Chemoselective Hydrogenation and Transfer Hydrogenation of Olefins and Carbonyls with the Cluster-Derived Ruthenium Nanocatalyst in Water

Article abstract of DOI:10.1002/cctc.201200448

Ion pairing of [H₃Ru₄(CO)₁₂]^- with the quaternary ammonium groups of water-soluble poly(diallyldimethylammonium chloride) gives the precursor of a nanocatalyst for hydrogenation and transfer hydrogenation reactions in water. In hydrogenation reactions, "on water" effect is seen for substrates such as cyclohexanones, methyl pyruvate, acetophenone, and safflower oil. With these substrates, higher turnover numbers are obtained in water than in methanol. The cluster-derived catalyst shows unique chemoselectivity, which is not seen either in a catalyst prepared through ion pairing of [RuCl₄]^- with the quaternary ammonium groups of the same polymer or in commercial (5%) Ru/Al₂O₃. In contrast to Ru/Al₂O₃, the [RuCl₄]^--derived catalyst, or many other ruthenium-based catalytic systems, the cluster-derived catalyst is totally inert toward the hydrogenation of -NO₂, -CN, and aromatic ring functionalities. In water, typical ketones and aldehydes could be reduced by using the cluster-derived catalyst and formate as the hydrogen donor. Industrially important cyano- and nitrobenzyl alcohols could thus be made from the corresponding aldehydes. High-resolution TEM data suggest that unique chemoselectivity is a result of highly crystalline ruthenium nanoparticles that consist mainly of Ru(111) crystal planes.

Full text of DOI:10.1002/cctc.201200448

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201200448
Chemoselective Hydrogenation and Transfer
Hydrogenation of Olefins and Carbonyls with the Cluster-
Derived Ruthenium Nanocatalyst in Water
[
a]
Arindam Indra, Prasenjit Maity, Sumit Bhaduri,* and Goutam Kumar Lahiri*
À
À
Ion pairing of [H Ru (CO) ] with the quaternary ammonium
mercial (5%) Ru/Al O . In contrast to Ru/Al O , the [RuCl ] -de-
2 3 2 3 4
3
4
12
groups of water-soluble poly(diallyldimethylammonium chlo-
ride) gives the precursor of a nanocatalyst for hydrogenation
and transfer hydrogenation reactions in water. In hydrogena-
tion reactions, “on water” effect is seen for substrates such as
cyclohexanones, methyl pyruvate, acetophenone, and safflow-
er oil. With these substrates, higher turnover numbers are ob-
tained in water than in methanol. The cluster-derived catalyst
shows unique chemoselectivity, which is not seen either in
rived catalyst, or many other ruthenium-based catalytic sys-
tems, the cluster-derived catalyst is totally inert toward the hy-
drogenation of ÀNO , ÀCN, and aromatic ring functionalities. In
2
water, typical ketones and aldehydes could be reduced by
using the cluster-derived catalyst and formate as the hydrogen
donor. Industrially important cyano- and nitrobenzyl alcohols
could thus be made from the corresponding aldehydes. High-
resolution TEM data suggest that unique chemoselectivity is
a result of highly crystalline ruthenium nanoparticles that con-
sist mainly of Ru(111) crystal planes.
À
a catalyst prepared through ion pairing of [RuCl4] with the
quaternary ammonium groups of the same polymer or in com-
Introduction
Among many of the require-
ments
of
environmentally
benign and economically viable
chemical processes, the use of
water rather than organic sol-
vents is an important theme of
[
1,2]
current research.
Water is
Scheme 1. Synthetic routes for the preparation of the water-soluble poly(diallyldimethylammonium chloride)
cheap and nonhazardous, and
a number of gases such as hy-
drogen have good solubility in
(PDADMAC]-supported catalysts (2–4).
[5]
water. Although catalytic hydrogenation reactions are widely
sis. Here the catalyst is dissolved/dispersed in one liquid
phase and the reactants and product(s) in the second liquid
phase. Recovery and recycling of the catalyst are achieved
through simple phase separation. With water-soluble/dispersi-
ble catalysts, another point of considerable interest is that
many organic reactions if performed in water show improve-
[
3]
performed in chemical industry, hydrogen gas under high
pressure is hazardous. Therefore, for selected reactions, transfer
hydrogenation with a suitable hydrogen donor should be ex-
[
4]
plored as an alternative. For both hydrogenation and transfer
hydrogenation reactions, homogeneous catalysts are generally
far more selective than heterogeneous ones but separation
from the product and recycling of the catalysts are often
difficult.
[1,6]
ments in rates and selectivities.
Nanoparticle catalysis in green solvents such as water has
shown tremendous potential, and systematic guidelines for the
further development of metal cores, stabilizers, and solvents
The need for efficient separation of the product and the cat-
alyst while maintaining the selectivity advantage of a soluble
catalyst has led to the concept of liquid–liquid biphasic cataly-
[7]
have been presented. Given the above-mentioned back-
ground and our own involvement with nanocatalysis over the
years, we wanted to evaluate the potential of the water-solu-
ble ruthenium precatalyst 2 (see Scheme 1) in a broad range
[
a] Dr. A. Indra, Dr. P. Maity, Prof. Dr. S. Bhaduri, Prof. Dr. G. K. Lahiri
[8–18]
of hydrogenation and transfer hydrogenation reactions.
We
Department of Chemistry
Indian Institute of Technology Bombay
Powai, Mumbai 400 076 (India)
Fax: (+91)22-25723480
also wanted to determine whether the literature reported “on
water” effect, that is, improved rate in water, could be ob-
served in these reactions. Thus, the main motivations for the
work reported here are as follows: 1) to evaluate the feasibility
of water as a reaction medium for hydrogenation and transfer
hydrogenation reactions with 2 as the precatalyst, 2) to investi-
E-mail: lahiri@chem.iitb.ac.in
sbhaduri@chem.iitb.ac.in
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201200448.
