Evaluation of achiral templates with fluxional bronsted basic substituents in enantioselective conjugate additions

Article abstract of DOI:10.1021/ol503275w

Enantioselective conjugate addition of malononitrile to pyrazolidinone-derived enoates proceeds in excellent yields and high enantioselectivities. A comparison of fluxional substituents with and without a Bronsted basic site and their impact on selectivity is detailed. Molecular sieves as an additive were found to be essential to achieve high enantioselectivity.

Full text of DOI:10.1021/ol503275w

Letter
pubs.acs.org/OrgLett
Evaluation of Achiral Templates with Fluxional Brønsted Basic
Substituents in Enantioselective Conjugate Additions
Shinya Adachi, Norihiko Takeda, and Mukund P. Sibi*
Department of Chemistry and Biochemistry, North Dakota State University, Fargo, North Dakota 58108-6050, United States
S
* Supporting Information
ABSTRACT: Enantioselective conjugate addition of malononi-
trile to pyrazolidinone-derived enoates proceeds in excellent
yields and high enantioselectivities. A comparison of fluxional
substituents with and without a Brønsted basic site and their
impact on selectivity is detailed. Molecular sieves as an additive
were found to be essential to achieve high enantioselectivity.
chiral templates play an important role in a wide variety of
acid complexes examined, the best result was obtained with
MgBr2·Et2O and indanol-derived ligand 7 (Table 1, entry 5, 36%
ee). We carried out further optimization studies using ligand 7 for
enantioselective malononitrile addition. Remarkably, the addi-
tion of molecular sieves (MS 4 Å) led to a substantial increase in
enantioselectivity (Table 1, entry 5 versus 6). Three alternate
Mg2+ Lewis acids were screened for the conjugate addition.
Mg(ClO4)2·6H2O as a Lewis acid was not effective and gave the
product in low yield and selectivity (Table 1 entry 7). However,
addition of MS 4 Å gave a huge increase in yield as well as
selectivity (Table 1, entry 7 versus 8). Similar trends of improved
selectivity with MS 4 Å were also observed with Mg(NTf2)2
(Table 1, entry 9 versus 10) and MgI2 (compare entries with 11
and 12). In the absence of MS 4 Å, Zn(NTf2)2 as a Lewis acid
gave no product (Table 1, entry 13). In contrast, addition of MS
4 Å led to a good yield of the product with no selectivity (Table 1,
entry 14).
We next examined MgBr2·Et2O/7-catalyzed reactions using
substrates 1a−d that contained the basic nitrogen atom at
different locations (Table 2). As noted previously, reaction with
1a showed a significant improvement in selectivity in the
presence of MS 4 Å (Table 2, entry 1 versus entry 2). Template
1b containing a 3-pyridylmethyl fluxional group gave the product
in low selectivity (Table 2, entry 3). Addition of MS 4 Å led to a
significant improvement in selectivity (Table 2, entry 4). More
notably, the sense of stereoinduction was opposite to that
obtained in the absence of molecular sieves.12 Reaction of 1c
showed a trend similar to that of reaction with 1a in the presence
and absence of MS 4 Å (entries 5 and 6, no change in the sense of
stereoinduction). Substrate 1d with a 4-isoquinolinylmethyl
fluxional substituent behaved similar to 1b in that the sense of
stereoinduction changed with the addition of MS 4 Å (Table 2,
entry 7 versus entry 8). Furthermore, 1d with a bigger fluxional
group than 1b gave higher level of selectivity for both
enantiomers (Table 2, entry 3 versus entry 7 and entry 4 versus
entry 8). These results indicate that the nature of the fluxional
1
Aasymmetric transformations.
In this context, we have
developed a class of templates, 3-pyrazolidinones, which
incorporate a fluxional substituent to provide enhanced
selectivity in enantioselective transformations.2 We have shown
that templates containing fluxional substituents were highly
effective in cycloaddition,3 addition of hydroxylamine to
enoates,4 and radical reactions.5 Suga and co-workers have
adapted our templates and have demonstrated their utility in
dipolar cycloadditions.6 We have extended the concept of
fluxional chirality and designed ligands,7 additives,8 and organo-
catalysts.9 In this work we report the utility of achiral templates
with fluxional substituents in conjugate addition of carbon
nucleophiles to enoates.10
The N-1 fluxional substituent in 3-pyrazolidinones is easily
varied, and its steric size has been shown to correlate with
enantioselectivity in chiral Lewis acid mediated transformations.
