P. K. Sharma et al. / Tetrahedron Letters 48 (2007) 8704–8708
8705
Having developed optimized conditions, we next investi-
gated the scope and limitations of the method (Table 2).
Monosubstituted terminal olefins were readily reduced
by this reagent (entries 1–8). The reactions for monosub-
stituted alkenes were complete in one hour except for
styrene (entry 6). Disubstituted olefins (entries 9 and
10) were also reduced completely, but more slowly com-
pared to monosubstituted olefins. Importantly, mono-
and disubstituted olefins were efficiently reduced while
more highly substituted olefins (entries 11–13) were
virtually inert to these reducing conditions. Interest-
ingly, in the case of 3-benzoyl-1-allylthymine (entry
16), the monosubstituted olefin was selectively reduced
over the endocyclic double bond of the heterocycle. It
is pertinent to mention here that this endocyclic double
bond in a similar thymidine analogue reacts with the
RuCl3/NaIO4 system and is dihydroxylated.9 Activated
trisubstituted alkenes (entries 14 and 15) were efficiently
reduced by this reagent system. The conditions
employed for the transformations are compatible with
a variety of functional groups. Benzyl ethers were not
affected under the reaction conditions, which make this
protocol potentially appealing for carbohydrate and
nucleoside chemistry. Esters, amides and alkyl/aryl
ethers were also found to be stable.
O
O
O
O
RuCl3xH2O/NaBH4
4
5
Scheme 2.
the explanation, we subjected alkene 1 to reduction with
NaBH4, but no reaction was observed even after over-
night stirring. This led us to believe that it might be
the small amount of active ruthenium species from the
first step, that was responsible for the hydrogenation
in the presence of NaBH4. On the basis of these obser-
vations, we treated alkene 1 with NaBH4 (1 equiv) and
RuCl3 · H2O10 (25 mol %), and were pleasantly sur-
prised to find that alkene 1 was cleanly hydrogenated
to 3 in excellent yield. The remarkable success of this
reaction protocol inspired us to apply it to various other
carbohydrate substrates (including allyl and vinyl sub-
stituents) as well as simple substrates such as styrene
and allylphenyl ether.
Optimization studies were performed using safrol (4) as
the substrate (Scheme 2). Complete conversion of 4 into
5 was observed while using RuCl3 · H2O down to
10 mol % along with two equivalents of NaBH4.
Although hydrogenation of 4 to 5 was complete even
while using one equivalent of NaBH4, the results were
most consistent in the majority of cases (Table 2) while
using two equivalents of NaBH4.
All these features in combination give the present
reducing agent a broad and general synthetic utility.
Ruthenium is neither poisonous nor explosive.14
RuCl3 · H2O has been used extensively in organic chem-
istry for oxidations/oxidative degradations of unsatu-
rated organic compounds.15,16 It has long been known
that transition metal salts catalyze the hydrolysis of
borohydride ions under aqueous conditions13 to gener-
ate hydrogen that can be used in combination with tran-
sition metal salts for the reductions. There are several
reports using ruthenium complexes, sometimes prepared
using long and tedious processes under anhydrous con-
ditions, for various reductive applications.17–19 Reduc-
tion of alkenes using NaBH4 as the reducing agent in
combination with dihydridoruthenium(II) complex has
also recently been reported.20 However, the direct use
of this relatively inexpensive reagent as a catalyst for
reductions is, to the best of our knowledge, reported
here for the first time. It is reasonable to assume that
the species responsible for the selective reduction of
mono- and disubstituted alkenes is most likely a ruthe-
nium hydride or active ruthenium in the presence of
hydrogen generated under the reaction conditions.
Subsequently, the influence of solvent on the reaction
course was investigated (Table 1). There was virtually
no difference in yield on using aqueous conditions or
non-aqueous conditions. However, the reaction time in-
creased in the absence of water (entries 3 and 6). It seems
reasonable when we consider that a proton source such
as water is required for the hydrolysis of NaBH4. Thus
longer reaction times in the absence of water is simply
a reflection that the amount of water present was
limited. We did not examine a larger series of different
solvents, as RuCl3 is known to accept a wide range of
solvents, whereas the scope and limitations of NaBH4
are well known.11,12 Concerning the influence of temper-
ature, we found that the addition of RuCl3 was exother-
mic with rapid evolution of hydrogen,13 and hence,
cooling was needed. However, the reactions proceeded
well at room temperature.
Table 1. Influence of solvent
Entrya
Solvent
Reaction time (h)
Conversionb (%)
5 (% Yield)c
1
2
3
4
5
6
THF (3 mL)/H2O (1 mL)
THF (3 mL)/H2O (0.5 mL)
THF (3 mL)
CH3CN (3 mL)/H2O (1 mL)
CH3CN (3 mL)/H2O (0.5 mL)
CH3CN (3 mL)
1
1
3
1
1
4
100
100
100
100
100
100
96
95
96
96
96
95
a All reactions were performed at room temperature on a 1 mmol scale using 10 mol % RuCl3 · H2O and 2 equiv NaBH4.
b Determined by NMR.
c Isolated yield.