Ruthenium(0) nanoclusters stabilized by a nanozeolite framework: Isolable, reusable, and green catalyst for the hydrogenation of neat aromatics under mild conditions with the unprecedented catalytic activity and lifetime

Article abstract of DOI:10.1021/ja101602d

The hydrogenation of aromatics is a ubiquitous chemical transformation used in both the petrochemical and specialty industry and is important for the generation of clean diesel fuels. Reported herein is the discovery of a superior heterogeneous catalyst, superior in terms of catalytic activity, selectivity, and lifetime in the hydrogenation of aromatics in the solvent-free system under mild conditions (at 25 °C and 42 ± 1 psig initial H₂ pressure). Ruthenium(0) nanoclusters stabilized by a nanozeolite framework as a new catalytic material is reproducibly prepared from the borohydride reduction of a colloidal solution of ruthenium(III)-exchanged nanozeolites at room temperature and characterized by using ICP-OES, XRD, XPS, DLS, TEM, HRTEM, TEM/EDX, mid-IR, far-IR, and Raman spectroscopy. The resultant ruthenium(0) nanoclusters hydrogenate neat benzene to cyclohexane with 100% conversion under mild conditions (at 25 °C and 42 ± 1 psig initial H₂ pressure) with record catalytic activity (initial TOF = 5430 h^-1) and lifetime (TTO = 177 200). They provide exceptional catalytic activity not only in the hydrogenation of neat benzene but also in the solvent-free hydrogenation of methyl substituted aromatics such as toluene, o-xylene, and mesitylene under otherwise identical conditions. Moreover, they are an isolable, bottleable, and reusable catalyst in the hydrogenation of neat aromatics. When the isolated ruthenium(0) nanoclusters are reused, they retain 92% of their initial catalytic activity even for the third run in the hydrogenation of neat benzene under the same conditions as those of the first run. The work reported here also includes (i) far-infrared spectroscopic investigation of nanozeolite, ruthenium(III)-exchanged-nanozeolite, and ruthenium(0) nanoclusters stabilized by a nanozeolite framework, indicating that the host framework remains intact after the formation of a nanozeolite framework stabilized ruthenium(0) nanoclusters; (ii) the poisoning experiments performed by using tricyclohexylphosphine (P(C₆H₁₁)₃) and 4-ethyl-2,6,7-trioxa-1-phosphabicyclo[2.2.2]octane PC₆H ₁₁O₃ to examine whether the ruthenium(0) nanoclusters are encapsulated in the cages or supported on the external surface of nanozeolite; (iii) a summary section detailing the main findings for the green chemistry ; and (iv) a review of the extensive literature of benzene hydrogenation, which is also tabulated as part of the Supporting Information.

Full text of DOI:10.1021/ja101602d

Published on Web 04/20/2010
Ruthenium(0) Nanoclusters Stabilized by a Nanozeolite
Framework: Isolable, Reusable, and Green Catalyst for the
Hydrogenation of Neat Aromatics under Mild Conditions with
the Unprecedented Catalytic Activity and Lifetime
†
¨
Mehmet Zahmakıran, Yal c¸ ın Tonbul, and Saim Ozkar*
Department of Chemistry, Middle East Technical UniVersity, Ankara 06531, Turkey
Received February 24, 2010; E-mail: sozkar@metu.edu.tr
Abstract: The hydrogenation of aromatics is a ubiquitous chemical transformation used in both the
petrochemical and specialty industry and is important for the generation of clean diesel fuels. Reported
herein is the discovery of a superior heterogeneous catalyst, superior in terms of catalytic activity, selectivity,
and lifetime in the hydrogenation of aromatics in the solvent-free system under mild conditions (at 25 °C
2
and 42 ( 1 psig initial H pressure). Ruthenium(0) nanoclusters stabilized by a nanozeolite framework as
a new catalytic material is reproducibly prepared from the borohydride reduction of a colloidal solution of
ruthenium(III)-exchanged nanozeolites at room temperature and characterized by using ICP-OES, XRD,
XPS, DLS, TEM, HRTEM, TEM/EDX, mid-IR, far-IR, and Raman spectroscopy. The resultant ruthenium(0)
nanoclusters hydrogenate neat benzene to cyclohexane with 100% conversion under mild conditions (at
-
1
25 °C and 42 ( 1 psig initial H
2
pressure) with record catalytic activity (initial TOF ) 5430 h ) and lifetime
(TTO ) 177 200). They provide exceptional catalytic activity not only in the hydrogenation of neat benzene
but also in the solvent-free hydrogenation of methyl substituted aromatics such as toluene, o-xylene, and
mesitylene under otherwise identical conditions. Moreover, they are an isolable, bottleable, and reusable
catalyst in the hydrogenation of neat aromatics. When the isolated ruthenium(0) nanoclusters are reused,
they retain 92% of their initial catalytic activity even for the third run in the hydrogenation of neat benzene
under the same conditions as those of the first run. The work reported here also includes (i) far-infrared
spectroscopic investigation of nanozeolite, ruthenium(III)-exchanged-nanozeolite, and ruthenium(0) nano-
clusters stabilized by a nanozeolite framework, indicating that the host framework remains intact after the
formation of a nanozeolite framework stabilized ruthenium(0) nanoclusters; (ii) the poisoning experiments
6 11 3
performed by using tricyclohexylphosphine (P(C H ) ) and 4-ethyl-2,6,7-trioxa-1-phosphabicyclo[2.2.2]octane
PC to examine whether the ruthenium(0) nanoclusters are encapsulated in the cages or supported
6
11 3
H O
on the external surface of nanozeolite; (iii) a summary section detailing the main findings for the “green
chemistry”; and (iv) a review of the extensive literature of benzene hydrogenation, which is also tabulated
as part of the Supporting Information.
6
7
Introduction
nanoclusters, rhodium(0) nanoclusters immobilized on silica,
8
Rh(COD) moiety anchored to polymers, rhodium(0) nanoclus-
ters entrapped in boehmite nanofibers, ruthenium(0) nanoclus-
ters immobilized into SBA-15, [(Cp*)
The complete hydrogenation of aromatics is one of the most
important and challenging transformations in the synthesis of
fine chemicals and intermediates and, traditionally, has been
9
10
+
Zr(CH
3
)
2
] activated
1
2
carried out at high temperature (>100 °C) and/or high pressure
2
(
50 atm H
2
). Literature review reveals that a range of
(3) A SciFinder literature search confirms that ca. 95% of the reports of
benzene hydrogenation catalysis employ high temperature and/or high
pressure with only 5% (90 of >1900) hits refining according to the
terms “benzene hydrogenation at room temperature”. Of the 95 hits,
17 report the complete (100%) hydrogenation of benzene without side
products at room temperature, and they were tabulated in Table S-1
in the Supporting Information.
homogeneous and heterogeneous catalysts have been used to
3
hydrogenate benzene. Among these, the catalysts that achieve
the complete hydrogenation of benzene without any side
2
products under mild conditions (e25 °C and e3 atm H ) are
4
the hydroxyalkylammonium halide stabilized rhodium(0) and
(
4) (a) Schulz, J.; Patin, H.; Roucoux, A. Chem. Commun. 1999, 535. (b)
Schulz, J.; Patin, H.; Roucoux, A. Chem.sEur. J. 2000, 6, 618. (c)
Schulz, J.; Patin, H.; Roucoux, A. AdV. Synth. Catal. 2003, 345, 222.
5) Mevellec, V.; Ramirez, E.; Philippot, K.; Chaudret, B.; Roucoux, A.
AdV. Synth. Catal. 2004, 346, 72.
