Asymmetric hydrogenation of α,β-unsaturated nitriles with base-activated iridium N,P ligand complexes

Article abstract of DOI:10.1002/anie.201402053

Although many chiral catalysts are known that allow highly enantioselective hydrogenation of a wide range of olefins, no suitable catalysts for the asymmetric hydrogenation of α,β-unsaturated nitriles have been reported so far. We have found that Ir N,P ligand complexes, which under normal conditions do not show any reactivity towards α,β-unsaturated nitriles, become highly active catalysts upon addition of N,N- diisopropylethylamine. The base-activated catalysts enable conjugate reduction of α,β-unsaturated nitriles with H₂ at low catalyst loadings, affording the corresponding saturated nitriles with high conversion and excellent enantioselectivity. In contrast, alkenes lacking a conjugated cyano group do not react under these conditions, making it possible to selectively reduce the conjugated C=C bond of an α,β-unsaturated nitrile, while leaving other types of C=C bonds in the molecule intact.

Full text of DOI:10.1002/anie.201402053

.
Angewandte
Communications
DOI: 10.1002/anie.201402053
Asymmetric Hydrogenation
Asymmetric Hydrogenation of a,b-Unsaturated Nitriles with Base-
Activated Iridium N,P Ligand Complexes**
Marc-Andrꢀ Mꢁller and Andreas Pfaltz*
Dedicated to the MPI fꢁr Kohlenforschung on the occasion of its centenary
Abstract: Although many chiral catalysts are known that allow
highly enantioselective hydrogenation of a wide range of
olefins, no suitable catalysts for the asymmetric hydrogenation
of a,b-unsaturated nitriles have been reported so far. We have
found that Ir N,P ligand complexes, which under normal
conditions do not show any reactivity towards a,b-unsaturated
nitriles, become highly active catalysts upon addition of N,N-
diisopropylethylamine. The base-activated catalysts enable
conjugate reduction of a,b-unsaturated nitriles with H2 at low
catalyst loadings, affording the corresponding saturated nitriles
with high conversion and excellent enantioselectivity. In
contrast, alkenes lacking a conjugated cyano group do not
react under these conditions, making it possible to selectively
borohydride and semicorrin cobalt catalysts led to the
saturated nitriles with only moderate enantioselectivity (up
to 69% ee).[5] Better enantioselectivities of up to 99% ee were
achieved with copper diphosphine catalysts and polymethyl-
hydrosiloxane (PMHS) as the reducing agent, but relatively
high catalyst loadings (3 mol%), a fourfold excess of PMHS,
and strongly basic work-up conditions (aq. NaOH) are
drawbacks of this method.[6] For rhodium-catalyzed hydro-
genation with hydrogen gas, high enantioselectivities have
only been observed for unsaturated nitriles with an additional
=
coordinating acetamido or carboxylate group at the C C
bond.[7]
Herein we report that chiral iridium N,P ligand complexes
can be activated by addition of a base, thus enabling the
hydrogenation of a,b-unsaturated nitriles with high enantio-
selectivities at low catalyst loadings.
=
reduce the conjugated C C bond of an a,b-unsaturated nitrile,
=
while leaving other types of C C bonds in the molecule intact.
A
symmetric hydrogenation is one of the most general, most
Our initial attempts to apply iridium-based complexes to
the asymmetric hydrogenation of a,b-unsaturated nitriles
under standard conditions, which had proved optimal for
widely applied reactions to generate enantiomerically
enriched products from prochiral starting materials.[1] Many
efficient chiral rhodium and ruthenium diphosphine catalysts
a
wide variety of functionalized and unfunctionalized
alkenes,[3] were unsuccessful. None of the many iridium
catalysts tested showed any reactivity towards this class of
substrate. Moreover, addition of a,b-unsaturated nitriles to
other substrates which were known to be fully hydrogenated
by iridium catalysts under standard conditions, resulted in
complete catalyst deactivation (Scheme 1A). Further inves-
tigation of the interaction of a,b-unsaturated nitriles with an
Ir/PHOX catalyst under a hydrogen atmosphere led to the
identification and isolation of the iridium(III) dihydride
complex 4 with two bound nitrile molecules (Scheme 1B;
BArF = tetrakis[bis-3,5-(trifluoromethyl)phenyl]borate).
