Graphical Abstract

Tetrahedron: Asymmetry

Graphical Abstract

To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.

Tetrahedron: Asymmetry

Stereochemistry Abstract

To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered. You may insert more abstracts by copying this box or by using the menu option to insert a stereochemistry abstract.

Tetrahedron: Asymmetry

Binap-AuTFA and Binap-AgTFA: Two efficient coinage metal complexes in the synthesis of chiral pyrrolidines through 1,3-dipolar cycloadddition of azomethine ylides

Carmen Nájera,a[*] María Martín-Rodríguez,a Feng-Liu Wu,b José M. Sansanoa

aDepartamento de Química Orgánica, Instituto de Síntesis Orgánica (ISO). Universidad de Alicante, 03080-Alicante (Spain)

bSchool of Chemistry and Molecular Biosciences, University of Queensland, Brisbane, Qld 4072, Australia.

Deticated to Prof. H. B. Kagan on the occasion of his 80 birhtday

Abstract—In this work, a comparison between chiral BINAP-AuTFA- and chiral BINAP-AgTFA-promoted catalytic enantioselective 1,3-dipolar cycloadditions of azomethine ylides and alkenes. Maleimides reacted smoothly in very good yields and enantioselections and trans-1,2-bis(phenylsulfonyl)ethylene reacted after longer reaction times but using smaller amounts of catalyst loading in order to achieve the highest enantioselectivities. In spite of the scarce induction of these two complexes when tert-butyl acrylate is used as dipolarophile, chiral gold(I) catalyst afforded, unexpectedly, an excellent enantioselection in the reaction of the iminoester precursor of key intermediate in the synthesis of hepatitis C virus inhibitor. © 2010 Elsevier Science. All rights reserved

Tetrahedron: Asymmetry

Coinage metals attract particular interest from synthetic organic chemists, because those metals become useful catalysts for synthesizing the core of many important drugs containing heterocyclic structures.[1] The main features of these metal complexes are the good chemoselectivity, good functional group compatibility, stability, traits that are crucial for application in complex molecular environments. One of the most representative examples concerns the synthesis of enantiomerically enriched pyrrolidines[2] through the catalytic enantioselective 1,3-dipolar cycloaddition[3] (1,3-DC) between azomethine ylide and alkenes. In fact, silver(I)[4],[5] and copper(I)[6] catalyzed 1,3-DC are very well known[7] and constitute the most reliable, sure and inexpensive enantioselective methodology to built up to four stereogenic centres of the resulting proline derivatives, in only one reaction step. In addition, they exhibit more versatility and wider scope than the analogous enantioselective organocatalysed 1,3-dipolar cycloadditions.3a,[8]

Chiral gold complexes have been employed in enantioselective activation of allenes and nucleophilic additions onto alkynes,[9] but they have not so extensively studied as catalysts in these cycloadditions. Only Toste et al. reported a very efficient enantioselective cycloaddtion of münchnones and electron-defficient alkenes employing (Sa)-Cy-SEGPHOS(AuOBz)2 (3.5 mol%). This transformation, followed by an ester/amide formation, furnished pyrrolines with very high enantioselection. [10]

In this communication, and continuing with the research line involving the chiral BINAP complex-promoted 1,3-DC, we survey the efficiency of (Ra)- or (Sa)-BINAP-AuTFA (TFA = trifluoroacetate anion) in the classical intermolecular 1,3-DC employing iminoesters and electrophilic alkenes, establishing a direct comparison with the analogous processes catalyzed by (Ra)- or (Sa)-Binap-AgTFA complexes.