5
iridium(0) nanoclusters, cyclodextrins stabilized rhodium(0)
(
†
On leave of absence from Dicle University, 21280, Diyarbakır, Turkey
(
(
1) (a) Larock, R. C. ComprehensiVe Organic Transformations; Wiley-
VCH: New-York, 1999. (b) Sheldon, R. A.; Kochi, J. K. Metal
Catalyzed Oxidation of Organic Compounds; Academic: New York,
(6) Nowicki, A.; Zhong, Y.; Leger, B.; Rolland, J. P.; Bricout, H.;
Monflier, E.; Roucoux, A. Chem. Commun. 2006, 296.
(7) Roucoux, A.; Philippot, K.; Payen, E.; Granger, P.; Dujardin, C.;
Nowicki, A.; Mevellec, V. New J. Chem. 2006, 30, 1214.
(8) Seeberger, M. H.; Jones, R. A. J. Chem. Soc., Chem. Commun. 1985,
6, 373.
1
981.
2) Augustine, R. L. Heterogeneous Catalysis for the Synthetic Chemistry;
Marcel Dekker: New York, 1996.
10.1021/ja101602d  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 6541–6549 9 6541

A R T I C L E S
Zahmakıran et al.
1
1
via super acidic sulfated alumina or sulfated zirconia, in-
alteration in the framework lattice or loss in the crystallinity of
the host material. Ruthenium(0) nanoclusters stabilized by a
nanozeolite framework were found to be the most active (initial
12
trazeolite ruthenium(0) nanoclusters, rhodium(0) nanoclusters
stabilized by polyhydroxylated ammonium chloride, and CNT-
13
1
4
-1
supported rhodium(0) nanoclusters (see Table S-1 in the
Supporting Information). As a common feature, all of these
catalytic systems, except Mark’s catalyst, involve transition
metal nanoclusters catalysts, which have attracted much attention
in organic synthesis due to their distinct catalytic activities for
various transformations. However, in their catalytic application
one of the most important problems is the aggregation of
nanoclusters into clumps and ultimately to the bulk metal,
despite using the best stabilizers, which leads to a momentous
decrease in catalytic activity and lifetime. The use of nanocluster
catalysts in systems with confined void spaces such as inside
mesoporous and microporous solids appears to be an efficient
turnover frequency (TOF) ) 5430 h ) and longest lifetime
(total turnovers, TTO ) 177 200) catalyst ever reported for the
1
1
complete hydrogenation of neat benzene under mild conditions
2
(at 25 °C and 42 ( 1 psig initial H pressure). They also provide
exceptional catalytic activity in the complete hydrogenation of
methyl substituted arenes such as toluene, o-xylene, and
mesitylene in the solvent-free systems under otherwise identical
conditions. Moreover, the ruthenium(0) nanoclusters exhibit high
durability throughout their catalytic use in the hydrogenation
reaction against agglomeration and leaching. This significant
property makes them a reusable catalyst in the hydrogenation
of aromatics without appreciable loss of their inherent activity.
More importantly, they fulfill the majority of the “Green
15
16
1
2,17
way of preventing aggregation.
Expectedly, our previous
study has shown that zeolite confined ruthenium(0) nanoclus-
ters provide record catalytic activity in the hydrogenation of
12
20
Chemistry” requirements which impose neat/solventless sys-
tems whenever possible. There are only seven reports for the
complete conversion of neat benzene to cyclohexane at tem-
2
neat benzene under mild conditions (22 °C, 42 ( 1 psig H ).
8
–14
Further improvement in the catalytic activity is rationally
expected by reducing the particle size of zeolite-Y, as the
diffusion of the substrate molecules through the cavities from
the external surface to the cages, where the nanoclusters reside,
can limit the reaction rate. A decrease in the particle size of
host material results in high external surface areas and, thus,
reduces the diffusion path length compared to the large size
zeolite crystals (g1 µm). With this anticipation, we employed
peratures e25 °C.
poisoning experiments performed using the large tricyclohexy-
lphosphine (P(C ) or the compact 4-ethyl-2,6,7-trioxa-1-
phosphabicyclo[2.2.2]octane (PC ) to determine the dis-
The work reported here also includes
6
11 3
H )
6
11 3
H O
tribution of ruthenium(0) nanoclusters between the cavities and
the external surface of a nanozeolite.
Experimental Section
1
8,19
the nanoparticles of zeolite-Y with a narrow size distribution
as the host for the stabilization of metal(0) nanoclusters.
Materials. Ruthenium(III) chloride trihydrate (RuCl
3
2
·3H O),
sodium borohydride (NaBH , 98%), tetramethylammonium hy-
4
Herein, we report the preparation and characterization of
ruthenium(0) nanoclusters stabilized by a nanozeolite framework
as a novel catalytic material. They were reproducibly prepared
from the borohydride reduction of a colloidal solution of
ruthenium(III)-exchanged nanozeolites at room temperature. The
composition, morphology, and structure of the nanozeolite
framework stabilized ruthenium(0) nanoclusters, as well as the
integrity and crystallinity of the host material, were investigated
by using ICP-OES, XRD, XPS, DLS, TEM, HRTEM, TEM/
EDX, mid-IR, far-IR, and Raman spectroscopy. The results of
these multiprong analyses combined with poisoning experiments
with phosphine ligands (Vide infra) reveal the formation of
ruthenium(0) nanoclusters within the zeolite cages as well as
on the external surface of a nanozeolite without causing an
droxide solution ((CH
3
)
4
NOH, 25% wt), aluminum propan-2-olate
, >98%), tetramethylammonium bromide ((CH NBr,
98%), aqueous 30% wt colloidal silica (LUDOX HS-30, SiO
(Al(OC H )
3 7 3
3 4
)
>
2
/
Na
2
O ) 90, % wt SiO
2
) 29.90, % wt Na
2
O ) 0.34), benzene
3
(
99%), toluene (99%), and d-Chloroform (CDCl
) were purchased
from Aldrich. Mesitylene (98%) and o-xylene (98%) were pur-
chased from Fluka. All catalyst reaction solutions were prepared
in an oxygen-free atmosphere (Labconco, drybox). Benzene,
toluene, mesitylene, and o-xylene were distilled over sodium under
argon and stored in the drybox. Ruthenium(III) chloride was
recrystallized from water, and the water content of RuCl ·xH O
was determined by TGA and found to be x ) 3. Deionized water
was distilled by a water purification system (Milli-Q System). All
glassware and Teflon coated magnetic stir bars were cleaned with
acetone, followed by copious rinsing with distilled water before
drying in an oven at 150 °C. Polypropylene bottles were washed
with doubly deionized water under ultrasonication.
3
2
(
9) Park, I. S.; Kwon, M. S.; Kim, N.; Lee, J. S.; Kang, K. Y.; Park, J.
Preparation of Ruthenium(0) Nanoclusters Stabilized by a
Nanozeolite Framework. The colloidal nanozeolites (FAU-type)
were prepared from a clear precursor solution with the molar
Chem. Commun. 2005, 5667.
(
(
10) Zhang, J.; Xie, Z.; Liu, Z.; Wu, W.; Han, B.; Huang, J.; Jiang, T.
Catal. Lett. 2005, 103, 59.
11) Nicholas, J. P.; Ahn, H.; Marks, T. J. J. Am. Chem. Soc. 2003, 125,
-
composition 1.00 Al
2
O
3
:4.36 SiO
O(2 Br ):0.048 Na O:249 H
follows: 38.3 g of H O, 26.2 g of TMAOH, 5.7 g of TMABr, 6.3 g
of Al(i-OPr) , and 13.1 g of Ludox HS-30) as described elsewhere.
