Diastereodivergent asymmetric sulfa-Michael additions of α-branched enones using a single chiral organic catalyst

Article abstract of DOI:10.1021/ja207847p

A significant limitation of modern asymmetric catalysis is that, when applied to processes that generate chiral molecules with multiple stereogenic centers in a single step, researchers cannot selectively access the full matrix of all possible stereoisome

Full text of DOI:10.1021/ja207847p

ARTICLE
pubs.acs.org/JACS
Diastereodivergent Asymmetric Sulfa-Michael Additions of
α-Branched Enones using a Single Chiral Organic Catalyst
Xu Tian,† Carlo Cassani,† Yankai Liu,† Antonio Moran,† Atsushi Urakawa,† Patrizia Galzerano,† Elena Arceo,†
and Paolo Melchiorre*,†,‡
†ICIQ, Institute of Chemical Research of Catalonia, Avenida Països Catalans 16-43007 Tarragona, Spain
‡ICREA, Instituciꢀo Catalana de Recerca i Estudis Avanc-ats, Passeig Lluís Companys 23-08010 Barcelona, Spain
S Supporting Information
b
ABSTRACT: A significant limitation of modern asymmetric catalysis is
that, when applied to processes that generate chiral molecules with
multiple stereogenic centers in a single step, researchers cannot selectively
access the full matrix of all possible stereoisomeric products. Mirror image
products can be discretely provided by the enantiomeric pair of a chiral
catalyst. But modulating the enforced sense of diastereoselectivity using a
single catalyst is a largely unmet challenge. We document here the
possibility of switching the catalytic functions of a chiral organic small
molecule (a quinuclidine derivative with a pendant primary amine) by
applying an external chemical stimulus, in order to induce diastereodivergent pathways. The strategy can fully control the
stereochemistry of the asymmetric conjugate addition of alkyl thiols to α-substituted α,β-unsaturated ketones, a class of carbonyls
that has never before succumbed to a catalytic approach. The judicious choice of acidic additives and reaction media switches the
sense of the catalyst’s diastereoselection, thereby affording either the syn or anti product with high enantioselectivity.
’ INTRODUCTION
step, chemists cannot selectively access the full matrix of
stereoisomers using a single chiral catalyst.
The catalytic stereoselective conjugate addition of carbonyl
compounds has historically offered a potent synthetic way of
producing valuable chiral molecules.1 Both metal-1aꢀc and
organic-1d,ebased approaches are now so reliable that synthetic
chemists can address even the most daunting of challenges
connected with the asymmetric catalytic β-functionalization of
simple unsaturated carbonyl compounds. In contrast, the activa-
tion of sterically hindered carbonyls is a persistent problem. In
particular, the catalytic activation of α,β-disubstituted unsatu-
rated ketones remains an elusive target.2 Conjugate additions to
linear α-branched enones generate two adjacent stereocenters
through an addition-protonation tandem sequence. All attempts
to design a stereocontrolled variant of this process must thus
address the challenge of diastereo- as well as enantioselectivity
(Figure 1). This requires the ability of a chiral catalyst to (i)
effectively activate the highly hindered keto-moiety, while (ii)
selectively shielding one of the two faces of the electrophilic
unsaturated π-system to forge the β-stereocenter with high
fidelity. Finally, (iii) strict control over the transient enolate/
enamine geometry is necessary to ensure a catalyst-directed
protonation event.
Mirror image products (complementary enantioselectivity)
can be individually provided by the enantiomeric pair of a chiral
catalyst.3 However, researchers are largely still not able to modulate
the sense of diastereoselectivity (control over the relative stereo-
chemistry) using a single chiral catalyst. A diastereochemical switch
generally requires the use of distinct chiral catalysts4,5 or ligands,6
the addition of different Lewis acids7 or tailored substrate modi-
fications.8 Less explored is the approach of changing the reaction
conditions to tune the functions of a single catalyst.9
Herein, we demonstrate the ability of a chiral organic small
molecule, the cinchona-based primary amine 5, to catalyze the
highly stereoselective sulfa-Michael addition (SMA) of different
alkyl thiols to a range of linear α-branched enones. We have
found that the catalytic function of the primary amine 5 can be
modulated to induce diastereodivergent pathways by applying an
external chemical stimulus. By judiciously choosing different
acidic additives and reaction media, we can switch the enforced
sense of diastereo-induction, thus allowing access to all possible
stereoisomeric products of the SMA reaction.
This article reports our efforts toward optimizing the syn- and
anti-SMA reaction conditions, and offers a rationalization for the
switchable selectivity of catalyst 5.
From a wider perspective, the complex stereoselectivity issues
inherent to this type of chemical transformation provide the
opportunity to face one of the most significant limitations of
asymmetric catalysis: when applied to processes that generate
chiral molecules with multiple stereogenic centers in a single
Received: August 18, 2011
Published: September 21, 2011
r
2011 American Chemical Society
17934
dx.doi.org/10.1021/ja207847p J. Am. Chem. Soc. 2011, 133, 17934–17941
|

Journal of the American Chemical Society
ARTICLE
’ RESULTS AND DISCUSSION
to engage in iminium ion formation with encumbered carbonyls
prompted us to explore its behavior with respect to the stereo-
selective SMA13 of α-branched α,β-unsaturated ketones. Considering
the synthetic usefulness of simple alkyl S-centered nucleophiles,14
we focused on the conjugate addition of benzyl mercaptan 1a to the
α-branched enone 2a as the model reaction. This transformation
generates two adjacent stereocenters through a conjugate addition-
protonation tandem sequence, leading to the syn-3a and anti-4a
products. Gratifyingly, catalyst 5 showed high efficiency in activating
the sterically hindered substrate 2a. Extensive optimization studies
are reported in Tables S1ꢀ5 within the Supporting Information,
with selected results summarized in Table 1.
Standard Reaction Optimization. Our laboratory and others,
independently, have recently established cinchona derivatives of
type 5 (chiral primary amines easily derived from natural sources) as
general and effective covalent binding activators of sterically con-
gested carbonyl compounds.10 In particular, we have demonstrated
the versatility of these organic molecules in the context of the stereo-
selective iminium-catalyzed SMA of alkyl thiols to both α,β-unsa-
turated enones11a and α-branched α,β-unsaturated aldehydes.11b
The unique ability of 60-hydroxy-9-amino-9-deoxyepiquinidine 512
The absence of any substantial diastereoselectivity in the pre-
sence of a strong base or an achiral iminium-forming catalyst
(entries 1ꢀ2) excluded the possibility of a selective pathway
induced by substrate control.15 In the absence of an acidic additive,
catalyst 5 afforded the selective formation of the syn product 3a
with a moderate level of enantiomeric excess (ee) (entry 3).
