Bimetallic Ru-Sn nanoparticle catalysts for the solvent-free selective hydrogenation of 1,5,9-cyclododecatriene to cyclododecene

Article abstract of DOI:10.1002/anie.200702274

(Figure Presented) Tin(n)y catalysts: Bimetallic RuSn supported nanoparticle catalysts (see picture; Ru green, Sn blue, O red), prepared from the carbonyl-cluster precursors [Ru₄ (μ⁴-SnPh) ₂(μ-SnPh₂)_4-x(CO)_12-x] (x = 0, 2, 3, 4) are shown to be active catalysts for the highly selective hydrogenation of 1,5,9-cyclododecatriene to cyclododecene under mild conditions.

Full text of DOI:10.1002/anie.200702274

Communications
DOI: 10.1002/anie.200702274
Heterogeneous Hydrogenation
Bimetallic Ru–Sn Nanoparticle Catalysts for the Solvent-Free Selective
Hydrogenation of 1,5,9-Cyclododecatriene to Cyclododecene**
Richard D. Adams,* Erin M. Boswell, Burjor Captain, Ana B. Hungria, Paul A. Midgley,
Robert Raja,* and John Meurig Thomas*
Cyclododecene (CD) is an important intermediate in the
chemical industry since it figures eminently in the synthesis of
dicarboxylic aliphatic acids, ketones, cyclic alcohols, lactones
and other useful materials, including 12-laurolactam and
dodecanedioic acid which are monomers used in the manu-
facture of nylon12, nylon612, copolyamides and polyesters all
of which have extensive applications.
bimetallic nanoparticles, consisting of clusters with a total
atom content of as little as two and as large as ten atoms.
Herein we describe the synthesis and structures of four
new tin-containing tetraruthenium cluster complexes, two of
which we have tested in their denuded form and have found to
be active catalysts for the highly selective hydrogenation of
CDT to CD. The compounds [Ru4(m4-SnPh)2(CO)12] (1),
[Ru4(m4-SnPh)2(m-SnPh2)2(m-CO)2(CO)8] (2), [Ru4(m4-SnPh)2-
Two of us previously reported[1] that cyclododecene may
be efficiently produced in a low-temperature, solvent-free
fashion by selectively hydrogenating 1,5,9-cyclododecatriene
(CDT) in the presence of a Ru6Sn nanoparticle catalyst. The
actual catalyst was derived[2] from the precursor carbonylate
[Ru6C(CO)16SnCl3]À, in which the chlorinated tin atom
bridges two of the six ruthenium atoms of the octahedral
cluster.[3] From in situ extended X-ray absorption fine struc-
ture (EXAFS) and FTIR studies[1] on the denuded active
Ru6Sn nanoparticle catalyst it was clear a residual chlorine
atom remained attached to the active catalytic entity (see
Figure 3 of Ref. [1]).
(m-SnPh2)3(m-CO)(CO)8]
(3),
and
[Ru4(m4-SnPh)2(m-
SnPh2)4(CO)8] (4) were obtained from the reaction of
[Ru4(CO)12(m-H)4] with Ph3SnH in octane solvent at reflux
(1258C; Scheme 1). All four compounds were characterized
Conscious of the known modifying effect of Sn on Ru (and
other) catalysts,[1,4] and also of the difficulties, such as
reproducibility, frequently associated with the presence of
chlorine in supported catalysts,[5] we have set about to prepare
a range of new Ru–Sn carbonyl complexes for use as catalyst
precursors that are free of Cl (and also of the carbidic
carbon), present in the previously reported active catalyst.[3]
In evolving a suitable method of preparation, we have arrived
at a means of systematically altering the Ru:Sn ratios in
Scheme 1.
1
by a combination of IR and H NMR spectroscopy, single-
crystal X-ray diffraction, and mass spectrometry.
Each compound contains an approximately square-planar
cluster of four ruthenium atoms with two quadruply bridging
SnPh ligands, one on each side of the Ru4 square (Scheme 1).
