Enantioselective 1,3-dipolar cycloaddition of nitrile imines to α-substituted and α,β-disubstituted α,β-unsaturated carbonyl substrates: A method for synthesizing dihydropyrazoles bearing a chiral quaternary center

Article abstract of DOI:10.1002/adsc.200600307

Dihydropyrazoles bearing a chiral quaternary center at the 5-position have been prepared by enantioselective 1,3-dipolar cycloaddition of nitrile imines to α-substituted- and α,β-disubstituted-α,β- unsaturated carbonyl substrates. Use of α,β-unsaturated carbonyl substrates with a l-benzyl-5,5-dimethylpyrazolidin-3-one auxiliary in conjunction with MgI₂ and a bisoxazoline ligand derived from (1R,2S)-(+)-cis-1-amino-2-indanol 6 proved optimal to obtain chiral dihydropyrazoles with high enantio-selectivity (up to 99% ee).

Full text of DOI:10.1002/adsc.200600307

COMMUNICATIONS
DOI: 10.1002/adsc.200600307
Enantioselective 1,3-Dipolar Cycloaddition of Nitrile Imines to a-
Substituted and a,b-Disubstituted a,b-Unsaturated Carbonyl Sub-
strates: A Method for Synthesizing Dihydropyrazoles Bearing a
Chiral Quaternary Center
Mukund P. Sibi,a,* Levi M. Stanley,a and Takahiro Soetaa
a
Department of Chemistry and Molecular Biology, North Dakota State University, Fargo, North Dakota 58105, USA
Fax : (+1)-701-231-8831; e-mail: mukund.sibi@ndsu.edu
Received: June 27, 2006; Accepted: August 26, 2006
Supportinginformation for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: Dihydropyrazoles bearinga chiral quater-
nary center at the 5-position have been prepared by
enantioselective 1,3-dipolar cycloaddition of nitrile
imines to a-substituted- and a,b-disubstituted-a,b-
unsaturated carbonyl substrates. Use of a,b-unsatu-
rated carbonyl substrates with a 1-benzyl-5,5-dime-
thylpyrazolidin-3-one auxiliary in conjunction with
MgI2 and a bisoxazoline ligand derived from
(1R,2S)-(+)-cis-1-amino-2-indanol 6 proved optimal
to obtain chiral dihydropyrazoles with high enantio-
selectivity (up to 99% ee).
number of enantioselective addition reactions utiliz-
ingunsubstituted and/or b-substituted pyrazolidinone
templates have been reported (Figure 1).[6] However,
Figure 1. Common templates for enantioselective addition
reactions.
Keywords: asymmetric synthesis; dihydropyrazoles;
1,3-dipolar cycloadditions; nitrile imines; pyrazoli-
dinones
the ability of pyrazolidinones to control rotamer ge-
ometry in a-substituted and a,b-disubstituted sub-
strates has not been thoroughly examined.
We have previously reported enantioselective ni-
Control of rotamer geometry is critical for enantiose- trile imine cycloadditions utilizing b-substituted a,b-
lective addition reactions to a,b-unsaturated carbonyl unsaturated oxazolidinone amides as dipolarophiles.[7]
compounds.[1] Successful examples of chiral Lewis The resulting4,5-dihydropyrazole products are of sig-
acid catalysis have been developed for a-unsubstitut- nificant interest as chiral buildingblocks. However,
ed substrates usingoxazolidinone as an achiral tem-
we were further interested in the development of a
plate. Substrates such as 1 are ideal for chiral Lewis method for the preparation of chiral dihydropyrazoles
acid catalysis as they lead to organized chelation and bearinga quaternary chiral center at the 5-position.