Under these conditions, the catalytic profile of amine 5 is dictated
by the tertiary bridgehead amine, which likely channels the SMA
reaction through a general base-catalyzed mechanistic pathway.16,17
We then sought to favor the iminium activation of the enone 2a
by addition of an acid cocatalyst that, by protonating the
quinuclidine moiety, would suppress the chemical handle neces-
sary for base catalysis,18 while providing the strongly acidic
conditions required for iminium ion formation.19 Combinations
Figure 1. Challenges arising from the catalytic activation of α-branched
enones.
Table 1. Selected Optimization Studiesa
entry
catalyst (mol %)
acid (mol %)
pKa
solvent
°C
conv (%)
dr syn-3a:anti-4a
ee (%) 3a/4a
1
DBU(20)
BnNH2 (30)
5 (20)
toluene
toluene
CHCl3
CHCl3
CHCl3
toluene
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
acetone
25
25
25
25
25
25
25
25
25
25
40
40
67
60
29
35
40
11
35
42
15
13
28
62
1:1.6
1.8:1
5:1
2
TFA (30)
3
65/49
77/17
82/17
81/78
79/55
88/59
72/86
<5/92
35/93
<5/98
4
5 (20)
CH3CO2H (20)
PhCO2H (40)
4.76
4.20
4.20
3.27
3.27
2.17
0.3
6.5:1
8.5:1
8:1
5
5 (20)
6
5 (20)
PhCO2H (40)
7
5 (20)
2-F-C6H4CO2H (20)
2-F-C6H4CO2H (40)
2-NO2-C6H4CO2H (40)
TFA (40)
7:1
8
5 (20)
6:1
9
5 (20)
1.3:1
1:1.5
1:1.3
1:6.2
10
11
12
5 (20)
5 (20)
(S)-6 (30)
<2.0
<2.0
5 (10)
(S)-6 (10)
a Both diastereomeric ratios (dr) and conversion were determined by 1H NMR analysis of the crude mixture. Enantiomeric excess (ee) values were
determined by GC analysis on commercially available chiral stationary phases. Reactions carried out using 2 equiv of 1a and [2a]0= 0.25M. Reaction
time: 16 h. pKa values determined in H2O. DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene; BnNH2 = benzylamine; TFA = trifuoroacetic acid; DPP =
diphenyl hydrogen phosphate.
17935
dx.doi.org/10.1021/ja207847p |J. Am. Chem. Soc. 2011, 133, 17934–17941

Journal of the American Chemical Society
ARTICLE
Table 2. Catalyst StructureꢀReactivity/Stereoselectivity Studies of the SMA Reactiona
entry
catalyst (mol %)
acid (mol %)
solvent
°C
conv (%)
dr syn-3a:anti-4a
ee (%) 3a/4a
1
5 (20)
8 (20)
5 (10)
8 (10)
5 (20)
5 (20)
5 (10)
5 (20)
2-F-C6H4CO2H (40)
2-F-C6H4CO2H (40)
(S)-6 (10)
CHCl3
CHCl3
acetone
acetone
acetone
CHCl3
acetone
acetone
acetone
acetone
acetone
acetone
acetone
acetone
25
25
40
40
25
40
40
40
40
40
40
40
40
40
42
26
6:1
7:1
88/59
77/n.d.
<5/98
2
3
62
1:6.2
1:1.7
2.1:1
1:1.3
1:1.7
4.5:1
4.2:1
1:5.2
1:6.1
1:4.1
1:3.9
1:2.5
4
(S)-6 (10)
25
n.d./97
30/83
5
2-F-C6H4CO2H (40)
(S)-6 (30)
35
6
28
35/93
7
TFA (10)
18
n.d./98
<5/<5
<5/<5
n.d./96
<5/98
n.d./92c
n.d./94c
n.d./94c
8
14
9
(S)-6 (40)
(R)-6 (10)
DPP (10)
(S)-6 (10)
(R)-6 (10)
DPP (10)
18
10
11
12
13
14
5 (10)
5 (10)
7 (10)
7 (10)
7 (10)
39b
52b
51b
43b
66b
a Both diastereomeric ratios (dr) and conversion were determined by 1H NMR analysis of the crude mixture. Enantiomeric excess (ee) values were
determined by HPLC or GC analysis on commercially available chiral stationary phases. Reaction time: 16 h. Reactions in chloroform carried out using 2
equiv of 1a and [2a]0 = 0.25M. Reactions in acetone carried out using 2 equiv of 1a and [2a]0 = 0.5M; n.d.= not determined. b Yield of the isolated
product after 60 h reaction time. c Reaction leading to the opposite enantiomer of anti-4a.
of benzoic acid derivatives with amine 5 effectively promoted the
SMA reaction, imparting a good level of syn diastereoselection
and moderate enantioinduction (entries 5ꢀ7). Evaluation of the
reaction media indicated chloroform as the best solvent. Notably,
a 1:2 ratio of amine 5 to ortho-fluorobenzoic acid offered an
effective catalytic system for achieving syn diastereoselectivity
while inferring high enantiomeric excess (ee = 88%, entry 8).