2
:2.39 (TMA)
2
O(2 OH ):1.2
4
325.
-
(
TMA)
2
2
2
O (weight composition is as
¨
(
(
12) Zahmakıran, M.; Ozkar, S. Langmuir 2008, 24, 7065.
13) Hubert, C.; Denicourt-Nowicki, A.; Gu e´ gan, J.-P.; Roucoux, A. Dalton
Trans. 2009, 7356.
2
21
3
(
(
14) Pan, H. B.; Wai, C. M. J. Phys. Chem. C 2009, 113, 19782.
15) (a) Schmid, G. In Nanoparticles: From Theory to Application; Wiley-
VCH: Weinheim, 2004. (b) Widegren, J. A.; Finke, R. G. J. Mol.
Catal. A: Chem. 2003, 191, 187. (c) Widegren, J. A.; Finke, R. G. J.
Mol. Catal. A: Chem. 2003, 198, 317.
After complete crystallization of the solution, the resulting colloidal
nanozeolite crystals were recovered by three repetitions of high
speed centrifugation (12 000 rpm for 1 h), decantation, and
redispersion in doubly distilled water at room temperature. Then,
to remove sodium defect sites, colloidal nanozeolites were ion-
¨
¨
(
16) Ozkar, S.; Finke, R. G. J. Am. Chem. Soc. 2002, 124, 5796. (b) Ozkar,
S.; Finke, R. G. Langmuir 2002, 18, 7653.
2
2
exchanged with 100 mL of 1 M NaCl solution at 50 °C for 12 h
(
17) (a) Seidel, A.; Loos, J.; Boddenberg, B. J. Mater. Chem. 1999, 9,
and recovered again by three cycles of high speed centrifugation
2
495. (b) Tang, Q.; Zhang, Q.; Wang, P.; Wang, Y.; Wan, H. Chem.
(
12 000 rpm for 1 h), decantation, and redispersion in doubly
¨
Mater. 2004, 16, 1967. (c) Zahmakıran, M.; Ozkar, S. Langmuir 2009,
¨
2
1
9
5, 2667. (d) Zahmakıran, M.; Ozkar, S. Appl. Catal., B 2009, 84,
¨
04. (e) C¸ alı s¸ kan, S.; Zahmakıran, M.; Ozkar, S. Appl. Catal., B 2010,
(20) Poliakoff, M.; Fitzpatrick, J. M.; Farren, T. R.; Anastas, P. T. Science
2002, 297, 807.
3, 387.
(
(
18) The nanosized zeolites are usually defined as the highly crystalline
zeolites that have a size of less than 1000 nm (see ref17).
(21) Holmberg, B. A.; Wang, H.; Norbeck, J. M.; Yan, Y. Microporous
Mesoporous Mater. 2003, 59, 13.
19) Tosheva, L.; Valtchev, V. P. Chem. Mater. 2005, 17, 2494.
(22) Breck, D. W. Zeolite Molecular SieVes; Krieger: Malabal, FL, 1984.
6
542 J. AM. CHEM. SOC. 9 VOL. 132, NO. 18, 2010

Ru(0) Nanoclusters Stabilized by Nanozeolite Framework
A R T I C L E S
distilled water at room temperature. Next, they were ion-exchanged
catalyst was weighed into a 40 mm × 20 mm borosilicate culture
tube containing a 5/16 in. × 5/8 in. Teflon coated magnetic stir
bar, and then the substrate was added with a gastight syringe. The
culture tube was then placed inside the Fischer-Porter (FP) pressure
bottle. The FP bottle was sealed, brought outside the drybox, placed
inside a constant temperature circulating water bath at 25 ( 0.1
°C, and connected via Swagelock TFE-sealed quick-connects to
the hydrogenation line which had already been evacuated for at
least 30 min to remove any trace oxygen and water present and
with Ru3 cations in 100 mL of aqueous solution of RuCl
+
(26.1
3
mg, 0.1 mmol) for 12 h at 50 °C. The solution was then cooled
down to room temperature, and ruthenium(III) cations were reduced
by the addition of 10 mmol of NaBH (378 mg). The reduction
4
was considered to be complete when no more hydrogen evolution
was observed (∼10 min). Then, this mixture was centrifuged at
5
000 rpm for 30 min, and the solid was separated from the solution
by decantation and washing with 200 mL of doubly distilled water.
The final product was dried under vacuum (10- Torr) at 90 °C for
3
2
then refilled with purified H at 42 ( 1 psig. Then, the FP bottle
6
h and stored in a glovebox.
was purged 10 times (15 s per purge, with stirring at >600 rpm).
The pressure in the FP bottle was then set to a constant 42 ( 1
Characterization of Nanozeolite Framework Stabilized Ru-
2
psig H , and the reaction was started. The reaction was followed
thenium(0) Nanoclusters. The chemical composition of the
samples was determined by inductively coupled plasma atomic
emission spectroscopy (ICP-AES; Varian-Vista) after each sample
was completely dissolved in the mixture of HNO /HCl with a 1:3
3
ratio. Powder X-ray diffraction (XRD) patterns were acquired on
by monitoring the pressure change in the FP bottle via an RS-232
module and using a Lab View 8.0 program on a PC. The pressure
vs time data were processed using Microsoft Office Excel 2003
and Origin 7.0. Reaction rates were determined from the slope of
2
the linear portion of the H uptake curve. The complete hydrogena-
a MAC Science MXP 3TZ diffractometer using Cu KR radiation
1
tion of each substrate was also confirmed by a H NMR spectrum
of the reaction solution taken at the end of the reaction.
(
(
wavelength 1.5406 Å, 40 kV, 55 mA). Dynamic light scattering
DLS) measurements in colloidal suspensions were performed with
Testing the Catalytic Activity of Ruthenium Free Nanozeo-
lite in the Hydrogenation of Neat Benzene. In a control
experiment the catalytic activity of ruthenium free nanozeolite (100
mg) was checked in the hydrogenation of 1.0 mL of benzene (under
a Malvern Zetasizer-Nano instrument equipped with a 4 mW
He-Ne laser (633 nm) and an avalanche photodiode detector.
Transmission electron microscopy (TEM) and high resolution TEM
(
(
HRTEM) analyses were performed on a JEM-2010F microscope
JEOL) operating at 200 kV. The TEM/EDX elemental analysis
42 ( 1 psig initial H pressure at 25 ( 0.1 °C), which was
2
was performed by using an energy dispersive X-ray (EDX) analyzer
KEVEX Delta series) mounted on the Hitachi S-800 modulated
to TEM. XPS analysis was performed on a Physical Electronics
800 spectrometer equipped with a hemispherical analyzer and
performed in the same way as described in the section “General
Procedure for the Hydrogenation of Aromatics Catalyzed by
Ruthenium(0) Nanoclusters Stabilized by a Nanozeolite Frame-
work”
(
5
using monochromatic Al KR radiation (1486.6 eV, the X-ray tube
Isolability, Bottleability, and Reusability of Ruthenium(0)
Nanoclusters Stabilized by a Nanozeolite Framework in the
Hydrogenation of Neat Benzene. After the first run of the
hydrogenation of 1.0 mL of benzene (11.2 mmol), catalyzed by
100 mg of ruthenium(0) nanoclusters stabilized by a nanozeolite
framework (with a ruthenium loading of 0.27% wt corresponding
to 2.7 µmol of Ru), the FP bottle was detached from the line, taken
into the drybox, and opened and the suspension in the culture tube
was transferred into a Schlenk tube, resealed, and connected to a
vacuum line. After the evaporation of volatiles, the solid residue
was weighed (90 and 78 mg for the second and third run,
respectively) and used again in the hydrogenation of 1.0 mL of
benzene (11.2 mmol) under the same conditions (25 ( 0.1 °C and
working at 15 kV, 350 W and pass energy of 23.5 keV). NMR
spectra were recorded on a Bruker Avance DPX 400 (400.1 MHz
1
for H). FTIR spectra were recorded under N
2
gas purging using a
Nicolet Magna-IR 750 spectrometer. The Raman spectrum of the
sample was recorded on a Bruker RFS-100/S series Raman
spectrometer equipped with a Nd:YAG laser at 1064 nm using the
FT-Raman technique.
