Stereochemical outcome of copper-catalyzed C-H insertion reactions. an experimental and theoretical study

Article abstract of DOI:10.1021/jo400415b

The combination of chiral preparative HPLC separation, VCD measurements, and theoretical calculations allows the unambiguous determination of the absolute configuration of the conformationally flexible products of copper-catalyzed carbene insertion reactions. DFT calculations were used to predict the stereochemical outcome of the copper-bis(oxazoline)-catalyzed C-H insertion reaction between methyl diazophenylacetate and tetrahydrofuran and also to predict the absolute configuration of the major stereoisomers derived from the same reaction with different cyclic ethers. These predictions were verified experimentally through NMR and VCD spectroscopy and allowed rationalization of the stereochemical outcome of these reactions without further derivatization of the products, which can be prblematic under certain conditions as described herein.

Full text of DOI:10.1021/jo400415b

Article
pubs.acs.org/joc
Stereochemical Outcome of Copper-Catalyzed C−H Insertion
Reactions. An Experimental and Theoretical Study
Gonzalo Jimenez-Oses,*,† Eugenio Vispe,‡ Marta Roldan,‡ Sergio Rodríguez-Rodríguez,‡,§
́
́
́
Pilar Lopez-Ram-de-Viu,‡ Luis Salvatella,‡ Jose A. Mayoral,‡ and Jose M. Fraile*,‡
́
́
́
†Department of Chemistry and Biochemistry, University of California, Los Angeles, Los Angeles, California 90095-1569,
United States
‡Instituto de Síntesis Química y Catal
isis Homogenea, Facultad de Ciencias, Universidad de Zaragoza-CSIC, 50009 Zaragoza, Spain
́ ́
S
* Supporting Information
ABSTRACT: The combination of chiral preparative HPLC
separation, VCD measurements, and theoretical calculations
allows the unambiguous determination of the absolute con-
figuration of the conformationally flexible products of copper-
catalyzed carbene insertion reactions. DFT calculations were
used to predict the stereochemical outcome of the copper-bis-
(oxazoline)-catalyzed C−H insertion reaction between methyl
diazophenylacetate and tetrahydrofuran and also to predict the
absolute configuration of the major stereoisomers derived from
the same reaction with different cyclic ethers. These predictions
were verified experimentally through NMR and VCD spectros-
copy and allowed rationalization of the stereochemical outcome
of these reactions without further derivatization of the products,
which can be prblematic under certain conditions as described herein.
of complex natural and, to a lesser extent, synthetic products
INTRODUCTION
■
and metal complexes with excellent results.2,5−34
Despite the great success achieved in the preparation of
enantiopure compounds, for which asymmetric catalysis has
played a key role, the determination of the absolute configuration
of each stereogenic center present in the new molecule remains
frequently elusive despite achieving a highly stereoselective
synthesis or an exhaustive purification.1 When X-ray diffraction
analysis or derivatization to known compounds is not feasible,
relative configurations and very scarcely absolute configurations
can be directly inferred by spectroscopic techniques like NMR,
polarimetry, and electronic or vibrational circular dichroism
(ECD2,3 or VCD,4 respectively). Together with the develop-
ment of these experimental techniques, the theoretical predic-
tion of magnetic, optical, electronic, and vibrational properties
of chemical compounds has emerged as a powerful tool in the
recent years. Particularly, VCD spectroscopy,4 which is based
on vibrational absorptions effected by polarized radiations,
benefits from many advantages, being applicable to virtually all
chiral compounds. VCD spectra are more complex and provide
much more structural information than ECD and are less influ-
enced by solvent effects.5 Additionally, the theoretical prediction of
VCD properties is performed in the ground (i.e., nonexcited)
electronic state and is less method-dependent than ECD.
Altogether, these advantages of VCD spectroscopy have contri-
buted to its rapid expansion within the organic community. In
this context, valuable efforts pioneered by Stephens et al. have
been performed recently to elucidate the absolute configuration
However, the theoretical prediction of any molecular property,
especially the spectroscopic ones, and the subsequent comparison
with experimental results suffer from a very serious drawback: it
can be performed straightforwardly only for quite rigid com-
pounds. In these situations, an excellent agreement between
experimental and theoretical calculations can be achieved.
Otherwise, the experimental values and spectra are the result of
the combination of the individual contributions from all the
conformers present in solution, which are generally explored
within the acquisition time scale. In these situations, theory can
provide reliable results only if all the possible conformers are
calculated and their properties are weighted with their relative
population under the experimental conditions.29,32,34 Despite
the increasing capability of modern theoretical methods to ac-
curately compute spectroscopic properties, even for large systems,
the high complexity arising from highly fluxional compounds has
seriously limited its application to real-life synthetic derivatives, and
this computational task is still challenging. Thus, an accurate descri-
ption of the conformational space of such compounds must be
done prior to the calculation of the spectroscopic properties. This
can be of particular importance when coexisting conformers display
opposite properties (vide infra).