We were pleased to observe that strong acids (pKa in water
<2.5, entries 9ꢀ11) could greatly alter the stereochemical out-
come of the reaction. The loss of diastereocontrol was accom-
panied by a remarkable increase in the enantiomeric excess of the
anti-4a adduct (>92% ee), while the syn-3a product distribution
was essentially racemic. We thus undertook an extensive opti-
mization (detailed within Tables S6ꢀ11 in the Supporting
Information) to identify a catalytic system capable of a high level
of anti diastereoinduction. By combining amine 5 with the
commercially available chiral phosphoric acid (S)-620 in a 1:1
ratio (10 mol %), switching the solvent to acetone,21 and heating
to 40 °C, we obtained the anti diastereoisomer 4a, essentially as a
single enantiomer (entry 12, table 1).22
Further Experimental Observations. The initial results
indicated that the pKa of the acidic additive and the reaction
media can be used to switch the sense of the catalyst’s diastereo-
selection, thereby affording either the enantioenriched syn or
anti product on demand. To gain more insight into the factors
responsible for the uncommon diastereoswitching behavior of
amine 5, we further studied the SMA reaction of 1a and enone 2a
(see Tables S6ꢀ11 and Figures S2ꢀ3 within the Supporting
Information, with selected results detailed in Table 2).
acetone, amine 8 promoted the SMA reaction yielding the anti-
diastereoisomer 4a with very high enantioselectivity but with
essentially no control over the relative stereochemistry (compare
entries 3 and 4, Table 2). This result underlines how the hydrogen-
bonding donor24 moiety at the 60 position of the cinchona scaffold
was essential to channel the reaction toward an anti-diastereo-
selective pathway.
The anti-diastereoselective pathway, however, is also strictly
dependent on the choice of solvent and the nature of the strong
acid used. The amine 5/ortho-fluorobenzoic acid combination,
which proved effective for promoting the syn-diastereoselective
reaction in chloroform, led to poor stereocontrol when used in
acetone (entry 5), while the amine 5/(S)-6 catalyst salt did not
afford relative stereocontrol in chloroform (entry 6). Moreover,
when a strong acid other than the phosphoric acid derivative was
used in combination with amine 5 (e.g., TFA, entry 7), a complete
loss of diastereoselectivity was observed, although the anti-4a
isomer was generated with very high enantiomeric purity. These
results suggest that the constructive cooperation of three crucial
parameters is necessary for inducing the anti-diastereoselectivity.
These are (i) the hydroxyl group at the 60 position of the catalyst
5, (ii) the nature of the reaction medium, and (iii) the strong
hydrogen-bonding ability of the phosphate anion. Each of these
factors, individually, is necessary but not alone sufficient for
switching the catalyst’s diastereochemical preference.
Within this context, it is intriguing to consider how combining
the amine 5 and the acid 6 results in a unique mechanistic pathway.
The discrete catalysts, when operating individually under general
base and Brønsted acid25 catalysis, respectively, still promoted
the SMA addition in acetone but induced a syn-selective racemic
pattern (entries 8 and 9).
Finally, to gain insight into the specific role played by the two
chiral entities of the 5/6 catalyst salt,26 we used the mismatched
combinations to promote the anti SMA reaction (compare
entries 3 and 10). The combination of amine 5 with (R)-6 gave
lower reactivity and slightly decreased stereoselectivity (96% ee
against 98% ee). This suggests that the chiral primary amine was
mainly dictating the enantioselectivity of the process, while the
First, we evaluated the influence of the hydroxyl group at the 60
position of the catalyst scaffold.23 The primary amine 8, directly
derived from natural quinidine, in combination with benzoic acid
promoted the syn SMA reaction of 1a and 2a in chloroform
leading to product syn-3a with slightly lower stereoselectivity
than when using catalyst 5 (compare entries 1 and 2, Table 2;
further results for the 8-catalyzed syn-selective SMA reaction are
reported in Supporting Information Table S5). In marked
contrast, when mixed with the chiral phosphoric acid (S)-6 in
17936
dx.doi.org/10.1021/ja207847p |J. Am. Chem. Soc. 2011, 133, 17934–17941

Journal of the American Chemical Society
ARTICLE
Figure 2. Switching the diastereoselectivity of chiral amine 5: providing
distinct chiral environments for the SMA reaction by exploiting the
catalyst’s conformational flexibility and the tunable nature of the ion pair
intermediate I. HꢀX: acidic additive.
chiral phosphoric acid cocatalyst was needed to induce an anti
diastereoselective pathway. The small influence of the phospho-
ric acid chiral backbone on the enantioselectivity prompted us to
investigate the effect of diphenyl hydrogen phosphate (DPP), an
achiral acid (entry 11). Indeed, DPP led to comparable results, in
terms of efficiency and stereoselectivity, to those obtained with
the chiral acid 6. This further supported the notion that, while the
hydrogen-bonding ability of the phosphate counteranion strongly
influenced the structural assembly of the iminium intermediate
(vide infra), its chirality was not essential for the stereochemical
outcome of the asymmetric transformation.
Interestingly, the opposite configuration of the SMA product
could be accessed by simply selecting the appropriate enantiomer
of the catalyst. Thus, combining the pseudoenantiomeric catalyst
7,27 derived from quinine, with (R)-6, (S)-6 or DPP afforded
anti-4a with opposite absolute configuration while maintaining a
high level of selectivity (entries 12ꢀ14).
Figure 3. Correlation between the catalyst’s functional switch and the
hydrogen-bonding ability of the phosphate anion. When bulkier phos-
phates than (S)-6 are used, a syn-selective pathway is dominant,
presumably due to a weaker binding ability of the anion. Interestingly,
the inductive effect of electron withdrawing groups (p-NO2 and CF3
moieties) at the 3- and 30-substituents of phosphoric acids preserved the
anti-diastereoselectivity. The pKa values are referred to DMSO, as given
in ref 34. †pKa value is not available: inductive effects should presumably
render this compound more acidic than 6.
conformational flexibility in solution,31 and three-dimensional
structural switches can be induced by different chemical stimuli,
such as a solvent change31a or protonation of the N-quinuclidine
moiety.32 Since the catalytic function of the cinchona scaffold is
intimately related to its spatial architecture, a conformational
change may induce a different catalytic behavior. It has already
been shown how modulation of the solvent polarity directly
reflects on the ability of a cinchona catalyst to induce a preferred
handedness in the product formed during an asymmetric trans-
formation.33 Along this line, we have carried out vibrational
circular dichroism (VCD), circular dichroism (CD) and nuclear
magnetic resonance (NMR) spectroscopic analyses of the 5/
ortho-fluorobenzoic acid and 5/diphenyl hydrogen phosphate
combinations (see Figures S7ꢀS28 in the Supporting Information).
These studies provided insight into the conformational behavior
of each catalytic system in chloroform and acetone. Interestingly,
we have found that, whereas the two catalytic salts did not show
clear chiroptical distinctions, a change of the solvent greatly
altered their ground-state conformations.