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 322 – 330 322

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
gate the scope for chemoselective hydrogenation and transfer
hydrogenation reactions with 2 and its reuse, and 3) to con-
duct structural studies on the active catalyst 3 with a view to
rationalize the observed chemoselectivity.
terization techniques and has already been reported in our ear-
[15]
lier paper. Here, we report the detailed characterization data
on the decarbonylated active catalyst 3. TEM and high-resolu-
tion TEM (HRTEM) images, particle size histogram, and UV/Vis,
XP, and energy-dispersive X-ray (EDX) spectra of 3 are shown
in Figures 1 and 2. TEM images show nearly monodisperse par-
ticles with an average diameter of 2.2 nm (Figure 1). HRTEM
images show good crystallinity of particles with predominantly
Ru(111) crystal planes as observed from the axis direction of
TEM measurement (Figure 1d).
In water, under the catalytic conditions, 2 generates the
active catalyst 3 and hydrogenates alkenes, aldehydes, and
ketone functionalities selectively. In contrast to many other
ruthenium-based catalysts, it is totally inert toward the hydro-
genation of ÀNO , ÀCN, and arene groups. With formate as
2
a hydrogen donor, precatalyst 2 also transfer hydrogenates ni-
trobenzaldehydes selectively. In hydrogenation reactions, the
water solution of the catalyst is recycled many times with
a negligible decrease in activity. With 3, higher activities are
observed in water than in methanol. Through a detailed char-
acterization of 3, we propose a plausible reason behind the
notable chemoselectivity of 3.
Results and discussion
Synthesis and characterization of the catalysts
In our earlier work, we have shown that ion pairing of anionic
carbonyl clusters with tetraalkyl ammonium groups of func-
tionalized inorganic oxides or organic polymers is a general
[8–14]
method for the synthesis of nanocatalyst precursors.
By
using water-soluble poly(diallyldimethylammonium chloride)
(
PDADMAC, 1) as the stabilizer, we have also reported plati-
[
15–18]
num- and ruthenium-based nanocatalysts.
The chemose-
lective hydrogenation of m-nitro- and m-cyanobenzaldehyde
in water and in methanol by using 2 and an analogue of 2
with MCM-41 as the support, respectively, has recently been
Figure 1. a,c,d) TEM images at different magnifications and b)particle size
histogram of 3.
[
15]
reported.
Addition of a methanolic so-
lution of K[H Ru (CO) ] to the
3
4
12
aqueous solution of 1 leads to
the formation of the cluster–
polymer
adduct
2
(see
Scheme 1 and the Experimental
Section). The anionic ruthenium
À
cluster displaces the Cl ions
and binds with the quaternary
nitrogen atoms of the polymer.
The active catalyst 3 is formed
on the hydrogen treatment of
water or the methanolic solu-
tion of 2. Although the hydro-
gen treatment removes the car-
bonyl groups, the Ru4 frame-
[
15]
work remains unaffected. The
detailed synthetic method is de-
[
15]
scribed in our previous paper
and also in the Experimental
Section.
The evidence for the structur-
al identity of 2 comes from
spectroscopic (IR, multinuclear
NMR) and microscopic charac-
Figure 2. a) UV/Vis spectrum (in water), b) XP spectrum, and c) energy-dispersive X-ray spectrum of 3. (The signal
for copper is from the sample holder.)
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 322 – 330 323

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
The presence of small rutheni-
um particles is also confirmed
by the typical UV/Vis spectrum
with a featureless but exponen-
tial decay profile along the
[
a]
Table 1. Selected data for the hydrogenation reaction of various substrates with 3, 4, and 5.
Entry
1
Substrate
Product
Catalyst
Solvent
Conversion
%]
TON (TOF)
À1
[
[h
]
3
4
3
2
H O
100
80
625 (156)
500 (125)
506 (126)
2
H O
higher
wavelength
(Fig-
[
19]
MeOH
81
ure 2a).
The Ru3p XP spec-
trum shows that Ru3p3/2 has
a binding energy of approxi-
mately 461 eV, which is in ac-
cordance with the zero-valent
state (Figure 2b). The EDX analy-
sis also confirms the presence of
ruthenium without any observa-
ble contamination from any
other metals (Figure 2c). All
these data along with our previ-
2
3
H
2
2
O
O
85
80
531 (133)
3
4
3
H
500 (125)
3
4
3
H
H
2
O
O
90
68
25
562 (140)
424 (106)
156 (39)
2
MeOH
5
6
3
H
2
2
O
O
100
30
625 (156)
[
15]
ous analysis indicate that the
loss of the carbonyl ligands
leads to the formation of small
and crystalline ruthenium nano-
particles in 3.
3
3
3
3
4
3
H
187 (47)
9 (2)
31 (8)
625 (156)
594 (148)
319 (78)
MeOH
1.5
H
2
H
2
H
2
O/EtOAc (1:1)
O
O
5
100
95
7
MeOH
51
The performance of the clus-
ter-derived catalyst 3 has been
8
9
3
3
H
2
O
100
80
625 (156)
500 (125)
compared with a RuCl -derived
3
H
2
2
O
O
catalyst and commercial (5%)
Ru/Al O . The RuCl -derived
1
0
3
3
3
3
H
53
10
40
98
331 (83)
62 (16)
250 (63)
142 (36)
2
3
3
MeOH
nanocatalyst
4
is prepared
H
H
2
O/EtOAc (1:1)
O
through slow addition of an
aqueous solution of RuCl3 to
a water solution of the polymer,
followed by removal of water to
minimum volume (see the Ex-
perimental Section). The orange
solid that separates out on stor-
age is converted to the active
catalyst 4 under hydrogen pres-
sure at 353 K.
