Enantioselective ruthenium-catalyzed 1,3-dipolar cycloadditions between C -carboalkoxy ketonitrones and methacrolein: Solvent effect on reaction selectivity and its rational

Article abstract of DOI:10.1021/jo5001737

A catalytic 1,3-dipolar cycloaddition between carboalkoxy ketonitrones and methacrolein under the effect of chiral ruthenium Lewis acid (R,R-1) was developed with high regio-, diastereo-, and enantiocontrol. The diastereochemical outcome of the cycloaddition reaction is marked by a significant solvent effect, and a divergent endo or exo control can be tuned by an appropriate choice of both the solvent and the N- and O-substituents of the ketonitrone. A rationale of the solvent effect, based on the computational study of the interactions between the methacrolein-Ru complex and its counteranion (SbF₆^-), is proposed to explain the selectivities obtained.

Full text of DOI:10.1021/jo5001737

Article
pubs.acs.org/joc
Enantioselective Ruthenium-Catalyzed 1,3-Dipolar Cycloadditions
between C‑Carboalkoxy Ketonitrones and Methacrolein: Solvent
Effect on Reaction Selectivity and Its Rational
Khalid B. Selim,†,‡ Arnaud Martel,† Mathieu Y. Laurent,† Jerome Lhoste,† Sandrine Py,§
́
̂
and Gilles Dujardin*,†
†LUNAM Universite
Cedex 9, France
́ ́ ́ ́
du Maine, Institut Molecules et Materiaux du Mans, UMR 6283 CNRS, Faculte des Sciences, 72085 Le Mans,
‡Department of Pharmaceutical Organic Chemistry, Faculty of Pharmacy, Mansoura University, Mansoura 35516, Egypt
§Dep
́ ́ ́
artement de Chimie Moleculaire (SERCO) UMR 5250, ICMG FR-2607, CNRS-Universite Joseph Fourier, BP 53, 38041
Grenoble Cedex 09, France
S
* Supporting Information
ABSTRACT: A catalytic 1,3-dipolar cycloaddition between
carboalkoxy ketonitrones and methacrolein under the effect of
chiral ruthenium Lewis acid (R,R-1) was developed with high
regio-, diastereo-, and enantiocontrol. The diastereochemical
outcome of the cycloaddition reaction is marked by a significant
solvent effect, and a divergent endo or exo control can be tuned
by an appropriate choice of both the solvent and the N- and O-
substituents of the ketonitrone. A rationale of the solvent effect,
based on the computational study of the interactions between
−
the methacrolein−Ru complex and its counteranion (SbF6 ), is
proposed to explain the selectivities obtained.
the opportunities gained by the use of C-ester-substituted
ketonitrones that permit the construction of isoxazolidines
containing up to two quaternary stereocenters.16
INTRODUCTION
■
1,3-Dipolar cycloaddition (1,3-DC) between nitrones and
ethylenic dipolarophiles is one of the most synthetically useful
pericyclic reactions, since the resulting isoxazolidines are
formed with up to 3 contiguous stereogenic centers and can
easily be converted into important chiral building blocks such
as complex β-lactams and β-amino acid derivatives (esters,
aldehydes, and alcohols).1 The efficiency of 1,3-DC requires a
control of both the regio- and diastereoselectivities in an
enantioselective manner. Major breakthroughs in asymmetric
1,3-DC between nitrones and ethylenic dipolarophiles have
been achieved upon activation by Lewis acids or organo-
catalysts.2 However, if these advances to date have concerned a
wide range of dipolarophiles including electron-rich (vinyl,
enols ethers and ketene ketals),3 neutral (allylic alcohols),4 and
electron-poor olefins (enals, ene-esters),5−15 variations of the
nitrone substrates have been more limited. Indeed, the most
efficient enantioselective reactions typically involve C-aryl-
substituted aldonitrones often bearing an aryl group as N-
substituent and in a lesser extent simple cyclic aldonitrones.
Actually, limited progress3f,4a,8g,15 has been made to develop
regio-, diastereo-, and facially controlled 1,3-DC reactions
involving other types of nitrones that could be of interest from
a synthetic point of view such as aldo- or ketonitrones
possessing functional C-substituent(s) (e.g., ester, amide) and/
or equipped with a deprotectable N-substituent. This is the
purpose of the approach developed in our group, highlighting
It must be pointed out that only rare examples of metal-
catalyzed, enantioselective 1,3-DC involving nitrones equipped
with an ester function have already been described. In the
aldonitrone (glyoxylate) series, reports are restricted to (i) the
case of organocatalyzed 1,3-DC reactions involving crotonalde-
hyde8g and (ii) the case of nitrone activation by a chelating
Lewis acid (CuBOX) for cycloaddition reactions involving vinyl
ethers.3f As a result of the inclination of such aldonitrones to
undergo quick Z/E isomerization at room temperature, a
mixture of cis/trans isomeric cycloadducts was obtained in both
cases.3f,8g
We recently developed aspartate- and alanine-derived N-
benzyl nitrones, which are configurationally stable E-ketoni-
trones, as new dipoles for controlling asymmetric 1,3-DC under
thermal conditions to form highly functionalized isoxazolidine
derivatives.17 In an organocatalyzed version, we demonstrated
that the diastereoselectivity could be completely controlled
when reacting crotonaldehyde with such a configurationally
stable E-ketonitrone. Enantioselectivities up to 95% and good
yields were achieved.15 In the present study, we now focus on
1,3-DC of these functional ketonitrones with methacrolein,
Received: January 24, 2014
Published: March 7, 2014
© 2014 American Chemical Society
3414
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
which shows the opposite regioselectivity as crotonaldehyde,
and we searched for an appropriate catalytic system. Indeed, as
previously shown by MacMillan’s group with α-aryl aldoni-
trones7 we found that the methodology using a chiral
imidazolidinone or another iminium precursor could not be
extended to methacrolein.
To improve the selectivity of the reaction, further
investigations were done by variation of the solvent and
temperature; we found an unusual effect of the reaction solvent
on selectivity and reactivity. First, changing the reaction solvent
from dichloromethane to chloroform resulted in a slight
increase in diastereocontrol, but conversion over the same
reaction time decreased dramatically (entry 4). The use of
dichloroethane gave similar results in comparison to dichloro-
methane (entry 5).
Effects of the coordinating properties of the solvent were
evaluated by performing the reaction in toluene and nitro-
methane. A weakly polar and noncoordinating solvent (e.g.,
toluene) led to inversion in diastereoselectivity to afford the exo
product as the major (entry 6), while a polar and weakly
coordinating solvent (e.g., nitromethane) switched off the
catalytic cycle to afford a sluggish conversion of the starting
material with a high exo selectivity (entry 7). These strong
solvent effects encouraged us to screen other solvents. In THF,
the endo selectivity was markedly enhanced (dr 93/7), while
enantioselectivity and reactivity were comparable to those in
dichloromethane (entry 8). tert-Butyl methylether (MTBE)
was found to be the optimal solvent in terms of selectivities and
reactivity, especially when the reaction was performed at −10
°C (endo/exo 97:3, 87% ee, 95% yield, entries 9, 10).
Decreasing the concentration of the reaction medium from
0.4 to 0.2 M of 2a gave a slight increase in endo- and
enantioselectivities (entry 11). Reactions performed in 2-
methylTHF and diethyl ether also exhibited enhanced
selectivities, albeit with a decreased reaction rate (entries 12,
13). Finally, decreasing the temperature to −20 °C did not
improve the selectivities (entry 14).
We thus turned our attention to metallocatalysis. Interest-
ingly, Kundig et al. have introduced a successful asymmetric
̈
activation of methacrolein as monodentate dipolarophile,
catalyzed by chiral iron and ruthenium Lewis acids.9 From
(Z)-diaryl aldonitrones, 1,3-DC reactions proceed with high
yields and total endo control but suffer from a lack of
regiocontrol: the 3,5-adduct is obtained as the major isomer
with moderate enantiomeric excesses (ee) in the case of
electron-deficient nitrones.9a−c This strategy depends on
inhibition of the coordination of nitrones to the Lewis acids,
which causes deactivation of the catalyst. Subsequently, many
developments have been achieved with other monocoordinat-
ing chiral Lewis acids based on CoIII,10 NiII,11 ZnII,12 TiIV,13 and
14,15
RhIII/IrIII
with valuable regiochemical outcomes, but as
already stated, all of these efforts concentrated on nitrones
derived from aromatic aldehydes. Interestingly, a total 3,5-
regiocontrol was recently observed, together with high
diastereo- and enantioselectivities, by Kundig’s group in the
reactions of N-alkyl C-aryl aldonitrones with methacrolein
using chiral ruthenium catalyst (R,R)-1 (Figure 1).9d,e
The uncatalyzed (background) reaction at 0 °C in dichloro-
methane, toluene, THF, or MTBE after 5 days resulted in the
same low conversion of 2a (9%), yielding an exo major product
(endo/exo 20:80) and showing no dependence on solvent
selection (Table 1, entry 15). Because the solvent does not
influence either conversion rate or selectivity in the uncatalyzed
reaction, the solvent effect in the Ru-catalyzed reaction might
arise from the interaction of the solvent with the methacrolein−
Ru complex or with the Ru-complex. In the related reactions
involving N-benzyl α-arylnitrones and methacrolein recently
Figure 1. 1,3-DC reaction studied with Ru-catalyst (R,R)-1.
In order to develop an enantioselective version of a 1,3-DC
reaction between ketonitrones 2 and methacrolein that would
lead to a N-deprotectable type of adduct possessing two
functionalized quaternary centers (Figure 1), we have achieved
in the present work a full experimental and theoretical study
with chiral monocationic complex (R,R)-1. This study affords
the first examples of enantioselective 1,3-DC reactions between
functionalized ketonitrones and methacrolein. The parameters
that influence the regio-, diastereo-, and enantiocontrolled
formation of the targeted isoxazolidines are discussed on the
basis of experiments and calculations, along with the evidence
of an unexpected solvent effect on the stereoselectivity of the
cycloaddition.
described by Kundig,9e the nature of the solvent was not
̈
investigated as a parameter influencing the exo/endo control. In
the present study, we evidence that a solvent effect occurs also
in the N-benzyl α-aryl-nitrone series (Table 2). Indeed, the
cycloaddition of aldonitrone 4 and methacrolein, reported with
catalyst (R,R-1) in dichloromethane at rt (Table 2, entry 1),9e
was tested at −10 °C, and variation of the solvent confirmed
our findings: compared to dichloromethane, the use of MTBE
as the solvent enhances both the endo- and enantioselectivity of
the cycloaddition (entries 2 vs 3).
RESULTS AND DISCUSSION
■
Under these optimized reaction conditions (MTBE, −10
°C), the scope and limitation of the 1,3-DC reactions of
methacrolein and C-carboxy ketonitrones were investigated by
variation of the C-alkyl, N- and O-substituents. N-Benzyl
methyl ester nitrones 2b (R1 = Et) and 2c (R1 = n-Pr) gave an
excellent diastereoselectivity with 84% ee and 86% ee,
respectively (Scheme 1), disclosing the successful extension
to nitrones bearing a bulkier C-substituent.
This procedure was also extended to nitrones bearing
different ester groups (Table 3). tert-Butyl ester nitrone 2d
afforded the highest enantioselectivity with good endo control
(92% ee, endo/exo 92:8, entry 1). Surprisingly, the reaction with
Cycloaddition Study. We started our investigation with
the reaction between methacrolein and nitrone 2a bearing an
easily removable N-benzyl group (Table 1). Racemic thermal
cycloaddition (neat, 90 °C, 2 days) gave 3,5-cycloadduct 3a
with a total conversion as two separated endo and exo isomers
in a 30:70 ratio. In the presence of 5 mol % of R,R-1 in
dichloromethane, nitrone 2a was treated with methacrolein at
room temperature for 5 days to afford with the same total 3,5-
regiocontrol the isoxazolidine 3a with minored diastereose-
lectivity and moderate enantioselectivities (entry 1). Lowering
the temperature and increasing the amount of the catalyst
improved the endo selectivity and ee’s (entries 2, 3).
3415
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
a
Table 1. Asymmetric 1,3-Dipolar Cycloaddition between Ketonitrone 2a and Methacrolein
b
,
c
d
e
e
entry
solvent
DCM
°C, days
yield (%)
endo/exo
ee (%) endo
ee (%) exo
f
1
rt, 5
0, 6
87 (100)
58 (72)
85 (87)
30 (52)
81 (87)
74 (84)
- (25)
46/54
47/53
70/30
78/22
72/28
39/61
18/82
93/7
66
71
78
78
76
74
-
70
75
75
72
75
68
-
f
2
DCM
3
DCM
0, 6
4
CHCl3
DCE
0, 5
5
0, 5
6
toluene
MeNO2
THF
0, 5
7
0, 4
8
0, 5
86 (86)
92 (95)
95 (97)
76
81
87
88
84
90
91
-
64
76
-
9
MTBE
MTBE
MTBE
2-MeTHF
Et2O
0, 5
90/10
97/3
10
−10, 5
−10, 6
−10, 6
−10, 6
−20, 6
0, 5
g
h
11
96 (96)
98/2
-
g
12
62 (62)
72 (74)
77 (78)
- (9)
95/5
-
g
13
g
88/12
97/3
-
14
MTBE
MTBE, DCM
-
i
j
15
20/80
-
a
b
c
d
1
Reactions were run with [2a] = 0.4 M. Isolated combined yield of exo and endo products. Conversions are in parentheses. Determined by H
e
f
g
NMR of the crude product. Determined by HPLC after reduction into the corresponding alcohol. 5 mol % of (R,R)-1. Reactions were run with
[2a] = 0.2 M. 90% yield of isolated pure 3a-endo. Reaction performed without catalyst Same result when the reaction is performed in toluene or
h
i
j
THF.
