by addition to methyl chloroformate (5 equiv) and final
hydrolysis [(a) LiOH, MeOH; (b) HClaq] furnished the
carboxylic acids 4a-c in g90% yield. Condensation of 4a-c
(2 equiv) with (R)-2,2′-diamino-1,1′-dinaphthyl (5) [(PhO)3P,
Py)8 gave the bis(amides) 6a-c (80-90%) which upon
reduction (AlH3‚N-methylpyrrolidine)9 and reductive methyl-
ation (NaBH4, CH2O) delivered the desired proligands 7a-c
(61-70%, Scheme 1).
(C6D6), the same reaction occurred in only 12 h at 60 °C
when conducted in the presence of thiophene (2 equiv). In
the latter case, the cyclization of 10a took place in 15 h at
60 °C in the presence of 5 mol % of 9a (>95% conversion)12
to provide 11a in 78% ee (Scheme 2).13 Significantly, the
Scheme 2
Scheme 1
cyclization of 10a at 30 °C [9a (5 mol %), C4H4S (10 mol
%), 23 d] proVided 11a in 89% ee and >95% conVersion.
These are, to our knowledge, the highest ee’s obtained for
the intramolecular hydroamination of 10a achieved to date.
The cyclization of several other aminoalkenes (e.g., 10b-
e) catalyzed by chelates generated from 7a-c and 8 in the
presence of thiophene were then conducted. These results
are summarized in Table 1.14 It is significant that asymmetric
hydroaminations could readily be achieved on a preparative
scale. In this context, cyclization of aminoalkenes 10a and
10d on a 3 mmol scale in the presence of the yttrium chelate
The preparation of both metallocene and nonmetallocene
complexes of the group 3 metals has frequently been
achieved via amine elimination using Ln[N(SiMe2H)2]3‚
(THF)2 by virtue of the enhanced kinetic basicity10 of these
amides relative to commercially available reagents such as
Y[N(TMS)2]3 (8).11 Invariably, the use of the former class
of amides leads to the presence of THF as a complexing
agent in the reaction medium. The possibility that auxiliary
ligands [e.g., THF, 4-DMAP, (n-Bu)3P, Et2S, and thiophene)
might influence the rates of complexation of 8 and hydro-
amination by 9a as well as alter the enantioselectivity of
C-N bond formation was gauged by the addition of these
ligands (2 equiv/equiv 8 and 7a) prior to amine elimination.
Of these auxiliary ligands, the addition of THF led to both
suppression of the rate of hydroamination and ee. In contrast,
the presence of thiophene increased the efficiency of pro-
ligand metalation to generate the putatiVe yttrium chelate,
presently formulated as 9a, but does not affect the rate of
hydroamination or the obserVed ee. In a typical set of
experiments, complexation of 7a with Y[N(TMS)2]3 (8) and
subsequent cyclization of amine 10a was conducted both in
the presence and absence of thiophene. Whereas chelate
formation between 7a and 8 alone required 24 h at 120 °C
(12) Conversions were based on 1H NMR integration relative to p-xylene
as the internal standard.
(13) Enantiomeric excesses and absolute configurations of the products
1
were determined by H NMR after conversion to the diastereomeric (-)-
O-acetylmandelate15 or Mosher’s acid16 derivatives following quantitative
vacuum transfer from the catalyst.
(14) General Procedure for Asymmetric Aminoalkene Hydro-
aminations. In an argon-filled glovebox, Y[N(TMS)2]3 (8) (9.1 mg, 0.016
mmol), the appropriate bis(thiol) proligand 7 (0.016 mmol), C6D6 (0.7 mL),
and thiophene (2.7 mg, 2.6 µL, 0.032 mmol) were introduced sequentially
into a J. Young NMR tube equipped with Teflon screw cap. The
homogeneous reaction mixture was maintained at 60 or 75 °C, respectively,
in a constant temperature oil bath until ligand attachment was judged
complete by the disappearance of the Y[N(TMS)2]3 resonance in the 1H
NMR spectrum, with concomitant production of (TMS)2NH. To the resulting
complex was added the appropriate aminoalkene 10 (0.32 mmol) and the
reaction mixture was subsequently heated at 30 or 60 °C in an oil bath to
1
achieve hydroamination. H NMR spectroscopy using a pulse delay of 10
s to avoid signal saturation was employed to monitor reaction progress.
The cyclic amines so produced were vacuum transferred along with the
C6D6 at 10-3 Torr to a 5 mL round-bottomed flask containing (R)-(-)-O-
acetylmandelic acid15 [or (R)-(+)-2-methoxy-2-(trifluoromethyl)phenylacetic
acid16 in the case of 11e] (0.32 mmol) at -78 °C. This transfer was
quantitated by washing the NMR tube with a small amount of CDCl3. The
resulting mixture was stirred at 22 °C for 2 h, and the volatile components
were removed in vacuo. The resulting diastereomeric salt was then dissolved
in CDCl3 and the enantiomeric excesses were determined by 1H NMR
spectroscopy. Procedures for representative asymmetric hydroaminations
conducted on a preparative (3 mmol) scale are provided in the Supporting
Information.
(7) Block, E.; Eswarakrishnan, V.; Gernon, M.; Ofori-Okai, G.; Saha,
C.; Tang, K.; Zubieta, J. J. Am. Chem. Soc. 1989, 111, 658.
(8) Adolfson, H.; Moberg, C. Tetrahedron: Asymmetry 1995, 6, 2023.
(9) Martlett, E. M.; Park, W. S. J. Org. Chem. 1990, 55, 2968.
(10) Eppinger, J.; Spiegler, M.; Hieringer, W.; Herrmann, W. A.;
Anwander, R. J. Am. Chem. Soc. 2000, 122, 3080.
(11) Available from Gelest, Inc., 612 William Leigh Drive, Tullytown,
PA 19007-6308.
1738
Org. Lett., Vol. 7, No. 9, 2005