Pretreatment of the Samples and Measurement Conditions
for the Far-IR Experiments. For the far-IR analyses, samples of
nanozeolite, ruthenium(III)-exchanged nanozeolite, and its reduced
form, nanozeolite framework stabilized ruthenium(0) nanoclusters,
were prepared by following the procedure described in the section
“
Preparation of Ruthenium(0) Nanoclusters Stabilized by Nanozeo-
2
42 ( 1 psig H initial pressure). The results were expressed as a
lite Framework”. The samples were pressed into self-supporting
wafers of 15 mm in diameter weighing approximately 15 mg by
using an applied pressure of 6-8 tons per square inch for up to
percentage of catalytic activity of ruthenium(0) nanoclusters retained
in the successive runs of neat benzene hydrogenation.
Determination of the Catalytic Lifetime of Ruthenium(0)
Nanoclusters Stabilized by a Nanozeolite Framework in the
Hydrogenation of Neat Benzene. The catalytic lifetime of
ruthenium(0) nanoclusters stabilized by a nanozeolite framework
was determined by measuring their total turnover number (TTO)
in the hydrogenation of neat benzene. This experiment was started
with a 50 mg amount of ruthenium(0) nanoclusters stabilized by a
nanozeolite framework (with a ruthenium loading of 0.27% wt
corresponding to 1.35 µmol of Ru) and 3 mL of benzene and
performed in the same way as described in the section “General
Procedure for the Hydrogenation of Aromatics Catalyzed by
Ruthenium(0) Nanoclusters Stabilized by a Nanozeolite Frame-
work”. When the complete hydrogenation of benzene had been
achieved the FP bottle was removed from the line, taken into the
drybox, and opened and more benzene (4 mL) was added into the
reaction solution; the reaction was continued in this way until no
hydrogen uptake was observed.
6
0 s. The wafers were clamped into a glass tube whose one side
was connected to a turbo-molecular vacuum pump and placed in a
furnace. The self-supporting wafers were dehydrated with the
following schedule, using an NX-(Mitsubishi) series temperature
controller: 0.5 h from 25 to 100 °C, 1 h at 100 °C, 4 h from 100
to 550 °C, and 4 h at 550 °C. After thermal treatment the samples
inside the glass tube were cooled down to room temperature, then
sealed, and transferred into the glovebox (0
2
< 5 ppm, H O < 1
2
ppm), where the sample was transferred into a sample holder
equipped with a high density polyethylene window. The level of
dehydration was checked by complementary mid-IR spectroscopy.
The degree of dehydration was judged by the flatness of the baseline
in the IR V(OH) stretching and δ(OH) deformation regions,
-
l
22
3
400-3700 and 1600-1650 cm , respectively. The far-IR spectra
were recorded on a Nicolet Magna-IR 750 spectrometer under N
purging by using a global source and DTGS detector. The spectral
2
-
1
resolution in all cases was 4 cm , and all of the spectra have been
baseline corrected.
6 11 3 6 11 3
P(C H ) and PC H O Poisoning of Ruthenium(0) Nano-
cluster Catalysts in the Hydrogenation of Neat Benzene. For
the catalyst poisoning experiments two separate stock solutions were
prepared by dissolving 18.7 mL of P(C H ) or 3.0 mg of PC H O
3
General Procedure for the Hydrogenation of Aromatics
Catalyzed by Ruthenium(0) Nanoclusters Stabilized by a
Nanozeolite Framework. All reaction mixtures were prepared in
6
11 3
6
11
in 10 mL of benzene. Then, 100 mg of catalyst (0.27% wt Ru
a nitrogen-filled oxygen-free drybox (0
following the procedure described elsewhere. In a drybox the
2
< 5 ppm, H
2
O < 1 ppm)
loaded) were transferred into a 22 mm × 175 mm culture tube
1
2
5
5
containing a /16 in × /
8
in. Teflon coated magnetic stir bar. To
J. AM. CHEM. SOC. 9 VOL. 132, NO. 18, 2010 6543

A R T I C L E S
Zahmakıran et al.
this was added an aliquot (40-80 µL) of P(C
6
H
11
)
3
or (216-720
(ii) there is no bulk metal formed at an observable size outside
the nanozeolite crystals; (iii) NFS-Ru(0) are highly dispersible
in aqueous medium and colloidal nanozeolites do not aggregate
at the end of the preparation of nanoclusters, unlike the results
6 11 3
µL) of PC H O stock solution by using a gastight syringe of 100
or 500 µL, respectively, and the total volume of solution was leveled
to 1.0 mL by the addition of benzene. The culture tube was then
placed in the FP bottle. The FP bottle was then sealed, taken outside
the drybox, placed in a constant temperature circulation water bath
thermostatted at 25 ( 0.1 °C, and attached to the hydrogenation
apparatus via the quick connects, and the solution was stirred for
25
of Mintovas’ work, in which the agglomeration of colloidal
nanozeolites has been observed in the synthesis of nanozeolite
confined copper(0) nanoclusters by a γ-radiolysis method; (iV)
the method used for the preparation of NFS-Ru(0) does not
cause any observable defects in the structure of colloidal
nanozeolite-Y, a fact which is also confirmed by XRD results.
The HR-TEM image of NFS-Ru(0) given in Figure 1c shows
the uniform distribution of ruthenium within the highly ordered
cavities of a nanozeolite. A TEM-EDX spectrum of the same
sample given in Figure 1d shows that ruthenium is the only
3
0 min. Next, the hydrogenation of benzene was carried out in the
same way as described in the section of “General Procedure for
the Hydrogenation of Aromatics Catalyzed by Ruthenium(0)
Nanoclusters Stabilized by a Nanozeolite Framework”. In each set
the initial rates were calculated in psig/h from the pressure vs time
2
3
data and used to obtain the percent catalytic activity retained
initial rate of benzene hydrogenation in the presence of poison
(
over the one in the absence of poison multiplied by 100).
26
element detected in addition to the zeolite framework elements
Leaching Test of Ruthenium(0) Nanoclusters Stabilized by
a Nanozeolite Framework. After the first run of the hydrogenation
of 1.0 mL of benzene, catalyzed by 150 mg of ruthenium(0)
nanoclusters stabilized by a nanozeolite framework (with a loading
of 0.27% wt Ru, corresponding to 2.7 µmol Ru), the FP bottle was
detached from the line, taken into the drybox, and opened and the
suspension in the culture tube was filtered; the filtrate was
transferred into a new culture tube, and 0.5 mL benzene was added.
The culture tube was placed into the FP bottle. The hydrogenation
of benzene was performed in the same way as described in the
section “General Procedure for the Hydrogenation of Aromatics
Catalyzed by Ruthenium(0) Nanoclusters Stabilized by a Nanozeo-
lite Framework”. No hydrogenation of benzene was observed after
and Cu from the grid. Although no bulk ruthenium metal is
observed outside the zeolite nanocrystals, the TEM-EDX and
ICP-OES analyses indicate the presence of ruthenium in the
samples. This implies that the ruthenium metal is mostly located
within the cavities of a colloidal nanozeolite.