Received: February 28, 2013
Published: May 23, 2013
© 2013 American Chemical Society
5851
dx.doi.org/10.1021/jo400415b | J. Org. Chem. 2013, 78, 5851−5857

The Journal of Organic Chemistry
Article
In this sense, the enantioselective insertion reaction of formal
carbenes into C−H bonds constitutes a major challenge in
current organic synthesis.35,36 Pioneering work by Davies et al.
has contributed to the rapid development of this chemistry, in
which high enantioselectivities have been reached using chiral
rhodium catalysts.37,38 Alternatively, our group has recently reported
an arguably green procedure based on the use of copper complexes
in the homogeneous or heterogeneous phase (Scheme 1).39−41
was established by NMR experiments on the basis of previous
studies reported by Davies et al.44 Again, the unlike (2R,αS and
2S,αR) isomers were the major products for the reaction with
THP. On the contrary, and due to the change in priority of the
substituents attached to C2, the like (2S,αS and 2R,αR) enantio-
mers were dominant when the reaction was conducted with 1,4-
dioxane. The diastereomeric ratios in all cases ranged from 50/50
to 75/25. In the case of the reaction with 1,3-dioxolane, only one
stereogenic center is formed.
Until now, no reliable structural information on the source of
the enantioselectivities of such processes has been available.
This is a major drawback not solved to date which limits the
high potential of these reactions. In this study, we combine
spectroscopic and theoretical techniques to establish, for the
first time, the stereochemical outcome of this reaction.
Scheme 1. Reactions between Methyl Diazophenylacetate
a
and Cyclic Ethers Catalyzed by Chiral Copper Complexes
RESULTS AND DISCUSSION
■
All attempts to obtain single monocrystals of any of the oily
products, either from enantioenriched mixtures of enantiomers
or even from pure enantiomers (see below), were unsuccessful.
In situations in which the only information available is a set of
peaks in a HPLC trace, one of the strategies to assess the
stereochemistry of each product, and subsequently the course
of the asymmetric induction, is to determine which compounds
have the same configuration at a given stereocenter.
a
Lap = Laponite (heterogeneous support).
For example, the absolute configuration of the major like-1
enantiomer may be determined by epimerization of the major
unlike-1 enantiomer, taking advantage of the presence of an
acidic hydrogen at the α position of these enolizable compounds.
This epimerization should lead to diastereomers with identical
configuration at C2. These experiments are thoroughly described
in the Supporting Information, and only the most significant
results will be described herein.
Hence, epimerization assays on Cα were performed on enantio-
enriched samples of unlike-1, where we assumed the configuration
of the major enantiomer is 1-(2R,αS) as proposed by Davies.42
Treatment with different bases and under different conditions led
to a mixture of the four stereoisomers. Under equilibrium
conditions, the unlike/like ratio was completely reversed with
respect to the catalytic results, showing that the like diastereomer
is more thermodynamically stable than the unlike product, which
is in good agreement with subsequent theoretical calculations
(vide infra). Strikingly, the enantiomeric purity was completely
lost, suggesting that not only the α position bearing the acidic
hydrogen but also the one adjacent to the endocyclic oxygen (C2)
was subjected to reaction upon basic treatment.
Good levels of stereoselection were achieved in the reaction
of methyl diazophenylacetate with tetrahydrofuran (THF) using
bis(oxazoline) (box) and azabis(oxazoline) (azabox) ligands
bearing phenyl and isopropyl groups (up to 64% enantiomeric
excess (ee) in the major diastereomer), the effect of immo-
bilizing the catalyst onto an inorganic support being remarkable
(up to 88% ee in the major diastereomer). However, only the
relative configuration of the Cα and C2 atoms of 1 could be
deduced from 1H NMR experiments for each of the dia-
stereomeric pairs (unlike and like). The absolute configuration
of each stereocenter could not be unambiguously determined,
and only the stereochemistry of the major unlike insertion
product could be tentatively assigned as (2R,αS) through the
qualitative coincidence of its specific optical rotation ([α]D25
−88.1) with previously reported values for the same ([α]D25
−74.0)42 and opposite ([α]2D5 = +68.4)43 enantiomers.
=
=
This methodology was extended to similar cyclic ethers, namely
tetrahydropyran (THP), 1,4-dioxane, and 1,3-dioxolane, to suc-
cessfully obtain compounds 2, 3 (obtained under both homo-
geneous and heterogeneous conditions), and 4 (obtained only
in heterogeneous phase), respectively (Scheme 1), with moderate
to good yields and stereoselectivities (up to 56−71% ee depending
on the substrate). To the best of our knowledge, the stereo-
selective preparation of these derivatives has not been reported
to date, precluding the comparison of their optical/spectroscopic
properties with reference data. As in the case of THF-derived
compounds, the relative configuration of each pair of enantiomers
Because the simultaneous enolization of the C2 position was
ruled out by additional assays performed in deuterated solvents,
a different mechanism for this transformation must be invoked.
The base-promoted retro-O-Michael reaction occurring in
related oxazolidines bearing alkyl esters and leading to α,β-
unsaturated compounds has been described.45−48
Scheme 2. Proposed Ring-Opening/Cyclization Mechanism for the Base-Promoted Racemization of Cα and C2 in Compound 1
5852
dx.doi.org/10.1021/jo400415b | J. Org. Chem. 2013, 78, 5851−5857

The Journal of Organic Chemistry
Article
Figure 1. (A−C) Overlay of the theoretical VCD spectra calculated for each of the 13 conformers of compound 4-(S) optimized at the B3LYP/
TZVP level. (A) Conformationally unweighted in the gas phase (global spectrum in green). (B) Conformationally weighted in the gas phase (global
spectrum in green). (C) Conformationally weighted in chloroform solution. (D) Plot of calculated Rt (CO st) vs ϕ dihedral angle for three different
rotamers of 4-(S) (θ ≃ 60° in blue, θ ≃ 180° in red, θ ≃ −60° in green). VCD calculations were performed at the B3LYP/TZVP level. In all cases,
the resolution (half-width at half-height) of the spectra is 8 cm−1.
