Enantiodivergent Cyclization by Inversion of the Reactivity in Ambiphilic Molecules

Article abstract of DOI:10.1002/anie.202006650

Inverting the reactivity of the functional groups in ambiphilic molecules provides a new synthetic strategy to perform late-stage enantiodivergence. Both enantiomers of the final compound can be obtained from a common chiral precursor. As a proof of concept, the synthesis of substituted five- and six-membered oxacycles is described. The key step is the cyclization of an ambiphilic linear precursor bearing a propargylic alcohol and an epoxide linked through an alkyl chain. Through a slight modification of these linear precursors and employing different reaction conditions, these functional groups can inverse their chemical reactivity, producing one enantiomer or another of the final product. This enantiodivergent cyclization involves three stereogenic centers that can undergo fully controlled retention or inversion of their configuration depending on the cyclization pathway that is activated. The cyclization provides late-stage enantiodivergence, enabling the synthesis of either enantiomers of the oxacycles from a common chiral substrate with total transfer of the enantiomeric purity.

Full text of DOI:10.1002/anie.202006650

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
International Edition Chemie
www.angewandte.org
Accepted Article
Title: Enantiodivergent cyclization by inversion of the reactivity in
ambiphilic molecules
Authors: Tomás Martín, Julio Rodríguez-López, Margarita Brovetto,
and Víctor S. Martín
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202006650
Link to VoR: https://doi.org/10.1002/anie.202006650

Angewandte Chemie International Edition
10.1002/anie.202006650
RESEARCH ARTICLE
Enantiodivergent cyclization by inversion of the reactivity in
ambiphilic molecules
[
a]
[a]
[a],[b]
and Tomás Martín*[a],[c]
Julio Rodríguez-López, Margarita Brovetto, Víctor S. Martín,*
Dedicated to Professor Julius Rebek, Jr. on occasion of his 75th birthday
[
a]
Instituto Universitario de Bio-Orgánica Antonio González, Universidad de La Laguna, Francisco Sánchez 2, 38206 La Laguna, Tenerife, Spain
E-mail: vmartin@ull.es; tmartin@ipna.csic.es
[
[
b]
c]
Departamento de Química Orgánica, Universidad de La Laguna, Francisco Sánchez s/n. Facultad de Farmacia, 38200 La Laguna, Tenerife, Spain
Instituto de Productos Naturales y Agrobiología, CSIC, Francisco Sánchez 3, 38206 La Laguna, Tenerife, Spain
Supporting information for this article is given via a link at the end of the document.
Abstract: Inverting the reactivity of the functional groups in ambiphilic
molecules provides a new synthetic strategy to carry out late-stage
enantiodivergence. Both enantiomers of the final compound can be
obtained from a common chiral precursor. As a proof of concept, the
synthesis of substituted five- and six-membered oxacycles is
described. The key step is the cyclization of an ambiphilic linear
precursor bearing a propargylic alcohol and an epoxide linked through
an alkyl chain. Through a slight modification of these linear precursors
and employing different reaction conditions, these functional groups
can inverse their chemical reactivity, producing one enantiomer or
another of the final product. This enantiodivergent cyclization involves
three stereogenic centers that can undergo fully controlled retention
or inversion of their configuration depending on the cyclization
pathway that is activated. The cyclization provides late-stage
enantiodivergence, enabling the synthesis of either enantiomers of
the oxacycles from a common chiral substrate with total transfer of the
enantiomeric purity.
catalyst, reagent, or reaction conditions. There are very few
examples of such enantiodivergent reactions, and usually they
occur with a single element of symmetry such as a stereogenic
center or an axis of chirality (Figure 1B).[5] Therefore, a great
challenge for synthetic organic chemists is to find
enantiodivergent reactions where
a
greater number of
Introduction
Continued demand for both enantiomers of a given compound by
pharmaceutical and chemical companies has prompted the
development of a large number of enantiodivergent processes,
starting from achiral substrates.[1] This approach usually requires
the use of both enantiomers or pseudo-enantiomers of a chiral
reagent, or ligand/catalyst.[2] However a serious drawback of this
approach is the fact that many natural chiral sources are
accessible in only one absolute configuration. There are some
examples where by varying the external stimuli it is possible to
obtain both enantiomers of a compound using a single enantiomer
of a chiral reagent or ligand/catalyst.[3] Another interesting
alternative to obtain both enantiomers is the enantiodivergent
strategy that starts from chiral substrates. In this regard, two main
approaches can be considered: one that sequentially modifies the
functional groups, without altering the stereochemistry of the
chiral elements present in the starting substrate (Figure 1A),[4] and
the other that implies changes in the stereochemistry, which can
be define as an enantiodivergent reaction.[5] In both cases, only
one enantiomer of the starting material is required, and the use of
chiral reagents or catalysts is optional.[6] Herein we define an
enantiodivergent reaction as any reaction which generates both
enantiomers of the final product starting from the same (or two
very similar) enantiomeric substrate(s), just by changing the
Figure 1. Examples of an enantiodivergent synthesis (A) and previous works
for enantiodivergent reactions (B), and our strategy for the enantiodivergent
cyclization by inversion of the reactivity in ambiphilic molecules (C).