exclusive reaction through the s-cis rotamer.[2] In con- Such a method would potentially allow control of the
trast, solutions with broad applicability for enantio- requisite tert-alkylamino chiral quaternary center
selective addition to substrates with substitution on present in a number of natural products.[8] We initially
the a-carbon are limited.[3,4] Our group has previously chose to evaluate methacrylate substrates in room
described the use of a,b-disubstituted acrylamides 2 temperature reactions catalyzed by Mg(NTf2)2/6. Sub-
G
as a general solution to control rotamer geometry, re- strates with 3,5-dimethylpyrazole and oxazolidinone
lieve A1,3 strain, reduce twisting, restore conjugation, auxiliaries gave products in low enantioselectivity
and improve reactivity in a variety of reactions.[5] We (Table 1, entries 1 and 2). We attribute the low selec-
now wish to report the use of a-substituted and a,b- tivity to a lack of rotamer control usingthese tem-
disubstituted pyrazolidinone templates 3 as additional plates since face shieldingof a-unsubstituted-b-substi-
substrates capable of rotamer geometry control in tuted oxazolidinones is high even at room tempera-
enantioselective nitrile imine cycloadditions.
A
ture. We were pleased to see significant enantioselec-
Adv. Synth. Catal. 2006, 348, 2371 – 2375
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
2371

Mukund P. Sibi et al.
Table 1. Template and chiral Lewis acid optimization for 1,3-dipolar cycloaddition of 5 to methacrylate substrates 4a–c.[a]
COMMUNICATIONS
Entry
Substrate
Lewis Acid
mol% CLA*
Temp [8C]
Product
Yield [%]
ee [%]
1
2
3
4
5
6
7
4a
4b
4c
4c
4c
4c
4c
4c
4c
4c
Mg
Mg
Mg
Mg
Mg
Mg
MgI2
MgI2
MgI2
MgI2
C
30
30
30
30
30
30
30
30
20
10
rt
rt
rt
7a
7b
7c
7c
7c
7c
7c
7c
7c
7c
88
78
67
97
63
80
79
82
80
69
0
22
70
77
36
87
74
98
99
94
À78
rt
À78
rt
À78
À78
À78
8[b]
9[b]
10[b]
[a]
Typical reactions were run for 2–4 h with 1.5 equivs. 5 and 1.5 equivs. NEt3.
1.2 equivs. 5 and 1.2 equivs. NEt3 were employed.
[b]
tivity (70% ee) when a substrate bearinga pyrazolidi-
none template was employed (entry 3). Use of Mg- of tiglate substrate 8a catalyzed by 30 mol% MgI2/6
(ClO4)2 led to a decrease in enantioselectivity (36% provided cycloadduct in 97% yield and 99% ee
disubstituted carbonyl compounds also. The reaction
ACHTREUNG
ee) and isolated yield (entry 5). MgI2 proved to be the (entry 1). Furthermore, catalyst loadingof 10 mol%
optimal Lewis acid for pyrazolidinone substrates, pro- led to no loss of selectivity although the isolated yield
viding 7c in 74% ee (entry 7). Significant increases in was lowered to 65% (entry 2). Product 9b, where
enantioselectivity were observed when Mg
Mg
ducted at À788C (entries 4, 6 and 8), giving product yields were moderate at À788C. Therefore, the reac-
in 77, 87, and 98% ee respectively. Given the optimal tion temperature was elevated to À208C (entry 4) re-
yield and enantioselectivity observed in reactions cat- sultingin a modest increase in isolated yield and no
alyzed by 30 mol% MgI2/6, we were interested in the loss of enantioselectivity. As seen with tiglate 8a, re-
C
G
2
possibility of loweringcatalytic loading. Catalyst load-
ingof 20 mol% gave product in 80% yield and 99%
activity is diminished significantly when 10 mol% cat-
alyst is used in combination with 8b (entry 5). In this
ee (entry 9), essentially identical results to the 30 case cycloadduct was isolated in 44% yield with 98%
mol% experiment. Isolated yields were slightly lower ee. a-Methyl cinnamate 8c proved to be more chal-
when 10 mol% of the catalyst was employed lenging in terms of reactivity. Optimal results were
(entry 10), but enantioselectivity remained high (94% obtained in the presence of 30 mol% catalyst by con-
ee). Overall, these results suggest that a combination ductingthe reaction at À208C for 24 h, providing
of MgI2/6 and the pyrazolidinone substrate 4c pro- product in 56% yield and 99% ee (entry 6). Reac-
vides effective control of rotamer geometry in a-sub- tions carried out at lower temperatures led to very
stituted a,b-unsaturated carbonyl systems.