The oxidation state of ruthenium and surface composition of
the NFS-Ru(0) were analyzed by X-ray photoelectron spectros-
copy. The survey scan XPS spectrum of the NFS-Ru(0) sample
(
see Figure S-3a in the Supporting Information) shows the
presence of ruthenium in addition to the zeolite framework
elements (O, Si, Al, Na, N, C). The deconvolution of high
resolution Ru 3d and 3p XPS spectra of the NFS-Ru(0) sample
8
h. Additionally, no ruthenium metal was detected in the filtrate
(
see Figure S-3b and c in the Supporting Information) gives
by ICP which had a detection limit of 24 ppb for Ru.
three prominent peaks at 285.3, 280.2, and 461.5 eV, readily
assigned to Ru(0) 3d3/2, Ru(0) 3d5/2, and Ru(0) 3p3/2, respec-
Results and Discussion
27
tively. The peaks observed at 287.6, 281.2, and 463.5 eV show
the presence of RuO , which may originate from the surface
oxidation of ruthenium(0) nanoclusters during the XPS sampling
Preparation and Characterization of Ruthenium(0) Nano-
clusters Stabilized by a Colloidal Nanozeolite Framework.
Nanozeolite framework stabilized ruthenium(0) nanoclusters,
hereafter referred to as NFS-Ru(0), are prepared as a novel
2
28,29
procedure.
The slight shifts (0.2-0.5 eV) observed in the
binding energies of NFS-Ru(0) with respect to ruthenium metal
might be attributed to both the quantum size effect and peculiar
2
2
material by a procedure comprising (i) the ion exchange of
30
Ru ions with the extra framework Na+ or TMA+ ions of
3+
nanozeolite at 50 °C for 12 h and then (ii) the reduction of
(
24) DLS analysis gives a unimodal particle size distribution for the NFS-
Ru(0) sample, a peak centered at 130 nm as shown in Figure S-2,
when the scattering intensity is expressed as number-weighted
dependence. It is noteworthy that the particle size measured by
considering only the nontouching particle in the TEM image of the
same sample is much smaller than that obtained by DLS which
accounts for the touching spots as one particle.
3
+
Ru ions in the cavities of a nanozeolite with sodium
borohydride in aqueous solution at room temperature, whereby
the Na+ ions reoccupy the cation sites left by the Ru3+ ions
upon reduction, as tracked by far-IR spectroscopy (Vide infra).
After centrifugation, filtering, copious washing with water, and
drying in vacuum at room temperature, NFS-Ru(0) were
obtained as gray powders and characterized by ICP-OES and
XRD. Comparison of the XRD patterns of nanozeolite and NFS-
Ru(0) (see Figure S-1 in the Supporting Information) clearly
shows that the host material remains intact with its FAU
(
(
25) Kecht, J.; Tahri, Z.; Waele, V. D.; Mostafavi, M.; Mintova, S.; Bein,
T. Chem. Mater. 2006, 18, 3373.
26) As judged by the LR
1
, LR
2
1 2
KR , and KR lines of Ru at 2.61, 2.56,
19.3, and 19.1 keV, respectively: Jones, I. P. Chemical Microanalysis
Using Electron Beams; The Institute of Materials: London, 1992.
27) (a) Wagner, C.; Riggs, W. M.; Davis, L. E.; Moulder, J. F.; Muilenberg,
G. E. Handbook of X-ray Photoelectron Spectroscopy; Physical
Electronic Division, Perkin-Elmer: Eden Prairie, MN, 1979; Vol. 55,
p 344. (b) Park, K. W.; Choi, J. H.; Kwon, B. K.; Lee, S. A.; Sung,
Y. E.; Ha, H. Y.; Hong, S. A.; Kim, H.; Lee, S. A.; Sung, Y. A.; Ha,
H. Y.; Hong, S. A.; Kim, H.; Wieckowski, A. J. Phys. Chem. B 2002,
(
2
2
framework during the preparation. The Bragg reflections of
NFS-Ru(0) (with a ruthenium content of 0.27% wt, determined
by ICP-OES analysis) are identical with those of the nanozeolite
framework.
1
06, 1869. (c) Nyholm, R.; Maartensson, N. Solid State Phys. 1980,
The high crystallinity, colloidal stability, the degree of
polydispersity, and typical morphological appearance of col-
loidal nanozeolites were investigated by TEM, HRTEM, and
DLS analyses. Inspection of the TEM images of the colloidal
NFS-Ru(0) samples with a metal loading of 0.27% wt Ru given
in Figure 1a and b reveals the following: (i) there exist only
crystals of colloidal nanozeolite-Y with a crystalline domain
size of ∼25 nm which is smaller than that found by DLS
13, L279. (d) Chakroune, N.; Viau, G.; Ammar, S.; Poul, L.; Veautier,
D.; Chehimi, M. M.; Mangeney, C.; Villain, C.; Fievet, F. Langmuir
2
005, 21, 6788.
(
28) Although oxidation of ruthenium(0) nanoclusters during the XPS
sampling procedure is a well known fact, the results obtained from
XPS analysis need to be tested by using other analytical techniques
2
to check whether RuO is really formed during the XPS sampling
procedure or in the preparation of NFS-Ru(0). For this purpose we
analyzed NFS-Ru(0) by Raman spectroscopy, which is a strong tool
to test the existence of RuO
2
. It was observed that the characteristics
2
4
Raman peaks of RuO (see ref 27b and c) in the range 400-1200
measurement (see Figure S-2 in the Supporting Information);
2
-
1
cm were missing in the Raman spectrum of NFS-Ru(0), which
confirms that the detection of RuO by XPS is a result of the surface
2
(
23) Wilkins, R. G. Kinetics and Mechanism of Reactions of Transition
Metal Complexes, 2nd ed.; VCH: New York, 1991.
oxidation of ruthenium(0) nanoclusters during the XPS sampling
procedure.
6
544 J. AM. CHEM. SOC. 9 VOL. 132, NO. 18, 2010

Ru(0) Nanoclusters Stabilized by Nanozeolite Framework
A R T I C L E S
Figure 1. (a and b) TEM images of NFS-Ru(0) with different magnifications, (c) high resolution TEM image of NFS-Ru(0), and (d) TEM-EDX spectrum
of NFS-Ru(0).
3
1
electronic properties of the zeolite matrix. The interaction of
ruthenium(0) nanoclusters with the framework oxygen of zeolite
is expected to induce a positive charge on the surface metal
which would increase the binding energies of ruthenium(0)
nanoclusters. A similar effect has also been observed for the
energies might be due to the fact that electrons in the core level
are strongly restricted by the atomic nucleus as observed in the
case of palladium(0) nanoclusters in zeolite-Y.
3
2
In the characterization of zeolite confined metal(0) nanoclus-
ters, far-IR spectroscopy is a direct probe for location of cation
3
1a
31c
33
zeolite encapsulated cobalt and platinum. In addition to
matrix effect, the high energy shifts in the ruthenium binding
sites in zeolite via their characteristic vibrational modes. The
combined use of frequencies and intensities of the far-IR
absorption bands due to the site specific cation translatory modes
allows us to secure metal cation vibrational assignments for sites
(
29) (a) Kim, H.; Popov, N. B. J. Power Sources 2002, 104, 52. (b) Liu,
H. L.; Yoon, S.; Cooper, S. L.; Cao, G.; Crow, J. E. Phys. ReV. B
34
I, II, III, and I′ in zeolite-Y (Figure 2a). The far-IR assignment
1
999, 60, 6980. (c) Lee, S. H.; Liu, P.; Cheong, H. M.; Tracy, C. E.;
Deb, S. K. Solid State Ionics 2003, 165, 217.