[b]
1
1
safflower oil
saturated fat
2
[
c]
12
3
3
H O
2
1.3
1.1
<1 (<0.1)
<1 (<0.1)
MeOH
[
c]
13
3
3
4
5
H O
2
1.8
5.5
55
100
1.1
ꢂ1 (<0.1)
ꢂ4 (<0.3)
34 (2.3)
MeOH
MeOH
MeOH
62 (4.1)
3
3
H
2
O
<1 (<0.1)
<1 (<0.1)
[
c]
1
4
MeOH
1
À2
[a] Reaction conditions: 20 mmol substrate, 3.2ꢁ10 mmol Ru catalyst, 5 mL solvent, T=353 K, P=40 bar H
2
,
À1
À2
t=4 h, 900 rpm stirring speed; [b] 1.37ꢁ10 mmol Ru catalyst; [c] 2 mmol substrate, 3.2ꢁ10 mmol Ru cata-
lyst, 5 mL solvent, T=353 K, P=50 bar H , t=15 h, 900 rpm stirring speed.
2
Hydrogenation in water
With 3 as the catalyst, hydroge-
nations of a wide range of sub-
strates containing olefinic and carbonyl functionalities have
been studied (see Table 1). For a few substrates, the catalytic
performances of 3, 4, and 5 are also compared. Unless stated
otherwise, in all the cases the reactions have been performed
in a biphasic medium consisting of a water solution of 3 and
the organic substrate. Good to excellent conversions are ob-
tained for ketones, a-ketoester (Table 1, entries 1–6), and ole-
fins (Table 1, entries 7–11). As the catalytic runs are of fixed du-
ration and ꢀ98% conversions are achieved for many sub-
strates, the turnover frequencies (TOFs) listed in Table 1 for
these substrates correspond to the lower limit.
ual increase in the steric bulk of the substrates. Similar steric
effects are observed in the hydrogenation of styrene deriva-
tives (Table 1, entries 7–10), in which the TOFs are in the fol-
lowing order: styreneꢁp-methyl styrene>a-methyl styrene.
À1
The TOF of acetophenone (47 h ) is notably less than that of
À1
cyclohexanone derivatives (>125 h ), probably because of
both steric and electronic factors. Similarly, the comparative
TOFs of a-methyl styrene (83), vinyl cyclohexane (125), and sty-
rene and p-methyl styrene (156) are best rationalized by invok-
ing a combination of steric and electronic factors.
To evaluate whether the synthesis of the ruthenium nanoca-
talyst by the carbonyl cluster route offers any practical advant-
age in terms of activity, the catalytic performances of 3, 4, and
5 have been compared for a few substrates. For cyclohexa-
For the cyclohexanone derivatives (Table 1, entries 1–3), the
TOFs are in the following order: c-hexanone>2-methyl-c-hexa-
none>2,6-dimethyl-c-hexanone. This must be due to the grad-
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 322 – 330 324

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
none, methyl pyruvate, and styrene (Table 1, entries 1, 4, and
discussed above would probably be the main factor behind
rate enhancement. However, the differences in rate enhance-
ment for different substrates must be attributed to additional
factors such as polarity and normal or trans-phase hydrogen
bonding.
7
1
), the turnover numbers (TONs) of 3 and 4 are approximately
.25, 1.3, and 1.05, respectively. Therefore, in terms of activity,
3
is marginally better than 4.
The hydrogenation of safflower oil (Table 1, entry 11) to the
corresponding saturated fat is an industrially significant reac-
Notably, with 3 as the catalyst, no ring hydrogenation is ob-
served with any of the aromatic substrates (Table 1, entries 6–8
and 10). The inertness of 3 toward arene hydrogenation is con-
firmed (Table 1, entries 12–14) by performing reactions under
increased hydrogen pressure [50 bar (1 bar=100 kPa)] and for
a longer reaction time (15 h). With benzene, phenol, and ani-
sole as substrates, ꢂ1.8% ring hydrogenation is observed in
water and <5.5% in methanol. In contrast, under identical
conditions, 55–100% ring hydrogenation is observed with 4
and 5 as catalysts and phenol as the substrate.
[
20]
tion.
In our earlier paper, we had reported the use of
a water-soluble, platinum carbonyl cluster-derived catalyst for
[
17]
effecting this reaction. With 3 as the catalyst, the reaction is
successful with full conversion. The ruthenium analysis of the
product, an important consideration from the actual applica-
tion point of view, shows no contamination by ruthenium. The
iodine value, a traditional way of measuring relative unsatura-
tion in oils and fats, decreases from 115 to 19, which indicates
complete saturation.
As mentioned in the Introduction, one of the motivations
for this work was to evaluate if water as a reaction medium
offers any special advantage with 3 as the catalyst. To answer
this question, hydrogenation of selected substrates has been
performed in methanol, in which 3 is fully soluble. Notably, the
active catalyst 3 is completely insoluble in DMF, THF, ethanol,
or acetonitrile, and this limits the study of hydrogenation reac-
tion only in methanol.
In our earlier paper, the chemoselective hydrogenation of
m-nitrobenzaldehyde and m-cyanobenzaldehyde to the corre-
sponding nitro- and cyanobenzyl alcohols with 3 as the cata-
[15]
lyst was reported. The chemoselectivity of 3 was in sharp
contrast to that of 4 and 5, both of which hydrogenate ÀNO ,
2
ÀCN, and ÀCHO functionalities indiscriminately. In view of
these results, the chemoselectivity of 3 toward ÀCHO in the
presence of olefinic functionality and towards p-nitrobenzalde-
hyde and p-cyanobenzaldehyde has also been evaluated, and
the results are shown in Table 2.
As shown in Table 1 (entries 1, 4, and 7) for cyclohexanone,
methyl pyruvate, and styrene, the TONs in methanol are 506,
1
56, and 319 whereas in water they are 625, 562, and 625,
In the hydrogenation of cinnamaldehyde and crotonalde-
hyde (Table 2, entries 1 and 2), low chemoselectivities are dem-
onstrated by 3. Cinnamaldehyde is more reactive than croto-
naldehyde, and it gives higher TON. However, high chemose-
lectivity is observed in the hydrogenation of m-vinylbenzalde-
hyde (Table 2, entry 3), in which the alkene functionality is se-
lectively hydrogenated. With 4 or 5 as the catalyst, no such
chemoselectivity is observed; both the alkene and the alde-
hyde functionalities are hydrogenated. Similarly, in the hydro-
genation of p-nitrobenzaldehyde (Table 2, entry 4), complete
chemoselectivity toward the reduction of the aldehyde func-
tionality is achieved with 3 but not with 4 or 5. For the latter
catalysts, both the functional groups are hydrogenated. With 3
as the catalyst in the hydrogenation of 3-nitrocyclohexene
(Table 2, entry 5), the nitro group remains untouched. The hy-
drogenation of 4-cyanobenzaldehyde with 3 results in 4-cyano-
benzyl alcohol, with the cyano group remaining untouched,
whereas 4 and 5 hydrogenate both the ÀCN and ÀCHO
groups (Table 2, entry 6).
respectively.
