A Functional-Group-Tolerant Catalytic trans Hydrogenation of Alkynes

Article abstract of DOI:10.1002/anie.201205946

Against the rules: During the hundred years following Sabatier's groundbreaking work on catalytic hydrogenation, syn delivery of the H atoms to the π system of a substrate remained the governing stereochemical rule. An exception has now be found with the

Full text of DOI:10.1002/anie.201205946

Angewandte
Chemie
DOI: 10.1002/anie.201205946
Hydrogenation
A Functional-Group-Tolerant Catalytic trans Hydrogenation of
Alkynes**
Karin Radkowski, Basker Sundararaju, and Alois Fꢀrstner*
The catalytic semihydrogenation of alkynes to Z alkenes is
widely practiced textbook knowledge.[1] Amongst the various
heterogeneous or homogenous catalysts capable of effecting
this transformation, the use of palladium on CaCO3 doped
with lead is particularly popular, and is commonly called
Lindlar catalyst after its discoverer;[2–4] variants using BaSO4
as the support material and quinoline as additive are also
popular. The major role of the catalyst poisons is to prevent
over-reduction of the alkene primarily formed,[5] whereas the
Z selectivity is an intrinsic feature: transition metal catalysts,
be they homogenous or heterogeneous, usually deliver the
two hydrogen atoms of H2 suprafacially to the p system of the
substrate by a sequence of hydrometalation/reductive elim-
ination as the elementary step (although the exact processes
on a metallic surface, where multimetallic sites may be
engaged, are still not understood in full detail).[1,5,6] The small
amounts of isomeric by-products that typically accompany the
Z alkenes are thought to derive from secondary processes.
Because of this well-established stereochemical course,
the formation of E alkenes by catalytic hydrogenation is
inherently difficult and no broadly applicable protocol is
known to date.[7,8] In fact, all commonly practiced methods for
the direct conversion of alkynes to E alkenes are stoichio-
metric in nature, with Birch-type reductions using dissolving
metals being the classical incarnation. The use of alkali metals
in liquid ammonia or amines, however, obviously prevents
applications to polyfunctionalized, base-labile and/or sensi-
tive substrates.[9,10] The arguably best catalytic alternative is
a two-step protocol, in which the alkyne is first subjected to
a ruthenium-catalyzed trans hydrosilylation[11] followed by
gentle protodesilylation of the resulting alkenylsilanes with
stoichiometric amounts of a suitable fluoride source.[12]
Although this indirect approach is largely superior to the
Birch reduction in terms of functional group compatibility
and has served our program well in the past,[13,14] we sought to
develop a more direct solution. Outlined herein are our
preliminary results on a rather unique ruthenium-catalyzed
trans-selective alkyne hydrogenation. As this method is
stereocomplementary to the classical Lindlar reduction and
tolerates a host of reducible functional groups, we believe that
it holds great promise for future applications.
The trans hydrosilylation alluded to above uses [Cp*Ru-
(MeCN)3]PF6 (1; Cp* = pentamethylcyclopentadienyl; see
Scheme 1) as precatalyst.[11,12] With the isolobal relationship
between R3Si and H in mind, one might envisage that this or
similar complexes could possibly effect trans-hydrogenation
reactions as well. In a pioneering NMR study, Bargon and co-
workers indeed demonstrated that the related cationic sorbic
4
+
=
ꢀ
=
acid complex [Cp*Ru(h -CH3CH CH CH CHCOOH)]
OTfꢀ (2) catalyzes the trans reduction of substrates such as
2-pentyne or 1-phenyl-1-propyne in [D4]methanol under
a hydrogen atmosphere (1 bar); terminal acetylenes were
found to be unreactive.[15] Parahydrogen-induced polarization
(PHIP) experiments demonstrated that the E alkene forma-
tion was intrinsic and not caused by posterior Z!E isomer-
ization. The PHIP spectra also implied a pair-wise delivery of
the hydrogen atoms of H2. To account for these results, the
authors proposed a mechanism involving m-bridged dinuclear
complexes, but emphasized the somewhat speculative nature
of their proposal.[15] This suggestion was also based on earlier
evidence from the rhodium series.[7]
Free alcohols, a diethyl acetal, and a conjugated ketone
were found to be compatible with this method, although no
yields of isolated products were reported.[15,16] Unfortunately,
however, our attempts to translate this methodology to the
semihydrogenation of the lactonic cycloalkyne 5, which serves
as a model compound in our ongoing studies on the develop-
ment of ever more effective catalysts for alkyne metathesis,[17]
only met with limited success (Scheme 1).