1
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202006650
RESEARCH ARTICLE
stereogenic centers are involved. The advantage of such an
approach is that it would allow both enantiomers to be obtained in
a more advanced stage in the synthesis of a desired compound.
We thought that this could be achieved through an ambiphilic
molecule (a molecule bearing electrophilic and nucleophilic sites)
where the reactivities of the functional groups could be inverted
depending on the reaction conditions. As a proof of concept, we
focus on the synthesis of cyclic ethers. We figured that an epoxide
and a propargylic alcohol placed in suitable positions of a linear
precursor constitute adequate functional groups to carry out this
aim. In this way, one could act as nucleophile and the other as an
electrophile or vice versa, providing an enantiodivergent route to
obtain both enantiomers of the final cyclic ether (Figure 1C). In
order to achieve this goal, it is essential that both cyclization
mechanisms occur in an orthogonal manner, since otherwise it
would result in partial or total racemization of the final product.
To the best of our knowledge, herein we present the first example
of an enantiodivergence process achieved by inversion of the
reactivity in ambiphilic molecules, as well as the first report
embracing more than one stereogenic center. This concept was
exemplified by the synthesis of substituted five- and six-
membered oxacycles through an enantiodivergent cyclization
step, where all the stereogenic centers of the starting material are
involved in the process.
propargylic position to give the tetrahydropyran cis-2 (path 1).
However, if this intermediate goes through an exo pathway, a
tetrahydrofuran is produced that could undergo ring expansion
and isomerization to give the tetrahydropyran ent-cis-2, i.e. the
enantiomer of the cis-2 (path 2). The third route is the nucleophilic
attack of the propargylic alcohol on the epoxide through an exo
pathway giving
a tetrahydrofuran that could undergo ring
expansion and isomerization to provide the tetrahydropyran cis-2
(path 3).[10] In order to achieve an enantiodivergent event in the
cyclization, it is necessary that the reaction takes place via path 2
or via path 3 exclusively in an orthogonal fashion. To limit the
number of possible routes in the cyclization step, we thought
about protecting the propargylic alcohol of 1 as an acetate, this
would prevent it from acting as a nucleophile. In this way, we
would ensure that only the Nicholas reaction occurs, where the
epoxide would act exclusively as a nucleophile. If the reaction
took place exclusively following path 2, we would obtain the
tetrahydropyran ent-cis-2. On the other hand, it would be
necessary to find the appropriate conditions for the nucleophilic
attack of propargylic alcohol on the epoxide following path 3 that
would lead to the enantiomeric tetrahydropyran cis-2 (Scheme 1).
Results and Discussion
To develop this idea, the epoxide is required to act as nucleophile
(
to attack the propargylic position, Figure 1C). Although this being
We began the study with the preparation of linear precursors, in
a non-trivial problem, we could fortunately resort to our previously
developed strategy for the synthesis of polysubstituted oxanes
and oxepanes. In those studies, we were able to perform this
transformation with a high degree of stereocontrol using a
Nicholas reaction, based on the intramolecular nucleophilic attack
of an epoxide to a carbocation generated by acid treatment of
hexacarbonyl dicobalt complexes of propargylic alcohols.[7,8] Thus,
a more thorough analysis of this Nicholas reaction with the cobalt
complex of epoxide 1 allowed us to disclose three possible routes
to the cyclization products (Scheme 1). Two of these routes have
an epoxonium ion as a common intermediate,[9] which can evolve
through an endo pathway followed by isomerization of the
which an epoxy-alcohol and a propargylic alcohol were linked
through an alkyl chain with two different lengths.[11] The linear
precursor 1 was prepared from the 2,3-epoxide alcohol 3, which
was obtained by a Katsuki-Sharpless asymmetric epoxidation of
the suitable allylic alcohol using the (R,R)-(+)-diisopropyl
tartrate.[12] The enantiomeric excess of epoxy-alcohol 3 was
determined by chiral HPLC using p-nitrobenzoate derivative 4
Scheme 2).[13] The next step was to use precursor 1 in the
(
cyclization step, and after several attempts we found that the exo-
cyclization of 1 through the nucleophilic attack of propargylic
alcohol on the epoxide was achieved quantitatively using catalytic
amounts of zinc chloride in dichloromethane at room temperature.