The superiority of water is more pronounced for substrates
acetophenone and a-methyl styrene (Table 1, entries 6 and 10).
For these two substrates, the TONs in water are approximately
2
1 and 5.5 times more than that obtained in methanol. Even
a mixture (1:1) of water and ethyl acetate gives notably higher
TONs than does methanol. For acetophenone and a-methyl
styrene, the TONs in the water/ethyl acetate mixture are ap-
proximately 3.5 and 4 times more than that obtained in metha-
nol. Notably, with the above-mentioned substrates, a two-
phase mixture is formed because of substrates’ low to negligi-
ble miscibility with water. In contrast, with methanol a single-
phase catalytic system is produced.
A plausible explanation for the observed beneficial effect of
water on the activity can lie in the relative solvating power of
water and methanol. Owing to the hydrophobic interaction
with the polymer chain, coordination by water to the coordina-
tively unsaturated catalytic sites can be weaker than that by
methanol. This would result in higher activity in the aqueous
medium, as observed in these experiments.
In our earlier work, we had shown that the results of catalyst
recycling experiments with 3 as the catalyst were substrate
[15]
In many organic reactions when water is used as the reac-
tion medium, depending on the solubility of the organic reac-
specific. Thus, in styrene hydrogenation recycling, there was
no decrease in activity; however, for m-nitrobenzaldehyde,
there was a substantial decrease (>30%). In this work, the re-
cyclability of catalyst 3 has been tested in the hydrogenation
of cyclohexanone and the results are similar to that for styrene.
Negligible variations in the TONs from batch to batch are ob-
served (Figure 3a). TEM images of 3 after the hydrogenation of
cyclohexanone were recorded, and the agglomeration of
ruthenium nanoparticles to a small extent was observed (see
Figure S1). The distribution of the particles remains uniform,
and particles within the size range of 3–9 nm can be observed.
[21]
tants, different rate-enhancing mechanisms can operate. For
compounds that are fully soluble in water, the increase in the
[
22]
rate is mainly caused by the hydrophobic effect. In certain
cases, in addition to the hydrophobic effect, polarity and hy-
drogen bonding contribute to the increased rates.
However, for compounds that are slightly to very sparingly
soluble in water, rate enhancement is due to the hydrophobic
[
23]
effect and/or trans-phase hydrogen bonding. As 3 is soluble
in water, it is proposed that the hydrophobic effect of the type
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 322 – 330 325

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
in water with various homoge-
neous, heterogeneous, and bi-
phasic catalytic systems has
[
a]
Table 2. Chemoselective hydrogenation of selected substrates with 3.
Entry
1
Substrate
Product
Catalyst
Conversion
Product selectivity
[%]
[
%]
[4,25]
been reported recently.
No transfer hydrogenation
was observed with isopropanol
as the hydrogen donor and 3 as
the catalyst; only negligible re-
duction of acetophenone was
achieved. Mixtures of NEt3–
HCOOH at various ratios also
did not give the desired reac-
tion. However, the use of
sodium formate as the hydride
donor gave reasonable conver-
sion and the reaction conditions
with sodium formate as the
donor were optimized. With
10 equivalents of HCOONa at
a
b
c
3
41
a: 18
b:
4
c: 19
d: 46
2
d
e
f
3
99
e:
8
f: 45
3
4
g
h
3
46
g: ꢀ96
4
5
100
100
h: 100
h: 100
353 K for 10 h, full conversion
with a TON of 184 could be ob-
tained for substrates such as cy-
clohexanone and acetophe-
none. Under these conditions, 3
acted as an efficient transfer hy-
drogenation catalyst and differ-
ent ketones and aldehydes
could be reduced with good
TONs (Table 3, entries 1–9).
i
3
4
5
3
3
4
5
100
90
i: 100
j: 100
j
k
100
100
42
j: 63, k: 37
ꢀ99
5
6
l
l: 100
For the cyclohexanone deriv-
atives, increasing the steric bulk
in the ortho position decreases
the TON (Table 3, entries 1–3).
Given the industrial importance
of nitro- and cyanobenzyl alco-
hols, we were mainly interested
in using transfer hydrogenation
for the reduction of this class of
m
n
100
100
m: 100
m: 100
À2
[
a] Reaction conditions: 20 mmol substrate, 3.2ꢁ10 mmol Ru catalyst, 5 mL water, T=353 K, P=40 bar H
2
,
t=4 h, 900 rpm stirring speed.
A direct ruthenium estimation before and after 5 batches,
corresponding to an overall TON of >3000, shows that the
total ruthenium loss from handling and/or leaching is ꢂ3.5%.
These results show that the hydrogenation of cyclohexanone
can be performed in a batch mode with nearly full conversion
for five batches and recovery and recycling of the catalyst do
not pose a problem. The time-monitored recycling experiment
for the hydrogenation of cyclohexanone shows that in the
second cycle, the decrease in activity is approximately 15% at
low conversion whereas a negligible decrease in activity is ob-
served from the second to the third cycle (see Figure S2).
substrates. For all the derivatives, the desired transformations
could be achieved (Table 3, entries 10–14). A decrease in TON
is observed in going from p-nitro derivative to m- or o-nitro
derivative. Thus, p-nitrobenzaldehyde gives 100% conversions,
whereas m- and o-nitro derivatives give 50 and 45% conver-
sions, respectively.
