Catalytic 1,4-selective hydrosilylation of pyridines and benzannulated congeners

Article abstract of DOI:10.1002/anie.201305028

Radically different! The hydrosilylation of pyridines and quinolines is strictly 1,4-selective and likely involves an ionic one-step rather than the established radical two-step hydride transfer from a ruthenium(II) hydride complex onto the respective pyridinium and quinolinium ion intermediates (see scheme; Ar^F=3,5-(CF₃)₂C₆H ₃). Even 4-substituted substrates react highly regioselectively. Isoquinolines yield the 1,2-reduced heterocycles.

Full text of DOI:10.1002/anie.201305028

Angewandte
Chemie
DOI: 10.1002/anie.201305028
Partial Reduction
Catalytic 1,4-Selective Hydrosilylation of Pyridines and Benzannulated
Congeners**
C. David F. Kçnigs, Hendrik F. T. Klare, and Martin Oestreich*
The NAD(P)H/NAD(P)+ redox cycle with its 1,4-dihydro-
pyridine/pyridinium ion interconversion is one of the funda-
mental transformations in biological systems.[1] Outside living
organisms, the forward reaction, that is, reduction through
oxidation of the 1,4-dihydropyridine, is straightforward, and
the use of 1,4-dihydropyridines[2] as reducing agents in
organocatalysis is a prime example of that.[3] The back
reaction poses, however, a remarkable challenge, and there
is, to date, no general method available for the partial
reduction of pyridines.[4,5] The notion that breaking the
aromaticity of pyridines by a Birch-type reduction to provide
an entry to synthetically useful building blocks was realized
close to a century ago.[6] Birch and Karakhanov later
investigated these reductions with solvated electrons more
systematically with limited success,[7] and it was only recently
that Donohoe and co-workers demonstrated the potential of
this method for a few selected systems.[8] Another obvious
approach is the partial hydrogenation of pyridines with
dihydrogen under transition-metal catalysis,[9] but the prob-
lem of overreduction of the more reactive enamine inter-
mediate remains unsolved. Recently, the Hill and Suginome
groups independently introduced a noteworthy alternative
strategy that allows for partial reduction of pyridines, either
by magnesium(II)-[10] or rhodium(I)-catalyzed[11] hydrobora-
tion.[12] Prior to these seminal contributions, homogeneous
hydrosilylation of pyridines[13] had been probed by Harrod
and co-workers with a titanocene(III) catalyst but the
chemoselectivity was moderate.[14] Aside from this isolated
report, the unsolved challenge had lain dormant for another
decade until Nikonov and co-workers disclosed a pyridine
hydrosilylation using cationic ruthenium(II) complexes [Cp-
(iPr3P)Ru(MeCN)2]+ XÀ [X = PF6 or B(C6F5)4; Cp = Cyclo-
pentadienyl].[15] The scope of this catalysis was, in the end,
relatively narrow, but a few pyridines reacted 1,4-selectively
at room temperature, and that certainly was a major step
forward. Despite these recent significant advances in pyridine
hydroboration[10,11] and hydrosilylation,[14,15] the latter meth-
ods are still far from being general.
Our laboratory together with the group led by Ohki and
Tatsumi demonstrated that the tethered cationic rutheni-
um(II) complex 1[16] (Scheme 1, top) cooperatively activates
[17]
À
À
Si H bonds.
The polar Ru S bond in 1 mediates the
À
heterolytic splitting of Si H bonds to give a ruthenium(II)
Scheme 1. The tethered cationic ruthenium(II) complex 1 with a polar
À
À
Ru S bond and its application to Si H bond activation by s-bond
metathesis (Si=triorganosilyl; ArF =3,5-bis(trifluoromethyl)phenyl).
hydride and a sulfur-stabilized silicon cation by s-bond
metathesis (Scheme 1, bottom). The catalytically generated
silicon electrophile in combination with the ruthenium(II)
hydride opened the door to several remarkable dehydrogen-
ative coupling reactions[17a–c] and is also able to catalyze the
hydrodefluorination of CF3 groups by a unique mecha-
nism.[17d]
We, therefore, entertained the idea that intermediate II,
À
formed from I and a triorganosilane in the Si H bond-
activation step (step a in Scheme 2), could transfer the silicon
cation onto the nitrogen atom of III (step b in Scheme 2). The
pyridinium cation IV would then be susceptible to nucleo-
philic attack of the hydride released from V (step c in
[*] Dipl.-Chem. C. D. F. Kçnigs, Dr. H. F. T. Klare,
Prof. Dr. M. Oestreich
Institut fꢀr Chemie, Technische Universitꢁt Berlin
Strasse des 17. Juni 115, 10623 Berlin (Germany)
E-mail: martin.oestreich@tu-berlin.de
Homepage: http://www.organometallics.tu-berlin.de
[**] This research was supported by the Technische Universitꢁt Berlin.
