Nanostructured RuO2 on MWCNTs: Efficient catalyst for transfer hydrogenation of carbonyl compounds and aerial oxidation of alcohols

Applied Catalysis A: General 484 (2014) 84–96

Contents lists available at ScienceDirect

Applied Catalysis A: General

journal homepage: w ww.elsevier.com/locate/apcata

Nanostructured RuO₂on MWCNTs: Efﬁcient catalyst for transfer

hydrogenation of carbonyl compounds and aerial oxidation of alcohols

M. Gopiraman^a, S. Ganesh Babu^a,b, R. Karvembu^c,∗∗, I.S. Kim^a,d,∗

^aNano Fusion Technology Research Lab, Interdisciplinary Graduate School of Science and Technology, National University Corporation, Shinshu University,

Ueda, Nagano 386 8567, Japan

^bSRM Research Institute, SRM University, Kattankulathur 603203, India

^cDepartment of Chemistry, National Institute of Technology, Tiruchirappalli 620015, India

^dInterdisciplinary Cluster for Cutting Edge Research (ICCER), Division of Frontier Fibers, Institute for Fiber Engineering (IFES), Shinshu University, Ueda,

Nagano 386 8567, Japan

a r t i c l e i n f o

a b s t r a c t

Article history:

Multiwall carbon nanotubes (MWCNTs)/ruthenium dioxide nanoparticles (RuO₂NPs) composite was pre-

pared by a straightforward ‘dry synthesis’ method. After being well characterized, the prepared composite

was used as a nanocatalyst (RuO₂/MWCNT) for the transfer hydrogenation of carbonyl compounds. The

excellent adhesion of RuO₂NPs on the anchoring sites of MWCNTs was conﬁrmed by TEM and Raman

analyses. The weight percentage (7.97 wt%) and the chemical state (+4) of Ru in RuO₂/MWCNT was con-

ﬁrmed by EDS and XPS analyses, respectively. It was found that the RuO₂/MWCNT has a higher speciﬁc

surface area of 189.3 m²g⁻¹. Initially the reaction conditions were optimized and then the scope of the

catalytic system was extended with a wide range of carbonyl compounds. The inﬂuence of the size of

RuO₂NPs on the transfer hydrogenation of carbonyl compounds was also studied. The RuO₂/MWCNT is

highly chemoselective, heterogeneous in nature, reusable and highly stable. Owing to the high stabil-

ity of the used catalyst (u-RuO₂/MWCNT), it was further calcinated at high temperature to obtain RuO₂

nanorods (NRs) hybrid MWCNTs. Then the hybrid material was used as a catalyst (r-RuO₂/MWCNT) for

the aerial oxidation of alcohols and the result was found to be good.

Received 31 December 2013

Received in revised form 23 June 2014

Accepted 26 June 2014

Available online 7 July 2014

Keywords:

MWCNTs

RuO₂nanostructures

Dry synthesis

Heterogeneous catalyst

Transfer hydrogenation

Aerial oxidation

1. Introduction

homogeneous catalysts due to their simple recovery and reusabil-

ity [6,7]. Particularly, supported RuNPs catalysts have been widely

Catalytic reduction of carbonyl compounds (C O) to their cor-

responding alcohols is an essential organic transformation in both

industrial and ﬁne chemical processes [1,2]. Several metal-based

catalytic systems have long been proposed to perform this trans-

formation [3,4]. However, due to simple and cost effective protocol,

most of the recent studies are focused on the transition metal-

catalyzed transfer hydrogenation of carbonyl compounds by using

2-propanol (i-PrOH) as a hydrogen donor [5]. From an econom-

ical point of view, heterogeneous metal nanoparticles (MNPs)

including ruthenium nanoparticles (RuNPs) have advantages over

employed for the reduction of carbonyl compounds due to their

high activity, selectivity, versatility and reusability [8–10]. Kantam

et al., [11] investigated MgO stabilized RuNPs catalyst for the trans-

fer hydrogenation of carbonyl compounds. Yamaguchi et al., [12]

prepared highly dispersed Ru(OH)_x/TiO₂composite and used as

a catalyst for the liquid-phase hydrogen transfer reactions. They

found that the Ru(OH)_x/TiO₂catalyst is highly effective and selec-

tive. In spite of higher catalytic activity, most of the common RuNPs

(supported on metal oxide and polymeric materials) often exhibit

less stability in high basic and acidic reaction conditions and, con-

sequently, the reusability and selectivity of the catalyst are highly

limited [13]. Therefore, developing an efﬁcient and stable catalyst

for the transfer hydrogenation of carbonyl compounds is still a

challenging task.

∗

Corresponding author at: Shinshu University, Interdisciplinary Graduate School

of Science and Technology, Nano Fusion Technology Research Lab, National Univer-

sity Corporation, Ueda, Nagano 386 8567, Japan. Tel.: +81 268 21 5139;

fax: +81 268 21 5482.

Corresponding author at: National Institute of Technology, Department of Chem-

istry, Tiruchirappalli 620015, India. Tel.: +91 431 2503636; fax: +91 431 2500133.