We propose that a ring-opening/cyclization cascade reaction
takes place together with the simple epimerization of Cα upon
treatment of ester 1 with a base (Scheme 2).
However, under milder conditions, the like-1 product was
obtained with a significant enantiomeric excess, 1-(2R,αR) being
the major product, probably due to the fact that the protonation of
enolate 1-(2R) is faster than the ring-opening process (Scheme 2).
From this identification of the like enantiomers, the major like-1
product of the insertion reaction can be tentatively assigned to
1-(2S,αS).
These results revealed that the configurational stability of
chiral activated cyclic ethers can be seriously affected under
certain conditions, precluding the experimental determination
of their absolute configuration by selective epimerization, deriva-
tization, and so forth. In these situations, spectroscopic and theore-
tical methodologies are required to confidently assess the stereo-
chemistry of the resulting products.
VCD Spectroscopy. Both enantiomers of the major unlike-
2 and like-3 pairs and both enantiomers of 4 were isolated by
semipreparative chiral HPLC (Experimental Section). The
VCD spectra of the six enantiopure compounds were collected
(see Experimental Section). In some cases, the spectra of two
enantiomers were not exact mirror images, which was attributed to
small concentration differences and baseline effects.49
The proper description of the dominant conformations in
such flexible compounds is a key issue in accurately reproducing
the observed spectroscopic properties. This fact is clearly
illustrated in Figure 1A, which shows the radically different
calculated spectra of all conformers of 4-(S). The geometries
and relative population of these conformers are shown in
Figure S2 in the Supporting Information. The relative
populations of each conformer derived from their relative free
Gibbs energies (ΔG) through a Boltzmann distribution at
25 °C were used to scale their corresponding spectra. The sum
of all contributions was used to obtain the final theoretical
spectra that were compared to those measured experimentally.
As can be seen in panels B and C of Figure 1, in these systems,
the contribution of solvation effects (chloroform) on the
relative population of each conformer and thus on the global
theoretical spectrum is negligible.
A detailed inspection of the individual spectra associated with
each conformer reveals that the positive or negative values of
the rotational strength (Rt) of each vibration depend not only
on the absolute configuration of the stereogenic carbon but also
on the relative orientation of each functional group. It is worth
noting that the free rotations or conformational changes of
some functional groups have a more pronounced effect on the
energy and intensity of their associated vibrations than a change
of configuration at remote stereogenic centers. To further
illustrate this observation, Figure 1D shows the high variability
of the calculated rotational strengths of the carbonyl stretching
vibration (CO st 1775−1800 cm−1) of compound 4-(S) along
the dihedral ϕ. In view of these results, only a careful evaluation
of the conformational energies and a subsequent weighting of
each spectrum can lead to satisfactory conclusions.
The experimental data were compared to those predicted
theoretically. In order to select the most convenient methodology,
three different levels of theory (B3LYP/6-31G(d), B3LYP/TZVP,
and B3PW91/TZVP) were previously tested. As can be seen in
Figure S3 in the Supporting Information, the B3LYP functional
achieved the best performance, in terms of both the energy and
intensity associated with the different vibrations. According to
previous results,21−26 a triple-ζ basis set is required to accurately
reproduce the experimental spectra.
In view of these preliminary results, a combinatorial evalua-
tion of both the relative energies and the VCD properties of all the
conceivable conformers of each enantiomer of unlike-2, like-3, and 4
5853
dx.doi.org/10.1021/jo400415b | J. Org. Chem. 2013, 78, 5851−5857

The Journal of Organic Chemistry
Article
Figure 2. (A) Experimental VCD spectra of both like-3 enantiomers (upper panel) and conformationally weighted theoretical VCD spectrum of
3-(2S,αS) (lower panel). (B) Experimental VCD spectra of both enantiomers of compound 4 (upper panel) and conformationally weighted
theoretical VCD spectrum of 4-(S) (lower panel). VCD calculations were performed at the B3LYP/TZVP level. Solvent (CHCl3) interference is
observed in the 1205−1225 cm−1 region. In all cases, the resolution (half-width at half-height) of the spectra is 8 cm−1.
Figure 3. Geometric features and reaction trajectories considered for the calculation of all TS’s of the Cu(I)-catalyzed reaction of methyl
diazophenylacetate with THF.
was done (Figure S3 in the Supporting Information). The
comparison of the conformationally weighted theoretical and
experimental VCD spectra of both enantiomers of like-3 and 4
reveals remarkably good agreement between both sets of data
(panels A and B of Figure 2, respectively) when bearing in
mind all the fluxionality drawbacks mentioned above.
The experiments and calculations performed throughout this
study allowed us to definitely establish the absolute configu-
ration of the major C−H insertion products as (2R,αS) in the
case of reaction with THP, (2S,αS) with 1,4-dioxane, and (S)
with 1,3-dioxolane. These results agree with the configuration
proposed by Davies for the major insertion product derived
from THF.42 Thus, a steady induction of asymmetry toward an
(S) configuration of Cα has been observed in the Cu(I)-
catalyzed reaction of methyl diazophenylacetate with cyclic
ethers in the presence of (S)-(bis)oxazolines.