Table 2. Asymmetric 1,3-Dipolar Cycloaddition between C-
Aryl Aldonitrone 4 and Methacrolein
Table 3. Asymmetric 1,3-Dipolar Cycloaddition between
Ketonitrones 2 and Methacrolein
a
b
c
entry mol % solvent °C, days yield (%) endo/exo ee (%) endo
d
1
2
3
5
10
10
DCM
DCM
MTBE
+5, 3
−10, 5
−10, 5
95
92
99
95/5
92/8
99/1
87
91
95
a
b
Isolated combined yield of exo and endo products. Determined by
c
1H NMR of the crude product. Determined by HPLC after reduction
d
to the corresponding alcohol. Reference 9e.
Scheme 1. Extension to Bulkier C-Substituents of the
Nitrone
a
b
Isolated combined yield of exo and endo products. Determined by
c
1H NMR of the crude product. Determined by HPLC after reduction
d
into the corresponding alcohol. Reaction performed in toluene.
e
Reaction performed in dichloromethane.
exo selectivity was observed (exo/endo dr 86:14), together with
moderate enantioselectivity and good yield (entry 3). In
contrast, the stereochemical outcome of the reaction performed
in dichloromethane (entry 4) was similar to that observed in
MTBE.
In order to investigate the possible influence of the N-
substituent of the nitrone, N-diphenylmethyl and N-methyl
ethyl ester nitrone 2e gave the corresponding isoxazolidine with
moderate diastereoselectivity and a slow conversion rate of 2e
(entry 2). When the weakly exo-selective conditions used in the
reaction of methyl ester nitrone 2a (Table 1, entry 6) were
applied to ethyl ester nitrone 2e (toluene at 0 °C), an increased
3416
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
nitrones were tested in a comparative way to N-benzyl nitrones.
Unfortunately, the N-diphenylmethyl-substituted nitrone was
found to be unreactive under these catalytic conditions.
Increasing the steric bulk at the nitrogen seems to inhibit the
Ru-catalyzed reaction. The N-Me nitrone 2f (methyl ester)
showed good reactivity, but its cycloaddition with methacrolein
exhibited only low endo selectivity (Table 3, entry 5).
Surprisingly, the N-Me nitrone 2g (ethyl ester) exhibited an
interesting inversion of diastereoselectivity in MTBE to afford
the exo product in a 85:15 exo:endo ratio, but in low yield and
with moderate enantioselectivity (entry 6). Because the use of
catalyst R,R-1 in toluene was found to favor the formation of an
exo cycloadduct, the reaction was next performed in toluene
and afforded the expected exo diastereomer in an improved
93:7 exo:endo ratio (entry 7). Unexpectedly, with this nitrone
dichloromethane also favors the exo product, which is produced
in high yield with 81% ee (entry 8).
absence of the anion from the geometry given by the X-ray
structure described by Kundig et al.9f The calculations were
̈
performed using Gaussian 0919 with LANL2DZ (ECP) basis
set for the ruthenium atom and 6-31+G(d) for the other atoms.
The geometry was optimized at this level, and then a single-
point calculation of the electronic energy20 (E) using a SMD
solvation model was performed with the closest described
solvent, based on our experimental data (i.e., diethylether). The
electronic energy difference (ΔE) corresponding to the
exchange of the molecule of acetone in the initial complex by
methacrolein or MTBE was calculated (Scheme 3).
Scheme 3. Energy Difference Associated with the Exchange
of Acetone in the Complex
Determination of the Absolute Configuration. The
absolute configuration of the major 3a-endo adduct has been
established as depicted in Scheme 2. 3a-endo (87% ee, 94% de)
Scheme 2. Determination of the Absolute Configuration of
the Major 3a-endo Cycloadduct
The calculated ΔE for these exchanges unambiguously
demonstrates the incapacity for MTBE to compete with
methacrolein in the reactive site, probably as a result of the
steric hindrance around ruthenium and the bulkiness of MTBE.
was converted into a diastereomeric mixture of amines through
condensation with (S)-(−)-α-methylbenzylamine, followed by
in situ reduction of the resulting imine. The amine 6 was
saponified and lactamized to furnish the diastereomerically pure
lactam 7 in 69% yield. X-ray analysis of the crystalline 7
unambiguously established the absolute configuration of the
stereogenic centers as 1R,5S, corresponding to the 3R,5S
configuration in the major endo 3a adduct precursor.
Figure 2. Structure of the complex-(R,R)-1 with one molecule of
DFT Study of Ru-Complex Solvent Interaction. The
results obtained experimentally demonstrate that the selectiv-
ities of these processes depend on a delicate balance between
different factors, such as electronic and/or steric structure of
reactants and solvent effects. The modification of selectivity
may be reached by three main ways (or by their combinations):
coordination of a dipolarophile to a ruthenium Lewis acid,
variation of substituents, and use of the appropriate solvent.
The substantial effect of the solvent on selectivity is one of
the most intriguing results. Owing to this effect, besides the
natural steric hindrance around Ru when methacrolein is
chelated, the possibility of a direct interaction of the solvent
with Ru was first examined. The geometries of the complexes
with acetone, methacrolein, and MTBE were calculated using
Truhlar’s DFT method M06,18 described as a well-suited
method in the M06 class for modeling transition metal
complexes. To facilitate the calculations and to avoid ambiguity
on the position of the anion, the complexes were calculated in
MTBE.
As displayed in both structures (Figures 2 and 3), the
ruthenium cation is fully surrounded by the ligands
represented, and when the molecule of methacrolein is
coordinating the ruthenium by its oxygen atom, no additional
ligand can be placed in the coordination sphere of the
ruthenium cation. As confirmed by the X-ray structure obtained
−
by Kundig, even the counterion (SbF ) can be pushed away
̈
6
from the ruthenium by the presence of one molecule of
methacrolein, and the latter locates very close to one of the
pentafluorophenyl rings of the complex.9f This possible
competition between the anion and methacrolein might be at
the origin of the counterion effect on reactivity first observed by
Kundig.9g Indeed, this group observed an increased reactivity of
̈
the complexes with the largest anions that are more easily
3417
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
Figure 3. Structure A of the complex-(R,R)-1 with one molecule of methacrolein and the proposed approach of the nitrone.
displaced by methacrolein, thus enhancing the availability of the
active frustrated Lewis pair.
somewhat intriguing. In addition, the facial selectivity observed
in the present case is similar to the one described previously by
Kundig and confirms the hypothesis previously formulated on a
DFT Study of Methacrolein Orientation in the Ru-
Complex. In order to investigate the preferred geometry of the
methacrolein−Ru complex, we considered the possible rotation
around the Ru−O bond (Figure 4), as well as the possible
contribution of s-cis conformers of methacrolein.
̈
favored approach on the re face.9f Structure A (Figure 3) gives
some clues to explain the high facial selectivity observed. The
main facial approach is most probably controlled by the natural
orientation of methacrolein when chelated to the ruthenium.
Methacrolein is slightly twisted to limit the steric interactions
between its methyl group and one of the phenyl groups of the
catalyst, thus placing this methyl group in a lipophilic pocket.
One of the phenyl rings will then hinder the si face, leading the
re face to be the most accessible as previously described.9 The
facial selectivity is therefore mainly the consequence of the
steric repulsion between a phenyl group of the catalyst and the
approaching nitrone. However, the methyl group of meth-
acrolein also plays a key role by twisting the methacrolein to
offer the re face (Figure 3).
Rationale of the Solvent Effect on the Selectivity. The
unprecedented influence of the solvent on the endo/exo
selectivity of the cycloaddition remained unexplained at this
stage and required further studies. The molecular electrostatic
potential of the methacrolein−(R,R)-1 complex A was then
calculated (Figure 5). Represented in blue is the partial
Figure 4. Representation of conformers A−D and relative electronic
energy in kJ/mol.
Conformer B was calculated to be less stable than A (Figure
4). In addition, the application of the SMD solvation model,
using diethylether as the solvent, contributed only slightly to
enhance this energy difference by 0.6 kJ/mol, suggesting that
the relative stability of the favored conformer should not be
significantly altered by a change of solvent. The two
corresponding conformers derived from the s-cis conformation
of methacrolein C and D (Figure 4) are also less stable than A
with SMD solvation applied.
Although conformer C is more stable than conformer B, it is
most likely unproductive due to the inaccessibility of the double
bond of methacrolein turned into the complex. This double
bond stays between one of the phenyl rings and one of the
pentafluorophenyl rings of the catalyst, and thus the
contribution of conformer C to the cycloaddition is expected
Figure 5. Molecular electrostatic potential of the methacrolein−(R,R)-
1 complex A.
to be negligible. The X-ray structure described by Kundig et al.
̈
corresponds, therefore, to the most stable conformer in
solution. If present, the alternative conformers B and C (Figure
4) would contribute to dramatically decrease the facial
selectivity of the reaction, the complex presenting the opposite
side of methacrolein.
electrostatic positive charge of the catalyst−methacrolein cation
complex, the region displaying a natural affinity for a Lewis
base, such as the solvent itself. This figure gives evidence that
the atoms directly linked to the ruthenium are most affected by
the strong withdrawing effect of the ruthenium cation. The
aldehydic hydrogen of methacrolein therefore appears as
partially positively charged and is one part of the complex
The quite homogeneous level of enantioselectivity we
observed, whatever the orientation of the exo/endo control, is
3418
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
that will interact with the oxygen of the solvent (MTBE) or the
−
anion (SbF6 ). This kind of interaction with the anion was
postulated by Kundig et al. to explain the influence of the
̈
counterion on reactivity and selectivity.9f The X-ray structure of
complex 1 described by Kundig agrees with the HOESY
̈
experiment performed on an acrylonitrile complex to prove the
presence of the anion close to a region analogous to the one
represented in blue in Figure 5.
In order to evaluate the energy associated with the
coordination of the solvent in this region, the methacrolein−
(R,R)-1 complex A was modeled in the presence of one
molecule of MTBE (Figure 6). The structure displays a
Figure 7. Methacrolein−(R,R)-1-MTBE-SbF6 complex.
The structure places the pentafluorophenyl ring oriented
toward the SbF6 anion. The key role played by the
pentafluorophenyl rings of the ligand is therefore most likely
related to their interactions with the polyfluorinated anion,
allowing the anion to be pushed away from the reactive site to
enhance the reactivity. In addition, a complex methacrolein−
(R,R)-1-SbF6 was calculated at the same level of theory with
SMD solvent model to be less stable by 3.3 kJ/mol than the
corresponding MTBE complex (Figure 7). This weak energy
difference corroborates the assumption of a competition
between the solvent and the counterion close to the
methacrolein molecule. This change in the position of the
anion in the transition state is probably crucial for the endo/exo
selectivity. Hence, in the presence of a weak Lewis base as a
solvent such as toluene, the anion might stand very close to the
methacrolein. In contrast, good Lewis bases such as MTBE
would push the anion away from this position and facilitate the
interaction between the aldehyde and the nitrone. Solvents with
an intermediate behavior, such as dichloromethane, would then
lead to a poor discrimination in the reaction paths and thus to
midway selectivities, i.e., changing the environment of the
methacrolein from an anion to a lipophilic group (MTBE)
would change dramatically the approach of the nitrone.27
From this computational study, a better understanding of the
solvent effect on the diastereoselectivity of the Ru-catalyzed
cycloaddition was made possible. The significant influence of
the solvent on the cycloaddition outcome described herein may
echo previous reports of Doyle et al.,14b who reported an
enhancement of reaction rates and selectivities when toluene
was used in place of halocarbon solvents with cationic chiral
dirhodium carboxamidates catalyzed nitrone cycloadditions.
From the sum of our experimental results, it is clear that the
exo/endo selectivity is also significantly influenced by other
parameters such as the nature of the N-substituent and of the
ester function of the nitrones. At this stage, only a putative
computational study of the transition state energies would allow
us to take into account the contribution of these different
parameters to the global stereochemical outcome of the Ru-
catalyzed cycloaddition. However, the present results have
already demonstrated that these different parameters and
solvent effects can operate in a cooperative way, as it became
thus possible to orientate the diastereoselectivity to syntheti-
cally useful endo or exo adducts in highly enantioenriched forms
simply by tuning the N- and O-substituents of the α-
carboxynitrone with the nature of the solvent.
Figure 6. Calculated structure of the full complex methacrolein−
MTBE-Ru-catalyst (R,R)-1 energy gained through the interaction with
MTBE.
proximity between the aldehydic hydrogen and the oxygen of
MTBE (2.23 Å), that is adequate to fulfill Desiraju’s criteria21
for existence of a weak hydrogen bond (confirmed by a Wiberg
bond index of 0.0148). Various data account for similar weak
C−H...O bonds, such as the one between the oxygen of MTBE
and a vinylic hydrogen (distance: 2.25 Å; Wiberg bond index:
0.0162).22 The evidence of analogous C−H...O interactions was
well established23 from small molecules crystals,24 and the
hydrogen bond nature of these interactions is widely
accepted.25
The significant stabilization gained by the formation of this
new complex can be related to the ability of the oxygen in
MTBE to act as a better Lewis base due to a stronger inductive
effect of the alkyl groups (Figure 6).26 The mass action of the
solvent in the reaction medium will obviously contribute to this
interaction with the oxygen of MTBE. At high dilution, MTBE
is then fully able to compete with the counterion. The solvent
would then displace the anion in this position away from the
reactive site.