As can be seen from the selected data shown in Table 1,
the hydrogenation of alkyne 5 in the presence of complexes
1 or 2 proceeded with appreciable E selectivity. However, the
sorbic acid complex 2 failed to bring about full conversion
even at a loading of 25 mol% and the reaction was not very
clean either (Table 1, entry 1). Although the commercial
acetonitrile adduct 1 was more effective,[16] the mass recovery
was poor in most cases (Table 1, entries 3–5) and the E/Z ratio
provided room for improvement. Careful inspection of the
crude mixtures indicated substantial oligomerization by
transesterification before and/or after the semireduction.
This competing pathway is attributed to an effective activa-
tion of the carboxyl groups by the evidently fairly Lewis acidic
ruthenium species derived from 1 under the reaction con-
ditions. This side reaction was more pronounced in MeOH
than in CH2Cl2, which was therefore chosen as the solvent for
further optimization.[18]
[*] K. Radkowski, Dr. B. Sundararaju, Prof. A. Fꢀrstner
Max-Planck-Institut fꢀr Kohlenforschung
45470 Mꢀlheim/Ruhr (Germany)
E-mail: fuerstner@kofo.mpg.de
[**] Generous financial support by the Max-Planck-Gesellschaft, the
Fonds der Chemischen Industrie, and the Alexander von Humboldt-
Stiftung (fellowship to B.S.) is gratefully acknowledged. We thank
the analytical departments of our Institute for excellent assistance.
Another important piece of information came from
experiments with the methoxide-bridged dimer 3. On treat-
ment with TfOH (Tf = trifluoromethanesulfonyl) in CH2Cl2,
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201205946.
Angew. Chem. Int. Ed. 2012, 51, 1 – 7
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
productivity, while assisting the trans-selective course of the
reduction.
As a consequence, we focused our efforts on the use of
commercial [Cp*Ru(cod)Cl] (4) as a defined precatalyst. At
a loading of 5 mol% and a hydrogen pressure of 10 bar,[21] this
complex cleanly effected the reduction of 5 to 6, to afford an
excellent E/Z ratio of no less than 98:2 (Table 1, entry 9).
Some over-reduction was noticed, but the alkane by-product
could be partly removed by flash chromatography. Impor-
tantly and in striking contrast to the experiments with 1 or 2,
the competing oligomerization was largely suppressed and the
mass recovery was therefore excellent (89% combined yield
of alkene and alkane). The reaction could be further
improved upon ionization of 4 with AgOTf[22] prior to the
actual hydrogenation step. This allowed the reaction time to
be reduced, without compromising the yield and selectivity
(Table 1, entries 10–11).[23] A control experiment confirmed
that the E selectivity is inherent to the reduction and not the
result of rapid Z!E isomerization.[24]
Scheme 1. Model reaction for the investigation of the trans-selective
alkyne hydrogenation effected by [Cp*Ru]-based precatalysts; for the
results, see Table 1.
This favorable outcome encouraged us to investigate the
scope of the method in more detail. As can be seen from the
results compiled in Table 2, a mixture of 4 (5.5 mol%) and
AgOTf (5 mol%) in CH2Cl2 allowed an assortment of
substrates to be reduced with excellent levels of E selectivity.
Importantly, the yields of the isolated products were high in
all cases investigated and the method was compatible with
a variety of polar substituents, including esters, amides, free
carboxylic acids, ketones,[18) a primary alcohol, and methyl
and silyl ethers, as well as an elimination-prone primary
tosylate. Moreover, reducible sites such as a nitro group, an
Table 1: Optimization of the trans-selective alkyne hydrogenation shown
in Scheme 1.[a]
Entry [Ru] mol% Additive Solvent
(mol%)
t
[h]
E/Z[b] Yield By-
[%][c] products[d]
[%]
1
2
3
4
5
6
7
2[e,h] 25
1[e,h] 25
MeOH 0.5 87:13 55[f]
MeOH 0.5 96:4 35
5
<2
1
1
10
5
MeOH
MeOH
CH2Cl2
1
1
1
94:6 45
94:6 48
88:12 55
2
12[g]
1[h] 25
11[g]
19[g]
1
1
3
2
5
CH2Cl2 0.5 85:15 86
CH2Cl2 58:42 n.d.