Among the carbon building blocks (CNTs, fullerenes, graphene

and carbon nanoﬁbers), multiwall carbon nanotubes (MWCNTs)

have played a signiﬁcant role as a support in various ﬁelds including

catalysis [14,15]. In fact, MWCNTs have unique and superior prop-

erties such as chemical inertness, high stability and huge surface

∗∗

E-mail addresses: kar@nitt.edu (R. Karvembu), kim@shinshu-u.ac.jp (I.S. Kim).

http://dx.doi.org/10.1016/j.apcata.2014.06.032

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

85

area [16]. Till date, several MWCNTs-supported MNPs catalysts

have been proposed. Particularly, MWCNTs-supported RuNPs have

shown more versatility in carrying out the selective catalytic pro-

cesses. Recently, Yu et al., [17] investigated the RuO₂·xH₂O/CNT

nanocatalysts for the aerobic oxidation of benzyl alcohol. Yang and

co-workers [18] have prepared the RuNPs/MWCNT composite by

‘wet synthesis’ method and employed as a nanocatalyst for the oxi-

dation of alcohols. They found that the RuNPs/MWCNT is highly

effective, stable and reusable. However, in the common ‘wet syn-

thesis’ method, several factors such as solvent, concentration of

metal precursor, reducing agent, time and temperature need to

be controlled carefully to obtain very good adhesion and homoge-

neous distribution of MNPs on MWCNTs [19–22]. Very recently, the

solventless bulk synthesis so called ‘dry synthesis’ has been attract-

ing greater interest due to its very simple protocol, better adhesion

of MNPs on carbon materials, and has an advantage of least param-

eters to control [20]. In our very recent course of investigation,

we prepared CuO/MWCNT [23], GNP-RuNRs [24], GNS–RuNPs [25]

and RuO₂/SWCNT [26] by a simple ‘dry synthesis’ method. Also, we

found that the resultant composites are highly active as nanocata-

lysts in the various organic transformations. Encouraged by these

results, we presumed that the RuO₂/MWCNT composite prepared

by ‘dry synthesis’ method would also exhibit a good catalytic activ-

ity for the reduction of carbonyl compounds. Herein, we report the

‘dry synthesis’ of RuO₂/MWCNT and its catalytic activity towards

reduction of carbonyl compounds. Chemoselectivity, heterogene-

ity and stability of the RuO₂/MWCNT were also examined. The

used catalyst was separated out from the reaction and reused in

the same reaction as well as in the aerial oxidation of alcohols.

The selective oxidation of alcohols is one of the very essential

transformations in organic synthesis and also the oxygenated prod-

ucts are extremely valuable in chemical industries [27–29]. Mainly,

Ru-catalyzed aerial oxidation of alcohols has been attracting a

great deal of attention due to its higher activity and selectivity

[30–32].

the RuO₂/MWCNT was measured by BET surface area analyzer

(Micromeritics–Pulse Chemisorb 2700). Prior to the measurement,

the sample was degassed for 2 h in N₂atmosphere at 200 ^◦C by using

the degassing unit [Micromeritics–Desorb 2300 A]. Shimadzu-2010

gas chromatograph (GC) was used to analyze the reaction mix-

ture.

2.2. Dry synthesis of RuO₂/MWCNT

In a typical experiment, 0.5 g of pure MWCNTs were chemi-

cally treated with a 3:1 mixture of conc. H₂SO₄and conc. HNO₃,

and then the mixture was sonicated at 40 ^◦C for 3 h in ultrasonic

bath. After cooling to 25 ^◦C, the solution was diluted with 500 mL

of deionized water and then vacuum-ﬁltered through a ﬁlter paper

of 0.65 ␮m porosity. The resultant solid mass (f-MWCNTs) was

frequently washed with deionized water until the pH became neu-

tral and then dried in vacuo at 60 ^◦C. Then, 0.13 g of Ru(acac)₃

was added into 0.5 g of f-MWCNTs and mixed well by a mortar

and pestle. The homogeneous mixture of f-MWCNTs and Ru(acac)₃

was obtained in 13–15 min. Finally, the mixture was calcinated

under N₂atmosphere at 350 ^◦C for 3 h in a mufﬂe furnace. Fig. 1

shows schematic illustration of the procedure for the preparation

of RuO₂/MWCNT.

2.3. Transfer hydrogenation of carbonyl compounds

In a typical procedure, a 5 mg (0.77 mol%) of RuO₂/MWCNT and

80 mg (2 mmol) of NaOH were stirred with 5 mL of i-PrOH taken in

an ace pressure tube equipped with a stirring bar. Then the sub-

strate (1 mmol) was added to the stirring solution and then the

mixture was heated at 82 ^◦C. The completion of the reaction was

monitored by GC. After the reaction, the catalyst was separated out

from the reaction mixture by simple centrifugation and the prod-

ucts and unconverted reactants were analyzed by GC without any

puriﬁcation. Selectivity of the product for each reaction was also

calculated. Finally, the separated RuO₂/MWCNT was washed well

with diethyl ether followed by drying in an oven at 60 ^◦C for 5 h and

it was reused for the subsequent transfer hydrogenation of carbonyl

compounds to investigate the reusability of the RuO₂/MWCNT.

2. Experimental

2.1. Materials and characterization

High purity MWCNTs with diameter ranging from 15 to 20 nm

were used. The MWCNTs were produced in large scale through

the optimal combination of chemical vapor deposition synthetic

method, and subsequent thermal treatment at 2800 ^◦C in an argon

atmosphere [33]. Ru(acac)₃(97%), H₂SO₄(98%), HNO₃(70%) and

HCl (70%) were purchased from Wako pure chemicals, Japan. All

other chemicals were purchased from Aldrich and used as received.

The surface morphology of the RuO₂/MWCNT was investigated

by TEM (JEM-2100 JEOL Japan) with accelerating voltage of 120 kV.

