[Rh2(COD)2(Dppm)(μ2-Cl)]BF4: Precursor for a selective hydrogenation catalyst and its recycling by silica entrapment

Article abstract of DOI:10.1021/om060172q

The synthesis of [Rh,₂(COD)₂(dppm)(μ₂- Cl)]BF₄ (1) (COD = 1,5-cyclooctadiene, dppm = bis-(diphenylphosphino) methane) from simple precursors is reported. This is a rare example of a dirhodium complex with an open [Rh₂(μ₂-dppm) (μ₂-Cl)] core. The complex has been used to affect the hydrogenation of styrene and benzo[b]thiophene with total selectivity and competitive rates of reaction. The recycling of the catalyst has been achieved by the entrapment of 1 in silica by a sol-gel method to produce a recyclable solid catalyst.

Full text of DOI:10.1021/om060172q

3912
Organometallics 2006, 25, 3912-3919
[Rh2(COD)2(Dppm)(µ2-Cl)]BF4: Precursor for a Selective
Hydrogenation Catalyst and Its Recycling by Silica Entrapment
Fabio Lorenzini,† Kenneth T. Hindle,† Steven J. Craythorne,† Alan R. Crozier,†
Fabio Marchetti,‡ Ciara´n J. Martin,§ Patricia C. Marr,*,† and Andrew C. Marr*,†
School of Chemistry and Chemical Engineering, The Queen’s UniVersity of Belfast, DaVid Keir Building,
Stranmillis Road, Belfast, BT9 5AG, U.K., Dipartimento di Chimica e Chimica Industriale, UniVersita` di
Pisa, Via Risorgimento, 35, 56124 Pisa, Italy, and School of Chemistry, UniVersity of Bristol,
Cantocks Close, Bristol, BS8 1TS, U.K.
ReceiVed February 22, 2006
The synthesis of [Rh2(COD)2(dppm)(µ2-Cl)]BF4 (1) (COD ) 1,5-cyclooctadiene, dppm ) bis-
(diphenylphosphino)methane) from simple precursors is reported. This is a rare example of a dirhodium
complex with an open [Rh2(µ2-dppm)(µ2-Cl)] core. The complex has been used to affect the hydrogenation
of styrene and benzo[b]thiophene with total selectivity and competitive rates of reaction. The recycling
of the catalyst has been achieved by the entrapment of 1 in silica by a sol-gel method to produce a
recyclable solid catalyst.
dioxide, and acetylenes in a bridging fashion.2,7 This phenom-
enon has relevance to catalytic processes occurring either on
transition-metal clusters or on metal surfaces, and a number of
A-frame complexes are capable of catalyzing homogeneous
hydrogenation and the water gas shift reactions.7c,8
Introduction
The synthesis of dinuclear metal complexes has attracted
much attention. Bimetallic rhodium complexes have been of
particular interest due to their potential to act as homogeneous
catalysts. Homobimetallic complexes can bind to unsaturated
substrates in modes not possible for their monometallic ana-
logues, and this can lead to new substrate reactivity and
cooperative effects. Stanley and co-workers1 have demonstrated
the remarkable homobimetallic cooperative activity of rhodium
catalysts in the hydroformylation reaction.
Bimetallic complexes have been frequently constructed using
a binucleating ligand as a framework. Diphosphinomethanes
have been employed to prepare a wide range of complexes,2
including heterobimetallic complexes.3 When reacting bis-
(diphenylphosphino)methane (dppm)4 with rhodium(I), there is
a strong tendency to form complexes in which two metal ions
are surrounded by two bridging ligands trans to each other
forming an “A-frame”.5,6 Rhodium complexes with two dppm
ligands arranged in a [Rh2(dppm)2] core have been rigorously
investigated.2,6 Of note is the ability of A-frame complexes to
coordinate small molecules such as carbon monoxide, sulfur
Despite concentrated effort for over 30 years, surprisingly
few homodinuclear complexes of rhodium containing only one
dppm as a bridging group have been synthesized and structurally
characterized.8,9 The absence of literature examples of these
species can be attributed to the lack of a convenient synthetic
precursor containing one dppm and a homobimetallic structure.
We now report the preparation and X-ray structural character-
ization of a new heterobridged homodinuclear complex of
rhodium containing one dppm and one chloride as bridging
groups (Figure 1). In this study we demonstrate the potential
of this complex by comparing its ability to provide selective
hydrogenation catalysts with well-established di- and triphos-
phine catalysts such as [RhCl(PPh3)3] and show how [Rh2-
(COD)2(dppm)(µ2-Cl)]BF4 (1) can be recycled by entrapping it
in a silica matrix.
(7) (a) Kubiak, C. P.; Eisenberg, R. J. Am. Chem. Soc. 1977, 99, 6129.
(b) Kubiak, C. P.; Eisenberg, R. Inorg. Chem. 1980, 19, 2726. (c) Mague,
J. T.; Sanger, A. R. Inorg. Chem. 1979, 18, 2060. (d) Olmstead, M. M.;
Lindsay, C. H.; Benner, L. S.; Balch, A. L. J. Organomet. Chem. 1979,
179, 289. (e) Cowie, M.; Mague, J. T.; Sanger, A. R. J. Am. Chem. Soc.
1978, 100, 3628.
* Corresponding author. E-mail: a.marr@qub.ac.uk; p.marr@qub.ac.uk.
† The Queen’s University of Belfast.
‡ Universita` di Pisa.
§ University of Bristol.
(1) (a) Aubrey, D. A.; Bridges, N. N.; Ezell, K.; Stanley, G. G. J. Am.
Chem. Soc. 2003, 125, 11180. (b) Matthews, R. C.; Howell, D. K.; Peng,
W. J.; Train, S. G.; Treleaven, W. D.; Stanley, G. G. Angew. Chem., Int.
Ed. 1996, 35, 2253. (c) Broussard, M.; Juma, E. B.; Train, S. G.; Peng, W.
J.; Laneman, S. A.; Stanley, G. G. Science 1993, 260, 1784. (d) Laneman,
S. A.; Stanley, G. G. AdV. Chem. Ser. 1992, 230, 349.
(2) (a) Puddephatt, R. J. Chem. Soc. ReV. 1983, 12, 99, and references
therein. (b) Chaudret, B.; Delavaux, B.; Poilblanc, R. Coord. Chem. ReV.
1988, 86, 191, and references therein.
(3) McEwan, D. M.; Pringle, P. G.; Shaw, B. L. Chem. Commun. 1982,
859.
(4) Abbreviations used: acac, acetyl acetonate; COD, 1,5-cyclooctadiene;
dppm, bis(diphenylphosphino)methane; TMS, tetramethylsilane; TOF,
turnover frequency; ICP-OES, inductively coupled plasma optical emission
spectrometry; BET, Brunner-Emmett-Teller; TEM, transmission electron
microscopy.
(8) Kubiak, C. P.; Woodcock, C.; Eisenberg, R. Inorg. Chem. 1982, 21,
2119.
