Ene-diene transmissive cycloaddition reactions with singlet oxygen: The vinylogous gem effect and its use for polyoxyfunctionalization of dienes

Article abstract of DOI:10.1021/jo5000434

The singlet oxygen reactivities and regioselectivities of the model compounds 1b-d were compared with those of the geminal (gem) selectivity model ethyl tiglate (1a). The kinetic cis effect is k_E/k_Z = 5.2 for the tiglate/angelate system 1a/1a′ without a change in the high gem regioselectivity. Further conjugation to vinyl groups enabled mode-selective processes, namely, [4 + 2] cycloadditions versus ene reactions. The site-specific effects of methylation on the mode selectivity and the regioselectivity of the ene reaction were studied for dienes 1e-g. A vinylogous gem effect was observed for the γ,δ-dimethylated and α,γ,δ-trimethylated substrates 1h and 1i, respectively. The corresponding phenylated substrates 1j-l showed similar mode selectivity, as monomethylated 1j exhibited exclusively [4 + 2] reactivity while the tandem products 12 and 14 were isolated from the di- and trimethylated substrates 1k and 1l, respectively. The vinylogous gem effect favors the formation of 1,3-dienes from the substrates, and thus, secondary singlet oxygen addition was observed to give hydroperoxy-1,2-dioxenes 19 and 20 in an ene-diene transmissive cycloaddition sequence. These products were reduced to give alcohols (16, 17, and 18) or furans (24 and 25), respectively, or treated with titanium(IV) alkoxides to give the epoxy alcohols 26 and 27. The vinylogous gem effect is rationalized by DFT calculations showing that biradicals are the low-energy intermediates and that no reaction path bifurcations compete.

Full text of DOI:10.1021/jo5000434

Article
pubs.acs.org/joc
Ene−Diene Transmissive Cycloaddition Reactions with Singlet
Oxygen: The Vinylogous Gem Effect and Its Use for
Polyoxyfunctionalization of Dienes
Angelika Eske, Bernd Goldfuss, Axel G. Griesbeck,* Alan de Kiff, Margarethe Kleczka, Matthias Leven,
Jorg-M. Neudorfl, and Moritz Vollmer
̈
̈
Department of Chemistry, University of Cologne, Greinstr. 4, D-50939 Cologne, Germany
S
* Supporting Information
ABSTRACT: The singlet oxygen reactivities and regioselec-
tivities of the model compounds 1b−d were compared with
those of the geminal (gem) selectivity model ethyl tiglate (1a).
The kinetic cis effect is kE/kZ = 5.2 for the tiglate/angelate
system 1a/1a′ without a change in the high gem regioselectivity. Further conjugation to vinyl groups enabled mode-selective
processes, namely, [4 + 2] cycloadditions versus ene reactions. The site-specific effects of methylation on the mode selectivity
and the regioselectivity of the ene reaction were studied for dienes 1e−g. A vinylogous gem effect was observed for the γ,δ-
dimethylated and α,γ,δ-trimethylated substrates 1h and 1i, respectively. The corresponding phenylated substrates 1j−l showed
similar mode selectivity, as monomethylated 1j exhibited exclusively [4 + 2] reactivity while the tandem products 12 and 14 were
isolated from the di- and trimethylated substrates 1k and 1l, respectively. The vinylogous gem effect favors the formation of 1,3-
dienes from the substrates, and thus, secondary singlet oxygen addition was observed to give hydroperoxy-1,2-dioxenes 19 and
20 in an ene−diene transmissive cycloaddition sequence. These products were reduced to give alcohols (16, 17, and 18) or
furans (24 and 25), respectively, or treated with titanium(IV) alkoxides to give the epoxy alcohols 26 and 27. The vinylogous
gem effect is rationalized by DFT calculations showing that biradicals are the low-energy intermediates and that no reaction path
bifurcations compete.
For 1,3-dienes or higher polyenes, additional chemical and
physical processes can compete with the ene process, including
[2 + 2] and [4 + 2] cycloadditions and physical quenching of
singlet oxygen.11 The latter process often dominates with
extended polyenes such as several carotenoids or tocopherols
that quench singlet oxygen with near-diffusion-controlled
rates.12 Under cellular conditions, however, this protective
behavior was recently questioned for β-carotene.13 Other
carotenoids and apocarotenoids show competing physical and
chemical quenching of singlet oxygen.14 Among these dienes,
crocin (Figure 1) represents an α,ω-dimethylated tetradeca-
heptaene-1,14-dicarboxylic acid that has been found to
INTRODUCTION
■
The singlet oxygen ene reaction is one of the synthetically most
versatile photooxygenation protocols.1 Not only are allylic
hydroperoxides (the initial products of the Schenck ene
1
reaction with O2) available by this process, but also, allylic
alcohols, epoxides, epoxy alcohols, 1,2-diols, and 1,2,3-triols can
be obtained via simple reductive or transition-metal-catalyzed
transformations.2 Several mechanistic scenarios have been
postulated for this reaction because numerous characteristics
of this process are not in agreement with a symmetry-allowed
concerted reaction course.3 Classical two-step processes
involving 1,4-biradical, 1,4-zwitterion, perepoxide, dioxetane,
or exciplex intermediates have been discussed.4 The results of
inter- and intramolecular isotope experiments with isotopically
labeled tetramethylethylenes provided evidence for suprafacial
1O2 attack and for a perepoxide intermediate.5 Also, the small
negative activation enthalpies and highly negative activation
entropies observed for the singlet oxygen ene reaction argue for
the participation of a reversibly formed exciplex as an
intermediate.6 Recently, the two-step no-intermediate mecha-
nism involving a bifurcating transition state with a perepoxide
structure was proposed by Houk and co-workers.7 Small
changes in the substrate structure can also open a low-energy
pathway via a real perepoxide intermediate.8 The valley ridge
model9 appears to be a powerful explanation of the cis effect
(vide infra) as well as deviations that result from geometrical
constraints of the allylic hydrogen atoms.10
Figure 1. Apocarotenoids with relevant structure motifs.
Received: January 23, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
chemically react with singlet oxygen with a rate constant of 0.56
× 108 M−1 s−1 in water.15 The corresponding bixines (Figure 1)
are γ-methylated octadecanonaenes with similar singlet oxygen
reactivity. Obviously, these compounds have multiple sites for
singlet oxygen reactions, and diverse primary and secondary
processes can be expected. However, no detailed studies of the
primary chemical reaction events or the structures of the first
peroxidic products with these apocarotenoids have been
reported. In an initial approach to this problem, we identified
three critical substrate motifs in these biological substrates: (a)
the tiglate-type structure with the potential ene-reactive α-
methyl group, (b) the α-methylated diene ester, and (c) the γ-
methylated diene ester motifs. This study aimed to investigate
the (chemical) singlet oxyen reactivity with these structural
motifs in order to determine the influence of β-vinylation of the
methacrylate system in type (a) and the introduction of methyl
groups at the α-, γ-, and δ-positions of enone esters as exist in
types (b) and (c).
The locants lone, twix, and twin are also used to describe the
regions in trisubstituted alkenes (Scheme 2). For the cis effect,
Scheme 2. Regioselectivity Trends in Singlet Oxygen Ene
Reactions: Medium and Gem Effects
preferential hydrogen interaction and subsequent abstraction
from the lone and twin positions is observed in solution. This
effect is nearly independent of the nature of the allylic
substituent X. Under restricted environment conditions,
however, this selectivity is altered: in zeolites, the twix
selectivity dominates strongly,19 whereas in microemulsions
and more noticeably in hydrogels, hydrogen abstraction from
the lone methylene group becomes prevalent.20 Another very
strong regiodirecting effect was observed for acceptor-
substituted 2-butenes. This phenomenon, termed the gem
effect, describes the regioselective ene reaction of 1O2 with α,β-
unsaturated carbonyl compounds and carbonyl analogues
(nitriles, sulfoxides, and sulfonates).21 There is only a little
environmental influence on this kind of selectivity effect.22
Tiglate Substrates and 1,3-Diene-1-carboxylates:
Reactivity and Selectivity Studies. In order to evaluate
the kinetic contribution of the cis effect, we compared the
reactivities of the E and Z isomers of 2-methyl-2-butenoic acid
ethyl ester (1a and 1a′, respectively). Both stereoisomers result
in the gem-type allylic hydroperoxide 2a with diastereoselectiv-
ities of >98% (Scheme 3), that is, there exists no cis effect in the
RESULTS
■
Cis and Gem Effects with Monoalkenes and Michael
Esters. The term cis effect describes a kinetic preference of
alkenes with two reactive groups at positions 1 and 2 in the cis
configuration in comparison with the corresponding trans
isomer. For singlet oxygen ene reactions, this definition is often
extended and also includes the regioselectivity of the hydrogen
transfer, meaning that CH activation is preferred for alkene
sites with the cis configuration of groups presenting allylic
hydrogens. This phenomenon, which was investigated in-depth
by extensive substrate variation and sophisticated kinetic
isotope and product isotope effects, has been interpreted as a
hint for the perepoxide intermediate model. Following this
model, in the rate-determining step of the singlet oxygen ene
1
reaction, a twofold interaction between O2 and the allylic
hydrogens accelerates the reaction by a factor of 3−4 for cis-2-
butene (A) versus trans-2-butene (B) (Scheme 1).16 This
Scheme 1. Regioselectivity Trends in Singlet Oxygen Ene
Reactions: Cis Effect
Scheme 3. Singlet Oxygen Ene Reactions with Tiglate and
Angelate Esters and Their Derivatives
primary interaction is also represented in the regioisomer ratios.
For example, for 2-methyl-2-butene (D), hydrogen transfer
from the higher-substituted site of the alkene prevails by 93%.
Two methyl groups in the cis configuration (C) give optimal cis
effects with respect to the total reactivity and regioselectivity.
The third (twin) methyl in D contributes only 7% of the
hydrogen transfer.17 For the diastereomeric allylic alcohols E
and F, the product cis effects are 100% for E and 92% for F,18
while the kinetic cis effect (kE/kF) is 2.2 for these substrates.
product-forming step. The kinetic cis effect was determined
from the pseudo-first-order substrate consumption curves
depicted in Figure 2 (a 1:1 mixture of alkene double-bond
1
isomers was reacted with O2 in CCl4). In agreement with the
oxygen uptake curves, the curve progression switch to second-
order behavior after 70% conversion and a kinetic cis effect of
5.2 were determined for 1a and 1a′. A subsequent decrease in
the kinetic cis effect is expected for a decreasing number of
allylic hydrogens on the β-alkyl group. The β-ethyl substrate 1b
B
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
shows a rate decrease by a factor of 40 for the ene reaction with
1e. Overall, the mode selectivity (i.e., ene vs [4 + 2] reaction
mode) is low for this substrate with no ene regiodifferentiation
possible here.