Mechanistic Considerations. We propose that the same
general mechanism (covalent-based organocatalysis through imi-
nium activation of α-branched enones) is operative in the two
stereodivergent chemical pathways.17,18 The conceptual frame-
work for rationalizing the diastereodivergent behavior of amine 5
is as follows: the addition of acidic additives and the change of the
reaction medium induce a three-dimensional structural modifi-
cation of the catalyst’s structure (Figure 2). Altering the chiral
catalyst conformation may in turn reflect on different catalyst-
substrate specific interactions and distinct transition state struc-
tures that could channel the transformation through diastereo-
divergent pathways.
(ii). Modular Nature of the Ion-Pair Molecular Assembly.
Amine 5, which bears a pendant primary amino moiety, can
effectively condense with hindered carbonyls12,22a in the presence of
an acidic additive to form a covalently bound reactive catalytic
species, the iminium ion intermediate (intermediate I in Figure 2).
The distinctive feature of such an ion pair assembly as a chiral
molecular catalyst is that its stereocontrolling ability can be fine-
tuned by structurally modifying both the cation and the anion.26
We investigated further to evaluate the influence of the
counteranions on the catalyst’s function. Previous observations
indicated that substituting the carboxylate with a phosphate
anion induces the functional change in the ion-pair molecular
assembly I. The results detailed in Figure 3 (see also Table S6
within the Supporting Information) show how the syn to anti
switch of the catalyst’s diastereoselection seems to strongly
depend on the hydrogen-bonding ability of the acid-anion.
Indeed, phosphates are strong H-bond acceptors. (S)-6 or
Changing the reaction conditions, including solvent and tem-
perature,28 or using light,29 has already been used in asymmetric
catalysis to modulate the chiral space in which an enantioselec-
tive catalytic reaction takes place, thus leading to enantiodiver-
gent reaction outcomes. This strategy has provided individual
access to either product enantiomers from a single enantiomer of
the chiral synthetic catalyst. The results reported above suggest
that the synthetic potential of chiral molecules whose catalytic
function changes in response to an external stimulus can be
further expanded to attack longstanding problems in chemical
synthesis: modulating the enforced sense of diastereoselectivity
on demand using a single chiral catalyst.
The following two factors explain the induced modification of
the chiral space where the sulfa-Michael reaction takes place
(Figure 2):
(i). Flexibility of the Cinchona Scaffold. Cinchona alkaloid
derivatives30 of type 5 are characterized by a high degree of
17937
dx.doi.org/10.1021/ja207847p |J. Am. Chem. Soc. 2011, 133, 17934–17941

Journal of the American Chemical Society
ARTICLE
DPP can induce a catalyst conformational change, promoting an
anti-selective reaction manifold. But when bulkier phosphates
(which bind more weakly to the catalyst) are used, a syn-selective
pathway for the SMA reaction is dominant. This is presumably
because of the reduced binding ability of the anion, which cannot
induce the catalyst conformational change.
Table 3. Syn-Selective SMA of α-Branched Enones: Nucleo-
phile and Electrophile Scopea
Together with the conformational flexibility of amine 5, the
use of structurally different counteranions provides distinct chiral
environments in which the enantioselective SMA reaction can
take place. This creates opportunities for designing a programmable
organocatalyst with multiple stereochemical preferences.
We are aware that important information on the active
catalytic species can be obtained by characterizing the covalent
reactive intermediate, the iminium ion assembly I generated by
condensation of the cinchona-based primary amine with the
carbonyl compound. This would provide a more reliable picture
of the mechanism. In general, however, the nature of the ion pairs
in iminium-catalyzed reactions is poorly understood. Its charac-
terization is a challenging and sought-after target for many
practitioners of organocatalysis.35,36 Our attempts to detect the
covalent active intermediate I by CD and NMR spectroscopic
studies have not produced important results to date.37 Further
extensive efforts are needed, combining spectroscopic and
theoretical approaches to detect, analyze, and characterize the
reactive intermediates involved in the catalysis. Investigations
along these lines are underway in our laboratories.
yield
(%)b
syn/
anti
ee
entry
R1
R2 R3
R4
3
(%)
1
2
3
4
5
6
7
8
9
Me
Me Me Ph
a
b
b
c
d
e
f
68
79
50
54
60
68
40
60
56
65
60
68
5.1:1
5.1:1
3.5:1
3.5:1
3.8:1
9.3:1
5.0:1
2.8:1
4.9:1
5.0:1
7.8:1
5.5:1
86
88
77c
85
89
87
81
73
90
89
87
61
4-NO2-C6H4 Me Me Ph
4-NO2-C6H4 Me Me Ph
Ph
Me Me Ph
Me Me Ph
4-Br-C6H4
4-NO2-C6H4 Et Me Ph
4-NO2-C6H4 Me Et Ph
Et
Ph Me Ph
g
h
i
Me
Me Me 4-MeO-C6H4
Me Me 4-Cl-C6H4
Me Me CHdCH2
Me Me CO2Et
10 Me
11 Me
12 Me
j
Reaction Scope of the syn-SMA. Having identified the
chemical stimuli (the acidic additive and the reaction medium)
that allow us to tune the diastereoselectivity of catalyst 5, we
explored the substrate scope of the SMA reaction.
Table 3 summarizes the results obtained in the syn-selective
sulfa-Michael addition/protonation sequence of α-branched enones
k
a Reactions carried out using 3 equiv of 1 and [2]0= 0.5M. Results
represent the average of two runs per substrate. For all the enones 2
used, E/Z ratio >95:5. ee value refers to the major syn-3 compound.
b Yield of the isolated product 3. c Reaction performed at 15 °C over 5
days with catalyst 7, leading to the opposite enantiomer: syn-(3S,4R)-3b.
