Hydrogenation of 2-benzylpyridine over alumina-supported Ru catalysts: Use of Ru3(CO)12 as a Ru precursor

Article abstract of DOI:10.1016/j.apcata.2017.08.038

Although Ru₃(CO)₁₂ becomes a popular precursor for supported Ru catalysts nowadays, the activities of the catalysts prepared by thermolysis of the supported Ru₃(CO)₁₂ under different atmospheres have been rarely compared. We herein report the preparation of alumina-supported Ru samples by thermal activation of Ru₃(CO)₁₂ in air, H₂ or N₂, followed by activity test in the hydrogenation of 2-benzylpyridine (BPy). When the supported Ru₃(CO)₁₂ was activated in air, RuO₂ particles of 12–15 nm diameters were produced by complete oxidation of carbonyl groups. In contrast, thermal activation in H₂ and N₂ induced the formation of highly dispersed Ru⁰ particles of 1.4–2.3 nm diameters. In such activations methane was produced, suggesting that direct hydrogenation of CO coordinated to the Ru surface complex occurred in H₂ while the coordinated CO reacted with ruthenium hydride species in N₂. In the activity test for BPy hydrogenation, the samples prepared in H₂ and N₂ showed superior H₂ storage efficiencies and higher rate constants compared to those prepared in air (reduced before the reaction). Additionally, the former samples were examined to be relatively stable even though exposed to ambient air for 7 days. Therefore, H₂ and N₂ gases are recommended for thermal activation of alumina-supported Ru₃(CO)₁₂.

Full text of DOI:10.1016/j.apcata.2017.08.038

Applied Catalysis A, General 547 (2017) 183–190
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Hydrogenation of 2-benzylpyridine over alumina-supported Ru catalysts:
Use of Ru (CO) as a Ru precursor
3
12
a
a
a,b,⁎
Tae Wan Kim , Jinho Oh , Young-Woong Suh
a
Department of Chemical Engineering, Hanyang University, Seoul 04763, Republic of Korea
Research Institute of Industrial Science, Hanyang University, Seoul 04763, Republic of Korea
b
A R T I C L E I N F O
A B S T R A C T
Keywords:
Although Ru
3
(CO)12 becomes a popular precursor for supported Ru catalysts nowadays, the activities of the
Hydrogenation
catalysts prepared by thermolysis of the supported Ru (CO) under different atmospheres have been rarely
3
12
2
-Benzylpyridine
compared. We herein report the preparation of alumina-supported Ru samples by thermal activation of
Ru (CO)12 in air, H or N , followed by activity test in the hydrogenation of 2-benzylpyridine (BPy). When the
supported Ru (CO)12 was activated in air, RuO particles of 12–15 nm diameters were produced by complete
oxidation of carbonyl groups. In contrast, thermal activation in H and N induced the formation of highly
dispersed Ru0 particles of 1.4–2.3 nm diameters. In such activations methane was produced, suggesting that
direct hydrogenation of CO coordinated to the Ru surface complex occurred in H while the coordinated CO
reacted with ruthenium hydride species in N . In the activity test for BPy hydrogenation, the samples prepared in
and N showed superior H storage efficiencies and higher rate constants compared to those prepared in air
reduced before the reaction). Additionally, the former samples were examined to be relatively stable even
though exposed to ambient air for 7 days. Therefore, H and N gases are recommended for thermal activation of
alumina-supported Ru (CO)12
Triruthenium dodecacarbonyl
Thermal activation
Alumina
3
2
2
3
2
2
2
2
2
H
2
2
2
(
2
2
3
.
1
. Introduction
dehydrogenation, which is confirmed by comparing the calculated
ΔGHYD values between diphenylmethane vs. 2-benzylpyridine (Table
S1; computation details in Supplementary materials). This is also ver-
ified by our preliminary results in the hydrogenation of the two sub-
strates over supported Pt, Pd, Rh, and Ru catalysts: the selectivity to the
full hydrogenation product is higher for diphenylmethane than for 2-
benzylpyridine over each metal catalyst (Table S2). Thus, a study is
necessary to find an active catalyst for the hydrogenation of N-con-
taining LOHC; in this work, 2-benzylpyridine (BPy) is selected as a
model LOHC. Note that BPy has a boiling point of 276 °C, a low melting
point of 9 °C, and a hydrogen storage capacity of 6.7 wt.%.
The most frequently used catalysts for the hydrogenation reaction
are supported noble metals such as Pt [9], Pd [10], Ru [11,12], and Rh
[13]. In case of benzene hydrogenation, the activity decreased in the
following order: Rh > Ru > Pt > Pd [14]. The similar order was
observed in the hydrogenation of 1-methyl-2-pyrroleethanol [15] and
9-ethylcarbazole [16]; the order was Rh > Ru > Pd > Pt. However,
2
In hydrogen storage regarded as an important issue in H economy
[
(
1], the recently emerging solution is a liquid organic hydrogen carrier
LOHC) to store and release H under relatively mild conditions
storage capacity of 5–8 wt.% [2–5]. Most LOHC
2
showing a high H
2
compounds consist of single or multi aromatic ring that can be hydro-
genated into cyclic ring using renewable hydrogen. The H -rich LOHC
of a liquid state is transported relatively easily and stored for a long
term. At a H unloading site, it is catalytically dehydrogenated into the
-lean form (also, a liquid) that will be reloaded at the original H
loading site or at a site to supply cheap H
2
2
H
2
2
2
.
The challenging issue in LOHC system is to improve the dehy-
drogenation efficiency because the reaction is endothermic and thus
favourable at high temperatures thermodynamically [5,6]; for instance,
more than 270 °C is necessary for the hydrogenated form of (di)ben-
zyltoluene [7]. To circumvent this intrinsic issue, the strategy of in-
troducing N atom into the aromatic ring of LOHC has been developed,
which significantly contributes to the decrease of dehydrogenation
enthalpy [4,6,8]. However, this will have a negative effect on the Gibbs
free energy of hydrogenation (ΔGHYD) that is a backward reaction of
there are very limited reports in literature for BPy hydrogenation: RuO
2
at 90–100 °C and 100 atm [17], and 5% Rh-on-carbon catalyst at
55–60 °C and 2.7 atm with 40% wt. ratio of the catalyst to the substrate
[18]. In our preliminary test for BPy hydrogenation, Ru/Al O showed
2 3
⁎
Corresponding author at: Department of Chemical Engineering, Hanyang University, Seoul 04763, Republic of Korea. Tel.: +82 222202329; fax: +82 22202294.