ꢀ
alkyl bromide, the N O bond of a Weinreb amide, an
TfOH
1
aromatic nitrile, and a terminal alkene remained intact. Even
a thioether moiety in vicinity to the alkyne did not poison the
catalyst; actually, the formation of the sulfide shown in
entry 11 (Table 2) worked particularly well, as virtually no
over-reduction was observed in this case but the high
E selectivity remained uncompromised. In contrast, sub-
strates containing a 1,3-diene or a 1,3-enyne unit could not
be hydrogenated (Table 2, entries 19 and 20), most likely
because of the tight coordination of such conjugated p sy-
stems to the active ruthenium species formed in situ (compare
the sorbic acid complex 2 and complex 11 shown in
Scheme 3); likewise and in analogy to the report by Bargon
and co-workers,[15] the attempted reduction of a terminal
alkyne did not work well. In all cases investigated, the
E configuration of the major product was assigned by NMR
spectroscopy and confirmed by direct comparison with the
corresponding Z alkene made by Lindlar reduction. It is
informative to look at the spectra (see the Supporting
Information), which show that the amount of over-reduced
and/or isomeric by-products present in the crude mixtures is
not higher than that obtained in the classical Lindlar
reductions using a commercial Pd(Pb)/CaCO3 sample. The
over-reduction of the substrate was typically in the range of 5–
15%; for reasons that are not entirely clear, however, it was
more significant in the reduction of tolane (Table 2, entry 12),
a propargylic silyl ether (Table 2, entry 7), and the strained
cycloalkynes that form the products shown in entries 14 and
18 (Table 2). Despite this minor drawback, the new method
(10)
TfOH/
COD (10)
8
3
5
CH2Cl2 2.5 82:18 82
8
9
10
4
4
5
5.5
CH2Cl2
1
98:2 89
15
AgOTf
(5)
AgOTf
(2)
CH2Cl2 0.5 98:2 89– 7–10[i]
96[i]
11
4
2.2
CH2Cl2 0.5 98:2 89
15
[a] Unless stated otherwise, all reactions were performed under H2
(10 bar) at ambient temperature. [b] Ratio observed in the crude product
prior to work-up. Determined by GC analysis. [c] Combined yield of the
isolated alkenes and alkane. [d] Sum of alkane and isomeric alkenes
formed by double bond migration, which co-elude as a single peak.
Determined by GC analysis. [e] At 558C. [f] Yield determined by GC
analysis; the reaction did not proceed further when stirred for additional
2 h. [g] GC–MS results suggest that the by-products largely consisted of
isomeric alkenes formed by double-bond migration. [h] At 1 bar H2
pressure. [i] Range observed in several independent reactions on
different scale (0.2–1.3 mmol); n.d.=not determined.
this complex releases unligated [Cp*Ru]+ into the solution.[19]
Whereas this free species on its own performed poorly
(Table 1, entry 7), the selectivity was largely restored and the
product yield good when 1,5-cyclooctadiene (COD) was
present (Table 1, entry 8).[20] These results suggested that
a chelating diene ligates the active species without interrupt-
ing the hydrogenation; in so doing, it tempers the Lewis
acidity of the Cp*RuII entity and hence ensures high
2
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 7
These are not the final page numbers!

Angewandte
Chemie
Table 2: trans-Selective reduction of internal alkynes.[a]
hydrogenation protocol for alkynes. As such, it is strictly
E/Z[b]
Yield [%][c]
96
stereocomplementary—but otherwise very well compara-
ble—to the classical repertoire.