Apparently, coordination of the two cyano groups leads to an
18-electron complex lacking a free coordination site, which is
necessary for catalytic activity.
=
are available and hydrogenate C C bonds bearing an
adjacent coordinating group with high enantioselectivity.[2]
More recently, iridium complexes derived from chiral N,P,
C,N or O,P ligands were developed to extend the application
range of asymmetric hydrogenation to olefins lacking a coor-
dinating group.[3] Although a very broad range of function-
alized and unfunctionalized olefins can be converted into
highly enantioenriched products using rhodium, ruthenium,
or iridium catalysts, asymmetric hydrogenation of a,b-unsa-
turated nitriles has met limited success so far. The challenges
imposed by this substrate class are related to the linear
geometry of the nitrile group, which keeps a coordinated
catalyst away from the C C bond, and the strong binding
affinity of the cyano group to transition-metal complexes
results in catalyst deactivation.
Hence, only a few examples of asymmetric reductions of
a,b-unsaturated nitriles have been reported so far, despite the
considerable synthetic value of the enantioenriched nitriles
produced in this way.[4] Conjugate reduction with sodium
=
Semeniuchenko et al. recently found that addition of N,N-
diisopropylethylamine (DIPEA) markedly enhanced the
reactivity of Ir/PHOX catalysts for the hydrogenation of
[*] M.-A. Mꢀller, Prof. Dr. A. Pfaltz
Department of Chemistry, University of Basel
St. Johanns-Ring 19, 4056 Basel (Switzerland)
E-mail: andreas.pfaltz@unibas.ch
[**] Support of this work by the Swiss National Science Foundation is
gratefully acknowledged.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201402053.
Scheme 1. Inhibition of the activated iridium catalyst by the substrate.
8668
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 8668 –8671

Angewandte
Chemie
electron-deficient olefins such as benzylidenemalononitrile or
a,b-unsaturated ketones,[8] and is in contrast to the general
experience that amine bases deactivate iridium catalysts.[3d,9]
These results prompted us to evaluate the effect of DIPEA on
the hydrogenation of 3a with various iridium catalysts
(Scheme 2).
and full conversion in methanol (see the Supporting Infor-
mation, Table 1).
To establish the scope of this catalyst system, we examined
a broad selection of a,b-unsaturated nitriles under optimized
reaction conditions (Scheme 3). Changing the double-bond
Scheme 2. Selected catalysts used in the hydrogenation of (E)-3-
phenylbut-2-enenitrile (3a).
Although the first experiment with catalyst A gave only
minimal conversion of less than 1%, further studies covering
a broad selection of iridium complexes (see the Supporting
Information) led to more reactive and more selective
catalysts. By far the best results were obtained with the
commercially available ThrePHOX complex C and the
SerPHOX analogue E. The reactivity and enantioselectivity
showed extreme variation from catalyst to catalyst. Even
small changes in the ligand structure had a strong impact on
catalyst performance (compare Avs. B or C vs. D). Both steric
and electronic factors seem to play a role, but no clear trends
could be deduced from the data.
We next studied the influence of the various reaction
parameters in the hydrogenation of 3a. An increase of the
hydrogen pressure from 50 to 100 bar resulted in full
conversion without affecting the enantioselectivity. Variation
of the amount of DIPEA added to the reaction mixture had
a very strong effect. The 4.5 equivalents originally chosen
were found to be optimal. Both an increase or reduction of
the amount of DIPEA impaired catalyst performance. With
1.5 equivalents of DIPEA the reaction still went to comple-
tion, but the ee value dropped from 94% to 88%, whereas
reduction to 0.1 equivalent led to a significant loss of ee value
and incomplete conversion. Raising the amount of DIPEA
from 4.5 to 18 equivalents also caused a drop in ee value and
conversion.
Scheme 3. Catalytic asymmetric hydrogenation of a,b-unsaturated
nitriles. [a] 2.0 mol% catalyst loading and 16 h reaction time.
The range of solvents tolerated proved to be remarkably
broad. To our surprise, even in strongly coordinating solvents
such as methanol or acetonitrile, which are known to inhibit
iridium catalysts such as C, catalytic activity was retained.