A comparable solvent effect on the mode selectivity was
observed for the α,δ-dimethylated substrate 1f (Scheme 5). The
Scheme 5. Solvent Dependence of the Singlet Oxygen Ene
Reaction with 1f
Figure 2. Singlet oxygen ene reaction with ethyl tiglate (1a) vs ethyl
angelate (1a′).
1
(R′ = H, R″ = Me) reacted slightly faster with O2, possibly
because 1,3-allylic strain positions the allylic hydrogens in the
optimal perpendicular orientation on both sides of the π-
system. This effect breaks down for one side of the π-system for
the β-isopropyl substrate 1c (R′ = R″ = Me) and completely for
the tert-butyl substrate 1d, resulting in a relative kinetic
situation of 1:1.2:0.6:0.05 for 1a:1b:1c:1d (Figure 3).
δ-methyl group was completely inert under the reaction
conditions even after prolonged irradiation time, and the ene
product 2f from α-hydrogen transfer was dominant in ethanol,
where the reaction could only be processed to 52% conversion.
The additional methyl group in 1f accelerates the singlet
oxygen reaction by a factor of 5 in comparison with 1e.
The picture changed substantially for the α,γ-dimethylated
substrate 1g (Scheme 6). The solvent effect on the mode
Scheme 6. Solvent Dependence of the Singlet Oxygen Ene
Reaction with 1g
Figure 3. Singlet oxygen ene reaction with β-alkylated E-configured
substrates 1a−d.
selectivity vanished nearly completely, and under all solvent
conditions endoperoxide 5 was obtained as the major product
with complete conversion in nonpolar solvents after less than
24 h. Compound 5 was crystallized from acetone and
characterized by X-ray structure analysis (Figure 4).24 Upon
comparison of the reactivities of the two methyl groups at the
α- and γ-positions, the α-hydrogen transfer product 2g was
observed and isolated as the respective allylic alcohol, whereas
A substantial electronic modification of the tiglate system
results from the incorporation of the β-vinyl group in 1e. This
compound shows solvent-dependent mode selectivity between
1
the [4 + 2] cycloaddition and the ene reaction with O2
(Scheme 4). The ene reaction, which is suspected to proceed
via a more polar transition state compared with the
asynchronous [4 + 2] process,23 became the dominant process
in polar protic solvents. The kinetic comparison with substrate
1b exhibiting the maximum kinetic cis effect (see Figure 2)
Scheme 4. Solvent Dependence of the Singlet Oxygen Ene
Reaction with 1e
Figure 4. Crystal structure of endoperoxide 5.
C
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
less than a 2% yield of the γ-hydrogen transfer product 6 was
observed by H NMR analysis.
The next diene substrate, γ,δ-dimethylated 1h, was
investigated as the pure E,E diastereoisomer as well as a 9:1
mixture of the E,Z and E,E diastereoisomers (Scheme 7). In
has the largest number of reaction possibilities, including three
different methyl groups for ene reactions as well as the 1,3-
diene moiety for [4 + 2] cycloaddition. Conformational reasons
cannot be responsible for the prevalence of the ene process
because no substantial difference in the s-cis/s-trans conforma-
tional ratio can be expected for 1g in comparison with E,E-1i.
Thus, this scenario of mode and ene regioselectivity is a perfect
example for a kinetic Curtin−Hammett situation. The
computational aspects of the strong vinylogous gem selectivities
of 1h and 1i are discussed later.
1
Scheme 7. Singlet Oxygen Ene Reactions with E,E(E,Z)-1h
and E,E(E,Z)-1i
1-Phenyl-1,3-dienes: Reactivity and Selectivity Stud-
ies. In order to evaluate the effect of conjugation preservation
(vide infra) as the decisive factor for regiocontrol in the gem
and vinylogous gem effects, substrates 1j−l with a phenyl group
instead of the carboxy group in 1a−i were studied. In the
absence of a kinetic cis effect, diene 1j prefers [4 + 2] reactivity
in a slow singlet oxygen reaction (Scheme 8).
Scheme 8. Singlet Oxygen Ene Reactions with 1j, E,Z-1k,
and E,Z-1l
both cases, rapid singlet oxygen addition delivered the allylic
hydroperoxide 7 as the sole product after >90% conversion of
the starting material. Even faster were the reactions of singlet
oxygen with the α,γ,δ-trimethylated substrates 1i (also applied
as the pure E,E diastereoisomer as well as a 4:1 mixture of the
E,Z and E,E isomers), which delivered allylic hydroperoxide 9.
In both cases, only trace amounts of the corresponding [4 + 2]
cycloadducts 8 and 10, respectively, were observed in the NMR
spectra. Upon further treatment of the substrate solutions with
singlet oxygen, the tandem products 19 and 20 were formed
(see later).
The most reactive substrate in this series of 1,3-diene
carboxylates is the E,E isomer of 1i (Table 1). Obviously, this
substrate profits from the kinetic cis effect as well as from the
highly nucleophilic diene structure. Remarkably, substrate 1i
The di- and trimethylated substrates 1k and 1l, however,
showed high vinylogous gem regioselectivity. The analogy to
the ester substrates 1h and 1i is apparent. A moderately higher
singlet oxygen reactivity and a decreased mode selectivity (i.e.,
more [4 + 2] cycloaddition products 13 and 15) were observed
in the crude reaction mixture from the phenylated substrates 1k
and 1l (Table 2). However, no primary ene adducts could be
Table 1. Singlet Oxygen Ene Reactions with 1,3-Diene
Carboxylates 1e−i in CDCl3
yields (%)
Table 2. Singlet Oxygen Ene Reactions with 1-Phenyl-1,3-
dienes 1j−l in CDCl3
Me subst.
pattern
t>90%
a
b
substrate
(h)
krel
α-ene γ-ene [4 + 2]
1e
α
120
48
22
12
28
6
5
16
30
56
35
6
−
−
1
44
65
93
<2
<2
<2
<2
yields (%)
1f
α,δ
α,γ
γ,δ
a
substrate Me subst. pattern t>90% (h)
α-ene
γ-ene
[4 + 2]
1g
1j
1k
1l
γ
48
24
10
−
−
<2
<5
>95
c
E,E-1h
E,Z-1h
E,E-1i
E,Z-1i
50
23
−
−
−
−
>98
>98
>98
>98
b
γ,δ (R = H)
α,γ,δ (R = Me)
93
7 (13)
cd
,
γ,δ
b
73
27 (15)
e
α,γ,δ
α,γ,δ
100
48
a
Time for >90% conversion as determined by 1H NMR analysis.
Based on tandem products (12 from 1k and 14 from 1l) from a
sequence of γ-ene followed by [4 + 2] cycloaddition.
de
,
13
b
a
Time needed for >90% conversion as determined by 1H NMR
analysis. The value for the fastest substrate, E,E-1i, was set to 100.
Corresponds to a kinetic cis effect of 2.2. Corrected to 100% E,Z.
Corresponds to a kinetic cis effect of 2.1.
b
c
d
observed from these substrates because the secondary singlet
oxygen [4 + 2] cycloadditions were faster than the primary ene
processes, so only the tandem products 12 and 14, respectively,
were isolated. The reaction times indicated in Table 2 thus
describe the quantitative addition of two singlet oxygen
molecules in a consecutive way for the substrates 1k and 1l.
Further Tandem and Oxyfunctionalization Reactions.
In most cases, the dienylic hydroperoxides from gem- or
e
D
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
for further singlet oxygenation (i.e., in tandem singlet oxygen
with combined reduction). Whereas the allylic alcohols 16e−g
were nonreactive, the diene hydroperoxides 7 and 9 did add a
second singlet oxygen molecule to give the 1,2-dioxenes 19 and
20, respectively, as did the diene carbinols 17 and 18 to give 21
and 22, respectively (Scheme 9).25 In Figure 5, three NMR
traces are shown that demonstrate the cleanness of these
tandem processes: photooxygenation of diene 1i and
subsequent reduction with dimethyl sulfide delivered diene
alcohol 18 after a short irradiation time. The subsequent slower
photooxygenation delivered endoperoxide 22 as a mixture of
diastereoisomers. In the case of the arylated substrates 1k and
1l, this rate difference was inverted, and thus, only the tandem
products 12 and 14, respectively, were obtained (vide supra).
The tandem process could likewise be performed without
isolation of the intermediate hydroperoxides by prolonged
irradiation of the sensitizer-containing reaction mixture, leading
to bisperoxides 19 and 20, respectively. Reduction of these
products with dimethyl sulfide gave the identical products as
the three-step oxygenation/reduction/oxygenation route in
comparable overall yields (Scheme 9).
Scheme 9. Reduction of Ene Products and Tandem
Processes with 1,3-Dienes 7 and 9
The reduction of 1,2-dioxenes with thiourea resulted in the
corresponding furans (Scheme 10).26 The tandem product 12
(from 1k) was reduced to hydroxy 1,2-dioxene 23 and
subsequently converted to furan 24, and accordingly, this
reaction sequence delivered furan 23 from substrate 1h via
tandem singlet oxygenation and one-pot reduction with
vinylogous-gem-selective ene reactions were not stable under
standard purification procedures. In these instances, reduction
to the stable diene carbinols was performed with dimethyl
sulfide, and these compounds were investigated as substrates
Figure 5. 1H NMR spectra of (1) diene E,E-1i, (2) dienol 18 (after 6 h of reaction time and reduction), and (3) tandem product 22
(diastereoisomeric mixture, 1:1).
E
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
product 7 (Figure 6). Both possible pathways (to 7 and 7′)
were calculated for each substrate employing the nonempirical
TPSS functional developed by Staroverov et al.28 to reach high
accuracy for the description of hydrogen abstraction. All of the
computations were carried out as unrestricted DFT to allow
spin contamination within the ground-state determinant
description of biradical intermediates and the TZVP basis set
from Ahlrich et al.29 was used.
Scheme 10. Secondary Transformations of 1,2-Dioxenes and
Allylic Hydroperoxides
It was theoretically evaluated by Maranzana et al. in 2005
that α,β-unsaturated carbonyl compounds are prone to form
biradical intermediates in Schenck ene-type reactions.30 Similar
reaction paths were also found for the substrates E,E-1h and
E,Z-1h. The reaction coordinate depicted in Figure 6,
corresponding to the transformation of 1h, starts with the
addition of singlet oxygen to generate the conjugated biradical
intermediate BRa. Alternatively, the nonconjugated biradical
BRc can be generated, from which eventually the regioisomeric
product 7′ results. However, it was found that the non-
conjugated intermediate BRc is not a stationary point on the
potential energy surface, so all further downstream products
originate in 7. The regioselectivity-determining hydrogen
abstraction follows the initial biradical stage. As depicted in
Figure 6, the activation barrier for the formation of the
experimentally found product 7 is almost zero (0.3 kcal/mol).