E-mail address: ywsuh@hanyang.ac.kr (Y.-W. Suh).
http://dx.doi.org/10.1016/j.apcata.2017.08.038
Received 5 April 2017; Received in revised form 8 August 2017; Accepted 22 August 2017
Available online 30 August 2017
0926-860X/ © 2017 Elsevier B.V. All rights reserved.

T.W. Kim et al.
Applied Catalysis A, General 547 (2017) 183–190
Table 1
0
BPy hydrogenation activities and Ru particles sizes of the supported Ru catalysts.
Entry
Catalyst
Ru precursor
Activation condition
BPy conv. (%)
CHPi sel. (%)
BPi sel. (%)
CHPy sel. (%)
H
2
storage eff. (%)
d (nm)
a,d
b,d
1
2
3
4
5
6
7
8
9
Pt/Al
Pd/Al
Rh/Al
Ru/Al
2
O
3
–
–
–
–
Ru
Ru
Ru
Ru
Ru
Ru
Ru
Ru
Ru
–
–
–
–
98.7
97.6
81.9
59.7
58.1
57.4
55.3
49.0
89.9
89.7
88.8
89.3
89.5
76.2
77.8
86.3
86.8
82.0
82.1
84.9
79.6
98.1
99.0
95.2
91.9
23.8
20.6
15.2
77.4
73.8
70.1
68.3
63.3
83.8
83.6
83.2
82.5
81.4
79.9
76.9
82.3
82.5
80.8
80.3
81.5
79.0
91.6
91.7
86.0
86.4
75.7
79.2
83.7
18.4
18.0
21.3
21.9
25.9
14.9
15.2
15.0
15.8
16.1
16.8
17.9
16.1
16.2
17.4
17.8
16.8
19.1
8.4
0.5
0.2
2.6
4.2
8.2
8.6
9.7
10.8
1.3
1.2
1.8
1.7
2.5
3.3
2.2
1.6
1.3
1.8
1.9
1.7
1.9
0.0
0.0
0.3
0.6
61.1
58.9
47.2
52.9
50.5
48.8
46.5
40.0
82.6
82.3
81.3
81.5
81.2
68.5
68.9
78.6
79.5
74.2
74.0
77.1
71.3
93.9
94.8
88.5
85.6
–
–
–
–
2
O
3
c,d
c,d
2
O
O
3
2
3
d
Ru(air-400)
Ru(air-500)
Ru(air-600)
Ru(air-700)
3
3
3
3
3
3
3
3
3
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
Air at 400 °C
Air at 500 °C
Air at 600 °C
Air at 700 °C
H
H
H
H
H
H
H
N
N
N
N
N
N
12.2
13.4
14.5
15.3
1.4
1.4
1.5
1.6
1.4
1.7
2.0
1.4
1.4
1.7
2.3
1.4
2.3
1.2
1.2
1.2
1.2
d
d
d
Ru(H
Ru(H
Ru(H
Ru(H
Ru(H
Ru(H
Ru(H
Ru(N
Ru(N
Ru(N
Ru(N
Ru(N
Ru(N
Ru(H
Ru(H
Ru(N
Ru(N
2
2
2
2
2
2
2
-400)
-500)
-600)
-700)
-500)
-500)
-700)
2
2
2
2
2
2
2
at 400 °C
at 500 °C
at 600 °C
at 700 °C
at 500 °C
at 500 °C
at 700 °C
at 400 °C
at 500 °C
at 600 °C
at 700 °C
at 500 °C
at 700 °C
at 500 °C
at 500 °C
at 500 °C
at 500 °C
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
d
Ru(NO)(NO
Ru(NO)(NO
3
)
)
3
3
3
2
2
2
2
2
2
2
2
2
2
-400)
-500)
-600)
-700)
-500)
-700)
-500)
-500)
-500)
-500)
Ru
Ru
Ru
Ru
Ru
Ru
Ru
Ru
Ru
Ru
3
3
3
3
3
3
3
3
3
3
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
(CO)12
2
2
2
2
2
2
d
d
e
f
H
H
N
N
2
2
8.3
13.7
13.0
e
f
2
2
a
b
c
d
e
f
5
5
5
wt.% Pt/Al
wt.% Pd/Al
wt.% Rh/Al
2
O
3
purchased from Sigma Aldrich.
purchased from TCI chemicals.
2
O
3
2
O
3
and 5 wt.% Ru/Al
2
O
3
purchased from Alfa Aesar.
2
Reduced by H at 250 °C for 3 h prior to the reaction.
Nominal Ru loading = 1 wt.%.
Nominal Ru loading = 3 wt.%.
the extremely high selectivity to the full hydrogenation product com-
pared to Pt/Al , Pd/Al , and Rh/Al (Table 1, entries 1–4).
However, the BPy conversion is significantly low over Ru/Al . When
we consider a very low price of Ru compared to Pt, Pd, and Rh, a highly
active Ru catalyst needs to be developed for the efficient BPy hydro-
genation.
3
when Ru (CO)12 is used as a Ru precursor, the results and discussion
2
O
3
2
O
3
2
O
3
herein will be useful for the hydrogenation of N-containing LOHC over
Ru catalyst and furthermore for the hydrogenation reactions de-
manding Ru as an active metal.
2 3
O
2. Experimental
3
Supported Ru catalysts have been generally prepared using RuCl as
a Ru precursor [12,19,20]. However, the chloride ion may remain in
2
.1. Preparation of Ru/Al
2 3
O samples by thermal decomposition of the
the final catalyst even after hydrogen reduction and additional washing
supported Ru (CO)12
3
−
[
21,22]. To remove any negative effect of Cl , strong reducing agents
such as hydrazine and sodium borohydride can be used [21–23],
however, the synthesis process becomes more complicated. Another
Supported Ru catalysts were prepared by impregnating the as-re-
2
−1
ceived γ-Al
of Ru (CO)12 (Sigma-Aldrich Chemical Co.) in tetrahydrofuran
Daejung Chemicals). After the suspension was stirred for 12 h at room
2 3
O support (Stream Chemicals, 216 m g ) with a solution
popular Ru precursor is Ru(NO)(NO
literature that Ru(NO)(NO is converted to Ru metal particles by
thermal activation in H [24–26]. The other Ru precursor is trir-
uthenium dodecacarbonyl Ru (CO)12 that we have used in this work. It
was reported that the higher Ru dispersion was attained over the cat-
alyst prepared by Ru (CO)12 among several Ru precursors [24,25]. In
previous reports thermal treatment of supported Ru (CO)12 catalysts
3 3
) . It is well established in the
3
3 3
)
(
2
temperature, the solvent was removed in a rotary evaporator (EYELA)
under reduced pressure at 45 °C. The obtained sample was dried at
05 °C for 10 h. Finally, it was thermally treated at 400, 500, 600 or
3
1
7
5
5
4
3
00 °C in stagnant air, a H flow (100 sccm) or a N flow (100 sccm) for
2
2
−1
3
h (ramping rate: 5 °C min ). Although the nominal Ru loading was
wt.% for all samples, the actual Ru loading was measured to be
.43 wt.% in average by ICP-OES analysis. The final samples were re-
was investigated under vacuum and hydrogen [27–45], and under inert
gas [46,47]. Nevertheless, the activities of the catalysts prepared by
thermolysis of the supported Ru
have been rarely compared.