The weight percentage and homogeneous distribution of RuO₂NPs

in the RuO₂/MWCNT were conﬁrmed by SEM–EDS [Hitachi (model-

3000H) scanning electron microscope]. The same ﬁeld of view

was then scanned using an EDS spectrometer to acquire a set

of X-ray maps for Ru, C and O using 1 ms point acquisition for

approximately one million counts. The interaction of RuO₂NPs

with MWCNTs was examined by Raman spectrometer (Hololab

5000, Kaiser Optical Systems Inc., USA) using argon laser at 532 nm

with a Kaiser holographic edge ﬁlter. WAXD experiments were

performed at room temperature using a Rotaﬂex RTP300 (Rigaku

Co., Japan) diffractometer at 50 kV and 200 mA. Nickel-ﬁltered

Cu K␣ radiation (10 < 2ꢀ < 80^◦) was used for the measurements.

To conﬁrm the chemical state of Ru in the RuO₂/MWCNT, XPS

spectrum was recorded in Kratos Axis-Ultra DLD model instru-

ment. Before the XPS analysis, the sample (RuO₂/MWCNT) was

irradiated under Mg K␣ ray source. The speciﬁc surface area of

2.4. Aerial oxidation of alcohols

Five milligram of r-RuO₂/MWCNT (0.68 mol%) was stirred with

3 mL of toluene taken in a round-bottomed ﬂask equipped with

a condenser and a magnetic stirrer. The substrate (1 mmol) was

added to the stirring solution and then the mixture was reﬂuxed

at 110 ^◦C under atmospheric pressure of air. The completion of the

reaction was checked by GC. After the reaction, the r-RuO₂/MWCNT

was separated out from the reaction mixture by simple centrifuga-

tion and the products and unconverted reactants were analyzed

by GC without any puriﬁcation. Selectivity of the product for each

reaction was also calculated.

2.5. Product conﬁrmation and analysis

In order to conﬁrm the formation of the product, samples of both

reactant and products were dissolved in ethyl acetate and then ana-

lyzed by GC. GC was equipped with 5% diphenyl and 95% dimethyl

siloxane, Restek-5 capillary column (0.32 mm dia, 60 m in length)

and a ﬂame ionization detector (FID). N₂was used as a carrier gas.

The initial column temperature was increased from 60 to 150 ^◦C at

the rate of 10 ^◦C/min and then to 220 ^◦C at the rate of 40 ^◦C/min. Dur-

ing the product analysis, the temperatures of the FID and injection

port were kept constant at 150 and 250 ^◦C, respectively.

86

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

Fig. 1. Schematic illustration for the preparation of RuO₂/MWCNT.

3. Results and discussion

In addition, the surface area per unit mass (S) of RuO₂NPs is cal-

culated to be 478.1 m²g⁻¹based on the equation S = 6000/(ꢁ × d)

[34], where d is the mean diameter, and ꢁ is the density of RuO₂

(6.97 g cm⁻³).

3.1. Characterization of RuO₂/MWCNT

The surface morphology of the RuO₂/MWCNT was investigated

in detail by using TEM analysis. Fig. 2 shows the TEM images

[Fig. 2(i–v)] of RuO₂/MWCNT and the particle size-distribution

histogram of RuO₂NPs [Fig. 2(vi)] in the RuO₂/MWCNT. As can

be seen from Fig. 2(i–v), ultra ﬁne and homogeneously dispersed

RuO₂NPs were externally attached on the surface of MWCNTs with

a very narrow particle size distribution. The size distribution his-

togram of RuO₂NPs [Fig. 2(vi)] conﬁrmed that the diameter of

RuO₂NPs ranges from 0.5 to 4 nm, with a mean diameter of 1.8 nm.

Worth mentioning that there were no free RuO₂NPs observed

in the background of the TEM images [Fig. 2(iv) and (v)], which

showed a complete utilization of the RuO₂NPs. Moreover, FE-

SEM images (Fig. S1 in Supporting information) were also taken

for the RuO₂/MWCNT and found that a very small RuO₂NPs are

well dispersed and externally attached on the surface of MWC-

NTs. Subsequently, the speciﬁc surface area of the RuO₂/MWCNT

was determined by BET analysis. Interestingly, a high speciﬁc sur-

face area of 189.3 m²g⁻¹with a pore volume of 0.898 cm³g⁻¹and

In order to determine the weight percentage of Ru in

RuO₂/MWCNT and to inspect the homogeneous distribution of

RuO₂NPs on MWCNTs, SEM–EDS and their corresponding elemen-

tal mapping images were taken for RuO₂/MWCNT (Fig. 3). The

weight percentage of Ru in RuO₂/MWCNT was 7.97 wt%, as deter-

mined by EDS analysis [Fig. 3(ii)]. Fig. 3(iv) and (v) shows that the

distribution of RuO₂NPs in RuO₂/MWCNT was homogeneous. As

seen from Fig. 3(iii)–(v), the RuO₂/MWCNT contains carbon, ruthe-

nium and oxygen elements only; it indicates that the proposed

method (dry synthesis) is reliable and effective.