M.O. is indebted to the Einstein Foundation (Berlin) for an endowed
professorship.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201305028.
Scheme 2. TentativeÀcatalytic cycle of the pyridine hydrosilylation
(counteranion BArF omitted for clarity).
4
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Scheme 2), affording VI (1,4-hydride transfer) and/or VII
(1,2-hydride transfer). The mechanisms of nucleophilic attack
at pyridinium and, likewise, quinolinium ions are still a matter
of debate.[18] As to the proposed hydride as nucleophile,
Norton and co-workers produced sound evidence that the 1,4-
reduction pathway of a ruthenium(II)-catalyzed hydrogena-
tion of an acyl pyridinium ion obeys a two-step radical
mechanism while 1,2-reduction occurs by a one-step ionic
mechanism.[9a] Also, Nikonovꢀs work included EPR measure-
ments that indicate the existence of a ruthenium-centered
paramagnetic intermediate.[15] We report here the application
of catalyst 1 to the regioselective hydrosilylation of pyridines
and its benzannulated congeners without any overreduction.
The new methodology tolerates challenging substitution
patterns and is shown to likely follow a different reaction
mechanism.
with NaBArF4 is not detrimental to catalytic turnover; in this
way it is not necessary to handle the oxygen-sensitive 16-
electron complex 1 in a glove-box (Table 1, entry 1, foot-
note [d]). Somewhat unexpectedly, MePh2SiH (4) was not as
reactive as in previous studies (Table 1, entry 2).[17] The
bulkier triorganosilanes 6 and 7 showed no conversion even
at elevated temperatures (Table 1, entries 4 and 5) but, again,
À
that is due to lack of reactivity in the Si H bond-activation
step.
The facile 1,4-hydrosilylation of pyridine compares well
with Nikonovꢀs finding[15] but we found our protocol to be
broadly applicable to 3-substituted pyridines 2b–2 f deco-
rated with either electron-donating or -withdrawing groups
(Table 2, entries 1–5). Even halides were tolerated (Table 2,
Table 2: Catalytic 1,4-selective hydrosilylation of pyridines.[a,b]
We began testing various triorganosilanes in the reduction
À
of the parent compound, pyridine (2a; Table 1). For the Si H
bond activation to occur at catalyst 1, these triorganosilanes
Table 1: Screening of suitable triorganosilanes as reducing agents.[a]
Entry
Substrate
Product
Yield
[%][c,d]
1
2
3
4
5
2b (R=Me)
8b
8c
8d
8e
8 f
84
98
96
76
76
2c (R=Ph)
2d (R=Br)
2e (R=Cl)
2 f (R=F)
À
Entry
Silane Si H
T
[8C]
t
[h]
Prod.
Yield [%][c]
1
2
3
4
5
Me2PhSiH (3)
MePh2SiH (4)
EtMe2SiH (5)
Et3SiH (6)
RT
45
RT
60
60
7
14
7
14
14
8a
9a
10a
11a
12a
94[d]
96
6
7
8
9
10
11
2g (R=Me)
2h (R=CF3)
2i (R=Et)
2j (R=iPr)
2k (R=Ph)
2l (R=Cl)
8g
8h
8i
8j
8k
8l
80
88
84
89[e]
75[f]
13[f]
[e]
–
[e]
Ph3SiH (7)
–
[a] All reactions were performed according to the General Procedure 1
(see the Supporting Information for details). [b] Conversion was
monitored by 1H NMR spectroscopy. [c] Yield of isolated product after
the catalyst had been removed by filtration through a short plug of
deactivated silica gel. [d] 91% yield with formation of catalyst 1 in situ
according to the General Procedure 2 (see the Supporting Information
for details). [e] No reaction.
[g]
–
12
13
14
2m (R=Me)
2n (R=Cl)
2o (R=F)
8m
8n
8o
80[h]
98
84
[a]–[c] See Table 1. [d] Formation of trace amounts of (Me2PhSi)2O as
a result of incomplete conversion due to hydrolysis of the remaining
(activated) silane. [e] Conversion, 15% (Me2PhSi)2O as contamination.