The formation of the disfavored product 7′ from the low-lying
biradical intermediate is clearly kinetically hindered since the
activation barrier for the corresponding hydrogen abstraction
and 1,2-oxygen migration (nonleast motion) is 7.7 kcal/mol
(TS2a). On the thermodynamic side, comparison of the
possible products 7 and 7′ does also show a clear preference
for the experimentally favored product 7 as well. Product 7,
which exhibits a larger conjugated double bond system is
energetically favored by 4.6 kcal/mol. The reaction coordinate
for the ene reaction of the double-bond-isomeric diene E,Z-1h
with singlet oxygen exhibits very similar properties. Again, there
is only one minimum biradical intermediate, BRb, which is
dimethyl sulfide and thiourea. Another useful transformation of
allylic hydroperoxides was also applied in this investigation,
namely, titanium(IV)-catalyzed oxygen transfer to give epoxy
alcohols.27 The primary singlet oxygen addition products with
the diene esters 1h and 1i (the allylic hydroperoxides 7 and 9,
respectively) were transformed in situ into the epoxy alcohols
26 and 27, respectively, by treatment of the crude reaction
mixture (after uptake of 1 equiv of oxygen) with catalytic
amounts of titanium tetraisopropoxylate. Thus, the tandem
products as well as the primary addition products are useful
substrates for further oxyfunctionalization.
Vinylogous Gem Effect: Theoretical Studies. The
reactions of dienes E,E-1h and E,Z-1h with singlet oxygen
were also investigated computationally by means of density
functional theory in order to rationalize the high vinylogous gem
regioselectivity that favors the formation of the conjugated
Figure 6. Computational analysis of the singlet oxygen ene reactions of E,E-1h and E,Z-1h. The experimentally determined regioselectivities
originate in the almost barrierless pathways including transition states TS1a and TS1b, respectively.
F
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 11. Substituent Dependence of 1O2 Addition to 1,3-
Dienes; Biradical Intermediates BRa,b and the s-cis/s-trans
Conformational Situation for Mono- and Dimethylated
Substrates
slightly lower in energy than BRa. There is also a very small
activation barrier of 0.2 kcal/mol, via TS1b, for the hydrogen
abstraction that leads to the favored product 7, whereas the
transition state TS2b leading to the disfavored product 7′ is
markedly higher in energy than TS2a. The downstream
structures derived from E,Z-1h differ from the intermediate
BRa and the transition states TS1a and TS2a only by slight
conformational distortions that even increase the selectivity for
the hydrogen abstraction for E,Z-1h (Figure 6). A conforma-
tional equilibrium at the biradical stage is most unlikely, since
every rotation barrier will be much higher in energy than TS1a
and TS1b and conformational equilibrium is eventually
established at the product hydroperoxide stage.
A closer look at the competing transition structures in the
regiodetermining step (Figure 7) elucidates why the favored
These effects and the theoretical calculations account for a two-
step reaction mechanism. In the first step, via a perepoxide-like
transition state, a biradical intermediate is formed that
undergoes hydrogen abstraction from a geminal or vinylogous
geminal position. If the first step is kinetically not favored,
either by the missing terminal methyl group or by the
nonexistent cis effect, the [4 + 2] cycloaddition reaction
prevails, most likely proceeding as an asynchronous concerted
cycloaddition.
Calculations showed that the regioselectivity of the singlet
oxygen ene reactions with the di- and trimethylated substrates
1h−l results from a biradical BRa that is converted by a nearly
barrier-free process to the corresponding hydrogen transfer
product with vinylogous gem selectivity.
Figure 7. Competing transition structures for hydrogen abstraction,
TS1a and TS2a (TPSS/TZVP).
pathway exhibits a much lower activation barrier. The bond
length between the peroxo group and the carbon framework in
TS1a is 1.514 Å. The corresponding oxygen−carbon distance in
the disfavored counterpart TS2a is markedly longer (1.748 Å).
This difference is due to a synchronous 1,2-shift of the peroxo
group that includes C−O bond cleavage. On the other hand, it
is apparent that the conjugated double bond system is
separated in the moment of the oxygen shift.
From the results of this study, a mechanistic rationale for the
singlet oxygen reaction with 1,3-diene carboxylates and the
corresponding 1-phenyl-1,3-dienes can be derived. The results
for the singlet oxygen reactivities of the E/Z isomers 1a/1a′ as
well as the double-bond isomers of the ester- and phenyl-
substituted dienes 1h,i and 1k,l, respectively, indicate that the
so-called cis effect leads to a kinetic preference for substrates
with two methyl groups in the cis configuration without
influencing the ene regioselectivity (Scheme 11). Irrespective of
the double-bond configuration, all of the substrates resulted in
high gem or vinylogous gem selectivity. Substrates bearing only
one methyl group at C-3 preferentially gave [4 + 2]
cycloaddition products (93% from 1g, 100% from 1j).
EXPERIMENTAL SECTION
■
General Aspects and Methods. meso-Tetraphenylporphyrin
(TPP) was purchased from Porphyrin Systems (Bremen, Germany).
The solvents for solution photooxygenation were puriss. and used as
1
purchased. H and 13C NMR spectra were recorded at 300, 500, or
1
600 MHz for H and 75.4, 125.7, or 150.9 MHz for 13C. Chemical
shifts are reported in parts per million relative to tetramethylsilane
with the solvent resonances of CDCl3 as the internal standards (center
line of the triplet at 77.0 ppm for 13C and the singlet at 7.24 ppm for
1H). Mass spectra and accurate mass determinations were obtained by
direct-inlet electrospray ionization (EI). Infrared spectra were
recorded with a Fourier transform (FT-IR) spectrometer. Samples
were scanned as neat liquids or dissolved in CH2Cl2. For all of the new
compounds, satisfactory elemental analyses and high-resolution mass
spectra as well as HPLC and/or GC analyses were obtained that
confirmed >95% purity. For the photooxygenation reactions, either a
halogen street lamp (150 W) or white LEDs were used.
Preparation of Compounds 1. General Procedure for Horner−
Wittig Reactions (GP-A). To a suspension of tert-butoxide (2 equiv) in
dry THF (2 mL per mmol of aldehyde) under argon was added slowly
triethyl phosphonoacetate (1 equiv). After the mixture was stirred for
15 min, the aldehyde (1 equiv) was added dropwise to the solution.
The mixture was allowed to stir for 2 h before the reaction was
CONCLUSION
■
The apparently decisive factor for directing the mode selectivity
(i.e., ene versus [4 + 2] reactivity) is the additional (and totally
unreactive) methyl group at the terminal diene carbon,
irrespective of whether it is in the Z or E configuration.
G
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
quenched by the addition of water, after which the mixture was
extracted three times with diethyl ether. The combined extracts were
washed with brine and dried with MgSO4. The solvent was evaporated
under reduced pressure, and the crude product was purified by column
chromatography.
purified by column chromatography [c-hex/EtOAc (3:1), Rf = 0.64] to
yield the product as a colorless oil (4.31 g, 88%). 1H NMR (300 MHz,
CDCl3) δ (ppm) = 1.30 (t, 3H, J = 7.1 Hz), 1.93 (s, 3H), 2.02 (d, 3H,
J = 1.2 Hz), 4.21 (q, 2H, J = 7.1 Hz), 5.07 (s, 1H), 5.21 (s, 1H), 7.11
(q, 1H, J = 1.2 Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) = 13.9
(CH3), 14.3 (CH3), 22.0 (CH3), 60.6 (CH2), 119.7 (CH2), 127.7
(Cq), 140.4 (CH), 140.8 (Cq), 168.8 (Cq).
General Procedure for Wittig Reactions (GP-B). Freshly prepared
1-ethoxycarbonylethylidene triphenylphosphane (1 equiv) was reacted
with the aldehyde (1 equiv) in toluene (1 mL per mmol of aldehyde)
at 100 °C for 24 h. The solvent was removed under reduced pressure,
and the crude product was dissolved in n-pentane, separated from the
solid by filtration, and purified by column chromatography.
(E)-Ethyl 2-Methylpent-2-enoate (1b).32 The crude product,
prepared by GP-B from 13.8 mmol of propionaldehyde, was purified
by column chromatography [c-Hex/EtOAc (20:1), Rf = 0.71] to yield
(2E,4E)-Ethyl 4-Methyl-2,4-hexadienoate (E,E-1h).37 The crude
product, prepared by GP-A from 23.8 mmol of (E)-2-methyl-2-
butenal, was purified by column chromatography [c-hex/EtOAc
(70:1), Rf = 0.55] to yield the product as a colorless oil (3.05 g,
1
83%). H NMR (300 MHz, CDCl3) δ (ppm) = 1.25 (t, 3H, J = 7.1
Hz), 1.72 (s, 3H), 1.76 (d, 3H, J = 7.2 Hz), 4.16 (q, 2H, J = 7.1 Hz),
5.73 (d, 1H, J = 17.5 Hz), 5.93 (q, 1H, J = 7.1 Hz), 7.27 (d, 1H, J =
15.7 Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) = 11.8 (CH3), 14.4
(CH3), 14.6 (CH3), 60.1 (CH2), 115.3 (CH), 133.7 (Cq), 136.31
(CH), 149.5 (CH), 167.7 (Cq).
1
the product (E/Z mixture 96:4) as a colorless oil (1.35 g, 69%). H
NMR (300 MHz, CDCl3) major diastereoisomer δ (ppm) = 1.02 (t,
3H, J = 7.6 Hz), 1.27 (t, 3H, J = 7.1 Hz), 1.80 (d, 3H, J = 1.2 Hz), 2.16
(m, 2H), 4.16 (q, 2H, J = 7.1 Hz), 6.72 (tq, 1H, J1 = 7.4 Hz, J2 = 1.2
Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) = 12.3 (CH3), 13.1
(CH3), 14.4 (CH3), 22.1 (CH2), 60.3 (CH2), 127.4 (CH), 144.1 (Cq),
168.4 (Cq).