3
(CO)12 under different atmospheres
ferred to as Ru(X–Y) in which X is the activation gas (air, H
Y is the activation temperature in a unit of °C (400, 500, 600 or 700).
Additionally, Ru(H -500) and Ru(N -500) samples with the nominal
2 2
or N ) and
In this work the supported Ru
at different temperatures ranging from 400 to 700 °C, where the
that has Lewis acid sites useful for hydro-
3 2
(CO)12 is decomposed in air, H , and
2
2
N
2
Ru loadings of 1 and 3 wt.% were prepared by the method described
above, where the actual Ru loadings were measured to be 0.72 and
.23 wt.% in average for each loading.
2 3
support material is γ-Al O
genation [48] and exhibits excellent thermal stability [49]. After the as-
2
prepared or reduced samples are tested in BPy hydrogenation at 150 °C
0
and 40 barg H
2
, the activity results and the size of Ru particles are
2
.2. Characterization of the prepared samples
compared in order to suggest the optimum decomposition condition
yielding an active Ru catalyst for BPy hydrogenation. Additionally, the
effect of post-treatment on the activities of Ru catalysts prepared under
Powder X-ray diffraction (PXRD) analysis was conducted with a
Rigaku miniFlex300 diffractometer using a Cu Kα radiation source
H
2
2
and N at 500 °C is investigated. Since this work focuses on the
(
1
30 kV and 10 mA). All diffraction patterns were recorded in 2θ range
preparation of a supported Ru catalyst active for BPy hydrogenation
−1
0–90° at a scan rate of 5° min
in a step of 0.02°. Fourier-transform
184

T.W. Kim et al.
Applied Catalysis A, General 547 (2017) 183–190
Scheme 1. Reaction pathway in the hydrogenation of 2-benzylpyridine
(BPy) into 2-(cyclohexylmethyl)-piperidine (CHPi) through two single-ring
hydrogenation intermediates, 2-benzylpiperidine (BPi) and 2-(cyclohex-
ylmethyl)pyridine (CHPy). Note that the pyridine ring is hydrogenated
much faster than the benzene ring.
infrared (FT-IR) spectra of the as-impregnated sample and the activated
samples were recorded with a Nicolet 6700 spectrometer (Thermo
decalin (27 g), and supported Ru sample (catalyst 36, 72, and 222 mg
for the actual Ru loading of 4.43, 2.23, and 0.72 wt.%, respectively)
were added into the reactor, where the Ru/substrate ratio was constant
at 0.089% in every run. This ratio was calculated as follows: [(catalyst
weight × actual Ru loading/Ru atomic mass)/(BPy weight/molecular
Scientific) equipped with
a
DTGS detector in the range
−1
−1
2200–1900 cm
at the resolution of 8 cm . Transmission electron
microscopy (TEM) images were taken at a JEOL JEM-2100F microscope
with an acceleration voltage of 200 kV, where the specimen was pre-
pared by dropping the sample in ethanol onto a copper grid and sub-
sequent drying in a vacuum at 40 °C. Ru composition was measured
with an inductively coupled plasma optical emission spectrometer (ICP-
OES) using an OPTIMA 8300 instrument (Perkin Elmer). For mea-
surement of actual Ru loading, the activated sample was in contact with
a mixture of aqua regia (5 ml) and distilled water (5 ml), followed by
pretreatment in a microwave digestion system (Topex, PreeKem).
Temperature-programmed desorption experiments coupled with a
mass spectrometry detector (TPD-MS) were performed to monitor the
2 2
weight of BPy)] × 100. After the reactor was purged with N , H was
fed into the reactor up to 40 barg that was maintained throughout the
reaction using a back pressure regulator. The reactor was then heated to
150 °C and as soon as the temperature reached 150 °C, the stirring
started with a rate of ca. 400 rpm (reaction time t = 0). After the re-
action for 2 h, the reactor was cooled to ambient temperature. The
reaction temperature and time were chosen by preliminary results.
When we conducted the hydrogenation experiments at 130, 150, and
170 °C, the activity difference at 150 °C appeared to be fairly good in
comparing the catalytic performance of Ru(H
Ru(air–500) samples (Fig. S1). Furthermore, when the BPy conversion
and H storage efficiency of the three samples at 150 °C were plotted
2 2
-500), Ru(N -500), and
decomposition behaviour of the supported Ru
3
(CO)12 precursor. For
experiments in an air flow (90 sccm), the sample (30 mg) was heated to
2
−
1
9
00 °C (ramping rate: 5 °C min ) in a NETZSCH TG209F1 instrument,
while the mass signals of m/z = 18 and 44 for H O and CO , respec-
tively, were measured using a NETZSCH QMS403C. For TPD-MS ex-
periments in a flow of H , N or He (30 sccm for all gases), the sample
50 mg) was heated to 900 °C (ramping rate: 5 °C min ) with a
against the reaction time, their increasing trend was changed around
120 min as the reaction approaches the maximum conversion (Fig. S2).
The reaction mixture was analysed with a Younglin YL6500 GC
equipped with a flame ionization detector and an HP-5 column
(30 m × 0.32 mm × 0.25 μm). The reaction pathway in BPy hydro-
genation is depicted in Scheme 1. The BPy conversion and selectivities
to CHPi, BPi, and CHPy were calculated by the following equations:
2
2
2
2
−
1
(
BELCAT-B instrument (BEL Japan, Inc.), while the mass signals of m/
z = 16, 18, 28, 30, and 44 were measured using a BEL-Mass spectro-
meter to investigate evolution of CH
CO , respectively.