The molecular structures of 1 and 4 are shown in Figures 1
and 2, respectively. Compound 1 contains twelve carbonyl
ligands like its parent compound, whereas in compounds 2–4,
two, three, and four of the CO ligands were replaced by SnPh2
groups that bridge the Ru–Ru edges of the Ru4 square. The
m4-SnPh stannylyne ligands present in these clusters are very
rare. In fact, there is only one reported example of a m4-SnPh
ligand; this was observed for the compound [Ru5(CO)11(h6-
C6H6)(m4-SnPh)(m3-CPh)].[6]
Compounds 1, 2, and 4 were chosen for our catalytic
investigations. The compounds were deposited (ca. 2% metal
loading) on Davison 923 silica mesopore (38 Š) and were
activated by heating to 2008C for 2 h in vacuum. Catalytic
tests were carried out as described in the Experimental
Section. A typical kinetic plot for the hydrogenation of CDT
at 393 K by the Ru4Sn6 catalyst supported on mesoporous
silica is shown in Figure 3.
[*] Prof. Dr. R. D. Adams, E. M. Boswell, Dr. B. Captain
Department of Chemistry and Biochemistry
University of South Carolina
Columbia, SC 29208 (USA)
Fax: (+1)803-777-6781
E-mail: adams@mail.chem.sc.edu
Prof. Dr. R. Raja
School of Chemistry, University of Southampton
Highfield, Southampton SO171BJ (UK)
E-mail: r.raja@soton.ac.uk
Dr. A. B. Hungria, Dr. P. A. Midgley, Prof. Sir J. M. Thomas
Department of Materials Science, University of Cambridge
Cambridge CB23QZ (UK)
Fax: (+44)1223-334-563
E-mail: jmt2@cam.ac.uk
[**] This research was supported by the Office of Basic Energy Sciences
of the U.S. Department of Energy under Grant No. DE-FG02-
00ER14980.
A comparison of the performance of the three new
bimetallic RuSn nanoparticle catalysts with one another, with
the Ru6Sn (chlorine-containing) catalyst, and with Ru5PtSn[7]
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
8182
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8182 –8185

Angewandte
Chemie
Figure 3. Kinetic plot for the hydrogenation of 1,5,9-cyclododecatriene
~
&
using Ru4Sn6 at 393 K ( conversion; 1,9-cyclododecadiene; cyclo-
*
dodecene; + cyclododecane). At 393 K, and under 30 bar H2 the
solvent-free conversion of 1,5,9-cyclododecatriene into cyclododecene
proceeds smoothly, and for the first 12 h almost exclusively, in the
presence of the Ru4Sn6 bimetallic nanoparticle catalyst supported on
mesoporous silica. n/mol% represents the conversion (%) into
product.
Figure 1. An ORTEP diagram of 1 (thermal ellipsoids set at 30%
probability). Selected bond lengths [Š]: Ru1–Ru2 2.9578(6), Ru1–Ru2*
2.9597(6), Ru1–Sn1 2.7135(5), Ru1–Sn1 2.7147(5), Ru2–Sn1 2.7153(6),
Ru2–Sn1* 2.7134(5).
Figure 4. Comparison of catalytic performance of Ru4Sn6, Ru4Sn4,
Ru4Sn2, and Ru5SnPt with the previously reported (chlorine-containing)
Ru6Sn. Reaction conditions: substrate 50 g, catalyst 25 mg (cluster
anchored on mesopore 2% metal loading), H2 pressure 30 bar,
T=373 K, t=8 h. n/mol% represents the conversion (%) into product.
Figure 2. An ORTEP diagram of 4 (thermal ellipsoids set at 30%
probability). Selected bond lengths [Š]: Ru1–Ru2 2.9578(6), Ru1–Ru2*
2.9597(6), Ru1–Sn1 2.7135(5), Ru1–Sn1 2.7147(5), Ru2–Sn1 2.7153(6),
Ru2–Sn1* 2.7134(5).
The catalysts were characterized both before and after
catalysis by scanning transmission electron microscopy
(STEM). An image showing the nanoparticles derived from
4 after catalysis is shown in Figure 6. It can be seen that the
metal particles are very small and are uniformly approx-
imately 1 nm in size, allowing for the enlargement in
appearance that results from electron optical effects.[8]
Energy dispersive X-ray (EDX) analysis by scanning electron
microscopy (SEM) shows that the composition of each of the
supported catalysts is very similar to the composition of its
molecular precursor.
is shown in Figure 4. Both in regard to degree of conversion
and with respect to selectivity towards formation of the
desirable CD, the Ru4Sn6 preparation surpasses the perfor-
mance of both Ru4Sn2 and Ru6Sn. It is also noteworthy that,
as seen in Figure 5, a substantial increase in conversion (with
close to 100% selectivity) occurs when the reaction is carried
out at 413 K.