[f] Conversion. [g] Formation of 1,4-dihydropyridine 8a (R=H) along
with (Me2PhSi)2O was observed. [h] 80% conversion after 12 h at 458C.
must fit into the pocket that the bulky thiolate ligand and the
ruthenium fragment create around the Ru S bond. From our
À
previous work, we already knew that the steric situation is
well-balanced with Me2PhSiH (3), MePh2SiH (4), and
EtMe2SiH (5) but not with Et3SiH (6) and Ph3SiH (7).[17]
We were then delighted to see that triorganosilanes 3–5
reacted with equimolar amounts of 2a in the presence of just
1.0 mol% of preformed ruthenium(II) complex 1, affording
1,4-dihydropyridines 8a–10a as the sole regioisomers without
overreduction (Table 1, entries 1–3). The reactions using
silanes 3 and 5 proceeded smoothly at room temperature
without the need of a solvent, and catalyst loadings as low as
0.1 mol% still promoted the hydrosilylation of 2a in 86%
yield, yet at a prolonged reaction time of 30 h. Notably, in situ
formation of the coordinatively unsaturated catalyst 1 from
the corresponding air-stable chloride complex by treatment
entries 3–5), and no debromination was observed (cf.
Ref. [15]). We note here that Lewis basic functional groups,
e.g., carboxyl and cyano groups, were not compatible with
catalyst 1 and the silicon electrophile. Remarkably, substitu-
tion para to the nitrogen atom neither thwarted the hydro-
silylation nor steered the regioselectivity from 1,4- to 1,2-
reduction. The 4-substituted pyridines 2g and 2h reacted
smoothly; 2i and 2j were too sterically demanding but were
still selectively transformed into the 4-substituted 1,4-dihy-
dropyridines (Table 2, entries 6–9). Phenyl substitution at C4
as in 2k nearly fully disrupted the hydrosilylation (Table 2,
entry 10).
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
In an independent experiment using deuterated silane
[2H]-3 and pyridine 2g, deuterium incorporation was found
exclusively in the 4-position (see the Supporting Information
for details). Dehalogenation was seen for the substrate with
a chloro substituent at C4 (Table 2, entry 11). Pyridines 2m–
2o with 3,5-substitution also reacted cleanly (Table 2,
entries 12–14). Only pyridines with ortho substitution would
not participate in the catalysis. We attribute this to sterics
hampering the transfer of the silicon cation from the sulfur-
stabilized silylium ion to the pyridine nitrogen atom (cf. step b
in Scheme 2).
Nevertheless, we subjected quinolines/acridine as well as
isoquinolines/phenanthridine to our method (Table 3). To our
surprise, quinolines 13a and 13b with and without a bromine
atom at C6 and even 2-phenyl-substituted quinoline 13c were
selectively reduced in 1,4-fashion (Table 3, entries 1–3). These
results, in terms of both regio- and chemoselectivity, were
highly unexpected in the light of the literature prece-
dence.[19,20] Acridine (14) reacted equally well (Table 3,
entry 4). With isoquinolines 15a–15d, 1,4-reduction would
require breaking the aromaticity of the annulated benzene,
and these heteroarenes including phenanthridine (16) yielded
the 1,2-reduced heterocycles instead (Table 3, entries 5–9).
Reactions of the more hindered 1- and 3-methyl-substituted
15c and 15d did not go to completion but, again, these are
particularly challenging substrates. Partial reduction of iso-
quinolines is difficult, and overreduction is usually
observed.[19]
The finding that 4-alkyl-substituted pyridines underwent
hydrosilylation with exclusive 1,4-selectivity suggested the
opportunity to distinguish between a two-step radical[9a,15] and
a one-step ionic 1,4-hydride transfer by “radical clocks”[21]
installed at C4. We chose the potential ring opening of the
cyclopropylmethyl radical (as in 2p) and the potential
spirocyclization of the hex-5-enyl radical (as in 2q) as
mechanistic probes (Figure 1). Both 2p and 2q reacted
Table 3: Catalytic 1,4-selective hydrosilylation of quinolines/acridine
(top) and 1,2-selective hydrosilylation of isoquinolines/phenanthridine
(bottom).[a,b,d]
Figure 1. Potential “radical clocks” to probe the 1,4-hydride transfer:
one-step ionic mechanism versus two-step radical mechanism.