(9) (a) Faraone, F.; Bruno, G.; Lo Schiavo, S.; Piraino, P. Organome-
tallics 1985, 4, 1098. (b) Faraone, F.; Bruno, G.; Lo Schiavo, S.; Bombieri,
G. Chem. Commun. 1984, 6. (c) Faraone, F.; Bruno, G.; Lo Schiavo, S.;
Piraino, P. Organometallics 1986, 5, 1400. (d) Faraone, F.; Bruno, G.; Lo
Schiavo, S.; Rotondo, E.; Piraino, P. Organometallics 1987, 6, 2502. (e)
Faraone, F.; Bruno, G.; Lo Schiavo, S.; Bombieri., G. J. Chem. Soc., Dalton
Trans. 1984, 533. (f) Faraone, F.; Bruno, G.; Lo Schiavo, S.; Nicolo`, F.;
Piraino, P. Organometallics 1985, 4, 2091. (g) Cano, M.; Heras, J. V.; Lobo,
M. A.; Pinilla, E.; Monge, M. A. Polyhedron 1994, 13, 1563. (h) County,
G. R.; Dickson, R. S.; Fallon, G. D.; Jenkins, S. M.; Johnson, J. J.
Organomet. Chem. 1996, 296, 279. (i) Chin, T. T.; Geiger, W. E.; Rheingold,
A. L. J. Am. Chem. Soc. 1996, 118, 5002. (j) Mann, B. E.; Meanwell, N.
J.; Spencer, C. M.; Taylor, B. F.; Maitlis, P. M. J. Chem. Soc., Dalton
Trans. 1985, 1555. (k) Oro, L. A.; Carmona, D.; Reyes, J. J. Organomet.
Chem. 1986, 302, 417. (l) Oro, L. A.; Pinillos, M. T.; Tiripicchio, A.;
Tiripicchio-Camellini, M. Inorg. Chim. Acta 1985, 99, L13.
(5) (a) Mague, J. T.; Mitchener, J. P. Inorg. Chem. 1969, 8, 119. (b)
Cowie, M.; Dwight, S. K. Inorg. Chem. 1980, 19, 2500.
(6) Hoffman, D. M.; Hoffmann, R. Inorg. Chem. 1981, 20, 3543.
10.1021/om060172q CCC: $33.50 © 2006 American Chemical Society
Publication on Web 06/29/2006

Dirhodium Catalyst for SelectiVe Hydrogenation
Organometallics, Vol. 25, No. 16, 2006 3913
Figure 1. Structure of the cationic core of [Rh2(COD)2(dppm)-
(µ2-Cl)]BF4 (1).
Figure 3. View of the structure of [Rh2(COD)2(µ2-dppm)(µ2-Cl)]+.
The ellipsoids of Rh, Cl, and P are at 30% probability. Selected
distances (Å) and angles (deg), Cyo ) centroids of the double bonds
of cyclooctadienes: Rh(1)-Cyo(1), 2.010(9); Rh(1)-Cyo(2),
2.130(9); Rh(1)-P(1), 2.297(3); Rh(1)-Cl, 2.411(3); Rh(2)-Cyo-
(3), 2.003(8); Rh(2)-Cyo(4), 2.121(8); Rh(2)-P(2), 2.301(3);
Rh(2)-Cl, 2.427(3); Cyo(1)-Rh(1)-Cyo(2), 86.6(4); Cyo(1)-
Rh(1)-P(1), 93.5(3); Cyo(1)-Rh(1)-Cl, 176.1(3); Cyo(2)-Rh-
(1)-P(1), 178.7(2); Cyo(2)-Rh(1)-Cl, 89.6(3); P(1)-Rh(1)-Cl,
90.3(1); Cyo(3)-Rh(2)-Cyo(4), 86.7(4); Cyo(3)-Rh(2)-P(2),
94.4(3); Cyo(3)-Rh(2)-Cl, 175.8(3); Cyo(4)-Rh(2)-P(2), 169.3(2);
Cyo(4)-Rh(2)-Cl, 89.2(3); P(2)-Rh(2)-Cl, 89.9(1); Rh(1)-Cl-
Rh2, 123.1(1); Rh(1)-P(1)-C(1), 110.4(4); P(1)-C(1)-P(2),
116.7(5); C(1)-P(2)-Rh(2), 117.5(3). The structure has been
deposited in the Cambridge database, CCDC 234930.
Figure 2. Observed and simulated 31P{1H} NMR spectra (CDCl3,
H3PO4) of [Rh2(COD)2(dppm)(µ2-Cl)]BF4 (1).
protons. Cooling the sample to -60 °C led to some broadening
of this signal, consistent with an exchange process, but
unfortunately we were unable to reach a point of coalescence.
In the absence of such motion, the methylene hydrogen
resonances often consist of two superimposed multiplets at δ
2.00-2.50.9f Resonances for the methylene protons of the dppm
ligand similar to that observed for 1 have been found for the
following complexes: [Rh2(η-C5H5)2(µ-CO)(µ-dppm)(µ-AgY)]
(Y ) CH3CO2, (C2H5)2NCS2),9c [{Rh2(η-C5H5)2(µ-CO)(µ-
dppm)}2{µ-Ag2(CO2)2CH2}],9c [Rh2(η-C5H5)2(µ-CO)(µ-dppm)-
(µ-AuPPh3)]X (X ) BF4, PF6),9f [Rh2(η-C5H5)2(µ-CO)(µ-
dppm)(µ-CuX)] (X ) Cl, I),9d and [Rh2(quin)2(CO)2(µ-dppm)]
(quin ) 2-quinaldinate, C9H6N-COO-).9g
The X-ray structure of the cation [Rh2(COD)2(µ2-dppm)(µ2-
Cl)]+ is shown in Figure 3. The two rhodium atoms are
approximately square-planar coordinated, if the centroids of the
ethenylidene groups of cyclooctadienes are considered as
ligands. The ligands around Rh(1) make an almost perfect plane,
and the metal atom deviates only 0.020 Å from it; the ligands
around Rh(2) are instead slightly more misaligned, and Rh(2)
deviates 0.098 Å from their mean plane. The two coordination
planes make a dihedral angle of 43.9°.
This angle is a consequence of the heavy puckering of the
six-membered ring Rh(1), Cl, Rh(2), P(2), C(1), P(1) mainly
resulting from the gap between the Rh(1)‚‚‚Rh(2) (4.253 Å)
distance and the bite of the diphosphine (P(1)‚‚‚P(2) (3.134 Å).
To check whether the crystal used for the structure determi-
nation corresponded to the main reaction product, the X-ray
diffraction pattern of a milled sample of the residual solid was
compared with the pattern calculated from the structure data of
1; the two patterns were almost coincident.
[Rh2(COD)2(µ2-dppm)(µ2-Cl)]BF4 (1) as a Precursor for
Homogeneous Hydrogenation. Hydrogenation of Styrene.
While testing the activity of 1 as a styrene hydroformylation
catalyst, the discovery of ethylbenzene in the product mixture
prompted us to investigate the activity toward the hydrogenation
of styrene. The hydrogenation of styrene can yield a mixture
Results and Discussions
Synthesis of [Rh2(COD)2(dppm)(µ2-Cl)]BF4 (1). The step-
wise addition of dppm in toluene and AgBF4 in acetone to a
solution of [Rh(COD)Cl]2 in dichloromethane (CH2Cl2) leads
to the formation of the binuclear complex [Rh2(COD)2(µ2-
dppm)(µ2-Cl)]BF4 (1) as an orange crystalline powder and
orange crystals. Complex 1 is air-stable as a solid, and its
solutions in dichloromethane and chloroform are stable for 1
day and 6 days, respectively.