The recyclability of 3 in transfer hydrogenation has been
tested with acetophenone as the substrate (Figure 3b). A nota-
ble decrease (ꢁ35%) in TON is observed in the second batch,
and again in the third batch there is a decrease in TON but
a little less (ꢁ15%). In the next two batches, the decrease in
TON is negligible. A direct ruthenium estimation before and
after five batches showed that the total ruthenium loss was
approximately 3.5%, that is, nearly the same as that for the hy-
drogenation reactions of cyclohexanone. However, in the hy-
drogenation of cyclohexanone with the recycled catalyst, there
was very little decrease in activity. The notable decrease in ac-
tivity in transfer hydrogenation on recycle in the second and
third batches is therefore not due to ruthenium loss.
Transfer hydrogenation in water
As mentioned earlier, one of the motivations for the work re-
ported here was to explore the possibility of using 3 as a trans-
fer hydrogenation catalyst. Transfer hydrogenation with isopro-
À
panol as the hydrogen donor and [H Ru (CO)12] as the preca-
3
4
[
24]
talyst was reported many years ago. Transfer hydrogenation
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 322 – 330 326

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Table 3. Transfer hydrogenation reaction with 3 in water in the presence
[a]
of sodium formate as the hydride donor.
Entry Substrate
Product
Conversion TON
%]
[
1
2
100
60
184
110
3
40
73
4
5
76
139
184
100
6
98
180
7
8
9
100
97
184
178
184
100
10
45
83
Figure 3. Recycling experiment with 3 for the a) catalytic hydrogenation of
cyclohexanone and b) transfer hydrogenation of acetophenone.
1
1
50
100
34
92
184
63
As mentioned earlier, similar observations were made in the
hydrogenation of m-nitrobenzaldehyde with the recycled cata-
[
15]
12
13
lyst 3. Kinetic analyses, ruthenium estimation, and other ex-
perimental data showed that the decrease in activity was due
to an inhibitory equilibrium between m-nitrobenzaldehyde
and 3. The inhibitory equilibrium leads to the formation of
a complex that does not take part in catalysis and causes the
1
4
56
103
[
15]
partial deactivation of 3 after the first batch.
À2
A plausible explanation for the decrease in TON in transfer
hydrogenation can lie in the partial deactivation of 3. It is
therefore proposed that there is an inhibitory equilibrium be-
tween formate ions and 3. The complex formed because of
this equilibrium causes partial deactivation of 3. The saturation
level of the deactivated complex is probably reached after two
batches, and consequently the decrease in TON is negligible in
the fourth and fifth batches.
[a] Reaction conditions: 3 mmol substrate, 30 mg (1.63ꢁ10 mmol Ru)
catalyst, 10 mL water, 30 mmol (1.98 g) HCOONa, T=353 K, t=10 h.
A number of mesoporous silica-supported, ruthenium-con-
taining mixed-metal carbonyl clusters are also highly active
[32]
ring hydrogenation catalysts.
Surprisingly, we find that 3
does not have ring hydrogenation activity, unlike 4, 5, or any
of these reported catalytic systems. Thus, the nearly complete
inertness of 3 toward ring hydrogenation and its high chemo-
selectivity toward ÀCHO reduction in the presence of the
Origin of chemoselectivity
ÀNO group are two of the most significant results presented
2
Ruthenium nanoparticles and/or colloids anchored on various
solid or soluble supports are efficient arene hydrogenation cat-
in this paper. An explanation for the origin of this chemoselec-
tivity is therefore proposed.
[
25–28]
alysts.
Catalytic arene hydrogenation with a homogeneous
Various techniques such as extended X-ray absorption fine
structure have shown that in inorganic oxide-supported car-
bonyl cluster-derived nanocatalysts, the essential integrity of
the original polyhedral structure is retained in the decarbony-
mononuclear ruthenium complex as well as with a cluster
complex was reported. However, in both the systems, finely di-
vided metallic ruthenium was generated under the reaction
[29–31]
[33]
conditions, and that was found to be the true catalyst.
lated form. By using high-angle annular dark-field scanning
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 322 – 330 327

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
TEM and kinetic analyses, we showed that on functionalized
mercial catalyst (5%) Ru/Al
Ruthenium trichlorides (RuCl ·nH O), resublimated iodine, mercu-
O was purchased from Alfa Aesar.
2 3
À
MCM-41-supported [H Ru (CO) ] -derived nanoparticles, the
3
2
3
4
12
[
8]
rous chloride, and carbon tetrachloride were purchased from S D
Fine-Chem Ltd. Safflower oil was taken from a local market. All
other chemicals were of reagent grade and used after proper distil-
lation prior to the catalytic experiments. Double distilled water was
used for the catalytic studies in water or biphasic water medium.
Hydrogen gas cylinder was supplied by BOC Ltd., India.
[H Ru (CO) ] and K[H Ru (CO) ] were synthesized according to the
tetrahedral Ru framework is retained. There is enough evi-
4
dence to show that in solution, subnanometer Rh4 clusters
with partial coordinative unsaturation are relatively robust cat-
alysts. By using in operando extended X-ray absorption fine
structure and kinetic data, Rh clusters in solution have recent-
4
[34]
ly been shown to be active in benzene hydrogenation.
4
4
12
38]
3
4
12
[
In nanocatalysis, the dependence of selectivity of a given re-
action on specific crystal planes, particle size, and nature of the
literature.
[
8,35–37]
stabilizer is well established.
We are unable to record
Instruments and general methods
high-angle annular dark field scanning TEM data of 3; however,
as mentioned earlier, HRTEM images show nearly monodis-
perse crystalline particles with predominantly Ru(111) crystal
planes (Figure 1). As TEM modifies small metal clusters on inor-
ganic supports during the imaging process, it is possible that
Preparation of the cluster-derived catalyst and subsequent manipu-
lation with it were performed by using standard Schlenk tech-
niques under nitrogen atmosphere (if necessary). Solvents were
dried by using standard methods (methanol over magnesium turn-
ings/iodine), distilled under nitrogen, and used immediately. The
starch solution was prepared according to the conventional
the agglomeration of Ru (ꢁ0.5 nm) clusters is brought about
4
by TEM. However, other evidences reported in our earlier work
strongly suggest that owing to the flexibility and folding of
PDADMAC, quasi micelle-type agglomerates of Ru4 are
1
method prior to titration for iodine value determination. H/
13
C NMR spectra were recorded on a 400 MHz Varian spectrometer.