(
30) Schmid, G., Ed. Clusters and Colloids; From Theory to Applications;
VCH Publishers: New York, 1994.
(32) Okitsu, K.; Yue, A.; Tanabe, S.; Matsumoto, H. Bull. Chem. Soc. Jpn.
2002, 75, 449.
(
31) (a) Guczi, L.; Bazin, D. Appl. Catal., A 1999, 188, 163. (b) Jiang,
Y.-X.; Weng, W.-Z.; Si, D.; Sun, S.-G. J. Phys. Chem. B 2005, 109,
(33) Baker, M. D.; Godber, J.; Ozin, G. A. J. Am. Chem. Soc. 1985, 107,
3033.
¨
7
637. (c) Fukuoko, A.; Higashimoto, N.; Sakamoto, Y.; Inagaki, S.;
(34) Ozkar, S.; McMurray, L.; Holmes, A. J.; Kuperman, A.; Ozin, G. A.
Fukushima, Y.; Ichikawa, M. Top. Catal. 2002, 18, 73.
J. Phys. Chem. 1991, 95, 9448.
J. AM. CHEM. SOC. 9 VOL. 132, NO. 18, 2010 6545

A R T I C L E S
Zahmakıran et al.
Figure 2. (a) Schematic view of the structure of the zeolite-Y (FAU framework); (b) Far-IR spectra of zeolite-Y (upper, black) and nanozeolite-Y powders
-7
(
down, red) taken from the self-supporting wafers dehydrated in vacuum (10 Torr) at 550 °C. Asterisks denote the bands attributed to the tetramethylammonium
+
cation (TMA ) translatory modes in the spectrum of nanozeolite-Y.
of the extra-framework cation sites in zeolite-Y has also been
2
3
35
confirmed by Na NMR spectroscopy. It has been shown that
there are normal modes of the charge balancing cations,
3
3
essentially decoupled from the lattice vibrations, and each
cation site has a vibrational mode with a characteristic frequency
in the far-IR region. The frequencies of cation translatory
33
vibrational modes depend on the mass and charge of the cation.
The far-IR spectra of nanozeolite-Y and zeolite-Y samples, both
thermally dehydrated, are depicted together in Figure 2b for
comparison. The far-IR spectrum of nanozeolite-Y shows two
lower frequency bands at 65 and 52 cm- which are readily
assigned to the translatory modes of the residual tetramethy-
1
+
36
lammonium cation, TMA . On passing from nanozeolite-Y
to ruthenium(III)-exchanged nanozeolite-Y, the site II and III
+
Na cation bands in the far-IR spectra (Figure 3a) lose intensity
-
1
while two lower frequency bands grow in at 69 and 55 cm ,
3
+
which can be assigned to the Ru cation translatory modes in
sites II and III. The higher frequency absorption band at 69
-
1
3+
cm is most likely due to the site II Ru cation mode, and the
-1
lower frequency absorption band at 55 cm is attributed to the
site III Ru . This result indicates that site II and site III Na
3
+
+
cations in the supercage (R-cage) are replaced by ruthenium(III)
cations. It is noteworthy that after ion exchange the absorption
+
+
bands of the remaining Na and TMA ions are shifted to higher
frequencies due to the change in binding strength affected by the
19,33,34
specific site population.
In the far-IR spectrum of NFS-Ru(0)
depicted in Figure 3b, one observes that the cation sites II and III
left vacant by Ru3 ions upon reduction are reoccupied by Na
+
+
cations coming from sodium borohydride. Obtaining the absorption
bands for NFS-Ru(0) essentially at the same frequency and intensity
as the ones in the spectrum of virgin nanozeolite-Y (Figure 3b)
indicates that the host material retains its integrity including the
cation site population after the generation of ruthenium(0) nano-
clusters within the supercages of a nanozeolite.
¨
(
35) (a) Jelinek, R.; Ozkar, S.; Ozin, G. A. J. Am. Chem. Soc. 1992, 114,
¨
4
907. (b) Jelinek, R.; Ozkar, S.; Ozin, G. A. J. Phys. Chem. 1992, 96,
949. (c) Jelinek, R.; Ozkar, S.; Pastore, H. O.; Malek, A.; Ozin, G. A.
¨
Figure 3. Far-IR spectra of (a) nanozeolite-Y (red) and ruthenium(III)-
exchanged nanozeolite-Y (green) as well as (b) nanozeolite-Y (red) and
NFS-Ru(0) (blue) formed from the sodium borohydride reduction of
ruthenium(III)-exchanged nanozeolite-Y.
5
¨
J. Am. Chem. Soc. 1993, 1145, 563. (d) Ozin, G. A.; Ozkar, S.;
Macdonald, P. J. Phys. Chem. 1990, 94, 6939.
(
36) Note that all the nanozeolite samples used in this study were subjected
to ion exchange without calcination and, therefore, may contain
+
residual TMA cations used as a template in the synthesis of
nanozeolite. The heavier TMA cations are expected to have lower
frequency translatory vibrational modes than those of Na cations.
The residual TMA cations, which are difficult to remove by thermal
+
Catalytic Activity of Ruthenium(0) Nanoclusters Stabilized
by a Nanozeolite Framework in the Hydrogenation of Neat
Aromatics under Mild Conditions. The catalytic activity of NFS-
Ru(0) was tested in the hydrogenation of aromatics (benzene,
+
+
dehydration in vacuum, are most probably in sites I and I′ (in sodalite
cages).
6
546 J. AM. CHEM. SOC. 9 VOL. 132, NO. 18, 2010

Ru(0) Nanoclusters Stabilized by Nanozeolite Framework
A R T I C L E S
Figure 4. Graphs of concentration of (a) benzene, (b) toluene, (c) o-xylene, and (d) mesitylene versus time for hydrogenation of 1.0 mL of neat substrate
catalyzed by 100 mg of NFS-Ru(0) (with a ruthenium loading of 0.27% wt corresponding to 2.7 µmol of Ru) at 25 ( 0.1 °C and 42 ( 1 psig initial H
pressure.
2
toluene, o-xylene, and mesitylene) in the solvent-free systems.
An important observation is that when the mostly hydrophilic
NFS-Ru(0) are dispersed in an organic medium they lose their
colloidal stability while their framework preserves its crystal-
linity as confirmed by XRD (see Figure S-4 in the Supporting
Information). Although the nanozeolites are gathering to form
clumps in the organic medium, ruthenium(0) nanoclusters retain
their stability and do not show any agglomeration into the bulk
ruthenium as evidenced by TEM analyses (see Figure S-5 in
the Supporting Information). The catalytic hydrogenation of
aromatics in the solvent-free system was started by agitating
the NFS-Ru(0) in the aromatic substrate at 25 ( 0.1 °C and 42
until the consumption of benzene, toluene, o-xylene, and
mesitylene with initial TOF values of 5430, 1800, 870, or
3
7
-
1
-1
460 h , respectively. The TOF of 5430 h for the complete
hydrogenation of neat benzene, a well-known test reaction in
arene hydrogenation, is the record value since the highest TOF
previously reported for the hydrogenation of neat benzene under
-1
mild conditions (e25 °C and e3 atm) is 1040 h (see Table
S-1 in the Supporting Information). It is also noteworthy that
12
NFS-Ru(0) provide a significant improvement in catalytic
activity in the hydrogenation of neat benzene compared to the
results reported for ruthenium(0) nanoclusters supported on
3
8
39
nonporous silica or mesoporous silica.
