Mimicking the HDS activity of ruthenium-based catalysts. Homogeneous hydrogenolysis of benzo[b]thiophene

Article abstract of DOI:10.1021/om9800804

The reaction of [(triphos)RuH(BH₄)] (1) in THF with KOBu^t yields the novel trihydride complex K[(triphos)RuH₃] (2) and BH₂OBu^t (triphos = MeC(CH₂PPh₂)₃). The ruthenate complex 2 can also be synthesized by hydrogenation (30 bar of H₂) in THF of the tris(acetonitrile) complex [(triphos)Ru(NCMe)₃](BPh₄)₂ (3) in the presence of a 5-fold excess Of BH₂OBu^t at 40 °C, This reaction produces a mixture Of NH₂Et, NHEt₂, NEt₃, and NH₃ as a result of MeCN hydrogenation, followed by amine redistribution reactions. Compound 2 is isolated in analytically pure form as [K(C₁₂H₂₄O₆)][(triphos)RuH₃] (2a) by recrystallization from THF/n-hexane in the presence of 18-crown-6 ether. In the presence of a strong base such as KOBu^t, both 1 and 3 are effective catalyst precursors for the homogeneous hydrogenolysis of benzo[b]thiophene (BT) to 2-ethylthiophenol (ETP) in THF under mild reaction conditions (>70 °C, 30 bar of H₂). The hydrogenolysis rate increases with the concentration of the base, which, depending on the catalyst precursor, may play up to three distinct roles in the catalytic reactions. It promotes the formation and stabilization of the catalytically active species (i.e. the 16e^- fragment [(triphos)RuH]^-) and speeds up the hydrogenolysis rate, delivering the ETP product into the solution as 2-ethylthiophenolate potassium salt. High-pressure ³¹P{¹H} and ¹H NMR experiments (HPNMR) in sapphire tubes sealed by titanium-alloy valves show that the interaction of 2 with BT at ≥ 70 °C in THF-d₈ selectively yields the dihydride thiolate complex K[(triphos)Ru(H)₂(o-S(C₆H₄)C₂H ₅)] (5) under H₂ and the vinylthiophenolate complex K[(triphos)Ru(η³-S(C₆H₆)CH=CH₂)] (6) under N₂. Compound 6 in THF transforms into 5 by treatment with H₂ even at room temperature. Under catalytic conditions at 70 °C, 5 and 2 are the only NMR-detectable species in equilibrium concentrations that depend on the temperature and on the base concentration. The hydrogenolysis mechanism is proposed to involve C-S insertion of ruthenium into the C₂-S bond of BT to give a 2-vinylthiophenolate ligand, followed by hydrogenation of the vinyl moiety and reductive elimination of the thiol. This latter step is accelerated by the strong Br?nsted base. The possible similarity in the hydrogenolysis reactions catalyzed by the present soluble complexes to those occurring in the hydrodesulfurization of fossil fuels over Ru-promoted heterogeneous catalysts is discussed.