The observed and simulated10 31P NMR spectra of 1 are
shown in Figure 2. The 31P{1H} NMR spectrum of 1 (in CDCl3,
relative to H3PO4) exhibits a pattern typical of an AA′XX′ spin
1
2
system centered at δ 27.9: JRh-P ) 150.35, JRh-P ) 2.74,
1JRh-Rh ) -0.07, 1JP-C-P ) 48.61 Hz. 1JRh-P and 1JRh-Rh values
are consistent with Rh(I) ions in a square-planar environment.9g,11
1JRh-Rh is essentially zero, and 1JRh-P is comparable to A-frame
complexes without intramolecular metal-metal bonds.12 The
values of the chemical shift and the coupling constants fall in
the range quoted for other related dinuclear rhodium complexes
with dppm as a bridging ligand to the two metal centers.9a,c,d,f,g,i
1
The H NMR spectrum of 1 (in CDCl3, relative to TMS)
shows signals attributable to the phenyl groups (δ 7.68-7.43),
COD (dCH, δ 4.77, 3.05; CH2, δ 2.13-1.58),13 and methylene
(δ 3.54) hydrogens. The methylene protons of the dppm ligand
resonate as a “triplet” centered at δ 3.54 (JP-H ) 9.41 Hz). We
suggest that this is due to a rapid conformational change in the
CP2Rh2 ring leading to fluxional equivalence of the PCH2P
(10) Budzelaar, P. H. M. gNMR V3.6; Cherwell Scientific Publishing,
IvorySoft, 1995.
(11) (a) Heras, J. V.; Cano, M.; Lobo, M. A.; Pinilla, E. Polyhedron
1989, 8, 167. (b) Uso`n, R.; Oro, L. A.; Sanau´, M.; Lahuerta, P.; Hildenbrand,
K. J. Inorg. Nucl. Chem. 1981, 43, 419.
(12) Sutherland, B. R.; Cowie, M. Inorg. Chem. 1984, 23, 1290.
(13) Slawin, A. M. Z.; Smith, M. B.; Woollins, J. D. J. Chem. Soc.,
Dalton Trans. 1996, 1283.

3914 Organometallics, Vol. 25, No. 16, 2006
Lorenzini et al.
Further experiments were conducted to probe the stability
and flexibility of the new catalyst. Under identical reaction
conditions (see Table 1 footnote) a catalytic run was carried
out using a very small quantity of [Rh2(COD)2(µ2-dppm)-
(µ2-Cl)]BF4 (1) (8.6 × 10-7 mol). After 6 h at 100 °C over 8.3
× 10-3 mol of styrene were converted selectively to ethyl
benzene, this corresponding to 9651 turnovers of a bimetallic
catalyst (4836 per rhodium). In addition the hydrogenation
activity of 1 was tested at room temperature and 28 bar (other
reaction conditions as quoted in Table 1); after 24 h full
conversion was observed.
The selectivity of the catalyst derived from 1 strongly points
to a catalytic cycle involving a Rh(I) molecular catalyst. Indeed
it has been suggested16 that the hydrogenation of the arene
functionality (absent for reactions catalyzed by 1) is diagnostic
of the formation of Rh(0). We suggest a dirhodium (dppmRh2)
molecular catalyst and evoke a cooperative effect in order to
explain the high activity and selectivity observed. The similarity
to [RhCl(PPh3)3] and [Rh(COD)(PPh3)2]BPh4 would suggest a
RhI/III cycle involving H2 oxidative addition, 1,2-migratory
insertion, and reductive elimination steps. Further work is being
conducted to understand the mechanism.
Figure 4. Common products of the hydrogenation of styrene.
of products including ethyl benzene, 1-ethyl-cyclohexene, and
ethyl cyclohexane (Figure 4).
The catalytic activity of 1 has been compared with that of
different catalyst precursors. Catalytic studies carried out on
the hydrogenation of styrene using 1 at 28 bar H2, 100 °C (Table
1) demonstrated that 1 acts as a precursor to a chemoselective
alkene hydrogenation catalyst. It should be noted that all
turnover frequencies (TOFs) are quoted per rhodium atom. The
catalysts tested were tolerant to CH2Cl2 and performed equally
well in acetone. CH2Cl2 was chosen for the experiments due to
difficulties in dissolving many of the precursors in acetone. Of
the precursors tested, only [RhCl(PPh3)3],14 [Rh(COD)(PPh3)2]-
BPh4,15 and 1 gave complete selectivity and total conversion in
2 h. Addition of dppm to [Rh(acac)(CO)2] failed to increase
the activity of this selective but sluggish catalyst. Increasing
the reaction time had no effect on the selectivity of the catalyst
derived from 1. The selectivities observed for the various
catalytic runs (1-11 Table 1) are plotted in Chart 1. It is
interesting to note that, starting from [Rh(COD)Cl]2 as the
source of rhodium, selectivity was relatively poor (72%). The
addition of 1 or 2 equiv of dppm did not improve this
considerably; however the addition of PPh3 did have an affect
on selectivity, as we would expect from the formation of a stable
homogeneous catalyst with 2 equiv of phosphine per metal.
Consistent with this hypothesis is the total (100%) selectivity
observed on addition of 4 equiv of PPh3. We conclude that the
catalytic system responsible for the selective catalyst on the
addition of 1 as a source of rhodium was not formed by the
separate addition of [Rh(COD)Cl]2 and dppm. Using a hetero-
geneous catalyst (5% Rh on carbon) hydrogenation of the arene
function was considerable, as expected for a Rh(0) catalyst.16
No comparison of turnover frequencies could be made for
these batch catalytic runs, as the reactions went to completion;
therefore a series of gas uptake experiments were conducted
for the three best catalytic systems: 1, [RhCl(PPh3)3], and
[Rh(COD)(PPh3)2]BPh4. The apparatus used (supplied by HEL
reactors) consisted of a heated, stirred autoclave attached to a
hydrogen cylinder through a solenoid control valve. This enabled
the rate of gas uptake to be measured at constant reaction
pressure. For catalyst concentrations of 8 × 10-4 mol‚dm-3 (the
concentration used in the batch autoclave runs) initial rates were
too fast to be measured reliably and the concentration of the
catalyst had to be reduced to a quarter to achieve measurable
rates. The initial rates are reported in Table 2.
Hydrogenation of 1-Hexene. The hydrogenation ability of
1 was compared with that of [Rh(acac)(CO)2], [Rh(acac)(CO)2]
+ dppm (1:1), and [RhCl(PPh3)3] (Table 3). All four catalysts
were totally selective for the hydrogenation of 1-hexene to
n-hexane.
Hydrogenation of benzo[b]thiophene. Attempts to increase
sustainability will lead to the need to use more diverse natural
materials as chemical feedstocks.17 Chemists will require an
abundance of catalysts capable of transforming difficult mol-
ecules into useful chemicals. As many of the molecules found
naturally are highly unsaturated and/or highly oxidized, hydro-
genation will be increasingly important.18 To probe the tolerance
of our new catalytic system to difficult substrates, we investi-
gated the hydrogenation of benzo[b]thiophene. A major con-
taminant of crude oil and coal, benzo[b]thiophene is an abundant
substrate for model studies.19 The hydrodesulfurization (HDS)
of crude feedstocks is required in order to prevent environmental
pollution and to improve the efficiency of catalytic reactions.