Table 4. Anti-Selective SMA of α-Branched Enones:
Nucleophile and Electrophile Scopea
entry
R1
R2
R3
R4
4
yield (%)
anti/syn
ee (%)
1
Me
Me
Me
Me
Me
Me
Me
Me
Et
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
a
80 (68)
43 (66)
77 (56)
48
6.2:1 (6.1:1)
3.9:1 (3.0:1)
5.7:1 (4.7:1)
2.8:1
98 (98)
94 (94)b
99 (96)
97b
2
a
b
b
c
d
e
f
3
4-NO2-C6H4
4-NO2-C6H4
CH2Bn
4
5
69
8.2:1
97
6
Ph
41 (44)
35
5.2:1 (5.1:1)
1.8:1
99 (95)
98
7
4-NO2-C6H4
4-MeO-C6H4
4-Br-C6H4
4-Cl-C6H4
2-thiophenyl
CO2Et
8
Me
Me
Me
Me
Me
Me
Me
Me
Me
42
4.2:1
99
9
g
h
i
59 (38)
58 (59)
44
6.5:1 (6.3:1)
5.3:1 (5.3:1)
4.2:1
99 (96)
99 (98)
98
10
11
12
13
14
15
16
j
56
5.2:1
96
Me
4-MeOC6H4
4-Cl-C6H4
CHdCH2
CO2Et
k
l
42 (68)
58 (71)
73
7.1:1 (7.0:1)
7.2:1 (4.2:1)
4.8:1
98 (98)
94 (97)
98
Me
Me
m
n
Me
50 (65)
2.4:1 (2.0:1)
92 (83)
a Results in parentheses refer to the reactions catalyzed by the amine 5/diphenyl phosphoric acid (DPP) combination. Reactions carried out using
2 equiv of 1 and [2]0 = 0.5 M. Results represent the average of two runs per substrate. For all the enones 2 used, E/Z ratio >95:5. When the
β-substituent of enones 2 is an aromatic group, 15 mol % of the phosphoric acid (S)-6 has been used. Other substrates perform better in the presence of
10 mol % of (S)-6. Ee value refers to the major anti-4 compound. Yield of isolated product 4. b Reaction performed with catalyst 7, leading to the opposite
enantiomer of anti-4.
17938
dx.doi.org/10.1021/ja207847p |J. Am. Chem. Soc. 2011, 133, 17934–17941

Journal of the American Chemical Society
ARTICLE
Figure 4. Switching diastereoselection in a single reaction flask and during the course of the reaction. The catalyst’s syn preference is inactivated when
the achiral phosphoric acid (DPP) is added to the reaction medium.38
using acombinationofamine5(20mol%) andortho-fluorobenzoic
acid (40 mol %) using chloroform as the reaction medium.
The method shows a good-to-high level of diastereoselectivity
and enantiocontrol over a wide substrate range. Both aliphatic
and aromatic substituents at the β position are well-tolerated
(entries 1ꢀ5).
The presence of a more sterically demanding group at the
branched R2 α-position or at the R3 position does not influence
the reactivity profile of the catalyst system: the corresponding
syn-adducts 3eꢀg are obtained in good chemical yield with a
high level of stereocontrol (entries 6ꢀ8). Remarkably, a broad
range of alkyl mercaptanes, containing either aromatic or vinyl
moieties, can also be used in the syn-selective SMA reaction
(entries 9ꢀ11). Moreover, a thioglycolate derivative is a suitable
nucleophile (entry 12).
Reaction Scope of the anti-SMA. The anti-selective SMA
protocol, induced by using the phosphoric acid (S)-6 (10 mol %,
1:1 with the amine 5) in acetone solvent, offers a wide substrate
scope for both the electrophilic and nucleophilic components,
affording the desired adducts 4 with synthetically useful diastereo-
meric ratios and excellent enantioselectivities (ee values ran-
ging from 92% to 99%). The results reported in Table 4
demonstrate that a variety of substituents at the β position can be
accommodated. Aromatics with diverse electronic properties
(entries 1ꢀ11), as well as an ester group (entry 12), are tolerated.
Moreover, using different alkyl thiols affords access to optically
active chiral sulfur compounds bearing removable S-protecting
groups (entries 13ꢀ16).14 Importantly, using the achiral acid
DPP in the anti-selective SMA reaction provided comparable
results in terms of efficiency and stereoselectivity. These results
are shown in parentheses in Table 4.
Switching Diastereoselection during the Course of the
Reaction. We then wondered if it would be possible to switch the
diastereoselectivity of catalyst 5 on demand during the course of
the reaction and in the same reaction flask. Initially, we used a
combination of 5 and ortho-fluorobenzoic acid to select the syn-
directing pathway for the reaction between thiol 1a and enone 2b
in chloroform, leading to the adduct syn-3b (Figure 4). Our
challenge was then to switch the catalytic behavior of 5 after the
syn-reaction reached completion. We reasoned that a stronger
acid, namely the achiral phosphoric acid DPP, would reset the
previous function, encoding the diastereo-switching information
within the catalyst. The use of 1:1 molar ratio of phosphoric acid
(pKa ≈ 1.9 in H2O) to amine 5 likely leads to a quantitative
association with the quinuclidine tertiary amine moiety, the most
basic site of the catalyst, thus displacing ortho-fluorobenzoic acid
(pKa = 3.27 in H2O) from the acidꢀbase equilibrium with 5.38
Indeed, after we added DPP as the designer acid, acetone (2:1
ratio to the original chloroform), and a different enone substrate
Figure 5. (a) Accessing the full matrix of the stereoisomers of the SMA
reaction. (b) X-ray structure of syn-3b. (c) X-ray structure of anti-4b.
ORTEP drawing at 50% probability.39
2a in the same flask, an anti-selective addition of thiol 1a to 2a
proceeded to afford anti-4a with a high level of stereocontrol,
testifying to the change in catalyst function. In this experiment,
both syn and anti products were synthesized in the same reaction
flask and using a single catalyst, and then individually isolated by
chromatographic purification.
Achieving the Full Matrix of Possible Stereoisomeric
Products of the SMA. Finally, the ability to selectively channel
the reaction manifolds toward complementary diastereochemical
outcomes using a single chiral catalyst27 was exploited to access
the full matrix of possible stereoisomeric products of the SMA
reaction (Figure 5a). When carried out in chloroform, the ortho-
fluorobenzoic acid/catalyst 5 combination induced a syn selec-
tive outcome of the SMA of 1a to enone 2b (bearing an aryl
β-substituent). In acetone, the phosphoric acid (S)-6 switched
the catalyst’s induction toward anti-selectivity. The same designer
acid-induced diastereo-switching behavior was observed with the
quasi-enantiomeric catalyst 7,27 derived from quinine, thus
allowing access to the full matrix of possible stereoisomeric
products of the SMA reaction.