This results in an equimolecular mixture of the tetrahydrofurans
cis-5 and trans-5. This mixture could be easily separated after
carrying out an acetylation of the secondary alcohols to afford the
diacetylated tetrahydrofurans cis-6 and trans-6. In order to
determine the enantiomeric purity of these compounds, the
benzoates cis-7 and trans-7 were prepared,[13] and it was found
that the enantiomeric purity of the epoxide 3 is maintained during
the cyclization process. On the other hand, to study the Nicholas
reaction, diacetate 8 was prepared from compound 1, which was
then transformed into the corresponding hexacarbonyl dicobalt
complex. Fortunately, the reaction could be stopped at the
tetrahydrofurans ent-cis-5 and ent-trans-5 using montmorillonite
K-10 as mild acidic conditions (vide infra).[14] These compounds
were obtained in an approximately equimolar ratio and their
subsequent acetylation of the secondary alcohols allowed us to
reach the diacetates ent-cis-6 and ent-trans-6 that were easily
separated.[15] Analysis of the specific rotation values revealed that
the latter are enantiomers of the diacetates cis-6 and trans-6,
respectively. Additionally, chiral HPLC analysis of the benzoates
ent-cis-7 and ent-trans-7 confirms that the reaction proceeded
with total transfer of the enantiomeric purity from epoxide 3
Scheme 1. Possible routes in the Nicholas reaction for the formation of
trisubstituted tetrahydropyrans. Ac = acetyl. TMS = trimethylsilyl.
Scheme 2).[13] This clearly indicates that the Nicholas reaction
proceeds through the epoxonium ion and subsequent exo-
(
2
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202006650
RESEARCH ARTICLE
Scheme 2. Enantiodivergent synthesis of disubstituted tetrahydrofurans. The enantiomeric excess (ee) was determined by chiral HPLC, Chiralpak IC-3. Ac = acetyl.
Bz = benzoyl. CAN = ceric ammonium nitrate. DMAP = 4-dimethylaminopyridine. TMS = trimethylsilyl.
cyclization following path 2 (Scheme 1). It should be noted that
the enantiodivergent event lies exclusively in the cyclization step,
whereas all of the subsequent transformation (decomplexation,
acetylation, benzoylation, etc.) are carried out to reach the final
compounds that allow direct comparison.
However, when the Nicholas reaction was carried out with the
linear precursor
3 2
8 using BF ·OEt as a Lewis acid, the
trisubstituted tetrahydropyran ent-cis-2 was obtained directly
after acetylation (Scheme 3).[16] Taking into account the
stereochemical outcome of the reaction, this compound derives
from the ring expansion undergone by the cobalt complexes of
the tetrahydrofurans ent-cis-5 and ent-trans-5 due to the attack
of the exocyclic secondary alcohol to the stabilized propargylic
carbocation and subsequent isomerization of the trans- to the cis-
isomer following path 2 in Scheme 1.[17] The driving force for the
isomerization process is the greater thermodynamic stability of
the cobalt complex of the cis-isomer compared to that of the trans-
isomer.[18] This result was also confirmed by treatment of the
hexacarbonyl dicobalt complexes of the tetrahydrofurans ent-cis-
5
and ent-trans-5 with BF
3
·OEt
2
,
which yielded the
tetrahydropyran ent-cis-2 after acetylation and decomplexation.
Similarly, by performing the same sequence of reactions on the
hexacarbonyl dicobalt complexes of the tetrahydrofurans cis-5
and trans-5, the tetrahydropyran cis-2 was obtained after
acetylation and decomplexation. Again, the specific rotation value
confirmed cis-2 being the enantiomer of ent-cis-2. In addition, the
chiral HPLC analysis of the benzoate derivatives cis-9 and ent-
cis-9 demonstrated that the enantiomeric purity of epoxide 3 is
maintained throughout both cyclization processes: the
nucleophilic attack of the propargylic alcohol on the epoxide and
the Nicholas reaction, where the propargylic carbocation is the
electrophile and the epoxide is the nucleophile (Scheme 3).[
Trying to extend the concept, the next step was to use a precursor
with an additional carbon atom in the alkyl chain that links the
epoxide and the propargylic alcohol. In this way, the disubstituted
Scheme 3. Enantiodivergent synthesis of trisubstituted tetrahydropyrans. The
enantiomeric excess (ee) was determined by chiral HPLC, Chiralpak IC-3. Ac =
13]
acetyl. Bz
= benzoyl. CAN = ceric ammonium nitrate. DMAP = 4-
dimethylaminopyridine. TMS = trimethylsilyl.
