A Designed Approach to Enantiodivergent Enamine Catalysis

Article abstract of DOI:10.1002/anie.201703919

The rational design and implementation of enantiodivergent enamine catalysis is reported. A simple secondary amine catalyst, 2-methyl-l-proline, and its tetrabutylammonium salt function as an enantiodivergent catalyst pair delivering the enantiomers of α-functionalized aldehyde products in excellent enantioselectivities. This novel concept of designed enantiodivergence is applied to the enantioselective α-amination, aldol, and α-aminoxylation/α-hydroxyamination reactions of aldehydes.

Full text of DOI:10.1002/anie.201703919

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: A Designed Approach to Enantiodivergent Enamine Catalysis
Authors: Mathew Vetticatt, Juliet Macharia, Victor Wambua, Yun Hong,
Lawrence Harris, Jennifer S Hirschi, and Gary B. Evans
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201703919
Angew. Chem. 10.1002/ange.201703919
Link to VoR: http://dx.doi.org/10.1002/anie.201703919
http://dx.doi.org/10.1002/ange.201703919

10.1002/anie.201703919
Angewandte Chemie International Edition
COMMUNICATION
A Designed Approach to Enantiodivergent Enamine Catalysis
Juliet Macharia,[a] Victor Wambua,[a] † Yun Hong,[a] † Lawrence Harris,[b] † Jennifer S. Hirschi,[a] Gary B.
Evans, [b] and Mathew J. Vetticatt* [a]
enantioselective synthesis.3 Organocatalytic reactions are
Abstract: The rational design and implementation of
classified based on ‘activation modes’ i.e. the mechanism of
enantiodivergent enamine catalysis is reported. A simple secondary
activation of the reacting substrate.4 One of these activation
amine catalyst, 2-methyl-L-proline, and its tetrabutylammonium salt
modes – enamine catalysis – effects the enantioselective α-
functionalization of carbonyls. In this mode of catalysis a simple
chiral organocatalyst such as L-proline (2) reacts with an
function as an enantiodivergent catalyst pair delivering opposite
enantiomers of α-functionalized aldehyde products in excellent
enantioselectivities.
This
novel
concept
of
‘designed
aldehdye (1) to form a transient enamine carboxylic acid
intermediate (3), which can react with a variety of electrophiles
(4) in a series of mechanistically related reactions to deliver an
array of enantioenriched α-functionalized aldehydes (5). Each
one of these reactions proceed via a key stereodetermining
transition state (TS) – known as the Houk-List TS (6) – where
the carboxylic acid proton directs the approach of 4 to one face
of the prochiral enamine via hydrogen (H)-bonding interactions.5
enantiodivergence’ is applied to the enantioselective α-amination,
aldol and α-aminoxylation/α-hydroxyamination reactions of
aldehydes.
The central paradigm in asymmetric catalysis dictates that
either enantiomer of a chiral catalyst/ligand can be utilized to
gain selective access to either antipode of the desired chiral
product. An obstacle in asymmetric synthesis is that many of
these optically active catalysts are obtained from ‘chiral pool’
molecules such as alkaloids, amino acids, sugars, terpenes, etc.
that are readily available as only one enantiomer.1 Access to the
other enantiomer is often quite difficult and sometimes requires
lengthy synthetic sequences. An attractive solution to this well-
known limitation of asymmetric catalysis is the concept of
enantiodivergent catalysis – an approach that utilizes a single
enantiomer of a chiral catalyst/ligand to gain selective access to
both enantiomers of the chiral product. Isolated examples have
garnered significant attention;2 however, a ‘designed’ approach
to enantiodivergent catalysis is an unaddressed goal.