Entry
Substrate
Product
Yield
[%][c]
1
2
3
97
94
92
slowly under the standard conditions (cf. Table 2) yet were
transformed cleanly into the corresponding 1,4-dihydropyr-
idines 8p and 8q (not shown). We found no spectroscopic
evidence for either the ring opening or the spirocyclization
(see the Supporting Information for details). These results
are, however, not totally conclusive as the hypothetical radical
intermediate is tertiary and resonance-stabilized, and rate
constants are not available for such systems but expected to
be rather low. The fact that the cyclopropyl group remains
intact, admittedly at low conversion, nevertheless indicates to
us that an ionic mechanism is operative.
4
5
97[e]
97
The present method provides a viable tool for the chemo-
and regioselective hydrosilylation of various pyridines and
related nitrogen-containing heterocycles. Using equimolar
amounts of substrate and silane at low catalyst loading, the
reactions are exceptionally clean (aside from (Me2PhSi)2O
contamination at incomplete silane consumption in a few
cases) and do not require complicated purification of partially
saturated heterocycles susceptible to oxidation. The pro-
nounced 1,4-selectivity in the hydrosilylation of pyridines and
quinolines is likely achieved in an ionic one-step hydride
transfer onto the pyridinium/quinolinium ion intermediate,
and that distinguishes the present work from previous reports
of a radical mechanism.
6
91
7
8
71[f]
75[g]
9
98
[a]–[d] See Table 2. [e] Reaction was performed at 458C. [f] Conversion.
Received: June 11, 2013
[g] Conversion, 27% (Me2PhSi)2O as contamination.
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[10] M. Arrowsmith, M. S. Hill, T. Hadlington, G. Kociok-Kçhn, C.
Weetman, Organometallics 2011, 30, 5556 – 5559.
[11] K. Oshima, T. Ohmura, M. Suginome, J. Am. Chem. Soc. 2012,
134, 3699 – 3702.
Keywords: homogeneous catalysis · hydrosilylation ·
.
À
N-heterocycles · regioselectivity · Si H bond activation
[12] Suginome and co-workers also reported a palladium(II)-cata-
lyzed silaboration of pyridines, affording 1,4- or 1,2-dihydropyr-
idines depending on the substitution pattern: K. Oshima, T.
Ohmura, M. Suginome, J. Am. Chem. Soc. 2011, 133, 7324 – 7327.
[13] For the heterogeneous hydrosilylation of pyridines, see: a) N. C.
Cook, J. E. Lyons, J. Am. Chem. Soc. 1965, 87, 3283 – 3284;
b) N. C. Cook, J. E. Lyons, J. Am. Chem. Soc. 1966, 88, 3396 –
3403.
[14] a) L. Hao, J. F. Harrod, A.-M. Lebuis, Y. Mu, R. Shu, E. Samuel,
H.-G. Woo, Angew. Chem. 1998, 110, 3314 – 3318; Angew. Chem.
Int. Ed. 1998, 37, 3126 – 3129; b) J. F. Harrod, R. Shu, H.-G. Woo,
E. Samuel, Can. J. Chem. 2001, 79, 1075 – 1085.
[15] a) D. V. Gutsulyak, A. van der Est, G. I. Nikonov, Angew. Chem.
2011, 123, 1420 – 1423; Angew. Chem. Int. Ed. 2011, 50, 1384 –
1387; for a highlight article, see: b) K. Osakada, Angew. Chem.
2011, 123, 3929 – 3930; Angew. Chem. Int. Ed. 2011, 50, 3845 –
3846.
[16] Y. Ohki, Y. Takikawa, H. Sadohara, C. Kesenheimer, B.
Engendahl, E. Kapatina, K. Tatsumi, Chem. Asian J. 2008, 3,
1625 – 1635.
[1] N. Pollak, C. Dçlle, M. Ziegler, Biochem. J. 2007, 402, 205 – 218.
[2] For reviews on the chemistry of dihydropyridines, see: a) U.
Eisner, J. Kuthan, Chem. Rev. 1972, 72, 1 – 42; b) D. M. Stout,
A. I. Meyers, Chem. Rev. 1982, 82, 223 – 243; c) R. Lavilla, J.
Chem. Soc. Perkin Trans. 1 2002, 1141 – 1156.
[3] For a review, see: a) S. G. Ouellet, A. M. Walji, D. W. C.
MacMillan, Acc. Chem. Res. 2007, 40, 1327 – 1339; for seminal
original contributions, see: b) M. Rueping, E. Sugiono, C. Azap,
T. Theissmann, M. Bolte, Org. Lett. 2005, 7, 3781 – 3783; c) S.
Hoffmann, A. M. Seayad, B. List, Angew. Chem. 2005, 117,
7590 – 7593; Angew. Chem. Int. Ed. 2005, 44, 7424 – 7427; d) R. I.