The relative and absolute configurations of the syn and anti
products 3b and 4b were unambiguously determined by anoma-
lous dispersion X-ray diffraction analysis (Figure 5b and 5c).39
17939
dx.doi.org/10.1021/ja207847p |J. Am. Chem. Soc. 2011, 133, 17934–17941

Journal of the American Chemical Society
ARTICLE
The opposite absolute configuration at the β carbonꢀsulfur
stereogenic center is further evidence of how the catalyst func-
tional switch drastically influences the stereochemical outcome
of the reaction. As the conjugate addition step (in contrast to the
protonation step) must be under the stereocontrol of the catalyst,
this evidence highlights the uncommon potential of the chiral
amine 5 to catalyze two enantio-, as well as diastereodivergent,
SMA reactions when properly influenced by external stimuli
(appropriate acid and solvent).40
Reyes, E. In Organocatalytic Enantioselective Conjugate Addition Reac-
tions; The Royal Society of Chemistry, Cambridge, 2010. (e) Tsogoeva,
S. B. Eur. J. Org. Chem. 2007, 1701–1716.
(2) For a chiral auxiliary-controlled asymmetric tandem sulfa-Michael/
MeerweinꢀPonndorfꢀVerley reduction of α-branched enones, see:
Nishide, K.; Ozeki, M.; Shigeta, Y.; Patra, P. K.; Hagimoto, Y.; Node, M.
Angew. Chem., Int. Ed. 2003, 42, 4515–4517.
(3) Generally, both enantiomers of the synthetic chiral catalysts can
be prepared easily, thereby enabling access to each product enantiomer
individually. See: Blaser, H. U., Schmidt, E., Eds. Asymmetric Catalysis on
Industrial Scale; Wiley-VCH: Weinheim, Germany, 2004.
(4) Wang, B.; Wu, F.; Wang, Y.; Liu, X.; Deng, L. J. Am. Chem. Soc.
2007, 129, 768–769.
’ CONCLUSION
We have documented the possibility of using a single chiral
catalyst to fully control the stereochemical outcome of the asym-
metric conjugate addition of alkyl thiols to α,β-disubstituted un-
saturated ketones,anelusiveclassofMichaelacceptors.Thejudicious
choice of a designer acidic additive and the reaction medium
switches the sense of the catalyst’s diastereoselection, establish-
ing a direct connection between the enantioselective catalyst and
the operator (the chemist). We are currently undertaking further
mechanistic investigations to fully understand the origin of the
diastereoswitching behavior. We believe that programming the
catalytic function of a catalyst using an external chemical stimulus
may provide new synthetic opportunities and conceptual pers-
pectives for successfully attacking major challenges connected
with the preparation of chiral molecules that cannot be addressed
by traditional approaches. We are currently seeking to develop
chiral catalysts that allow for reversible function-switching.
(5) For strategies that use two cyclic specific aminocatalysts to
achieve the modular control of the enforced sense of enantio- and
diastereoinduction in tandem reactions, see: (a) Simmons, B.; Walji,
A. M.; MacMillan, D. W. C. Angew. Chem., Int. Ed. 2009, 48, 4349–4353.
(b) Chi, Y.; Scroggins, S. T.; Frꢀechet, J. M. J. J. Am. Chem. Soc. 2008,
130, 6322–6323.
(6) Yan, X.-X.; Peng, Q.; Li, Q.; Zhang, K.; SounYao, J.; Hou, X.-L.;
Wu, Y.-D. J. Am. Chem. Soc. 2008, 130, 14362–14363.
(7) (a) Lu, G.; Yoshino, T.; Morimoto, H.; Matsunaga, S.; Shibasaki,
M. Angew. Chem., Int. Ed. 2011, 50, 4382–4385. (b) Nojiri, A.; Kumagai,
N.; Shibasaki, M. J. Am. Chem. Soc. 2009, 131, 3779–3784.
(8) Huber, J. D.; Leighton, J. L. J. Am. Chem. Soc. 2007, 129,
14552–14553.
(9) For an example in which the relative amount of the chiral ligand,
with respect to a Lewis acid, induced a diastereodivergent pathway, see:
(a) Bandini, M.; Cozzi, P. G.; Umani-Ronchi, A. Angew. Chem., Int. Ed.
2000, 39, 2327–2330. For an example of solvent-induced diastereo-
switching, see: (b) Northrup, A. B.; MacMillan, D. W. C. Science 2004,
305, 1752–1755. For a recent example based on the use of acidic
additives: (c) Gao, J.; Bai, S.; Gao, Q.; Liu, Y.; Yang, Q. Chem. Commun.
2011, 47, 6716–6718.
(10) For reviews focusing on the development and use of cinchona-
based primary amines, see: (a) Jiang, L.; Chen, Y.-C. Catal. Sci. Technol.
2011, 1, 354–365. (b) Bartoli, G.; Melchiorre, P. Synlett 2008, 1759–
1771. For general reviews on aminocatalysis, see: (c) MacMillan,
D. W. C. Nature 2008, 455, 304–308. (d) List, B. Angew. Chem., Int.
Ed. 2010, 49, 1730–1734. (e) Barbas, C. F., III Angew. Chem., Int. Ed.
2008, 47, 42–47. (f) Melchiorre, P.; Marigo, M.; Carlone, A.; Bartoli, G.
Angew. Chem., Int. Ed. 2008, 47, 6138–6171. For a comprehensive review
on iminium catalysis: (g) Erkkil€a, A.; Majander, I.; Pihko, P. M. Chem.
Rev. 2007, 107, 5416–5470.
’ ASSOCIATED CONTENT
S
Supporting Information. Complete experimental proce-
b
dures, optimization studies, compound characterization, VCD,
CD, and NMR conformational analyses, HPLC traces, and NMR
spectra. This material is available free of charge via the Internet at
http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
pmelchiorre@iciq.es
(11) Sulfa-Michaelreactionofenones:(a)Ricci, P.;Carlone, A.; Bartoli,
G.; Sambri, L.; Melchiorre, P. Adv. Synth. Catal. 2008, 350, 49–53. Tandem
thio-amination reaction of α-branched enals: (b) Galzerano, P.; Pesciaioli,
F.; Mazzanti, A.; Bartoli, G.; Melchiorre, P. Angew. Chem., Int. Ed. 2009,
48, 7892–7894.
Author Contributions
X.T. and C.C. contributed equally to this work.