RE
RE
RE
RE
10 mol% 2
9 mol% DBU
O
RE
NH
N
NH
N
O
O
10 mol% 2
N
N
+
H
O
O
H
H
RE
N
via
N
via
O
3a'
OH
3a
(S)-5a 46% ee
1a
4a
RE = -CO2Et
(R)-5a 85% ee
We envisioned that the rational design of an enantiodivergent
reaction within a particular activation mode (such as enamine
catalysis) could lead to a diverse array of mechanistically related
enantiodivergent chemistries – a valuable addition to the rich
toolbox of enantioselective synthesis – and provide “proof-of-
principle” for the logical development of enantiodivergent
catalysis. The basis for our work was a 2010 study by
Blackmond and Armstrong, which reports an unusual reversal of
enantioselectivity in the α-amination reaction of propanal (1a)
O
O
Y
E
YH
E
YH
E
O
O
N
4
N
O
+H2O
O
+
OH
H
N
H
H
-H2O
OH
R
R
R
2
R
1
2
3
5
typical electrophiles
CO2R
and
diethylazodicarboxylate
(4a,
RE=CO2Et)
in
O
PG
Houk-List
TS
O
Ph
N
dichloromethane.6 Utilizing 10 mol% of 2 as catalyst, 85% ee of
the R-product 5a was obtained. Addition of 9 mol% of a tertiary
amine base, such as diazobicycloundecane (DBU), to the
reaction mixture results in an unexpected reversal in facial
selectivity, now delivering the (S)-product 5a in 46% ee. In a
series of follow-up kinetic and NMR studies, Blackmond and
Armstrong established that the key intermediate in the reaction
involving DBU additive is a prolinate species – enamine
carboxylate 3a´.7 Sunoj and co-workers used density functional
theory (DFT) calculations to show that (S)-5a is the favored
product when prolinate 3a´ is the key intermediate in the
catalytic cycle.8 However, this study did not address the origin of
this reversed selectivity in prolinate catalysis compared to
proline catalysis.
N
Y
E
N
O
H
N
O
N
RO2C
CO2R
Y
E
RO2C
R
R
6
The field of asymmetric organocatalysis – catalysis by small
organic molecules – has witnessed explosive growth in the past
two decades and is currently one of the main branches of
[a]
J. Macharia, V. Wambua, Y. Hong, Dr. J. S. Hirschi, Prof. Dr. M. J.
Vetticatt
Department of Chemistry
Binghamton University
Binghamton, New York 13902 (USA)
E-mail: vetticatt@binghamton.edu
Homepage: http://www.vetticattgroup.com/
The origin of enantioselective formation of (R)-5a in proline
catalysis is well understood based on the Houk-List TS model. A
detailed transition state analysis of the origin of modest
enantiodivergence (46% ee of (S)-5a) in the Blackmond
prolinate reaction is the first step in our enantiodivergent catalyst
design strategy. This analysis will provide design principles for a
prolinate derivative that catalyzes a highly (S)-selective reaction.
Based on the mechanism for proline catalysis, the protonated
analogue of this ‘designed’ catalyst will mediate the (R)-selective
reaction via a TS similar to 6. Thus, TS analysis will lead to a
truly enantiodivergent catalyst – defined here as a single
[b]
Dr. L. Harris, Prof. Dr. G. B. Evans
Ferrier Research Institute
Victoria University of Wellington
Wellington, NZ
E-mail: gary.evans@vuw.ac.nz
†
The authors contributed equally to this manuscript
This material is based on work supported by startup funding at
Binghamton University from the SUNY Research Foundation and
the National Science Foundation through XSEDE resources
provided by the XSEDE Science Gateways program.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201703919
Angewandte Chemie International Edition
COMMUNICATION
enantiomer of chiral catalyst that delivers >90% ee for both
enantiomeric products. We report herein, the rational design of a
simple L-proline derivative that catalyzes the perfectly
enantiodivergent α-amination reaction of aldehydes. From this
“proof-of-principle” example, we extend this approach to other
reactions within the enamine catalysis framework.
incorporation of an H-bond donor, such as a hydroxyl group, at
C-4 of L-prolinate and trans to the carboxylate. This prolinate
catalyst (7´, Figure 3) is easily generated from commercially
available 4-trans-hydroxy L-proline (7) and
a base. The
hypothesis is that while the carboxylate moiety of 7´ destabilizes
‘front attack’ of the electrophile (TSR-anti and TSS-syn), the potential
of the 4-hydroxyl group to form a stronger H-bond in TSS-anti
compared to TSR-syn (since 4a can H-bond better to the hydroxyl
group while approaching the anti enamine due to proximity) will
lead to increased selectivity for (S)-5a.12 Accordingly, transition
structures analogous to those shown in Figure 1 were located
using 7´ as the catalyst. Consistent with our hypothesis the
energy difference between TSS-anti and TSR-syn increased from
0.9 kcal/mol (2´ as catalyst) to 1.4 kcal/mol (7´ as catalyst),
leading to an increase in the predicted selectivity from 63% ee to
78% ee (S) (Figure 3).