With an optimal chiral Lewis acid/template system elevated temperatures led to significant dipole dimeri-
in hand, we were interested in evaluatingwhether this zation and lower yields of the desired product 9c. Cy-
minimal amounts of cycloaddition, while reactions at
system would be effective when a,b-disubstituted a,b- cloadditions to cycloalkene substrates 8d and 8e gave
unsaturated carbonyl compounds were used as dipo- cycloadducts in 98% and 99% ee, respectively (en-
larophiles. Results in Table 2 clearly demonstrate that tries 7 and 8). Enantioselectivity remains high (99%
the pyrazolidinone template functions well with a,b- ee) in cycloaddition to 8e when 10 mol% chiral Lewis
2372
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2371 – 2375

Enantioselective 1,3-Dipolar Cycloaddition of Nitrile Imines
COMMUNICATIONS
Table 2. 1,3-Dipolar cycloaddition of 5 to a,b-disubstituted substrates 8a–e.[a]
Entry
Substrate
R1
R2
mol% CLA*
Temp [8C]
Product
Yield [%]
ee [%]
1
2
3
4
8a
8a
8b
8b
8b
8c
8d
8e
8e
Me
Me
Me
Me
Me
Me
-(CH2)3-
-(CH2)4-
-(CH2)4-
Me
Me
Et
Et
Et
30
10
30
30
10
30
30
30
10
À78
À78
À78
À20
À78
À20
À78
À20
À78
9a
9a
9b
9b
9b
9c
9d
9e
9e
97
65
69
75
44
56
83
63
33
99
99
99
99
98
99
98
99
99
5
6[b]
7
Ph
8
9
[a]
Typical reactions were run for 2–4 h with 1.2 equivs. 5 and 1.2 equivs NEt3.
Reactions were run for 24 h.
[b]
acid is employed, but a large decrease in yield (33%) ground reaction, not because of a lack of rotamer ge-
was observed (entry 9). This trend is interestingas 10
mol% chiral Lewis acid experiments with cyclopen-
ometry control.
To further evaluate the utility of the pyrazolidinone
tene 8d resulted in high yields, but a noticeable loss template/chiral Lewis acid system in nitrile imine cy-
of enantioselectivity.[9] At this time we believe the un- cloadditions, we chose to screen three additional hy-
catalyzed background reaction between the nitrile drazonoyl bromides (Table 3). Reactions of nitrile
imine generated from 5 and substrate 8d to be signifi- imine derived from 10a to substrates 4c and 8a gave
cant, while the background reaction with other a,b- cycloadducts in very high enantioselectivities (93–
disubstituted substrates screened is negligible.[10] Thus, 99% ee) and yields (92–96%), even when chiral
the drop in selectivity is a result of racemic back- Lewis acid loadings were lowered to 10 mol% (en-
Table 3. Reactions of additional hydrazonoyl bromides with substrate 4c or 8a.[a]
Entry Substrate R1
Dipole R2
R3
mol% CLA* Temp [8C] Product Yield [%] ee [%]
1
2
3
4
4c
4c
8a
8a
4c
4c
4c
H
H
10a
10a
Ph
Ph
Ph
Ph
4-BrC6H4
4-BrC6H4
4-BrC6H4
4-BrC6H4
30
10
30
10
À78
À78
À78
À78
À78
À20
À20
11a
11a
11b
11b
11c
11c
11d
93
92
96
96
48
61
67
99
93
99
99
80
79
67
Me 10a
Me 10a
H
H
H
5
10b
10b
10c
BnOCH2- 4-MeOC6H4 30
BnOCH2- 4-MeOPh
Ph Bn
6
30
30
7[b]
[a]
[b]
Typical reactions were run for 2–4 h with 1.2 equivs. 10a–c and 1.2 equivs. NEt3.