XPS spectra were recorded for MWCNTs (a), f-MWCNTs (b)

and RuO₂/MWCNT (c) to study the formation of oxygen functional

groups on MWCNTs, and the chemical state of Ru in RuO₂/MWCNT

(Figs. 4 and 5). As expected, all the three samples (a–c) demon-

strated a C 1s peak and O 1s peak at 284.4 and 532.5 eV, respectively

(Fig. S2 in Supporting information). Referring Fig. 4(i), the bind-

ing energy (BE) of the C C and C H bonds was observed at

284.5–285 eV and the peaks at 285.1, 285.5, 286.6 and 288.5 eV

˚

a BJH desorption average pore diameter of 186.98 A was found.

were attributed to C OH,

C O C , C O and COOH groups,

Fig. 2. ((i)–(v)) TEM images of RuO₂/MWCNT and (vi) the particle size distribution of RuO₂NPs.

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

87

Fig. 3. (i) SEM image and (ii) EDS spectrum of RuO₂/MWCNT, and corresponding EDS mapping of (iii) C, (iv) Ru and (v) O.

Fig. 4. XPS spectra of f-MWCNTs; magniﬁed (i) C 1s and (ii) O 1s peaks.

respectively [35]. Moreover, the deconvolution of O 1s spectra

of f-MWCNTs [Fig. 4(ii)] resulted in ﬁve peaks located at 529.6,

530.7, 531.5, 532.2 and 533.5 eV, which were assigned to the C O,

the surface of MWCNTs. The two main reasons for the function-

alization of MWCNTs are; (i) to make the MWCNTs hydrophilic

for the homogenous decoration of RuO₂NPs on MWCNTs, and (ii)

to create additional nucleation centers for the better adhesion of

RuO₂NPs on MWCNTs. Moreover, these functional groups play a

COOH, C OH,

C

O

C

and H₂O, respectively [36]. This result

conﬁrmed the successful creation of oxygen functional groups on

Fig. 5. XPS spectra of RuO₂/MWCNT; magniﬁed (i) C 1s and (ii) Ru 3p peaks.

88

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

Fig. 6. (i) XRD pattern of RuO₂/MWCNT and (ii) Raman spectra of f-MWCNTs (a) and RuO₂/MWCNT (b).

bridging role between the RuO₂NPs and MWCNTs; consequently,

good dispersion and strong attachment of RuO₂NPs on MWCNTs

were achieved [37]. The XPS spectrum of the RuO₂/MWCNT in Ru

3p region [Fig. 5(ii)] showed BE of Ru 3p_3/2at 462.5 eV and Ru 3p_1/2

at 485.2 eV, which corresponded to the photoemission from RuO₂

(Ru⁴⁺) [37]. In Fig. 5 (i), the XPS spectrum of the RuO₂/MWCNT show

Ru 3d_5/2peak at 280.8 eV which is also attributed to the photoemi-

ssion from RuO₂[37]. The overlapping of the C 1s and the Ru 3d_3/2

peaks at ∼285 eV makes it complicated to assign the BE of Ru 3d_3/2

[Fig. 5(i)]. Fig. 6 (i) shows the XRD patterns of RuO₂/MWCNT. The

diffraction peaks were observed at 26.5^◦and 42.4^◦, corresponding

to the (0 0 2) and (1 0 0) crystal planes of MWCNTs, respectively,

which attributed to the hexagonal graphite structures of MWCNTs

[24]. In addition to that, RuO₂/MWCNT showed signature patterns

at 27.5^◦, 34.9^◦, 39.9^◦and 54.5^◦corresponding to the typical crys-

tal faces (1 1 0), (1 0 1), (2 0 0) and (2 2 0) of RuO₂(JCPDS 21-1172),

respectively [24]. The broadness of the diffraction peaks conﬁrmed

that the RuO₂NPs in RuO₂/MWCNT are in nanocrystalline nature.

The nature of interaction between RuO₂NPs and MWCNTs was

investigated by comparing the Raman spectra of f-MWCNTs (a) and

RuO₂/MWCNT (b) [Fig. 6(ii)]. As expected, samples a and b showed

two characteristic peaks at 1345 and 1580 cm⁻¹, corresponding

to the sp³- and sp²-hybridized carbons which authenticated the

disordered graphite (D band) and the ordered state graphite (G

band) of MWCNTs [38]. Well known that, the ratio of D and G

bands (I_D/I_G) intensities is often used as a diagnostic tool to eval-

uate the defect concentration in MWCNTs. Therefore, I_D/I_Gwas

calculated for the samples a and b [Fig. 6(ii)]. The data revealed

that RuO₂NPs were strongly attached on the surface of MWCNTs

as the I_D/I_Gratio was reasonably high for RuO₂/MWCNT (1.4579)

in comparison to that of f-MWCNTs (1.4113). In addition, negative

shift was also observed in the D band (1347 to 1341 cm⁻¹) and G

band (1580 to 1576 cm⁻¹) for RuO₂/MWCNT, which indicates that

RuO₂NPs interacted strongly on the surface of the MWCNTs [39].

The absence of peaks around 1700 cm⁻¹in the Raman spectra of

a and b suggested that the present process produces fairly pure

MWCNTs.