(2E,4Z)-Ethyl 4-Methyl-2,4-hexadienoate (E,Z-1h). The crude
product, prepared by GP-A from 3.51 mmol of (Z)-2-methyl-2-
butenal, was purified by column chromatography [c-hex/EtOAc
(70:1), Rf = 0.4] to yield the product as a colorless oil (373 mg,
(E)-Ethyl 2,4-Dimethylpent-2-enoate (1c).32 The crude product,
prepared by GP-B from 5.6 mmol of isobutyraldehyde, was purified by
column chromatography [c-Hex/EtOAc (20:1), Rf = 0.68] to yield the
69%). H NMR (300 MHz, CDCl3) δ (ppm) = 1.28 (t, 3H, J = 7.1
1
Hz), 1.81 (s, 3H), 1.83 (d, 3H), 4.19 (q, 2H, J = 7.1 Hz), 5.78 (q, 1H,
J = 6.5 Hz) 5.84 (d, 1H, J = 15.7 Hz), 7.75, (d, 1H, J = 15.7 Hz); 13C
NMR (75 MHz, CDCl3) δ (ppm) = 13.8 (CH3), 14.4 (CH3), 20.0
(CH3), 60.3 (CH2), 117.9 (CH), 131.6 (Cq), 133.8 (CH), 140.8
(CH), 167.7 (Cq).
1
product (E/Z mixture 97:3) as a yellow oil (610 mg, 70%). H NMR
(300 MHz, CDCl3) major diastereoisomer δ (ppm) = 0.97 (d, 6H, J =
5.7 Hz), 1.23 (t, 3H, J = 7.1 Hz), 1.77 (d, 3H, J = 1.3 Hz), 2.70−2.48
(m, 1H), 4.16 (q, 2H, J = 7.1 Hz), 6.51 (dq, 1H, J1 = 9.7 Hz, J2 = 1.3
Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) = 12.2 (CH3), 14.3
(CH3), 22.3 (CH3), 28.1 (CH), 60.6 (CH2), 125.8 (Cq), 149.0 (CH),
168.5 (Cq).
(E,E)-Ethyl 2,4-Dimethylhexa-2,4-dienoate (E,E-1i).38 The crude
product, prepared by GP-B from 23.8 mmol of (E)-2-methyl-2-
butenal, was purified by column chromatography [c-hex/EtOAc
(10:1), Rf = 0.65] to yield the product as a colorless oil (2.20 g,
(E)-Ethyl 2,4,4-Trimethylpent-2-enoate (1d).33 To a precooled (0
°C) suspension of sodium hydride (60 wt % in mineral oil, 520 mg,
13.0 mmol) in ethyl propionate (40 mmol) was added ethanol (0.7
mmol). Afterward, pivaldehyde (10 mmol, 1.1 mL) was added in
portions over 30 min. After 30 min of stirring at 0 °C, the mixture was
diluted with hexane, and the reaction was quenched by slow addition
of saturated sodium bicarbonate, after which the mixture was further
diluted with water. The layers were separated, and the aqueous phase
was extracted three times with hexane. The combined extracts were
dried with MgSO4. Hexane was evaporated under reduced pressure,
and the crude product was purified by column chromatography [c-
hex/EtOAc (8:1), Rf = 0.61] to yield the product as a colorless oil
(1.27 g, 75%). 1H NMR (300 MHz, CDCl3) δ (ppm) = 1.15 (s, 9H),
1.27 (t, 3H, J = 7.1 Hz), 1.93 (d, 3H, J = 1.3 Hz), 4.15 (q, 2H, J = 7.1
Hz), 6.77 (q, 1H, J = 1.4 Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) =
13.4 (CH3), 14.4 (CH3), 30.2 (CH3), 33.0 (C), 60.7 (CH2), 126.8
(Cq), 151.1 (CH), 169.4 (Cq).
55%). H NMR (300 MHz, CDCl3) δ (ppm) = 1.28 (t, 3H, J = 7.1
1
Hz), 1.72 (d, 3H, J = 6.9 Hz), 1.81 (s, 3H), 1.98 (s, 3H), 4.18 (q, 2H, J
= 7.1 Hz), 5.69 (q, 1H, J = 7.0 Hz), 7.10 (s, 1H); 13C NMR (75 MHz,
CDCl3) δ (ppm) = 14.0 (CH3), 14.1 (CH3), 14.4 (CH3), 16.0 (CH3),
60.6 (CH3), 125.0 (Cq), 130.9 (CH), 133.2 (Cq), 143.0 (CH), 169.3
(Cq).
(E,Z)-Ethyl 2,4-Dimethylhexa-2,4-dienoate (E,Z-1i). The crude
product, prepared by GP-B from 14.7 mmol of (Z)-2-methyl-2-
butenal, was purified by column chromatography [c-hex/EtOAc
(20:1), Rf = 0.72] to yield the product as a colorless oil (373 mg,
1
69%). H NMR (300 MHz, CDCl3) δ (ppm) = 1.31 (t, 3H, J = 7.1
Hz), 1.54 (d, 3H), 1.81 (s, 3H), 1.82 (s, 3H), 4.47 (q, 2H, J = 7.1 Hz),
5.47 (q, 1H, J = 6.9 Hz), 7.13 (s, 1H); 13C NMR (75 MHz, CDCl3) δ
(ppm) = 14.1 (CH3), 14.4 (CH3), 15.1 (CH3), 16.0 (CH3), 60.7
(CH3), 125.4 (CH), 128.1 (Cq), 132.2 (Cq), 138.8 (CH), 168.5 (Cq).
(E)-2-Methyl-4-phenyl-1,3-butadiene (1j).39 The crude product,
prepared by GP-A from 14.3 mmol of acrolein, was purified by column
chromatography [c-hex/EtOAc (100:1), Rf = 0.45] to yield the
product as a colorless oil (1.75 g, 85%). 1H NMR (300 MHz, CDCl3)
δ (ppm) = 2.00 (s, 3H), 5.12 (s, 2H), 6.56 (d, 1H, J = 16.4 Hz), 6.91
(d, 1H, J = 16.4 Hz), 7.25 (t, 1H, J = 7.3 Hz), 7.3 (t, 2H, J = 7.3 Hz),
7.56 (d, 2H, J = 7.8 Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) = 18.7
(CH3), 117.5 (CH2), 126.6 (CH), 127.6 (CH), 126.7 (CH), 128.8
(CH), 131.9 (CH), 134.0 (Cq), 145.9 (Cq).
(E)-Ethyl 2-Methyl-2,4-pentadienoate (1e).34 The crude product,
prepared by GP-B from 38 mmol of acrolein, was purified by column
chromatography [c-hex/EtOAc (3:1), Rf = 0.70] to yield the product
1
(E/Z mixture 96:4) as a colorless oil (4.25 g, 80%). H NMR (300
MHz, CDCl3) major diastereoisomer δ (ppm) = 1.31 (t, 3H, J = 7.1
Hz), 1.96 (d, 3H, J = 0.6 Hz), 4.22 (q, 2H, J = 7.1 Hz), 5.45 (d, 1H, J
= 10.0 Hz), 5.56 (d, 1H, J = 16.8 Hz), 6.66 (m, 1H), 7.17 (dd, 1H, J1 =
11.4 Hz, J2 = 0.6 Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) = 12.7
(CH3), 14.4 (CH3), 60.7 (CH2), 124.1 (CH2), 128.3 (Cq), 132.3
(CH), 138.3 (CH), 168.4 (Cq).
(1E,3E)-3-Methyl-1-phenyl-1,3-pentadiene (1k).40 The crude
product, prepared by GP-A from 23.8 mmol of (E)-2-methyl-2-
butenal, was purified by column chromatography [c-hex/EtOAc
(20:1), Rf = 0.53] to yield the product as a colorless oil (2.82 g,
(2E,4E)-Ethyl 2-Methyl-2,4-hexadienoate (1f).35 The crude prod-
uct, prepared by GP-B from 38 mmol of crotonaldehyde, was purified
by column chromatography [c-hex/EtOAc (3:1), Rf = 0.61] to yield
1
75%). H NMR (300 MHz, CDCl3) δ (ppm) = 1.80 (d, 3H, J = 7.0
Hz), 1.87 (s, 3H), 5.72 (q, 1H, J = 7.0 Hz), 6.45 (d, 1H, J = 16.1 Hz),
6.82 (d, 1H, J = 16.1 Hz), 7.15−7.44 (m, 5H); 13C NMR (75 MHz,
CDCl3) δ (ppm) = 12.3 (CH3), 14.3 (CH3), 125.5 (CH), 126.4 (Cq),
127.0 (CH), 128.4 (CH), 128.8 (CH), 134.2 (CH), 134.9 (Cq), 138.3
(Cq).
1
the product (E/Z mixture 93:7) as a colorless oil (4.11 g, 70%). H
NMR (300 MHz, CDCl3) δ (ppm) = 1.27 (t, 3H, J = 7.1 Hz), 1.84 (d,
3H, J = 6.8 Hz), 1.89 (s, 3H), 4.17 (q, 2H, J = 7.1 Hz), 6.05 (dq, 1H, J1
= 14.4 Hz, J2 = 6.8 Hz), 6.33 (ddd, 1H, J1 = 14.4 Hz, J2 = 11.3 Hz, J3 =
1.5 Hz), 7.12 (d, 1H, J = 11.2 Hz); 13C NMR (75 MHz, CDCl3) δ
(ppm) = 12.7 (CH3), 14.4 (CH3), 18.8 (CH3), 60.5 (CH2), 125.0
(Cq), 127.5 (CH), 137.0 (CH), 138.5 (CH), 168.7 (Cq).
(2E,4E)-4-Methyl-2-phenyl-2,4-hexadiene (1l).40 The crude prod-
uct, prepared by GP-B from 23.8 mmol of (E)-2-methyl-2-butenal, was
purified by column chromatography [c-hex/EtOAc (20:1), Rf = 0.55]
(E)-Ethyl 2,4-Dimethyl-2,4-pentadienoate (1g).36 The crude
product, prepared by GP-B from 32 mmol of methacrolein, was
to yield the product as a colorless oil (2.74 g, 73%). H NMR (300
1
MHz, CDCl3) δ (ppm) = 1.83 (d, 3H, J = 6.9 Hz), 1.91 (s, 3H), 2.27
H
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(s, 3H), 5.72 (q, 1H, J = 6.9 Hz), 6.32 (s, 1H), 7.26−7.54 (m, 5H);
13C NMR (75 MHz, CDCl3) δ (ppm) = 13.9 (CH3), 16.9 (CH3), 17.6
(CH3), 125.0 (CH), 126.0 (CH), 126.7 (CH), 128.3 (CH), 132.0
(CH), 134.0 (Cq), 134.0 (Cq), 144.7 (Cq).