(
2
1 psig H , and the progress of the reaction was followed by
monitoring the hydrogen uptake which can be converted to the
concentration loss of the aromatic substrate by using the
stoichiometry. The complete hydrogenation of aromatics was
(
37) The TOF and TTO values reported herein are those typically reported.
TOF ) (mol of H consumed/total mol of catalyst loading)/time; TTO
2
1
) (TOF) (time). That is, the reported values of TOF and TTO are not
also confirmed by checking the H NMR spectra of the reaction
corrected for the amount of metal that is on the surface of the catalyst
and/or the actual number of active sites. For the calculation of initial
TOF values the conversions of substrates within the first hour were
considered.
solution at the end of the reaction. Figure 4 shows the plots of
concentration loss of aromatics vs time for the hydrogenation
of neat aromatics (benzene, toluene, o-xylene, and mesitylene)
catalyzed by NFS-Ru(0). For all of the substrates the hydro-
genation starts immediately without an induction period as a
preformed catalyst is used. A linear hydrogenation continues
(
38) Ruthenium(0) nanoclusters supported on silica provide a TOF value
-
1
of 864 h in the hydrogenation of benzene to cyclohexane under
forcing conditions (at 140 °C and 870 psig H pressure). Ning, J.;
Xu, J.; Liu, J.; Lu, F. Catal. Lett. 2006, 109, 175.
2
J. AM. CHEM. SOC. 9 VOL. 132, NO. 18, 2010 6547

A R T I C L E S
Zahmakıran et al.
linearly with the increasing concentration of phosphine and the
hydrogenation stops when 0.50 equiv of P(C H O ) per
6 11 3
ruthenium is added to benzene (Figure 5). The observation that
the benzene hydrogenation is completely poisoned by less than
1
equiv of phosphine is compelling evidence for a heterogeneous
4
2
6 11 3
catalysis. In the experiments with P(C H ) the activity of
the catalyst in the hydrogenation of neat benzene also decreases
initially with the increasing phosphine/ruthenium molar ratio
up to 0.15 and then remains unchanged beyond this ratio. The
dissimilarity of two phosphine ligands in poisoning the NFS-
Ru(0) catalyst in the hydrogenation of neat benzene arises from
their size difference. While the small phosphine P(C
6 11 3
H O )
(
kinetic diameter is 4.6 Å) can readily enter the supercages
2
2
through the 7.4 Å aperture of nanozeolite, the large phoshine
P(C (kinetic diameter is 10.9 Å) cannot enter the nanozeo-
lite cages, thus remaining outside. Therefore, the large P(C
can poison only the ruthenium(0) nanoclusters on the external
surface of the nanozeolite, while the small P(C ) deacti-
Figure 5. Plots of the relative catalytic activity of NFS-Ru(0) retained
6 11 3
H )
versus mol poison/total mol Ru ratio for the P(C
poisoning in the hydrogenation of 1.0 mL of neat benzene catalyzed by
00 mg of NFS-Ru(0) (with a ruthenium loading of 0.27% wt corresponding
to 2.7 µmol Ru) at 25 ( 0.1 °C and 42 ( 1 psig initial H pressure.
6 11 3 6 11 3
H ) and P(C H O )
6 11 3
H )
1
6
11 3
H O
2
vates all the ruthenium(0) nanoclusters in the cavities and on
the external surface of the zeolite. The results of these
experiments given in Figure 5 indicate that the NFS-Ru(0)
sample has 70% of the catalytically active ruthenium(0)
nanoclusters standing on the external surface of the nanozeolite
while the remaining 30% are in the cavities of the nanozeolite.
Catalytic Lifetime of Nanozeolite Framework Stabilized
Ruthenium(0) Nanoclusters in the Hydrogenation of Neat
Benzene under Mild Conditions. The catalytic lifetime of NFS-
Ru(0) in the hydrogenation of neat benzene was determined by
measuring the total turnover number (TTO) at 25 ( 0.1 °C and
In the series of benzene, toluene, o-xylene, and mesitylene,
the hydrogenation rate decreases with the increasing number
of methyl substituents, because of the electronic effect on the
40
aromatic ring due to the addition of methyl groups. Addition-
ally, the observation of the slowest hydrogenation rate for
mesitylene can also be explained by the Lennard-Jones kinetic
2
2
diameter of mesitylene (7.6 Å), which is larger than the
2
2
supercage aperture of nanozeolite-Y (7.4 Å).
P(C and PC Poisoning of Nanozeolite Frame-
6
H
11
)
3
6 11 3
H O
work Stabilized Ruthenium(0) Nanoclusters in the Hydrogena-
tion of Neat Benzene. In working with the zeolite confined metal
particles, the crucial question is whether the nanoclusters are on
the surface or in the cavities of the host material. To investigate
the distribution of ruthenium(0) nanoclusters on the external surface
or within the cages of nanozeolite-Y, a series of poisoning
experiments for the hydrogenation of neat benzene catalyzed by
NFS-Ru(0) were performed by using two different phosphine
4
2
2 ( 1 psig initial H pressure. When all of the benzene was
converted to cyclohexane, more benzene was added into the
solution and the reaction was continued in this way until no
hydrogen uptake was observed. It was found that NFS-Ru(0)
provide 177 200 turnovers in benzene hydrogenation over 105 h
before deactivation (see Figure S-6 in the Supporting Informa-
tion). The aVerage TOF value during this lifetime experiment
-1
is 1690 h which is approximately one-third of the initial TOF
ligands as poison: tricyclohexylphosphine, P(C ), and 4-ethyl-
H )
6 11 3
-1
value of 5430 h indicating that the nanoclusters are deactivated
as the catalytic reaction proceeds. The decrease in the catalytic
activity of ruthenium(0) nanoclusters over time may be attributed
to the gathering of nanozeolite particles to form clumps in the
organic medium as evidenced by TEM analyses (Vide supra)
which decreases the accessibility of ruthenium(0) nanoclusters
confined within the cages of nanozeolites.
2
,6,7-trioxa-1-phosphabicyclo[2.2.2]octane, P(C H O ), having
6 11 3
41
kinetic diameters of 10.9 Å and 4.6 Å, respectively. The
hydrogenation of neat benzene catalyzed by NFS-Ru(0) was
conducted in the presence of phosphine ligand in various
phosphine/ruthenium molar ratios at 25.0 ( 0.1 °C. The initial
rate of benzene hydrogenation was determined in each of the
independent experiments at various phosphine/ruthenium ratios
and divided by the hydrogenation rate of neat benzene in the
absence of phosphine at 25.0 ( 0.1 °C to obtain the catalytic
activity retained (initial rate of benzene hydrogenation in the
presence of poison over the one in the absence of poison). Figure
Nevertheless, this is the highest TTO value ever reported in
the hydrogenation of neat benzene, 70 times larger than the
1
2
previous best value.
Isolability and Reusability of Nanozeolite Framework
Stabilized Ruthenium(0) Nanoclusters in the Hydrogenation
of Neat Benzene under Mild Conditions. Isolability and reus-
ability of NFS-Ru(0) were also tested in the hydrogenation of
neat benzene. After the complete hydrogenation of neat benzene,
the catalyst was isolated as black powders and bottled under a
nitrogen atmosphere. The isolated NFS-Ru(0) are redispersed
in benzene and yet an active catalyst in benzene hydrogenation.
They retain 92% of their initial activity for the third run in the
hydrogenation of neat benzene with complete conversion into
cyclohexane (see Figure S-7 in the Supporting Information).