Entry
1
Major Product
t [h]
0.5
98:2
The unorthodox stereochemical course of this hydro-
genation raises important questions as to the actual mecha-
nism. Although a final answer cannot be given at this point,
several findings bear relevant information. First, it must be
noticed that a solution of 4 and AgOTf in CD2Cl2, when kept
in the absence of the alkyne substrate under H2 atmosphere
(10 bar) in an NMR pressure tube for 1–5 h, gives rise to
a mixture containing several discrete species. Three distinct
signals in the high-field region (dH (CD2Cl2): ꢀ4.96, ꢀ8.02,
ꢀ13.42 ppm) indicate that more than one ruthenium hydride
complex is formed. Some reasonable candidates are shown in
Scheme 2; however, they could not be unambiguously iden-
2
3
0.5
4
95:5
96:4
88
66[d]
4
2.5
93:7
95
5
6
0.5
0.5
95:5
97:3
87
60[d]
7
4
93:7
96 (33)[c,d]
8
9
0.5
2.5
87:13
96:4
95 (21)[c]
86
10
11
12
0.5
1.5
21
97:3
97:3
95:5
67
Scheme 2. Known ruthenium hydride (or hydrogen) complexes that
might possibly form in situ from 4/AgOTf under the conditions of the
catalytic hydrogenation.
88[e]
82 (34)[c]
tified by comparison with literature data, because the
reported spectra were recorded in a different solvent or at
a significantly different temperature. Therefore we decided to
prepare these complexes by unambiguous routes and indi-
vidually test their catalytic performance.
13
0.5
98:2
89[f]
14
15
0.5
96:4
97:3
64 (27)[c,d]
81[d]
16[g]
The conspicuous signal below ꢀ13 ppm suggested that the
dimeric dihydride species 7 might be generated from the
precatalyst, if the COD ligand is hydrogenated off (or
replaced by solvent prior to reaction with H2).[25] Recalling
that Bargon and co-workers had proposed a bimetallic
mechanism,[15] this species seemed a particularly hot candi-
date. Therefore complex 7 was prepared according to
a literature route[26] but it was found to afford only a dis-
appointingly low selectivity (E/Z = 55:45) when used as
catalyst for the reduction of 5 to 6 under standard conditions
(10 bar H2, CH2Cl2, RT).
Likewise, our control experiments exclude that the
monohydride complex 8[27] is the operative E-selective
catalyst. This species, for which a convenient new synthesis
route was found using sodium formate as the hydride source
(Scheme 3), resulted only in poor conversion and an unfav-
orable E/Z ratio of 45:55.
Should any complex 8 be formed from [Cp*Ru(cod]OTf
and H2, however, one equivalent of TfOH must also be
generated. Therefore we checked whether the acid plays any
role in the catalytic process. To this end, preformed hydride 8
was protonated with TfOH in CD2Cl2 at ꢀ788C to give the
known cationic dihydrogen complex 9 (Scheme 3).[28] This
species is only stable below ꢀ408C, at which temperature it
did not effect any noticeable hydrogenation of the model
substrate 5. In contrast, when a mixture containing 9
16
17
3
92:8
91:9
80
85
34[g]
18
1[g]
97:3
77 (26)[c,d]
19
20
n.r.
n.r.
[a] Unless stated otherwise, the reactions were performed with 4
(5.5 mol%) and AgOTf (5 mol%) in CH2Cl2 at RTunder H2 (10 bar initial
pressure); for representative procedures, see the Supporting Informa-
tion. [b] Determined by GC analysis. [c] Combined yield of alkenes and
the alkane (5–15%, as determined by GC analysis); if the amount of
alkane by-product (and/or isomeric alkenes formed by double-bond
migration) exceeds 15%, the measured GC% are indicated in brackets.
[d] Alkene isomers formed by double-bond migration were detected.
[e] No alkane was detected. [f] 1.3 mmol scale, see the Supporting
Information. [g] Using the neutral complex 4 without AgOTf; n.r.=no
reaction.
constitutes the first practical, efficient, functional-group-
tolerant, broadly applicable, and highly E-selective semi-
Angew. Chem. Int. Ed. 2012, 51, 1 – 7
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
toluene had been found unsuitable as (co)solvent in the
present hydrogenation.[18]
In any case, the method described above seems to be the
first practical and truly functional-group-tolerant procedure
for the catalytic semihydrogentaion of alkynes to E alkenes,
which is stereocomplementary to the classical Lindlar reduc-
tion or its synthetic equivalents. The active catalyst is
generated in situ from [Cp*RuCl(cod)] (4) and AgOTf
under hydrogen pressure. Although no detailed picture of
the operative mechanism can yet be drawn, the available data
suggest that neither the mononuclear nor the dinuclear
hydride complexes, [Cp*RuH(cod)] and [Cp*Ru(m-
H)4RuCp*], respectively, account for the observed results.