−
The full structure with SbF6 anion close to its initial
position was modeled at the same level of calculation (Figure
−
7), placing the SbF6 anion close to one of the pentafluor-
ophenyl rings and to two of the hydrogen atoms of the
cyclopentadienyl ring.
3419
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
chromatography was performed with G/UV254 plates, and the
products were observed under UV light or with KMnO4 stain. NMR
N−O Bond Cleavage. Finally, the synthetic utility of the
highly functionalized 5-formyl isoxazolidines 3, which contain
two quaternary centers of controlled configuration, was
demonstrated by transformation of the pure endo cycloadduct
3a,28 obtained with a high enantioselectivity, into the amino
alcohol 10 (Scheme 4). To this aim, isoxazolidine 3a-endo was
1
(200 or 400 MHz for H and 100 MHz for 13C) was measured in
CDCl3 unless otherwise mentioned, and chemical shifts and coupling
constants are presented in parts per million relative to Me4Si and
hertz, respectively. Abbreviations are as follows: s, singlet; d, doublet; t,
triplet; q, quartet; m, multiplet; br, broad. Proton and carbon
assignments were established using COSY, HMQC, HSQC, and
DEPT experiments. Relative stereochemistry of cycloadducts was
established using NOESY experiments. IR spectroscopy of oil and
solid samples were measured as neat liquid films and KBr pellets,
respectively. The wave numbers of maximum absorption peaks of IR
spectroscopy are presented in cm−1. MeNO2 and MTBE were distilled
from CaH2 before use. All reactions were performed under argon
atmosphere unless otherwise mentioned. Reagents were purchased
from commercial suppliers and used without purification. Reactions at
−10 and −20 °C were performed using a bath cooled by cryogenic
flow. High-resolution mass spectra were performed on a GC TOF
mass spectrometer. R,R-1 catalyst9f and nitrones 2a,17b 2b,17b and
2e15a were prepared according to the reported procedures. For HPLC
analysis, racemic samples of cycloadducts were prepared under the
microwave conditions as reported in our previous paper.15a
Scheme 4. Synthesis of 3-Amino-5-hydroxy Lactone 10
Typical Procedures for the Preparation of Nitrones. (E)-N-(1-
Methoxy-1-oxopentan-2-ylidene)-1-phenylmethanamine
Oxide (2c). A mixture of methyl 2-oxopentanoate (2.48.g, 19 mmol),
BnNHOH (2.57 g, 21 mmol), and CH2Cl2 (25 mL) was stirred at rt
for 36 h and then concentrated in vacuo. The residue was purified by
column chromatography (cyclohexane/EtOAc = 9:1) to give
compound 2c (2.91 g, 65% yield) as a pale yellow oil. IR: 3034,
reduced to afford alcohol 8, which was converted into the
bicyclic compound 9 after sequential saponification and
lactonization. Ring cleavage through N−O bond reduction
with Raney-Ni resulted in the protected amino alcohol 10
under conditions allowing concomitant N-Boc protection in
high yields (74% overall yield from adduct 3a-endo for the 3
steps).
1
2961, 2874, 1709, 1525, 1455, 1321, 1190, 1135, 734, 708. H NMR
(400 MHz, CDCl3): 7.47 (2H, dd, J = 7.6, 1.7 Hz, 2ArH), 7.36−7.31
(3H, m, 3ArH), 5.64 (2H, s, CH2), 3.82 (3H, s, CH3), 2.71−2.67 (2H,
CONCLUSION
m, CH2), 1.59−1.49 (2H, m, CH2), 0.94 (3H, t, J = 7.4 Hz, CH3). 13
C
■
In this work, a catalytic 1,3-dipolar cycloaddition between E-
configured ketonitrones containing an ester function as one of
their C-substituents and methacrolein was developed with total
3,5-regiocontrol and with high diastereo- and enantioselectiv-
ities, using the chiral ruthenium monocationic catalyst (R,R-1).
This mono-complexing Lewis acid was able to ensure the
enantioselective access to isoxazolidines possessing two
quaternary centers from functional ketonitrones through a
selective activation of the enal and, moreover, in a
diastereodivergent manner.
NMR (100 MHz, CDCl3): 14.1 (CH3), 18.3 (CH2), 30.4 (CH2), 52.6
(CH3), 67.6 (CH2), 128.4 (CH), 128.6 (CH × 2), 128.7 (CH × 2),
134.2 (C), 141.5 (C), 163.3 (CO). HRMS (ESI+): m/z calcd for
C13H17NO3 236.1281 [M + H]+, found 236.1273; 258.1101 [M +
Na]+, found 258.1098; 274.0840 [M + K]+, found 274.0835.
(E)-N-(1-(tert-Butoxy)-1-oxopropan-2-ylidene)-1-phenylme-
thanamine Oxide (2d). A mixture of tert-butyl pyruvate (1.1 mL, 10
mmol) and BnNHOH (1.76 g, 11 mmol) in CH2Cl2 (20 mL) was
stirred at rt for 16 h and then concentrated in vacuo. The residue was
dissolved in a mixture of CH2Cl2 (50 mL), and the organic phase was
washed with H2O (20 mL × 3), dried over MgSO4, and concentrated
in vacuo to give crude oil. Column chromatography (cyclohexane/
EtOAc = 9:1) gave compound 2d (0.895 g, 36% yield) as a colorless
oil. IR: 2978, 2933, 1705, 1531, 1456, 1369, 1308, 1133, 844. 1H NMR
(400 MHz, CDCl3): 7.48 (2H, dd, J = 6.8, 3.2 Hz, Ph), 7.33 (3H, dd, J
= 5.0, 2.0 Hz, Ph), 5.63 (2H, s, CH2), 2.21 (3H, s, CH3), 1.51 (9H, s,
t-Bu). 13C NMR (100 MHz, CDCl3): 15.8 (CH3), 28.0 (CH3 × 3),
66.8 (CH2), 83.2 (C), 128.4 (CH), 128.5 (CH × 2), 128.9 (CH × 2),
134.3 (C), 139.3 (C), 161.9 (CO). HRMS (ESI+): m/z calcd for
C14H19NO3 250.1438 [M + H]+, found 250.1438; 272.1257 [M +
Na]+, found 272.1256.
Indeed, the Ru-catalyzed 1,3-dipolar cycloadditions described
herein displayed an unreported solvent effect on the
diastereoselectivity that allows us to propose a novel vision
on the mode of action for the catalyst (R,R-1), which behaves
as a frustrated Lewis pair. From the DFT calculations
performed, the solvent effect could be rationalized as the
consequence of the competition occurring between the
counterion of the catalyst and the solvent in the close-shell of
complexed methacrolein. Moreover, the invariable facial
selectivity ensured by the catalytic system could also be
elucidated. By changing the solvent and the N- and O-
substituents on the nitrone, the selectivity of the reaction could
be controlled to give either endo or exo product, with good to
excellent enantioselectivities. Finally, a representative [3 + 2]
cycloadduct could be easily transformed into a highly
enantioenriched lactone (e.g., 10) and lactam (e.g., 7) featuring
two quaternary centers, as first insights to the synthetic
potential displayed by the class of heterocycles that are
obtained from this methodology.
(E)-N-(1-Methoxy-1-oxopropan-2-ylidene)methanamine
Oxide (2f). A mixture of methyl pyruvate (2.3 mL, 20 mmol),
MeNHOH·HCl (1.84 g, 22 mmol), NaOAc·3H2O (3.26 g, 24 mmol),
and MeOH (30 mL) was stirred at rt for 14 h and then concentrated
in vacuo. The residue was dissolved in a mixture of CH2Cl2 (50 mL),
and the organic phase was washed with H2O (25 mL), dried over
MgSO4, and concentrated in vacuo to give crude oil. Column
chromatography (cyclohexane/EtOAc = 3:1) gave compound 2f (2.52
g, 96% yield) as a colorless oil at rt that solidified to a white wax in
fridge. IR: 2957, 1713, 1543, 1439, 1309, 1194, 1140, 1034, 806, 751.
1H NMR (400 MHz, CDCl3): 4.18 (3H, q, J = 1.3 Hz, CH3), 3.84
(3H, s, CH3), 2.24 (3H, q, J = 1.3 Hz, CH3). 13C NMR (100 MHz,
CDCl3): 14.9 (CH3), 52.6 (CH3), 53.4 (CH3), 138.0 (C), 163.0 (C
O). HRMS (ESI+): m/z calcd for C5H9NO3 132.0655 [M + H]+,
found 132.0661; 154.0475 [M + Na]+, found 154.0477.
EXPERIMENTAL SECTION
General Information. All melting points are uncorrected. Column
chromatography was performed using 60 μm silica gel. Thin-layer
■
3420
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
(E)-N-(1-Ethoxy-1-oxopropan-2-ylidene)methanamine
Oxide (2g). A mixture of ethyl pyruvate (2.2 mL, 20 mmol),
MeNHOH·HCl (1.84 g, 22 mmol), NaOAc·3H2O (3.26 g, 24 mmol),
and MeOH (30 mL) was stirred at rt for 14 h and then concentrated
in vacuo. The residue was dissolved in a mixture of CH2Cl2 (50 mL),
and the organic phase was washed with H2O (25 mL), dried over
MgSO4, and concentrated in vacuo to give crude oil. Column
chromatography (cyclohexane/EtOAc = 3:1) gave compound 2g (2.60
g, 90% yield) as a colorless oil. IR: 2983, 2940, 1713, 1542, 1445, 1369,
1304, 1143, 1024, 750. 1H NMR (400 MHz, CDCl3): 4.29 (2H, q, J =
7.1 Hz, CH2), 4.17 (3H, q, J = 1.1 Hz, CH3), 2.25 (3H, q, J = 1.1 Hz,
CH3), 1.36 (3H, t, J = 7.1, Hz, CH3). 13C NMR (100 MHz, CDCl3):
14.1 (CH3), 15.0 (CH3), 53.1 (CH3), 61.8 (CH2), 138.4 (C), 162.6
(CO). HRMS (ESI+): m/z calcd for C6H11NO3 146.0812 [M +
H]+, found 146.0808.
cooled to −10 °C and stirred for 5 min before the addition of
methacrolein (0.04 mL, 0.5 mmol). The reaction mixture was
quenched after 7 days by adding acetone (0.5 mL) at −10 °C and
stirred at room temperature for 20 min before the addition of pentane
(10 mL) to precipitate the catalyst. Filtration over Celite, washing with
pentane, and concentration in vacuo gave a crude oil. Conversion and
1
diastereoselectivity were determined by H NMR of the crude to be
94% and endo/exo 99:1, respectively. Column chromatography
(cyclohexane/EtOAc = 19:1 to 15:1) gave 66.4 mg, 91% total yield
and pure 3b-endo adduct was isolated in 59.0 mg, 81% yield as a pale
yellow oil. [α]20 = −3.7 (c 1.11, CHCl3) with 84% ee for pure 3b-
D
endo. Rf = 0.46 (cyclohexane/EtOAc = 6:1). IR: 3031, 2973, 2881,
1
1731, 1497, 1455, 1304, 1241, 1157, 1073, 739, 698. H NMR (400
MHz, CDCl3): 9.51 (s, β-CHO, 1H), 7.37 (dd, J = 6.9, 1.4 Hz, Harom,
2H), 7.32 (t, J = 6.9 Hz, Harom, 2H), 7.25 (tt, J = 6.9, 1.4 Hz, Harom,
1H), 4.05 (d, J = 14.4 Hz, CH2Ph, 1H), 4.02 (d, J = 14.4 Hz, CH2Ph,
1H), 3.74 (s, ester Me, 3H), 3.18 (d, J = 13.0 Hz, 4β- CH2, 1H), 2.08
(dq, J = 14.8, 7.4 Hz, 3α-CH2, 1H), 2.02 (d, J = 13.0 Hz, 4α-CH2, 1H),
1.74 (dq, J = 14.8, 7.4 Hz, 3α-CH2, 1H), 1.30 (s, 5α-Me, 3H), 0.96 (t,
J = 7.4 Hz, 3α-Me, 3H). 13C NMR (100 MHz, CDCl3): 9.4 (3-CH3),
20.7 (5-CH3), 25.5 (3-CH2), 46.1 (4-CH2), 52.1 (ester CH3), 55.1
(CH2Ph), 74.0 (3-C), 83.9 (5-C), 127.1 (CH), 128.0 (CH × 2), 128.3
(CH × 2), 138.3 (C), 171.4 (CO), 204.1 (CHO). HRMS (DCI
+NH3CH4): m/z calcd for C16H21NO4 292.1543 [M + H]+, found
292.1543. Enantioselectivity was determined by chiral HPLC analysis
after reduction of 3b-endo with NaBH(OAc)3 to the corresponding
alcohol 8b-endo.