(E)-Ethyl 3-Hydroxy-2-methylenehex-4-enoate (16f). The crude
product, prepared by GP-C and GP-D from 3.25 mmol of 1f, was
purified by column chromatography [c-hex/EtOAc (3:1), Rf = 0.56] to
1
yield the product as a yellow oil (90 mg, 16%). H NMR (300 MHz,
CDCl3) δ (ppm) = 1.29 (t, 3H, J = 7.1 Hz), 1.69 (d, 3H, J = 6.4 Hz),
2.73 (s, 1H), 4.21 (q, 2H, J = 7.1 Hz), 4.87 (d, 1H, J = 6.5 Hz), 5.56
(dd, 1H, J1 = 15.3 Hz, J2 = 6.5 Hz), 5.77−5.65 (m, 1H), 5.79 (s, 1H),
6.20 (s, 1H); 13C NMR (75 MHz, CDCl3) δ (ppm) = 13.7 (CH3),
17.7 (CH3), 60.9 (CH2), 72.1 (CH), 124.9 (CH2), 128.0 (CH), 130.7
(CH), 141.9 (Cq), 166.5 (Cq); IR (film) ν (cm−1) = 3427, 2922, 2363,
1707, 1696, 1636.
Photooxygenation, Reduction, and Oxygen Transfer. Gen-
eral Procedure for Photooxygenations (GP-C). A 3 mL aliquot of a
CDCl3 stock solution, 2 × 10−4 M in the photosensitizer TPP and 0.2
M in the corresponding diene, was added to a Schlenk tube equipped
with a magnetic stirrer. During irradiation with a 50 W white LED
lamp at room temperature, oxygen was bubbled through the solution.
The reaction was controlled by TLC and NMR spectroscopy. After
complete conversion, the solvent was evaporated under reduced
pressure at 5 °C. The conversion and isomeric ratio were determined
from the crude reaction mixture.
Ethyl 3,6-Dimethyl-3,6-dihydro-1,2-dioxine-3-carboxylate (4).
The crude product, prepared by GP-C and GP-D from 3.25 mmol
of 1f, was purified by column chromatography [c-hex/EtOAc (3:1), Rf
1
General Procedure for Reductions (GP-D). To the hydroperoxide-
containing solution from GP-C, dimethyl sulfide (5 equiv with respect
to the calculated substrate amount in the photooxygenation reaction)
was added at 0 °C. The mixture was allowed to stir for 12 h at room
temperature in the dark. The solvent was removed under reduced
pressure, and the crude product was purified by column chromatog-
raphy. Compounds 2b−d from the kinetic studies were characterized
by NMR analyses only.
= 0.64] to yield the product as a yellow oil (55 mg, 9%). H NMR
(300 MHz, CDCl3) δ (ppm) = 1.13 (d, 3H, J = 6.8 Hz), 1.27 (t, 3H, J
= 7.1 Hz), 1.38 (s, 3H), 4.22 (q, 2H, J = 7.1 Hz), 4.78 (m, 1H, J = 6.8
Hz), 5.86 (dd, 1H, J1 = 10.2 Hz, J2 = 0.6 Hz), 6.03 (dd, 1H, J1 = 10.1
Hz, J2 = 2.0 Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) = 14.2, 17.1,
22.0, 60.1, 73.8, 80.1, 127.4, 129.9, 171.8; IR (film) ν (cm−1) = 3018,
2358, 1726, 1683, 1445, 1369; HRMS (EI) calcd for C9H14O4 m/z
141.0552 [M − C2H5O]+, found 141.0550.
Ethyl 3-Hydroxy-4-methyl-2-methylenepent-4-enoate (16g). The
crude product, prepared by GP-C and GP-D from 6.49 mmol of 1g,
was purified by column chromatography [c-hex/EtOAc (3:1), Rf =
0.53] to yield the product as a yellow oil (76 mg, 7%). 1H NMR (300
MHz, CDCl3) δ (ppm) = 1.30 (t, 3H, J = 7.1 Hz), 1.72 (s, 3H), 2.75
(s, 1H), 4.22 (q, 2H, J = 7.1 Hz), 4.90 (s, 1H), 4.97 (s, 1H), 5.08 (s,
1H), 5.83 (s, 1H), 6.30 (s, 1H); 13C NMR (75 MHz, CDCl3) δ (ppm)
= 14.3 (CH3), 19.0 (CH3), 61.1 (CH2), 74.8 (CH), 112.6 (CH2),
126.1 (CH2), 140.6 (Cq), 144.7 (Cq), 166.7 (Cq); IR (film) ν (cm−1)
= 3445, 3078, 2979, 2936, 1910, 1710, 1627; HRMS (EI) calcd for
C8H13O2 m/z 141.091 [M − CH2O]+, found 141.091.
Ethyl 3,5-Dimethyl-3,6-dihydro-1,2-dioxine-3-carboxylate (5).
The crude product, prepared by GP-C and GP-D from 6.49 mmol
of 1g, was purified by column chromatography [c-hex/EtOAc (3:1), Rf
= 0.70] to yield the product as a white solid (354 mg, 37%). 1H NMR
(300 MHz, CDCl3) δ (ppm) = 1.28 (t, 3H, J = 7.1 Hz), 1.39 (s, 3H),
1.74 (s, 3H), 4.09 (d, 1H, J = 16.0 Hz), 4.31−4.14 (m, 2H), 4.62 (d,
1H, J = 16.0 Hz), 5.77 (m, 1H); 13C NMR (75 MHz, CDCl3) δ (ppm)
= 14.2 (CH3), 18.0 (CH3), 22.4 (CH3), 61.5 (CH2), 72.4 (CH2), 80.4
(Cq), 121.6 (CH), 132.7 (Cq), 171.9 (Cq); IR (film) ν (cm−1) = 2988,
1715, 1473, 1447, 1368; HRMS (EI) calcd for C9H14O4 m/z 141.0552
[M − C2H5O]+, found 141.056. Anal. Calcd for C9H14O4: C, 58.05; H,
7.58. Found: C, 57.93; H, 7.57.
(E)-Ethyl 5-Hydroperoxy-4-methylenehex-2-enoate (7). Photo-
oxygenation of 180 mg (1.12 mmol) of 1h for 3.5 h according to
GP-C led to the exclusive formation of hydroperoxide 7 (characterized
by NMR and IR only). 1H NMR (300 MHz, CDCl3) δ (ppm) = 1.31
(t, 3H, J = 7.2 Hz), 1.37 (d, 3H, J = 6.6 Hz), 4.22 (q, 2H, J = 7.1 Hz),
4.82 (q, 1H, J = 6.5 Hz), 5.59 (2H, d, J = 4.6 Hz), 6.13 (d, 1H, J = 16.1
Hz), 7.30 (d, 1H, J = 16.1 Hz), 8.70 (s, 1H); 13C NMR (75 MHz,
CDCl3) δ (ppm) = 14.5 (CH3), 18.5 (CH3), 60.9 (CH2), 81.2 (CH),
119.5 (CH), 123.1 (CH2), 143.5 (CH), 144.6 (Cq), 167.3 (Cq); IR
(film) ν (cm−1) = 3284 (s), 2983 (m), 2935 (m), 2841 (s), 1702 (s),
1618 (s).
Ethyl 3-Hydroperoxy-2-methylenebutanoate (2a).21 1H NMR
(300 MHz, CDCl3) δ (ppm) = 0.89 (t, 3H, J = 7.4 Hz), 1.26 (t, 3H, J
= 7.1 Hz), 1.81−1.46 (m, 2H), 4.18 (q, 2H, J = 7.1 Hz), 4.92 (t, 1H, J
= 6.6 Hz), 5.86 (s, 1H), 6.34 (s, 1H), 9.10 (s, 1H); 13C NMR (75
MHz, CDCl3) δ (ppm) = 14.0 (CH3), 18.5 (CH3), 61.0 (CH2), 79.2
(CH), 125.7 (CH2), 140.6 (Cq), 166.4 (Cq).
Ethyl 3-Hydroperoxy-2-methylenepentanoate (2b). 1H NMR
(300 MHz, CDCl3) δ (ppm) = 1.26 (t, 3H, J = 7.1 Hz), 1.29 (d,
3H, J = 6.6 Hz), 4.18 (q, 2H, J = 7.1 Hz), 4.92 (q, 1H, 3JH,H = 6.6 Hz),
5.88 (s, 1H), 6.30 (s, 1H), 9.16 (s, 1H); 13C NMR (75 MHz, CDCl3)
δ (ppm) = 9.8 (CH3), 14.1 (CH3), 26.0 (CH2), 61.4 (CH2), 84.6
(CH), 126.1 (CH2), 139.6 (Cq), 166.5 (Cq).
1
Ethyl 3-Hydroperoxy-4-methyl-2-methylenepentanoate (2c). H
NMR (300 MHz, CDCl3) δ (ppm) = 0.93 (d, 3H, J = 3.2 Hz), 0.95 (d,
3H, J = 3.1 Hz), 1.26 (t, 3H, J = 7.1 Hz), 2.01−1.89 (m, 1H), 4.18 (q,
2H, J = 7.1 Hz), 4.68 (d, 1H, J = 6.6 Hz), 5.87 (s, 1H), 6.42 (s, 1H),
8.48 (s, 1H); 13C NMR (75 MHz, CDCl3) δ (ppm) = 14.1 (CH3),
17.9 (CH3), 19.1 (CH3), 31.4 (CH), 60.2 (CH2), 84.6 (CH), 126.4
(CH2), 139.6 (Cq), 166.7 (Cq).
Ethyl 3-Hydroperoxy-4,4-dimethyl-2-methylenepentanoate (2d).
1H NMR (300 MHz, CDCl3) δ (ppm) = 0.89 (s, 9H), 1.25 (t, 3H, J =
6.8 Hz), 4.21 (q, 2H, J = 7.1 Hz), 4.82 (s, 1H), 5.84 (s, 1H), 6.43 (s,
1H), 8.55 (s, 1H); 13C NMR (75 MHz, CDCl3) δ (ppm) = 14.1
(CH3), 25.9 (CH3), 35.4 (Cq), 61.5 (CH2), 89.4 (CH), 124.8 (CH2),
138.8 (Cq), 167.4 (Cq).