In an effort to extend the role of the fluxional substituent, we
were interested in incorporating a Brønsted basic site into the
fluxional substituent. We hypothesized that the fluxional
substituent with a Brønsted basic site could play several roles:
(1) provide steric shielding, (2) act as a base and generate a
nucleophile, (3) potentially deliver the nucleophile, and (4)
impact the chiral Lewis acid substrate complex. Four fluxional
substituents containing pyridine and quinoline rings were chosen
for the study. Malononitrile was the reagent of choice for the
conjugate additions to evaluate the effect of fluxional substituent
with a Brønsted basic site. The study also included fluxional
substituents lacking the Brønsted basic site as controls. Results
from conjugate addition of malononitrile to 3-acylpyrazolidi-
nones mediated by chiral Lewis acids are detailed here.
We began our work using an acceptor derived from 3-
pyrazolidinone with a 2-pyridylmethyl substituent.11 Optimiza-
tion of reaction conditions for the addition of malononitrile 2 to
pyrazolidinone derived enoate 1a (R = 2-pyridylmethyl) using
different chiral Lewis acids was undertaken initially (Table 1).
Ligands 4−7 were evaluated in combination with different Lewis
acids. Chiral Lewis acids derived from ligands 4−6 were not
effective (Table 1, entries 1−4). Among the several chiral Lewis
Received: November 11, 2014
Published: December 9, 2014
© 2014 American Chemical Society
6440
dx.doi.org/10.1021/ol503275w | Org. Lett. 2014, 16, 6440−6443

Organic Letters
Letter
a
substituent with Brønsted site significantly impacts the
enantioselectivity. In all cases, molecular sieves were effective,
and the best result was obtained when 1d was used (Table 2,
entry 8, ee = 88%).
Table 1. Screening of Chiral Lewis Acids and Additives
Next, reactions with 1e and 1f, substrates devoid of any
Brønsted basic site in the fluxional group were examined (Table
3). Additionally, reaction with pyridine as an additive to mimic
Table 3. Control Reactions with Benzyl and Naphthyl
a
Fluxional Groups
b
c
entry
substrate
MS 4 Å
Py
yield (%)
ee (%)
1
2
3
4
5
6
7
1e
−
−
+
−
+
69
87
75
84
22
61
83
89
85
(+) 44
(−) 02
(+) 90
(+) 80
(+) 36
(−) 15
(+) 96
(+) 97
(+) 94
b
c
entry
Lewis acid
ligand
MS 4 Å
yield (%)
ee (%)
−
+
1
MgBr2·Et2O
Ni(ClO4)2·6H2O
Co(ClO4)2·6H2O
Sc(OTf)3
4
5
5
6
7
7
7
7
7
7
7
7
7
7
−
−
−
−
−
+
73
80
80
0
0
7
+
2
1f
−
−
+
−
+
3
14
0
4
−
−
+
5
MgBr2·Et2O
MgBr2·Et2O
Mg(ClO4)2
Mg(ClO4)2
Mg(NTf2)2
Mg(NTf2)2
MgI2
88
92
18
91
87
93
88
82
0
36
79
5
d
8
+
6
9
+
7
−
+
a
Unless otherwise noted, 0.20 equiv of MgBr2·Et2O, 0.22 equiv of 7,
and 4.0 equiv of 2 was used. Isolated yields after purification by
column chromatography. ee values were determined by chiral HPLC.
0.10 equiv of MgBr2·Et2O, 0.11 equiv of 7, and 1.5 equiv of 2 were
used.
8
78
b
d
9
−
+
16
c
10
11
12
13
14
82
d
d
−
+
12
MgI2
80
0
Zn(NTf2)2
Zn(NTf2)2
−
+
79
5
reactions with 1a−d was also carried out. Reaction of 1e with a
benzyl fluxional group in the absence of both MS 4 Å and
pyridine gave the product in good yield and modest selectivity
(Table 3, entry 1). Addition of 1 equiv of pyridine increased
chemical efficiency, but the selectivity decreased (Table 3. entry
2). Reaction in the presence of MS 4 Å led to a higher yield and
a
b
4.0 equiv of malononitrile was used. Isolated yields after purification
by column chromatography. ee values were determined by chiral
c
d
HPLC. Enriched with the opposite enantiomer.
a
Table 2. Effect of Fluxional Substituent
very high selectivity (Table 3, entry 1 versus entry 3).13
A
combination of MS 4 Å and pyridine led to a slight decrease in
selectivity (Table 3, entry 4). Compound 1f with a bulky
naphthyl fluxional substituent was evaluated next by conducting
reactions with and without additives (Table 3, entries 5−9). The
trend was similar to that observed for 1e in that the optimal
results were obtained in the presence of MS 4 Å. Notably, we
obtained the highest selectivity (Table 3, entry 7, ee = 96%) for
3f. These results clearly demonstrate that the size of the fluxional
group has a significant impact on selectivity. The catalyst loading
could be reduced to 10 mol % without loss of selectivity (Table 3,
entry 8, ee = 97%). The above results demonstrate that
enantioselectivity is highly dependent on the structure of the
fluxional group R on the N-1 nitrogen.