Phosphazene base P1-t-Bu [tert-butylimino-tris(dimethylamino)phosphorane].
Adv. Synth. Catal. 2006, 348, 2371 – 2375
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2373

Mukund P. Sibi et al.
COMMUNICATIONS
tries 1–4). Hydrazonoyl bromide 10b derived from respondingexperiment with Et N. This trend is fur-
3
benzyloxyacetaldehyde and 4-methoxyphenylhydra- ther demonstrated when hydrazonoyl chloride 12c de-
zine gave product in 48% yield and 80% ee when re- rived from cinnamaldehyde is employed (entries 6
actions were conducted at À788C (entry 5). A modest and 7). Results with P1-t-Bu phosphazene base at
increase in yield with little loss of selectivity was ob- À788C illustrate the dramatic increase in both yield
served when this reaction was carried out at À208C and selectivity observed under these conditions. Final-
(entry 6). Results with hydrazonoyl bromide 10c de- ly, cycloaddition of nitrile imine derived from benzy-
rived from benzylhydrazine and benzaldehyde were loxyacetaldehyde with Et3N gave cycloadduct 13d in
disappointing, providing product with only 67% yield 64% yield and 91% ee.
and 67% ee (entry 7).
In summary, we have demonstrated a pyrazolidin-
In many cases the use of hydrazonoyl chlorides in one template/chiral Lewis acid combination that ef-
place of hydrazonoyl bromides is desirable due to fectively controls rotamer geometry in nitrile imine
their increased stability and ease of preparation in se- cycloadditions to a-substituted a,b-unsaturated and
lected cases. As such, we evaluated a number of hy- a,b-disubstituted a,b-unsaturated substrates. This
drazonoyl chlorides in nitrile imine cycloadditions to Lewis acid/template combination provides dihydro-
substrate 4c (Table 4). Our previous work suggested pyrazole cycloadducts containinga chiral quaternary
dehydrohalogenation of hydrazonoyl chlorides with center at the 5-position in moderate to high yields (up
common tertiary amine bases was inefficient at to 97%) and excellent enantioselectivity (up to 99%
À788C, but proceeds at À208C with Et3N. In fact, cy- ee). Further applications of rotamer geometry control
cloaddition of the nitrile imine derived from 12a gave with these systems and applications to natural product
product in 77% yield and 93% ee (entry 1). Use of targets are currently in progress in our laboratory.
P1-t-Bu phosphazene base at À208C led to decrease
in both isolated yield and enantioselectivity
(entry 2).[11] However, use of P1-t-Bu phosphazene
Experimental Section
base at À788C provided product 13a in 96% and
96% ee (entry 3). These results are comparable to
those obtained for cycloaddition of 5 to 4c. A similar
trend was observed in cycloadditions of nitrile imine
derived from 12b (entries 4 and 5). P1-t-Bu phospha-
zene base at À788C once again provided optimal re-
sults with product 13b isolated in 94% yield and 95%
ee, markinga drastic increase in yield versus the cor-
Representative Procedure for Enantioselective
Cycloaddition of Nitrile Imines (Table 2, Entry 1)
Substrate 8a (0.0573 g, 0.2 mmol), MgI2 (0.0167 g,
0.06 mmol), ligand 6 (0.024 g, 0.066 mmol), and 4 Š molecu-
lar sieves (0.10 g) were dissolved in CH2Cl2 (2 mL) and al-
lowed to stir at room temperature for 30–60 min. The sub-
strate/chiral Lewis acid complex was cooled to À788C in a
Table 4. Reactions of various hydrazonoyl chlorides with 4c.[a]
Entry
Dipole
R1
R2
Base
Temp [8C]
Product
Yield [%]
ee [%]
1
2
3
4
5
6
7
8
12a
12a
12a
12b
12b
12c
12c
12d
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Et3N
À20
À20
À78
À20
À78
À20
À78
À20
13a
13a
13a
13b
13b
13c
13c
13d
77
64
96
47
94
39
78
64
93
83
96
91
95
89
94
91
P1-t-Bu[b]
P1-t-Bu[b]
Et3N
4-MeOC6H4
4-MeOC6H4
Ph
Ph
Ph
Ph
P1-t-Bu[b]
Et3N
PhCH=CH-
PhCH=CH-
BnOCH2-
P1-t-Bu[b]
Et3N
[a]
Reactions usingEt N to generate the nitrile imine were run for 24 h with 1.5 equivs. 12a–d and 1.5 equivs. NEt3. Reac-
3
tions usingP -t-Bu base to generate the nitrile imine were run for 2–4 h with 1.5 equivs. of 12a–c and 1.5 equivs. of P1-tBu
1
base.
