A. Pfaltz and A. Baeza
high H2 pressure, but the enantioselectivities were consider-
ably higher (99% ee for catalysts A2 and B).[7] As observed
for the N-Boc derivative, complex B proved to be the most
reactive catalyst, giving high conversion under the condi-
tions shown in Scheme 6. The need for more forcing condi-
Scheme 8. Asymmetric hydrogenation of 3-subsituted N-tosyl indoles 9a.
i) H2 (100 bar), C (2.5 mol%), CH2Cl2, 608C, 30 h.
In summary, our results show that cationic Ir catalysts
with chiral N,P ligands are efficient catalysts for the asym-
metric hydrogenation of N-protected indoles. The ee values
are in the same range as those reported for Ru and Rh com-
plexes with the (S,S)-(R,R)-PhTRAP ligand, which were the
only suitable catalysts known for this substrate class up to
now.[5] However, in contrast to Rh- and Ru-catalyzed hydro-
genations, no base additives are needed. A further advant-
age of the Ir catalysts is their air and moisture stability. The
results obtained with N-Boc-, N-acetyl-, and N-tosylindoles
demonstrate that the protecting group influences both the
reactivity and enantiomeric excess. With the right combina-
tion of catalyst and protecting group, high yields and excel-
lent enantioselectivities can be achieved for various 2- and
3-substituted indoles.
Scheme 6. Asymmetric hydrogenation of 2-subsituted N-tosyl indole 7a.
i) H2 (100 bar), B (2 mol%), CH2Cl2, 258C, 30 h.
tions could be due to coordination of the catalyst to the sul-
fone group, which could lower the activity of the catalyst,
but on the other hand cause the remarkable increase in
enantioselectivity (Scheme 6).
Next, other 2-substituted N-tosyl-indoles were examined.
As before, 2-phenyl and 2-carboethoxyindole 7b and 7c
were subjected to hydrogenation under the conditions used
for 7a. The best results were obtained with catalyst C but,
despite high enantioselectivities, the conversions remained
low. However, with higher catalyst loading and sequential
catalyst addition (see above), we were able to successfully
hydrogenate substrate 7c with high conversion and excellent
enantioselectivity when the temperature was raised to 608C
(Scheme 7). However, when the same conditions were ap-
Experimental Section
General procedure for the iridium-catalyzed asymmetric hydrogenation
of N-protected indoles: A solution of the N-protected indole (0.2 mmol)
and the iridium complex (1–2.5 mol%) in dry dichloromethane (1 mL,
previously filtered over basic alumina) under an inert atmosphere was
placed in an autoclave, which was sealed, purged with hydrogen, and
pressurized to the desired hydrogen pressure. After the mixture had been
stirred at the desired temperature for the indicated time, the solvent was
evaporated, and the catalyst was removed by filtration through a short
silica gel column (3ꢁ1 cm) with pentane/diethyl ether (1:1, 25 mL) as
eluent to give the corresponding N-protected indoline after evaporation
of the solvent.
When the reaction was carried out with sequential addition of catalyst,
the same procedure was followed, but after 8 h at the reported tempera-
ture, the autoclave was allowed to cool down to 30–408C and opened
after the hydrogen pressure had been released. Then a solution of the iri-
dium complex in the corresponding dry solvent (100 mL, previously fil-
tered over basic alumina) was syringed into the reaction vessel. The auto-
clave was sealed again, purged with hydrogen, and repressurized. Then
stirring was continued at the reported temperature for an additional 16 h.
Scheme 7. Asymmetric hydrogenation of 2-subsituted N-tosyl indoles 7b
and 7c. i) H2 (100 bar), C (4 mol%), solvent, 24 h.
plied to the 2-phenylindole 7b, conversion only reached
55%. Even at 1108C in chlorobenzene the conversion did
not exceed 70%. On the other hand, the enantiomeric
excess remained at 98% despite the high temperature
(Scheme 7).
Among the 3-substituted N-tosylindoles, the 3-methyl de-
rivative 9a reacted with perfect enantioselectivity (>99%
ee) when using catalyst C under standard conditions, but the
conversion was moderate (52%). Therefore, the reaction
was carried out at elevated temperature (Scheme 8). Most
gratifyingly, almost quantitative conversion was achieved at
608C, while the ee remained at very high level of 98%. N-
Tosylindoles with carboxylic ester groups at C-3 (9b, R=
CO2Me; 9c, R=CH2CO2Et) showed extremely low reactivi-
ty and gave only very low or no conversion at all, even
under forcing conditions.
For catalyst screening, reactions were performed on a 0.1 mmol scale.
A preparative experiment was also carried out by using 2-methyl-N-Boc-
protected indole 1a and catalyst A2. The general procedure was fol-
lowed, but using a 0.4m solution (instead of 0.2m) of 1a (578 mg,
2.5 mmol) in chlorobenzene (6.25 mL) and 1 mol% of A2 catalyst
(38 mg). After the mixture had been stirred at 258C under H2 (50 bar)
for 6 h. The solvent was evaporated, and the catalyst was removed by fil-
tration through a silica gel column (2ꢁ12 cm) with pentane/diethyl ether
(1:1, 100 mL) as eluent. The pure indoline 2a was obtained in 94% yield
(544 mg) with 92% ee.
2038
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2036 – 2039