Theoretical Studies on the C−H Insertion Mechanism.
Aside from the experimental verification of the absolute con-
figuration of the reaction products, the theoretical prediction of
the most favorable pathways is a powerful tool to gain insights
into the stereochemical course of a given reaction and the
asymmetry induction mechanisms, particularly in metal-
carbenoid processes in which the stereoselective step is hidden
by the rate-limiting step such as in this type of C−H insertion
reactions (i.e., formation of the carbenoid).50,51 To this end, we
performed a theoretical analysis of all the possible reaction
pathways for a model C−H insertion reaction. In our previous
reports, we described the copper-catalyzed C−H insertion of
diazo compounds into several cyclic ethers in the presence of
box and azabox ligands.39−41 For the calculations, we selected
THF as the model substrate and the box(Ph) ligand bearing
phenyl groups at the oxazoline rings (Figure 3).
A total of 16 transition states (TS’s), resulting from the
combination of four geometric features including stereoisomers
and conformers (Figures 2 and S3 in the Supporting Information),
were located and characterized. The contribution of each TS was
quantified by a Boltzmann distribution at 66 °C based on B3LYP-
D3/6-31G(d) zero-point corrected electronic energies (Table 1).
The most stable arrangement for each stereoisomeric
approximation is shown in Figure 4. All calculated geometries
and a more detailed description of their geometric features are
summarized in Figure S4 and Table S1 in the Supporting
Information. All transition structures show a quasi-linear
hydride transfer from THF C2 to the copper−carbene complex
Cα (C2...H...Cα = 147−161°), leading to two formal moieties
(oxolan-2-yl carbocation and a neutral organocopper species),
which recombine in a highly asynchronous but concerted way
as previously reported by theoretical studies on alkane C−H
carbene insertion reactions catalyzed by copper52,53 and
rhodium.54 The presence of an incipient carbocation in the
TS’s agrees with several Hammett studies on metal-catalyzed
C−H carbene insertion reactions.55−57
Systematic differences between pro-αR and pro-αS approx-
imations can be observed. Thus, whereas the THF attack took
place through the catalyst equatorial plane (Na−CuCα···C2
torsion angle absolute value > 168°) in the pro-αS approximation,
a significant deviation is found for the pro-αR attack (Na−Cu
Cα···C2 torsion angle = 145−164°).
5854
dx.doi.org/10.1021/jo400415b | J. Org. Chem. 2013, 78, 5851−5857

The Journal of Organic Chemistry
Article
Such systematic αR vs αS differences are attributed in part to
the steric repulsions occurring between the ester group and the
oxazoline substituent. Such steric repulsions are originated by
the Cα partial sp2 to sp3 rehybridization, which is analogous to
our reaction model of a cyclopropanation reaction catalyzed by
the same copper complex.58,59 These destabilizing contacts confer
on all pro-αR TS’s a substantial conformational stress, reducing
the stereodiscrimination between the pro-2R and pro-2S pathways,
which is higher in the less hindered pro-αS approximations.
Significantly, the experimental trend is reproduced only when
an empirical correction to dispersion (D3)60 is added to the
energy of each TS. Otherwise, the (2S,αS) stereoisomer is pre-
dicted to be favored by about 2 kcal mol−1, which is in dis-
agreement with the experimental results. This result reinforces
the importance of van der Waals interactions in systems in
which a number of hydrophobic contacts can take place.
Particularly, π−π interactions between the oxazoline and
carbene phenyl rings, and also CH−π interactions between
the other oxazoline phenyl ring and the tetrahydrofuran moiety,
are apparent (3.2−3.4 Å) in the pro-αS approximations and
probably contribute to the stabilization of these pathways.
As summarized in Table 1, the calculated contributions of
each stereoisomer are in good agreement with the product
distribution obtained experimentally in the presence of box(Ph)
ligand. However, this fairly good correlation is obtained only
when all TS’s are calculated and included in the Boltzmann
distribution, because several reactive approximations are nearly
isoenergetic and contribute almost equally to the final mixture.
This issue is of major importance when evaluating the mecha-
nisms of asymmetry induction, especially when the stereo-
selectivity is not excellent. In such cases, overlooking just one
reaction pathway can change the predicted stereoselectivity
dramatically.