Temperature-programmed reduction (TPR) experiments were con-
ducted in an AutoChem 2910 instrument (Micromeritics) by ramping to
4 2 2 2
, H O, CO, NO/NO /N O, and
BPy mole
BPy conversion (%) = 1 −
CHPi selectivity (%) =
× 100.
2
BPy+BPi + CHPy+CHPi moles
CHPi mole
−1
× 100.
600 °C (rate: 5 °C min ) in a 10% H
2
/Ar flow (50 sccm), after the
BPi+CHPy+CHPi moles
sample (50 mg) was in-situ pretreated in two different ways: (1) the
−1
sample was heated at 300 °C (ramping rate: 5 °C min ) for 1 h and
then cooled to 50 °C, where the stream was an Ar flow (50 sccm) in the
whole run and (2) the sample was heated at 400 °C (ramping rate:
BPi mole
BPi selectivity (%) =
× 100.
BPi+CHPy+CHPi moles
−1
0
CHPy mole
5
°C min ) for 1 h in a 10% O
2
/He flow (50 sccm) to oxidize Ru metal
CHPy selectivity (%) =
× 100.
to RuO and then cooled to 50 °C in an Ar flow (50 sccm).
2
BPi+CHPy+CHPi moles
CO chemisorption was carried out with a BELCAT-B instrument
BEL Japan, Inc.), where the sample (50 mg) was pretreated at 250 °C
for 1 h (ramping rate: 5 °C min ) in a 10% H /Ar flow (30 sccm). After
2
Furthermore, the hydrogen storage efficiency was calculated as
(
follows:
−
1
cooling to 30 °C in a He flow (30 sccm), 5% CO/He gas was repeatedly
H2 storage efficiency (%) = [CHPi mole × 6H + (BPi + CHPy)
2
injected as a pulse until the peak area became saturated. The diameter
of Ru particles (d) was calculated using a ChemMaster program (de-
moles × 3H ]/(BPy mole × 6H ) × 100.
2
2
0
tails in the Supporting Information), where a chemisorption stoichio-
metry CO/Ru was assumed to be 1:1.
3. Results and discussion
3.1. Ru(air) samples prepared by thermal decomposition of the supported
2.3. Hydrogenation activity test
Ru
3
(CO)12 in air
The hydrogenation reaction experiment was performed in a Parr
Fig. 1a shows PXRD patterns of the as-prepared Ru(air) samples,
reactor (volume 100 ml). Typically, the substrate BPy (3 g), the solvent
2
indicating the major reflections corresponding to RuO (PDF #43-1290)
185

T.W. Kim et al.
Applied Catalysis A, General 547 (2017) 183–190
Fig. 2. Mass signals detected in the decomposition of the supported Ru precursor under
(
a) air, (b) H
represent emission of H
NO /N O (purple), respectively. (For interpretation of the references to colour in this
2
, (c) He, and (d) N
2
, where the traces of m/z = 2, 18, 44, 16, 28, and 30
2
(grey), H
2
O (black), CO (blue), CH (red), CO (green), and NO/
2
4
2
2
figure legend, the reader is referred to the web version of this article.)
Fig. 1. PXRD patterns of (a) Ru(air), (b) Ru(H
2
), and (c) Ru(N
2
) samples with the re-
flections of Ru0 (PDF #06-0663), RuO
(PDF #43-1290), and γ-Al O (PDF #10-0425) at
2
2 3
the bottom.
49–58% and the selectivity to CHPi of 63–74%, resulting in the H
2
storage efficiency of 40–50% (Table 1, entries 5–8). Their low activities
0
are due to larger Ru particle sizes (i.e., lower Ru dispersions) found by
with the minor reflections of γ-Al
previously, the species [Ru(CO)
oxidation at 20 °C or vacuum decomposition above 100 °C, where X
represents an oxygen atom of the alumina lattice in this structure
2
O
3
(PDF #10-0425). As revealed
CO chemisorption experiments. RuO
clearly observed in a high-magnification TEM image of Ru(air-500)
Fig. 3a).
In order to suppress the formation of large-sized RuO
supported Ru (CO)12 was decomposed in air at 200 °C, yielding the Ru
particle size of ca. 3.1 nm. However, the activity test showed the BPy
conversion of 53.1%, CHPi selectivity of 75.9%, and H storage effi-
ciency of 46.6%, which is slightly lower than the activity of Ru(air-
00). Since this is the unexpected result, Ru(air-200) was characterized
2
particles of large sizes were
2
X
2
]
n
is formed on Al surface by air
2 3
O
(
2
particles, the
[
27–41]. For the as-impregnated R−u
3 2 3
(CO)12/Al O sample, IR bands
0
1
3
were detected at 2073 and 2001 cm (Fig. S3), corresponding to Ru
B
species among the reported Ru surface species. When the surface spe-
cies (Al –O) Ru(CO) is thermally decomposed in air, the reaction takes
place as follows:
2
s
2
2
4
3
[(Al
s
–O)
2
Ru(CO)
2
] + 3H
2
O
(ads) + 6CO
O + 12CO
2
(ads) + 4.5O
2
→
by TPR-MS: a very huge TPR peak below 200 °C, similar to Ru(air-400),
and additional hydrogen consumption around 350 °C (Fig. S6). In the
3(Al –O–Al
s
s
) + 3RuO
2
+ 3H
2
2
4
accompanying mass traces of m/z = 15 for CH , the peak was detected
This reaction was confirmed by TGA result (Fig. S4) that the mea-
sured mass loss (13.0%) was very close to the theoretical mass loss
13.8%, calculated from the Ru loading and the mass loss of γ-Al at
00 °C) assumed to be valid.
Thermal decomposition in an air flow was investigated by TPD-MS
experiment. Fig. 2a clearly shows CO emission in the range
00–250 °C, which means that the prepared Ru(air) samples are free of
around 350 °C. Since this was not detected for Ru(air-400), the result is
explained by the fact that a little amount of carbonyl species still re-
mains on the surface of Ru(air-200). Therefore, a low activation tem-
(
8
2 3
O
perature under air favours the formation of smaller RuO
leads to incomplete decomposition of carbonyl species.
2
particles but
2
1
Ru
3
(CO)12. This was in agreement with the FT-IR spectrum of Ru(air-
3.2. Ru(H
2
) samples prepared by thermal decomposition of the supported
4
2
00) sample exhibiting no IR bands of CO in the region
3
Ru (CO)12 in H
2
−1
200–1900 cm
(ca. 170 and 464 °C) are considered, larger RuO
would be formed by thermal decomposition in air at 400–700 °C. TPD-
MS experiment also identified the release of H O (m/z = 18) in the
range 100–250 °C by desorption of the adsorbed H O molecules.