TEM images were captured by using Philips JEM 2000FX and JEOL
JEM 2100F FEG transmission electron microscopes. TEM samples
were dispersed in ethanol by sonication and then a drop was
placed over a hydrophilic carbon-coated copper grid followed by
drying in open air for 2 h. UV/Vis/near-IR measurements were per-
formed on a Perkin–Elmer Lambda 950 spectrophotometer. % Con-
version and distribution of the products were determined either
by using a gas chromatograph equipped with a Shimadzu GC-14A
flame ionization detector by using a CBP20M-25-025 capillary
column and standard commercial samples as references or by
[
15]
formed.
The agglomerates of coordinatively unsaturated tetrahedral
Ru clusters are therefore expected to provide uniform (111)
4
crystal planes. However, if the reduction of ÀNO , ÀCN, or aro-
2
matic ring functionality required crystal planes other than
(111), then 3 would be inert toward these transformations.
Conclusions
1
H NMR analysis. The bulk ruthenium content of catalyst solutions
Ion pairing of a ruthenium carbonyl cluster anion with the qua-
ternary ammonium groups of a water-soluble polymer gives
an efficient ruthenium nanocatalyst (3) for hydrogenation and
transfer hydrogenation reactions. In hydrogenation for repre-
sentative substrates such as safflower oil, higher TONs are ob-
tained in water than in methanol. The cluster-derived catalyst
was determined on a Plasma Lab 8440 ICP-AES instrument. XPS
measurement was performed on a JEOL JPS-9010 MC instrument
(
model 11120-0404) with the MgK source. The sample (catalyst 3)
a
was pressed to form pellets, followed by vacuum drying before
XPS measurement. The spectrum was calibrated with respect to
the C1s peak, which was adjusted to 285 eV. The EDX spectrum
was recorded during TEM studies with a JEOL-JEM-2100F FEG-TEM
instrument.
shows chemoselectivity unique to both catalysts prepared
À
through ion pairing of [RuCl ] with the quaternary ammonium
4
groups of the same polymer and commercial Ru/Al O . The
2
3
Synthesis of catalyst 2
cluster-derived catalyst is totally inert toward the reduction of
the ÀNO group and aromatic rings. By using the cluster-de-
2
Commercially available 20% aqueous solution (2 mL) of PDADMAC
À1
rived catalyst and transfer hydrogenation reactions, industrially
important cyano- and nitrobenzyl alcohols are made from the
corresponding aldehydes. High-resolution TEM data suggest
that unique chemoselectivity is a result of highly crystalline
ruthenium nanoparticles that consist mainly of Ru(111) crystal
planes.
(molecular mass=100000–200000 gmol ) was added dropwise to
a freshly prepared methanolic solution (10 mL) of K[H Ru (CO) ]
3
4
12
(0.128 g, 0.16 mmol), and the mixture was stirred at RT for 24 h.
The volume of the solution was reduced to 5 mL; the solid mass
was precipitated out with acetone, washed thoroughly with ace-
tone, and dried in vacuum. The orange solid material was used as
precatalyst 2 to generate the active catalyst 3 through the hydro-
gen treatment. Although the solubility of 1–3 was checked in dif-
ferent solvents such as DMF, THF, ethanol, and acetonitrile, water
and methanol were found to be the only two solvents in which
they were soluble.
Experimental Section
Materials
À1
PDADMAC (molecular mass=100000–200000 gmol ), 20 wt% so-
lution in water, trirutheniumdodecacarbonyl, methyl pyruvate, sty-
rene, cyclohexanone, acetophenone, cinnamaldehyde, crotonalde-
hyde, cyclopentanone, p-nitrobenzaldehyde, p-cyanobenzaldehyde,
m-vinylbenzaldehyde, and all styrene derivatives were purchased
from Aldrich and used as received. Benzene, phenol, sodium for-
mate, and anisole were obtained from Merck India Ltd. The com-
Synthesis of catalyst 4
Commercially available 20% aqueous solution (3 mL) of PDADMAC
(molecular mass=100000–200000 gmol ) was added dropwise to
an aqueous solution (10 mL) of RuCl ·nH O (100 mg) and stirred at
RT for 24 h. The volume of the solution was reduced under low
À1
3
2
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 322 – 330 328

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
pressure, and the solid mass was precipitated out with acetone.
The collected residue was washed several times with acetone and
dried in vacuum. The solid material was used as precatalyst 4.
Chem. Rev. 2009, 109, 643–710; g) C. J. Li, Chem. Rev. 2005, 105, 3095–
3
165; h) R. A. Sheldon, Green Chem. 2005, 7, 267–278; i) H. U. Blaser, M.
Studer, Green Chem. 2003, 5, 112–117; j) U. M. Lindstrçm, Chem. Rev.
002, 102, 2751–2772; k) F. Joo, Aqueous Organometallic Catalysis,
2
Vol. 23, Kluwer Academic Pub., Dordrecht, 2001; l) Organic Synthesis in
Water (Ed.: P. A. Grieco), Blackie, London, 1998; m) Aqueous-phase Or-
ganometallic Catalysis: Concepts and Applications (Eds.: B. Cornils, W. A.
Herrmann), Wiley-VCH, Weinheim, 1998.
Catalytic hydrogenation reaction
All the hydrogenation reactions were performed in a 200 mL
Teflon-coated stainless steel autoclave fitted with a heat controller
and a magnetic stirrer. Hydrogenated products were identified ini-
tially by using authentic commercial samples of the expected
products.
[
3] a) D. S. Wang, Q. A. Chen, S. M. Lu, Y. G. Zhou, Chem. Rev. 2012, 112,
2557–2590; b) M. J. Beier, J. M. Andanson, T. Mallat, F. Krumeich, A.