Table 1. Calculated B3LYP-D3/6-31G(d) Energies and
Stereoisomeric Distribution at 66 °C Derived from C−H
Insertion Transition States TS1
ΔEZPE
cumulative
mol %
experimental
a
structure
(kcal mol−1
)
mol %
0.3
ratio
TS1-(2R,αS)
alkoxy_O-in
3.2
0.0
1.9
0.1
5.2
1.0
5.1
0.4
5.1
2.3
4.7
1.6
3.4
5.5
3.9
2.3
66.5
27.4
4.5
50.9
25.2
10.8
13.1
TS1-(2R,αS)
alkoxy_O-out
35.2
2.0
29.1
0.0
8.2
0.0
19.2
0.0
1.1
0.0
3.4
0.2
0.0
0.1
1.2
TS1-(2R,αS)
oxy_O-in
TS1-(2R,αS)
oxy_O-out
TS1-(2S,αS)
alkoxy_O-in
TS1-(2S,αS)
alkoxy_O-out
TS1-(2S,αS)
oxy_O-in
TS1-(2S,αS)
oxy_O-out
TS1-(2R,αR)
alkoxy_O-in
TS1-(2R,αR)
alkoxy_O-out
TS1-(2R,αR)
oxy_O-in
TS1-(2R,αR)
oxy_O-out
TS1-(2S,αR)
alkoxy_O-in
1.5
TS1-(2S,αR)
alkoxy_O-out
TS1-(2S,αR)
oxy_O-in
TS1-(2S,αR)
oxy_O-out
a
Obtained with box(Ph) ligand in homogeneous phase.39
Finally, and as a complement to the kinetic stereoselectivity
predicted through the calculated TS’s, the thermodynamic
stability of the products was evaluated at the same theoretical
level. In sum, up to 12 unlike and 10 like conformers were cal-
culated for each stereoisomer. A very low unlike/like calculated
ratio (0.97) was obtained through a Boltzmann distribution of
the relative electronic energies of all conformers (Table S2 in
the Supporting Information). This value is opposite to the
kinetic distribution obtained both experimentally and theoret-
ically, and in a good agreement with the results obtained during
prolonged treatment of 1 with base.
In summary, the computational investigation of all
conceivable reaction pathways for the copper-carbene insertion
into THF confirmed the high propensity of (S)-bis(oxazoline)
ligands to promote pro-αS asymmetric induction due to steric
interactions between the carbene ester group and the alkyl/aryl
substituent of the oxazoline. The unlike/like diastereomeric
ratio was also predicted correctly, as was the high enantio-
selectivity in favor of the (2R,αS) enantiomer.
CONCLUSIONS
■
When highly flexible products are obtained, the spectroscopic
determination of their absolute configuration is a very chal-
lenging task. In such cases, VCD spectroscopy in combination
with rigorous quantum mechanics calculations provides a very
useful methodology.
Figure 4. B3LYP-D3/6-31G(d) minimum energy geometries of the
four stereoisomeric TS’s for the Cu−carbene complex insertion
reaction with THF (TS1). Distances are given in angstroms, and van
der Waals interactions are qualitatively depicted as orange lines.
In this work, a synergistic experimental and theoretical study
on the copper-catalyzed C−H insertion of diazo compounds into
cyclic ethers allowed the confident elucidation of the stereochemical
5855
dx.doi.org/10.1021/jo400415b | J. Org. Chem. 2013, 78, 5851−5857

The Journal of Organic Chemistry
Article
Computational Methods. All calculations reported in this paper
were performed within the density functional theory. The B3LYP
functional65,66 was used for all geometry optimizations. The 6-31G(d)
and TZVP basis sets were used for geometry optimizations and the
calculation of frequencies and VCD properties.67 BSSE corrections
were not considered in this work. All the stationary points located were
characterized by the correct number and nature of their imaginary
frequencies. Scaled frequencies were not considered. Bulk solvent effects
were considered implicitly by performing single-point energy calculations
on the gas phase optimized geometries, through the SMD polarizable
continuum model of Cramer and Thrular68 as implemented in Gaussian
09. The internally stored parameters for chloroform were used to calculate
solvation free energies (ΔGsolv). Electronic energies including Grimme’s
empirical correction to dispersion60 and Gibbs free energies (ΔG) were
used throughout the discussion. In all cases, the Gaussian 09 suite of
programs was used.69 Full details on calculated structures and energies can
be found in the Supporting Information.
course of this reaction, which has been found to be general for a
series of related substrates.
Following the same trend exhibited by bis(oxazoline)−Cu(I)
complexes in the asymmetric cyclopropanation of olefins, the
highly asynchronous concerted carbene insertion takes place
preferably by the re face of the Cu(I)−carbene system. The
source of the observed stereoselection arises from a balance of
stereoelectronic factors in which the repulsive interactions between
the ester group and the alkyl/aryl groups of the oxazoline rings,
and also the attractive van der Waals interactions between the
substrates and the ligand, play major roles.
In epimerization studies, performed by basic treatment of
activated cyclic ethers, the absolute configuration of all the
stereogenic centers present on the molecule is affected through
a proposed ring-opening/cyclization process, leading to a com-
plete loss of asymmetry. These findings are a caveat for chem-
ical derivatization, a common strategy for the elucidation of
stereochemical induction.
ASSOCIATED CONTENT
■
S
* Supporting Information
Detailed description of specific experiments and theoretical
models; NMR and HPLC data of compounds 2−4; tables of
geometric features, electronic energies, and enthalpies, entropies,
and Gibbs free energies for the different conformations of the
structures considered in this work; calculated geometries of the
structures discussed in this paper; and additional figures and tables.
This material is available free of charge via the Internet at http://
pubs.acs.org.
EXPERIMENTAL SECTION
■
Bisoxazolines,61 azabisoxazolines,62−64 and supported catalysts39−41,62
were prepared as described in the literature. The copper-catalyzed reac-
tions of methyl diazophenylacetate with cyclic ethers and the reaction
products have been described elsewhere.39−41
Preparative Chiral HPLC. Compound 2. Column: Chiralpak IA
2 × 25 cm. Eluent: n-hexane/ethanol (99.8/0.2), 16 mL/min. Sample:
90 mg of unlike-2 + like-2 previously purified by silica gel column
chromatography. Sample concn: 100 mg/mL. Injections: 0.1 mL.