3
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202006650
RESEARCH ARTICLE
Scheme 5. Enantiodivergent synthesis of disubstituted tetrahydropyrans. The
enantiomeric excess (ee) was determined by chiral HPLC, Chiralpak IC-3. Ac =
acetyl. Boc = tert-butoxycarbonyl. Bz = benzoyl. CAN = ceric ammonium nitrate.
DMAP = 4-dimethylaminopyridine. MS = molecular sieves. TMS = trimethylsilyl.
Scheme 4. Enantiodivergent synthesis of disubstituted tetrahydropyrans. The
enantiomeric excess (ee) was determined by chiral HPLC, Chiralpak IC-3. Ac =
acetyl. Bz
= benzoyl. CAN = ceric ammonium nitrate. DMAP = 4-
dimethylaminopyridine. TMS = trimethylsilyl.
tetrahydropyrans controls the final stereochemistry of this process,
a Nicholas cyclization of linear precursor 14 was carried out using
montmorillonite K-10 as an acid. After acetylation and
decomplexation, a mixture of tetrahydropyrans ent-cis-12 and
ent-trans-12, in a ratio of 3.6:1.0, was obtained. Considering that
montmorillonite K-10 was unable to isomerize neither the
hexacarbonyl dicobalt complex of the tetrahydropyran trans-12
nor the hexacarbonyl dicobalt complex of the monoacetate of the
trans-tetrahydropyran[19] (the cyclization product of 11 with
tetrahydropyran could be obtained directly. The linear precursor
1 was prepared from the 2,3-epoxide alcohol 10. In this case the
R,R)-(+)-diethyl tartrate was used as a chiral auxiliary in the
1
(
Katsuki-Sharpless asymmetric epoxidation.[12] The enantiomeric
excess of epoxide alcohol 10 was determined by chiral HPLC
using the p-nitrobenzoate derivative 4 (Scheme 4).[13] After
several attempts, we found that BF
3
·OEt
2
at low temperature is
BF
3
·OEt
2
) it is clear that the stereochemical course is dictated by
·OEt and not by the cyclization. It is
able to activate the epoxide of precursor 11 and produce the
attack of the propargylic alcohol to give an equimolecular mixture
of the tetrahydropyrans cis-12 and trans-12 after acetylation.[16]
In the next stage, we pondered to perform an isomerization to the
cis-compound via alkyne-hexacarbonyl dicobalt complexes under
acidic conditions, considering that the complex of this
stereoisomer is thermodynamically more stable.[18] While the
relatively mild acid montmorillonite K-10 failed, complete
isomerization to the tetrahydropyran cis-12 was observed using
the isomerization using BF
3
2
necessary to point out that the specific rotation value of ent-cis-
12 and the chiral HPLC analysis of the benzoate derivative ent-
cis-13 showed a slight loss of enantiomeric purity with respect to
that of the starting epoxide 10. Such loss of enantiomeric purity
may be consistent with the fact that the cyclization reaction does
not exclusively occur via Nicholas pathway (epoxide as
nucleophile), as it is possible that 3.5% of 14 react through the
epoxide acting as an electrophile. We have two possible
explanations for this fact: first, moisture in the reaction triggers a
Nicholas reaction occurs with water to generate the propargylic
alcohol, which acts as a nucleophile in the cyclization. However,
this possibility was ruled out because when the reaction was
3 2
BF ·OEt . The preparation of the dibenzoate cis-13 derivative
allowed us to confirm that the enantiomeric purity of the starting
epoxide 10 was maintained during the cyclization process
(
Scheme 4).[13] To avoid competition between the two cyclization
mechanisms, that would result in a loss of the enantiomeric purity
of the final product, we decided again to protect the propargylic
alcohol of 11 as an acetate, ensuring that only the Nicholas
reaction will take place. Thus, starting from linear precursor 14
performed in the presence of 4Å MS, a similar loss of
enantiomeric purity was observed. Second, the acetate in the
primary alcohol is not fully capable of driving the Nicholas reaction.
3 2
In this case, the BF ·OEt , instead of coordinating to the
and following a sequence of cobalt complex formation, cyclization, propargylic acetate to form a stabilized carbocation, activates the
acetylation and decomplexation, we were able to obtain the
tetrahydropyran ent-cis-12 in good yields. To determine whether
the cyclization or the isomerization of the obtained
epoxide and as a result the oxygen in the acetate of the
propargylic alcohol acts as a nucleophile generating the small
amount of the undesired enantiomer.