TSS-syn
syn-3a'
si-approach
TSS-anti
anti-3a'
si-approach
TSR-syn
syn-3a'
re-approach
TSR-anti
anti-3a'
re-approach
CO2
CO2
CO2
CO2
4a
N
N
N
N
4a
4a
4a
(S)-5a
(R)-5a
Figure 1. TS analysis of the prolinate catalyzed α-amination reaction of
propanal (1a) and diethylazodicarboxylate (4a, RE= CO2Et)
Transition structures for the approach of 4a to the re- and si-
face of anti-3a´ and syn-3a´ (Figure 1) were identified using
disfavors syn-enamine
HO
formation: making
TSS-syn and TSR-syn
energetically inaccesible
directs electrophile approach
from face opposite
the carboxylate: favors
TSS-anti and TSR-syn
Me
B3LYP/6-31+G** calculations with
a polarizable continuum
X+
CO2
N
H
model (PCM) for dichloromethane as implemented by Gaussian
09.9,10,11 Approach of 4a from the same face as the carboxylate –
re-face of anti-3a´ (TSR-anti) and si-face of syn-3a´ (TSS-syn) – is
disfavored due to repulsive electrostatic interactions between
the carboxylate and the developing negative charge on 4a
during carbon-carbon bond formation. This unfavorable
approach is highlighted by the obvious proximity of the areas of
increased electron density (red) in the electrostatic potential map
of TSR-anti (Figure 2). Consistent with the Sunoj study,8 the lowest
energy structures leading to the two enantiomers of product
result from the approach of 4a from the face opposite to the
HO
9'
Me
X+
X+
CO2
N
H
CO2
N
H
7'
TSS-anti
8'
destabilizes electrophile
approach from same
face as carboxylate: disfavors
O
Y
N
TSR-anti and TSS-syn
E
O
X+
CO2
N
H
Key TS in
prolinate catalysis
X+= DBUH+
2'
Relative E+zpe in kcal/mol
predicted ee (%)
catalyst
carboxylate moiety. The electrostatic potential map for TSS-anti
,
TSS-syn TSS-anti TSR-syn TSR-anti
the lowest energy transition structure, involves approach of 4a to
the si-face of anti-3a´ and exhibits areas of increased electron
density on opposite sides of the pyrrolidine ring – minimizing the
electrostatic repulsion between areas of increased electron
density (Figure 2). The modest enantioselectivity observed in
this reaction is attributable to the 0.9 kcal/mol energy difference
between TSS-anti and TSR-syn (Figure 3). This energy difference
corresponds to a predicted ee of 63% for (S)-5a, which is in
reasonable agreement with the experimental 46% ee (S).
0.0
0.0
0.0
0.0
0.9
1.4
5.2
5.3
3.4
7.6
2.1
3.3
2'
7'
8'
9'
5.8
12.7
11.2
11.6
63 (S)
78 (S)
95 (S)
99 (S)
Figure 3. Catalyst design considerations for highly enantioselective prolinate
catalysis. Bolded and underlined values represent energies of lowest energy
transition structures for each enantiomer. Transition structures modeled
without counterion using E-Y = 4a (RE = -CO2Et)
The second strategy to obtain high selectivity for (S)-5a
involves a prolinate catalyst with an additional substituent at the
2-position, 2-methyl-L-prolinate (8´). We predicted the second
substituent at the 2-position would essentially shut down the
formation of the syn-enamine pathway (due to unfavorable steric
interactions) thereby destabilizing TSR-syn and TSS-syn (Figure 3).