Typical Procedures for the Asymmetric 1,3-Dipolar Cyclo-
addition. (+)-(3R,5S)-Methyl 2-Benzyl-5-formyl-3,5-dimethyli-
soxazolidine-3-carboxylate (3a-endo) (Table 1, entry11). In a 10
mL tube, (R,R-1) (35.0 mg, 0.025 mmol) and methacrolein (0.04 mL,
0.5 mmol) were charged and suspended in MTBE (0.7 mL). The
mixture was cooled to −10 °C and stirred for 5 min before the
addition of 2a (52 mg, 0.25 mmol) in MTBE (0.5 mL) via canula. The
reaction mixture was stirred at −10 °C and quenched after 6 days by
adding acetone (0.5 mL) at −10 °C and stirred at room temperature
for 2 min before the addition of pentane (10 mL) to precipitate the
catalyst. Filtration over Celite, washing with pentane, and concen-
tration in vacuo gave a crude oil. Conversion and diastereoselectivity
1
were determined by H NMR of the crude to be 96% and endo/exo
(−)-(3R,5S)-Methyl 2-Benzyl-5-formyl-5-methyl-3-propyli-
soxazolidine-3-carboxylate (3c-endo) (Scheme 1). In a 10 mL
tube, 2c (59.0 mg, 0.25 mmol) and (R,R-1) (35.0 mg, 0.025 mmol)
were charged and suspended in MTBE (0.6 mL). The mixture was
cooled to −10 °C and stirred for 5 min before the addition of
methacrolein (0.04 mL, 0.5 mmol). The reaction mixture was
quenched after 6 days by adding acetone (0.5 mL) at −10 °C and
stirred at room temperature for 2 min before the addition of pentane
(10 mL) to precipitate the catalyst. Filtration over Celite, washing with
pentane, and concentration in vacuo gave a crude oil. Conversion and
98:2, respectively. Column chromatography (cyclohexane/EtOAc =
19:1 to 9:1) gave 66.7 mg, 96% total yield and pure 3a-endo adduct
was isolated in 62.6 mg, 90% yield as a pale yellow oil. [α]25D = +84.9
(c 1.01, CHCl3) with 88% ee for pure 3a-endo. Rf = 0.55 (cyclohexane/
EtOAc = 3:1). IR: 2982, 2953, 1736, 1497, 145.5, 1265, 1152, 1076.
1H NMR (400 MHz, CDCl3): 9.47 (s, β-CHO, 1H), 7.40 (dd, J = 7.1,
1.4 Hz, Harom, 2H), 7.34−7.30 (tt, J = 7.1, 1.4 Hz, Harom, 2H), 7.26
(tt, J = 6.1, 1.4 Hz, Harom, 1H), 4.04 (d, J = 14.4 Hz, CH2Ph, 1H),
3.89 (d, J = 14.4 Hz, CH2Ph, 1H), 3.71 (s, ester Me, 3H), 3.19 (d, J =
13.0 Hz, 4β-CH2, 1H), 1.94 (d, J = 13.0 Hz, 4α-CH2, 1H), 1.50 (s, 3α-
Me, 3H), 1.27 (s, 5α-Me, 3H). 13C NMR (100 MHz, CDCl3): 18.5 (3-
CH3), 19.5 (5-CH3), 49.8 (4-CH2), 52.4 (ester CH3), 55.2 (CH2Ph),
69.1 (3-C), 83.7 (5-C), 127.2 (CH), 128.19 (CH × 2), 128.23 (CH ×
2), 138.0 (C), 172.5 (CO), 204.7 (CHO). HRMS (FIID): m/z
calcd for C15H19NO4 277.1314 [M]+, found 277.1325. Enantiose-
lectivity was determined by chiral HPLC analysis after reduction of 3a-
endo with NaBH(OAc)3 to the corresponding alcohol 8a-endo.
(+)-(3R,5R)-Methyl 2-Benzyl-5-formyl-3,5-dimethylisoxazoli-
dine-3-carboxylate (3a-exo) (Table 1, entry 6). In toluene, 0.5
mmol scale, 0.4 M. Conversion and diastereoselectivity were
determined by 1H NMR of the crude to be 84% and endo/exo
61:39, respectively. Column chromatography (cyclohexane/EtOAc =
1
diastereoselectivity were determined by H NMR of the crude to be
96% and endo/exo 99:1, respectively. Column chromatography
(petroleum ether/EtOAc = 19:1 to 9:1) gave 68.6 mg, 90% total
yield with 86% ee as a colorless oil. [α]20D = −8.8 (c 1.32, CHCl3) for
pure 3c-endo. Rf = 0.41 (cyclohexane/EtOAc = 6:1). IR: 3031, 2973,
1
2881, 1731, 1497, 1455, 1304, 1241, 1157, 1073, 739, 698. H NMR
(400 MHz, CDCl3): 9.50 (s, β-CHO, 1H), 7.37 (dd, J = 6.9, 1.4 Hz,
Harom, 2H), 7.32 (t, J = 6.9 Hz, Harom, 2H), 7.26 (tt, J = 6.9, 1.4 Hz,
Harom, 1H), 4.04 (s, CH2Ph, 2H), 3.74 (s, ester Me, 3H), 3.18 (d, J =
13.0 Hz, 4β- CH2, 1H), 2.02 (d, J = 13.0 Hz, 4α-CH2, 1H), 2.00 (dt, J
= 12.6, 4.7 Hz, 3α-CH2, 1H), 1.66 (dt, J = 12.6, 4.7 Hz, 3α-CH2, 1H),
1.42−1.27 (m, 3α-CH2, 2H), 1.30 (s, 5α-Me, 3H), 0.98 (t, J = 7.3 Hz,
3α-Me, 3H). 13C NMR (100 MHz, CDCl3): 14.5 (3-CH3, 3-CH2),
18.5 (5-CH3), 34.9 (3-CH2), 46.6 (4-CH2), 52.1 (ester CH3), 55.1
(CH2Ph), 73.4 (3-C), 84.0 (5-C), 127.1 (CH), 128.0 (CH × 2), 128.3
(CH × 2), 138.3 (C), 171.5 (CO), 204.2 (CHO). HRMS (DCI+):
m/z calcd for C17H23NO4 306.1700 [M + H]+, found 306.1715.
Enantioselectivity was determined by chiral HPLC analysis after
reduction of 3c-endo with NaBH(OAc)3 to the corresponding alcohol
8c-endo.
19:1 to 9:1) gave 103 mg, 74% total yield as a colorless oil. [α]20
=
D
+131.3 (c 1.05, CHCl3) with 68% ee for exo and 74% ee for endo in a
major 3a-exo mixture (61:39). Rf = 0.61 (cyclohexane/EtOAc = 3:1).
IR: 3032, 2984, 2952, 2808, 1732, 1497, 1455, 1374, 1250, 1165, 1069,
740, 699. 1H NMR (400 MHz, CDCl3): 9.41 (s, α-CHO, 1H), 7.36−
7.29 (m, Harom, 4H), 7.26 (tt, J = 6.1, 1.7 Hz, Harom, 1H), 4.02 (d, J
= 14.7 Hz, CH2Ph, 1H), 3.82 (s, ester Me, 3H), 3.71 (d, J = 14.7 Hz,
CH2Ph, 1H), 2.68 (d, J = 12.7 Hz, 4β- CH2, 1H), 2.50 (d, J = 12.7 Hz,
4α-CH2, 1H), 1.40 (s, 3α-Me, 3H), 1.30 (s, 5β-Me, 3H). 13C NMR
(100 MHz, CDCl3): 17.9 (5-CH3), 21.0 (3-CH3), 47.4 (4-CH2), 52.0
(ester CH3), 55.3 (CH2Ph), 69.6 (3-C), 83.7 (5-C), 127.2 (CH),
128.1 (CH × 2), 128.3 (CH × 2), 137.9 (C), 172.4 (CO), 206.1
(CHO). HRMS (FIID): m/z calcd for C15H19NO4 277.1314 [M]+,
found 277.1320. Enantioselectivity was determined by chiral HPLC
analysis after reduction of 3a-exo with NaBH(OAc)3 to the
corresponding alcohol 8a-exo.
(+)-(3R,5S)-tert-Butyl 2-Benzyl-5-formyl-3,5-dimethylisoxa-
zolidine-3-carboxylate (3d-endo) (Table 3, entry 1). In a 10
mL tube, 2d (62.3 mg, 0.25 mmol) and (R,R-1) (35.0 mg, 0.025
mmol) were charged and suspended in MTBE (1.2 mL). The mixture
was cooled to −10 °C and stirred for 5 min before the addition of
methacrolein (0.04 mL, 0.5 mmol). The reaction mixture was
quenched after 6 days by adding acetone (0.5 mL) at −10 °C and
stirred at room temperature for 2 min before the addition of pentane
(10 mL) to precipitate the catalyst. Filtration over Celite, washing with
pentane, and concentration in vacuo gave a crude oil. Conversion and
(−)-(3R,5S)-Methyl 2-Benzyl-3-ethyl-5-formyl-5-methylisox-
azolidine-3-carboxylate (3b-endo) (Scheme 1). In a 10 mL
tube, 2b (55.4 mg, 0.25 mmol) and (R,R-1) (35.0 mg, 0.025 mmol)
were charged and suspended in MTBE (1.2 mL). The mixture was
1
diastereoselectivity were determined by H NMR of the crude to be
100% and endo/exo 92:8, respectively. Column chromatography
(cyclohexane/EtOAc = 19:1) gave 79 mg, 99% total yield and pure
3421
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
3d-endo adduct was isolated in 62.8 mg, 79% yield as a white solid, mp
47−48 °C: [α]20D = +97.5 (c 0.63, CH2Cl2) with 92% ee for pure 3d-
endo. Rf = 0.62 (cyclohexane/EtOAc = 6:1). IR: 3032, 2978, 2932,
C), 83.6 (5-C), 127.1 (CH), 128.0 (CH × 2), 128.2 (CH × 2), 138.0
(C), 171.8 (CO), 206.2 (CHO). HRMS (CI + NH3CH4): m/z
calcd for C16H21NO4 292.1543 [M + H]+, found 292.1543.
Enantioselectivity was determined by chiral HPLC analysis after
reduction of 3e-exo with NaBH(OAc)3 to the corresponding alcohol
8e-exo.
1
1732, 1497, 1456, 1370, 1309, 1256, 1148, 735. H NMR (400 MHz,
CDCl3): 9.46 (s, β-CHO, 1H), 7.40 (d, J = 7.0 Hz, Harom, 2H), 7.33
(t, J = 7.0 Hz, Harom, 2H), 7.26 (m, Harom, 1H), 4.05 (d, J = 14.5
Hz, CH2Ph, 1H), 3.85 (d, J = 14.5 Hz, CH2Ph, 1H), 3.18 (d, J = 12.8
Hz, 4β- CH2, 1H), 1.88 (d, J = 12.8 Hz, 4α-CH2, 1H), 1.46 (s, 3α-Me
and t-Bu, 12H), 1.25 (s, 5α-Me, 3H). 13C NMR (100 MHz, CDCl3):
18.3 (3-CH3), 19.3 (5-CH3), 27.9 (ester t-Bu), 50.0 (CH2), 55.1
(CH2Ph), 69.6 (3-C), 81.9 (t-Bu C), 83.7 (5-C), 127.1 (CH), 128.1
(CH × 2), 128.3 (CH × 2), 138.4 (C), 171.2 (CO), 205.0 (CHO).
HRMS (CI + NH3CH4): m/z calcd for C18H25NO4 320.1856 [M +
H]+, found 320.1868. Enantioselectivity was determined by chiral
HPLC analysis after reduction of 3d-endo with NaBH(OAc)3 to the
corresponding alcohol 8d-endo.
(+)-(3R,5S)-Methyl 5-Formyl-2,3,5-trimethylisoxazolidine-3-
carboxylate (3f-endo) (Table 3, entry 5). In a 10 mL tube, 2f
(33 mg, 0.25 mmol) and (R,R-1) (35.0 mg, 0.025 mmol) were charged
and suspended in MTBE (1.2 mL). The mixture was cooled to −10
°C and stirred for 5 min before the addition of methacrolein (0.04 mL,
0.5 mmol). The reaction mixture was quenched after 6 days by adding
acetone (0.5 mL) at −10 °C and stirred at room temperature for 2
min before the addition of pentane (10 mL) to precipitate the catalyst.