Baiker, ACS Catal. 2012, 2, 337–340; c) J. H. Xie, S. F. Zhu, Q. L. Zhou,
Chem. Rev. 2011, 111, 1713–1760; d) P. D. Burton, T. J. Boyle, A. K. Datye,
J. Catal. 2011, 280, 145–154; e) N. Yan, C. Zhao, C. Luo, P. J. Dyson, H.
Liu, Y. Kou, J. Am. Chem. Soc. 2006, 128, 8714–8715.
Catalytic hydrogenation reactions were performed in an autoclave
under a hydrogen pressure of 40 bar at 353 K (or as stated in
Tables 1 and 2) for 4 h with stirring at 900 rpm. A definite amount
of precatalyst 2 or some other catalyst was dissolved in water
[
[
4] A. Robertson, T. Matsumoto, S. Ogo, Dalton Trans. 2011, 40, 10304–
10310.
5] a) J. F. B. Hall, X. Han, M. Poliakoff, R. A. Bourne, M. W. George, Chem.
Commun. 2012, 48, 3073–3075; b) P. J. Debouttiꢃre, Y. Coppel, A. D.
Nowicki, A. Roucoux, B. Chaudret, K. Philippot, Eur. J. Inorg. Chem. 2012,
(
5 mL) in a 10 mL beaker and a substrate (20 mmol) was added to
this solution. Then, the beaker was kept in the autoclave under
a hydrogen pressure of 40 bar at 353 K for 4 h with stirring at
1229–1236; c) Y. Lan, M. Zhang, W. Zhang, L. Yang, Chem. Eur. J. 2009,
1
5, 3670–3673; d) A. Behr, M. Urschey, V. A. Brehme, Green Chem. 2003,
9
00 rpm. After the reaction, the product was extracted with ethyl
5, 198–204.
[6] a) X. P. Fu, L. Liu, D. Wang, Y. J. Chen, C. J. Li, Green Chem. 2011, 13,
1
acetate and then analyzed by using either GC or H NMR.
5
3
49–553; b) V. V. Fokin, K. B. Sharpless, Angew. Chem. 2001, 113, 3563–
565; Angew. Chem. Int. Ed. 2001, 40, 3455–3457; c) N. Jiang, A. J. Ra-
Catalytic transfer hydrogenation reaction
gauskas, Tetrahedron Lett. 2006, 47, 197–200; d) J. C. Morita, H. Nakat-
suji, T. Misaki, Y. Tanabe, Green Chem. 2005, 7, 711–715; e) J. Mao, B.
Wan, F. Wu, S. Lu, Tetrahedron Lett. 2005, 46, 7341–7344; f) X. Xu, T.
Hirao, J. Org. Chem. 2005, 70, 8594–8596; g) S. Narayan, J. Muldoon,
M. G. Finn, V. V. Fokin, H. C. Kolb, K. B. Sharpless, Angew. Chem. 2005,
117, 3339–3343; Angew. Chem. Int. Ed. 2005, 44, 3275–3279; h) J. E.
Klijn, J. B. F. N. Engberts, Nature 2005, 435, 746–747; i) M. D. Hughes, Y.-
J. Xu, P. Jenkins, P. Mcmorn, P. London, D. I. Enache, A. F. Carley, G. A.
Attard, G. J. Hutchings, F. King, E. H. Stitt, P. Johnston, K. Griffin, C. J.
Kiely, Nature 2005, 437, 1132–1135.
Transfer hydrogenation reactions were performed in a 25 mL two-
neck round bottom flask fitted with a condenser and an oil bath at
3
3
53 K for 10 h with stirring (900 rpm). In the actual experiment,
0 mg (1.63ꢁ10 mmol of ruthenium) of precatalyst 2 or an
À2
equivalent amount of other catalysts was dissolved in water
10 mL) with substrate (3 mmol) and HCOONa (30 mmol, 1.98 g).
(
After the reaction, the product was extracted with ethyl acetate
and then analyzed by using either a gas chromatograph or
[
[
[
7] N. Yan, C. Xiao, Y. Kou, Coord. Chem. Rev. 2010, 254, 1179–1218.
8] A. Indra, M. Doble, S. Bhaduri, G. K. Lahiri, ACS Catal. 2011, 1, 511–518.
9] A. Indra, S. Basu, D. G. Kulkarni, C. S. Gopinath, S. Bhaduri, G. K. Lahiri,
Appl. Catal. A 2008, 344, 124–130.
10] S. Basu, M. Mapa, C. S. Gopinath, M. Doble, S. Bhaduri, G. K. Lahiri, J.
Catal. 2006, 239, 154–161.
11] S. Basu, H. Paul, C. S. Gopinath, S. Bhaduri, G. K. Lahiri, J. Catal. 2005,
1
a H NMR spectrometer.
In all cases, the product was separated easily from the aqueous so-
lution of 3 through solvent extraction with ethyl acetate. The aque-
ous catalyst solution was reused immediately for recycling
experiments.
[
[
[
[
2
29, 298–302.
12] S. Bhaduri, G. K. Lahiri, P. Munshi, D. Mukesh, Catal. Lett. 2000, 65, 61–
5.
13] S. Bhaduri, V. S. Darshane, K. Sharma, D. Mukesh, Chem. Commun. 1992,
738–1740.
Iodine values of safflower oil and its hydrogenated product
6
The iodine values of safflower oil and its hydrogenated product
were estimated by using the Hꢂbl method, as reported in the
1
[
[
14] S. Bhaduri, K. R. Sharma, Dalton Trans. 1984, 2309–2313.
15] A. Indra, N. Maity, P. Maity, S. Bhaduri, G. K. Lahiri, J. Catal. 2011, 284,
[39]
literature.
1
76–183.
16] P. Maity, C. S. Gopinath, S. Bhaduri, G. K. Lahiri, Green Chem. 2009, 11,
54–561.
17] P. Maity, S. Basu, S. Bhaduri, G. K. Lahiri, Adv. Synth. Catal. 2007, 349,
955–1962.