Among the polar protic and aprotic solvents that were tested,
only DMSO was found to be unsuitable. The highest ee value
was observed in DMF, but the reaction did not go to
completion. Full conversion and excellent enantioselectivities
of 94–95% ee were achieved in alcoholic solvents and in
water.[10] The remarkable water tolerance of this catalyst
system is an attractive feature enhancing its scope and
practicality. By lowering the temperature to 08C the enantio-
selectivity could be further improved, thus reaching 98% ee
geometry of 3a from trans to cis caused a large drop in
conversion and enantioselectivity, with a reversal of the
absolute configuration (3b). Introduction of an ortho sub-
stituent onto the phenyl group, which destabilizes a coplanar
=
arrangement of the C C bond and the aromatic p-system,
also reduced the reactivity and ee value, but to a lower extent
(3c). Electron-donating or electron-withdrawing groups in
the meta or para position of the aryl substituent had no or only
small effects (3d–i, 3l–p). The substrates 3j,k, having larger
=
alkyl substituents at the C C bond, gave full conversion with
similar or even higher enantioselectivity compared to the
methyl-substituted compound 3a. Overall this catalyst system
showed high functional-group tolerance, as exemplified by
Angew. Chem. Int. Ed. 2014, 53, 8668 –8671
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 8669

.
Angewandte
Communications
the fluoro-, chloro-, bromo-, acetamido-, carbomethoxy-,
cyano-, and nitro-substituted compounds which were all
successfully hydrogenated.
which is known to give full conversion with catalyst C in
CH2Cl2 (Scheme 4A).[12] In pure methanol some reactivity
remained, but conversions were significantly lower than in
CH2Cl2 and the ee value fell from 99% to 11%. Addition of
DIPEA quenched the reaction almost completely. Notably,
when a 1:1 mixture of (E)-3-phenylbut-2-enenitrile (3a) with
1 was subjected to these reaction conditions, the unsaturated
nitrile was fully hydrogenated, while 1 did not react (Sche-
me 4B). Apparently, DIPEA enhances the reactivity of the
The b,b-dialkyl-substituted acrylonitriles 3q and 3r
exhibited lower reactivity and required 2 mol% of catalyst
and longer reaction time for full conversion at 08C. The
enantioselectivities were also lower than in the hydrogenation
of the phenyl methyl analogue 3a. Interestingly, the cis-
isomer 3r afforded a higher ee value than the trans-isomer 3q,
contrary to the results obtained with 3a and 3b. The best
results were achieved in the hydrogenation of 3r at 08C,
which led to the saturated nitrile with 82% ee and full
conversion. At À158C the ee value was increased to 88%, but
conversion reached only 51%.
=
catalyst towards the electrophilic C C bond of an a,b-
unsaturated nitrile while inhibiting hydrogenation of more-
[12]
=
electron-rich, less polarized C C bonds.
Thus, selective
hydrogenation of a,b-unsaturated nitriles containing addi-
=
tional C C bonds becomes possible, as demonstrated by the
=
We next investigated to what extent the catalyst loading
and the amount of DIPEA could be reduced without affecting
the enantioselectivity and conversion. In methanol containing
0.23 equivalents of DIPEA as little as 0.05 mol% of catalyst
was necessary to achieve 99% to full conversion within
18 hours. With the exception of 3d (91% ee), the enantiose-
lectivities for 3a, 3g, 3i, 3j, 3n, and 3o remained at the same
levels as those observed in reactions with 1 mol% of catalyst.
At longer reaction times of 112 hours, 96.5% conversion with
95% ee was observed for 3a on a 1.5 mmol scale using only
0.005 mol% catalyst (corresponding to 19300 turnovers). In
methanol with 1 mol% catalyst, the amount of DIPEA could
be reduced to 0.1 equivalents without affecting the ee value
and conversion. In this solvent, even in the absence of
DIPEA, the reaction went to completion, but the ee value
dropped from 95% to 86% (see the Supporting Information,
Table 2).