Ethyl 3-Hydroxy-2-methylenepent-4-enoate (16e). The crude
product, prepared by GP-C and GP-D from 5.71 mmol of 1e, was
purified by column chromatography [c-hex/EtOAc (3:1), Rf = 0.48] to
yield the alcohol as a yellow oil (120 mg, 14%) and the endoperoxide
1
[c-hex/EtOAc (3:1), Rf = 0.61] as a colorless oil (96 mg, 10%). H
NMR (300 MHz, CDCl3) δ (ppm) = 1.24 (t, 3H, J = 7.1 Hz), 3.03 (s,
1H), 4.17 (q, 2H, J = 7.1 Hz), 4.89 (d, 1H, J = 4.8 Hz), 5.12 (d, 1H, J
= 10.5 Hz), 5.27 (d, 1H, J = 17.2 Hz), 5.78 (s, 1H), 5.89 (m, 1H), 6.19
(s, 1H); 13C NMR (75 MHz, CDCl3) δ (ppm) = 13.8 (CH3), 60.5
(CH2), 71.8 (CH), 115.6 (CH2), 125.1 (CH2), 137.9 (CH), 141.1
(Cq), 170.8 (Cq); IR (film) ν (cm−1) = 3427, 2922, 2363, 1707, 1696,
1636; HRMS (EI) calcd for C7H11O2 m/z 127.075 [M − CH2O]+,
found 127.074.
(E)-Ethyl 5-Hydroperoxy-2-methyl-4-methylenehex-2-enoate (9).
Photooxygenation of 200 mg (1.19 mmol) of 1i for 3 h according to
GP-C led to the exclusive formation of hydroperoxide 9 (characterized
by NMR and IR only). 1H NMR (300 MHz, CDCl3) δ (ppm) = 1.25
(d, 3H, J = 6.6 Hz), 1.29 (t, 3H, J = 7.1 Hz), 2.00 (s, 3H), 4.21 (q, 2H,
J = 7.0 Hz), 4.82 (q, 1H, J = 6.6 Hz), 5.3 (1H, s), 5.51 (1H, s), 7.12 (t,
1H, J = 1.2 Hz), 8.42 (s, 1H). 13C NMR (150 MHz, CDCl3) δ (ppm)
= 14.3 (CH3), 14.3 (CH3), 17.9 (CH3), 61.1 (CH2), 84.3 (CH), 118.7
(CH2), 131.1 (Cq) 135.4 (CH), 144.4 (Cq), 168.5 (Cq); IR ν (cm−1) =
3222, 2972, 2875, 2361, 1707, 1621.
Ethyl 3-Methyl-3,6-Dihydro-1,2-dioxine-3-carboxylate (3). 1H
NMR (300 MHz, CDCl3) δ (ppm) = 1.27 (t, 3H, J = 7.1 Hz), 1.39
(s, 3H), 4.21 (q, 2H, J = 7.1 Hz), 4.29−4.23 (ddd, 1H, J1 = 16.6 Hz, J2
= 3.8 Hz, J3 = 1.2 Hz), 4.75 (dt, 1H, J1 = 16.6 Hz, J2 = 1.2 Hz), 6.01
(ddd, 1H, J1 = 10.2 Hz, J2 = 3.8 Hz, J3 = 1.2 Hz), 6.09 (dt, 1H, J1 =
10.2 Hz, J2 = 1.6 Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) = 13.7
(CH3), 22.0 (CH3), 61.6 (CH2), 69.3 (CH2), 80.6 (Cq), 124.7 (CH),
127.7 (CH), 171.6 (Cq); IR (film) ν (cm−1) = 3564, 2986, 2358, 1727,
1699, 1683, 1538, 1445, 1371. Anal. Calcd for C8H12O4: C, 55.81; H,
7.02. Found: C, 55.57; H, 7.03.
3,6-Dihydro-5-methyl-3-phenyl-1,2-dioxine (11). Photooxygena-
tion of 100 mg (0.69 mmol) of 1j for 2 h according to GP-C led to the
formation of 120 mg (98%) of endoperoxide 11. 1H NMR (300 MHz,
CDCl3) δ (ppm) = 1.83 (s, 3H), 4.42 (d, 1H, J = 15.8 Hz), 4.57 (d,
I
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
1H, J = 15.8 Hz), 5.59 (s, 1H), 5.80 (s, 1H), 7.35−7.42 (m, 5H); 13
C
(ppm) = 14.1/14.1 (CH3), 16.7/17.0 (CH3), 21.9/22.1 (CH3), 61.8/
61.8 (CH2), 68.8/68.8 (CH2), 80.3/80.3 (Cq), 80.8/81.3 (CH),
123.7/124.6 (CH), 136.8/136.9 (Cq), 171.9/171.9 (Cq); IR (film) ν
NMR (75 MHz, CDCl3) δ (ppm) = 18.38 (CH3), 73.13 (CH2), 80.52
(CH), 120.90 (CH), 128.7 (CH), 128.8 (CH), 128.9 (Cq), 132.6
(Cq), 137.9 (Cq); HRMS (EI) calcd for C11H12O2 m/z 176.0837 [M]+,
found 176.0839.
1
(cm ) = 3440, 2988, 2358, 2330, 1732, 1716, 1681, 1635.
Ethyl 5-(1-Hydroxyethyl)-3,6-dihydro-1,2-dioxine-3-carboxylate
(21). The crude product, prepared by GP-D from 150 mg of 19,
was purified by column chromatography [c-hex/EtOAc (2:1), Rf =
3,6-Dihydro-5-(1-hydroperoxyethyl)-3-phenyl-1,2-dioxine (12).
Photooxygenation of 100 mg (0.50 mmol) of 1k for 12 h according
to GP-C led to the formation of 105 mg (95%) of hydro-
peroxyendoperoxide 12 (dr = 50:50) (characterized by NMR and
1
0.45] to yield the product as a colorless oil (107 mg, 70%). H NMR
(600 MHz, CDCl3) diastereoisomers A/B δ (ppm) = 1.31−1.34 m,
(3H/3H), 1.36 (d, 2H, J = 6.5 Hz)/1.36 (d, 2H, J = 6.5 Hz), 4.25−
4.32 (m, 2H/2H), 4.38 (q, 1H, J = 6.5 Hz)/4.44 (q, 1H, J = 6.5 Hz),
4.48 (d, 1H, J = 16.4 Hz/4.54 (d, 1H, J = 16.6 Hz), 4.77 (d, 1H, J =
16.4 Hz)/4.83 (d, 1H, J = 16.6 Hz), 4.97 (s, 1H)/5.00 (s, 1H), 6.06−
6.08 (m, 1H)/6.08−6.10 (m, 1H); 13C NMR (150 MHz, CDCl3)
diastereoisomers A/B δ (ppm) = 14.3/14.3 (CH3), 21.7/21.9 (CH3),
61.9/61.9 (CH2), 67.9/68.3 (CH), 69.3/69.6 (CH2), 76.6/76.7 (CH),
115.2/115.8 (CH), 141.7/141.8 (Cq); IR (film) ν (cm−1) = 3415,
2988, 2940, 2333, 2330, 2299, 1744, 1700, 1688; HRMS (EI) calcd for
C9H14O5 m/z 202.0832 [M]+, found 202.0831.
1
IR only). H NMR (600 MHz, CDCl3) diastereomeres A/B δ (ppm)
= 1.37 (d, 3H, J = 6.7 Hz)/1.45 (d, 3H, J = 6.7 Hz), 4.56−4.91 (m,
3H/3H), 5.58 (s, 1H)/5.69 (s, 1H), 6.06 (s, 1H)/6.10 (s, 1H), 7.32−
7.41 (m, 5H/5H), 8.37 (s, 1H/1H); 13C NMR (150 MHz, CDCl3) δ
(ppm) = 16.9/17.0 (CH3), 69.3/69.3 (CH2), 80.2/80.4 (CH), 81.6/
81.9 (CH), 124.0/124.6 (CH), 128.4−129.3 (Caryl); IR ν (cm−1) =
3418, 2982, 2893, 2328, 1636.
3,6-Dihydro-5-(1-hydroperoxyethyl)-3-methyl-3-phenyl-1,2-diox-
ine (14). Photooxygenation of 100 mg (0.50 mmol) of 1l for 10 h
according to GP-C resulted in the formation of hydroperoxyendoper-
oxide 14 (105 mg, 90%; dr = 68:32) (characterized by NMR and IR
only). 1H NMR (600 MHz, CDCl3) diastereoisomers A/B δ (ppm) =
1.33−1.41 (m, 6H), 1.57 (s, 3H)/ 1.58 (s, 3H), 4.42−4.81 (m, 3H/
3H), 6.27 (s, 1H)/6.33 (s, 1H), 7.23−7.53 (m, 5H/5H), 8.28 (sb, 1H/
1H); 13C NMR (150 MHz, CDCl3) δ (ppm) = 16.7/17.0 (CH3),
26.0/26.3 (CH3), 68.7/68.9 (CH2), 80.8/81.0 (Cq), 81.5/81.9 (CH),
125.5/125.9 (CH), 127.6/127.7 (CH), 128.2/128.4 (CH), 134.5/
134.6 (Cq); IR ν (cm−1) = 3888, 3000, 2912, 1620.
(E)-Ethyl 5-Hydroxy-4-methylenehex-2-enoate (17). The crude
product, prepared by GP-D from 180 mg of 7, was purified by column
chromatography [c-hex/EtOAc (2:1), Rf = 0.28] to yield the product
as a colorless oil (86 mg, 45%). 1H NMR (300 MHz, CDCl3) δ (ppm)
= 1.27 (t, 3H, J = 7.2 Hz), 1.34 (d, 3H, J = 6.5 Hz), 2.37 (s, 1H,), 4.18
(q, 2H, J = 7.1 Hz), 4.56 (q, 1H, J = 6.4 Hz), 5.42 (1H, s), 5.60 (1H,
s), 6.01 (d, 1H, J = 16.2 Hz), 7.27 (d, 1H, J = 16.2 Hz); 13C NMR (75
MHz, CDCl3) δ (ppm) = 14.4 (CH3), 23.1 (CH3), 60.7 (CH2), 67.2
(CH), 118.9 (CH), 120.9 (CH2), 144.1 (CH), 148.7 (Cq), 167.3 (Cq);
IR (film) ν (cm−1) = 3430, 2981, 2929, 2841, 1688, 1626; HRMS (EI)
calcd for C9H14O3 m/z 170.0942 [M]+, found 170.0926.