This indicates that NFS-Ru(0) are an isolable, bottleable,
redispersible, and reusable catalyst in the hydrogenation of
5
shows the plots of the percent catalytic activity retained in
the hydrogenation of neat benzene versus the phosphine/
ruthenium molar ratio for both of the phosphine poisons. In the
hydrogenation experiments performed in the presence of
6 11 3
P(C H O ), the catalytic activity of NFS-Ru(0) decreases almost
(
(
(
39) Ruthenium(0) nanoclusters supported on mesoporous silica provide a
1
-
TOF value of 3600 h in the hydrogenation of benzene to cyclohexane
pressure). Su,
under forcing conditions (at 110 °C and 580 psig H
2
F.; Lv, L.; Yin, F.; Tao-Liu, T.; Cooper, A. I.; Song-Zhao, A. J. Am.
Chem. Soc. 2007, 129, 14213.
40) (a) March, J. AdVanced Organic Chemistry: Reactions, Mechanisms,
and Structure, 4th ed.; Wiley-Interscience: New York, 1992. (b)
Fonseca, G. S.; Silveira, E. T.; Gelesky, M. A.; Dupont, J. AdV. Synth.
Catal. 2005, 347, 847. (c) Bahaman, M. V.; Vannice, M. A. J. Catal.
1
991, 127, 251.
(42) (a) Lin, Y.; Finke, R. G. Inorg. Chem. 1994, 33, 4891. (b) Hornstein,
B. J.; Aiken, J. D., III.; Finke, R. G. Inorg. Chem. 2002, 41, 1625.
41) Li, X.; Ozin, G. A.; Ozkar, S. J. Phys. Chem. 1991, 95, 4463.
6
548 J. AM. CHEM. SOC. 9 VOL. 132, NO. 18, 2010

Ru(0) Nanoclusters Stabilized by Nanozeolite Framework
A R T I C L E S
aromatics in a solvent-free system under mild conditions.
Additionally, no ruthenium was detected (by ICP, detection limit
(5) The solvent-free hydrogenation of aromatics catalyzed
by NFS-Ru(0) is “relatively green” in terms of its environmental
impact as it fulfills 7 of the 12 requirements of green chemistry
20
24 ppb for Ru) in the filtrate obtained by filtration of the reaction
mixture after the first run of hydrogenation. The results of this
leaching test confirms the retaining of ruthenium within the
zeolite matrix (no ruthenium passes into the solution during the
suction filtration). A control experiment was also performed to
show that the hydrogenation of benzene is completely stopped
by the removal of NFS-Ru(0) from the reaction solution.
including the following: (i) it is 100% selective and minimizes
byproducts or waste; (ii) it maximizes the incorporation of all
reactants into the products; (iii) it is solventless (i.e., uses neat
aromatics as the substrate/solvent); (iV) it requires relatively low
energy as it occurs under mild conditions of 25.0 ( 0.1 °C and
e3 atm of pressure; (V) it is catalytic, not stoichiometric; (Vi)
it does not use any blocking, protecting/deprotecting group; (Vii)
Conclusions
1
real-time monitoring is easy by measuring the H
NMR, or GC analysis, for example.
2
uptake, H
The main findings of this work as well as implications or
predictions can be summarized as follows:
Acknowledgment. Partial support by Turkish Academy of
Sciences and TUBITAK (2214- Research Fellowship for M.Z) is
acknowledged. M.Z. thanks all the members of the Compact
Chemical Process Research Team at the National Institute of
Advanced and Industrial Science and Technology (AIST) in
Tsukuba.
(
1) The colloidal nanozeolite framework stabilized ruthe-
nium(0) nanoclusters were prepared for the first time as a new
material and characterized by using ICP-OES, XRD, DLS, TEM,
HRTEM, TEM-EDX, XPS, mid-IR, far-IR, and Raman spec-
troscopy. The results of these multiprong analyses reveal the
formation of ruthenium(0) nanoclusters within the framework
of well-dispersed colloidal nanozeolite particles whereby the
host material retains its integrity including the cation site
population after the generation of ruthenium(0) nanoclusters
within the supercages of the nanozeolites.
Supporting Information Available: Table S-1: The top 17
catalysts in benzene hydrogenation at room temperature. Figure
S1: X-ray diffraction patterns of nanozeolite and nanozeolite
framework stabilized ruthenium(0) nanoclusters with a ruthe-
nium content of 0.27% wt. Figure S2: Particle size distribution
and its fit obtained from the dynamic light scattering analysis
of the nanozeolite framework stabilized ruthenium(0) nanoclus-
ters with a ruthenium content of 0.27% wt. Figure S3: (a) XPS
survey scan spectrum, (b) high resolution Ru 3d XPS spectrum
and its simulated peak fitting in the region 279-292 eV, (c)
high resolution Ru 3p XPS spectrum and its simulated peak
fitting in the region 470-450 eV of the nanozeolite framework
stabilized ruthenium(0) nanoclusters with a Ru content of 0.27%
wt. Figure S4: X-ray diffraction pattern of nanozeolite frame-
work stabilized ruthenium(0) nanoclusters recovered after the
hydrogenation of 1.0 mL of benzene starting with 100 mg of
NFS-Ru(0). Figure S5: TEM images of the nanozeolite frame-
work stabilized ruthenium(0) nanoclusters with a Ru content
of 0.27% wt dispersed in neat benzene at room temperature.
Figure S6: Graph of benzene consumption versus time for the
hydrogenation of neat benzene catalyzed by the nanozeolite
framework stabilized ruthenium(0) nanoclusters. Figure S7: The
percentage of catalytic activity and conversion versus the
number of catalytic runs for the hydrogenation of 1.0 mL of
benzene (11.2 mmol), catalyzed by NFS-Ru(0) (0.27% wt Ru,
(
2) The dispersion of nanozeolite framework stabilized
ruthenium(0) nanoclusters in the aromatic substrates acts as a
superb catalyst in terms of activity, selectivity, and lifetime in
the solvent-free hydrogenation of aromatics under mild condi-
tions. They provide a record TTO of 177 200 and TOF up to
1
5
420 h- in the hydrogenation of neat benzene with 100%
conversion and selectivity to cyclohexane at 25 ( 0.1 °C and
with an initial H pressure of 42 ( 1 psig. Recall that the prior
highest TTO and TOF values are 2420 mol H
2
-
1
2
·(mol Ru) and
040 h , respectively. Moreover, they show also exceptional
-1
12
1
catalytic activity in the solvent-free hydrogenation of toluene,
o-xylene, and mesitylene with 100% conversion and TOF values
-1
of 1800, 870, and 580 h , respectively.
3) The complete hydrogenation of neat benzene is achieved
(
even in successive runs performed by redispersing the NFS-
Ru(0) isolated after the previous run. When reused, NFS-Ru(0)
retain 92% of their initial catalytic activity even for the third
run with a complete hydrogenation of benzene to cyclohexane.
Thus, they are an isolable, bottleable, and reusable catalyst in
the hydrogenation of neat benzene at 25 ( 0.1 °C with an initial
H
2
pressure of 42 ( 1 psig.
(
4) The poisoning experiments using small and large phos-
phines were performed to address the crucial issue of whether
the nanoclusters are on the surface or in the cavities of the host
material. The results of these experiments reveal that 70% of
active ruthenium(0) nanoclusters are on the external surface of
nanozeolite particles while only 30% of ruthenium(0) nano-
clusters are in the supercages of the zeolite.
1
00 mg, 2.7 µmol of Ru) at 25 ( 0.1 °C with an initial H
2
pressure of 42 ( 1 psig. This material is available free of charge
via the Internet at http://pubs.acs.org.
JA101602D
J. AM. CHEM. SOC. 9 VOL. 132, NO. 18, 2010 6549