Filtration over Celite, washing with pentane, and concentration in
vacuo gave a crude oil. Conversion and diastereoselectivity were
1
determined by H NMR of the crude to be 95% and endo/exo 68:32,
(+)-(3R,5S)-Ethyl 2-Benzyl-5-formyl-3,5-dimethylisoxazoli-
dine-3-carboxylate (3e-endo) (Table 3, entry 2). In a 10 mL
tube, 2e (55.3 mg, 0.25 mmol) and (R,R-1) (35.0 mg, 0.025 mmol)
were charged and suspended in MTBE (1.2 mL). The mixture was
cooled to −10 °C and stirred for 5 min before the addition of
methacrolein (0.04 mL, 0.5 mmol). The reaction mixture was
quenched after 5 days by adding acetone (0.5 mL) at −10 °C and
stirred at room temperature for 10 min before the addition of pentane
(10 mL) to precipitate the catalyst. Filtration over Celite, washing with
pentane, and concentration in vacuo gave a crude oil. Conversion and
respectively. Column chromatography (cyclohexane/EtOAc = 19:1 to
6:1) gave 44.6 mg, 89% total yield as a pale yellow oil. [α]20D = +124.9
(c 0.55, CHCl3) for a mixture of endo /exo (92:8) with 80% ee for 3f-
endo. Rf = 0.56 (cyclohexane/EtOAc = 3:1). IR: 2956, 1737, 1640,
1
1519, 1475, 1375, 1291, 1193, 1156, 1091, 982. H NMR (400 MHz,
CDCl3): 9.63 (s, β-CHO, 1H), 3.71 (s, ester Me, 3H), 3.17 (d, J =
13.0 Hz, 4β- CH2, 1H), 2.70 (s, N-Me, 3H), 1.92 (d, J = 13.0 Hz, 4α-
CH2, 1H), 1.39 (s, 3α-Me, 3H), 1.31 (s, 5α-Me, 3H). 13C NMR (100
MHz, CDCl3): 17.9 (3-CH3), 19.9 (5-CH3), 38.2 (N−CH3), 49.4
(CH2), 52.4 (ester CH3), 69.5 (3-C), 83.7 (5-C), 172.4 (CO),
204.2 (CHO). HRMS (DCI + NH3CH4): m/z calcd for C9H15NO4
202.1074 [M + H]+, found 202.1083. Enantioselectivity was
determined by chiral HPLC analysis after reduction of 3f-endo with
NaBH(OAc)3 to the corresponding alcohol 8f-endo.
(3R,5R)-Methyl 5-Formyl-2,3,5-trimethylisoxazolidine-3-car-
boxylate (3f-exo). Rf = 0.64 (cyclohexane/EtOAc = 3:1). IR: 2954,
2887, 1732, 1447, 1376, 1289, 1255, 1192, 1169, 1094. 1H NMR (400
MHz, CDCl3): 9.60 (s, α-CHO, 1H), 3.76 (s, ester Me, 3H), 2.64 (d, J
= 12.8 Hz, 4β-CH2, 1H), 2.62 (s, N-Me, 3H), 2.44 (d, J = 12.8 Hz, 4α-
CH2, 1H), 1.33 (s, 5β-Me, 3H), 1.29 (s, 3αβ-Me, 3H). 13C NMR (100
MHz, CDCl3): 18.1 (3-CH3), 20.7 (5-CH3), 38.5 (N−CH3), 47.0
(CH2), 51.9 (ester CH3), 70.3 (3-C), 84.0 (5-C), 172.3 (CO),
204.6 (CHO). HRMS (CI + NH3CH4): m/z calcd for C9H15NO4:
202.1074 [M + H]+, found 202.1088.
1
diastereoselectivity was determined by H NMR of the crude to be
43% and endo/exo 63:37, respectively. Column chromatography
(cyclohexane/EtOAc = 19:1 to 9:1) gave 30.7 mg, 42% total yield
as a pale yellow oil. [α]20 = +72.6 (c 0.49, CH2Cl2) with 84% ee for
D
pure 3e-endo. Rf = 0.40 (cyclohexane/EtOAc = 6:1). IR: 2974, 2882,
1
1731, 1497, 1456, 1373, 1241, 1156, 1007, 738. H NMR (400 MHz,
CDCl3): 9.47 (s, β-CHO, 1H), 7.40 (dd, J = 6.9, 2.4 Hz, Harom, 2H),
7.34−7.30 (tt, J = 6.9, 1.4 Hz, Harom, 2H), 7.26 (tt, J = 6.9, 1.4 Hz,
Harom, 1H), 4.17 (q, J = 7.1 Hz, ester CH2, 2H), 4.05 (d, J = 14.4 Hz,
CH2Ph, 1H), 3.88 (d, J = 14.4 Hz, CH2Ph, 1H), 3.19 (d, J = 13.0 Hz,
4β- CH2, 1H), 1.93 (d, J = 13.0 Hz, 4α-CH2, 1H), 1.50 (s, 3α-Me,
3H), 1.27 (t, J = 7.1 Hz, ester Me, 1H), 1.267 (s, 5α-Me, 3H). 13C
NMR (100 MHz, CDCl3): 14.1 (5-CH3), 18.5 (3-CH3), 19.5 (ester
CH3), 49.8 (4-CH2), 55.2 (CH2Ph), 61.4 (ester CH2), 69.1 (3-C),
83.7 (5-C), 127.2 (CH), 128.19 (CH × 2), 128.25 (CH × 2), 138.1
(C), 172.0 (CO), 204.8 (CHO). HRMS (CI + NH3CH4): m/z
calcd for C16H21NO4 292.1543 [M + H]+, found 292.1557.
Enantioselectivity was determined by chiral HPLC analysis after
reduction of 3e-endo with NaBH(OAc)3 to the corresponding alcohol
8e-endo.
(+)-(3R,5R)-Ethyl 5-Formyl-2,3,5-trimethylisoxazolidine-3-
carboxylate (3g-exo) (Table 3, entry 8). In a 10 mL tube, 2g
(36.3 mg, 0.25 mmol) and (R,R-1) (35.0 mg, 0.025 mmol) were
charged and suspended in CH2Cl2 (1.2 mL). The mixture was cooled
to 0 °C and stirred for 5 min before the addition of methacrolein (0.04
mL, 0.5 mmol). The reaction mixture was quenched after 5.5 days by
adding acetone (0.5 mL) at 0 °C and stirred at room temperature for 2
min before the addition of pentane (10 mL) to precipitate the catalyst.
Filtration over Celite, washing with pentane, and concentration in
vacuo gave a crude oil. Conversion and diastereoselectivity were
(+)-(3R,5R)-Ethyl 2-Benzyl-5-formyl-3,5-dimethylisoxazoli-
dine-3-carboxylate (3e-exo) (Table 3, entry 3). In a 10 mL
tube, 2e (55.3 mg, 0.25 mmol) and (R,R-1) (35.0 mg, 0.025 mmol)
were charged and suspended in toluene (1.2 mL). The mixture was
cooled to 0 °C and stirred for 5 min before the addition of
methacrolein (0.04 mL, 0.5 mmol). The reaction mixture was
quenched after 5.5 days by adding acetone (0.5 mL) at −10 °C and
stirred at room temperature for 2 min before the addition of pentane
(10 mL) to precipitate the catalyst. Filtration over Celite, washing with
pentane, and concentration in vacuo gave a crude oil. Conversion and
1
determined by H NMR of the crude to be 100% and endo/exo 7:93,
respectively. Column chromatography (cyclohexane/EtOAc = 19:1 to
9:1) gave 46 mg, 86% total yield and pure 3g-exo adduct was isolated
in 41.3 mg, 77% yield as a pale yellow oil. [α]20 = +166.8 (c 0.71,
D
CH2Cl2) with 81% ee for pure 3g-exo. Rf = 0.38 (cyclohexane/EtOAc
= 4:1). IR: 2984, 2939, 1728, 1447, 1376, 1254, 1179, 1094, 1021. 1H
NMR (400 MHz, CDCl3): 9.61 (s, α-CHO, 1H), 4.23 (q, J = 7.1 Hz,
ester CH2, 2H), 2.65 (d, J = 12.8 Hz, 4β- CH2, 1H), 2.64 (s, N-Me,
3H), 2.44 (d, J = 12.8 Hz, 4α-CH2, 1H), 1.35 (s, 5β-Me, 3H), 1.32 (t, J
= 7.1 Hz, ester Me, 3H), 1.30 (s, 3α-Me, 3H). 13C NMR (100 MHz,
CDCl3): 14.3 (ester CH3), 18.0 (5-CH3), 20.6 (3-CH3), 38.4 (N−
CH3), 46.9 (CH2), 61.0 (ester CH2), 70.0 (3-C), 83.8 (5-C), 171.7
(CO), 204.7 (CHO). HRMS (DCI + NH3CH4): m/z calcd for
C10H17NO4 216.1230 [M + H]+, found 216.1237. Enantioselectivity
was determined by chiral HPLC analysis after reduction of 3g-exo with
NaBH4 to the corresponding alcohol 8g-exo.
1
diastereoselectivity were determined by H NMR of the crude to be
86% and endo/exo 14:86, respectively. Column chromatography
(cyclohexane/EtOAc = 19:1 to 9:1) gave 62 mg, 85% total yield
and pure 3e-exo adduct was isolated in 47.0 mg, 65% yield as a
colorless oil. [α]20 = +159.2 (c 0. 97, CHCl3) with 66% ee for pure
D
3e-exo. Rf = 0.44 (cyclohexane/EtOAc = 6:1). IR: 2982, 2936, 2806,
1
1728, 1497, 1455, 1374, 1288, 1250, 1175, 1068, 739, 699. H NMR
(400 MHz, CDCl3): 9.42 (s, α-CHO, 1H), 7.37−7.30 (m, Harom,
4H), 7.26 (m, Harom, 1H), 4.29 (q, J = 7.1 Hz, ester CH2, 2H), 4.04
(d, J = 14.7 Hz, CH2Ph, 1H), 3.74 (d, J = 14.7 Hz, CH2Ph, 1H), 2.69
(d, J = 12.7 Hz, 4β- CH2, 1H), 2.49 (d, J = 12.7 Hz, 4α-CH2, 1H), 1.40
(s, 3α-Me, 3H), 1.36 (t, J = 7.1 Hz, ester Me, 1H), 1.31 (s, 5β-Me,
3H). 13C NMR (100 MHz, CDCl3): 14.3 (ester CH3), 17.9 (5-CH3),
20.9 (3-CH3), 47.3 (4-CH2), 55.1 (CH2Ph), 61.1 (ester CH2), 69.4 (3-
(−)-(3R,5S)-Methyl 2-Benzyl-3,5-dimethyl-5-((((S)-1-
phenylethyl)amino)methyl)isoxazolidine-3-carboxylate (6).
To a mixture of 3a (33.0 mg, 0.120 mmol, endo/exo = 97:3, 87%
3422
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
ee), Na2SO4 (20 mg) in CH2Cl2 (1.0 mL) was added a solution of (S)-
α-(−)-methyl benzylamine (14.6 mg, 0.120 mmol) in CH2Cl2 (1.0
mL) at room temperature. The reaction mixture was stirred for 20 h,
then NaBH(OAc)3 (102 mg, 0.480 mmol) and glacial acetic acid (29
mg, 0.480 mmol) were successively added. After stirring at room
temperature for 4 h, the reaction mixture was diluted with CH2Cl2 (5
mL) and quenched with H2O (1 mL) at 0 °C. The mixture was stirred
at room temperature for 10 min, then the aqueous layer was extracted
with CH2Cl2 (10 mL × 3) and the combined organic layers were dried
over Na2SO4, filtered, and concentrated in vacuo. The resulting crude
was purified by column chromatography (cyclohexane/EtOAc, 3:1 to
1:1) to afford compound 6 (38 mg, 83%, dr 97.5:2.5) as a colorless oil.
[α]25D = −18.6 (c 1.02, CHCl3) along with minor isomer (6 mg, 13%)
in 96% global yield. IR: 2953, 2927, 1738, 1495, 1453, 1267, 1128,
1028. 1H NMR (400 MHz, CDCl3): 7.40 (d, J = 7.3 Hz, Harom, 2H),
7.32 (t, J = 7.3 Hz, Harom, 2H), 7.28−7.24 (m, Harom, 3H), 7.23−
7.17 (m, Harom, 3H), 4.00 (d, J = 14.4 Hz, CH2Ph, 1H), 3.83 (d, J =
14.4 Hz, CH2Ph, 1H), 3.69 (s, Me ester, 3H), 3.67 (q, J = 6.6 Hz, 5-
CHPh, 1H), 3.10 (br s, NH, 1H), 2.86 (d, J = 11.5 Hz, 5β-CH2, 1H),
2.74 (d, J = 12.7 Hz, 4β-CH2, 1H), 2.22 (d, J = 11.5 Hz, 5α-CH2, 1H),
1.87 (d, J = 12.7 Hz, 4α-CH2, 1H), 1.46 (s, 3α-Me, 3H), 1.30 (s, 5α-
Me, 3H), 1.19 (d, J = 6.6 Hz, Me, 3H). 13C NMR (100 MHz, CDCl3):
19.1 (3-CH3), 24.1 (5-CH3), 24.9 (CH3), 51.3 (4-CH2), 52.0 (ester
CH3), 54.7 (5-CH2), 55.2 (CH2Ph), 58.5 (CHPh), 69.3 (3-C), 79.7
(5-C), 126.6 (CH), 126.7 (CH × 2), 126.8 (CH), 128.09 (CH × 2),
128.13 (CH × 2), 128.2 (CH × 2), 139.0 (C), 146.1 (C), 173.5 (C
O). HRMS (ESI+): m/z calcd for C23H30N2O3 383.2329 [M + H]+,
found 383.2312.
layers were washed with brine (10 mL), dried over MgSO4, filtered,
and concentrated in vacuo to give a crude oil. Column chromatography
(cyclohexane/EtOAc = 6:1 to 2:1) gave 31.2 mg, 78% total yield with
91% ee as a colorless oil. [α]20D = +45.9 (c 0.59, CHCl3) for pure 8a-
endo. IR: 3429, 2978, 2952, 1732, 1497, 1455, 1376, 1271, 1195, 1147,
1051, 739. 1H NMR (400 MHz, CDCl3): 7.38 (d, J = 7.1 Hz, Harom,
2H), 7.34−7.29 (tt, J = 7.1, 1.4 Hz, Harom, 2H), 7.24 (tt, J = 7.2, 1.4
Hz, Harom, 1H), 3.98 (d, J = 14.4 Hz, CH2Ph, 1H), 3.82 (d, J = 14.4
Hz, CH2Ph, 1H), 3.74 (s, ester Me, 3H), 3.52 (d, J = 3.5 Hz, CH2,
2H), 3.26 (s, OH, 1H), 2.96 (d, J = 12.7 Hz, 4β-CH2, 1H), 1.95 (d, J =
12.7 Hz, 4α-CH2, 1H), 1.50 (s, 3α-Me, 3H), 1.25 (s, 5α-Me, 3H). 13
C
NMR (100 MHz, CDCl3): 19.0 (3-CH3), 22.7 (5-CH3), 49.0 (4-
CH2), 52.2 (ester CH3), 54.9 (CH2Ph), 69.1 (5-CH2), 69.8 (3-C),
80.2 (5-C), 127.1 (CH), 128.0 (CH × 2), 128.3 (CH × 2), 138.1 (C),
173.2 (CO). HRMS (FI/FD): m/z calcd for C15H21NO4 279.1471
[M]+, found 279.1477. Enantioselectivity was determined by chiral
HPLC analysis, Chiralpak As-H, isooctane/i-PrOH = 99:1, 1 mL/min,
254 nm; tR (min): 25.3 (minor) and 32.9 (major) with 91% ee for 8a-
endo.