Our results demonstrate not only the superiority of using
bimetallic carbonyl complexes as precursors to provide highly
Angew. Chem. Int. Ed. 2007, 46, 8182 –8185
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8183

Communications
5.4 mg (7%) of blue [Ru4(m4-SnPh)2(m-SnPh2)4(CO)8] (4). Spectral
1
data for 1: IR(nCO): n˜ = 2045(vs), 2001(s) cmÀ1. H NMR (C6D6): d =
7.00–7.08 (m, 6H), 7.58–7.62 ppm (m, 4H). EI/MS: m/z 1131. The
isotope pattern is consistent with the presence of four ruthenium and
two tin atoms. Spectral data for 2: IR(nCO): n˜ = 2057(s), 2021(vs),
1998(s), 1983(s,sh), 1959(w), 1844(w), 1821(w) cmÀ1. 1H NMR (C6D6):
d = 6.90–7.20 (m, 10H), 7.21–7.40 (m, 16H), 7.66–7.71 ppm (m, 4H).
EI/MS: m/z 1621. The isotope pattern is consistent with the presence
of four ruthenium and four tin atoms. Spectral data for 3: IR(nCO): n˜ =
2058(w), 2039(m), 2017(m), 2002(vs), 1989(s), 1958(m),
1822(w) cmÀ1 1H NMR (C6D6): d = 7.04–7.08 (m, 3H), 7.28–7.42
.
(m, 28H), 7.65–7.72 ppm (m, 9H). EI/MS: m/z 1866. The isotope
pattern is consistent with the presence of four ruthenium and five tin
atoms. Spectral data for 4: IR(nCO): n˜ = 1996(vs), 1962(s) cmÀ1
.
Figure 5. The superiority of the Ru4Sn6 bimetallic catalyst over the
1H NMR (C6D6): d = 6.87–6.89 (m, 4H), 7.20–7.32 (m, 30H), 7.63–
7.66 ppm (m, 16H). EI/MS: m/z 2111. The isotope pattern is
consistent with the presence of four ruthenium and six tin atoms.
Crystal data for 1: Ru4Sn2O12C24H10, Mr = 1131.98, triclinic, space
previously reported Ru6Sn in the production of the cyclododecene from
the cyclodecatriene is illustrated (as a function of temperature).
Reaction conditions: substrate approximately 50 g, catalyst approxi-
mately 25 mg (cluster anchored on mesopore approximately 2% metal
loading), H2 pressure approximately 30 bar, t=8 h. n/mol% represents
the conversion (%) into product.
¯
group P1, a = 9.1416(4), b = 9.6670(4), c = 9.7105(4) Š, a = 74.889(1),
b = 66.258(1), g = 86.839(1)8, V= 757.16(6) Š3, Z = 1, T= 294 K,
MoKa = 0.71073 Š, 2Vmax = 56.628, GOF = 1.073. The final R1(F2)
was 0.0358 for 3202 reflections I > 2s(I). Crystal data for 2:
¯
Ru4Sn4O10C46H30, Mr = 1621.74, triclinic, space group P1, a =
11.8757(6), b = 12.9166(7), c = 18.0535(9) Š, a = 80.993(1), b =
81.988(1), g = 66.009(1)8, V= 2490.0(2) Š3, Z = 2, T= 294 K,
MoKa = 0.71073 Š, 2Vmax = 56.708, GOF = 1.024. The final R1(F2)
was 0.0354 for 10078 reflections I > 2s(I). Crystal data for 3:
¯
Ru4Sn5O9C57H40, Mr= 1866.62, triclinic, space group P1, a =
13.3973(3), b = 13.8172(3), c = 17.8555(4) Š, a = 89.312(1), b =
89.351(1), g = 64.805(1)8, V= 2990.55(11) Š3, Z = 2, T= 294 K,
MoKa = 0.71073 Š, 2Vmax = 56.608, GOF = 1.021. The final R1(F2)
was 0.0405 for 10690 reflections I > 2s(I). Crystal data for 4:
¯
Ru4Sn6O8C68H50, Mr = 2111.50, triclinic, space group P1, a =
11.9551(5), b = 12.3520(5), c = 12.6818(5) Š, a = 78.933(1), b =
70.662(1), g = 75.589(1)8, V= 1699.12(12) Š3, Z = 1, T= 294 K,
MoKa = 0.71073 Š, 2Vmax = 56.588, GOF = 0.988. The final R1(F2)
was 0.0322 for 6452 reflections I > 2s(I). CCDC-659926–CCDC-
659929 contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif
Figure 6. High-angle annular dark-field (HAADF) images of Ru4Sn6
nanoclusters on Davison 38 Š silica after CDT hydrogenation catalysis.