The enantioselectivity will then result from
a competition
between TSS-anti and TSR-anti. Since the carboxylate moiety of 8´
destabilizes TSR-anti (vide supra), TSS-anti becomes the dominant
pathway using this catalyst. Consistent with this hypothesis,
transition structures located using 8´ as catalyst reveal a 2.1
kcal/mol difference in energy between TSS-anti and TSR-anti
,
leading to a prediction of 95% ee for (S)-5a. Also consistent with
our proposal, the transition structures involving syn-3a´ are
energetically inaccessible using 8´ as catalyst (Figure 3). The
third and final strategy involves a combination of the above two
described approaches i.e. the use of 4-hydroxy-2-methyl-L-
prolinate (9´) as a catalyst for the highly selective formation of
(S)-5a. With the carboxylate group destabilizing approach of 4a
from the ‘front side’, the 2-methyl group virtually eliminating both
syn transition states, and the 4-trans-hydroxy group stabilizing
Figure 2. Electrostatic potential surfaces demonstrate the origin of facial
selectivity in prolinate catalysis
This transition state analysis provides vital clues for the
design of
a
prolinate catalyst that will deliver high
enantioselectivity for (S)-5a. The first strategy involves
This article is protected by copyright. All rights reserved.

10.1002/anie.201703919
Angewandte Chemie International Edition
COMMUNICATION
4a via H-bonding, the only viable transition state using 9´ as
catalyst should be TSS-anti. This hypothesis is confirmed by our
calculations, which predict a 3.3 kcal/mol energy difference
between TSS-anti and TSR-anti, a predicted 99% ee for (S)-5a
(Figure 3).
using
the
2-methyl-L-proline/2-methyl-L-prolinate
(8/8´[X=N(Bu)4]) catalyst pair. Using dibenzylazodicarboxylate
(4a, RE = CO2Bn) as the electrophile, reactions were performed
using propanal, isovaleraldehyde and 3-phenylpropanal (Tables
2, entries 1-3 & 6-8). Diisopropyl and diethyl azodicarboxylates
were then tested as electrophiles for the reaction with 3-
phenylpropanal as the aldehyde (Table 2, entries 4-5 & 9-10). In
reactions utilizing 8 as the catalyst, good isolated yields of (R)-
10a were obtained in >90% ee for each of the reactant pairs,
while for reactions utilizing 8´ X=N(Bu)4 as catalyst, >90% ee for
(S)-10a were obtained in good yields for all except one substrate
pair (entry 7).14 Comparison of entries 1-5 to entries 6-10
establish that catalyst pair 8/8´[X=N(Bu)4] functions as near-
perfect enantiodivergent catalysts for the α-amination reaction
for all tested reactant pairs.
Table 1. Optimization of enantioselective prolinate catalysis
O
O
RE
RE
1. prolinate catalyst
N
R
N
H
O
+
N
N
H
2. NaBH4, MeOH
3. 0.5 M NaOH
RE
R
1
4a
(S)-10a
R = Bn RE = -CO2Et
catalysta
entry
yieldb (%)
eec (%)
27 (S)
54 (S)
1
2
58
59
2'[X=DBUH]
Next, we sought to establish the generality of ‘designed
enantiodivergence’ using the enantiodivergent catalyst pair 8/8´
to test another reaction within the enamine catalysis framework.
In 2002, Jorgensen and co-workers reported the L-proline
catalyzed reaction between propanal (1, R=Me) and
7'[X=DBUH]
3
4
5
6
7
61
51
61
60
84 (S)
80 (S)
94 (S)
93 (S)
8'[X=DBUH]
8'[X=N(Me)4]
8'[X=P(Ph)3Et]
8'[X=N(Bu)4]
8'[X=N(Bu)4] @(-20°C)
diethylketomalonate
(4b),
giving
high
yields
and
94 (S)
93 (S)
92 (S)
enantioselectivities of the aldol product (S)-5b.15 We performed
this reaction using both 8 and 8´[X=N(Bu)4] as catalyst in
dichloromethane. Reactions employing 8 as catalyst result in
82% yield and 93% ee for (S)-5b and the switch to 8´[X=N(Bu)4]
results in a complete reversal in enantioselectivity, giving 84%
yield and 82 % ee for (R)-5b (Figure 4a). This result suggests
that this novel approach to ‘designed’ enantiodivergence has
potentially broad scope in proline-enamine catalysis.