Rather, the cationic dihydrogen complex [Cp*Ru(H2)-
(cod)]OTf (9) and/or products derived from this labile
entity seem to play a role. We intend to scrutinize this
notion, with the hope of identifying ligand frameworks that
impart more stability onto the active species and hence better
lend themselves to mechanistic investigations.
Scheme 3. Fate of the cationic [Cp*Ru(cod)] fragment.
Received: July 25, 2012
Published online: && &&, &&&&
Keywords: alkenes · alkynes · hydrogenation · ruthenium ·
trans reduction
(10 mol%) and alkyne 5 was allowed to reach ambient
temperature under H2 (10 bar), product 6 was quantitatively
formed with excellent selectivity (E/Z = 95:5).[29] This result
corresponds closely to the outcome under the standard
conditions used in the preparative experiments (Table 1,
entry 10), thus suggesting that either the cationic dihydrogen
complex 9 itself is in the catalytic cycle and effects the
trans hydrogenation before it decomposes, or that this com-
plex first transforms into another species, which then serves as
the actual catalyst.[30]
.
[1] a) A. M. Kluwer, C. J. Elsevier in Handbook of Homogeneous
Hydrogenation, Vol. 1 (Eds.: J. G. de Vries, C. J. Elsevier),
Wiley-VCH, Weinheim, 2007, pp. 375 – 411; b) H.-U. Blaser, A.
Schnyder, H. Steiner, F. Rçssler, P. Baumeister in Handbook of
Heterogeneous Catalysis, Vol. 7, 2nd ed. (Eds.: G. Ertl, H.
Knçzinger, F. Schꢀth, J. Weitkamp), Wiley-VCH, Weinheim,
2008, pp. 3284 – 3308; c) S. Siegel in Comprehensive Organic
Synthesis Vol. 8 (Eds.: B. M. Trost, I. Fleming), Pergamon,
Oxford, 1991, pp. 417 – 442; d) H. Takaya, R. Noyori in Com-
prehensive Organic Synthesis, Vol. 8 (Eds.: B. M. Trost, I.
Fleming), Pergamon, Oxford, 1991, pp. 443 – 469; e) P. N.
Rylander, Hydrogenation Methods, Academic Press, London,
1985; f) ꢁ. Molnꢂr, A. Sꢂrkꢂny, M. Varga, J. Mol. Catal. A 2001,
173, 185 – 221; g) E. N. Marvell, T. Li, Synthesis 1973, 457 – 468.
[2] a) H. Lindlar, Helv. Chim. Acta 1952, 35, 446 – 450; b) H.
Lindlar, R. Dubuis, Org. Synth. 1966, 46, 89 – 91.
[3] a) D. J. Cram, N. L. Allinger, J. Am. Chem. Soc. 1956, 78, 2518 –
2524; b) J.-J. Brunet, P. Caubere, J. Org. Chem. 1984, 49, 4058 –
4060; c) J. Rajaram, A. P. S. Narula, H. P. S. Chawla, S. Dev,
Tetrahedron 1983, 39, 2315 – 2322; d) K. R. Campos, D. Cai, M.
Journet, J. J. Kowal, R. D. Larson, P. J. Reider, J. Org. Chem.
2001, 66, 3634 – 3635; e) M. Gruttadauria, L. F. Liotta, R. Noto,
G. Deganello, Tetrahedron Lett. 2001, 42, 2015 – 2017; f) Y. Lee,
Y. Motoyama, K. Tsuji, S.-H. Yoon, I. Mochida, H. Nagashima,
ChemCatChem 2012, 4, 778 – 781.
[4] For other functional-group-tolerant Z-selective semihydrogena-
tion methods, see the following for leading references and
literature cited therein: a) R. R. Schrock, J. A. Osborn, J. Am.
Chem. Soc. 1976, 98, 2143 – 2147; b) M. W. van Laren, C. J.
Elsevier, Angew. Chem. 1999, 111, 3926 – 3929; Angew. Chem.
Int. Ed. 1999, 38, 3715 – 3717; c) M. Sodeoka, M. Shibasaki, J.
Org. Chem. 1985, 50, 1147 – 1149; d) C. A. Brown, V. K. Ahuja, J.