acetophenone was investigated with different bases such as KOH,

NaOH, K₂CO₃or (CH₃)₃COK. Among them, NaOH was found to

be the efﬁcient since it gave an excellent yield of 96% (Table 1,

entries 1–4). The amount of base also played a crucial role in the

present RuO₂/MWCNT system (Table 1 , entries 2, 5–7); a 2 mmol

of base was required to achieve the excellent yield (Table 1, entry

2). Similarly, the amount of catalyst was optimized; as a con-

sequence, a very lower conversion of 13% was obtained in the

absence of RuO₂/MWCNT (Table 1, entry 8). Whereas addition of

5.0 mg (0.71 mol%) of RuO₂/MWCNT yielded 96% of the product

(Table 1, entry 2). Lesser amount of catalyst (2.5 mg; 0.24 mol%)

was ineffective (27%) (Table 1, entry 9) which may be due to the

presence of insufﬁcient number of active sites (RuO₂) for the sub-

strates. Furthermore, increase of the amount of the catalyst (7.5

and 10 mg; 0.48 and 0.96 mol%) showed no signiﬁcant change in

the yield (Table 1, entries 10–11). Hence, 5.0 mg (0.77 mol%) was

found to be the optimum amount of the catalyst. Subsequently,

reaction temperature was optimized. At the lower reaction tem-

peratures (27, 50 and 70 ^◦C), the reactions were very slow, which

gave very poor yield (Table 1, entries 12–14). But excellent yield

was obtained (96%) when the reaction is stirred at the temperature

of 82 ^◦C (Table 1, entry 2). This might be due to faster adsorption

of the substrates on RuO₂active sites at 82 ^◦C. Progress of the reac-

tion was monitored at every 10 min time interval (Table 1, entries

15–23). No further increase in the yield after 45 min concluded that

45 min is the optimum reaction time (Table 1, entry 2). Worth men-

tioning that under the optimized reaction conditions, the present

RuO₂/MWCNT system achieved an excellent yield of 96% with a

good turnover number of 125 and turnover frequency of 167 h⁻¹

.

The optimized reaction conditions were adopted to extend the sub-

strate scope.

3.3. Extension of scope

Benzaldehyde, substituted benzaldehydes and other substi-

tuted aldehydes were converted into the corresponding primary

alcohols in excellent to moderate yields (Table 2, entries 1–10).

The yield of the products was moderately affected by the sub-

stituent on the aromatic ring, but the selectivity was maintained.

Initially, benzaldehyde was converted into 95% of benzyl alco-

hol with 100% selectivity after 45 min (Table 2 , entry 1). But

HSi(OMe)₃/LiOMe catalytic system yielded only 85% of benzyl

alcohol even after 20 h [40]. Electron withdrawing substituent

in the phenyl ring retarded the reactivity. In evidence to this,

4-bromo (Table 2, entry 2), 2-bromo (Table 2, entry 3), 4-chloro

3.2. Optimization of reaction conditions

In order to get an effective conversion of carbonyl com-

pounds to alcohols, reaction parameters such as base, amount

of base, amount of catalyst, temperature and reaction time were

optimized (Table 1). The conversion of acetophenone (model sub-

strate) was checked through GC. At ﬁrst, transfer hydrogenation of

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

89

Table 1

Optimization of the reaction conditions for the transfer hydrogenation of acetophenone^a.

O

OH

RuO₂/MWCNT

.

c

Entry

Base

Amount of base (mmol)

Amount of catalyst (mol%)

Temperature (^◦C)

Time (min)

Yield^b(%)

TON/(TOF h⁻¹

)

1

2

3

4

5

6

7

8

KOH

2

0.77

0

0.24

0.48

0.96

0.77

82

25

50

70

82

45

0

10

20

30

40

60

70

80

90

26

96

15

39

41

56

96

11

27

49

96

15

37

52

0

29

45

59

78

96

97

34/45

125/167

20/27

51/68

53/71

73/97

125/167

–

112/149

102/136

100/133

20/27

48/64

68/91

–

38/224

58/77

NaOH

K₂CO₃

(CH₃)₃COK

NaOH

1

1.5

2.5

2

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

77/154

101/151

125/125

125/107

126/95

126/84

2

a

All the reactions were performed with 1.0 mmol (117.0 ␮L) of acetophenone. A 5 mL aliquot of i-PrOH was used in all the reactions.

GC yield.

TON/TOF [TON = the amount of product (mol)/the amount of active sites; TOF = TON/time (h)].

b

c

(Table 2, entry 4) and 4-nitro benzaldehydes (Table 2, entry 6)

4-bromo-1-phenylethanol and 92% of 4-methoxy-1-phenylethanol

only after 48 and 36 h, respectively [42]. Acetophenones having

substituent at the ortho, meta or para position were also subjected

to transfer hydrogenation. 3-Nitroacetophenone (Table 3, entry 4),

2-methylacetopheone (Table 3, entry 5) and 4-methylacetophone

(Table 3, entry 6) produced corresponding 1-phenylethanol with

excellent conversion, selectivity and yield. Particularly, in the

reduction of 2-methylacetopheone to 1-(o-tolyl)ethanol (Table 3,

entry 5), the present catalytic system gave a better yield of 95%

were found to be less reactive compared to unsubstituted benz-

aldehyde in the present catalytic system. In contrast, the electron

releasing substituent increased the reactivity. For example, 4-

methylbenzaldehyde was hydrogenated to form 4-methylbenzyl

alcohol in 97% yield with 100% selectivity (Table 2, entry 5).

Aldehydes with bulky substituent namely 2-naphthaldehyde and

1-pyrenecarboxaldehyde were reduced to corresponding primary

alcohols with lesser selectivity (82 and 80%). The former one was

converted into 2-naphthalenemethanol (51%) after 45 min (Table 2,

entry 7), whereas the later required 120 min to achieve 54% of

the corresponding reduced product (Table 2 , entry 8). 100% selec-

tivity was achieved when 3-phenylpropanal was hydrogenated to

3-phenylpropan-1-ol after 120 min (Table 2 , entry 9). N,N-dimethyl

benzaldehyde yielded 75% of N,N-dimethylbenzyl alcohol after

45 min with 95% selectivity (Table 2, entry 10).