64
53
61
8
9
8'[X=N(Bu)4] (3 mol%)
8'[X=N(Bu)4]
(10 mol% AcOH additive)
8 + 10 mol% [N(Bu)4]+[OAc]-
10
60
51
75 (S)
96 (S)
11 9'[X=N(Bu)4]
a 10 mol% catalyst at 0°C unless otherwise noted b isolated yield
of derivatized product 10a c determined by chiral HPLC using
Chiralcel AD-H column
Table 2. Scope of enantiodivergent α-amination reaction
O
We then tested our computational predictions using the
prolinate-catalyzed α-amination of 3-phenylpropanal (1, R=Bn)
and diethylazodicarboxylate (4a, RE = CO2Et) as a model
reaction (Table 1). The computational trend is qualitatively
reproduced with an experimentally observed increase in ee for
(S)-5a upon going from 2´ to 7´ to 8´ (Table 1, entries 1-3), when
10 mol% of catalyst is used as a pre-formed DBU salt in
reactions performed at 0°C in dichloromethane. The 84% ee
obtained using 2-methyl-L-prolinate (8´, [X=DBUH]) is the
highest reported enantioselectivity to date, for this reaction,
using a prolinate catalyst. We proceeded to explore the effect of
varying the counter-ion of 8´. While the tetramethylammonium
salt of 8´ gives slightly reduced ee (entry 4), the
tetrabutylammonium and ethyltriphenylphosphonium salts of 8´
give 93 and 94% ee for (S)-5a (entries 5-6). Based on these
results and considering the cost and ease of preparation, we
chose 8´[X=N(Bu)4] as the catalyst of choice for further
optimization of reaction conditions. Lowering the temperature to
O
RE
RE
N
R
1. 10 mol% catalyst, 0°C
N
H
N
+
N
O
H
RE
2. NaBH4, MeOH
3. 0.5 M NaOH
R
1
4a
10a
RE
entry
R
yielda (%)
time (h)
ee (%)
catalyst = 8, CH2Cl2
94c (R)
96c (R)
1
2
Me
i-Pr
Bn
CO2Bn
CO2Bn
CO2Bn
CO2iPr
CO2Et
62
64
20
21
92b (R)
92b (R)
93b (R)
3
4
5
69
63
63
16
20
23
Bn
Bn
catalyst = 8' [X = N(Bu)4], CHCl3
91c (S)
85c (S)
6
7
70 (88)
66 (81)
Me
i-Pr
Bn
1
CO2Bn
CO2Bn
CO2Bn
CO2iPr
CO2Et
0.75
91b (S)
96b (S)
92b (S)
8
68 (83)
59
1
1
9
-20°
enantioselectivity. Lowering the catalyst loading to 3 mol%
results in slightly diminished yields but identical
C
leads to
a
minimal increase in yield and
Bn
10
68
0.5
Bn
a isolated yield of derivitized product 10a; yields in parenthesis are
isolated yield of NaBH4 reduced 5a. ee determined by chiral HPLC
using bChiralcel AD-H column or cChiralcel OD-H column.
enantioselectivity. Addition of 10 mol% acetic acid as additive
gives almost identical results compared to entry 6; however, the
reaction is complete in 10 minutes as compared to the 0.5 h for
entry 6. In-situ generation of the prolinate salt 8´[X=N(Bu)4] by
using 10 mol% each of 8 and tetrabutylammonium acetate
resulted in similar yield but slightly diminished enantioselectivity
(entry 10). Finally, consistent with our theoretical predictions,
catalyst 9´[X=N(Bu)4] gives the highest ee (entry 11). However,
the protonated analogue (9) produces only 30% ee and sluggish
reaction rates for the formation of (R)-5a.13 We therefore
proceeded with 8/8´[X=N(Bu)4] as the catalyst pair of choice to
examine the substrate scope of our newly discovered
enantiodivergent chemistry.
Finally, we tested the reaction of propanal (1, R=Me) with the
ambident electrophile nitrosotoluene (4c, Ar=2-tolyl). Proline and
other enamine catalysts possessing an acidic functionality
typically give α-aminoxylation product (5c from O-addition).16 On
the other hand, enamine catalysts lacking an acidic functional
group deliver the α-hydroxyamination product (5c´ from N-
addition).17 Based on these observations in the literature we
speculated that in the reaction of 1 and 4c, the catalyst pair 8/8´
would not only switch the facial selectivity of α-addition of 4c but
also the chemoselectivity. The rationale is that since
possesses a carboxylic acid proton it will facilitate O-attack and
since 8´ lacks this acidic functional group, N-addition will result.
8
To establish the substrate scope, we performed the α-
amination of five distinct aldehyde-alkyl diazodicarboxylate pairs
This article is protected by copyright. All rights reserved.