Org. Chem. 1973, 38, 2226 – 2230; e) P. Hauwert, G. Maestri,
J. W. Sprengers, M. Catellani, C. J. Elsevier, Angew. Chem. 2008,
120, 3267 – 3270; Angew. Chem. Int. Ed. 2008, 47, 3223 – 3226;
f) H. S. La Pierre, J. Arnold, F. D. Toste, Angew. Chem. 2011,
123, 3986 – 3989; Angew. Chem. Int. Ed. 2011, 50, 3900 – 3903;
To check this latter possibility, the fate of complex 9a (X =
OTf) was further investigated (Scheme 3). It has previously
been reported that its sister compound 9b (X = BF4) converts
into the 1,3,5-cyclooctatriene complex 10b (X = BF4).[28]
Surprisingly though, we observed a different outcome: thus,
9a was found to afford complex 11a as the only product
detectable by NMR spectroscopy, when a solution in CD2Cl2
was allowed to warm from ꢀ788C to ambient temperature.
This remarkable transformation, during which the COD ring
transforms into an acyclic h4,h2-1,3,7-octatriene ligand by
[31,32]
ꢀ
C C bond cleavage,
was also observed upon stirring of
the cationic species generated from [Cp*RuCl(cod)] (4) and
AgOTf for 4 h in CH2Cl2 under Ar. Moreover, we confirmed
that it is not the escorting counterion that accounts for this
unusual result, as ionization of 4 with AgBF4 furnished the
analogous complex 11b (X = BF4) together with the arene
complex 12b derived thereof.[33]
Because the complexes 11 and 12 form on a timescale
similar to that of the actual hydrogenation reaction, we tested
whether they exert any catalytic activity. In fact, the use of the
isolated complex 11a (5 mol%) resulted in an E-selective
hydrogenation of the model substrate 5, albeit the E/Z ratio
(86:14) was somewhat less favorable than that obtained with
the in situ mixture 4/AgOTf (Table 1, entry 10). In contrast,
the arene adduct 12b was inactive, which likely explains why
4
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 7
These are not the final page numbers!

Angewandte
Chemie
g) B. M. Choudary, G. V. M. Sharma, P. Bharathi, Angew. Chem.
1989, 101, 506 – 507; Angew. Chem. Int. Ed. Engl. 1989, 28, 465 –
466.
Stade, A. Kondoh, G. Seidel, R. Goddard, A. Fꢀrstner, Chem.
Eur. J. 2012, 18, 10281 – 10299; c) A. Fꢀrstner, C. Mathes, C. W.
Lehmann, Chem. Eur. J. 2001, 7, 5299 – 5317.
[5] For leading studies on the structure of Lindlar catalysts and the
effect of the additives, see: a) A. B. McEwen, M. J. Guttieri,
W. F. Maier, R. M. Laine, Y. Shvo, J. Org. Chem. 1983, 48, 4436 –
4438; b) R. Schlçgl, K. Noack, H. Zbinden, A. Reller, Helv.
Chim. Acta 1987, 70, 627 – 679.
[18] Acetone could also be used, whereas no reaction was observed in
toluene or CH2Cl2/toluene.
[19] D. Rondon, B. Chaudret, X.-D. He, D. Labroue, J. Am. Chem.
Soc. 1991, 113, 5671 – 5676.
[20] Phosphines and arenes instead of COD shut the reaction off.
[21] To ensure high rates, H2 pressures of ꢁ 2 bar were necessary; the
isomer ratio did not change much with pressure.
[6] W. A. Maier, Angew. Chem. 1989, 101, 135 – 146; Angew. Chem.
Int. Ed. Engl. 1989, 28, 135 – 145.
[7] Dinuclear hydridorhodium complexes were shown to be intrinsi-
cally selective catalysts for the semihydrogenation of alkynes to
E alkenes; unfortunately, the lifetime of the catalyst clusters is
very limited and Z-alkene formation takes over after a few
minutes under catalytic conditions. It seems that this method has
not found any applications to advanced synthesis, see: a) R. R.
Burch, E. L. Muetterties, R. G. Teller, J. M. Williams, J. Am.
Chem. Soc. 1982, 104, 4257 – 4258; b) R. R. Burch, A. J. Shuster-
man, E. L. Muetterties, R. G. Teller, J. M. Williams, J. Am.
Chem. Soc. 1983, 105, 3546 – 3556.
[8] For an E-selective transfer hydrogenation of unfunctionalized
alkynes with an iridium catalyst, see: a) K. Tani, A. Iseki, T.