(Fig. S3). When the Hüttig and Tammann tempera-
tures of RuO
2
2
particles
For the samples subjected to H
pical reflections of γ-Al were only detected without Ru -related
reflections (PDF #06-0663) in the PXRD patterns (Fig. 1b). Thus, the as-
prepared Ru(H ) samples were tested in BPy hydrogenation without
additional H reduction. All Ru(H ) samples exhibited the similar BPy
conversion of ca. 89% and also the similar CHPi selectivity of
83 ± 1%, affording the enhanced H storage efficiency of 82 ± 1%
(Table 1, entries 9–12). Additionally, when Ru(H -500) was tested after
reduction at 250 °C in the same manner as Ru(air) samples, the BPy
2
activation at 400–700 °C, the ty-
0
2 3
O
2
2
2
0
Therefore, Ru(air) samples were reduced to Ru by H
h prior to the BPy hydrogenation reaction. Note that the reduction of
RuO is complete below 250 °C, confirmed by TPR experiments (Fig.
S5). The reduced Ru(air) samples showed the BPy conversion of
2
at 250 °C for
2
2
3
2
2
2
186

T.W. Kim et al.
Applied Catalysis A, General 547 (2017) 183–190
2 2
Fig. 3. High-resolution TEM images of the as-prepared samples; (a) Ru(air-500), (b) Ru(H -500), and (c) Ru(N -500).
hydrogenation activity was almost similar to that of the as-prepared Ru
(
H
2
-500) (Table 1, entry 13). On the other hand, Ru(NO)(NO
used as Ru precursor and then Al -supported Ru(NO)(NO
decomposed at 500 or 700 °C in H
little lower performance (H storage efficiency of ca. 69%) compared to
the corresponding Ru (CO)12-based ones (Table 1, entries 14 and 15).
3
)
)
3 3
3
was
was
2
O
3
2
. The resulting samples exhibited a
2
3
This is consistent to the previous results [24,25].
In thermal decomposition of the as-impregnated sample in a H
2
flow, the emission of CH
4
(m/z = 16) was observed in the range
2
00–400 °C (Fig. 2b). This is in agreement with no CO band in the re-
−1
gion 2200–1900 cm
2 4
for Ru(H -400) (Fig. S3). CH is produced by
direct methanation of carbon monoxide at discrete ruthenium sub-
carbonyl sites. [29]. This explains no release of CO in TPD experiment.
On the other hand, CO
which is related with the release of CO
Al surface. Also, H O was also emitted in the range 100–600 °C. H
emission at 100–250 °C is due to desorption of adsorbed H O molecules
and the emission at 250–600 °C is possibly due to dehydroxylation of γ-
Al
2
emission was observed in the range 50–250 °C,
2
adsorbed on the Ru-free part of
2
O
3
2
2
O
2
2 3
O .
Based on the above results, it is presumed that the following reac-
tion takes place in H
2
:
Fig. 4. Temperature-programmed reduction profiles of the as-prepared (a) Ru(H
2
) and (b)
3[(Al
–O)
s 2
Ru(CO)
2
] + 3H
2
O
(ads) + 6CO
2
2
(ads) + (45–4a)H →
Ru(N ) samples (black curves), where TPR profiles of the samples oxidized in a 10% O /
2
2
3
(Al –O–Al
s
s
) + 3Ru + (24–2a)H
2
O + aCO + (12–a)CH
2
4
He flow at 400 °C for 1 h are added for comparison (red curves). The numerical values
indicate the reduction degree of the samples. (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)
For confirmation, separate TPR and TPO experiments were con-
ducted for the as-prepared Ru(H ) samples. The amount of H con-
sumed by TPR was calculated from the peak area in the range
2
2
3.3. Ru(N
2
) samples prepared by thermal decomposition of the supported
5
1
0–150 °C (black curves in Fig. 4a). The low reduction degree of
3
Ru (CO)12 in N
2
0
4–25% for Ru(H
2
) samples means the formation of Ru by decom-
(CO)12 in H . The weak H reduction is
explained by partial oxidation of Ru(H ) samples by exposure to am-
position of the supported Ru
3
2
2
PXRD patterns for the samples subjected to N
2
treatment at
00–700 °C are shown in Fig. 1c. In all patterns, Ru -related reflections
were not visible as for Ru(H ) samples. However, very small reflections
of RuO (marked with red circles) were detected in the PXRD patterns
of Ru(N -600) and Ru(N -700).
TPD-MS experiment was conducted in a He flow, where He instead
of N was used because CO has the same m/z of 28 as N . Along with the
emission of H O and CO , CH and CO emission were detected in the
range 250–450 and 250–600 °C, respectively. The experiment in a N
flow also confirmed the CH emission without any formation of NO,
NO , and N O evidenced by the baseline mass trace of m/z = 30
Fig. 2d). Basset et al. pointed out that the water on the hydrated η-
Al support is the hydrogen source for CH formation and it also
takes parts in catalytic water gas shift reaction (CO
O → CO + H ) explaining H and CO formation [27]. In terms
of CH formation, Bell et al. demonstrated that the reaction between the
supported Ru (CO)12 cluster and nucleophilic hydroxyl groups on the
2
0
4
bient air before use. This is confirmed by TPR experiments for the
corresponding oxidized samples: a huge difference between black and
red curves indicates the existence of Ru particles in Ru(H
2
0
2
2
) samples
2
2
(
Fig. 4a). Additionally, the reduction degree decreases for both the as-
prepared and oxidized samples as the activation temperature increases.
This is due to strong interaction of unsaturated Ru atoms with the
0
2
2
2
2
4
oxygen present in the alumina lattice [28]. TPO experiments for Ru(H
00) and Ru(H
-700) also examined the formation of Ru0 particles by
oxygen consumption from 250 to 550 °C (Fig. S7).
2
-
2
5
2
4
0
2
2
The sizes of supported Ru particles were calculated by CO chemi-
(
0
sorption experiments for Ru(H
1
thermal stability of Ru (Hüttig temperature of 509 °C and Tammann
temperature of 1031 °C). These much smaller Ru particles are re-
2
) samples (Table 1). Ru particles of
2
O
3
4
.5 ± 0.1 nm were found, which is associated with relatively high
0
+
H
2
2
2
2
2
0
4
sponsible for the higher BPy hydrogenation activity compared to Ru
air) samples. Nanometer-sized Ru0 particles were confirmed in the
3
(
support surface occurs, resulting in the formation of ruthenium-hydride
species that reacts with some fractions of CO in Ru surface species [28].