Sulfur-containing molecules such as benzo[b]thiophene are toxic
environmental pollutants in their own right, classified as
polynuclear aromatic hydrocarbons, and furthermore chemical
oxidation at sulfur can lead to the formation of oxides that cause
acid rain. The presence of sulfur in substrates for catalysis (either
chemical or automotive) often leads to irreversible poisoning
of catalytic centers.20 Heterogeneous catalysts are particularly
vulnerable to irreversible sulfur poisoning, as sulfur attached
to the metal surface at defect sites cannot be easily hydrogenated
(to H2S) and thus removed (unlike nitrogen and carbon).20
Molecular catalysts have the potential to offer invaluable
information on how to avoid sulfur poisoning, as the reactivity
at the catalytic centers can be systematically altered and the
effects on recycling assessed.
The rates measured are more than 100 times faster than those
estimated for [Rh(acac)(CO)2] (acac ) acetylacetonate) and
[Rh(acac)(CO)2] + dppm (Table 1). The three catalysts behaved
similarly under hydrogenation conditions. The solutions exhib-
ited considerably higher rates of gas uptake for the conditions
quoted compared with milder conditions. Projecting from the
initial rate data, we can surmise that the batch autoclave runs
reported in Table 1, entries 1, 2, and 3, were complete after
less than 10 min and ethyl benzene was not prone to any further
hydrogenation for catalysts 1, [RhCl(PPh3)3], and [Rh(COD)-
(PPh3)2]BPh4.
The activity of 1 as a catalyst for the hydrogenation of benzo-
[b]thiophene was screened against a number of established
homogeneous hydrogenation catalysts. The results are sum-
marized in Tables 4-8. Rates observed at 100 °C (Table 4)
were consistent with values reported in the literature (initial rates
0.3-6.2 turnovers h-1) for a range of catalysts including [RhCl-
(14) Young, J. F.; Osborn, J. A.; Jardine, F. H.; Wilkinson, G. J. Chem.
Soc., Chem. Commun. 1965, 131.
(15) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1971, 93, 2397.
(16) Weddle, K. S.; Aiken, J. D., III; Finke, R. G. J. Am. Chem. Soc.
1998, 120, 5653.
(17) Technology vision 2020, The US Chemical Industry, ACS, 1996.
(18) Poliakoff, M. Nature 2000, 407, 938.
(19) Angelici, R. J. Acc. Chem. Res. 1988, 21, 387.
(20) McAllister, B.; Hu, P. J. Chem. Phys. 2005, 122, 84709.

Dirhodium Catalyst for SelectiVe Hydrogenation
Organometallics, Vol. 25, No. 16, 2006 3915
Table 1. Homogeneous Hydrogenationa of Styrene
test no.
catalyst
[Rh]/10-4 mol dm-3
conversion/%
selectivity/%
TOF/102 h-1
1
2
3
4
5
6
7
8
9
[Rh2(COD)2(dppm)(Cl)]BF4
RhCl(PPh3)3
[Rh(COD)(PPh3)2]BPh4
[Rh(COD)Cl]2
[Rh(COD)Cl]2 + 1dppm
[Rh(COD)Cl]2 + 2dppm
[Rh(COD)Cl]2 + 2PPh3
[Rh(COD)Cl]2 + 4PPh3
[Rh(acac)(CO)2]
8.0
6.6
8.6
11.1
11.1
16.2
13.6
20.4
6.9
100
100
100
100
100
100
100
100
34
100
100
100
71
61
76
g5.0
g6.0
g4.6
>3.6
>3.6
>2.4
>2.9
g1.9
2.0
94
100
100
100
59
10
11
[Rh(acac)(CO)2] + 1dppm
Rh on carbon (5%)
6.9
6.8
31
100
1.8
g5.8
a All reactions in CH2Cl2, T ) 100 °C, P(100 °C) ) 28 bar, [styrene] ) 0.793 mol dm-3; the reaction was run for 2 h at stir rate 1200 rpm; selectivities
are quoted as % ethyl benzene.
Chart 1. Selectivity Data for the Hydrogenationa of Styrene
Rh{(η3-S(C6H4)-CHdCH2] (TRIPHOS ) MeC(CH2PPh2)3).25
In every case the only hydrogenation product observed was 2,3-
dihydrobenzo[b]thiophene (Figure 5). The rate of reaction,
reported as the turnover frequency (turnovers h-1), is at the
upper end of the literature range, and it should be noted that
this approaches rates reported by Sa´nchez-Delgado at 170 °C
and 115 bar pH2.21
for Catalysts 1-11b
The most active catalyst measured was [Rh2(COD)2(dppm)-
(Cl)]BF4 (1). The extent of hydrogenation catalyzed by 1 as a
function of time was calculated by doing separate batch
autoclave runs for 12, 15, 24, and 71 h at [Rh]:[substrate] ratio
1:20 (Table 5). Clearly there is a loss of activity as a function
of time, and this may be due to changes in the composition of
the solution (increased 2,3-dihydrobenzo[b]thiophene, decreased
benzo[b]thiophene) or catalyst deactivation.
Similar results were obtained at 125 °C (34 bar). At this
temperature the effect of altering the [Rh]:[substrate] ratio was
investigated (Table 6). For a 2 h catalytic run changes in rate
are entirely consistent with first-order variance with respect to
substrate concentration. From this we would expect a decrease
in rate as a function of time, and we deduce substrate
concentration is rate limiting (non-zero order). We conclude
that turnover frequencies calculated from batch runs at longer
reaction times are misleading, as the rate of reaction slows
during the run due to the reduction in the concentration of benzo-
[b]thiophene. At longer reaction times (136 h) the selectivity
of the catalytic system is lost, and this may be an indicator of
catalyst decomposition. The product mixture after long reaction
times is complex, containing a mixture of unreacted benzo[b]-
thiophene (BT), 2,3-dihydrobenzo[b]thiophene (DHBT), 2-me-
thylbenzo[b]thiophene (2-M-BT), 3-methylbenzo[b]thiophene
(3-M-BT), 5-methylbenzo[b]thiophene (5-M-BT), 2,7-dimeth-
ylbenzo[b]thiophene (2,7-dM-BT), 2-ethylbenzo[b]thiophene (2-
Et-BT), and 3,4-dihydro-2H-1-benzo[b]thiopyran (3,4-dH-BTP)
(Figure 6). These side products are the result of C-C bond
forming reactions, and the origin of the additional carbon atoms
is likely to be the CH2Cl2 solvent. Such side reactions are more
prevalent at 150 °C for most catalytic systems, and work at
this temperature enabled us to quantify the side products.
a Conversions are given in Table 1. b1: [Rh2(COD)2(dppm)(Cl)]BF4,
2: RhCl(PPh3)3, 3: [Rh(COD)(PPh3)2]BPh4, 4: [Rh(COD)Cl]2, 5:
[Rh(COD)Cl]2 + 1dppm, 6: [Rh(COD)Cl]2 + 2dppm, 7: [Rh(COD)Cl]2
+ 2PPh3, 8: [Rh(COD)Cl]2 + 4PPh3, 9: [Rh(acac)(CO)2], 10:
[Rh(acac)(CO)2] + 1dppm, 11: Rh on carbon (5%).