dispersed supported naked bimetallic catalysts, but also the
beneficial effects of the tin modifier on enhancing the
selectivity. There is clearly much scope to enhance further
the catalytic performance of Ru–Sn bimetallic catalysts (by
exploring other ratios of Ru:Sn and other structures of
precursor entities (from 1:1 to 1:5 of Sn:Ru)). Moreover, it is
likely that addition of Pt to form trimetallic nanoparticle
catalysts, as was done in the case of Ru5PtSn,[7] will yield
particularly powerful new, solvent-free hydrogenation cata-
lysts for which there will be much demand in the emerging
hydrogen economy.[9]
Conversion of 3 into 4. Ph3SnH (5.7 mg, 0.003 mmol) was added
to a suspension of 3 in nonane (25 mL). The reaction mixture was
heated to reflux for 2 h, after which the solvent was removed in
vacuum. The residue was extracted with methylene chloride and
separated by TLC over silica gel using a 3:1 (v/v) hexane/methylene
chloride solvent mixture to yield 1.1 mg (16%) of 4.
Preparation of the catalysts: 1 (13.0 mg) was dissolved in of
CH2Cl2 (10 mL). Davison mesoporous silica (400 mg; Grace Davison,
designated Davison 923, having a pore diameter of 38 Š) was added
to this solution and the solvent was removed under a slow stream of
N2. The support (with anchored cluster) was activated (decarbony-
lated) by calcination in vacuum by heating to 2008C over a period of
approximately 45 min and then maintained at 2008C for an additional
2 h. Similarly compound 4 was anchored on Davison mesoporous
silica.
Catalysis: The catalytic testing was carried out with a high-
pressure stainless reactor (Cambridge Reactor Design) lined with
polyether ether ketone (PEEK). Nanoparticle Ru–Sn catalysts
supported on Davison mesoporous silica (mean diameter 38 Š)
(25 mg) were activated (473 K, 2 h) in the presence of hydrogen
(0.5 MPa) prior to the hydrogenation of the 1,5,9-cyclododecatriene.
The catalysts were introduced into the reactor using a roboter
controlled, custom-built catalyst delivery unit, in order to minimize
exposure to air. The reactor was then depressurized and cooled to
room temperature, before introducing the reactant (50 g) and internal
standard (octane, 2.5 g). After introducing the reactant and internal
Experimental Section
[Ru4(CO)12(m-H)4] with Ph3SnH at 1258C: Ph3SnH (53 mg,
0.151 mmol) was added to a solution of [Ru4(CO)12(m-H)4] (25 mg,
0.033 mmol) in distilled octane (20 mL). The reaction mixture was
heated to reflux for 20 min, after which the solvent was removed in
vacuum. The residue was extracted with methylene chloride and
separated by thin-layer chromatography (TLC) over silica gel using a
3:1 (v/v) hexane/methylene chloride solvent mixture to yield in order
of elution 1.7 mg (4%) of lilac [Ru4(m4-SnPh)2(CO)12] (1), 1.3 mg
(2%) of purple [Ru4(m4-SnPh)2(m-SnPh2)2(m-CO)2(CO)8] (2), 2.8 mg
(4%) of purple [Ru4(m4-SnPh)2(m-SnPh2)3(m-CO)(CO)8] (3), and
8184
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8182 –8185

Angewandte
Chemie
standard, the reactor was purged three times with dry nitrogen prior
to the introduction of H2. The reaction vessel was pressurized with
hydrogen (30 bar, dynamic) and heated to the desired temperature
with continued stirring (1200 rpm). During the reaction, small
aliquots were removed by using a mini-robot autosampler to enable
the kinetics to be studied (see Figure 3). The products of the reaction
were analyzed with gas chromatography (G.C. Varian Model 3400
CX) employing a HP-1 capillary column (25 m ” 0.32 mm) and flame
ionization detector. The identity of the products was confirmed by
LC-MS (Shimadzu, QP 8000).
Both the Ru4Sn6 and Ru4Sn2 bimetallic, supported catalysts, were
reused six or seven times for the hydrogenation of the parent CDT
without any significant loss in catalytic activity or selectivity, and our
standard procedure[1a] was employed to test for (and establish) any
leaching, which was infinitesimal.