Ethyl 5-(1-Hydroxyethyl)-3-methyl-3,6-dihydro-1,2-dioxine-3-car-
boxylate (22). The crude product, prepared by GP-D from 200 mg of
20, was purified by column chromatography [c-hex/EtOAc (2:1), Rf =
1
0.50] to yield the product as a colorless oil (180 mg, 75%). H NMR
(600 MHz, CDCl3) diastereoisomers A/B δ (ppm) = 1.27 (t, 3H/3H,
J = 7.1 Hz), 1.30 (d, 3H, J = 3.8 Hz)/1.32 (d, 3H, J = 3.8 Hz), 1.39 (s,
3H)/1.40 (s, 3H), 2.35 (s, 1H/1H), 4.15−4.26 (m, 2H/2H), 4.31 (q,
1H, J = 6.6 Hz)/4.38 (q, 1H, J = 6.6 Hz), 4.32 (d, 1H, J = 14.8 Hz)/
4.38 (d, 1H, J = 14.8 Hz), 4.71 (d, 1H, J = 16.3 Hz)/4.78 (d, 1H, J =
16.3 Hz), 5.95 (s, 1H)/5.95 (s, 1H); 13C NMR (150 MHz, CDCl3)
diastereoisomers A/B δ (ppm) = 14.2/14.2 (CH3), 21.6/21.9 (CH3),
22.1/22.2 (CH3), 61.7/61.7 (CH2), 67.5/68.3 (CH), 68.8/69.2
(CH2), 80.3/80.4 (Cq), 120.4/121.5 (CH), 140.4/140.5 (Cq),
171.9/171.9 (Cq); IR (film) ν (cm−1) = 3420 (sb), 2982 (w), 2936
(w), 2905 (w), 2353 (w), 2328 (w), 1732 (s), 1715 (s), 1645 (m);
HRMS (ESI) calcd for C10H16O5 m/z 239.0889 [M + Na]+, found
239.0887.
1-(3,6-Dihydro-6-phenyl-1,2-dioxin-4-yl)ethanol (23). The crude
product, prepared by GP-D from the reduction of 200 mg (1.00
mmol) of hydroperoxide 12, was purified by column chromatography
[c-hex/EtOAc (3:1), Rf = 0.3] to yield the product as a colorless oil
(89 mg, 43%; dr = 68:32). 1H NMR (500 MHz; CDCl3)
diastereoisomers A/B δ (ppm) = 1.36 (d, 3H, J = 6.5 Hz)/1.42 (d,
3H, J = 6.7 Hz), 2.34 (s, 1H/1H), 4.33−4.45 (m, 1H/1H), 4.58−4.89
(m, 2H/2H), 5.56−5.63 (m, 1H/1H), 5.94−6.02 (m, 1H), 7.33−7.38
(m, 5H/5H); 13C NMR (125 MHz; CDCl3) diastereoisomers A/B δ
(ppm) = 22.0/22.0 (CH3), 68.0/68.4 (CH), 69.4/69.6 (CH2), 80.2/
80.3 (CH), 120.1/120.6 (CH), 128.4/128.5 (CH), 128.6/128.7 (CH),
128.9/130.0 (CH), 129.0/129.0 (CH), 137.2/137.4 (Cq), 140.2/140.2
(Cq); IR (film) ν (cm−1) = 3404, 3086, 3062, 3032, 2972, 2929, 2887,
2362, 2343, 1701, 1653, 1618; HRMS (ESI) calcd for C12H14O3 m/z
229.0835 [M + Na]+, found 229.0837.
(E)-Ethyl 5-Hydroxy-2-methyl-4-methylenehex-2-enoate (18).
The crude product, prepared by GP-D from 226 mg of 9, was
purified by column chromatography [c-hex/EtOAc (2:1), Rf = 0.33] to
1
yield the product as a colorless oil (117 mg, 49%). H NMR (300
MHz, CDCl3) δ (ppm) = 1.22 (d, 3H, J = 6.5 Hz), 1.25 (t, 3H, J = 7.1
Hz), 1.93 (d, 3H, J = 1.5 Hz), 2.71 (s, 1H), 4.15 (q, 2H, J = 7.1 Hz),
4.31 (q, 1H, J = 6.4 Hz), 5.02 (1H, s), 5.43 (1H, s), 7.04 (t, 1H, J = 1.2
Hz); 13C NMR (75 MHz, CDCl3) δ (ppm) = 14.3 (CH3), 14.3
(CH3), 22.5 (CH3), 60.9 (CH2), 70.6 (CH), 115.1 (CH2), 130.4 (Cq),
136.4 (CH), 148.4 (Cq), 168.4 (Cq); IR (film) ν (cm−1) = 3427, 2980,
2933, 2875, 2358, 2341, 1708, 1693, 1631; HRMS (EI) calcd for
C10H16O3 m/z 184.1091 [M]+, found 184.1095.
Ethyl 5-(1-Hydroperoxyethyl)-3,6-dihydro-1,2-dioxine-3-carboxy-
late (19). Photooxygenation of 100 mg (0.46 mmol) of 1i for 36 h
according to GP-C led to the exclusive formation of hydro-
1-(5-Phenylfuran-3-yl)ethanol (24). To a solution of compound 23
(200 mg, 1 mmol) in MeOH (5 mL) was added thiourea (2 mmol),
and the mixture was stirred for 48 h. The solvent was evaporated
under reduced pressure, and the crude product was purified by column
chromatography [c-hex/EtOAc (2:1), Rf = 0.25] to yield the product
1
peroxyendoperoxide 19 (86 mg, 75%; dr = 50:50). H NMR (300
MHz, CDCl3) diastereoisomers A/B δ (ppm) = 1.24−1.37 (m, 6H/
6H), 4.13−5.08 (m, 5H/5H, CH2−CH3, CH2−O−O, CH−OOH),
6.12 (s, 1H/1H); 13C NMR (75 MHz, CDCl3) diastereoisomers A/B
δ (ppm) = 14.1/14.2 (CH3), 16.2/16.2 (CH3), 61.8/61.9 (CH2),
69.1/69.2 (CH2), 76.4/76.5 (CH), 81.0/81.2 (CH), 118.5/118.8
(CH), 138.2/138.3 (Cq), 168.4/168.6 (Cq); IR (film) ν (cm−1) =
3429, 2978, 2349, 1731, 1720, 1672.
Ethyl 5-(1-Hydroperoxyethyl)-3-methyl-3,6-dihydro-1,2-dioxine-
3-carboxylate (20). Photooxygenation of 100 mg (0.61 mmol) of 1i
for 26 h according to GP-C led to the exclusive formation of 112 mg of
hydroperoxyendoperoxide 20 (dr = 50:50). 1H NMR (600 MHz,
CDCl3) diastereoisomers A/B δ (ppm) = 1.22−1.30 (m, 6H/6H),
1.37 (s, 3H)/1.40 (s, 3H), 4.14−4.21 (m, 2H/2H), 4.35 (d, 1H, J =
16.2 Hz)/4.36 (d, 1H, J = 16.3 Hz), 4.51−4.56 (m, 1H/1H), 4.70−
4.78 (m, 1H/1H), 5.99 (s, 1H)/6.00 (s, 1H), 9.04 (br s, 1H)/9.07 (br
s, 1H, OOH); 13C NMR (150 MHz, CDCl3) diastereoisomers A/B δ
1
as a yellowish oil (116 mg, 62%). H NMR (300 MHz, CDCl3) δ
(ppm) = 1.54 (d, 3H, J = 6.4 Hz), 1.99 (s, 1H), 4.90 (q, 1H, J = 6.4
Hz), 6.71 (s, 1H), 7.28 (t, 1H, J = 7.0 Hz), 7.40 (t, 2H, J = 7.0 Hz),
7.42 (s, 1H), 7.68 (d, 2H, J = 7.0 Hz); 13C NMR (75 MHz, CDCl3) δ
(ppm) = 24.1 (CH3), 63.2 (CH), 103.9 (CH), 123.9 (Cq), 127.6
(CH), 128.8 (CH), 130.8 (Cq), 132.5 (Cq), 138.0 (CH); IR (film) ν
(cm−1) = 3398, 2999, 2897, 1761; HRMS (EI) calcd for C12H12O2 m/
z 188.0842 [M]+, found 188.0847.
Ethyl 4-(1-Hydroxyethyl)furan-2-carboxylate (25). To an ice-
cooled solution of compound 21 (190 mg, 1 mmol) in MeOH (5 mL)
was added thiourea (2 mmol), and the mixture was stirred for 24 h.
The solvent was evaporated under reduced pressure, and the crude
product was purified by column chromatography [c-hex/EtOAc (1:2),
Rf = 0.4] to yield the product as a colorless oil (62 mg, 34%). 1H NMR
J
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(300 MHz, CDCl3) δ (ppm) = 1.31 (t, 3H, J = 7.1 Hz), 1.43 (d, 3H, J
= 6.4 Hz), 2.17 (s, 1H), 4.29 (q, 2H, J = 7.1 Hz), 4.81 (q, 1H, J = 6.4
Hz), 7.11 (s, 1H), 7.45 (s, 1H); 13C NMR (75 MHz, CDCl3) δ (ppm)
= 14.3 (CH3), 24.1 (CH3), 61.1 (CH2), 62.7 (CH), 116.4 (CH), 132.5
(Cq), 142.1 (Cq), 142.2 (CH), 158.8 (Cq); IR (film) ν (cm−1) = 3390,
2998, 2970, 1782, 1520, 1454; HRMS (EI) calcd for C9H12O4 m/z
184.0732 [M]+, found 184.0735.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Deutsche Forschungsgemeinschaft (DFG)
(Project Gr-881-13/1) for financial support.
(E)-Ethyl 3-(2-(1-Hydroxyethyl)oxiran-2-yl)acrylate (26).25 Hydro-
peroxide 7 (219 mg, 1.17 mmol) was reacted with 10 mol % titanium
tetraisopropoxide (TTIP) in 20 mL of CDCl3 for 12 h at −15 °C. The
solvent was removed under reduced pressure, and the crude product
was purified by column chromatography [c-hex/EtOAc (3:1), Rf = 0.3]
REFERENCES
■
(1) (a) Alberti, M. N.; Orfanopoulos, M. Tetrahedron 2006, 62,
10660−10675. (b) Clennan, E. L.; Pace, A. Tetrahedron 2005, 61,
6605−6691. (c) Clennan, E. L. Tetrahedron 2000, 56, 9151−9179.
(d) Stratakis, M.; Orfanopoulos, M. Tetrahedron 2000, 56, 1595−
1615.