After the successful reduction of aldehydes to primary alco-

hols, ketones were converted into secondary alcohols under the

same reaction conditions (Table 2, entries 1–15). Acetophenone

was hydrogenated to 1-phenylethanol (96%) with 100% selectivity

after 45 min (Table 3, entry 1). But HSi(OMe)₃/LiOCMe₂CMe₂OLi

[40] and Au/TiO₂catalytic systems [41] required 20 and 4 h,

respectively, for the complete reduction of acetophenone. In con-

trast to the reduction of benzaldehydes, acetophenones bearing

electron withdrawing or electron releasing substituent in the

phenyl ring yielded the corresponding substituted 1-phenylethanol

more readily. In witness to this statement, 4-bromoacetophenone

and 4-methoxyactophenone were converted into their corre-

sponding secondary alcohols namely 4-bromo-1-phenylethanol

(Table 3, entry 2) and 4-methoxy-1-phenylethanol (Table 3, entry

3) in 92 and 88% yield, respectively, with excellent selectivity

after 45 min. But RuCs-␤(IMP-carbonyl) catalyst yielded 98% of

(100% selectivity) with an excellent TON (123) and TOF (148 h⁻¹

)

values after only 50 min. Whereas Ru(OH)_x/TiO₂system gave

82% (with TON/TOF of 82/27 h⁻¹) of the same product only after

3 h even at 90 ^◦C under Ar atmosphere [12]. Substituted ben-

zophenone speciﬁcally 4-methoxybenzophenone was reduced to

4-methoxydiphenylmethanol (73%) after 45 min (Table 3, entry

7); moderate yield of the product may be due to the pres-

ence of the bulky substituent [43]. Scope of this system was

also extended to aliphatic and alicyclic carbonyl compounds.

Aliphatic ketones yielded the corresponding alcohols in good to

excellent yields. Almost 100% selectivity was achieved in all the

cases with the yield ranging from 95 to 98% (Table 3, entries

8–10). Notably, in the reduction of less reactive 2-butanone to

2-butanol (Table 3, entry 8), the present RuO₂/MWCNT system

showed a better yield of 95% (98% selectivity) when compared

to GNPs-RuO₂NRs system [24]. The better yield is due to the

higher surface area of RuO₂/MWCNT (189.3 m²g⁻¹) than the GNPs-

RuO₂NRs (65.17 m²g⁻¹). Similarly, 2-octanone gave 2-octanol in

an excellent yield of 98% (100% selectivity) just after 1 h (Table 3,

entry 11) whereas Ru(OH)_x/TiO₂system exhibited 92% yield even

after 2 h [12]. In contrast to Fe@SiO₂Ru nanocatalyst [44], ali-

cyclic alcohols with different ring size were produced from the

alicyclic ketones with 100% selectivity, but they took longer

90

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

Table 2

RuO₂/MWCNT-catalyzed transfer hydrogenation of aldehydes^a.

Entry

Substrate

Product

Time (min)

45

Conv.^b(%)

Sel.^b(%)

100

Yield^b(%)

TON (TOF h⁻¹

123/160

)

OH

O

1

95

OH

O

Br

2

3

45

88

76

91

79

76

103/134

99/129

O

100

Cl

4

5

45

63

97

88

51

97

66/86

OH

O

100

126/167

O

OH

O₂N

6

7

45

79

69

100

82

79

51

103/134

66/86

O

OH

O

8

9

120

74

89

80

54

89

70/35

OH

O

120

100

116/58

N

10

45

80

95

75

97/126

a

Reaction conditions: Substrate (1 mmol), RuO₂/MWCNT (0.77 mol%), NaOH (2 mmol), i-PrOH (5 mL), 82 ^◦C.

Determined by GC analysis.

b

reaction times. As the ring size increases, decreased reactivity was

respectively, with very good selectivity (Table 4, entries 3 and

4). In the same way 4-(2-pyridyl)benzaldehyde gave 77% of 4-(2-

pyridyl)benzyl alcohol with 100% selectivity after 100 min (Table 4,

entry 5). Worth mentioning that RuO₂/MWCNT system achieved

good yields (51 to 98%) with high TON (127 to 66) and TOF (246

to 35 h⁻¹) values in the transfer hydrogenation of carbonyl com-

pounds.

observed (Table 3, entries 12–14). The reactivity order is cyclohex-

anone (92%) > cycloheptanone (88%) > cyclododecanone (77%). But

nickel/aluminosilicate catalyst yielded only 58% of cyclohexanone

under similar reaction conditions [45]. 100% Selectivity was accom-

plished when carbonyl functionality was present in the side chain

of the alicyclic ring (Table 3, entry 15).

In order to illustrate the synthetic utility of the present catalytic

system, some heterocyclic aldehydes and ketones were subjected

to transfer hydrogenation (Table 4). 2-Acetylthiophene yielded the

corresponding alcohol (97%) after 90 min with 100% selectivity

(Table 4, entry 1). RhCl(PPh₃)₃catalytic system gave only 4% of the

same product [46]. Likewise, the present system was found to be

effective in the reduction of 2-furfuraldehyde which gave 74% of the

product (Table 4, entry 2). Only 50% of the product was achieved

from the reduction of 2-furfuraldehyde with Pt decorated Al₂O₃

catalyst [47]. 3-Formylindol and 5-formylindol were effectively

reduced to indole-3-methanol (88%) and indole-5-methanol (96%),

3.4. Chemoselectivity of RuO₂/MWCNT

Referring Table 5, the present RuO₂/MWCNT catalytic

system is highly chemoselective in nature. 4-Acetylbenzald-

ehyde was chemoselectively reduced to 1-(4-(hydroxymethyl)

phenyl)ethanone (Table 5, entry 1). The phenyl ring containing

aldehyde and ester functional groups was subjected to transfer

hydrogenation and only the aldehyde was selectively reduced to

primary alcohol without affecting the ester functionality (Table 5,

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

91

Table 3

RuO₂/MWCNT-catalyzed transfer hydrogenation of ketones^a.