Table 2. Initial Rate of Hydrogenationa of Styrene by Gas
Uptake
[Rh]/
rate/
TOF/
catalyst
10-4 mol dm-3 mol dm-3 h-1 102 h-1
[Rh2(COD)2(dppm)(Cl)]BF4
RhCl(PPh3)3
[Rh(COD)(PPh3)2]BPh4
1.96
1.97
1.90
5.24
6.32
5.07
270
320
270
a All reactions in CH2Cl2, T ) 100 °C, P(100
) 27 bar, [styrene] )
°C)
0.793 mol dm-3, the reaction was stirred at 1000 rpm.
Table 3. Homogeneous Hydrogenationa of 1-Hexene
[Rh]/
conversion/
%
TOF/
catalyst
10-4 mol dm-3
102 h-1
[Rh(acac)(CO)2]
[Rh(acac)(CO)2] + dppm
[RhCl(PPh3)3]
7.3
7.3
6.9
7.6
32
14
100
100
1.6
0.7
g5.3
g4.8
[Rh2(COD)2(dppm)(Cl)]BF4
Results for catalytic runs at 150 °C (45 bar) employing a
range of catalysts are displayed in Tables 7 and 8. At 150 °C
the new catalyst precursor 1 forms a much more active and
selective hydrogenation catalyst than [RhCl(PPh3)3], [Rh(COD)-
(PPh3)2]BPh4, [RuHCl(CO)(PPh3)3], or [RuCl2(PPh3)3]. Total
a All reactions in CH2Cl2, T ) 100 °C, P(100 °C) ) 28 bar, [1-hexene] )
0.727 mol dm-3; the reaction was run for 2 h at stir rate 1200 rpm. The
reaction using [Rh2(COD)2(dppm)(Cl)]BF4 was also performed in acetone;
results were identical.
(PPh3)3],21,22 [Cp*Rh(MeCN)3](BF4)2],23 [Rh(COD)(PPh3)2]-
PF6,21
[RuHCl(CO)(PPh3)3],21
[OsHCl(CO)(PPh3)3],21
(23) Baralt, E.; Smith, S. J.; Hurwitz, I.; Horva´th, I. T.; Fish, R. H. J.
Am. Chem. Soc. 1992, 114, 5187
[Ir(COD)(PPh3)2]PF6,21 [RuCl2(PPh3)3],21,24 and (TRIPHOS)-
(24) Fish, R. H.; Tan, J. L.; Thormodsen, A. D. Organometallics 1985,
4, 1743.
(25) Bianchini, C.; Herrera, V.; Jimenez, M. V.; Meli, A.; Vizza, F. J.
Am. Chem. Soc. 1995, 117, 8567.
(21) Sa´nchez-Delgado, A.; Gonza´lez, E. Polyhedron 1989, 8, 1431.
(22) Fish, R. H.; Tan, J. L.; Thormodsen, A. D. J. Org. Chem. 1984, 49,
4500

3916 Organometallics, Vol. 25, No. 16, 2006
Table 4. Homogeneous Hydrogenationa of Benzo[b]thiophene
Lorenzini et al.
[Rh]/
[substrate]/
[substrate]/
[catalyst]
convn/
%
rate/
TOF/
h-1
catalyst
10-4 mol dm-3
10-2 mol dm-3
10-4 mol dm-3 h-1
1b
13.4
3.8
9.6
4.9
4.6
7.27
2.01
5.34
2.80
2.67
54
53
56
57
58
19
10
17
5
69.1
10.1
45.4
7.0
5.2
2.7
4.7
1.4
1.5
RhCl(PPh3)3
[Rh(COD)(PPh3)2]BPh4
RuHCl(CO)(PPh3)3
RuCl2(PPh3)3
5
6.7
b
a All reactions in CH2Cl2, T ) 100 °C, P(100
) 28 bar; the reaction was run for 2 h at stir rate 1200 rpm. 1 ) [Rh2(COD)2(dppm)(Cl)]BF4.
°C)
Table 5. Hydrogenationa of Benzo[b]thiophene Using
[Rh2(COD)2(dppm)(Cl)]BF4 for Different Reaction Times
material had a BET of 139.5 ( 0.25 m2/g and a metal loading
of 0.06 wt % (by inductively coupled plasma optical emission
spectrometry, ICP-OES). The 31P solid-state NMR spectrum of
the material produced is shown in Figure 7. Comparison with
Figure 2 shows a great correlation between the resonances,
suggesting that [Rh2(COD)2(dppm)(µ2-Cl)]BF4 is entrapped
intact.
Catalytic Testing of [Rh2(COD)2(dppm)(µ2-Cl)]BF4 (1)
Entrapped in Silica. Silica-entrapped 1 was tested as a reusable
catalyst for the hydrogenation of styrene (Chart 2) and benzo-
[b]thiophene (Chart 3) at 100 °C. Due to the dramatic differences
in rate for the two reactions, each styrene experiment was
conducted for 30 min and each benzo[b]thiophene experiment
for 21 h.
rate/
[Rh]/10-4
mol dm-3 10-2 mol dm-3
[substrate]/
time/ conversion/ 10-4 mol TOF/
h
%
dm-3 h-1
h-1
8.0
11.2
9.0
1.69
2.38
1.81
2.22
12
15
24
71
27
33
40
47
3.8
5.2
3.0
1.5
0.5
0.5
0.3
0.1
11.4
a All reactions in CH2Cl2, T ) 100 °C, P(100 °C) ) 28 bar, stir rate 1200
rpm.
selectivity could be maintained for 12 h using a substrate-to-
catalyst ratio of 51:1; under similar conditions (55:1) all other
catalysts tested gave considerable amounts of side products
(Table 8). The highest rate of hydrogenation measured for the
2 h run was 0.0127 mol dm-3 h-1 using 1, corresponding with
To catalyze the hydrogenation of styrene, 0.0152 g of a 0.06%
weight percentage rhodium catalyst prepared by the sol-gel
entrapment of [Rh2(COD)2(dppm)(µ2-Cl)]BF4 (1) in silica was
used. For comparison with the homogeneous catalysts the
amount of metal added corresponds to 8.9 × 10-8 mol of
rhodium and a concentration of 8.1 × 10-6 mol dm-3 in each
run. After each run the product solution was removed and
analyzed for organic product composition and rhodium con-
centration (in order to screen for catalyst leaching). In all cases
the catalyst was 100% selective, yielding only ethyl benzene.
Running the reaction for 2 h gave 100% conversion to ethyl
benzene with no ring hydrogenation. Over the 10 runs the
catalyst completed more than 6.7 × 105 turnovers, corresponding
to the hydrogenation of 6.9 cm3 of styrene. In all but one solution
the leaching rate was beyond the limit of detection (<0.003%).
The solution of run 3 contained Rh consistent with leaching of
0.003% (Rh concentration 2.4 × 10-10 mol dm-3). The
conversion of styrene (Chart 2) varies from 72% in the first
two runs to 67% in the final run, a 7% reduction in overall
activity; most of this reduction occurs between the second and
third run. The erosion of rate cannot be explained purely by
metal leaching, which is less than 0.03% (2.7 × 10-11 mol Rh)
over the 10 runs. BET analysis of the catalyst after treatment
under reaction conditions for 6 h reveals that there has been a
considerable change in the surface area of the matrix. The
effective surface area was reduced from 139.5 m2/g in the freshly
prepared catalyst to 72.3 m2/g after use. This helps to explain
the reduction in activity between runs 2 and 3 (Chart 2). We
suggest that the reduction in surface area is associated with a
closing of the structure, causing a small loss of rhodium
(0.003%) and making the remaining active catalytic centers less
available to the substrate. We are currently investigating
entrapment methods that yield more robust materials.27 Trans-
mission electron microscopy (TEM) was used to analyze the
homogeneity of the used catalyst (Figure 8). Rhodium could
be detected by EDX metal analysis, but no Rh(0) nano- or
microparticles were observed. This result in addition to the
complete selectivity observed suggests that the entrapped metal
remains in a molecular form.
a TOF of 7.2 h-1
.