4
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202006650
RESEARCH ARTICLE
we can affirm that the reduction occurred with an enantiomeric
excess >99%. Therefore, compound 19 is
diastereoisomers in the epoxide with a diastereomeric excess
de) of 93.1%, which corresponds to the enantiomeric excess that
derives from the asymmetric epoxidation of compound 10
Scheme 6).[11]
Cyclization of compound 19 with BF
a mixture of
(
(
3
2
·OEt at low temperature
generated mostly cis-tetrahydropyran with the primary alcohol as
tert-butyl carbonate, which was converted to cyclic carbonate cis-
17 by treatment with sodium hydride. The enantiomeric excess of
this compound was determined through its dibenzoate derivative
cis-13, giving >99.5%.[13] This is because the other
diastereoisomer of 19 (<4%) provided the trans-isomer in the
cyclization stage, which was probably lost in the purification
process. On the other hand, the Nicholas reaction carried out on
linear precursor 20 leads to the opposite enantiomer of the cis-
17, that is, the cyclic carbonate ent-cis-17. The dibenzoate
derivative ent-cis-13 confirmed that the enantiomeric purity from
epoxide 10 is conserved throughout the cyclization, indicating that
the process runs exclusively through the Nicholas reaction where
the epoxide acts as a nucleophile.[13]
Scheme 5. Enantiodivergent synthesis of disubstituted tetrahydropyrans. The
enantiomeric excess (ee) was determined by chiral HPLC, Chiralpak IC-3. Ac =
acetyl. Boc = tert-butoxycarbonyl. Bz = benzoyl. CAN = ceric ammonium nitrate.
DMAP = 4-dimethylaminopyridine. MS = molecular sieves. TMS = trimethylsilyl.
Ts = tosyl.
Conclusion
In summary, we have introduced the concept of
enantiodivergence by inverting the reactivity of specific functional
groups in ambiphilic precursors. This strategy opens a new
approach for the synthesis of either enantiomer of a given
In order to solve this problem, the influence of the protecting group
in the primary alcohol on the stereochemical course of the
cyclization step was studied by changing the acetate group to a
Boc group. The introduction of the Boc protecting group is known
compound from
a common chiral substrate. Using linear
precursors that possess a propargylic alcohol and an epoxide
linked through an alkyl chain, we have developed an
enantiodivergent cyclization step, leading to both enantiomers of
substituted oxacycles. In this way, depending on the reaction
conditions, in one case the epoxide acts as an electrophile and
the propargylic alcohol as a nucleophile, and in the other case the
opposite occurs, the epoxide acts as a nucleophile and the
propargylic alcohol as an electrophile. Therefore, all three
stereogenic centers present in the substrate participate in the
processes, undergoing fully controlled retention or inversion of
their configuration, depending on the reaction conditions used. It
is noteworthy that the enantiomeric purity of the linear precursor
is completely transferred to the final products throughout both
cyclization pathways. The herein developed method allows
enantiodivergence to be carried out at a late stage of the
synthesis of a desired compound, holding great potential to
dramatically decrease the synthetic effort when both enantiomers
of a target compound are required.
to generate
a terminating nucleophile for acid-promoted
cascades.[20] In this case, when the Nicholas reaction was
performed with linear precursor 16, with the propargylic alcohol
as acetate and the primary alcohol as Boc, the cyclic carbonate
ent-cis-17 was obtained exclusively (Scheme 5). Gratifyingly,
after transformation of ent-cis-17 to the dibenzoate ent-cis-13 we
found that the enantiomeric purity of epoxide 10 was completely
transferred.[13] On the other hand, when compound 15 was treated
with BF
3
·OEt
2
at low temperature, the cis- and trans-
tetrahydropyrans were obtained with the primary alcohol
protected as tert-butyl carbonate, which were transformed into the
cyclic carbonates cis-17 and trans-17 with sodium hydride.
Following the isomerization protocol of the trans-tetrahydropyran
to the cis-tetrahydropyran described above, the cyclic carbonate
cis-17 was obtained in good yields. Again, the transformation of
the compound cis-17 to the dibenzoate cis-13 confirms that the
cyclization takes place with
a complete transfer of the
[
13]
enantiomeric purity of epoxide 10. With these results, it can be
stated that we developed an enantiodivergent cyclization for the
synthesis of disubstituted tetrahydropyrans, where all three
stereogenic centers are involved in the process.