Entry

Substrate

Product

Time (min)

Conv.^b(%)

Sel.^b(%)

Yield^b(%)

TON (TOF h⁻¹

)

O

OH

1

45

96

100

96

125/167

O

OH

Br

2

3

45

95

88

97

92

88

120/160

114/152

O

OH

H₃CO

O₂N

H₃CO

O₂N

100

OH

O

4

5

6

45

50

45

81

95

91

86

100

67

95

91

87/116

123/148

118/157

OH

O

OH

O

OH

O

7

8

9

45

30

45

86

97

87

98

73

95

95/127

123/246

123/164

O

OH

O

10

11

45

60

97

98

100

97

98

126/168

127/127

O

OH

O

12

13

90

96

88

96

92

88

120/80

114/76

O

OH

100

OH

14

120

77

100

77

100/50

92

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

Table 3 (Continued)

Entry

Substrate

Product

Time (min)

Conv.^b(%)

Sel.^b(%)

Yield^b(%)

TON (TOF h⁻¹

)

OH

O

15

90

74

100

74

96/64

a

Reaction conditions: substrate (1 mmol), RuO₂/MWCNT (0.77 mol%), NaOH (2 mmol), i-PrOH (5 mL), 82 ^◦C.

Determined by GC analysis.

b

Table 4

RuO₂/MWCNT-catalyzed transfer hydrogenation of heterocyclic carbonyl compounds^a.

Entry

Substrate

Product

Time (min)

Conv.^b(%)

Sel.^b(%)

Yield^b(%)

TON (TOF h⁻¹

)

S

O

OH

1

2

90

97

78

100

96

97

74

126/84

96/64

H

O

OH

H

N

O

OH

3

4

120

100

95

96

93

88

96

114/57

125/75

H

N

H

N

O

N

HO

N

100

OH

O

5

100

77

100

77

100/60

a

Reaction conditions: Substrate (1 mmol), RuO₂/MWCNT (0.77 mol%), NaOH (2 mmol), i-PrOH (5 mL), 82 ^◦C.

Determined by GC analysis.

b

Table 5

Chemoselective transfer hydrogenation of carbonyl compounds catalyzed by RuO₂/MWCNT^a.

Entry

Substrate

Product

Time (min)

Conv.^b(%)

Sel.^b(%)

Yield^b(%)

TON (TOF h⁻¹

)

O

OH

O

1

2

45

80

79

84

64

79

83/111

O

OH

H₃COOC

100

103/137

O

OH

HOOC

3

60

77

100

77

100/100

a

Reaction conditions: substrate (1 mmol), RuO₂/MWCNT (0.77 mol%), NaOH (2 mmol), i-PrOH (5 mL), 82 ^◦C.

Determined by GC analysis.

b

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

93

Fig. 7. (i and ii) TEM images of RuO₂/MWCNT, (iii) EDS spectrum of RuO₂/MWCNT and (iv) Ru 3p peaks of RuO₂/MWCNT.

entry 2). Similar trend was observed with the phenyl ring having

ketone and acid functional groups (Table 5, entry 3).

b-RuO₂/MWCNT catalyst (Table 6, entry 3). Same kind of behavior

was observed with benzaldehyde (Table 6, entry 4) and heterocyclic

carbonyl compound (Table 6, entry 5). Hence, it is inferred that the

excellent activity of RuO₂/MWCNT compared to b-RuO₂/MWCNT

was mainly due to the ultra-ﬁne structure of the RuO₂NPs.

3.5. Effect of particle size on catalytic activity

Particle size and surface area are important factors in any

nanocatalytic system. In order to determine the effect of the RuO₂

particle size on catalytic efﬁciency in terms of yields, another cata-

lyst with RuO₂particle size of around 5 to 12 nm (b-RuO₂/MWCNT)

was prepared by calcinating the mixture [f-MWCNT and Ru(acac)₃]

under air atmosphere at 375 ^◦C for 3 h and its catalytic performance

was compared with the RuO₂/MWCNT. At ﬁrst, b-RuO₂/MWCNT

nanocatalyst was characterized by TEM analysis. It was found that

the particle size of RuO₂NPs was around 5 to 12 nm with the mean

diameter of 9 nm [Fig. 7(i) and (ii)]. The weight percentage of Ru in

b-RuO₂/MWCNT was found to be 8.01 by using SEM–EDS analysis

[Fig. 7(iii)]. XPS analysis revealed that Ru was present as RuO₂and

hence it showed 3p_1/2peak at 485 eV and 3p_3/2peak at 462 eV. The

surface area per unit mass (S) of b-RuO₂/MWCNT was found to be

88.6 m²g⁻¹. After the characterization, b-RuO₂/MWCNT was used

as a catalyst for the transfer hydrogenation of some carbonyl com-

pounds (Table 6). Acetophenone was reduced to 1-phenylethanol in

moderate yield of 65% (Table 6, entry 1) whereas the RuO₂/MWCNT

system gave an excellent yield (96%) of the desired product (Table 3,

entry 1). The reactivity of aliphatic carbonyl compounds was found

to be lesser in the b-RuO₂/MWCNT system (Table 6, entry 2)