The increased instances of C-C coupling reactions at higher
temperatures are consistent with a high kinetic barrier for this
reaction, and increased selectivity with increased substrate
concentration is consistent with a rate of hydrogenation that is
substrate-concentration dependent. The high selectivity exhibited
by 1 compared to the other catalysts tested may be due to the
presence of a bimetallic active core that restricts the modes of
binding of benzo[b]thiophene. Baralt et al.23 and Sa´nchez-
Delgado et al.26 have noted that benzo[b]thiophene can bind to
rhodium via S-bonded, η2, η,4 η5, or η6 modes, and these
additional modes may provide lower energy reaction pathways
for side reactions by mechanisms such as ring slippage.
To test the recycling of catalysts derived from 1, we prepared
a silica-entrapped catalyst.
Entrapment of 1 in a Silica Matrix. Previously we reported
the facile recycling of a totally selective solid hydrogenation
catalyst made by the sol-gel entrapment of [RhCl(PPh3)3] by
conventional sol-gel and ionic-liquid-mediated routes.27 To
entrap28 a molecular catalyst, the metal complex is added to
the sol-gel recipe for the solid oxide material, and on formation
of the oxide, the catalyst becomes imprisoned (entrapped) within
the solid matrix. Provided a material of the correct physical
nature (porosity, surface area) is produced, the result is a catalyst
with molecular reactivity that can be recycled many times in
catalytic experiments. Blum and Avnir29,30 have reported the
advantages of this method for producing catalysts that have a
high resistance to poisoning and the use of entrapment to enable
the combination of incompatible catalysts within the same
reaction vessel and facilitate cascade processes.
We were able to entrap 1 in silica by a conventional sol-gel
method similar to that reported previously.27 The resultant
(26) Sa´nchez-Delgado, R. A.; Herrera, V.; Rinco´n, L.; Andriollo, A.;
Martin, G. Organometallics 1994, 13, 553.
(27) Craythorne, S. J.; Crozier, A. R.; Lorenzini, F.; Marr, A. C.; Marr,
P. C. J. Organomet. Chem. 2005, 690, 3518.
(28) Blum, J.; Avnir, D.; Schumann, H. Chemtech 1999, 29, 32.
(29) Gelman, F.; Blum, J.; Avnir, D. J. Am. Chem. Soc. 2002, 124, 14460.
(30) Gelman, F.; Blum, J.; Avnir, D. New J. Chem. 2003, 27, 205.
Taking 72% conversion in 30 min as a measure of the activity
of the catalyst, the rate was 1.14 mol dm-3 h-1 and the TOF

Dirhodium Catalyst for SelectiVe Hydrogenation
Organometallics, Vol. 25, No. 16, 2006 3917
Table 6. Effects of Altering [Rh]:[Substrate] Ratio on Rates of Hydrogenationa of Benzo[b]thiophene Catalyzed by
[Rh2(COD)2(dppm)(Cl)]BF4 at 125 °C
[Rh]/10-4
mol dm-3
[substrate]/10-2
mol dm-3
[substrate]/
[catalyst]
time/
h
conversion/
%
rate/10-4
TOF/
h-1
mol dm-3 h-1
8.8
11.6
9.4
8.4
11.4
0.85
2.32
5.20
0.79
2.37
10
20
55
10
21
2
2
2
12
12
20
23
18
42
49
8.5
26.4
46.8
2.8
1.0
2.3
5.0
0.3
0.9
9.7
a All reactions in CH2Cl2, T ) 125 °C, P(125
) 34 bar, stir rate 1200 rpm.
°C)
Table 7. Hydrogenationa of Benzo[b]thiophene at 150 °C
[Rh]/10-4
mol dm-3
[substrate]/
[substrate]/
[catalyst]
time/
h
convn
(select.)/%
rate/10-4
catalyst
10-2 mol dm-3
mol dm-3h-1
TOF/h-1
1b
1
8.2
7.8
17.6
11.4
8.2
6.0
8.4
4.2
9.7
0.78
2.09
9.38
5.76
2.18
2.29
0.77
2.29
5.34
3.03
2.77
10
27
53
51
27
38
9
55
55
55
55
2
2
2
12
12
14
15.5
2
22 (100)
26 (100)
27 (100)
40 (100)
25 (80)
22 (42)
30 (80)
8 (27)
8.6
27.2
126.6
19.2
3.6
1.5
1.2
2.4
13.5
1.5
1.0
3.5
7.2
1.7
0.4
0.3
0.1
0.6
1.4
0.3
0.5
1c
1
1
1
1
RhCl(PPh3)3
[Rh(COD)(PPh3)2]+
RuHCl(CO)(PPh3)3
RuCl2(PPh3)3
2
2
2
13 (39)
7 (14)
7 (25)
5.5
5.0
2.4
a All reactions in CH2Cl2, T ) 150 °C, P(150 °C) ) 45 bar, stir rate 1200 rpm. Selectivity is quoted as percentage 2,3-dihydrobenzo[b]thiophene; rate is
quoted for hydrogenation only. b 1 ) [Rh2(COD)2(dppm)(Cl)]BF4. c The conversion for a reaction run in toluene was far poorer (7% compared with 27%
in CH2Cl2).
Table 8. Percentage Composition of the Product Solutions Yielded from the Hydrogenationa of Benzo[b]thiophene at 150 °Cb
catalyst
DBT
2-M-BTc
3-M-BT
5-M-BT
2,7-dM-BT
2-Et-BT
3,4-dH-BTP
1d
100
27
39
14
25
0
43
36
41
33
0
3
9
3
2
0
18
16
35
33
0
6
0
0
0
0
2
0
5
5
0
1
0
2
2
RhCl(PPh3)3
[Rh(COD)(PPh3)2]BPh4
RuHCl(CO)(PPh3)3
RuCl2(PPh3)3
a All reactions in CH2Cl2, T ) 150 °C, P(150 °C) ) 45 bar; the reaction was run for 2 h at stir rate 1200 rpm. b Further reaction conditions are quoted in
Table 7, entries 3, 8, 9, 10, and 11. c Abbreviations are explained in Figure 6. d 1 ) [Rh2(COD)2(dppm)(Cl)]BF4.
Figure 5. Hydrogenation of benzo[b]thiophene to 2,3-dihydroben-
zo[b]thiophene.
Figure 7. Solid-state 31P spectrum of [Rh2(COD)2(dppm)(µ2-Cl)]-
BF4 (1) entrapped in silica.