(3R,5R)-Methyl 2-Benzyl-5-(hydroxymethyl)-3,5-dimethyli-
soxazolidine-3-carboxylate (8a-exo). The same procedure as in
8a-endo. Colorless oil, IR: 3431, 2978, 2938, 1731, 1497, 1455, 1375,
1
1272, 1167, 1055, 738. H NMR (400 MHz, CDCl3): 7.36−7.30 (m,
Harom, 4H), 7.24 (m, Harom, 1H), 4.02 (d, J = 14.6 Hz, CH2Ph, 1H),
3.81 (s, ester Me, 3H), 3.65 (d, J = 14.6 Hz, CH2Ph, 1H), 3.56 (d, J =
11.0 Hz, 5β-CH2, 1H), 3.47 (d, J = 11.0 Hz, 5α-CH2, 1H), 2.86 (br s,
OH, 1H), 2.67 (d, J = 12.5 Hz, 4β-CH2, 1H), 2.23 (d, J = 12.5 Hz, 4α-
CH2, 1H), 1.48 (s, 3α-Me, 3H), 1.29 (s, 5β-Me, 3H). 13C NMR (100
MHz, CDCl3): 20.7, 20.8 (3-CH3, 5-CH3), 47.6 (4-CH2), 51.9 (ester
CH3), 55.4 (CH2Ph), 70.3 (3-C and 5-CH2), 80.2 (5-C), 127.2 (CH),
128.0 (CH × 2), 128.4 (CH × 2), 138.1 (C), 172.7 (CO). HRMS
(ESI+): m/z calcd for C15H21NO4 280.1543 [M + H]+, found
280.1542, 302.1363 [M + Na]+, found 302.1367, 318.1102 [M + K]+,
found 318.1107. Enantioselectivity for entry was determined by chiral
HPLC analysis, Chiralpak As-H, isooctane/i-PrOH = 99:1, 1 mL/min,
254 nm; tR (min): 19.1 (minor) and 23.4 (major) with 68% ee for 8a-
exo.
(−)-(1R,5S)-7-Benzyl-1,5-dimethyl-3-[(S)-1-phenylethyl]-6-
oxa-3,7-diazabicyclo[3.2.1]octan-2-one (7). In a 10 mL vial, to a
solution of 6 (84 mg, 0.22 mmol, dr 88:12) in a mixture of THF/water
(2:1, 3.0 mL) was added LiOH (11 mg, 0.44 mmol) at room
temperature. The mixture was stirred for 5 h at room temperature
before concentration in vacuo, then quenched by adding an aqueous
solution of citric acid (160 mg in 5 mL), and stirred at room
temperature for 10 min. Extraction of the aqueous layer with CH2Cl2
(15 mL × 4), followed by drying the combined organic layers with
MgSO4, filtration, and concentration in vacuo afforded the crude
carboxylic acid. The crude was dissolved in CH2Cl2 (3 mL) and
treated with DCC (68.0 mg, 0.33 mmol) and DMAP (13.5 mg, 0.11
mmol) at room temperature, and the reaction mixture was stirred for
14 h. The mixture was concentrated in vacuo and treated with Et2O to
precipitate the urea derivatives. After filtration, washing with Et2O, and
evaporation in vacuo, the resulting crude was purified by column
chromatography (petroleum ether/EtOAc = 5:1 to 3:1) to afford
compound 7 (53 mg, 69% yield) as a diastereomerically pure white
(−)-(3R,5S)-Methyl 2-Benzyl-3-ethyl-5-(hydroxymethyl)-5-
methylisoxazolidine-3-carboxylate (8b-endo). The same proce-
dure as in 8a-endo to give 20.0 mg, 96% yield as a colorless oil. [α]20
D
= −43.9 (c 0.78, CHCl3) with 84% ee for pure 8b-endo. IR: 3465,
2969, 2875, 1728, 1497, 1455, 1307, 1239, 1157, 1053, 735. 1H NMR
(200 MHz, CDCl3): 7.39−7.20 (m, Harom, 5H), 4.00 (d, J = 14.7,
CH2Ph, 1H), 3.90 (d, J = 14.7 Hz, CH2Ph, 1H), 3.79 (s, ester Me,
3H), 3.56 (s, 5-CH2, 2H), 2.94 (d, J = 12.9 Hz, 4β-CH2, 1H), 2.01 (d,
J = 12.9 Hz, 4α-CH2, 1H), 2.00 (dq, J = 14.3, 7.4 Hz, 3-CH2, 1H), 1.76
(dq, J = 14.3, 7.4 Hz, 3-CH2, 1H), 1.28 (s, 5α-Me, 3H), 1.00 (t, J = 7.4
Hz, 3α-Me, 3H). 13C NMR (100 MHz, CDCl3): 9.3 (3-Et CH3), 23.5
(5-CH3), 26.3 (3-Et CH2), 44.5 (4-CH2), 52.0 (ester CH3), 54.8
(CH2Ph), 67.6 (5-CH2), 74.4 (3-C), 80.6 (5-C), 126.9 (CH), 127.8
(CH × 2), 128.2 (CH × 2), 138.6 (C), 172.9 (CO). HRMS (ESI
+): m/z calcd for C16H23NO4 294.1700 [M + H]+, found 294.1709.
Enantioselectivity was determined by chiral HPLC analysis after
reduction of 3b-endo with NaBH(OAc)3 to the corresponding alcohol,
Chiralpak As-H, isooctane/i-PrOH = 98:2, 1 mL/min, 254 nm; tR
(min): 18.0 (minor) and 28.3 (major) with 84% ee for 8b-endo.
(−)-(3R,5S)-Methyl 2-Benzyl-5-(hydroxymethyl)-5-methyl-3-
propylisoxazolidine-3-carboxylate (8c-endo). The same proce-
solid, mp 123−124 °C. [α]20 = −179.5 (c 1.0, CHCl3), along with
D
two isomers (16 mg, 21% yield, dr 70:30) in global yield 90%. IR:
3030, 2976, 2935, 1646, 1496, 1454, 1274, 1207, 1170, 1029, 910, 699.
1H NMR (400 MHz, CDCl3): 7.38−7.22 (m, Harom, 10H), 6.11 (q, J
= 7.1 Hz, CH, 1H), 4.08 (d, J = 14.4 Hz, CH2Ph, 1H), 3.70 (d, J =
14.4 Hz, CH2Ph, 1H), 3.12 (dd, J = 12.0, 1.1 Hz, 4β-CH2, 1H), 2.65
(d, J = 12.0 Hz, 4α-CH2, 1H), 2.39 (dd, J = 11.5, 1.3 Hz, 8β-CH2, 1H),
2.12 (d, J = 11.5 Hz, 8α-CH2, 1H), 1.62 (d, J = 7.1 Hz, CHMe, 3H),
1.46 (s, 1α-Me, 3H), 1.28 (s, 5α-Me, 3H). 13C NMR (100 MHz,
CDCl3): 15.9 (CH3), 17.7 (1-CH3), 21.8 (5-CH3), 46.3 (8-CH2), 49.7
(CH), 53.9 (4-CH2), 57.5 (CH2Ph), 67.0 (1-C), 76.4 (5-C), 127.1
(CH), 127.50 (CH × 2), 127.54 (CH), 128.3 (CH × 2), 128.60 (CH
× 2), 128.63 (CH × 2), 138.0 (C), 139.6 (C), 169.1 (CO). HRMS
(ESI+): m/z calcd for C22H26N2O2 351.2067 [M + H]+, found
351.2070.
dure as in 8a-endo to give 25.0 mg, 92% yield as a colorless oil. [α]20
D
= −44.6 (c 0.52, CHCl3) with 86% ee for pure 8c-endo. IR: 3447,
1
2960, 2930, 2873, 1729, 1455, 1224, 1157, 1053, 911. H NMR (400
MHz, CDCl3): 7.36 (d, J = 6.9 Hz, Harom, 2H), 7.33−7.28 (tt, J = 6.9,
1.4 Hz, Harom, 2H), 7.23 (tt, J = 7.1, 1.4 Hz, Harom, 1H), 4.00 (d, J =
14.7 Hz, CH2Ph, 1H), 3.91 (d, J = 14.7 Hz, CH2Ph, 1H), 3.78 (s, ester
Me, 3H), 3.54 (s, 5-CH2, 2H), 2.99 (s, OH, 1H), 2.94 (d, J = 12.9 Hz,
4β-CH2, 1H), 2.00 (d, J = 12.9 Hz, 4α-CH2, 1H), 1.96 (dt, J = 13.2,
4.7 Hz, 3-CH2, 1H), 1.69 (dt, J = 13.2, 4.7 Hz, 3-CH2, 1H), 1.45−1.34
(m, 3-CH2, 2H), 1.27 (s, 5α-Me, 3H), 0.97 (t, J = 7.3 Hz, 3α-Me, 3H).
13C NMR (100 MHz, CDCl3): 14.6 (3-CH3), 18.4 (5-CH3), 23.5 (3-
CH2), 35.8 (3-CH2), 45.0 (4-CH2), 52.0 (ester CH3), 54.8 (CH2Ph),
Typical Procedures for the Determination of Enantioselec-
tivity of Adduct 3. (+)-(3R,5S)-Methyl 2-Benzyl-5-(hydroxy-
methyl)-3,5-dimethylisoxazolidine-3-carboxylate (8a-endo). In
a 10 mL vial, to a solution of 3a-endo (40.0 mg, 0.144 mmol, Table 1,
entry 14) in THF (2 mL) were added NaBH(OAc)3 (122.0 mg, 0.576
mmol) and AcOH (36 μL, 0.576 mmol) at room temperature. The
reaction mixture was quenched after 17 h by adding water (2.0 mL)
and stirred at room temperature for 10 min before extraction with
CH2Cl2 (10 mL × 4) and EtOAc (10 mL × 1). The combined organic
3423
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
1
67.6 (5-CH2), 73.8 (3-C), 80.6 (5-C), 126.9 (CH), 127.8 (CH × 2),
128.2 (CH × 2), 138.6 (C), 172.9 (CO). HRMS (ESI+): m/z calcd
for C17H25NO4 308.1856 [M + H]+, found 308.1853; 330.1676 [M +
Na]+, found 330.1697; 346.1415 [M + K]+, found 346.1421.
Enantioselectivity was determined by chiral HPLC analysis after
reduction of 3c-endo with NaBH(OAc)3 to the corresponding alcohol,
Chiralpak As-H, isooctane/i-PrOH = 99:1, 1 mL/min, 230 nm; tR
(min): 32.6 (minor) and 41.1 (major) with 86% ee for 8c-endo.
(+)-(3R,5S)-tert-Butyl 2-Benzyl-5-(hydroxymethyl)-3,5-dime-
thylisoxazolidine-3-carboxylate (8d-endo). The same procedure
1740, 1458, 1377, 1273, 1202, 1165. H NMR (400 MHz, CDCl3):
3.75 (s, ester Me, 3H), 3.63 (d, J = 11.3 Hz, 5-CH2, 1H), 3.52 (d, J =
11.3 Hz, 5-CH2, 1H), 2.96 (d, J = 12.8 Hz, 4β-CH2, 1H), 2.63 (s,
NCH3, 3H), 1.95 (d, J = 12.8 Hz, 4α-CH2, 1H), 1.63 (br s, OH, 1H),
1.40 (s, 3α-Me, 3H), 1.28 (s, 5α-Me, 3H). 13C NMR (100 MHz,
CDCl3): 23.0 (3-CH3), 29.7 (5-CH3), 37.9 (N−CH3), 49.0 (4-CH2),
52.3 (ester CH3), 70.1 (3-C and 5-CH2), 80.3 (5-C), 173.2 (CO).
HRMS (ESI+): m/z calcd for C9H17NO4 204.1230 [M + H]+, found
204.1221. Enantioselectivity was determined by chiral HPLC analysis
after reduction of 3f-endo with NaBH(OAc)3 to the corresponding
alcohol, Chiralpak As-H, isooctane/i-PrOH = 99:1, 1 mL/min, 254
nm; tR (min): 23.2 (minor) and 29.2 (major) with 80% ee for 8f-endo.