when compared to RuO₂/MWCNT system. Similarly, reduction of

alicyclic ketone produced lower yield of alicyclic alcohol with

3.6. Heterogeneity and reusability of RuO₂/MWCNT

In order to understand the heterogeneous nature of the

RuO₂/MWCNT, a hot ﬁltration test was performed. In a typical test,

reduction of acetophenone was carried out under optimized con-

ditions for 20 min and the yield of 1-phenylethanol determined by

GC was 45%. Subsequently, the RuO₂/MWCNT was ﬁltered out from

the reaction mixture (after 20 min) and then the reaction was con-

tinued without the catalyst for 60 min. The progress of the reaction

was monitored at each 10 min intervals; the results are shown in

Fig. 8(i). It can be seen that there was no signiﬁcant change in the

yield in the absence of the catalyst even after extended reaction

time (after 60 min), which indicated that the reaction proceeded

mainly due to the catalytic effect of RuO₂/MWCNT and also there

was no leaching of Ru from the RuO₂/MWCNT.

Since reusability is one of the important features of the nanocat-

alysts, it was also tested for the RuO₂/MWCNT. After the ﬁrst

cycle, the catalyst was separated out from the reaction mixture

by simple centrifugation, washed well with diethyl ether, dried at

150 ^◦C, and then reused in the second cycle. Likewise, the catalytic

cycle was repeated for 8 times and the yield monitored by GC is

shown in Fig. 8(ii). Interestingly, the RuO₂/MWCNT system gave an

94

M. Gopiraman et al. / Applied Catalysis A: General 484 (2014) 84–96

Table 6

Catalytic activity of b-RuO₂/MWCNT in the transfer hydrogenation of carbonyl compounds^a.

Entry

Substrate

Product

Time (min)

Conv.^b(%)

Sel.^b(%)

Yield^b(%)

TON (TOF h⁻¹

)

O

OH

1

2

45

60

65

55

100

65

49

84/112

64/64

OH

O

94

O

3

4

120

60

47

72

100

47

72

61/31

94/94

O

OH

S

O

OH

5

120

39

92

31

40/20

a

Reaction conditions: Substrate (1 mmol), RuO₂/MWCNT (0.78 mol%), NaOH (2 mmol), i-PrOH (5 mL), 82 ^◦C.

Determined by GC analysis.

b

excellent yield of 87% (100% selectivity) with high TON (113) and

The resultant hybrid (r-RuO₂/MWCNT) was well characterized and

it was used as a catalyst for the aerial oxidation of alcohols. Fig. 9

shows the TEM images [Fig. 10(i-iii)], SEM and the correspond-

ing EDS spectrum [Fig. 10(iv) and (v)] and XPS spectrum [Fig.

10(vi)] of r-RuO₂/MWCNT. The TEM images conﬁrmed the exter-

nal attachment of very ﬁne and well dispersed RuO₂NRs on the

MWCNTs. The length and diameter of RuO₂NRs were found in

the range of 20–40 and 7–14 nm, respectively. The weight per-

centage of Ru in the r-RuO₂/MWCNT was found to be 6.92 from

SEM–EDS analysis. In the XPS spectrum of r -RuO₂/MWCNT [Fig.

10(vi)], the BE of Ru 3p_3/2at 462.4 eV and Ru 3p_1/2at 485.1 eV

were attributed to the photoemission from RuO₂(Ru⁴⁺) [48]. It

was found that the r-RuO₂/MWCNT is highly efﬁcient for the

aerial oxidation of alcohols (Table 7). In the preliminary test, for-

tunately, 1-phenylethanol showed a very high conversion of 98%

with 100% selectivity (Table 7, entry 1). Inspired by this result,

the scope of the catalytic system was extended to a wide range

of aromatic, aliphatic, alicyclic and heterocyclic alcohols (Table 7).

TOF (151 h⁻¹) values at the 8^thcycle, which indicated its good

reusability. Moreover, the reused catalyst (u-RuO₂/MWCNT) was

further investigated by TEM, SEM–EDS and XPS analyses (Fig. S3 in

Supporting information). In comparison to the fresh RuO₂/MWCNT,

no signiﬁcant change in the morphology, size, shape, chemical state

and weight percentage of Ru was observed for the u-RuO₂/MWCNT.

Hence, it is obvious that the RuO₂/MWCNT is physically as well as

chemically stable.

3.7. Versatility of RuO₂/MWCNT

Inspired by the high stability of the RuO₂/MWCNT, the u-

RuO₂/MWCNT was further applied in oxidation reactions. Initially,

the u-RuO₂/MWCNT was washed with diethyl ether and dried in

an oven at 60 ^◦C for 5 h. Then the washed u-RuO₂/MWCNT was cal-

cinated under N₂atmosphere at 450 ^◦C for 6 h in a mufﬂe furnace

to obtain ruthenium oxide nanorods (RuO₂NRs) hybrid MWCNTs.

Fig. 8. (i) Heterogeneity and (ii) recyclability of RuO₂/MWCNT for transfer hydrogenation of acetophenone. [Reaction conditions: 117 ␮L of acetophenone (1 mmol), 80 mg

of NaOH (1 mmol), 5 mg of RuO₂/MWCNT (0.77 mol %) and 5 mL of i-PrOH at 82 ^◦C for 45 min].