One of the best test to prove the ability of a chiral catalyst in a silver(I) catalysed enantioselective 1,3-dipolar cycloaddition is the transformation of azomethine ylides (generated from iminoesters 1) and N-methylmaleimide (NMM) in enantiomerically enriched prolines 2 (Scheme 1). The chiral gold complexes were prepared in situ from (Me2S)AuCl and the corresponding amount of the chiral diphosphane ligand, followed by the anion interchange with the corresponding silver salt (approx. 1 h). The resulting suspension was filtered through a celite path and the solution was evaporated to yield the titled complexes.10 The anion interchange was necessary because the initial complex Binap-AuCl was inactive in terms of enantiodiscrimination. The reactions carried out with diisopropylethylamine (DIPEA) afforded cleaner reaction crudes than the analogous transformations performed with triethylamine (Table 1, entries 1 and 2). The anion interchange on the original (Sa)-Binap-AuCl complex was done with several silver salts (Table 1, entries 3-8) running the cycloaddition in the presence of DIPEA. The best results were achieved when using the benzoate or TFA anions (high conversions and 74% ee each, Table 1, entries 7 and 8). The chiral gold(I)-TFA complex was selected because higher enantiodiscriminations were achieved and the reaction products were obtained with high purity. Another bases, such as Et3N, DABCO or DBU did not improve the results achieved employing DIPEA as base (Table 1, 9-11). Other solvents like THF, Et2O, and DCM did not improve the result described in the reaction run with toluene.

A very important feature of these carboxylate anions is the weak basicity, which was enough to promote the identical enantioselective cycloadditions in the absence of base associated with an unexpected increment of the enantioslectivity (Table 1, entries 12-18). TFA anion was the most suitable internal base to promote this reaction in high conversions, good enantioselectivities and affording very clean crude reaction products. A lower catalyst loading (5 mol%) decreased the conversion and the enantioselection of the process (Table 1, entry 17), whilst the formation of the gold(I) complex overnight did not affect to the final result of the reaction (Table 1, entry 18).

Scheme 1.

Table 1. Optimization of the chiral gold(I)-catalyzed 1,3-DC between iminoester 1a and NMM.

Entry / Gold(I) catalyst
(10 mol%) / Base
(10 mol%) / Conv.
(%)a,b / ee
(%)c
1 / (Sa)-Binap-AuCl / Et3N / >95 / rac.
2 / (Sa)-Binap-AuCl / DIPEA / >95 / rac.
3 / (Sa)-Binap-AuCl/AgSbF6 / DIPEA / <10 / ___
4 / (Sa)-Binap-AuCl/AgClO4 / DIPEA / >95 / 60
5 / (Sa)-Binap-AuCl/AgOAc / DIPEA / >95 / 62
6 / (Sa)-Binap-AuCl/AgOTf / DIPEA / >95 / 59
7 / (Sa)-Binap-AuCl/AgOBz / DIPEA / >95 / 74
8 / (Sa)-Binap-AuCl/AgTFA / DIPEA / >95 / 74
9 / (Sa)-Binap-AuCl/AgTFA / Et3N / >95 / 92d
10 / (Sa)-Binap-AuCl/AgTFA / DABCO / <50 / ___d
11 / (Sa)-Binap-AuCl/AgTFA / DBU / <30 / ___d
12 / (Sa)-Binap-AuCl/AgOAc / _____ / >95 / 70
13 / (Sa)-Binap-AuCl/AgOBz / _____ / >95 / 94d
14 / (Sa)-Binap-AuCl/AgTFA / _____ / >95 / 94
15 / (Ra)-Binap-AuCl/AgTFA / _____ / >95 / ent-94
16 / (Sa)-Binap-(AuCl)2/AgTFA / _____ / >95 / rac.
17 / (Sa)-Binap-AuCl/AgTFAe / _____ / <90 / 60
18 / (Sa)-Binap-AuCl/AgTFAf / _____ / >96 / 94

a Determined by 1H NMR of the crude samples.

b The observed endo:exo ratio was always >98:2 (1H NMR)

c Determined by chiral HPLC analysis (Daicel, Chiralpak AS).

d Notable amounts of unidentified side products were observed (1H NMR).

e The reaction was performed with 5 mol% of gold(I) complex.

f The anion interchange was allowed overnight instead of 1 h.