(+)-(3R,5R)-Ethyl 5-(Hydroxymethyl)-2,3,5-trimethylisoxazo-
lidine-3-carboxylate (8g-exo). The same procedure as in 8a-endo to
as in 8a-endo to give 11.9 mg, 95% yield as a colorless oil. [α]20
=
D
+42.8 (c 0.1, CHCl3) with 92% ee for pure 8d-endo. IR: 3438, 2977,
1
2935, 1728, 1497, 1455, 1369, 1256, 1147, 1053, 735. H NMR (200
MHz, CDCl3): 7.35−7.13 (m, Harom, 5H), 3.95 (d, J = 14.3, CH2Ph,
1H), 3.70 (d, J = 14.3 Hz, CH2Ph, 1H), 3.49 (d, J = 11.5 Hz, 5-CH2,
1H), 3.42 (d, J = 11.5 Hz, 5-CH2, 1H), 3.31 (br s, OH, 1H), 2.87 (d, J
= 12.6 Hz, 4β-CH2, 1H), 1.84 (d, J = 12.6 Hz, 4α-CH2, 1H), 1.42 (s,
ester t-Bu, 9H), 1.40 (s, 3α-Me, 3H), 1.16 (s, 5α-Me, 3H). 13C NMR
(100 MHz, CDCl3): 18.9 (3-CH3), 22.6 (5-CH3), 28.1 (t-Bu CH3 ×
3), 49.1 (4-CH2), 55.0 (CH2Ph), 69.4 (5-CH2), 70.2 (3-C), 80.1 (5-
C), 82.0 (t-Bu C), 127.1 (CH), 128.0 (CH × 2), 128.4 (CH × 2),
138.5 (C), 172.0 (CO). HRMS (ESI+): m/z calcd for C18H27NO4
322.2013 [M + H]+, found 322.2005; 344.1832 [M + Na]+, found
344.1837. Enantioselectivity was determined by chiral HPLC analysis
after reduction of 3d-endo with NaBH4 to the corresponding alcohol,
Chiralpak As-H, isooctane/i-PrOH = 99:1, 1 mL/min, 218 nm; tR
(min): 8.9 (minor) and 11.3 (major) with 92% ee for 8d-endo.
(+)-(3R,5S)-Ethyl 2-Benzyl-5-(hydroxymethyl)-3,5-dimethyli-
soxazolidine-3-carboxylate (8e-endo). The same procedure as in
8a-endo to give 10.2 mg, 60% yield as a colorless oil. [α]20D = +33.8 (c
0.32, CHCl3) with 84% ee for pure 8e-endo. IR: 3445, 2980, 2936,
give 10.1 mg, 65% yield as a colorless oil. [α]20 = +100.2 (c 0.19,
D
CHCl3) with 81% ee for pure 8g-exo. IR: 3414, 2976, 2937, 2874,
1
1725, 1447, 1376, 1274, 1179, 1066. H NMR (200 MHz, CDCl3):
4.23 (q, J = 7.1 Hz, ester CH2, 2H), 3.66 (d, J = 11.0 Hz, 5-CH2, 1H),
3.54 (br s, OH, 1H), 3.53 (d, J = 11.0 Hz, 5-CH2, 1H), 2.65 (d, J =
12.5 Hz, 4β-CH2, 1H), 2.60 (s, NCH3, 3H), 2.29 (d, J = 12.5 Hz, 4α-
CH2, 1H), 1.39 (s, 3α-Me, 3H), 1.32 (t, J = 7.1 Hz, ester Me, 3H),
1.31 (s, 5α-Me, 3H). 13C NMR (100 MHz, CDCl3): 14.3 (Me CH2),
20.3 (3-CH3), 20.8 (5-CH3), 38.2 (N−CH3), 47.6 (4-CH2), 60.9 (Et
CH2), 70.7 (3-C), 71.3 (5-CH2), 80.3 (5-C), 171.9 (CO). HRMS
(ESI+): m/z calcd for C10H19NO4 218.1387 [M + H]+, found
218.1380. Enantioselectivity was determined by chiral HPLC analysis
after reduction of 3g-exo with NaBH4 to the corresponding alcohol,
Chiralpak As-H, isooctane/i-PrOH = 99:1, 1 mL/min, 254 nm; tR
(min): 14.0 (minor) and 20.7 (major) 81% ee for 8g-exo.
(−)-(1R,5S)-7-Benzyl-1,5-dimethyl-3,6-dioxa-7-azabicyclo-
[3.2.1]octan-2-one (9). In a 25 mL flask, to a solution of 8a-endo (80
mg, 0.286 mmol) in a mixture of THF/water (2:1, 4.5 mL) was added
LiOH (13.7 mg, 0.572 mmol) at room temperature. The mixture was
stirred for 1 h at room temperature, and then the reaction mixture was
diluted with CH2Cl2 (4 mL), quenched by adding 1 N HCl (2 mL) at
0 °C, and stirred at room temperature for 10 min. Extraction of the
aqueous layer with CH2Cl2 (10 mL × 4), followed by drying the
combined organic layers with MgSO4, filtration, and concentration in
vacuo afforded the crude carboxylic acid. The crude was dissolved in
CH2Cl2 (4 mL) and treated with DCC (89.0 mg, 0.427 mmol) and
DMAP (17.5 mg, 0.143 mmol) at room temperature, and the reaction
mixture was stirred for 2 h. The mixture was concentrated in vacuo and
treated with Et2O to precipitate the urea derivatives. After filtration,
washing with Et2O, and evaporation in vacuo, the resulting crude solid
was purified by column chromatography (petroleum ether/EtOAc =
4:1) leading to compound 9 (63.0 mg, 89% yield) as a white solid, mp
1
2875, 1733, 1497, 1455, 1375, 1268, 1167, 1051, 735. H NMR (200
MHz, CDCl3): 7.34−7.17 (m, Harom, 5H), 4.14 (q, J = 7.1 Hz, ester
CH2, 2H), 3.93 (d, J = 14.5 Hz, CH2Ph, 1H), 3.73 (d, J = 14.5 Hz,
CH2Ph, 1H), 3.46 (s, CH2, 2H), 2.90 (d, J = 12.7 Hz, 4β-CH2, 1H),
1.89 (d, J = 12.7 Hz, 4α-CH2, 1H), 1.44 (s, 3α-Me, 3H), 1.23 (t, J =
7.1 Hz, ester Me, 3H), 1.18 (s, 5α-Me, 3H). 13C NMR (100 MHz,
CDCl3): 14.2 (Et CH3), 18.9 (3-CH3), 22.7 (5-CH3), 49.0 (4-CH2),
55.0 (CH2Ph), 61.3 (Et CH2), 69.4 (5-CH2), 69.7 (3-C), 80.2 (5-C),
127.1 (CH), 128.0 (CH × 2), 128.3 (CH × 2), 138.3 (C), 172.8 (C
O). HRMS (ESI+): m/z calcd for C16H23NO4 294.1700 [M + H]+,
found 294.1703; 316.1519 [M + Na]+, found 316.1529. Enantiose-
lectivity was determined by chiral HPLC analysis after reduction of 3e-
endo with NaBH4 to the corresponding alcohol, Chiralpak As-H,
isooctane/i-PrOH = 99:1, 1 mL/min, 210 nm; tR (min): 20.1 (minor)
and 26.5 (major) with 84% ee for 8e-endo.
90−91 °C. [α]20 = −175.0 (c 0.53, CHCl3) with 88% ee. IR: 2982,
(+)-(3R,5R)-Ethyl 2-Benzyl-5-(hydroxymethyl)-3,5-dimethyli-
soxazolidine-3-carboxylate (8e-exo). The same procedure as in
8a-endo to give 23.0 mg, 99% yield as a colorless oil. [α]20D = +85.3 (c
0.93, CHCl3) with 66% ee for pure 8e-exo. IR: 3435, 2980, 2935, 2873,
D
2935, 1739, 1455, 1378, 1272, 1139, 1035. 1H NMR (400 MHz,
CDCl3): 7.37 (dd, J = 7.1, 1.7 Hz, Harom, 2H), 7.35−7.30 (tt, J = 7.1,
1.7 Hz, Harom, 2H), 7.27 (tt, J = 6.1, 1.6 Hz, Harom, 1H), 4.25 (dd, J
= 11.2, 2.0 Hz, 4β-CH2, 1H), 4.16 (d, J = 11.2 Hz, 4α-CH2, 1H), 4.03
(dd, J = 14.1 Hz, CH2Ph, 1H), 3.77 (d, J = 14.1 Hz, CH2Ph, 1H), 2.51
(dd, J = 11.9, 2.0 Hz, 8β-CH2, 1H), 2.37 (d, J = 11.9 Hz, 8α-CH2, 1H),
1.45 (s, 1α-Me, 3H), 1.35 (s, 5α-Me, 3H). 13C NMR (100 MHz,
CDCl3): 17.8 (1-CH3), 18.9 (5-CH3), 45.4 (8-CH2), 57.8 (CH2Ph),
67.3 (1-C), 76.7 (5-C), 78.9 (4-CH2), 127.4 (CH), 128.4 (CH × 2),
128.6 (CH × 2), 137.1 (C), 170.1 (CO). HRMS (ESI+): m/z calcd
for C14H17NO3 248.1281 [M + H]+, found 248.1272; 270.1101 [M +
Na]+, found 270.1094. Enantioselectivity was determined by chiral
HPLC analysis, Chiralpak As-H, isooctane/i-PrOH = 97:3, 1 mL/min,
230 nm; tR (min): 33.5 (minor) and 36.8 (major) ; 88% ee.
1
1725, 1497, 1455, 1375, 1260, 1167, 1060, 737. H NMR (200 MHz,
CDCl3): 7.29−7.14 (m, Harom, 5H), 4.21 (q, J = 7.1 Hz, ester CH2,
2H), 3.97 (d, J = 14.6 Hz, CH2Ph, 1H), 3.61 (d, J = 14.6 Hz, CH2Ph,
1H), 3.49 (d, J = 11.0 Hz, 5-CH2, 1H), 3.40 (d, J = 11.0 Hz, 5-CH2,
1H), 2.78 (br s, OH, 1H), 2.61 (d, J = 12.5 Hz, 4β-CH2, 1H), 2.16 (d,
J = 12.5 Hz, 4α-CH2, 1H), 1.41 (s, 3α-Me, 3H), 1.29 (t, J = 7.1 Hz,
ester Me, 3H), 1.22 (s, 5α-Me, 3H). 13C NMR (100 MHz, CDCl3):
14.4 (Et CH3), 20.7, 20.8 (3-CH3, 5-CH3), 47.6 (4-CH2), 55.3
(CH2Ph), 61.1 (Et CH2), 70.2, 70.4 (5-CH2, 3-C), 80.2 (5-C), 127.2
(CH), 128.0 (CH × 2), 128.4 (CH × 2), 138.2 (C), 172.2 (CO).
HRMS (ESI+): m/z calcd for C16H23NO4 294.1700 [M + H]+, found
294.1701; 316.1519 [M + Na]+, found 316.1517. Enantioselectivity
was determined by chiral HPLC analysis after reduction of 3e-exo with
NaBH(OAc)3 to the corresponding alcohol, Chiralpak As-H,
isooctane/i-PrOH = 99:1, 1 mL/min, 254 nm; tR (min): 16.3
(minor) and 19.9 (major) with 66% ee for 8e-exo.
(+)-tert-Butyl ((3R,5S)-5-Hydroxy-3,5-dimethyl-2-oxotetra-
hydro-2H-pyran-3-yl)carbamate (10). In a 25 mL flask, to a
solution of 9 (57 mg, 0.23 mmol, 88% ee) in 2-propanol (3 mL) were
added Boc2O (0.251 g, 1.15 mmol) and Raney-Ni (0.419 g) in 2-
propanol (2 mL) at room temperature. Hydrogen gas (balloon) was
then charged to the reaction flask. After stirring for 1 h at 80 °C,
Raney-Ni was filtered over Celite and then washed with methanol, and
the solvent was evaporated in vacuo to afford the crude residue.
Column chromatography (CH2Cl2/MeOH = 30:1) gave compound
(+)-(3R,5S)-Methyl 5-(Hydroxymethyl)-2,3,5-trimethylisoxa-
zolidine-3-carboxylate (8f-endo). The same procedure as in 8a-
endo to give 14.4 mg, 75% yield as a colorless oil. [α]20 = +12.2 (c
D
0.05, CHCl3) with 80% ee for pure 8f-endo. IR: 3385, 2925, 2854,
3424
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
10 (56 mg, 94% yield) as white amorphous solid, mp 107−108 °C.
(5) (a) Gothelf, K. V.; Thomsen, I.; Jørgensen, K. A. J. Am. Chem.
Soc. 1996, 118, 59. (b) Gothelf, K. V.; Jørgensen, K. A. Chem.
Commun. 2000, 1449.