Retention times: 2-(2R,αS), 8.4 min; 2-(2S,αR), 9.5 min; like product
(enantiomer 1), 10.4 min; like product (enantiomer 2), 12.1 min.
Yields: 20 mg of 2-(2R,αS) [α]2D5 = −76.4° (c 0.55, CCl4); 30 mg of 2-
(2S,αR) [α]2D5 = +78.2° (c 0.75, CCl4). Compound 3. Column:
Chiralpak IB 2 × 25 cm. Eluent: n-hexane/acetone (98/2), 20 mL/
min. Sample: 206 mg of unlike-3 + like-3 previously purified by silica
gel column chromatography. Sample concn: 160 mg/mL. Injections:
0.1 mL. Retention times: unlike-3 (enantiomer 1), 15.4 min; unlike-3
(enantiomer 2), 16.2 min; 3-(2R,αR), 17.4 min; 3-(2S,αS), 21.0 min.
Yields: 40 mg of 3-(2R,αR) [α]2D5 = +50.6° (c 0.575, CHCl3); 48 mg of
3-(2S,αS) [α]2D5 = −46.6° (c 0.585, CHCl3). Compound 4. Column:
Chiralpak IB 2 × 25 cm. Eluent: n-hexane/acetone (99/1), 20 mL/
min. Sample: 118 mg. Sample concn: 50 mg/mL. Injections: 0.1 mL.
Retention times: 4-(R), 32.4 min; 4-(S), 34.2 min. Yields: 10 mg of
4-(R) [α]2D5 = −40.4° (c 0.65, CHCl3); 20 mg of 4-(S) [α]2D5 = +41.4°
(c 0.95, CHCl3).
Racemization Assays. Method 1a. Compound 1-(2R,αS) (40 mg,
0.18 mmol) was dissolved in dry methanol (1 mL), and NaOMe (0.5
M in methanol, 0.72 mL, 0.36 mmol) was added; the mixture was
stirred at room temperature for 24 h. Method 1b. Compound
1-(2R,αS) (10.5 mg, 0.048 mmol) was dissolved in dry methanol
(0.5 mL), and NaOMe (0.5 M in methanol, 76 μL, 0.038 mmol) was
added; the mixture was stirred at room temperature for 24 h. Method
1c. Compound 1-(2R,αS) (9.8 mg, 0.045 mmol) was dissolved in dry
methanol-d4 (0.7 mL), and solid NaOMe (2 mg, 0.036 mmol) was
added; the mixture was stirred at room temperature for 24 h. Method
2. Compound 1-(2R,αS) (10 mg, 0.045 mmol) was dissolved in dry
toluene (0.5 mL), and 1,3,5-triazabicyclodec-5-ene (TBD) bound to
polystyrene (13.8 mg, 0.036 mmol) was added; the mixture was stirred
at room temperature for 5 days. In all experiments, the reaction was
monitored by HPLC (Chiralpak IB 0.46 × 25 cm) using n-hexane/
isopropanol (99.5/0.5, 0.8 mL/min) as eluent. Retention times: 1-(2R,αS),
12.0 min; 1-(2S,αR), 13.1 min; 1-(2S,αS), 16.1 min; 1-(2R,αR), 17.3 min.
VCD Spectroscopy. Experiments were carried out in solution on a
Jasco FVS-4000 spectrometer using a cell equipped with BaF2
windows and a 0.1 mm spacer. Samples were dissolved in CCl4 or
CHCl3 to get an IR absorbance close to 0.4, and then 2000 scans were
collected in the range of 900−1800 cm−1 at a resolution of 8 cm−1.
Then spectra were corrected using the solvent as baseline.
AUTHOR INFORMATION
Corresponding Author
*E-mail: G.J.-O., gjimenez@chem.ucla.edu; J.M.F., jmfraile@
unizar.es.
■
Present Address
§S.R.-R.: Procter & Gamble, Innovation Center, Strombeek
Bever, Belgium.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was made possible by the generous financial support
of the Ministerio de Economia y Competitividad (MINECO,
■
́
Project CTQ2011-28124). We also thank the Supercomputing
Center of Galicia (CESGA) for computer support.
REFERENCES
■
(1) Chiral Analysis; Busch, K. W., Busch, M. A., Eds.; Elsevier:
Amsterdam, 2006.
(2) Bringmann, G.; Bruhn, T.; Maksimenka, K.; Hemberger, Y. Eur. J.
Org. Chem. 2009, 2009, 2717−2727.
(3) Mori, T.; Inoue, Y.; Grimme, S. J. Org. Chem. 2006, 71, 9797−
9806.
(4) Stephens, P. J.; Devlin, F. J.; Cheeseman, J. R. VCD Spectroscopy
For Organic Chemists; CRC Press: Boca Raton, FL, 2012.
(5) Nicu, V. P.; Baerends, E. J.; Polavarapu, P. L. J. Phys. Chem. A
2012, 116, 8366−8373.
(6) Devlin, F. J.; Stephens, P. J.; Cheeseman, J. R.; Frisch, M. J. J. Am.
Chem. Soc. 1996, 118, 6327−6328.
(7) Aamouche, A.; Devlin, F. J.; Stephens, P. J. Chem. Commun. 1999,
361−362.
