Communications
Table 1: Transfer hydrogenation of alkynes with catalyst 1.[a]
acetylene to determine the catalystꢀs selectivity towards
internal and terminal alkynes. The reaction profiles of the
transfer hydrogenation of these mixtures look very similar to
the respective individual reactions. The main difference is
seen in the first hour: whereas phenylacetylene reaches 50%
conversion in the first half hour, 1-phenyl-1-propyne seems to
lag somewhat, with less than 5% conversion after the first half
hour (it would normally be approximately 20%). Once about
90% of the terminal alkyne has been converted (after 1 h),
the catalyst also begins to hydrogenate the internal alkyne,
after which the same product distributions are obtained. We
can therefore conclude that the catalyst has a preference for
terminal alkynes over internal alkynes.
Hydrogenation of alkynes containing several other func-
tionalities was also carried out. These experiments revealed
that neither alcohols nor ketones inhibit catalytic activity and
lead, with good to excellent chemoselectivity, to semihydro-
genation of the alkyne functionality only (Table 1, entries 15–
19). Since the reactions are performed in MeCN, it seems safe
to assume that the catalyst is also compatible with nitrile
functionalities. The a,b-unsaturated ketone 4-phenyl-3-
butyne-2-one initially gives the Z alkene, but during the
time required to reach full conversion it isomerizes to give the
thermodynamically more stable E alkene as the main prod-
uct, probably via the enolate (Table 1, entry 17). No hydro-
genation of ketones to alcohols was observed, which is
remarkable because most transfer-hydrogenation reactions
(based on RuII catalysts) specifically lead to reduction of
carbonyl functionalities. To our knowledge, this is therefore
the first example of a homogeneous transfer-hydrogenation
catalyst that specifically reduces alkynes in the presence of
ketones.
Entry
Substrate
Solvent Conv. [%] t [h][b] Z alkene/
E alkene/
alkane[c]
1
2
3
4
THF
MeCN >99
THF >99
MeCN >99
>99
4
12
4
91/7/2
96/4/1
98/2/–
95/–/5
8
5
MeCN
35
24
7
98/1/1[d]
98/–/2[d]
6
7
8
9
10
11
THF
>99
MeCN >99
THF 80
MeCN >99
THF
MeCN
<24[e] >99/–/–[d]
24[f]
<24[e]
<2
–/99/–
93/3/4
50/50[f]
>99/–[f]
>99
85
24
12
MeCN >99
<24
>99/–[d,f]
13
14
THF
96[g]
2[e] 88/12[f,h]
7
MeCN >99[g]
95/5[f,h]
15
16
MeCN >99
MeCN >99
<24[e] >99/–/–[d]
<24[e] 91/9/–[d]
17
18
THF
MeCN
>99
52
6
7/93/–[d,i]
24 75/25/–[d,i]
24 69/–/31[d,i]
19
MeCN
90
20
21
22
23
24
25
THF
MeCN
THF
MeCN >99
THF un-
MeCN known
29[f]
88
>99
1
46/54/–[d,i]
24 59/22/7[d,i]
<3
<24
24
several
products
Z and E
(di)enes[d]
[a] Reaction conditions: 160 mm of substrate, 1.6 mm of catalyst, and
800 mm of HCO2H/NEt3 in the specified solvent at reflux. [b] Time at full
conversion determined by extrapolation. [c] Product distribution as
determined by GC and 1H NMR spectroscopy. [d] Product distribution as
determined by 1H NMR spectroscopy. [e] The exact time for full
conversion is not known but no more starting material was present
after 24 h. [f] Product distribution is depicted as alkene/alkane ratio.
[g] Conversion ceases after the specified time then the alkenes slowly
isomerize. [h] A double-bond shift occurs after 7 h. 2-, 3-, and 4-Octenes
were identified by 1H NMR spectroscopy but could not be separated
from the over-reduction product by GC. [i] No reduction of carbonyl
functionalities was observed.
Hydrogenation of the electron-poor alkyne dimethyl
butynedioate (Table 1, entries 20 and 21) was less successful,
thereby setting some boundaries to the selectivity of our
catalyst. This could have been expected as the products of this
reaction are known to form relatively stable alkene com-
plexes with low-valent palladium, which could deactivate the
active species; the formation of palladacyclic compounds may
also play a role. In fact, dimethyl fumarate has previously
been used to isolate reactive species, such as Pd0(NHC)
complexes.[8b] Note that this deactivation is much more
of substrates. Both aromatic (Table 1, entries 1–7) and simple pronounced in the less strongly coordinating solvent THF
aliphatic (Table 1, entry 9) internal alkynes are readily hydro- than in MeCN.
genated to the desired Z alkene with good to excellent
stereoselectivity and, importantly, hardly any over-reduction
to alkane. We note that performing the reaction in the more
strongly coordinating solvent MeCN results in a longer
reaction time but generally gives a higher selectivity for the
The limits of the chemoselectivity of the catalyst are
reached when an enyne such as 1-ethynylcyclohexene or a
diyne such as diphenylbutadiyne (Table 1, entries 22–25) are
employed. The enyne initially behaves the same as 1-octyne—
the alkyne moiety is semihydrogenated—and the alkene
desired product. Simple terminal alkynes (Table 1, entries 10– produced is slowly hydrogenated only when most of the
14) react equally well but tend to give minor amounts of by-
products (< 5%). The initial chemoselectivity towards
alkynes over alkenes is equally good but in THF over-
reduction of the alkene starts after about 90% conversion.
The results are essentially the same for terminal aliphatic
alkenes in MeCN, although no over-reduction is observed for
styrene or p-tolylstyrene.
alkyne has been hydrogenated. Isomerization to the more
stable 3-ethylidenecyclohexene, which in turn is partly hydro-
genated, is observed and this leads to a rather complex
mixture of several C8H12/14 isomers. Transfer hydrogenation of
the diyne also gives a mixture of compounds. Both Z and
E alkenes (enynes, dienes) are detected by NMR spectrosco-
py whereas the fully reduced product is clearly not detected.
To further explore the chemoselectivity we performed the The Pd0(NHC) complex employed is therefore a very good to
reaction with mixtures of 1-phenyl-1-propyne and phenyl-
excellent chemo- and stereoselective catalyst for the semi-
3224
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3223 –3226