2
TEM image of Ru(H -500) sample (Fig. 3b).
187

T.W. Kim et al.
Applied Catalysis A, General 547 (2017) 183–190
Although the exact chemical reaction could not be written due to
complexity of the involved reactions, it is evident that highly dispersed
0
Ru particles can be formed by thermal decomposition of the supported
Ru
3
(CO)12 in N
The observation of tiny amounts of RuO
00) by PXRD could be associated with strong interaction of co-
2
.
2
2 2
in Ru(N -600) and Ru(N -
7
ordinatively unsaturated Ru atoms with the oxygen present in the
alumina lattice. This is supported by the fact that RueRu bonds are
broken and individual Ru atoms can be incorporated into the lattice to
form surface RuAl
another evidence for Ru incorporation, the supported Ru
decomposed at 800 °C in N . As a result of activity test, the H
efficiency of Ru(N -800) was significantly dropped to 30.8% from
9.5% of Ru(N -500), where the nominal Ru loading was 5 wt.% (Fig.
S8). This drop was also found for Ru(N -500) of 1 wt.% Ru nominal
loading. The results are understood because larger Ru particles were
produced at 800 °C. However, the PXRD pattern of Ru(N -800) reveals
were
2
O
4
[50] or RuAl alloys [51]. In order to search for
(CO)12 was
storage
3
2
2
2
7
2
2
0
Fig. 5. Measurement for the rate constant in BPy hydrogenation over Ru(air-500), Ru(H
00), and Ru(N -500), where the rate dependence was assumed to be the first order in the
substrate.
2
-
2
5
2
2 3
another clue for the activity decline: the reflections of γ-Al O
significantly less intense and instead a very large hump was observed in
the 2θ range 10–40°. The most related phase might be considered to be
Ru
4
Al13 alloy (PDF #18-0056). This is believed to be a result of ru-
thenium incorporation into Al
2 3
O , although further study is required
for obtaining more direct evidences.
2
TPR and TPO experiments were also conducted for Ru(N ) samples.
The reduction degrees of the as-prepared Ru(N
2
) samples were sig-
nificantly different from those of the corresponding oxidized samples
0
(
Fig. 4b), as for Ru(H
former Ru(N ) samples. TPO experiments also showed the oxygen
consumption from 250 to 550 °C (Fig. S7), elucidating the existence of
2
) samples. That is, Ru particles exist in the
2
0
Ru particles in Ru(N
In BPy hydrogenation the catalytic performance of Ru(N
was a little lower than or comparable to that of Ru(H ) samples due to
similar sizes of Ru particles (Table 1, entries 16–19). This indicates
2
) samples.
2
) samples
2
0
0
that for Ru(N
2
) samples Ru nanoparticles would be dispersed well on
2 2
Fig. 6. BPy hydrogenation activities of (a) Ru(H -500) and (b) Ru(N -500) samples in the
four repeated cycles.
the alumina support, which was confirmed in the TEM image of Ru(N
2
-
5
00) sample (Fig. 3c). It should be noted here that a very low amount of
0
2 2 2
RuO detected for Ru(N -600) and Ru(N -700) is reduced to Ru under
the reaction condition. This was identified by the activity test over the
reduced Ru(N -500) and Ru(N -700) samples showing only slightly
lower H storage efficiencies than the as-prepared ones (Table 1, entries
0 and 21).
2
2
2
2
2 2
3.4. Activity and stability comparison of Ru(air), Ru(H ), and Ru(N )
samples
Using the BPy conversions measured at the beginning of the hy-
drogenation reaction, the first-order rate constant (k) was estimated for
Ru(H
two samples were used as prepared whereas the last sample was re-
duced by H at 250 °C for 3 h before use. Fig. 5 shows that Ru(H -500)
and Ru(N -500) have very similar rate constants (0.8 ± 0.01 h
which are 2.3-fold higher than 0.35 h
are clearly explained with respect to the size of Ru particles: 13.4 nm
for Ru(air-500) vs. 1.4 nm for Ru(H -500) and Ru(N -500). Ad-
ditionally, we tried to compare the activities of Ru(H -500) and Ru(N
00) samples with different nominal Ru loadings of 1 and 3 wt.%. They
2 2
-500), Ru(N -500), and Ru(air-500) samples. Note that the first
2
2
−
1
2
)
−1
for Ru(air-500). These results
0
Fig. 7. BPy hydrogenation activities of Ru(air-500) samples subjected to different pre-
treatments; the as-prepared sample (left section), the samples reduced at 100–350 °C for
2
2
2
2
-
3
0 min (middle section), and the sample reduced at 250 °C for 3 h (right section).
5
were superior to the corresponding samples with 5 wt.% nominal
0
loading, due to smaller Ru particle sizes (Table 1, entries 22–25). Also,
about 15% compared to the original activity, but the activity in further
runs remained constant for both samples (Fig. 6). This indicates that Ru
(H -500) and Ru(N -500) are relatively stable in the reaction although
2 2
there is a little activity loss between the first and second runs. There-
fore, the two samples are believed to be used repeatedly without pre-
treatment for regeneration.
there is very little difference between the activities of Ru(H
Ru(N -500). In this regard H or N is recommended rather than air for
thermal decomposition of the supported Ru (CO)12
Since catalyst stability is a very important issue in repeated hy-
drogen storage and release, the recycle runs were performed three times
2
-500) and
2
2
2
3
.
2 2
for Ru(H -500) and Ru(N -500) samples. Note that the recovered cat-
alyst by filtration of the reaction mixture was used without any pre-
treatment for the next run. The activity in the second run decreased by
188

T.W. Kim et al.
Applied Catalysis A, General 547 (2017) 183–190
2 2
Fig. 8. BPy hydrogenation activities of Ru(H -500) and Ru(N -500) sam-
ples subjected to different pretreatments; (a) the as-prepared samples (left
section), the Ru(H -500) samples oxidized at 100–300 °C for 30 min
middle section), and the Ru(N -500) sample oxidized at 300 °C for 30 min
right section), and (b) the Ru(H -500) and Ru(N -500) samples exposed
2
(
(
2
2
2
under air at room temperature for 3 or 7 days.