18] P. Maity, S. Basu, S. Bhaduri, G. K. Lahiri, J. Mol. Catal. A-Chem. 2007, 270,
17–122.
[
[
[
[
Acknowledgements
5
1
This work was financially supported by Reliance Industries Ltd.,
Mumbai, and Council of Scientific and Industrial Research, New
Delhi, India.
1
19] a) J. Wang, H. F. M. Boelens, M. B. Thathagar, G. Rothenberg, ChemPhys-
Chem 2004, 5, 93–98; b) X. Yan, H. Liu, K. Y. Liew, J. Mater. Chem. 2001,
1
1, 3387–3391.
Keywords: chemoselectivity · hydrogenation · nanocatalyst ·
ruthenium · transfer hydrogenation · water-soluble polymer
[
20] a) M. Y. Jung, Y. L. Ha, J. Agric. Food Chem. 1999, 47, 704–708; b) B. L. A.
Prabhavathi Devi, M. S. L. Karuna, K. N. Rao, P. S. Saiprasad, R. B. N.
Prasad, J. Am. Oil Chem. Soc. 2003, 80, 1003–1005.
[
[
1] M. O. Simon, C. J. Li, Chem. Soc. Rev. 2012, 41, 1415–1427.
2] a) H. Syska, W. A. Herrmann, F. E. Kuehn, J. Organomet. Chem. 2012, 703,
[21] R. N. Butler, A. G. Coyne, Chem. Rev. 2010, 110, 6302–6337.
[22] R. Breslow, Acc. Chem. Res. 1991, 24, 159–164.
5
6–62; b) D. B. Bagal, Z. S. Qureshi, K. P. Dhake, S. R. Khan, B. M. Bhan-
[23] Y. S. Jung, R. A. Marcus, J. Am. Chem. Soc. 2007, 129, 5492–5502.
[24] S. Bhaduri, K. Sharma, D. Mukesh, J. Chem. Soc. Dalton Trans. 1993,
1191–1199.
[25] a) J. Dimroth, U. Schedler, J. Keilitz, R. Haag, R. Schomckera, Adv. Synth.
Catal. 2011, 353, 1335–1344; b) Z. Weng, S. Muratsugu, N. Ishiguro, S.
Ohkoshi, M. Tada, Dalton Trans. 2011, 40, 2338–2347.
age, Green Chem. 2011, 13, 1490–1494; c) G. Liu, H. Gu, Y. Sun, J. Long,
Y. Xu, H. Li, Adv. Synth. Catal. 2011, 353, 1317–1324; d) Y. Q. Sun, G. H.
Liu, H. Y. Gu, T. Z. Huang, Y. L. Zhang, H. X. Li, Chem. Commun. 2011, 47,
2583–2585; e) I. Nieto, M. S. Livings, J. B. Sacci, L. E. Reuther, M. Zeller,
E. T. Papish, Organometallics 2011, 30, 6339–6342; f) K. H. Shaughnessy,
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 322 – 330 329

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[
[
[
26] M. Zahmakıran, Y. Tonbul, S. ꢄzkar, J. Am. Chem. Soc. 2010, 132, 6541–
549.
27] M. Zahmakıran, Y. Tonbul, S. ꢄzkar, Appl. Catal. B-Environ. 2010, 96,
33–540.
28] a) M. Takasaki, Y. Motoyama, K. Higashi, S. H. Yoon, I. Mochida, H. Naga-
shima, Chem. Asian J. 2007, 2, 1524–1533; b) M. Takasaki, Y. Motoyama,
S. H. Yoon, I. Mochida, H. Nagashima, J. Org. Chem. 2007, 72, 10291–
[34] E. Bayram, J. C. Linehan, J. L. Fulton, J. A. S. Roberts, N. K. Szymczak, T. D.
Smurthwaite, S. ꢄzkar, M. Balasubramanian, R. G. Finke, J. Am. Chem.
Soc. 2011, 133, 18889–18902.
[35] E. Schmidt, A. Vargas, T. Mallat, A. Baiker, J. Am. Chem. Soc. 2009, 131,
12358–12367.
6
5
[36] C. A. Witham, W. Huang, C. K. Tsung, J. N. Kuhn, G. A. Somorjai, F. D.
Toste, Nat. Chem. 2010, 2, 36–41.
10293.
[37] D. Y. Murzin, J. Catal. 2010, 276, 85–91.
[
[
[
[
29] G. Sꢂss-Fink, M. Faure, T. R. Ward, Angew. Chem. 2002, 114, 105–108;
[38] J. W. Koepke, J. R. Johnson, S. A. R. Knox, H. D. Kaesz, J. Am. Chem. Soc.
1975, 97, 3947–3972.
Angew. Chem. Int. Ed. 2002, 41, 99–101.
30] C. M. Hagen, L. Vieille-Petit, G. Laurenczy, G. Sꢂss-Fink, R. G. Finke, Orga-
nometallics 2005, 24, 1819–1831.
[39] C. Paquot, A. Hautfenne, Standard Methods for the Analysis of Oils, Fats
th
and Derivatives, 7 ed., Blackwell Scientific Publications Ltd., Oxford,
31] J. A. Widegren, M. A. Bennett, R. G. Finke, J. Am. Chem. Soc. 2003, 125,
1987, pp. 88.
10301–10310.
32] R. Raja, T. Khimyak, J. M. Thomas, S. Hermans, B. F. G. Johnson, Angew.
Chem. 2001, 113, 4774–4778; Angew. Chem. Int. Ed. 2001, 40, 4638–
4642.
Received: July 6, 2012
[
33] a) C. A. Jaska, I. J. Manners, J. Am. Chem. Soc. 2004, 126, 9776–9785;
b) R. Rousseau, G. K. Schenter, J. L. Fulton, J. C. Linehan, M. H. Engelhard,
T. Autrey, J. Am. Chem. Soc. 2009, 131, 10516–10524.
Revised: September 7, 2012
Published online on November 19, 2012
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 322 – 330 330