(8) Aamouche, A.; Devlin, F. J.; Stephens, P. J.; Drabowicz, J.;
Bujnicki, B.; Mikolajczyk, M. Chem.Eur. J. 2000, 6, 4479−4486.
(9) Aamouche, A.; Devlin, F. J.; Stephens, P. J. J. Am. Chem. Soc.
2000, 122, 7358−7367.
(10) Stephens, P. J.; Davlin, F. J. Chirality 2000, 12, 172−179.
5856
dx.doi.org/10.1021/jo400415b | J. Org. Chem. 2013, 78, 5851−5857

The Journal of Organic Chemistry
Article
(11) Stephens, P. J.; Aamouche, A.; Devlin, F. J.; Superchi, S.;
Donnoli, M. I.; Rosini, C. J. Org. Chem. 2001, 66, 3671−3677.
(12) Devlin, F. J.; Stephens, P. J.; Scafato, P.; Superchi, S.; Rosini, C.
Tetrahedron: Asymmetry 2001, 12, 1551−1558.
(43) Suematsu, H.; Katsuki, T. J. Am. Chem. Soc. 2009, 131, 14218−
14219.
(44) Davies, H. M. L.; Ren, P. D. Tetrahedron Lett. 2001, 42, 3149−
3151.
(45) Jimenez-Oses, G.; Aydillo, C.; Busto, J. H.; Zurbano, M. M.;
Peregrina, J. M.; Avenoza, A. J. Org. Chem. 2007, 72, 5399−5402.
(46) Aydillo, C.; Jimenez-Oses, G.; Busto, J. H.; Peregrina, J. M.;
Zurbano, M. M.; Avenoza, A. Chem.Eur. J. 2007, 13, 4840−4848.
(47) Reider, P. J.; Conn, R. S. E.; Davis, P.; Grenda, V. J.; Zambito, A.
J.; Grabowski, E. J. J. J. Org. Chem. 1987, 52, 3326−3334.
(48) Park, H.-g.; Lee, J.; Kang, M. J.; Lee, Y.-J.; Jeong, B.-S.; Lee, J.-
H.; Yoo, M.-S.; Kim, M.-J.; Choi, S.-h.; Jew, S.-s. Tetrahedron 2004, 60,
4243−4249.
(13) Devlin, F. J.; Stephens, P. J.; Scafato, P.; Superchi, S.; Rosini, C.
Chirality 2002, 14, 400−406.
(14) Devlin, F. J.; Stephens, P. J.; Osterle, C.; Wiberg, K. B.;
Cheeseman, J. R.; Frisch, M. J. J. Org. Chem. 2002, 67, 8090−8096.
(15) Freedman, T. B.; Cao, X. L.; Dukor, R. K.; Nafie, L. A. Chirality
2003, 15, 743−758.
(16) Cere, V.; Peri, F.; Pollicino, S.; Ricci, A.; Devlin, F. J.; Stephens,
P. J.; Gasparrini, F.; Rompietti, R.; Villani, C. J. Org. Chem. 2005, 70,
664−669.
(49) Kuppens, T.; Bultinck, P.; Langenaeker, W. Drug Discovery
Today: Technol. 2004, 1, 269−275.
(17) Stephens, P. J.; McCann, D. M.; Devlin, F. J.; Flood, T. C.;
Butkus, E.; Stoncius, S.; Cheeseman, J. R. J. Org. Chem. 2005, 70,
3903−3913.
(50) Hansen, J.; Autschbach, J.; Davies, H. M. L. J. Org. Chem. 2009,
74, 6555−6563.
(18) Devlin, F. J.; Stephens, P. J.; Besse, P. Tetrahedron: Asymmetry
2005, 16, 1557−1566.
(51) Garcia, J. I.; Jimenez-Oses, G.; Mayoral, J. A. Chem.Eur. J.
2011, 17, 529−539.
(19) Devlin, F. J.; Stephens, P. J.; Bortolini, O. Tetrahedron:
Asymmetry 2005, 16, 2653−2663.
(52) Braga, A. A. C.; Maseras, F.; Urbano, J.; Caballero, A.; Díaz-
́
Requejo, M. M.; Perez, P. J. Organometallics 2006, 25, 5292−5300.
(53) Bonge, H. T.; Hansen, T. Eur. J. Org. Chem. 2010, 2010, 4355−
4359.
(20) Carosati, E.; Cruciani, G.; Chiarini, A.; Budriesi, R.; Ioan, P.;
Spisani, R.; Spinelli, D.; Cosimelli, B.; Fusi, F.; Frosini, M.; Matucci,
R.; Gasparrini, F.; Ciogli, A.; Stephens, P. J.; Devlin, F. J. J. Med. Chem.
2006, 49, 5206−5216.
(54) Nakamura, E.; Yoshikai, N.; Yamanaka, M. J. Am. Chem. Soc.
2002, 124, 7181−7192.
(21) Stephens, P. J.; McCann, D. M.; Devlin, F. J.; Smith, A. B., III J.
Nat. Prod. 2006, 69, 1055−1064.
(55) Wang, J.; Chen, B.; Bao, J. J. Org. Chem. 1998, 63, 1853−1862.
(56) Davies, H. M. L.; Jin, Q. H.; Ren, P. D.; Kovalevsky, A. Y. J. Org.
Chem. 2002, 67, 4165−4169.