10.1002/anie.201703919
Angewandte Chemie International Edition
COMMUNICATION
Consistent with our assumption, catalyst 8 gives 66% yield and
99% ee of (R)-aminoxylation product (5c, Ar=2-tolyl) as the
exclusive product; while catalyst 8´[X=N(Bu)4] gives a 64% yield
and 98% ee for enantio- and chemodivergent (S)-
hydroxyamination product (5c´, Ar=2-tolyl) as the exclusive
product (Figure 4b). Though chemodivergence has previously
been achieved for this reaction using the Hayashi/Jørgensen
catalyst (in the presence and absence of external acid
additive),16c,17b this is the first example where both enantio- and
chemodivergence is simultaneously achieved using a single
chiral catalyst.
Yamaguchi, S. Tanaka, K. Nagasawa, Org. Biomol. Chem. 2013, 11,
2780. (k) S. Mortezaei, N. R. Catarineu, X. Duan, C. Hu, J. W. Canary,
Chem. Sci. 2015, 6, 5904.
[3]
Science of Synthesis: Asymmetric Organocatalysis I & II; B. List, K.
Maruoka, Eds., Thieme: Stuttgart, Germany, 2012.
[4]
[5]
D. W. C. Macmillan, Nature, 2008, 455, 304.
S. Bahmanyar, K. N. Houk, H. J. Martin, B. List, J. Am. Chem. Soc.
2003, 125, 2475.
[6]
[7]
D. G. Blackmond, A. Moran, M. Hughes, A. Armstrong, J. Am. Chem.
Soc. 2010, 132, 7598.
(a)
O
OH
O
O
+
10 mol% catalyst
RE
RE
H
RE
RE
4b
H
(a) J. E. Hein, A. Armstrong, D. G. Blackmond, Org. Lett. 2011, 13,
4300. (b) J. E. Hein, J. Burés, Y-H. Lam, M. Hughes, K. N. Houk, A.
Armstrong, D. G. Blackmond, Org. Lett. 2011, 13, 5644. (c) J. Burés, A.
Armstrong, D. G. Blackmond, Chem. Sci. 2012, 3, 1273.
CH2Cl2 , 0°C
RE = CO2Et
R
5b
1
catalyst = 8' [X = N(Bu)4], 40 min
catalyst = 8, 6h
82% yield
93% ee (S)
84% yield
82% ee (R)
[8]
[9]
A. K. Sharma, R. B. Sunoj, Chem. Comm. 2011, 47, 5759.
A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
(b)
Ar
OH
O
NH
O
O
O
O
N
10 mol% catalyst
or
Ar
+
N
H
CH2Cl2
Ar=2-tolyl
Ar
[10] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999.
1
4c
(R)-5c
(S)-5c'
[11] M. J. Frisch, et al. Gaussian 09; Gaussian, Inc.: Wallingford, CT, 2013.
catalyst = 8, 0°C, 30 min
5c:5c' > 99:1
66% yield
catalyst = 8' [X = N(Bu)4], -20°C, 3h
5c:5c' > 1:99
64% yield
[12] A similar strategy has been used in C. Palomo, S. Vera, I. Velilla, A.
Mielgo, E. Gomez-Bengoa, Angew. Chem. Int. Ed. 2006, 45, 5984;
Angew. Chem. 2006, 118, 6130.
99% ee [(R)-5c]
98% ee [(S)-5c']
Figure 4. Prototypical examples of an (a) enantiodivergent aldol reaction of
diethylketomalonate with propanal and (b) enantio- and chemodivergent
addition of nitrosotoluene to propanal
[13] We believe that the catalyst
9 forms a stable product iminium
carboxylate by intramolecular H-bonding with the trans-hydroxyl proton
leading to slow hydrolysis and product release. This likely leads to
epimerization and erosion of enantioselectivity. Increasing the
concentration of propanal and/or addition of acetic acid additive
resulted in an increase in ee to 65%.
In conclusion, we have developed a robust catalyst system
that delivers enantiodivergent products in
a series of α-
functionalization reactions of aldehydes. Rational design and
execution of such enantiodivergent methodologies is
unprecedented. We believe that this transition state analysis-
based approach can be extrapolated to other ‘activation modes’
within asymmetric organocatalysis and other areas of
enantioselective catalysis. This approach will be of particular
importance in reactions where both enantiomers of the chiral
catalyst are not readily available.