Control experiments to assess the role of additives using 1e as
a substrate were undertaken (Table 4). Reaction in the presence
of MgBr2 only (no ligand or additive) gave none of the conjugate
addition product (Table 4, entry 1) suggesting a base is required
for the generation of the nucleophile. Reaction with MS 4 Å
resulted in low yield of the product (Table 4, entry 2), indicating
that MS 4 Å could deprotonate malononitrile, but the process is
not efficient.14 Reaction using pyridine as an additive gave good
yield of the product (Table 4, entry 3) suggesting that
deprotonation by pyridine is efficient. As noted previously,
b
c
entry
R
MS 4 Å
yield (%)
ee (%)
1
2
3
4
5
6
7
8
1a
−
+
88
92
73
73
83
76
82
87
(+) 36
(+) 78
(−) 27
(+) 76
(+) 39
(+) 79
(−) 59
(+) 88
1b
1c
1d
−
+
−
+
−
+
a
Unless otherwise noted, 0.20 equiv of MgBr2·Et2O, 0.22 equiv of 7,
and 4.0 equiv of 2 were used. Isolated yields after purification by
column chromatography. ee values were determined by chiral HPLC.
b
c
6441
dx.doi.org/10.1021/ol503275w | Org. Lett. 2014, 16, 6440−6443

Organic Letters
Letter
a
breadth and scope studies used the template with a
naphthylmethyl group. Increasing the size of the β-substituent
from a methyl (1f) to an ethyl group (1g) resulted in diminished
reactivity (2 h versus 21 h for completion of reaction, Table 5,
entry 2 and 3). Remarkably, the enantioselectivity for the product
3g was the highest for the all the reactions (Table 5, entry 3, ee =
99%). Compounds 1h and 1i with a β-CH2-i-Pr and CH2OBn
substituent, respectively, also gave high enantioselectivity (Table
5, entries 4 and 5, ee = 94 and 95%). In contrast to these β-alkyl
substrates, cinnamate (1j) was somewhat less selective (Table 5,
entry 6). Interestingly, the use of template with a benzyl fluxional
group was effective, and the product was obtained with 93% ee
(Table 5, entry 7). In contrast, substrates with a p-substituent on
the phenyl ring (1l, 1m, and 1n) gave low selectivities (Table 5,
entries 8−10).
Table 4. Control Experiments To Assess Impact of Additives
b
c
ligand
7
MS
4 Å
yield
ee
entry
additive (equiv)
(%)
(%)
1
−
−
−
+
+
+
+
+
+
+
+
−
+
0
20
75
69
66
80
89
82
80
51
92
0
0
2
3
−
−
−
−
+
Py (1 equiv)
0
4
44
46
65
90
42
80
41
88
5
H2O (0.5 equiv)
6
2-picoline (1 equiv)
7
2-picoline (1 equiv)
To gain an understanding of the differences in selectivity with
changes in the template as well as the role of MS 4 Å, compound
3l was converted to a known compound 8 by a Lewis acid
mediated transesterification (Scheme 1). This established that 3l
8
−
+
3-picoline (1 equiv)
9
3-picoline (1 equiv)
10
11
−
+
2,6-di-t-Bu-pyridine (1 equiv)
2,6-di-t-Bu-pyridine (1 equiv)
a
Unless otherwise noted, 0.20 equiv of MgBr2·Et2O, 0.22 equiv of 7,
Scheme 1. Determination of Absolute Stereochemistry
b
and 4.0 equiv of 2 were used. Isolated yields after purification by
column chromatography. ee values were determined by chiral HPLC.
c
enantioselective reaction in the absence of MS 4 Å gave the
addition product in modest selectivity (Table 4, entry 4).
Addition of water did not lead to a decrease of enantioselectivity
(Table 4, entry 5). Pyridine derivatives were explored next as
additives. These additives showed a similar trend to the reactions
with pyridine in that MS 4 Å was essential for obtaining high
selectivity (Table 4, entries 6−11).