Supporting Information. Material and general methods; detailed
1H, 13C and 2D NMR
experimental procedures and spectral data; copies of
spectra; chiral HPLC analysis.
Finally, to prove that the reactivity inversion of the functional
groups occurs only in the cyclization step (without the need of
subsequent isomerization), we turned our attention towards linear
precursor 19 where all three stereocenters are well-defined.
Compound 19 was obtained from the oxidation of the propargylic
alcohol of compound 15 to the corresponding acetylenic ketone,
and its subsequent asymmetric reduction through an asymmetric
transfer hydrogenation carried out with Noyori’s catalyst 18.[21]
Considering the proton NMR spectrum of the benzoate of
propargylic alcohol 19, as well as its chromatogram in chiral HPLC,
Acknowledgements
This research was supported by the Spanish MINECO and
MCIU/AEI
(CTQ2014-56362-C2-1-P,
CTQ2014-59649-P,
PGC2018-094503-B-C22,
and
PGC2018-094503-B-C21),
cofinanced by the European Regional Development Fund (ERDF).
5
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202006650
RESEARCH ARTICLE
Conflict of interest
Arenas, J. Iglesias-Sigüenza, R. Fernández, J. M. Lassaletta, J. Am.
Chem. Soc. 2016, 138, 12053-12056.
[
[
7]
8]
F. R. P. Crisóstomo, T. Martín, V. S. Martín, Org. Lett. 2004, 6, 565-568.
This strategy was also used in the formal synthesis of the (+)-muconin:
F. R. P. Crisóstomo, R. Carrillo, L. G. León, T. Martín, J. M. Padrón, V.
S. Martín, J. Org. Chem. 2006, 71, 2339-2345, and for the total synthesis
of the teurilene: J. Rodríguez-López, F. P. Crisóstomo, N. Ortega, M.
López-Rodríguez, V. S. Martín, T. Martín, Angew. Chem. 2013, 125,
The authors declare no conflict of interest.
Keywords: enantiodivergent reactions • ambiphilic molecules •
cyclic ethers • epoxides • Nicholas reaction
3747-3750; Angew. Chem. Int. Ed. 2013, 52, 3659-3662.
[
9]
Some examples where epoxonium ions are proposed as reaction
intermediates. a) H. Duddeck, V. Wiskamp, D. Rosenbaum, J. Org.
Chem. 1981, 46, 5332-5336; b) V. S. Kumar, D. L. Aubele, P. E.
Floreancig, Org. Lett. 2002, 4, 2489-2492; c) S. Wan, H. Gunaydin, K. N.
Houk, P. E. Floreancig, J. Am. Chem. Soc. 2007, 129, 7915-7923; d) Y.
Xie, P. E. Floreancig, Angew. Chem. 2013, 125, 653-656; Angew. Chem.
Int. Ed. 2013, 52, 625-628.
[
1]
2]
I. P. Beletskaya, C. Nꢀjera, M. Yus, Chem. Rev. 2018, 118, 5080-5200.
Selected examples: a) T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc.
[
1
980, 102, 5974-5976; b) Y. Gao, R. M. Hanson, J. M. Klunder, S. Y. Ko,
H. Masamune, K. B. Sharpless, J. Am. Chem. Soc. 1987, 109, 5765-
780; c) E. N. Jacobsen, I. Markó, W. S. Mungall, G. Shröder, K. B.
5
Sharpless, J. Am. Chem. Soc. 1988, 110, 1968-1970; d) J. Borges-
González, A. Feher-Voelger, F. P. Crisóstomo, E. Q. Morales, T. Martín,
Adv. Synth. Catal. 2017, 359, 576-583; e) J. Borges-González, I. García-
Monzón, T. Martín, Adv. Synth. Catal. 2019, 361, 2141-2147.
[
10] It is well-known that the cyclization reactions involving opening of
epoxides favored 5-exo-tet over 6-endo-tet processes, generating mostly
tetrahydrofurans. J. E. Baldwin, J. Chem. Soc., Chem. Commun. 1976,
[
3]
a) J. Escorihuela, M. I. Burguete, S. V. Luis, Chem. Soc. Rev. 2013, 42,
734-736.
5595-5617; b) D. G. Blackmond, A. Moran, M. Hughes, A. Armstrong, J.
[11] See the Supporting Information for further details.
Am. Chem. Soc. 2010, 132, 7598-7599; c) Y. Sohtome, S. Tanaka, K.
Takada, T. Yamaguchi, K. Nagasawa, Angew. Chem. 2010, 122, 9440-
[12] Reference 2a and T. Kasutki, V. S. Martín, Organic Reactions, Vol. 48
(Ed.: L. A. Paquette), Wiley, New York, 1996, pp. 1-299.