[14] The α-amination product 5a is derivatized to 10a for ease of HPLC
analysis, but if the derivitization is stopped at the stage of the product
alcohol, excellent isolated yields are obtained (yields for product alcohol
shown in parenthesis for entries 6-8 in Table 2).
[15] A. Bøgevig, N. Kumaragurubaran, K. A. Jørgensen, Chem. Comm.
2002, 620.
Keywords: organocatalysis • enantiodivergence • prolinate
catalysis • rational design • asymmetric catalysis
[16] (a) S. P. Brown, M. P. Brochu, C. J. Sinz, D. W. C. MacMillan, J. Am.
Chem. Soc. 2003, 125, 10808. (b) T. Kano, A. Yamamoto, H. Mii, J.
Takai, S. Shirakawa, K. Maruoka, Chem. Lett. 2008, 37, 250. (c) A.
Mielgo, I. Velilla, E. Gomez-Bengoa, C. Palomo, Chem. Eur. J. 2010,
16, 7496.
[1]
[2]
H. U. Blaser, Chem. Rev. 1992, 92, 935.
Reviews: (a) M. Bartók, Chem. Rev. 2010, 110, 1663. (b) Y. H. Kim,
Acc. Chem. Res. 2001, 34, 955. (c) M. P. Sibi, M. Liu, Curr. Org. Chem.
2001, 5, 719. (d) G. Zanoni, F. Castronovo, M. Franzini, G. Vidari, E.
Giannini, Chem. Soc. Rev. 2003, 32, 115. (e) T. Tanaka, M. Hayashi,
Synthesis. 2008, 21, 3361. Selected examples: (f) J. Wang, B. L.
Feringa, Science 2011, 331, 1429. (g) J. L. Stymiest, V. Bagutski, R. M.
French, V. K. Aggarwal, Nature 2008, 456, 778. (h) M-Q. Hua, H-F. Cui,
L. Wang, J. Nie, J-A. Ma, Angew. Chem. Int. Ed. 2010, 49, 2772;
Angew. Chem. 2010, 122, 2832. (i) S. Mortezaei, N. R. Catarineu, J. W.
Canary, J. Am. Chem. Soc. 2012, 134, 8054. (j) Y. Sohtome, T.
[17] (a) T. Kano, M. Ueda, J. Takai, K. Maruoka, J. Am. Chem. Soc. 2006,
128, 6046. (b) C. Palomo, S. Vera, I. Velilla, A. Mielgo, E. Gomez-
Bengoa, Angew. Chem. Int. Ed. 2007, 46, 8054; Angew. Chem. 2007,
119, 8200.
This article is protected by copyright. All rights reserved.

10.1002/anie.201703919
Angewandte Chemie International Edition
COMMUNICATION
COMMUNICATION
Juliet Macharia, Victor Wambua, Yun
Hong, Lawrence Harris, Jennifer S.
Hirschi, Gary B. Evans, and Mathew J.
Vetticatt*
O
CO2Bn
CO2Bn
CO2Bn
+
CO2Bn
CO2Bn
N
H
H
HN
N
CO2X
N
N
HN
N
N
N
O
O
O
BnO2C
BnO2C
CO2Bn
H
H
H
91% ee (S)
Enantiodivergent
94% ee (R)
X = H
X = N(Bu)4
Ar
Ar
N
N
prolinate catalysis
N
proline catalysis
Ar
CO2X
OH
O
O
O
HN
N
Page No. – Page No.
O
Ar
H
98% ee (S)
99% ee (R)
Enantio- and chemodivergent
A Designed Approach to
Enantiodivergent Enamine Catalysis
The rational design and implementation of enantiodivergent enamine catalysis is
reported. A simple secondary amine catalyst, 2-methyl-L-proline, and its
tetrabutylammonium salt function as an enantiodivergent catalyst pair delivering
opposite enantiomers of α-functionalized aldehyde products in excellent
enantioselectivities. This novel concept of ‘designed enantiodivergence’ is applied
to the enantioselective α-amination, aldol and α-aminoxylation/α-hydroxyamination
reactions of aldehydes.
This article is protected by copyright. All rights reserved.