A series of iminoesters and maleimides were tested following the best reaction conditions described in entry 14 of Table 1 and directly compared with identical transformations carried out with AgTFA (Scheme 2 and Table 2). In all of the examples described in this Table 2 the endo:exo diastereoselectivity was very high (>98:2, determined by 1H NMR spectroscopy) independently of the central metal nature (10 mol%). In general the base-assisted reaction was complete in 16 h giving rise to elevated chemical yields and better enantioselections with the chiral silver complex. However, the absence of base is much more beneficial for the reactions run with chiral gold(I) complex affording both excellent yields and enantioselections (Table 2, compare entries 1-6). It was remarkable the result obtained when NPM was employed as dipolarophile. A racemic product 2ac could be only obtained with the silver(I) catalyst, whilst a 80% ee of this cycloadduct was obtained in the gold(I)-promoted cycloaddition (Table 2, entries 5 and 6). For other arylideneaminoesters 1b-d the behaviour was very similar obtaining a higher enantiodiscrimination for the gold(I)-catalysed processes (Table 2, entries 7-9). When substrate 1e (derived from 2-naphthalenecarbaldehyde) the enantioselectivity of the silver(I)-catalysed process was higher than the ee generated by the analogous reaction developed by the (Sa)-BinapAuTFA complex (Table 2, entry 10).

Scheme 2.

The insertion of a bulky substituent at the α-position of the 1,3-dipole precursor was next evaluated. Thus, methyl benzylideneiminophenylalaninate 3 was allowed to react with NMM under the standard reaction conditions (Scheme 3). The reaction performed with the gold(I) complex needed 24 h more than the corresponding reaction using the analogous silver(I) complex for achieving almost total conversions. The high enantioselection showed by (Sa)-BinapAuTFA complex (99% ee) versus the 65% ee induced by (Sa)-BinapAgTFA complex justified the importance of this complex in this cycloaddtition.

Scheme 3.

According to our experience with the results obtained from the application of chiral Binap-silver(I) complexes in the enantioselective 1,3-DC of azomethine ylide and electrophilic alkenes,5j,k,m we also tested the efficiency of the two complexes in the enantioselective cycloaddition of azomethine ylides and trans-1,2-bis(phenylsulfonyl)ethylene as synthetic equivalent of acetylene (Scheme 4 and Table 3). The reaction performed with gold(I) catalyst offer the best enantioselectivities of cycloadducts 5 using a 5 mol% loading and identical quivalents of DIPEA (Table 3, entries 2, 5 and 8). Lower enantiomeric excesses were determined when higher amounts of catalyst and DIPEA (10 mol%, Table 3, entries 1, 4, and 7) were used, and definitively no reaction was observed in absence of base (Table 3, entries 3, 6, and 9).

Tetrahedron: Asymmetry

Table 2. 1,3-DC between iminoglycinates 1 and maleimides.

(Sa)-BinapAuTFA / (Sa)-BinapAgTFA
Entry / 1 / Ar / R1 / Base / 2 / Yield (%)a,b / ee (%)c / Yield (%)a,b / ee (%)c
1 / 1a / Ph / Me / DIPEA / 2aa / quant. / 70 / quant. / 99
2 / 1a / Ph / Me / _____d / 2aa / 90 / 99 / quant. / 99
3 / 1a / Ph / Et / DIPEA / 2ab / quant. / 70 / 90 / 99
4 / 1a / Ph / Et / _____d / 2ab / quant. / 99 / 91 / 99
5 / 1a / Ph / Ph / DIPEA / 2ac / 90 / 64 / 88 / rac.
6 / 1a / Ph / Ph / _____d / 2ac / 92 / 80 / quant. / rac.
7 / 1b / 2-MeC6H4 / Me / _____d / 2ba / 86 / 88 / 90 / 70
8 / 1c / 2-ClC6H4 / Me / _____d / 2ca / 88 / 99 / 92 / 85
9 / 1e / 4-(MeO)C6H4 / Me / _____d / 2da / 95 / >99 / quant. / 99
10 / 1f / 2-Naphthyl / Me / _____d / 2ea / 94 / 91 / quant. / 99

a Isolated yields after flash chromatography (silica gel).