9443; Angew. Chem. Int. Ed. 2010, 49, 9254-9257; d) H. Lee, B. Y. Lee,
[
13] The benzoate or p-nitrobenzoate derivatives were prepared exclusively
with the purpose to determine the enantiomeric excess using a UV
detector coupled to chiral HPLC.
J. Yun, Org. Lett. 2015, 17, 764-766; e) J. Macharia, V. Wambua, Y.
Hong, L. Harris, J. S. Hirschi, G. B. Evans, M. J. Vetticatt, Angew. Chem.
2017, 129, 8882-8886; Angew. Chem. Int. Ed. 2017, 56, 8756-8760; f) A.
[14] a) F. R. P. Crisóstomo, R. Carrillo, T. Martín, V. S. Martín, Tetrahedron
Lett. 2005, 46, 2829-2832; b) N. Ortega, T. Martín, V. S. Martín, Org. Lett.
Martinez-Cuezva, M. Marin-Luna, D. A. Alonso, D. Ros-Ñiguez, M.
Alajarin, J. Berna, Org. Lett. 2019, 21, 5192-5196; g) J. Zhang, X. Huang,
R. K. Zhang, F. H. Arnold, J. Am. Chem. Soc. 2019, 141, 9798-9802; h)
M. Dommaschk, J. Echavarren, D. A. Leigh, V. Marcos, T. A. Singleton,
Angew. Chem. 2019, 131, 15097-15100; Angew. Chem. Int. Ed. 2019,
2006, 8, 871-873; c) N. Ortega, T. Martín, V. S. Martín, Eur. J. Org. Chem.
2009, 554-563; d) N. Ortega, V. S. Martín, T. Martín J. Org. Chem. 2010,
75, 6660-6672; e) J. Rodríguez-López, N. Ortega, V. S. Martín, T. Martín,
Chem. Commun. 2014, 50, 3685-3688.
58, 14955-14958; i) T. Hostmann, J. J. Molloy, K. Bussmann, R. Gilmour,
[
15] To determine if the stereochemical control in the formation of the
equimolecular mixture comes from the cyclization or isomerization
Org. Lett. 2019, 21, 10164-10168; j) C. Onneken, K. Bussmann, R.
Gilmour, Angew. Chem. 2020, 132, 338-342; Angew. Chem. Int. Ed.
2 6
reaction of the propargylic position, the ent-trans-6·Co (CO) was
2020, 59, 330-334.
treated with montmorillonite K-10, and after 5 hours (same time used to
complete the cyclization reaction of 8) and subsequent decomplexation,
a mixture of the compound ent-trans-6 and ent-cis-6 was obtained in
proportions 1.3:1.0, respectively. This result indicates that the
stereochemical course of this process is controlled mainly by the
isomerization reaction and not by the cyclization.
[
4]
a) S. Takano, M. Yanase, T. Sugihara, K. Ogasawara, J. Chem. Soc.,
Chem. Commun. 1988, 1538-1540, b) H. Kotsuki, Y. Ushio, I. Kadota, M.
Ochi, J. Org. Chem. 1989, 54, 5153-5161; c) T. Hudlicky, J. D. Price, F.
Rulin, T. Tsunoda, J. Am. Chem. Soc. 1990, 112, 9439-9440; d) M. E.
Fox, C. Li, J. P., Jr. Marino, L. E. Overman, J. Am. Chem. Soc. 1999,
121, 5467-5480; e) N. K. Garg, D. D. Caspi, B. M. Stoltz, J. Am. Chem.
[
16] The acetylation step is carried out to facilitate structural determination.
17] Ring contraction from oxepanes to tetrahydropyrans has been observed
with epoxides as trapping agents in the transposition of allylic alcohols
Soc. 2005, 127, 5970-5978; f) N. Toyooka, D. Zhou, H. Nemoto, Y.
Tezuka, S. Kadota, N. R. Andriamaharavo, H. M. Garraffo, T. F. Spande,
J. W. Daly, J. Org. Chem. 2009, 74, 6784-6791; g) A. K. Dubey, A.
Chattopadhyay, Tetrahedron: Asymmetry 2011, 22, 1516-1521; h) S. J.
Pérez, M. A. Purino, D. A. Cruz, J. M. López-Soria, R. M. Carballo, M. A.
Ramírez, I. Fernández, V. S. Martín, J. I. Padrón, Chem. Eur. J. 2016,
[
2 7
with catalytic amounts of Re O . See reference [9d].