[b]
Phosphazene base P1-t-Bu [tert-butylimino-tris(dimethylamino)phosphorane].
2374
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2371 – 2375

Enantioselective 1,3-Dipolar Cycloaddition of Nitrile Imines
COMMUNICATIONS
dry ice/acetone bath. Followingcoolingfor 5 min, hydrazo-
noyl bromide 5 (0.1039 g, 0.24 mmol) in CH2Cl2 (2 mL) was
added to the reaction. After an additional 5 min NEt3
(0.34 mL, 0.24 mmol) was added to the reaction. The reac-
tion was allowed to stir for 2–4 h or until startingmaterial
was consumed (TLC). Followingcompletion of the reaction,
4 Š molecular sieves were removed by filtration through a
short pad of Celite. The reaction mixture was concentrated
onto silica gel (2 g) and purified by column chromatography
on an ISCO Combiflash system (100% hexanes to 40%
EtOAc/hexanes gradient) to furnish 9a as a tan foam; yield:
0.1241 g(0.194 mmol, 97%); mp 84–86 8C; The enantiomer-
ic purity was determined by HPLC analysis (254 nm, 258C)
tR = 11.2 min (major); tR = 18.6 min (minor) [Chiracel
AD-H (0.46 cm ”25 cm)(from Daicel Chemical Ind., Ltd.)
hexane/i-PrOH, 90/10, 1.0 mLminÀ1] as 99% ee. [a]D25: À3.68
(c 0.54, CHCl3); 1H NMR (CDCl3, 500 MHz): d=1.14 (s,
3H), 1.25 (s, 3H), 1.26 (d, J=7.0 Hz, 3H), 2.42 (d, J=
17.5 Hz, 1H), 2.69 (d, J=17.5 Hz, 1H), 4.06 (d, J=14.0 Hz,
1H), 4.08 (q, J=7.0 Hz, 1H), 4.19 (d, J=14.0 Hz, 1H), 6.93
(d, J=9.5 Hz, 2H), 7.25–7.30 (m, 5H), 7.41–7.43 (m, 2H),
7.50–7.56 (m, 4H); 13C NMR (125 MHz, CDCl3): d=12.3,
15.6, 25.7, 27.6, 44.0, 47.8, 57.0, 61.3, 75.1, 112.6, 117.5, 122.7,
127.8, 128.1, 128.5, 129.2, 131.2, 131.8, 131.9, 137.5, 142.5,
150.2, 169.8, 173.9; exact mass calcd. for C30H30Br2N4O2Na+:
659.0628, found: 659.0622.
Barnes, J. S. Johnson, T. Lectka, P. von Matt, S. J.
Miller, J. A. Murry, R. D. Norcross, E. A. Shaughnessy,
K. R. Campos, J. Am. Chem. Soc. 1999, 121, 7582; c) S.
Castellino, J. Org. Chem. 1990, 55, 5197.
[3] For recent enantioselective additions to methacrolein,
see: a) T. Kano, T. Hashimoto, K. Maruoka, J. Am.
Chem. Soc. 2006, 128, 2174; b) S. Hong, E. J. Corey, J.
Am. Chem. Soc. 2006, 128, 1346; c) S. Cabrera, R.
Gomez Arrayas, J. C. Carretero, J. Am. Chem. Soc.
2005, 127, 16394; d) M. Shirahase, S. Kanemasa, Y.
Oderaotoshi, Org. Lett. 2004, 126, 2716.