We carried out a small breadth and scope study by varying the
enoyl substrate, and these results are tabulated in Table 5. As
demonstrated previously, substrate 1f with a naphthylmethyl
fluxional group gave higher selectivity than 1e with a benzyl
fluxional group (Table 5, entry 1 versus entry 2). Thus, our initial
has the (S)-configuration.10a−c Similarly, compounds 3a−f were
also converted to a single methyl ester 9 by transesterfication.
Interestingly, regardless of the nature of the fluxional substituent,
the face selectivity of the conjugate addition was identical as
evidenced by the formation of the ester 9 with (S)-configuration.
We propose two tentative models to account for reactions
using different templates and the reversal of selectivity in the
presence and absence of MS 4 Å. In the presence of MS 4 Å,
reactions occur from a tetrahedral or cis octahedral organization
of the Lewis acid/ligand/substrate complex (Figure 1A). On the
basis of control experiments, either MS 4 Å, basic nitrogen in the
template, or bromide counterion15 assists in the formation of the
nucleophile. The ligand is the major contributor of face shielding
with some reinforcement by the fluxional group, and since the
basic fluxional groups gave predominantly the same enantiomer,
this suggests that ion-pair directed addition was not occurring,
a
Table 5. Substrate Scope
substrate
b
c
entry
1
R1
R2
time (h) yield (%) ee (%)
d
1
1e
1f
Ph
Me
Me
Et
2
2
75
89
88
88
80
86
86
87
86
77
90
97
99
94
95
71
93
49
53
42
2
1-Naph
1-Naph
1-Naph
1-Naph
1-Naph
Ph
3
1g
1h
1i
21
21
21
21
21
21
21
68
4
CH2iPr
CH2OBn
Ph
5
6
1j
7
1k
1l
Ph
8
Ph
p-FC6H4
p-ClC6H4
p-MeOC6H4
9
1m
1n
Ph
10
Ph
a
Unless otherwise noted, 0.10 equiv of MgBr2·Et2O, 0.11 equiv of 7,
b
and 1.5 equiv of 2 were used. Isolated yields after purification by
c
column chromatography. ee values were determined by chiral HPLC.
d
0.20 equiv of MgBr2·Et2O, 0.22 equiv of 7, and 4.0 equiv of 2 were
used.
Figure 1. Stereochemical models.
6442
dx.doi.org/10.1021/ol503275w | Org. Lett. 2014, 16, 6440−6443

Organic Letters
Letter
(9) (a) Ma, G.; Deng, J.; Sibi, M. P. Angew. Chem., Int. Ed. 2014, 53,
11818. (b) Ma, G.; Sibi, M. P. Org. Chem. Front. 2014, 1, 1152.
(10) Selected examples of enantioselective conjugate addition of
malononitrile: (a) Taylor, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 2003,
125, 11204. (b) Taylor, M. S.; Zalatan, D. N.; Lerchner, A. M.; Jacobsen,
E. N. J. Am. Chem. Soc. 2005, 127, 1313. (c) Hoashi, Y.; Okino, T.;
Takemoto, Y. Angew. Chem., Int. Ed. 2005, 44, 4032. (d) Ooi, T.; Ohara,
D.; Fukumoto, K.; Maruoka, K. Org. Lett. 2005, 7, 3195. (e) Wang, J.; Li,
H.; Zu, L.; Jiang, W.; Xie, H.; Duan, W.; Wang, W. J. Am. Chem. Soc.
contrary to our original hypothesis. This accounts for the
observed stereochemical outcome of the reaction and the impact
of the size of the fluxional group on selectivity: the larger the
substituent the higher the selectivity. In the absence of MS 4 Å,
reactions occur from both a cis octahedral and trans octahedral
organization of the reactive complex with interstitial water
molecules still coordinated to the metal ion (Figure 1B). Since
these two organizations shield opposite faces, the selectivity is
modest. In the case of substrates 1b and 1d, the Brønsted base
has the potential to form a H-bond with water and leading to a
higher proportion of the trans octahedral complex and thus
leading to an inversion in stereochemistry. Geometrical
constraints may make this type of interaction less effective for
compounds 1a and 1c.
In summary, we have developed enantioselective conjugate
addition of malononitrile to pyrazolidinone-derived enoates. The
efficiency of this reaction was found to be highly dependent on
both the structure of the substituent on the fluxional nitrogen
and additives such as MS. The results suggest that inclusion of
donor atom in the fluxional group may allow for interesting
organization of chiral Lewis acid/substrate complex and ensuing
selectivity issues.