reduction of 3s (Scheme 4C). The cyano-substituted C C
bond was almost fully hydrogenated to afford 93% of the
monohydrogenation product 5s with 96% ee, and only 6% of
the fully hydrogenated product 5s’. The high preference for
reduction of the trisubstituted C C bond is remarkable in
view of the very high reactivity of terminal C C bonds under
=
=
normal hydrogenation conditions.[11]
Obviously, this base-modified catalyst system must oper-
ate through a different mechanism than under standard base-
free conditions. In view of the pronounced Brønsted acidity of
cationic iridium hydride complexes,[13] a deprotonated neutral
iridium(I) monohydride may be postulated as a reactive
intermediate which is generated by deprotonation of a dihy-
dride complex such as 4 (Scheme 1). A neutral hydride
complex would be expected to be less electrophilic and,
consequently, to release a bound nitrile more easily, thus
opening up a free coordination site, which is required for the
reaction. Moreover, the hydride would be more nucleophilic
than hydrides in a cationic dihydride complex, thus facilitat-
We wondered how standard substrates lacking a nitrile
group behaved under these reaction conditions in methanol
with and without addition of DIPEA. We therefore studied
the hydrogenation of the typical alkene 1 shown in Scheme 4,
=
ing hydride transfer to the electrophilic C C bond of an a,b-
unsaturated nitrile. However, experimental evidence for such
a mechanism is difficult to obtain because of the high
reactivity and sensitivity of iridium hydride species,[14] and
attempts to characterize intermediates other than
(Scheme 1) were unsuccessful so far.
4
Considering the dramatic influence of an added external
base, we wondered whether replacement of the BArF ion by
a basic weakly coordinating anion would bring about a similar
effect.
Indeed complex F, having the sterically hindered 2,6-
ditert-butyl-4-nitrophenolate anion, showed high catalytic
activity in the hydrogenation of 3a in CH2Cl2 and afforded
essentially the same conversion and enantioselectivity as the
corresponding BArF salt E in the presence of DIPEA
(Table 1). High conversion was achieved down to 0.2 mol%
catalyst loading. When the catalyst loading was further
reduced, the ee value remained high, however, conversions
were much lower than that with catalyst E and 0.23 equiv-
alents of DIPEA. This can be explained by degradation of the
phenolate complex F, and was observed over time in solution.
Although the maximum achievable turnover numbers are
lower than those with the DIPEA-modified catalyst systems,
the phenolate complex F offers the advantage to carry out
hydrogenations without addition of excess base under nearly
neutral conditions.
Scheme 4. Reactivity and selectivity studies of a,b unsaturated nitriles
and other alkenes.
8670 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 8668 –8671

Angewandte
Chemie
Table 1: Comparison of catalyst F and E/DIPEA in the hydrogenation of
(E)-3-phenylbut-2-enenitrile (3a).
[1] a) D. J. Ager, A. H. M. de Vries, J. G. de Vries, Chem. Soc. Rev.
2012, 41, 3340 – 3380; b) J. M. Brown in Hydrogenation of
Functionalized Carbon-Carbon bonds, Springer, Berlin, 1999;
c) N. B. Johnson, I. C. Lennon, P. H. Moran, J. A. Ramsden, Acc.
Chem. Res. 2007, 40, 1291 – 1299.
[2] a) G. Erre, S. Enthaler, K. Junge, S. Gladiali, M. Beller, Coord.
Chem. Rev. 2008, 252, 471 – 491; b) A. J. Minnaard, B. L. Feringa,
L. Lefort, J. G. de Vries, Acc. Chem. Res. 2007, 40, 1267 – 1277;
c) W. Zhang, Y. Chi, X. Zhang, Acc. Chem. Res. 2007, 40, 1278 –
1290; d) W. Tang, X. Zhang, Chem. Rev. 2003, 103, 3029 – 30700.
[3] a) J. J. Verendel, O. Pꢀmies, M. Diꢁguez, P. G. Andersson, Chem.
Rev. 2014, DOI: 10.1021/cr400037u; b) Y. Zhu, K. Burgess, Acc.
Chem. Res. 2012, 45, 1623 – 1636; c) D. Woodmansee, A. Pfaltz,
Top. Organomet. Chem. 2011, 34, 31 – 76; d) S. J. Roseblade, A.
Pfaltz, Acc. Chem. Res. 2007, 40, 1402 – 1411.
[4] a) F. F. Fleming, Nat. Prod. Rep. 1999, 16, 597 – 606; b) F. F.
Fleming, L. Yao, P. C. Ravikumar, L. Funk, B. C. Shook, J. Med.
Chem. 2010, 53, 7902 – 7917; c) R. A. Michelin, M. Mozzon, R.