Yamagata, Chem. Commun. 1999, 1821 – 1822; a catalyst formed
from [RhCl3(pyridine)3] and NaBH4 in DMF allowed for the
trans hydrogenation of diphenylacetylene to (E)-stilbene,
although functionalized alkynes gave the corresponding Z al-
kenes, see: b) P. Abley, F. J. McQuillin, J. Chem. Soc. D 1969,
1503 – 1504.
[22] AgPF6 and AgNTf2 gave similar results, whereas AgBF4,
AgSbF6, and AgB(ArF)4 (ArF = 3,5-(CF3)2C6H3) led to lower
selectivity and/or (significantly) lower yields.
[23] The rate difference between reactions catalyzed by the neutral
complex 4 or by the cationic species formed from 4/AgOTf was
actually much more pronounced in the cases shown in Table 2,
entries 14 and 15, and 16 and 17.
[24] This finding concurs with the conclusions drawn by Bargon and
co-workers, see Ref. [15]. Specifically, when the Z alkene
derived from alkyne 5 by Lindlar reduction (Z/E = 93:7) was
exposed to the standard reaction conditions (4 (5.5 mol%),
AgOTf (5.5 mol%), CH2Cl2, H2 (10 bar)), no isomerization
whatsoever was noticed after 30 min; even after 15 h, the Z/E
ratio was still ꢁ 84:16.
[25] COD ligated to [Cp*Ru] is kinetically rather labile, see: M. O.
Albers, D. J. Robinson, A. Shaver, E. Singleton, Organometallics
1986, 5, 2199 – 2205.
[26] a) H. Suzuki, H. Omori, D. H. Lee, Y. Yoshida, M. Fukushima,
M. Tanaka, Y. Moro-oka, Organometallics 1994, 13, 1129 – 1146;
for the reaction of this complex with diphenylacetylene to give m-
alkyne-bridged complexes, see: b) H. Omori, H. Suzuki, T.
Kakigano, Y. Moro-oka, Organometallics 1992, 11, 989 – 992;
c) review: H. Suzuki, Eur. J. Inorg. Chem. 2002, 1009 – 1023.
[27] U. Kçlle, B.-S. Kang, G. Raabe, C. Krꢀger, J. Organomet. Chem.
1990, 386, 261 – 266. This paper reports a hydride shift for
complex 8 of dH = ꢀ5.2 ppm in C6D6, whereas we record dH =
ꢀ5.74 in C6D6 and ꢀ6.2 in CD2Cl2.
[9] a) K. N. Campbell, L. T. Eby, J. Am. Chem. Soc. 1941, 63, 216 –
219; b) R. A. Benkeser, G. Schroll, D. M. Sauve, J. Am. Chem.
ˇ
Soc. 1955, 77, 3378 – 3379; c) I. Kovꢂrovꢂ, L. Steinz, Synth.
Commun. 1993, 23, 2397 – 2404; d) review: D. J. Pasto in
Comprehensive Organic Synthesis, Vol. 8 (Eds.: B. M. Trost, I.
Fleming), Pergamon, Oxford, 1991, pp. 471 – 488.
[10] For the use of (over)stoichiometric CrII, which is more functional
group tolerant, see: a) C. E. Castro, R. D. Stephens, J. Am.
Chem. Soc. 1964, 86, 4358 – 4363; b) A. B. Smith III, P. A.
Levenberg, J. Z. Suits, Synthesis 1986, 184 – 189; c) E. M. Car-
reira, J. Du Bois, J. Am. Chem. Soc. 1995, 117, 8106 – 8125.
[11] The term “trans hydrosilylation” of an alkyne is used here to
denote a reaction in which the H and the R3Si unit of R3SiH end
up trans to each other; because of the formalism of nomencla-
ture, however, the resulting alkenylsilanes are correctly termed
cis configured.
[28] G. Jia, W. S. Ng, C. P. Lau, Organometallics 1998, 17, 4538 – 4540.
[29] 6% (as determined by GC) of alkane were concomitantly
formed.