b The observed endo:exo ratio was always >98:2 (1H NMR)

c Determined by chiral HPLC analysis.

d The reaction needed 48 h for completion.

e The reaction was performed with 5 mol% of gold(I) complex.

f The anion interchange was allowed overnight instead of 1 h.

Tetrahedron: Asymmetry

Scheme 4.

The silver(I) catalyst operated in the absence of base and employing a 10 mol% of caltalyst loading, and the reduction of this amount did not produce such as significant changes in enantioselections as gold(I) catalyst did (Table 3, compare entries 1-9). Taking in account all the possible combinations the best enantioselections were obtained in the presence of (Sa)-BinapAuTFA complex (5 mol%).

Tetrahedron: Asymmetry

Table 3. 1,3-DC between iminoglycinates 1 and trans-1,2-bis(phenylsulfonyl)ethylene.

(Sa)-BinapAuTFA / (Sa)-BinapAgTFA
Entry / 1 / Ar / Catalyst / DIPEA / 5 / Yield (%)a,b / ee (%)c / Yield (%)a,b / ee (%)c
1 / 1a / Ph / 10 mol% / 10 mol% / 5a / 80 / 86 / quant. / 92
2 / 1a / Ph / 5 mol% / 5 mol% / 5a / 74 / 99 / 80 / 86
2 / 1a / Ph / 10 mol% / _____ / 5a / _____ / _____ / 76 / 96
3 / 1d / 4-MeC6H4 / 10 mol% / 10 mol% / 5d / 81 / 88 / 79 / 96
4 / 1d / 4-MeC6H4 / 5 mol% / 5 mol% / 5d / 78 / 99 / 79 / 96
5 / 1d / 4-MeC6H4 / 10 mol% / _____ / 5d / _____ / _____ / 74 / 96
6 / 1g / 3-Pyridyl / 10 mol% / 10 mol% / 5g / 73 / 96 / 75 / 92
7 / 1g / 3-Pyridyl / 5 mol% / 5 mol% / 5g / 73 / 99 / 75 / 98
8 / 1g / 3-Pyridyl / 10 mol% / _____ / 5g / _____ / _____ / 78 / 96

a Isolated yields after flash chromatography (silica gel).

b The observed endo:exo ratio was always >98:2 (1H NMR)

c Determined by chiral HPLC analysis.

Tetrahedron: Asymmetry

One of the main drawbacks of the Binap-silver(I) catalysed 1,3-DC was the reaction with acrylates. In spite of using the best reaction conditions (see above) of Table 1, the reaction of 1,3-dipole precursors 1 with tert-butyl acrylate only produced racemic mixtures of cycloadducts. However, when iminoesters 6 or 7 (appropriate starting materials to synthesize hepatitis C virus inhibitors)[11] were allowed to react with tert-butyl acrylate catalysed by (Sa)-Binap-AuTFA, at rt, for 48h, in the presence of base (Et3N, 10 mol%, rather than DIPEA) intermediates 8 or 9 were isolated with different success (Scheme 5a, and b). Whilst thienyl derivative 6 furnished a racemic endo-cycloadduct 8, thiazole derivative 4 generated proline endo-6 with a relative good enantiomeric excess (Scheme 5c). The best encouraging result was obtained when this last reaction was performed at 0 ºC furnishing endo-cycloadduct 9 in good isolated yield (88%) and excellent enantioselectivity (99% ee). In both of the reported examples, the unique diastereoisomer identified in the crude product 1H NMR spectra was the endo-isomer.