Bertani, Coord. Chem. Rev. 1996, 147, 299 – 338; d) V. Laksh-
mana Rao, A. Saxena, K. N. Ninan, J. Macromol. Sci. Polym.
Rev. 2002, 42, 513 – 540.
[5] M. Misun, A. Pfaltz, Helv. Chim. Acta 1996, 79, 961 – 972.
[6] a) D. Lee, D. Kim, J. Yun, Angew. Chem. 2006, 118, 2851 – 2853;
Angew. Chem. Int. Ed. 2006, 45, 2785 – 2787; b) K. Yoo, H. Kim,
J. Yun, Chem. Eur. J. 2009, 15, 11134 – 11138.
[7] a) M. Ma, G. Hou, T. Sun, X. Zhang, W. Li, J. Wang, X. Zhang,
Chem. Eur. J. 2010, 16, 5301 – 5304; b) M. J. Burk, P. D. de Kon-
ing, T. M. Grote, M. S. Hoekstra, G. Hoge, R. A. Jennings, W. S.
Kissel, T. V. Le, I. C. Lennon, T. A. Mulhern, J. A. Ramsden,
R. A. Wade, J. Org. Chem. 2003, 68, 5731 – 5734.
[8] a) V. Semeniuchenko, V. Khilya, U. Groth, Synlett 2009, 271 –
275; b) V. Semeniuchenko, T. E. Exner, V. Khilya, U. Groth,
Appl. Organomet. Chem. 2011, 25, 804 – 809.
[9] a) R. H. Crabtree, H. Felkin, G. E. Morris, J. Organomet. Chem.
1977, 141, 205 – 215; b) P. Schnider, PhD thesis, University of
Basel, 1996; c) D. G. Blackmond, A. Lightfoot, A. Pfaltz, T.
Rosner, P. Schnider, N. Zimmermann, Chirality 2000, 12, 442 –
449.
[10] In H2O a dispersion of the substrate was formed.
[11] F. Menges, A. Pfaltz, Adv. Synth. Catal. 2002, 344, 40 – 44.
[12] Hydrogenation of an unsaturated ester, ketone, and nitro
analogue of nitrile 3a showed conversions below 46% under
the reaction conditions denoted in Scheme 3 at RT.
Entry Cat. Cat. loading [mol%] DIPEA t [h] Conv. [%][a] ee [%][b]
[equiv]
1
2
3
4
5
6
E
F
E
F
E
F
1.0
1.0
0.2
0.2
0.05
0.05
4.5
0
0.9
0
0.23
0
4
4
4
4
16
72
>99
>99
95
92
74
94 (+)
94 (+)
95 (+)
92 (+)
91 (+)
94 (+)
7
[a] Determined by GC on an achiral stationary phase. [b] Determined by
HPLC on a chiral stationary phase.
In summary, we have found that addition of an external
sterically hindered amine base or the presence of a basic
counterion in the catalyst dramatically enhances the catalytic
activity of iridium N,P ligand complexes in the hydrogenation
of a,b-unsaturated nitriles. While these catalysts show no
reactivity towards a,b-unsaturated nitriles under normal
base-free conditions, the base-activated catalyst systems
allow smooth hydrogenation at low catalyst loadings, thus
affording the corresponding saturated nitriles with high
conversion and excellent enantioselectivity. In contrast,
other alkenes lacking a conjugated cyano group do not react
under these reaction conditions. So it becomes possible to
=
selectively reduce the cyano-substituted C C bond of an a,b-
=
unsaturated nitrile, thus leaving less electrophilic C C bonds
in the molecule intact. Thus the catalyst systems described
herein enable enantio- and chemoselective hydrogenations,
for which no suitable catalysts were available before.
Received: February 3, 2014
Published online: March 20, 2014
[13] Y. Zhu, Y. Fan, K. Burgess, J. Am. Chem. Soc. 2010, 132, 6249 –
6253.
[14] a) S. Gruber, M. Neuburger, A. Pfaltz, Organometallics 2013, 32,
4702 – 4711; b) S. Gruber, A. Pfaltz, Angew. Chem. 2014, 126,
1927 – 1931; Angew. Chem. Int. Ed. 2014, 53, 1896 – 1900.
Keywords: asymmetric catalysis · hydrogenation · iridium ·
N,P ligands · nitriles
.
Angew. Chem. Int. Ed. 2014, 53, 8668 –8671
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8671