[30] The notion that the cationic hydrogen complex 9 itself is the
active catalyst or an immediate precursor to it is corroborated by
a stoichiometric control experiment. When substrate 5 was
exposed to one equivalent of 9 (generated from 7 and TfOH)
under Ar, alkene 6 was formed with excellent E/Z selectivity
(93:7). However, the yield was low because product 13 was
produced in equal amounts. This compound derives from
a ruthenium-catalyzed cycloaddition process that has precedent
in the literature, see: B. M. Trost, K. Imi, A. F. Indolese, J. Am.
Chem. Soc. 1993, 115, 8831 – 8832. In fact, reaction of 5 with
[12] a) B. M. Trost, Z. T. Ball, T. Jçge, J. Am. Chem. Soc. 2002, 124,
7922 – 7923; b) B. M. Trost, Z. T. Ball, J. Am. Chem. Soc. 2005,
127, 17644 – 17655; c) B. M. Trost, M. R. Machacek, Z. T. Ball,
Org. Lett. 2003, 5, 1895 – 1898; d) review: B. M. Trost, Z. T. Ball,
Synthesis 2005, 853 – 887.
[13] a) A. Fꢀrstner, K. Radkowski, Chem. Commun. 2002, 2182 –
2183; b) F. Lacombe, K. Radkowski, G. Seidel, A. Fꢀrstner,
Tetrahedron 2004, 60, 7315 – 7324; c) A. Fꢀrstner, M. Bonne-
kessel, J. T. Blank, K. Radkowski, G. Seidel, F. Lacombe, B.
Gabor, R. Mynott, Chem. Eur. J. 2007, 13, 8762 – 8783.
[14] a) K. Lehr, R. Mariz, L. Leseurre, B. Gabor, A. Fꢀrstner, Angew.
Chem. 2011, 123, 11575 – 11579; Angew. Chem. Int. Ed. 2011, 50,
11373 – 11377; b) K. Micoine, A. Fꢀrstner, J. Am. Chem. Soc.
2010, 132, 14064 – 14066.
[15] D. Schleyer, H. G. Niessen, J. Bargon, New J. Chem. 2001, 25,
423 – 426.
[16] The selective trans hydrogenation of more robust peptidic
alkyne substrates, with the aid of [Cp*Ru(MeCN)3]PF6 as
catalyst, is described in a patent, although no yields are given,
see: A. Robinson, S. Goodwin, WO 2011/146974A1, May 25,
2010.
[17] a) J. Heppekausen, R. Stade, R. Goddard, A. Fꢀrstner, J. Am.
Chem. Soc. 2010, 132, 11045 – 11057; b) J. Heppekausen, R.
Angew. Chem. Int. Ed. 2012, 51, 1 – 7
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
COD (1.1 equiv) in the presence of [Cp*RuCl(cod)] (4,
5 mol%) delivered 13 in 54% yield.
[31] 11 is a known complex previously prepared by coupling of two
butadiene units in the coordination sphere of ruthenium, see: K.
Itoh, K. Masuda, T. Fukahori, K. Nakano, K. Aoki, H.
Nagashima, Organometallics 1994, 13, 1020 – 1029.
[33] 11 might convert into 12 by isomerization/electrocyclization/
aromatization. The conversion of COD to ethylbenzene with
modified (cyclopentadienyl)ruthenium templates has precedent
in the literature, although no mechanism was proposed, see: B.
Dutta, R. Scopelliti, K. Severin, Organometallics 2008, 27, 423 –
429.
[32] For the cleavage of norbornadiene within the coordination
sphere of [Cp*Ru], see: H. Suzuki, T. Kakigano, H. Fukui, M.
Tanaka, Y. Moro-oka, J. Organomet. Chem. 1994, 473, 295 – 301.
6
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 7
These are not the final page numbers!

Angewandte
Chemie
Communications
Hydrogenation
K. Radkowski, B. Sundararaju,
A. Fꢁrstner*
&&&& — &&&&
A Functional-Group-Tolerant Catalytic
trans Hydrogenation of Alkynes
Against the rules: During the hundred
years following Sabatier’s groundbreak-
ing work on catalytic hydrogenation,
syn delivery of the H atoms to the p sy-
stem of a substrate remained the gov-
erning stereochemical rule. An exception
has now be found with the use of cationic
[Cp*Ru] templates, which accounts for
the first practical, functional-group-toler-
ant, broadly applicable and highly E-
selective semihydrogenation method for
alkynes.
Angew. Chem. Int. Ed. 2012, 51, 1 – 7
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7
These are not the final page numbers!