[
18] In the trans isomer, the bulkiest substituent, the hexacarbonyl dicobalt
complex, is placed in the equatorial position and the other substituents
are located in the axial positions, generating three 1,3-diaxial interactions.
However, in the cis isomer, all substituents are placed at equatorial
positions and, therefore, the cis complex is more stable than the trans,
analogously to what happens in disubstituted tetrahydropyrans. a) S.
Tanaka, T. Tsukiyama, M. Isobe, Tetrahedron Letters 1993, 34, 5757-
22,15529-15535.
[
5]
a) E. S. Lewis, C. E. Boozer, J. Am. Chem. Soc. 1952, 74, 308-311; b)
C. Strohmann, M. Bindl, V. C. Fraaß, J. Hörnig, Angew. Chem. 2004,
116, 1029-1032; Angew. Chem. Int. Ed. 2004, 43, 1011-1014; c) T.
Kawabata, S. Matsuda, S. Kawakami, D. Monguchi, K. Moriyama, J. Am.
Chem. Soc. 2006, 128, 15394-15395; d) J. L. Stymiest, V. Bagutski, R.
M. French, V. K. Aggarwal, Nature, 2008, 456, 778-782; e) M. Sasaki, T.
Takegawa, K. Sakamoto, Y. Kotomori, Y. Otani, T. Ohwada, M.
Kawahata, K. Yamaguchi, K. Takeda, Angew. Chem. 2013, 125, 13194-
5760; b) S. Tanaka, M. Isobe, Tetrahedron 1994, 50, 5633-5644.
[
19] The final product of the Nicholas cyclization of 14 (before acetylation and
decomplexation) is the hexacarbonyl dicobalt complex of
tetrahydropyran monoacetate, and therefore we need to assure that this
complex does not isomerize with montmorillonite K-10. This could be
done using its enantiomer coming from the cyclization of 11 with
13198; Angew. Chem. Int. Ed. 2013, 52, 12956-12960; f) R. J. Armstrong,
M. Nandakumar, R. M. P. Dias, A. Noble, E. L. Myers, V. K. Aggarwal,
3 2
BF ·OEt , in this way an equimolecular mixture of the tetrahydropyrans
Angew. Chem. 2018, 130, 8335-8340; Angew. Chem. Int. Ed. 2018, 57,
cis and trans monoacetates is obtained. The hexacarbonyl dicobalt
complexes in this mixture were treated with montmorillonite K-10 for four
days and only a slight change in the cis:trans ratio was observed.
8
203-8208; g) Y. Ando, D. Tanaka, R. Sasaki, K. Ohmori, K. Suzuki,
Angew. Chem. 2019, 131, 12637-12643; Angew. Chem. Int. Ed. 2019,
8, 12507-12513.
5
[
20] a) F. E. McDonald, F. Bravo, X. Wang, X. Wei, M. Toganoh, J. R.
[
6]
Within this category, those cases in which the element of chirality can
change its configuration without bond-breaking or bond-forming were
excluded. An example for this is the dynamic kinetic resolution (DKR)
that occurs in the biaryl and binaphthyl systems: a) K. Mori, T. Itakura, T.
Akiyama, Angew. Chem. 2016, 128, 11814-11818; Angew. Chem. Int.
Ed. 2016, 55, 11642-11646; b) P. Ramírez-López, A. Ros, A. Romero-
Rodríguez, B. Do, W. A. Neiwert, K. I. Hardcastle, J. Org. Chem. 2002,
67, 2515-2523; b) F. Bravo, F. E. McDonald, W. A. Neiwert, B. Do, K. I.
Hardcastle, Org. Lett. 2003, 5, 2123-2126; c) J. Tanuwidjaja, S.-S. Ng,
T. F. Jamison, J. Am. Chem. Soc. 2009, 131, 12084-12085; d) B. S.
Underwood, J. Tanuwidjaja, S.-S. Ng, T. F. Jamison, Tetrahedron 2013,
69, 5205-5220; e) reference [8].
6
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202006650
RESEARCH ARTICLE
[21] K. Matsumura, S. Hashiguchi, T. Ikariya, R. Noyori, J. Am. Chem. Soc.
1997, 119, 8738-8739.
7
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202006650
RESEARCH ARTICLE
Entry for the Table of Contents
Two for one: Inverting the reactivity of the functional groups in ambiphilic molecules provides a new way to carry out late-stage
enantiodivergence. Thus, starting from a common chiral substrate both enantiomers of the final product can be obtained. This
concept was exemplified by the synthesis of substituted five- and six-membered oxacycles.
8
This article is protected by copyright. All rights reserved.