[4] For enantioselective addition to methacrylates and ti-
glates, see: a) K.-S. Yang, K. Chen, Org. Lett. 2002, 4,
1107; b) B.-J. Li, L. Jiang, M. Liu, Y.-C. Chen, L.-S.
Ding, Y. Wu, Synlett 2005, 603; c) H. Doi, T. Sakai, M.
Iguchi, K.-I. Yamada, K. Tomioka, J. Am. Chem. Soc.
2003, 125, 2886; d) H. Doi, T. Sakai, M. Iguchi, K.-I.
Yamada, K. Tomioka, Chem. Commun. 2004, 1850.
[5] For work on a,b-disubstituted acrylamides from the
Sibi group, see: a) M. P. Sibi, N. Prabagaran, S. G.
Ghorpade, C. P. Jasperse, J. Am. Chem. Soc. 2003, 125,
11796; b) M. P. Sibi, G. Petrovic, J. Zimmerman, J. Am.
Chem. Soc. 2005, 127, 2390; c) M. P. Sibi, Z. Ma, K.
Itoh, N. Prabagaran, C. P. Jasperse, Org. Lett. 2005, 7,
2349.
[6] For enantioselective additions utilizingpyrazolidinone
templates, see: a) M. P. Sibi, K. Itoh, C. P. Jasperse, J.
Am. Chem. Soc. 2004, 126, 5366; b) M. P. Sibi, Z. Ma,
C. P. Jasperse, J. Am. Chem. Soc. 2004, 126, 718;
c) M. P. Sibi, N. Prabagaran, Synlett 2004, 2421; d) M. P.
Sibi, L. Venkatraman, M. Liu, C. P. Jasperse, J. Am.
Chem. Soc. 2001, 123, 8444; e) M. P. Sibi, M. Liu, Org.
Lett. 2001, 3, 4181; f) H. Nakano, N. Tsugawa, R.
Fugita, Tetrahedron Lett. 2005, 46, 5677.
Acknowledgements
We thank the National Science Foundation (NSF-CHE-
0316203) for financial support. L. M.S. thanks the North
Dakota Space Grant Consortium for a fellowship.
[7] M. P. Sibi, L. M. Stanley, C. P. Jasperse, J. Am. Chem.
References
Soc. 2005, 127, 8276.
[8] S. H. Kang, S. Y. Kang, H.-S. Lee, A. J. Buglass, Chem.
Rev. 2005, 105, 4537.
[9] Reactions utilizingsubstrate 8d in place of substrate 8e
under conditions identical (10 mol% chiral Lewis acid)
to those used in Table 2, entry 9 gave product in 63%
yield and 48% ee.
[1] For a general discussion, see: Principles of Asymmetric
Synthesis (Eds.: R. E. Gawley, J. Aube), Pergamon,
Oxford, 1996; see also: a) K. Nishimura, K. Tomioka, J.
Org. Chem. 2002, 67, 431; b) J. M. Hawkins, M. Nambu,
S. Loren, Org. Lett. 2003, 5, 4293; c) D. M. Birney,
K. N. Houk, J. Am. Chem. Soc. 1990, 112, 4127;
d) M. P. Sibi, J. B. Sausker, J. Am. Chem. Soc. 2002,
124, 984; e) M. Avalos, R. Babiano, J. L. Brao, P.
Cintas, J. L. Jimenez, J. C. Palacios, M. A. Silva, J. Org.
Chem. 2000, 65, 6613.
[10] Reaction of nitrile imine derived from 5 to substrate 8d
at À788C with no chiral Lewis acid present showed
1
>50% conversion by H NMR, while an identical reac-
tion with with substrate 8e showed <5% conversion.
[11] Although isolated yields are lower in this case, no start-
ingmaterial remains.
[2] a) D. A. Evans, S. J. Miller, T. Lectka, P. von Matt, J.
Am. Chem. Soc. 1999, 121, 7559; b) D. A. Evans, D. M.
Adv. Synth. Catal. 2006, 348, 2371 – 2375
ꢀ 2006 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2375