complexes such as [BinapAuTFA]2 afforded modest results
for the cycloaddition of methyl benzylideneglycinate and
nitrostyrene (up to 80:20 dr and 85:15 er).9 On the other
hand, benzophenone-derived N-(diphenylmethylene) glyci-
nates have also been employed as azomethine ylide precur-
sors in the presence of chiral silver catalysts10 and organo-
catalysts.11 In general, only glycinate derived imino esters
have been employed as azomethine ylide precursors except
in the case of the ligand 7 and Cu(OTf)2 which catalyzed the
1,3-DC with the corresponding alaninate. In many of these
examples the elucidation of the reaction pathways has been
studied by both DFT calculations12 and experimental results.4e
Initially, we selected (Sa)-Monophos 8 and (Sa,R,R)-9
(Figure 2) as chiral phosphoramidites for the preliminary
catalyzed 1,3-DC between methyl N-benzylideneglycinate
10a and β-nitrostyrene 11a, in toluene as solvent, at rt for
17 h (25 °C, Table 1). When (Sa)-Monophos 8 Cu(OTf)2
3
was used as the catalyst mainly racemic endo-2a was
obtained (Table 1, entry 1). However, in the case of
(Sa,R,R)-9 Cu(OTf)2, 88/11 dr and excellent enantioselec-
3
tion >99:1 were obtained for the exo-diastereoisomer 2a
(Table 1, entry 2). When using the enantiomeric ligand
(Ra,S,S)-9 the corresponding enantiomer exo-2a was
mainly isolated (Table 1, entry 3). By contrast, the complex
formed by phosphoramidite (Sa,S,S)-9 and Cu(OTf)2
demonstrated to be a mismatched combination because
the reaction gave the opposite diastereoselection with no
enantioselection (Table 1, entry 4). Cu(OTf)2 was the most
appropriate copper(II) salt rather than Cu(OAc)2 in terms of
both diastereo- and enantioselection (Table 1, compare
entries 2 and 5). Copper(I) bromide did not afford the
expected results, while Cu(OTf) C6H6 showed the same
3
result that was obtained in the reaction run with Cu(OTf)2
(Table 1, entries 6 and 7). We selected the catalyst formed by
Cu(OTf)2 because reactions involving copper(I) usually re-
quire an inert atmosphere and degassed solvents in order to
avoid dismutation. The presence of an external base is crucial
for the reaction success, triethylamine being more adequate
than DIPEA and DABCO (Table 1, compare entry 2 with
entries 8ꢀ10). The solvent effect was also dramatic because
almost racemic mixtures of the product 2a were isolated
when Et2O, MeCN, or DCM was employed, although in
the last example the diastereomeric exo/endo ratio was the
highest achieved in this transformation and in very good
yields (Table 1, entries 11ꢀ14). Unexpectedly, in all the cases,
cycloadduct endo-2a was obtained in racemic form.
Smaller amounts of a catalyst loading (3 mol %) in the
reaction gave a lower yield (55%) and an enantioselectivity
similar of that of 2a (not included in Table 1). The absolute
configuration of exo-cycloadduct 2a was established ac-
cording to the retention times in HPLC using chiral
columns and comparison with the data obtained for the
same known product.4c,9
Figure 1. Useful nitro-substituted prolines 1 and 2 and pre-
viously reported chiral catalysts for the enantioselective 1,3-
DC of imino esters and nitroalkenes.
We envisaged that the use of chiral phosphoramidites 8
and 9,13 as monodentate privileged ligands,14 could be a
good alternative to the described sophisticated ligands for
copper salts15 to be used as chiral catalysts in the general
asymmetric 1,3-DC of azomethine ylides, derived from
R-amino acids, and nitroalkenes.
The reaction of nitrostyrene 11a and imino ester 10a was
studied at lower temperatures. At ꢀ80 °C a 1:1 mixture of
the corresponding exo-cycloadduct-2a and the syn-imino
ester 12a was obtained. After acidic treatment at ꢀ80 °C
and simple extractive workup, the corresponding syn-
amino ester 13a and exo-2a hydrochloride were isolated
(Scheme 1). Diastereomeric ratios of 13a and enantiomeric
ratios of both exo/endo-2a and syn-13a were independent
of the working temperature.16
Figure 2. Employed chiral phosphoramidites.
(8) Arai, T.; Yokoyama, N.; Mishiro, A.; Sato, H. Angew. Chem., Int.
Ed. 2010, 49, 7895.
ꢀ ꢀ
(9) Martı
´
n-Rodrıguez, M.; Najera, C.; Sansano, J. M.; de Cozar, A.;
´
Cossıo, F. P. Chem.;Eur. J. 2011, 17, 14224.
´
(10) Imae, K.; Konno, T.; Ogata, K.; Fukuzawa, S.-I. Org. Lett.
2012, 14, 4410.
(11) (a) Xie, J.; Yoshida, K.; Takasu, K.; Takemoto, Y. Tetrahedron
Lett. 2008, 49, 6910. (b) Xue, M.-X.; Zhang, X.-M.; Gong, L.-Z. Synlett
2008, 691.
(14) The versatility and efficiency of the chiral phosphoramidite
ꢀ
Retamosa, M. G.; Sansano, J. M. Angew. Chem., Int. Ed. 2008, 47, 6055.
9ꢀAgClO4 complex was described first by our group: (a) Najera, C.;
ꢀ
(b) Najera, C.; Retamosa, M. G.; Martı
´
n-Rodrı
ꢀ
´
J. M.; de Cozar, A.; Cossıo, F. P. Eur. J. Org. Chem. 2009, 5622.
´
guez, M.; Sansano,
ꢀ
(12) For a review, see: de Cozar, A.; Cossı
2011, 13, 10858.
(13) (a) Teichert, J. F.; Feringa, B. L. Angew. Chem., Int. Ed. 2010, 49,
2486. (b) Privileged Chiral Ligands and Catalysts; Zhou, Q.-L., Ed.; Wiley-
VCH: New York, 2011.
´
o, F. P. Phys. Chem. Chem.
(15) Chiral phosphoramidite 9ꢀCuOTf has been isolated and char-
acterized (IR, 1H NMR, 13C NMR, 31P NMR, 19F NMR, and HRMS)
by: Maksymowicz, R. M.; Roth, P. M. C.; Fletcher, S. P. Nat. Chem.
2012, 4, 649.
(16) See Supporting Information for more details.
B
Org. Lett., Vol. XX, No. XX, XXXX