Figure 6. Side products from the hydrogenation of benzo[b]-
[b]thiophene (Chart 3). For comparison with the homogeneous
thiophene in CH2Cl2.
catalysts this would correspond to a concentration of 5.97 ×
1.4 × 105 h-1. This TOF is 5 times higher than that achieved
10-5 mol dm-3. In all cases the only product detected was 2,3-
homogeneously (Table 2). In all runs the TOF is between 1.3
dihydrobenzo[b]thiophene. This is consistent with the homo-
geneous catalyst testing at 100 °C. There is a 5% variation in
and 1.4 × 105 h-1. This result illustrates the role of the matrix
in reinforcing molecular reactivity and contrasts with many other
the weight of benzo[b]thiophene added to the catalyst throughout
methods of heterogenizing homogeneous catalysts that lead to
the 10 runs, and therefore a small variance in conversion, such
a reduction in activity.31 High activity was also observed27 for
as that shown in the first five catalytic runs (first 105 h), can
the hydrogenation of styrene catalyzed by [RhCl(PPh3)3]
be expected. After the fifth run there is a noticeable drop in
performance, and the overall trend over the 10 runs is
downward. No rhodium leaching could be detected by ICP-
entrapped by an ionic-liquid-mediated method (TOF 6.84 × 104
h-1).
A 0.1024 g sample of the 0.06% weight percentage rhodium
OES analysis of the product solutions. Assuming 20% conver-
catalyst was used repeatedly for the hydrogenation of benzo-
sion to be the true activity of the catalyst before decomposition,
(31) Cole-Hamilton, D. J. Science 2003, 299, 1702.
this represents a rate of 5.06 × 10-4 mol dm-3 h-1 and a TOF

3918 Organometallics, Vol. 25, No. 16, 2006
Lorenzini et al.
Chart 2. Percentage Conversion for the Hydrogenationa of
Styrene Catalyzed by Recycled Entrapped 1
degassed prior to use and saturated with inert or reactant gas.
15
[Rh(COD)Cl]2,32 [RhCl(PPh3)3],14 and [Rh(COD)2(PPh3)2]BPh4
were prepared by established routes. [Rh(acac)(CO)2], [RuHCl(CO)-
(PPh3)3], [RuCl2(PPh3)3], and benzo[b]thiophene were purchased
from Aldrich. All other reagents were used as commercially
1
available. H and 31P NMR spectra were obtained on a Bruker
1
Advance DPX 300 spectrometer. Chemical shifts of the H NMR
spectrum are reported on CDCl3 solutions relative to Me4Si.
Chemical shifts of the 31P NMR spectra are reported on CDCl3
solutions relative to 85% H3PO4. The GC-MS were performed by
ASEP, The Queen’s University of Belfast; the GC column used
was a Restek rtx5 60 m within a Perkin-Elmer Turbo-Mass GC
mass spectrometer. Solid-state NMR experiments were performed
on a Bruker Avance 400 NMR spectrometer operating at 100.61
MHz for 31P. High-power decoupling experiments were carried out
using a standard HPDEC pulse sequence. The samples were spun
at magic angle at a rate of 14.031 kHz. A 1P π/2 pulse time of 3.5
µs and a recycle delay of 10 s between scans were used for all the
samples. Surface area was calculated by the BET method on a
Micromeritics ASAP 2010.
a All runs in CH2Cl2, 28 bar H2, 100 °C, [styrene] ) 0.793 mol dm-3
,
each run was 30 min, stir rate 1200 rpm.
Chart 3. Percentage Conversion for the Hydrogenationa of
Benzo[b]thiophene Catalyzed by Recycled Entrapped 1
The catalytic tests were conducted using a Parr Series 5500 high-
pressure stainless steel 50 cm3 compact laboratory reactor stirred
by a magnetically driven four-blade impeller. Reaction time was
started once the reactor had heated to the reaction temperature.
[Rh2(COD)2(dppm)(µ2-Cl)]BF4 (1). A solution of dppm (0.207
g, 0.538 mmol) in toluene (3 cm3) was added to a solution of
[Rh(COD)Cl]2 (0.255 g, 0.516 mmol) in CH2Cl2 (4 cm3). After
stirring for 5 min, a solution of AgBF4 (0.192 g, 0.958 mmol) in
acetone (9 cm3) was added. A white solid (AgCl) instantly
precipitated from the brown solution. The resulting mixture was
stirred for 30 min. The solid was then filtered and washed with
CH2Cl2 (30 cm3), until the CH2Cl2 (then added to the filtrate)
remained colorless. The volume of the filtrate was then reduced
under reduced pressure, and the product precipitated by cooling to
-18 °C or adding diethyl ether. 1 was obtained as an orange
crystalline powder or orange crystals. The product was filtered,
washed with diethyl ether (2 × 7 cm3), and dried in vacuo (P ) 4
× 10-2 Torr) at room temperature. The total yield of the product
was 0.413 g (86%); 1H NMR (in CDCl3, relative to TMS) δ 7.68-
7.43 (20H, m, H phenyl), 4.77, 3.05 (8H, m, dCH COD), 2.13-
1.58 (16H, m, CH2 COD), 3.54 ppm (2H, t, PCH2P); 31P{1H} NMR
(in CDCl3, relative to H3PO4) δ 27.9 ppm (AA′XX′ system, 1JRh-P
) 150.35, JRh-P ) 2.74, JRh-Rh ) -0.07, JP-C-P ) 48.61 Hz).
Anal. Calcd for C41H46BClF4P2Rh2: C, 53.02; H, 4.99; Cl, 3.82;
F, 8.18. Found: C, 53.37; H, 4.94; Cl + F, 12.15. FAB-MS (m/z):
841 [M+] (calcd 841). Samples of the powder and crystals were
used for X-ray diffraction studies.
Crystal data for 1, C41H46BClF4P2Rh2: Mr ) 928.80, monoclinic,
space group P21/n (no. 14), a ) 10.220(3) Å, b ) 14.987(4) Å, c
) 25.454(9) Å, â ) 96.67(1)°, V ) 3872(2) Å3, T ) 293(2) K, Z
) 4, µ(Mo KR) ) 1.053 mm-1, 4844 independent reflections
measured, dcalc ) 1.593 g cm-3, R1 ) 0.0600, wR2 ) 0.1160 (for
I > 2σ(I)), and 447 refined parameters. Orientational disorder was
present in the anion. The structure has been deposited in the
Cambridge database, CCDC 234930.
a All runs in CH2Cl2, 28 bar H2, 100 °C, [benzo[b]thiophene] ) 0.05
mol dm-3, each run was 21 h, stir rate 1200 rpm.
of 8.5 h-1, consistent with the best result obtained in homoge-
neous solution (7.2 at 150 °C, see Table 7). The successful
operation of catalyst 1 for 100 h is consistent with observations
under homogeneous conditions in which substrate-limited
activity was maintained for 71 h. We attribute the loss of activity
after this to decomposition or rearrangement of the catalyst. This
is not accompanied by a loss of selectivity.
2
1
1
Conclusion
We have reported a straightforward route for the preparation
of the rare [(µ-dppm)Rh2] functionality and have demonstrated
that [Rh2(COD)2(dppm)(µ2-Cl)]BF4 (1) acts as a precursor for
a chemo- and regioselective hydrogenation catalyst. Rates of
reaction are similar to or greater than those obtained with catalyst
precursors containing g2 phosphines per metal atom. The
entrapment of 1 within a silica matrix by the sol-gel method
yields a catalyst with identical selectivity and enhanced activity
in the hydrogenation of styrene and benzo[b]thiophene that can
be recycled at least 10 times.