(22) Stephens, P. J.; Devlin, F. J.; Gasparrini, F.; Ciogli, A.; Spinelli,
D.; Cosimelli, B. J. Org. Chem. 2007, 72, 4707−4715.
(57) Mbuvi, H. M.; Woo, L. K. Organometallics 2008, 27, 637−645.
(58) Fraile, J. M.; Garcia, J. I.; Martinez-Merino, V.; Mayoral, J. A.;
Salvatella, L. J. Am. Chem. Soc. 2001, 123, 7616−7625.
(59) Fraile, J. M.; Garcia, J. I.; Gil, M. J.; Martinez-Merino, V.;
Mayoral, J. A.; Salvatella, L. Chem.Eur. J. 2004, 10, 758−765.
(60) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104.
́ ́ ̌
(23) Stephens, P. J.; Pan, J.-J.; Devlin, F. J.; Urbanova, M.; Hajícek, J.
J. Org. Chem. 2007, 72, 2508−2524.
(24) Krohn, K.; Gehle, D.; Dey, S. K.; Nahar, N.; Mosihuzzaman, M.;
Sultana, N.; Sohrab, M. H.; Stephens, P. J.; Pan, J.-J.; Sasse, F. J. Nat.
Prod. 2007, 70, 1339−1343.
(25) Stephens, P. J.; Pan, J.-J.; Krohn, K. J. Org. Chem. 2007, 72,
7641−7649.
(61) Cornejo, A.; Fraile, J. M.; Garcia, J. I.; Gil, M. J.; Martinez-
Merino, V.; Mayoral, J. A.; Pires, E.; Villalba, I. Synlett 2005, 2321−
2324.
(62) Fraile, J. M.; Garcia, J. I.; Herrerias, C. I.; Mayoral, J. A.; Reiser,
O.; Socuellamos, A.; Werner, H. Chem.Eur. J. 2004, 10, 2997−3005.
(63) Glos, M.; Reiser, O. Org. Lett. 2000, 2, 2045−2048.
(64) Werner, H.; Vicha, R.; Gissibl, A.; Reiser, O. J. Org. Chem. 2003,
68, 10166−10168.
(26) Stephens, P. J.; Pan, J. J.; Devlin, F. J.; Krohn, K.; Kurtan, T. J.
Org. Chem. 2007, 72, 3521−3536.
(27) Lattanzi, A.; Scettri, A.; Zanasi, R.; Devlin, F. J.; Stephens, P. J. J.
Org. Chem. 2010, 75, 2179−2188.
(28) Ji, Z.; Santamaria, R. N.; Garzon
114, 3591−3601.
(29) Abbate, S.; Lebon, F.; Lepri, S.; Longhi, G.; Gangemi, R.;
Spizzichino, S.; Bellachioma, G.; Ruzziconi, R. ChemPhysChem 2011,
12, 3519−3523.
́
, I. L. J. Phys. Chem. A 2010,
(65) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
(66) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(67) Harihara, Pc.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213−222.
(68) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B
2009, 113, 6378−6396.
(30) Pandith, A. H.; Islam, N.; Syed, Z. F.; Rehman, S.-u.; Bandaru,
S.; Anoop, A. Chem. Phys. Lett. 2011, 516, 199−203.
(31) Sato, H.; Uno, H.; Nakano, H. Dalton Trans. 2011, 40, 1332−
1337.
(69) Frisch, M. J.; et al. Gaussian 09; Gaussian, Inc.: Wallingford, CT,
2009.
(32) Moreno, J. R. A.; Urena, F. P.; Gonzal
́
ez, J. J. L.; Gam
Martínez-Haya, B. Tetrahedron: Asymmetry 2012, 23, 294−299.
(33) Wu, T.; Zhang, X.-P.; Li, C.-H.; Bour, P.; Li, Y.-Z.; You, X.-Z.
Chirality 2012, 24, 451−458.
́
ez, F.;
̃
̌
(34) Zhu, P.; Yang, G.; Poopari, M. R.; Bie, Z.; Xu, Y. ChemPhysChem
2012, 13, 1272−1281.
(35) Davies, H. M. L.; Beckwith, R. E. J. Chem. Rev. 2003, 103,
2861−2903.
(36) Doyle, M. P.; Duffy, R.; Ratnikov, M.; Zhou, L. Chem. Rev.
2010, 110, 704−724.
(37) Davies, H. M. L.; Loe, O. Synthesis 2004, 2595−2608.
(38) Davies, H. M. L.; Nikolai, J. Org. Biomol. Chem. 2005, 3, 4176−
4187.
(39) Fraile, J. M.; Garcia, J. I.; Mayoral, J. A.; Roldan, M. Org. Lett.
2007, 9, 731−733.
(40) Fraile, J. M.; Mayoral, J. A.; Ravasio, N.; Roldan, M.; Sordelli, L.;
Zaccheria, F. J. Catal. 2011, 281, 273−278.
(41) Fraile, J. M.; Lopez-Ram-de-Viu, P.; Mayoral, J. A.; Roldan, M.;
Santafe-Valero, J. Org. Biomol. Chem. 2011, 9, 6075−6081.
(42) Davies, H. M. L.; Hansen, T.; Churchill, M. R. J. Am. Chem. Soc.
2000, 122, 3063−3070.
5857
dx.doi.org/10.1021/jo400415b | J. Org. Chem. 2013, 78, 5851−5857