3
.5. Post-treatment of Ru(air), Ru(H
2
), and Ru(N
2
) samples
and N
position of Ru
totally due to the formation of 1–2 nanometer-sized Ru particles in H
and N . This originates from direct hydrogenation of CO coordinated to
the ruthenium surface complex in H or selective methanation of the
2
are suitable atmospheres, compared to air, for thermal decom-
3
(CO)12 supported on Al . The activity enhancement is
2 3
O
0
For Ru(air) samples, H
2
reduction should be conducted in order to
into Ru active for the BPy hydrogenation. However,
when the as-prepared Ru(air-500) sample was tested in BPy hydro-
genation without pre-reduction, the H storage efficiency was estimated
to be 39.1%. Although the activity of the Ru(air-500) sample reduced at
50 °C for 3 h was higher than that of the as-prepared Ru(air-500), this
2
0
transform RuO
2
2
2
2
coordinated CO with ruthenium hydride species (formed by interaction
between RueCO and nucleophilic hydroxyl groups on the support
0
2
surface) in N
and Ru(N
tion due to strong interaction between Ru atoms and the lattice oxygen
in the alumina support. In contrast, RuO is formed upon thermal de-
composition of the supported Ru (CO)12 in air. Due to high-tempera-
ture oxidation, RuO particles become large and after H
particles are larger than those in Ru(H ) and Ru(N ) samples. This ex-
2
. The so-obtained Ru particles in the as-prepared Ru(H
2
)
0
means that some RuO
genation reaction condition.
2
would be reduced to Ru under the hydro-
2
) samples are considered to be stable under ambient condi-
0
Thus, Ru(air-500) was reduced by pure H
00–350 °C with the interval of 50 °C prior to the activity test (Fig. 7).
2
for 30 min in the range
2
1
3
0
Even though the 30-min reduction is assumed to be not enough for
2
2
reduction, Ru
0
complete transformation of RuO
duction at 150–300 °C exhibited the comparable H
50.1–52.4%) to the Ru(air-500) sample reduced at 250 °C for 3 h. The
lower hydrogenation activities of the reduced samples at 100 and
50 °C are a result of insufficient reduction of RuO
and sintering of Ru0
particles, respectively. Therefore, H reduction would be conducted in
the range 150–300 °C for at least 30 min prior to the hydrogenation
2
to Ru , the samples obtained by re-
2
2
2
storage efficiencies
plains lower activities of Ru(air) samples in BPy hydrogenation.
Consequently, the results in this work will be utilized as useful re-
commendation towards preparing an active Ru catalyst by thermal
decomposition of Ru (CO)12 for the hydrogenation of N-containing
3
compounds including BPy.
(
3
2
2
activity test, when the supported Ru
air.
3
(CO)12 is thermally decomposed in
Acknowledgements
On the other hand, Ru(H
prepared, as confirmed in Table 1. However, if they are prepared ex-
situ, the effect of oxidation condition needs to be investigated. When Ru
-500) sample was oxidized in air at 100–300 °C for 30 min, the H
storage efficiency decreased from 82.3% for the as-prepared sample to
0.2% for the sample oxidized at 300 °C (Fig. 8a). This is absolutely due
2
) and Ru(N
2
) samples can be used as
This research was financially supported by the Korea Institute of
Energy Technology Evaluation and Planning under the Ministry of
Trade, Industry & Energy, Republic of Korea (KETEP-2015-
3030041160), as well as by the Basic Science Research Program
through the National Research Foundation of Korea under the Ministry
of Education, Republic of Korea (NRF-2016R1A6A1A03013422). The
authors are grateful for partial support by the research fund of Hanyang
University (HY-2011-00000000227).
(
H
2
2
5
0
to the transformation of some Ru to RuO
temperatures, which was similarly found for Ru(N
nally, Ru(H -500) and Ru(N -500) samples were exposed to ambient air
for 3 or 7 days, and then tested in the BPy hydrogenation without ad-
ditional H reduction. This post-treatment is for addressing the stability
of these samples. As shown in Fig. 8b, the H storage efficiencies of the
2
by partial oxidation at high
2
-500) sample. Fi-
2
2
Appendix A. Supplementary data
2
2
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.apcata.2017.08.038.
air-exposed samples were estimated to be 72.5–74.5%. The activity
results were a little lower than those of the as-prepared samples but
were significantly higher than those of Ru(H
dized at 150–300 °C. This indicates that Ru(H
2
-500) samples pre-oxi-
-500) and Ru(N -500)
References
2
2
samples are relatively stable under the ambient condition, although the
[1] D. Teichmann, W. Arlt, P. Wasserscheid, R. Freymann, Energy Environ. Sci. 4
(2011) 2767–2773.
prolonged exposure to air affects their hydrogenation activities in a
0
[2] M. Markiewicz, Y.Q. Zhang, A. Bösmann, N. Brückner, J. Thöming, P. Wasserscheid,
little extent. This is also an evidence for strong interaction of Ru atoms
S. Stolte, Energy Environ. Sci. 8 (2015) 1035–1045.
with the oxygen in the alumina lattice.
[3] K. Müller, J. Völkl, W. Arlt, Energy Technol. 1 (2013) 20–24.
[
[
4] T. He, Q. Pei, P. Chen, J. Energy Chem. 24 (2015) 587–594.
5] D. Geburtig, P. Preuster, A. Bösmann, K. Müller, P. Wasserscheid, Int. J. Hydrogen
Energy 41 (2016) 1010–1017.
4. Conclusions
[
[
6] A. Moores, M. Poyatos, Y. Luo, R.H. Crabtree, New J. Chem. 30 (2006) 1675–1678.
7] N. Brückner, K. Obesser, A. Bösmann, D. Teichmann, W. Arlt, J. Dungs,
P. Wasserscheid, ChemSusChem 7 (2014) 229–235.
From the superior catalytic performance and higher rate constant of
Ru(H ) and Ru(N ) samples in BPy hydrogenation, it is believed that H
2 2 2
189

T.W. Kim et al.
Applied Catalysis A, General 547 (2017) 183–190
[8] F. Sotoodeh, B.J.M. Huber, K.J. Smith, Appl. Catal., A: Gen. 419–420 (2012) 67–72.
[30] A. Zecchina, E. Guglielminotti, A. Bossi, M. Camia, J. Catal. 74 (1982) 225–239.
[31] E. Guglielminotti, A. Zecchina, A. Bossi, M. Camia, J. Catal. 74 (1982) 240–251.