Silica Entrapment of 1. Tetraethoxysilane (TEOS) (2.42 g,
0.0107 mol) was added to ethanol (1.73 g, 0.03604 mol) and stirred
for 20 min in a plastic container. Four drops of water (distilled)
were added and stirred in for a further 10 min. 1 (0.096 g, 0.000103
mol) was dissolved in the minimum quantity of dichloromethane
(0.5 cm3). This catalyst solution was added to the mixture and stirred
for 60 s. Four drops of benzylamine were added, and the mixture
was vigorous stirred until the gel point was reached. The magnetic
flea was removed, and the sample was left to age for 1 week, after
which time the sample was ground into a powder with a mortar
Experimental Section
General Procedures. All preparations were carried out in
standard Schlenk tubes under a dinitrogen atmosphere. Elemental
analyses were performed by ASEP, The Queen’s University of
Belfast. Solid samples of the synthesized complexes were stored
at room temperature. CH2Cl2 and diethyl ether were purified and
dried by standard procedures. Toluene and acetone were thoroughly
(32) Schenck, T. G.; Downes, J. M.; Milne, C. R. C.; MacKenzie, P.
B.; Boucher, H.; Whelan, J.; Bosnich, B. Inorg. Chem. 1985, 24, 2334.

Dirhodium Catalyst for SelectiVe Hydrogenation
Organometallics, Vol. 25, No. 16, 2006 3919
Figure 8. TEM images of silica-entrapped 1 after catalysis.
and pestle. Soxhlet extraction was carried out with CH2Cl2 to
remove any catalyst that was not entrapped, and the resultant pale
yellow powder was dried in a desiccator over silica gel. The product
contained 0.06% weight percentage rhodium by ICP-OES, 31P NMR
δ 28.3 ppm.
ICP-OES Analysis of the Entrapped Catalyst. A mixture of
concentrated HCl (11.5 M, 2.5 cm3) and concentrated HNO3 (16
M, 2.5 cm3) was heated and added to the sample to be digested
(0.01 g). The solution was stirred until homogeneous, and then
distilled water was added to make the volume up to 100 cm3. Once
cooled, a 10 cm3 aliquot was submitted for analysis on a
ThermoElemental IRIS Duo.
ICP-OES Analysis of Product Solutions. All volatile organics
were removed under vacuum, and the resultant residue was
dissolved by digesting in a preheated mixture of concentrated HCl
(11.5 M, 0.25 cm3) and HNO3 (16 M, 0.25 cm3). Once homoge-
neous, distilled water was added to make the volume up to 10 cm3.
Once cooled, the solution was submitted for analysis on a
ThermoElemental IRIS Duo.
Homogeneous Catalytic Experiments. Batch Catalytic Runs.
Hydrogenation of Styrene and 1-Hexane. The catalyst and
modifier (if required, see Tables 1 and 3 for rhodium concentration)
were dissolved in CH2Cl2 (10 cm3) and injected into the autoclave.
Styrene (1 cm3, 0.0087 mol) or 1-hexane (1 cm3, 0.0080 mol) was
injected, and the autoclave was pressurized to 20 bar with H2. The
stirrer was started (stir rate 1200 rpm), and the autoclave was heated
to 100 °C (internal temperature). The pressure at 100 °C was 28
bar.
10-6 mol) in styrene (1 cm3, 0.0087 mol) and CH2Cl2 (10 cm3)
was charged to the reactor under N2. The reactor vessel was then
purged with H2 (BOC, 99.995% purity) for 2 min at a flow rate of
50 cm3 min-1, after which the magnetically driven gas entrainment
impeller was set to 1000 rpm and the vessel pressurized to 10 bar
of H2. The temperature was then gradually increased until a constant
temperature of 100 °C was attained. The reactor was then
immediately pressurized to 27 bar of H2, which corresponded to
reaction t ) 0. The mass flow controller constantly replaced
consumed H2, maintaining the reactor pressure. Rate data were
calculated from the rate of uptake of H2 to the solution.
Silica-Entrapped Catalyst Recycle. Hydrogenation of Styrene.
A 0.0152 g sample of the 0.06% Rh in silica catalyst was weighed
into the autoclave. The autoclave and catalyst were evacuated and
flushed with nitrogen and CH2Cl2 (10 cm3), and styrene (1 cm3,
0.0087mol) was injected. The autoclave was pressurized to 20 bar
with H2 and the stirrer started (stir rate 1200 rpm). The apparatus
was heated to 100 °C (internal temperature) for 30 min, rapidly
cooled, and degassed, and the liquid products were removed and
analyzed for rhodium content by ICP-OES and organic content by
GC and GC-MS.
The autoclave containing the solid catalyst was sealed, evacuated,
and flushed with nitrogen. Further aliquots of CH2Cl2 (10 cm3) and
styrene (1 cm3, 0.0087 mol) were injected. The autoclave was
pressurized, stirred, heated, cooled, degassed, and analyzed as
before. This procedure was repeated until 10 catalytic runs had been
completed.
Hydrogenation of Benzo[b]thiophene. A 0.1024 g sample of
the 0.06% Rh on silica catalyst was weighed out into the autoclave.
A solution of benzo[b]thiophene (0.0713 g, 5.31 × 10-4mol) in
CH2Cl2 (10 cm3) was injected. The autoclave was pressurized to
20 bar with H2 and the stirrer started (1200 rpm). The autoclave
was heated to 100 °C for 21 h and then cooled and degassed. Then
the liquids were removed and analyzed for Rh content and product
composition.
After the reaction time the autoclave was cooled rapidly and
degassed and the resultant solution analyzed by GC and GC-MS.
Hydrogenation of Benzo[b]thiophene. The benzo[b]thiophene
and catalyst were dissolved in CH2Cl2 (10 cm3) and injected into
the autoclave (for rhodium and benzo[b]thiophene concentrations
see Tables 4-7). The autoclave was pressurized to 20 bar with
H2. The stirrer was started (stir rate 1200 rpm), the autoclave was
heated to 100, 125, or 150 °C, and the resultant pressure was 28,
34, or 45 bar, respectively.
The autoclave containing the solid catalyst was resealed and
evacuated, then flushed with nitrogen. The procedure was repeated
nine further times using 0.0683, 0.0720, 0.0708, 0.0718, 0.0698,
0.0711, 0.0719, 0.0720, and 0.0709 g of benzo[b]thiophene,
respectively.
After the reaction time the autoclave was cooled rapidly and
degassed and the resultant solution analyzed by GC and GC-MS.
Gas Uptake Experiments. The hydrogen uptake experiments
were carried out in a 50 cm3 stainless steel Hazards Evaluation
Laboratories (HEL) reactor system, equipped with on-line control
and analysis of reaction temperature, pressure, and gas uptake. The
gas uptake and reactor pressure control utilized a Brooks mass flow
controller, and the temperature control employed the use of an
external oil heating jacket (Julabo F32). The temperature was
monitored in the reaction solution.
Acknowledgment. We thank Prof. Brian Vincent, University
of Bristol, for his help, The McClay Trust (F.L.) and the EPSRC
for funding, ASEP and QUESTOR (The Queen’s University
Belfast) for analysis, and Johnson Matthey PLC for the loan of
precious metals.
A solution of 1 (0.0010 g 2.2 × 10-6 mol), [RhCl(PPh3)3] (0.0020
g, 2.2 × 10-6 mol), or [Rh(COD)(PPh3)2]BPh4 (0.0022 g, 2.1 ×
OM060172Q