1
to yield the product as a colorless oil (121 mg, 51%; dr = 88:12). H
NMR (300 MHz, CDCl3) major diastereoisomer δ (ppm) = 1.22−
1.30 (m, 6H), 2.40 (s, 1H), 2.65 (d, 1H, J = 5.5 Hz), 3.06 (d, 1H, J =
5.5 Hz), 3.88 (q, 1H), 4.17 (q, 2H, J = 7.1 Hz), 6.07 (d, 1H, J = 15.6
Hz), 7.05 (d, 1H, J = 15.6 Hz); 13C NMR (75 MHz, CDCl3) major
diastereoisomer δ (ppm) = 14.3 (CH3), 18.9 (CH), 53.4 (CH2), 60.8
(CH2), 66.8 (CH), 123.1 (CH), 142.8 (CH), 166.1 (Cq); IR ν (cm−1)
= 3458, 2982, 2936, 2907, 1718, 1659; HRMS (ESI) calcd for
C9H14O4 m/z 209.0784 [M + Na]+, found 209.0783.
(2) (a) Adam, W.; Braun, M.; Griesbeck, A. G.; Lucchini, V.; Staab,
E.; Will, B. J. Am. Chem. Soc. 1989, 111, 203−212. (b) Adam, W.;
Nestler, B. J. Am. Chem. Soc. 1993, 115, 7226−7231.
(3) (a) Alberti, M. N.; Orfanopoulos, M. Chem.Eur. J. 2010, 16,
9414−9421. (b) Stephenson, L. M.; Grdina, M. J.; Orfanopoulos, M.
Acc. Chem. Res. 1980, 13, 419−425.
(4) Maranzana, A.; Ghigo, G.; Tonachini, G. Chem.Eur. J. 2003, 9,
(E)-Ethyl 3-(2-(1-Hydroxyethyl)oxiran-2-yl)-2-methylacrylate
(27).25 Hydroperoxide 9 (200 mg, 1.00 mmol) was reacted with 10
mol % TTIP in 20 mL of CDCl3 for 12 h at −15 °C. The solvent was
removed under reduced pressure, and the crude product was purified
by column chromatography [c-hex/EtOAc (2:1), Rf = 0.36] to yield
the product as a colorless oil (90 mg, 45%; dr = 90:10). 1H NMR (300
MHz, CDCl3) major diastereoisomer δ (ppm) = 1.27 (d, 3H, J = 6.4
Hz), 1.30 (t, 3H, J = 7.1 Hz), 1.90 (d, 3H, J = 1.4 Hz), 2.73 (d, 1H, J =
5.3 Hz), 3.12 (d, 1H, J = 5.3 Hz), 3.88 (q, 1H, J = 6.2 Hz), 4.20 (q,
2616−2626.
(5) Orfanopoulos, M.; Stephenson, L. M. J. Am. Chem. Soc. 1980,
102, 1417−1418.
(6) Hurst, J. R.; Wilson, S. L.; Schuster, G. B. Tetrahedron 1985, 41,
2191−2197.
(7) Singleton, D. A.; Hang, C.; Szymanski, M. J.; Meyer, M. P.;
Leach, A. G.; Kuwata, K. T.; Chen, J. S.; Greer, A.; Foote, C. S.; Houk,
K. N. J. Am. Chem. Soc. 2003, 125, 1319−1328.
1
2H, J = 7.1 Hz), 6.89 (s, 1H); H NMR (300 MHz, CDCl3) minor
(8) Leach, A. G.; Houk, K. N.; Foote, C. S. J. Org. Chem. 2008, 73,
8511−8519.
diastereoisomer δ (ppm) = 1.27 (d, 3H, J = 6.4 Hz), 1.30 (t, 3H, J =
7.1 Hz), 1.90 (d, 3H, J = 1.4 Hz), 2.73 (d, 1H, J = 5.3 Hz), 3.12 (d,
1H, J = 5.3 Hz), 3.88 (q, 1H, J = 6.2 Hz), 4.20 (q, 2H, J = 7.1 Hz),
6.89 (s, 1H); 13C NMR (75 MHz, CDCl3) major diastereoisomer δ
(ppm) = 14.2 (CH3), 14.2 (CH3), 18.5 (CH3), 49.9 (CH2), 61.1
(CH2), 66.9 (CH), 133.8 (CH), 134.0 (CH), 167.4 (Cq); 13C NMR
(75 MHz, CDCl3) minor diastereoisomer δ (ppm) = 13.9 (CH3), 13.9
(CH3), 19.6 (CH3), 51.7 (CH2), 61.0 (CH2), 68.4 (CH), 134.5 (CH),
134.5 (Cq), 167.5 (Cq); IR ν (cm−1) = 3470, 2984, 2940, 2925, 1735,
1671, 1655, 1643.
(9) Rehbein, J.; Carpenter, B. K. Phys. Chem. Chem. Phys. 2011, 13,
20906−20922.
(10) Griesbeck, A. G.; Goldfuss, B.; Leven, M.; de Kiff, A.
Tetrahedron Lett. 2013, 54, 2938−2941.
(11) Schweitzer, C.; Schmidt, R. Chem. Rev. 2003, 103, 1685−1757.
(12) Di Mascio, P.; Devasagayam, T. P. A.; Kaiser, S.; Sies, H.
Biochem. Soc. Trans. 1990, 18, 1054−1056.
(13) Bosio, G. N.; Breitenbach, T.; Parisi, J.; Reigosa, M.; Blaikie, F.
H.; Pedersen, B. W.; Silva, E. F. F.; Mar
Chem. Soc. 2013, 135, 272−279.
́
tire, D. O.; Ogilby, P. R. J. Am.
(14) Speranza, G.; Manitto, P.; Monti, D. J. Photochem. Photobiol., B
1990, 8, 51−61.
COMPUTATIONAL DETAILS
■
All of the computations in this work were carried out with the program
package TURBOMOLE 6.331 employing the unrestricted TPSS
functional (exchange and correlation)28 and the contracted TZVP
basis from the Ahlrich group.29 All of the structure geometries were
fully optimized using quasi-Newton−Raphson methods and rational-
ized by analytic frequency calculations. Zero-point energy corrections
were omitted for all activation barriers, since the harmonic
approximation caused errors in the vibrational contributions from
the intermediates BRa and BRb (see Scheme 11) in range of 0.3 kcal/
mol. The multipole-accelerated resolution of identity approximation
was used to reduce the computational cost without producing
qualitative errors; the grid size for numerical integration was set to m5.
(15) Manitto, P.; Speranza, G.; Monti, D.; Gramatica, P. Tetrahedron
Lett. 1987, 28, 4221−4224.
(16) Wilkinson, F.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref.
Data 1995, 24, 663−1021.
(17) Adam, W.; Krebs, O.; Orfanopoulos, M.; Stratakis, M. J. Org.
Chem. 2002, 67, 8395−8399.
(18) Griesbeck, A. G.; Adam, W.; Bartoschek, A.; El-Idreesy, T. T.
Photochem. Photobiol. Sci. 2003, 2, 877−881.
(19) (a) Li, X.; Ramamurthy, V. J. Am. Chem. Soc. 1996, 118, 10666−
10667. (b) Robbins, R. J.; Ramamurthy, V. Chem. Commun. 1997,
1071−1072. (c) Stratakis, M.; Froudakis, G. Org. Lett. 2000, 2, 1369−
1372.
(20) (a) Griesbeck, A. G.; Cho, M. Tetrahedron Lett. 2009, 50, 121−
123. (b) Griesbeck, A. G.; Uhlig, J.; Sottmann, T.; Belkoura, L.; Strey,
R. Chem.Eur. J. 2012, 18, 16161−16165.
ASSOCIATED CONTENT
* Supporting Information
NMR spectra of new compounds, thermal ellipsoid plot of the
X-ray diffraction structure of 5, and tables of total energies and
atom coordinates from the theoretical calculations. This
material is available free of charge via the Internet at http://
pubs.acs.org.
■
S
(21) (a) Adam, W.; Griesbeck, A. G. Synthesis 1986, 1050−1052.
(b) Orfanopoulos, M.; Foote, C. S. Tetrahedron Lett. 1985, 26, 5991−
5994.
(22) Stensaas, K. L.; Bajaj, A.; Al-Turk, A. Tetrahedron Lett. 2005, 46,
715−718.
(23) Gollnick, K.; Griesbeck, A. G. Tetrahedron 1984, 40, 3235−
3250.
AUTHOR INFORMATION
Corresponding Author
*Fax: + 49 221 470 3083. Tel: + 49 221 470 5057. E-mail:
griesbeck@uni-koeln.de.
■
(24) Crystal structure of endoperoxide 5: deposition number CCDC
945446; C9H14O4; a = 6.0161(4) Å, b = 14.424(2) Å, c = 21.643(3) Å;
space group Pbca.
(25) Griesbeck, A. G.; de Kiff, A. Org. Lett. 2013, 15, 2073−2075.
K
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(26) Yunus, K.; Balci, M. Tetrahedron 2003, 59, 2063−2066.
(27) Adam, W.; Richter, M. J. Acc. Chem. Res. 1994, 27, 57−62.
(28) Staroverov, V. N.; Scuseria, G. E.; Tao, J. M.; Perdew, J. P. J.
Chem. Phys. 2003, 119, 12129−12137.
(29) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97,
̈
2571−2577.
(30) Maranzana, A.; Canepa, C.; Ghigo, G.; Tonachini, G. Eur. J. Org.
Chem. 2005, 3643−3649.
(31) Ahlrichs, R.; Bar, M.; Haser, M.; Horn, H.; Kolmel, C. Chem.
̈
̈
̈
Phys. Lett. 1989, 162, 165−169.
(32) Yadav, J. S.; Pratap, T. V.; Rajender, V. J. Org. Chem. 2007, 72,
5882−5885.
(33) Loh, T. G.; Cao, G. Q.; Pei, J. Tetrahedron Lett. 1998, 39, 1457−
1460.
(34) Ni, Y.; Montgomery, J. J. Am. Chem. Soc. 2006, 128, 2609−2614.
(35) Piers, E.; Jung, G. L.; Ruediger, E. H. Can. J. Chem. 1987, 65,
670−682.
(36) Kemper, J.; Studer, A. Angew. Chem., Int. Ed. 2005, 44, 4914−
4917.
(37) Ahmed, M. M.; Mortensen, M. S.; O’Doherty, G. A. J. Org.
Chem. 2006, 71, 7741−7746.
(38) Moses, J. E.; Baldwin, J. E.; Bruckner, S.; Eade, S. J.; Adlington,
̈
R. M. Org. Biomol. Chem. 2003, 1, 3670−3684.
(39) Lebel, H.; Guay, D.; Paquet, V.; Huard, K. Org. Lett. 2004, 6,
3047−3050.
(40) Uriac, P.; Bonnic, J.; Huet, J. Tetrahedron 1985, 41, 5051−5060.
L
dx.doi.org/10.1021/jo5000434 | J. Org. Chem. XXXX, XXX, XXX−XXX