Storer, D. E. Carrera, Y. Ni, D. W. C. MacMillan, J. Am. Chem.
Soc. 2006, 128, 84 – 86.
[4] a) J. G. Keay, Adv. Heterocycl. Chem. 1986, 39, 1 – 77; b) J. G.
Keay in Comprehensive Organic Synthesis, Vol. 8 (Eds.: B. M.
Trost, I. Fleming), Pergamon, Oxford, 1991, pp. 579 – 602.
[5] For a review discussing unconventional ways to regenerate 1,4-
dihydropyridines from pyridinium ions, see: F. Hollmann,
I. W. C. E. Arends, K. Buehler, ChemCatChem 2010, 2, 762 –
782.
[6] a) B. D. Shaw, J. Chem. Soc. 1925, 215 – 216; b) B. D. Shaw, J.
Chem. Soc. 1937, 300 – 302; c) A. J. Birch, J. Chem. Soc. 1947,
1270; d) S. Danishefsky, A. Nagel, J. Chem. Soc. Chem.
Commun. 1972, 373.
[17] a) H. F. T. Klare, M. Oestreich, J.-i. Ito, H. Nishiyama, Y. Ohki,
K. Tatsumi, J. Am. Chem. Soc. 2011, 133, 3312 – 3315; b) C. D. F.
Kçnigs, H. F. T. Klare, Y. Ohki, K. Tatsumi, M. Oestreich, Org.
Lett. 2012, 14, 2842 – 2845; c) C. D. F. Kçnigs, M. F. Mꢁller, N.
Aiguabella, H. F. T. Klare, M. Oestreich, Chem. Commun. 2013,
49, 1506 – 1508; d) T. Stahl, H. F. T. Klare, M. Oestreich, J. Am.
Chem. Soc. 2013, 135, 1248 – 1251.
[7] A. J. Birch, E. A. Karakhanov, J. Chem. Soc. Chem. Commun.
1975, 480 – 481.
[8] a) T. J. Donohoe, A. J. McRiner, P. Sheldrake, Org. Lett. 2000, 2,
3861 – 3863; b) T. J. Donohoe, A. J. McRiner, M. Helliwell, P.
Sheldrake, J. Chem. Soc. Perkin Trans. 1 2001, 1435 – 1445.
[9] a) A. P. Shaw, B. L. Ryland, M. J. Franklin, J. R. Norton, J. Y.-C.
Chen, M. Lynn Hall, J. Org. Chem. 2008, 73, 9668 – 9674 (one
example of a ruthenium(II)-catalyzed 1,4-reduction of an acyl
pyridinium ion, 58% conv.); b) Q.-A. Chen, M.-W. Chen, C.-B.
Yu, L. Shi, D.-S. Wang, Y. Yang, Y.-G. Zhou, J. Am. Chem. Soc.
2011, 133, 16432 – 16435 (one example of a ruthenium(II)-
catalyzed 1,4-reduction of a Hantzsch ester, 69% conv.).
[18] For a review addressing this issue, see: I. S. Poddubnyi, Khim.
Geterotsikl. Soedin. 1995, 774 – 815 (Chem. Heterocycl. Compd.
1995, 31, 682 – 714), and references therein.
[19] A. M. Voutchkova, D. Gnanamgari, C. E. Jakobsche, C. Butler,
S. J. Miller, J. Parr, R. H. Crabtree, J. Organomet. Chem. 2008,
693, 1815 – 1821.
[20] S. J. Geier, P. A. Chase, D. W. Stephan, Chem. Commun. 2010,
46, 4884 – 4886.
[21] a) D. Griller, K. U. Ingold, Acc. Chem. Res. 1980, 13, 317 – 323;
b) M. Newcomb, Tetrahedron 1993, 49, 1151 – 1176.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Partial Reduction
Radically different! The hydrosilylation of
pyridines and quinolines is strictly 1,4-
selective and likely involves an ionic one-
step rather than the established radical
two-step hydride transfer from a ruthe-
nium(II) hydride complex onto the
respective pyridinium and quinolinium
ion intermediates (see scheme; ArF =3,5-
(CF3)2C6H3). Even 4-substituted sub-
strates react highly regioselectively. Iso-
quinolines yield the 1,2-reduced hetero-
cycles.
C. D. F. Kçnigs, H. F. T. Klare,
M. Oestreich*
&&&& — &&&&
Catalytic 1,4-Selective Hydrosilylation of
Pyridines and Benzannulated Congeners
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!