[α]20 = +5.5 (c 1.07, CHCl3). IR: 3354, 2979, 2933, 1760, 1705,
D
1
1519, 1455, 1368, 1249, 1166, 1090, 961, 758. H NMR (400 MHz,
CDCl3): 4.98 (s, NH, 1H), 3.70 (d, J = 12.1 Hz, 6β-CH2, 1H), 3.49 (d,
J = 12.1 Hz, 6α-CH2, 1H), 3.33 (br s, OH, 1H), 2.43 (d, J = 13.3 Hz,
4β-CH2, 1H), 2.39 (d, J = 13.3 Hz, 4α-CH2, 1H), 1.55 (s, 3α-Me, 3H),
1.52 (s, 5α-Me, 3H), 1.44 (s, t-Bu, 9H). 13C NMR (100 MHz,
CDCl3): 23.9 (5-CH3), 25.2 (3-CH3), 28.3 (t-Bu), 42.7 (4-CH2), 58.2
(3-C), 68.6 (6-CH2), 80.5 (C, t-Bu), 83.7 (5-C), 154.3 (CO, Boc),
178.1 (CO, lactone). HRMS (FI/FD): m/z calcd for C12H21NO5
282.1312 [M + Na]+, found 282.1315; 298.1051 [M + K]+, found
298.1055.
(6) (a) Sibi, M. P.; Ma, Z.; Jasperse, C. P. J. Am. Chem. Soc. 2004,
126, 718. (b) Suga, H.; Nakajima, T.; Itoh, K.; Kakehi, A. Org. Lett.
2005, 7, 1431. (c) Evans, D. A.; Song, H.-J.; Fandrick, K. R. Org. Lett.
2006, 8, 3351.
(7) Jen, W. S.; Wiener, J. J. M.; MacMillan, D. W. C. J. Am. Chem. Soc.
2000, 122, 9874.
(8) For organocatalyzed 1,3-DC between aldonitrones and enals, see:
(a) Karlsson, S.; Hogberg, H.-E. Tetrahedron: Asymmetry 2002, 13,
̈
923. (b) Karlsson, S.; Hogberg, H.-E. Eur. J. Org. Chem. 2003, 15,
̈
2782. (c) Puglisi, A.; Benaglia, M.; Cinquini, M.; Cozzi, F.; Celentano,
G. Eur. J. Org. Chem. 2004, 3, 567. (d) Lemay, M.; Trant, J.; Ogilvie,
W. W. Tetrahedron 2007, 63, 11644. (e) Chow, S. S.; Nevalainen, M.;
Evans, C. A.; Johannes, C. W. Tetrahedron Lett. 2007, 48, 277.
ASSOCIATED CONTENT
* Supporting Information
Additional experimental procedures and spectroscopic data.
Cartesian coordinates (in xyz format) and energies of
optimized structures. Results of the Quantum Theory of
Atoms In Molecules (QTAIM) study. This material is available
free of charge via the Internet at http://pubs.acs.org.
■
S
(f) Weselin
(g) Weselinski, Ł.; Słyk, E.; Jurczak, J. Tetrahedron Lett. 2011, 52, 381.
(9) (a) Viton, F.; Bernardinelli, G.; Kundig, E. P. J. Am. Chem. Soc.
́
ski, Ł.; Stępniak, P.; Jurczak, J. Synlett 2009, 14, 2261.
́
̈
2002, 124, 4968. (b) Bad
E. P.; Viton, F. Chem. Asian J. 2008, 3, 1298. (c) Bad
Brinkmann, Y.; Viton, F.; Kundig, E. P. Pure Appl. Chem. 2008, 5,
̆
oiu, A.; Bernardinelli, G.; Mareda, J.; Kundig,
̈
̆
oiu, A.;
̈
AUTHOR INFORMATION
Corresponding Author
*E-mail: gilles.dujardin@univ-lemans.fr
Notes
1013. (d) Bad
̆
oiu, A.; Bernardinelli, G.; Kundig, E. P. Synthesis 2010,
2207. (e) Badoiu, A.; Kundig, E. P. Org. Biomol. Chem. 2012, 10, 114.
(f) Kundig, E. P.; Saudan, C. M.; Bernardinelli, G. Angew. Chem. Int.
■
̈
̆
̈
̈
Ed 1999, 38, 1120. (g) Anil Kumar, P. G.; Pregosin, P. S.; Vallet, M.;
Bernardelli, G.; Jazzar, R. F.; Viton, F.; Kundig, E. P. Organometallics
̈
The authors declare no competing financial interest.
2004, 23, 5410.
(10) Mita, T.; Ohtsuki, N.; Ikeno, T.; Yamada, T. Org. Lett. 2002, 4,
2457.
(11) Shirahase, M.; Kanemasa, S.; Oderaotoshi, Y. Org. Lett. 2004, 6,
675.
ACKNOWLEDGMENTS
Financial support of this study was provided by ANR Grant
(BS07- 005-01).
■
(12) (a) Kano, T.; Hashimoto, T.; Maruoka, K. J. Am. Chem. Soc.
2005, 127, 11926. (b) Hashimoto, T.; Omote, M.; Kano, T.; Maruoka,
K. Org. Lett. 2007, 9, 4805.
REFERENCES
■
(1) (a) Martin, J. N.; Jones, R. C. F. In Synthetic Applications of 1,3-
Dipolar Cycloaddition Chemistry Toward Heterocycles and Natural
Products; Padwa, A., Pearson, W. H., Eds.; Wiley and Sons: Hoboken,
NJ, 2003; Chapter 1, p 1. (b) Gothelf, K. V. In Cycloaddition Reactions
in Organic Synthesis; Kobayashi, S., Jørgensen, K. A., Eds.; Wiley-VCH:
Weinheim, Germany, 2002; Chapter 6, p 211. (c) Kanemasa, S. In
Cycloaddition Reactions in Organic Synthesis; Kobayashi, S., Jørgensen,
K. A., Eds.; Wiley-VCH: Weinheim, Germany, 2002; Chapter 7, p 249.
(2) (a) Frederickson, M. Tetrahedron 1997, 53, 403. (b) Gothelf, K.
V.; Jørgensen, K. A. Chem. Rev. 1998, 98, 863. (c) Gothelf, K. V.;
Jørgensen, K. A. Chem. Commun. 2000, 1449. (d) Karlsson, S.;
(13) (a) Carmona, D.; Lamata, M. P.; Viguri, F.; Rodriguez, R.; Oro,
L. A.; Balana, A. I.; Lahoz, F. J.; Tejero, T.; Merino, P.; Franco, S.;
Montesa, I. J. Am. Chem. Soc. 2004, 126, 2716. (b) Carmona, D.;
Lamata, M. P.; Viguri, F.; Rodriguez, R.; Oro, L. A.; Lahoz, F. J.;
Balana, A. I.; Tejero, T.; Merino, P. J. Am. Chem. Soc. 2005, 127,
13386. (c) Carmona, D.; Lamata, M. P.; Viguri, F.; Rodriguez, R.;
Fischer, T.; Lahoz, F. J.; Dobrinovitch, I. T.; Oro, L. A. Adv. Synth.
Catal. 2007, 349, 1751.
(14) (a) Wang, Y.; Wolf., J.; Zavalij, P.; Doyle, M. P. Angew. Chem.,
Int. Ed. 2008, 47, 1439. (b) Wang, X.; Weigl.; Doyle, M. P. J. Am.
Chem. Soc. 2011, 133, 9572.
(15) (a) Selim, K. B.; Beauchard, A.; Lhoste, J.; Martel, A.; Laurent,
M. Y.; Dujardin, G. Tetrahedron: Asymmetry 2012, 23, 1670.
Hogberg, H.-E. Org. Prep. Proced. Int. 2001, 33, 103. (e) Pellissier, H.
Tetrahedron 2007, 63, 3235. (f) Stanley, L. M.; Sibi, M. P. Chem. Rev.
2008, 108, 2887. (g) Nguyen, T. B.; Martel, A.; Gaulon, C.; Dhal, R.;
̈
́
(b) Weselinski, Ł.; Kalinowska, E.; Jurczak, J. Tetrahedron: Asymmetry
2012, 23, 264.
́
Dujardin, G. Org. Prep. Proced. Int. 2010, 42, 387. (h) Najera, C.;
Sansano, J. M.; Yus, M. J. Braz. Chem. Soc. 2010, 21, 377. (i) Kissane,
M.; Maguire, A. R. Chem. Soc. Rev. 2010, 39, 845. (j) Pellissier, H.
Tetrahedron 2012, 68, 2197. (k) Xinga, Y.; Wang, N.-X. Coord. Chem.
Rev. 2012, 256, 938. (l) Nguyen, T. B.; Martel, A.; Gaulon, C.; Dhal,
R.; Dujardin, G. Org. Prep. Proced. Int. 2012, 44, 1.
(16) For reviews on the formation of quaternary stereocenters, see:
(a) Seebach, D. Angew. Chem. Int. Ed. 2011, 50, 96. (b) Clayden, J.;
Donnard, M.; Lefranc, J.; Tetlow, D. J. Chem. Commun. 2011, 47,
4624. (c) Hawner, C.; Alexakis, A. Chem. Commun. 2010, 46, 7295.
(d) Shibasaki, M.; Kanai, M. Chem. Rev. 2008, 108, 2853. (e) Trost, B.
M.; Jiang, C. Synthesis 2006, 3, 369. (f) Christoffers, J.; Baro, A. Adv.
Synth. Catal. 2005, 347, 1473. (g) Quaternary Stereocenters: Challenges
and Solutions for Organic Synthesis; Christoffers, J., Baro, A., Eds.;
Wiley-VCH: Weinheim, Germany, 2005. (h) Douglas, C. J.; Overman,
L. E. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5363.
(17) (a) Nguyen, T. B.; Martel, A.; Dhal, R.; Dujardin, G. Org. Lett.
2008, 10, 4493−4496. (b) Nguyen, T. B.; Beauseigneur, A.; Martel, A.;
Dhal, R.; Laurent, M.; Dujardin, G. J. Org. Chem. 2010, 75, 611.
(18) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215.
(19) Frisch, M. J., et al.; Gaussian 09, Revision B.01; Gaussian, Inc.:
Wallingford, CT, 2010.
(3) (a) Seerden, J.-P. G.; Scholte op Reimer, A. W. A.; Scheeren, H.
W. Tetrahedron Lett. 1994, 35, 4419. (b) Seerden, J.-P. G.; Kuepers, M.
M. M.; Scheeren, H. W. Tetrahedron: Asymmetry 1995, 6, 1441.
(c) Seerden, J.-P. G.; Boeren, M. M. M.; Scheeren, H. W. Tetrahedron
1997, 53, 11843. (d) Simonsen, K. B.; Bayon, P.; Hazell, R. G.;
Gothelf, K. V.; Jørgensen, K. A. J. Am. Chem. Soc. 1999, 121, 3845.
(e) Simonsen, K. B.; Jørgensen, K. A.; Hu, Q.-S.; Pu, L. Chem.
Commun. 1999, 811. (f) Jensen, K. B.; Hazell, R. G.; Jørgensen, K. A. J.
Org. Chem. 1999, 64, 2353. (g) Jensen, K. B.; Roberson, M.;
Jørgensen, K. A. J. Org. Chem. 2000, 65, 9080. (h) Jiao, P.; Nakashima,
D.; Yamamoto, H. Angew. Chem., Int. Ed. 2008, 47, 2411.
(i) Yanisagawa, A.; Izumiseki, A.; Sugita, T.; Kushihara, N.; Yoshida,
K. Synlett 2011, 16, 107.
(20) Owing to the size of the calculated complex, enthalpy and free
Gibbs energy were not calculated and only the electronic energy is
given.
(4) (a) Ding, X.; Taniguchi, K.; Hamamoto, Y.; Sada, K.; Fujinami,
S.; Ukaji, Y.; Inomata, K. Bull. Chem. Soc. Jpn. 2006, 79, 1079.
3425
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426

The Journal of Organic Chemistry
Article
(21) Desiraju, G. R. Acc. Chem. Res. 1991, 24, 290.
(22) Additional data are given in the Supporting Informations.
(23) (a) Desiraju, G. R. Chem. Commun. 2005, 2995. (b) Desiraju, G.
R.; Steiner, T. The Weak Hydrogen Bond in Structural Chemistry and
Biology; Oxford University Press: Oxford, 1999 and references therein.
(24) (a) Taylor, R.; Kennard, O. J. Am. Chem. Soc. 1982, 104, 5063.
(b) Steiner, T. Crystallogr. Rev. 1996, 6, 1.
(25) (a) Scheiner, T.; Kar, T. J. Phys. Chem. A 2008, 112, 11854.
(b) Castellano, R. K. Curr. Org. Chem. 2004, 8, 845. (c) Koch, U.;
Popelier, P. L. A. J. Phys. Chem. 1995, 99, 9747. (d) Steiner, T.;
Desiraju, G. R. Chem. Commun. 1998, 891. (e) Yang, H.; Wong, M. W.
J. Am. Chem. Soc. 2015, 135, 5808.
(26) Laurence, C.; Graton, J.; Berthelot, M.; Besseau, F.; Le Questel,
J.-Y.; Luco̧ n, M.; Ouvrard, C.; Planchat, A.; Renault, E. J. Org. Chem.
2010, 75, 4105.
(27) (a) Ess, D. H.; Jones, G. O.; Houk, K. N. Adv. Synth. Catal.
2006, 348, 2337. (b) Kuznetsov, M. L.; Nazarov, A. A.; Kozlova, L. V.;
Kukushkin, V. Y. J. Org. Chem. 2007, 72, 4475.
(28) For each of the adducts 3a−g, the major diastereomer (endo or
exo) could be isolated as a pure diastereomer after flash
chromatography in yields that match those calculated from the
crude product ratio.
3426
dx.doi.org/10.1021/jo5001737 | J. Org. Chem. 2014, 79, 3414−3426