The conversions and selectivities were determined as defined by
the following equations and the yields were normalized with respect
to the response factors obtained as above: Conv. (%) = [(mol of
initial substrateÀmol of residual substrate)/(mol of initial sub-
strate)] ” 100. Sel. (%) = [(mol of individual product)/(mol of total
products)] ” 100.
For the internal standard GC method, the response factor (RF),
and mol% of individual products were calculated using the following
equations: RF = (mol product/mol standard) ” (area standard/area
product). Mol% product = RF ” mol standard ” (area product/area
standard) ” 100/mol sample.
Keywords: bimetallic clusters · electron microscopy ·
heterogeneous catalysis · hydrogenation · nanoparticles
.
[1] S. Hermans, R. Raja, J. M. Thomas, B. F. G. Johnson, G. Sankar,
D. Gleeson, Angew. Chem. 2001, 113, 1251; Angew. Chem. Int. Ed.
2001, 40, 1211.
[2] J. M. Thomas, B. F. G. Johnson, R. Raja, G. Sankar, P. A. Midgley,
Acc. Chem. Res. 2003, 36, 20.
[3] S. Hermans, B. F. G. Johnson, Chem. Commun. 2000, 1955.
[4] a) B. A. Riguetto, C. E. C. Rodrigues, M. A. Morales, E. Baggio-
Saitovitch, L. Gengembre, E. Payen, C. M. P. Marques, J. M. C.
Bueno, Appl. Catal. A 2007, 318, 70; b) G. W. Huber, J. W.
Shabaker, J. A. Dumesic, Science 2003, 300, 2075; c) M. Guidotti,
V. Dal Anto, A. Gallo, E. Gianotti, G. Peli, R. Psaro, I. Sordelli,
Catal. Lett. 2006, 112, 89; d) M. S. Holt, W. L. Wilson, J. H.
Nelson, Chem. Rev. 1989, 89, 11; e) J. N. CoupØ, E. Jord¼o, M. A.
Fraga, M. J. Mendes, Appl. Catal. A 2000, 199, 45; f) S. M. Santos,
A. M. Silva, E. Jordao, M. A. Fraga, Catal. Commun. 2004, 5, 377;
g) M. Toba, S. Tanaka, S. Niwa, F. Mizukami, Z. Koppany, L.
Guczi, K.-Y. Cheah, T.-S. Tang, Appl. Catal. A 1999, 189, 243.
[5] a) B. Coq, F. Figueras, C. Moreau, P. Moreau, M. Warawdekar,
Catal. Lett. 1993, 22, 189; b) R. A. van Santen, M. Neurock,
Molecular Heterogeneous Catalysis: A Conceptual and Computa-
tional Approach, Wiley-VCH, New York, 2006.
[6] R. D. Adams, B. Captain, W. Fu, M. D. Smith, Inorg. Chem. 2002,
41, 5593 – 5601.
Electron microscopy. High-angle annular dark-field (HAADF)
images were recorded on a 200 kV FEI Tecnai STEM/TEM electron
microscope. Specimens were prepared by crushing the catalyst
between glass slides and depositing the resulting powder onto a
copper grid supporting a perforated carbon film. Deposition was
achieved by dipping the grid directly into the sample powder to avoid
possible contact with any solvent. X-ray energy dispersive spectra
(XEDS) were acquired at 15 kV on a JEOL 5800 LV Scanning
Electron Microscope equipped with a ultra-thin window (UTW) X-
ray detector.
[7] A. B. Hungria, R. Raja, R. D. Adams, B. Captain, J. M. Thomas,
P. A. Midgley, V. Golvenko, B. F. G. Johnson, Angew. Chem. 2006,
118, 4900 – 4903; Angew. Chem. Int. Ed. 2006, 45, 4782 – 4785.
[8] P. D. Nellist, S. J. Pennycook, Adv. Electron. Electron Phys. 2000,
113, 147 – 203.
[9] a) G. W. Crabtree, M. S. Dresselhaus, M. V. Buchanan, Phys.
Today 2004, 57, 39 – 44; b) J. M. Ogden, Phys. Today 2002, 55, 69 –
75; c) J. Ogden, Sci. Am. 2006, 295, 94 – 101; d) S. Satyapal, J.
Petrovic, G. Thomas, Sci. Am. 2007, 296, 81 – 87; e) P. P. Edwards,
V. L. Kuznetsov, W. I. F. David, Philos. Trans. R. Soc. London
2007, 365, 1043.
Received: May 23, 2007
Revised: August 1, 2007
Published online: September 20, 2007
Angew. Chem. Int. Ed. 2007, 46, 8182 –8185
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8185