[32] E. Guglielminotti, A. Zecchina, A. Bossi, M. Camia, J. Catal. 74 (1982) 252–265.
[33] J.G. Goodwin Jr., C. Naccache, Appl. Catal. 4 (1982) 145–152.
[34] K. Asakuar, M. Yamada, Y. Iwasawa, H. Kuroda, Chem. Lett. 14 (1985) 511–514.
[35] Y. Iwasawa, Adv. Catal. 35 (1987) 187–264.
[9] S. Lu, W.W. Lonergan, J.P. Bosco, S. Wang, Y. Zhu, Y. Xie, J.G. Chen, J. Catal. 259
(
2008) 260–268.
10] R. Rahi, M. Fang, A. Ahmed, R.A. Sánchez-Delgado, Dalton Trans. 41 (2012)
4490–14497.
11] Y.P. Sun, H.Y. Fu, D.I. Zhang, R.X. Li, H. Chen, X.J. Li, Catal. Commun. 12 (2010)
88–192.
12] M. Tang, J. Deng, M. Li, X. Li, H. Li, Z. Chen, Y. Wang, Green Chem. 18 (2016)
082–6090.
[
[
[
1
1
[36] K. Asakura, Y. Iwasawa, M. Yamada, J. Chem. Soc., Faraday Trans. 1: Phys. Chem.
Condens. Phases 84 (1988) 2457–2464.
[37] K. Asakura, K.K. Bando, Y. Iwasawa, J. Chem. Soc., Faraday Trans. 86 (1990)
2645–2655.
[38] K. Asakura, Y. Iwasawa, J. Chem. Soc., Faraday Trans. 86 (1990) 2657–2662.
[39] S. Dobos, I. Böszörményi, V. Silberer, L. Guczi, J. Mink, Inorg. Chim. Acta 96 (1985)
L13–L15.
6
[
[
[
[
13] G.Y. Fan, J. Wu, Catal. Commun. 31 (2013) 81–85.
14] H. Greenfield, Acad. Sci. 214 (1973) 233–242.
15] L. Hegedüs, T. Máthe, A. Tungler, Appl. Catal., A: Gen. 147 (1996) 407–414.
16] K.M. Eblagon, D. Rentsch, O. Friedrichs, A. Remhof, A. Zuettel, A.J. Ramirez-
Cuesta, S.C. Tsang, Int. J. Hydrogen Energy 35 (2010) 11609–11621.
17] M. Freifelder, G.R. Stone, J. Org. Chem. 26 (1961) 3805–3808.
18] M. Freifelder, R.M. Robinson, G.R. Stone, J. Org. Chem. 27 (1962) 284–286.
19] J. Xu, X. Su, H. Duan, B. Hou, Q. Lin, X. Liu, X. Pan, G. Pei, H. Geng, Y. Huang,
T. Zhang, J. Catal. 333 (2016) 227–237.
[40] T. Mizushima, K. Tohji, Y. Udagawa, A. Ueno, J. Am. Chem. Soc. 112 (1990)
[
[
[
7887–7893.
[41] E.E. Platero, F. Ruiz de Peralta, Catal. Lett. 48 (1997) 159–163.
[42] X. Yang, W. Zhang, C. Xia, X. Xiong, X. Mu, B. Hu, Catal. Commun. 11 (2010)
867–870.
[
20] T.N. Phaahlamohlaka, D.O. Kumi, M.W. Dlamini, L.L. Jewell, N.J. Coville, Catal.
[43] C. Yu, X. Meng, G. Chen, P. Zhao, RSC Adv. 6 (2016) 22633–22638.
[44] J. Iwamoto, M. Itoh, Y. Kajita, M. Saito, K.I. Machida, Catal. Commun. 8 (2007)
941–944.
[45] S. Muratsugu, S. Kityakarn, F. Wang, N. Ishiguro, T. Kamachi, K. Yoshizawa,
O. Sekizawa, T. Uruga, M. Tada, Phys. Chem. Chem. Phys. 17 (2015) 24791–24802.
[46] D. Szmigiel, W. Raróg-Pilecka, E. Miśkiewicz, M. Gliński, M. Kielak, Z. Kaszkur,
Z. Kowalczyk, Appl. Catal., A: Gen. 273 (2004) 105–112.
[47] K. Sato, K. Imamura, Y. Kawano, S.I. Miyahara, T. Yamamoto, S. Matsumura,
K. Nagaoka, Chem. Sci. 8 (2017) 674–679.
[48] J. Wang, H. Chen, Q. Li, React. Kinet. Catal. Lett. 69 (2000) 277–284.
[49] M. Trueba, S.P. Trasatti, Eur. J. Inorg. Chem. 17 (2005) 3393–3403.
[50] L. Yongxue, W. Shuguang, Y. Pinliang, X. Qin, React. Kinet. Catal. Lett. 40 (1989)
53–58.
Today 275 (2016) 76–83.
[21] B. Lin, R. Wang, J. Lin, S. Du, X. Yu, K. Wei, Catal. Commun. 8 (2007) 1838–1842.
[22] B. Lin, R. Wang, J. Lin, J. Ni, K. Wei, Catal. Commun. 12 (2011) 553–558.
[23] D.A. Echeverri, J.M. Marín, G.M. Restrepo, L.A. Rios, Appl. Catal., A: Gen. 366
(2009) 342–347.
[24] J. Álvarez-Rodríguez, A. Guerrero-Ruiz, I. Rodríguez-Ramos, A. Arcoya-Martín,
Catal. Today 107–108 (2005) 302–309.
[
[
25] F. Lu, C. Yu, X. Meng, J. Zhang, G. Chen, P. Zhao, RSC Adv. 6 (2016) 73810–73816.
26] M. Shirai, Y. Murakami, N. Hiyoshi, N. Minura, A. Yamaguchi, O. Sato, J. Mol.
Catal. A: Chem. 388–389 (2014) 148–153.
[27] A.K. Smith, A. Theolier, J.M. Basset, R. Ugo, D. Commereuc, Y. Chauvin, J. Am.
Chem. Soc. 100 (1978) 2590–2591.
[
[
28] V.L. Kuznetsov, A.T. Bell, Y.I. Yermakov, J. Catal. 65 (1980) 374–389.
29] D.A. Hucul, A. Brenner, J. Am. Chem. Soc. 103 (1981) 217–219.
[51] O. Okada, M. Ipponmatsu, M. Kawai, K. Aika, T. Onishi, Chem. Lett. 13 (1984)
1041–1044.
190