2006, 128, 12652. (f) Yang, K. S.; Nibbs, A. E.; Turkmen, Y. E.; Rawal, V.
̈
H. J. Am. Chem. Soc. 2013, 135, 16050.
(11) For details on synthesis, experimental conditions, and character-
ization data, see the Supporting Information.
(12) (a) Gothelf, K. V.; Hazell, R. G.; Jørgensen, K. A. J. Org. Chem.
1998, 63, 5483. (b) Sibi, M. P.; Liu, M. Curr. Org. Chem. 2001, 5, 719.
(c) Tanaka, T.; Hayashi, M. Synlett 2008, 3361.
(13) Use of MS 3 Å or MS 5 Å instead of MS 4 Å gave nearly identical
results.
(14) Molecular sieves can act as a base. See: (a) Palomo, C.; Pazos, R.;
Oiarbide, M.; García, J. M. Adv. Synth. Catal. 2006, 348, 1161. (b) Evans,
D. A.; Mito, S.; Seidel, D. J. Am. Chem. Soc. 2007, 129, 11583.
(c) Hasegawa, M.; Ono, F.; Kanemasa. Tetrahedron Lett. 2008, 49, 5220.
(d) García, J. M.; Maestro, M. A.; Oiarbide, M.; Odriozola, J. M.; Razkin,
J.; Palomo, C. Org. Lett. 2009, 11, 3826.
(15) The bromide counterion may be formed after the substrate
coordinates the chiral Lewis acid.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures and characterization data for new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: mukund.sibi@ndsu.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank NSF CHM-0709061 and NDSU for financial support.
■
REFERENCES
■
(1) (a) Palomo, C.; Oiarbide, M.; Kardak, B. G.; Garcia, J. M.; Linden,
A. J. Am. Chem. Soc. 2005, 127, 4154. (b) Sibi, M. P.; Chen, J.; Stanley, L.
Synlett 2007, 298. (c) Sibi, M. P.; Gustafson, B.; Coulomb, J. Bull. Korean
Chem. Soc. 2010, 31, 541−542. (d) Lee, S.; Kim, S. Org. Lett. 2008, 10,
4255.
(2) (a) Sibi, M. P.; Venkatraman, L.; Liu, M.; Jasperse, C. P. J. Am.
Chem. Soc. 2001, 123, 8444. (b) Sibi, M. P.; Stanley, L. M.; Nie, X.;
Venkatraman, L.; Liu, M.; Jasperse, C. P. J. Am. Chem. Soc. 2007, 129,
395. (c) Corminboeuf, O.; Quaranta, L.; Renaud, P.; Liu, M.; Jasperse,
C. P.; Sibi, M. P. Chem.Eur. J. 2003, 9, 28.
(3) (a) Sibi, M. P.; Ma, Z.; Jasperse, C. P. J. Am. Chem. Soc. 2004, 126,
718. (b) Sibi, M. P.; Itoh, K.; Jasperse, C. P. J. Am. Chem. Soc. 2004, 126,
5366. (c) Sibi, M. P.; Stanley, L. M.; Soeta, T. Adv. Synth. Catal. 2006,
348, 2371. (d) Sibi, M. P.; Stanley, L. M.; Soeta, T. Org. Lett. 2007, 9,
1553. (e) Sibi, M. P.; Rane, D.; Stanley, L. M.; Soeta, T. Org. Lett. 2008,
10, 2971.
(4) Sibi, M. P.; Liu, M. Org. Lett. 2001, 3, 4181.
(5) Sibi, M. P.; Yang, Y.-H.; Lee, S. Org. Lett. 2008, 10, 5349.
(6) Suga, H.; Adachi, Y.; Fujimoto, K.; Furihata, Y.; Tsuchida, T.;
Kakehi, A.; Baba, T. J. Org. Chem. 2009, 74, 1099.
(7) (a) Sibi, M. P.; Zhang, R.; Manyem, S. J. Am. Chem. Soc. 2003, 125,
9306. (b) Sibi, M. P.; Stanley, L. M. Tetrahedron: Asymmetry 2004, 15,
3353.
(8) Sibi, M. P.; Manyem, S.; Palencia, H. J. Am. Chem. Soc. 2006, 128,
13660.
6443
dx.doi.org/10.1021/ol503275w | Org. Lett. 2014, 16, 6440−6443