(12) For the first use of catalyst 5, see: (a) Chen, W.; Du, W.; Duan,
Y.; Wu, Y.; Yang, S.-Y.; Chen, Y.-C. Angew. Chem., Int. Ed. 2007, 46, 7667–
7670. See also: (b) Bencivenni, G.; Galzerano, P.; Mazzanti, A.; Bartoli, G.;
Melchiorre, P. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 20642–20647.
(13) For a review, see: (a) Enders, D.; L€uttgen, K.; Narine, A.
Synthesis 2007, 959–980. For recently published asymmetric catalytic
SMA of enones, see: (b) Dai, L.; Wang, S.-X.; Chen, F.-E. Adv. Synth.
Catal. 2010, 352, 2137–2141. (c) Rana, N. K.; Selvakumar, S.; Singh,
V. K. J. Org. Chem. 2010, 75, 2089–2091. See also: (d) Marigo, M.;
Schulte, T.; Franzꢀen, J.; Jørgensen, K. A. J. Am. Chem. Soc. 2005, 127,
15710–15711. For theoretical investigations, see: (e) Krenske, E. H.;
Petter, R. C.; Zhu, Z.; Houk, K. N. J. Org. Chem. 2011, 76, 5074–5081.
(14) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 3rd ed.; Wiley-VCH: New York, 1999; Chapter 6.
(15) Control experiments, detailed in Figures S29ꢀ30 of the
Supporting Information, indicated the anti-4a adduct as the thermo-
dynamic product. No epimerization was observed when the products
syn-3a and anti-4a were mixed with the catalyst 5 under the reaction
conditions.
’ ACKNOWLEDGMENT
This work is dedicated to Prof. Carlo Melchiorre on the
occasion of his retirement. Research support from the Institute of
Chemical Research of Catalonia (ICIQ) Foundation and from
MICINN (grant CTQ2010-15513 and Consolider Ingenio
2010, grant CSD2006-0003) is gratefully acknowledged. Y.L. is
grateful for a Marie Curie fellowship (grant FP7-PEOPLE-2010-
IIF n.273088). The authors thank Eduardo Escudero-Adꢀan for
X-ray crystallographic analysis and Grace Fox for proofreading
the manuscript.
’ REFERENCES
(1) (a) Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds. Comprehensive
Asymmetric Catalysis; Springer: Berlin, 1999. (b) Catalytic Asymmetric
Conjugate Reactions; Cꢀordova, A., Ed.; Wiley-VCH Verlag GmbH:
Weinheim, 2010. (c) Christoffers, J.; Koripelly, G.; Rosiak, A.; R€ossle,
M. Synthesis 2007, 1279–1300. (d) Vicario, J. L.; Badía, D.; Carrillo, L.;
17940
dx.doi.org/10.1021/ja207847p |J. Am. Chem. Soc. 2011, 133, 17934–17941

Journal of the American Chemical Society
ARTICLE
(16) In the early 1980s, Wynberg’s pioneering work showed how the
basic bridgehead nitrogen in the quinuclidine core of the cinchona
derivatives can be used in the context of the asymmetric SMA reaction to
activate thiols via a general base catalysis mechanism, see: (a) Hiemstra,
H.; Wynberg, H. J. Am. Chem. Soc. 1981, 103, 417–430. For an asymmetric
SMA of alkyl thiols catalyzed by cinchona derivatives, see: (b) Liu, X.; Sun,
B.; Wang, B.; Wakem, M.; Deng, L. J. Am. Chem. Soc. 2009, 131, 418–419.
(17) Amine 5 is characterized by a dichotomous catalytic behavior.
This organic molecule can execute its catalytic functions using both
covalent and noncovalent-based organocatalysis. The result obtained
when using amine 5 as a general base activator (entry 3, Table 1)
suggested that a chiral molecule, which can use completely distinct
modes of substrate activation (in this case, covalent and noncovalent
catalysis), could provide an appropriate strategy for inducing mecha-
nistically unrelated and (dia)stereodivergent reaction pathways. It is a
remarkable coincidence that the free base catalyst 5 (a general base
catalyst) gives rise to similar selectivity with respect to the 5/ortho-
fluorobenzoic acid system (operating under iminium activation, com-
pare entries 3 and 8 in Table 1).
diastereoisomers), allowing access to both antipodes of the chiral
product. They act as the two enantiomers of a single chiral catalyst.
(28) For recent examples: (a) Sohtome, Y.; Tanaka, S.; Takada, K.;
Yamaguchi, T.; Nagasawa, K. Angew. Chem., Int. Ed. 2010, 49, 9254–
9257. (b) Messerer, M.; Wennemers, H. Synlett 2011, 499–502.
(29) Wang, J.; Feringa, B. L. Science 2011, 331, 1429–1432.
(30) Song, C. E. Cinchona Alkaloids in Synthesis and Catalysis; Wiley-
VCH: Weinheim, Germany, 2009.
(31) (a) B€urgi, T.; Baiker, A. J. Am. Chem. Soc. 1998, 120, 12920–
12926. (b) Dijkstra, G. D. H.; Kellogg, R. M.; Wynberg, H.; Svendsen,
J. S.; Marko, I.; Sharpless, K. B. J. Am. Chem. Soc. 1989, 111, 8069–8076.
(c) Urakawa, A.; Meier, D. M.; R€uegger, H.; Baiker, A. J. Phys. Chem. A
2008, 112, 7250–7255.
(32) Olsen, R. A.; Borchardt, D.; Mink, L.; Agarwal, A.; Mueller, L. J.;
Zaera, F. J. Am. Chem. Soc. 2006, 128, 15594–15595.
(33) Aune, M.; Gogoll, A.; Matsson, O. J. Org. Chem. 1995, 60,
1356–1364.
(34) Christ, P.; Lindsay, A. G.; Vormittag, S. S.; Neud€orfl, J.-M.;
Berkessel, A.; O’Donoghue, A. C. Chem.—Eur. J. 2011, 17, 8524–8528.
(35) For the characterization of covalent reactive intermediates of
organocatalysis formed by condensation of chiral secondary amines with
aldehydes, see: (a) Seebach, D.; Groselj, U.; Badine, D. M.; Schweizer,
W. B.; Beck, A. K. Helv. Chim. Acta 2008, 91, 1999–2034. (b) Schmid,
M. B.; Zeitler, K.; Gschwind, R. M. J. Am. Chem. Soc. 2011, 133, 7065–
7074.
