tetrahedron
1985, 41, 2147−2154.
(38) Campagnole, M.; Bourgeois, M.-J.; Montaudon, E. Tetrahedron
2002, 58, 1165−1171.
(39) Balci, M. Chem. Rev. 1981, 81, 91−108.
REFERENCES
(40) Jin, H.-X.; Liu, H.-H.; Zhang, Q.; Wu, Y. J. Org. Chem. 2005, 70,
4240−4247.
■
(1) Yadav, J. S.; Babu, R. S.; Sabitha, G. ARKIVOC 2003, 125−139.
(2) Macreadie, P.; Avery, T.; Greatrex, B.; Taylor, D.; Macreadie, I.
Bioorg. Med. Chem. Lett. 2006, 16, 920−922.
(41) Ponce, M. A.; Ramirez, J. A.; Galagovsky, L. R.; Gros, E. G.;
Erra-Balsells, R. Photochem. Photobiol. Sci. 2002, 1, 749−756.
(42) Takahashi, K.; Yamaguchi, Y.; Hayashi, A. Acta Crystallogr., Sect.
C: Cryst. Struct. Commun. 1991, C47, 2581−2583.
(43) Schreiber, S. L.; Claus, R. E.; Reagan, J. Tetrahedron Lett. 1982,
23, 3867−3870.
(3) Avery, T. D.; Macreadie, P. I.; Greatrex, B. W.; Robinson, T. V.;
Taylor, D. K.; Macreadie, I. G. Bioorg. Med. Chem. 2007, 15, 36−42.
(4) Ann Casteel, D. Nat. Prod. Rep. 1999, 16, 55−73.
(5) Greatrex, B. W.; Taylor, D. K. J. Org. Chem. 2004, 69, 2577−
2579.
(6) Greatrex, B. W.; Kimber, M. C.; Taylor, D. K.; Fallon, G.;
Tiekink, E. R. T. J. Org. Chem. 2002, 67, 5307−5314.
(7) Brown, R. C.; Sefton, M. A.; Taylor, D. K.; Elsey, G. M. Aust. J.
Grape Wine Res. 2006, 12, 115−118.
(8) Avery, T. D.; Taylor, D. K.; Tiekink, E. R. T. J. Org. Chem. 2000,
65, 5531−5546.
(44) Kawamura, S.-i.; Yamakoshi, H.; Nojima, M. J. Org. Chem. 1996,
61, 5953−5958.
(45) Adam, W.; Erden, I. Angew. Chem., Int. Ed. Engl. 1978, 17, 210.
(46) Pryor, W. A.; Giamalva, D.; Church, D. F. J. Am. Chem. Soc.
1985, 107, 2793−2797.
(47) Pryor, W. A.; Giamalva, D.; Church, D. F. J. Am. Chem. Soc.
1983, 105, 6858−6861.
(48) Munaf Kharbuli, A.; Duncan Lyngdoh, R. H. THEOCHEM
2008, 860, 150−160.
(9) Avery, T. D.; Fallon, G.; Greatrex, B. W.; Pyke, S. M.; Taylor, D.
K.; Tiekink, E. R. T. J. Org. Chem. 2001, 66, 7955−7966.
(10) Kimber, M. C.; Taylor, D. K. J. Org. Chem. 2002, 67, 3142−
3144.
(49) Iwai, T.; Fujihara, T.; Tsuji, Y. Chem. Commun. 2008, 6215−
6217.
(50) Fristrup, P.; Kreis, M.; Palmelund, A.; Norrby, P.-O.; Madsen, R.
(11) O’Shea, K. E.; Foote, C. S. J. Org. Chem. 1989, 54, 3475−3477.
(12) Kleinpeter, E. In Compr. Heterocycl. Chem. III; Katritzky, A. R.,
Ed.; Elsevier: UK, 2008; Vol. 8, pp 677−738.
J. Am. Chem. Soc. 2008, 130, 5206−5215.
(51) Doughty, D. H.; Pignolet, L. H. J. Am. Chem. Soc. 1978, 100,
7083−7085.
(52) Curtin, D. Y.; Kauer, J. C. J. Org. Chem. 1960, 25, 880−885.
(53) Berman, J. D.; Stanley, J. H.; Sherman, W. V.; Cohen, S. G. J.
Am. Chem. Soc. 1963, 85, 4010−4013.
(54) During the examination of this work, a reviewer suggested that
we may wish to consider an alternative heterolytic pathway for
formation of compounds of type 22 from 8c that involved the loss of
CO2 during the reaction sequence. While the examiner and we concur
(13) Robinson, T. V.; Taylor, D. K.; Tiekink, E. R. T. J. Org. Chem.
2006, 71, 7236−7244.
(14) Valente, P.; Avery, T. D.; Taylor, D. K.; Tiekink, E. R. T. J. Org.
Chem. 2009, 74, 274−282.
(15) Zvarec, O.; Avery, T. D.; Taylor, D. K.; Tiekink, E. R. T.
Tetrahedron 2010, 66, 1007−1013.
(16) Adam, W.; Bloodworth, A. J.; Eggelte, H. J.; Loveitt, M. E.
Angew. Chem., Int. Ed. Engl. 1978, 17, 209.
3818
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819

The Journal of Organic Chemistry
Article
that such an alternative cannot explain the formation of all of the
unusual products of similar type to 22, it is worthy of consideration.
Thus, Richardson demonstrated that 2-alkylperoxyalkanols undergo a
relatively facile base-catalyzed fragmentation with loss of a form-
aldehyde/aldehyde moiety ( Richardson, W. H.; Heeson, T. C. J. Org.
Chem. 1972, 37, 3416−3419. ). In our study, the necessary alcohol for
the Richardson-type fragmentation could be generated from either
hydration or reversible nucleophilic addition to the aldehyde-
substituted 1,2-dioxines. This mechanism can also occur equally well
on the products of nonsymmetrical ozonolysis since one end will
always contain a peroxy aldehyde.
(55) Ramesh, P.; Reddy, V. L. N.; Reddy, N. S.; Venkateswarlu, Y. J.
Nat. Prod. 2000, 63, 1420−1421.
(56) Fuchter, M. J.; Hoffman, B. M.; Barrett, A. G. M. J. Org. Chem.
2006, 71, 724−729.
(57) Suzuki, M.; Ohtake, H.; Kameya, Y.; Hamanaka, N.; Noyori, R.
J. Org. Chem. 1989, 54, 5292−5302.
(58) Moreau, J. L.; Couffignal, R. J. Organomet. Chem. 1985, 294,
139−144.
(59) Petrova, O. V.; Mikhaleva, A. I.; Sobenina, L. N.; Trofimov, B.
A. Russ. J. Org. Chem. 2010, 46, 452−454.
(60) Huang, J. T.; Su, T. L.; Watanabe, K. A. J. Org. Chem. 1991, 56,
4811−4815.
3819
dx.doi.org/10.1021/jo3001518 | J. Org. Chem. 2012, 77, 3808−3819