(18) General base catalysis is inhibited when an acidic additive is
present. A natural cinchona alkaloid, such as quinidine, is an active yet
nonselective catalyst for the SMA reaction (operating via general base
catalysis). However, the addition of benzoic acid resulted in a completely
inactive system. See Figure S2 within the Supporting Information for
more details.
(19) (a) Evans, G. J. S.; White, K.; Platts, J. A.; Tomkinson, N. C. O.
Org. Biomol. Chem. 2006, 4, 2616–2627. (b) Hine, J. Acc. Chem. Res.
1978, 11, 1–7.
(20) For a review on the use of chiral binaphthol-derived phosphoric
acid in catalytic asymmetric reactions, see: (a) Akiyama, T. Chem. Rev.
2007, 107, 5744–5758. (b) Rueping, M.; Kuenkel, A.; Atodiresei, I.
Chem. Soc. Rev. 2011, 40, 4539–4549.
(36) In the present case, the formation of the iminium intermediate
is greatly hampered by the relatively low nucleophilicity of the primary
amine moiety within the catalyst 5 and the steric hindrance of the carbonyl
reactive center of the enone. The characterization of the iminium ion
assembly I is further complicated by the flexibility of the cinchona scaffold.
Indeed, there is a dearth of information about the active conformer of the
cinchona-based primary amine catalyst.
(21) During the extensive optimization studies, we tested acetone as
the reaction medium, fully aware of the possible secondary pathways that
such a solvent might initiate. This was crucial, because acetone greatly
improved the diastereo-induction of the process. However, we always
detected a small amount (from 10 to 20%) of a byproduct derived from
the SMA addition of thiols 2 to the in situ-formed enone (4-methylpent-
3-en-2-one) generated by 5-catalyzed self-aldol-condensation of ace-
tone. More details are reported in Supporting Information Figure S4.
(22) The cinchona-based primary amine-phosphoric acid combina-
tion has recently emerged as a powerful catalyst for the functionalization
of α-branched enals, see: (a) Bergonzini, G.; Vera, S.; Melchiorre, P.
Angew. Chem., Int. Ed. 2010, 49, 9685–9688. (b) Lifchits, O.; Reisinger,
C. M.; List, B. J. Am. Chem. Soc. 2010, 132, 10227–10229.
(37) During the CD spectroscopic studies on the two catalytic salts
(detailed in the Supporting Information) we tried to detect the iminium
ion formation after addition of a large excess of an aliphatic α-branched
enone. This attempt, however, did not produce any appreciable change
in the CD spectra. Moreover, it was not possible to detect the covalent
intermediate by NMR spectroscopy.
(38) A control experiment (Supporting Information Figure S6)
revealed that catalyst 5 (20 mol%), when mixed with 40 mol% of
ortho-fluorobenzoic acid and 20 mol% of 6 in acetone, is programmed
for an anti-directing function. Indeed, the reaction between 1a and 2a
afforded the anti adduct 4a with 4.6:1 dr and 97% eeanti. As reported in
Supporting Information Figure S5, the chiral acid (S)-6 can also be used
to switch the diastereoselectivity of catalyst 5 during the course of the
reaction, in a similar experiment to that detailed in Figure 4.
(39) Crystallographic data for compounds 3b and 4b are available
free of charge from the Cambridge Crystallographic Data Centre,
accession numbers CCDC 804889 and CCDC 804888, respectively.
(40) The fact that the products are not diastereomeric at the α but at
the β carbon (the site of the initial nucleophilic attack) is rather
intriguing. This inspired us to carefully consider an alternative explana-
tion for the observed stereochemical outcome, specifically that the
switch of the catalyst functions is connected with completely unrelated
mechanisms of catalysis. In ref 17, we have already commented on the
potential of amine 5 to use completely distinct modes of catalysis for
activating the reagents of the SMA reaction and, more importantly, on
the possibility of controlling its catalytic functions by applying an
external stimulus: indeed, an acidic additive could be used to modulate
the catalyst behavior by switching the catalytic potential from base
catalysis (activation of the thiol) to iminium activation of the enone.
However, as commented in ref 18, a general base catalysis mechanism
under acidic conditions (in the presence of the 2-F-benzoic acid) is not
very plausible. At the present stage of investigation, we consider that the
more plausible mechanistic picture is the one described within the main
text, where the same general mechanism (iminium activation) is
operative in the two stereodivergent chemical pathways.
(23) Marcelli, T.; van Maarseveen, J. H.; Hiemstra, H. Angew. Chem.,
Int. Ed. 2006, 45, 7496–7504.
(24) For an excellent overview on the importance of attractive
noncovalent interactions in asymmetric catalysis, see: Knowles, R. R.;
Jacobsen, E. N. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 20678–20685.
(25) Reviews on Brønsted acid catalysis: (a) Kampen, D.; Reisinger,
C. M.; List, B. Top. Curr. Chem. 2010, 291, 395–456. (b) Doyle, A. G.;
Jacobsen, E. N. Chem. Rev. 2007, 107, 5713–5743. (c) Yamamoto, H.;
Payette, N. In Hydrogen Bonding in Organic Synthesis; Pihko, P. M., Ed.;
Wiley-VCH: Weinheim, Germany, 2009; pp 73ꢀ140.
(26) The foundations for rationalizing the influence of chiral coun-
teranion on both the reactivity and stereoselectivity of iminium catalysis
have been established. List and co-workers have recently introduced
asymmetric counterion directed catalysis (ACDC) as an efficient
strategy for enantioselective transformations that proceed via cationic
species, including iminium-ion intermediates, see: (a) Mayer, S.; List, B.
Angew. Chem., Int. Ed. 2006, 45, 4193–4195. (b) Martin, N. J. A.; List, B.
J. Am. Chem. Soc. 2006, 128, 13368–13369. For the combination of
cinchona-based primary amines with chiral acids, see ref 22 and also: (c)
Bartoli, G.; Bosco, M.; Carlone, A.; Pesciaioli, F.; Sambri, L.; Melchiorre,
P. Org. Lett. 2007, 9, 1403–1405.
(27) The cinchona catalysts 5 and 7 derived from natural quinidine
and quinine constitute a pseudoenantiomeric pair (formally they are
17941
dx.doi.org/10.1021/ja207847p |J. Am. Chem. Soc. 2011, 133, 17934–17941