Full text of DOI:10.1021/om9800804

2636
Organometallics 1998, 17, 2636-2645
Mim ick in g th e HDS Activity of Ru th en iu m -Ba sed
Ca ta lysts. Hom ogen eou s Hyd r ogen olysis of
Ben zo[b]th iop h en e
Claudio Bianchini,* Andrea Meli, Simonetta Moneti, and Francesco Vizza
Istituto per lo Studio della Stereochimica ed Energetica dei Composti di Coordinazione,
ISSECC-CNR, Via J . Nardi 39, I-50132 Firenze, Italy
Received February 6, 1998
The reaction of [(triphos)RuH(BH4)] (1) in THF with KOBut yields the novel trihydride
complex K[(triphos)RuH3] (2) and BH2OBut (triphos ) MeC(CH2PPh2)3). The ruthenate
complex 2 can also be synthesized by hydrogenation (30 bar of H2) in THF of the
tris(acetonitrile) complex [(triphos)Ru(NCMe)3](BPh4)2 (3) in the presence of a 5-fold excess
of BH2OBut at 40 °C. This reaction produces a mixture of NH2Et, NHEt2, NEt3, and NH3 as
a result of MeCN hydrogenation, followed by amine redistribution reactions. Compound 2
is isolated in analytically pure form as [K(C12H24O6)][(triphos)RuH3] (2a ) by recrystallization
from THF/n-hexane in the presence of 18-crown-6 ether. In the presence of a strong base
such as KOBut, both 1 and 3 are effective catalyst precursors for the homogeneous
hydrogenolysis of benzo[b]thiophene (BT) to 2-ethylthiophenol (ETP) in THF under mild
reaction conditions (g70 °C, 30 bar of H2). The hydrogenolysis rate increases with the
concentration of the base, which, depending on the catalyst precursor, may play up to three
distinct roles in the catalytic reactions. It promotes the formation and stabilization of the
catalytically active species (i.e. the 16e- fragment [(triphos)RuH]-) and speeds up the
hydrogenolysis rate, delivering the ETP product into the solution as 2-ethylthiophenolate
1
potassium salt. High-pressure 31P{1H} and H NMR experiments (HPNMR) in sapphire
tubes sealed by titanium-alloy valves show that the interaction of 2 with BT at g70 °C in
THF-d8 selectively yields the dihydride thiolate complex K[(triphos)Ru(H)2(o-S(C6H4)C2H5)]
(5) under H2 and the vinylthiophenolate complex K[(triphos)Ru(η3-S(C6H6)CHdCH2)] (6)
under N2. Compound 6 in THF transforms into 5 by treatment with H2 even at room
temperature. Under catalytic conditions at 70 °C, 5 and 2 are the only NMR-detectable
species in equilibrium concentrations that depend on the temperature and on the base
concentration. The hydrogenolysis mechanism is proposed to involve C-S insertion of
ruthenium into the C2-S bond of BT to give a 2-vinylthiophenolate ligand, followed by
hydrogenation of the vinyl moiety and reductive elimination of the thiol. This latter step is
accelerated by the strong Brønsted base. The possible similarity in the hydrogenolysis
reactions catalyzed by the present soluble complexes to those occurring in the hydrodes-
ulfurization of fossil fuels over Ru-promoted heterogeneous catalysts is discussed.
In tr od u ction
production of the so-called city diesel.2 Catalysts com-
prising late transition metals such as platinum and
palladium are definitely turning out to be most efficient
for the deep HDS (<15 ppm sulfur) of naphthas after
oil has been hydrotreated over Co(Ni)-Mo-S phases.3
The high cost of the noble metals and the need to
simplify the industrial process motivate the current
research efforts aimed at developing a new generation
of promoted Mo(W) catalysts.
Catalysts based on Co(Ni)-Mo sulfides supported on
γ-Al2O3 have been used extensively over the last 50
years to remove sulfur from fossil fuel feedstocks via
hydrodesulfurization (HDS) (eq 1).1
CxHyS + 2H2 cat.8 CxHy+2 + H2S
(1)
Among the various approaches to rational catalyst
design, modeling studies of the HDS chemistry of
transition-metal compounds as either metal sulfides1,4
The increasing demand for low-sulfur fuels and the
need to process heavy oil supplies are precluding
conventional catalysts, particularly with regard to the
(1) (a) Topsøe, H.; Clausen, B. S.; Massoth, F. E. Hydrotreating
Catalysis; Springer-Verlag: Berlin, 1996. (b) Scherzer, J .; Gruia, A. J .
Hydrocracking Science and Technology; Marcel Dekker: New York,
1996. (c) Gates B. C. Catalytic Chemistry; Wiley: New York, 1992;
Chapter 5, p 390. (d) Prins, R.; deBeer, V. H. J .; Somorjai, G. A. Catal.
Rev.-Sci. Eng. 1989, 31, 1. (e) van den Berg, J . P.; Lucien, J . P.;
Germaine, G.; Thielemans, G. L. B. Fuel Process. Technol. 1993, 35,
119.
(2) Cooper, B. H.; Donnis, B. B. L. Appl. Catal., A 1996, 137, 203.
(3) (a) Mignard, S.; Marchal, N.; Kasztelan, S. Bull. Soc. Chim. Belg.
1995, 104, 259. (b) Maxwell, J . E.; Naber, E.; deJ ong, K. P. Appl. Catal.,
A 1994, 113, 153. (c) Navarro, R.; Pawelec, B.; Fierro, J . L. G.;
Vasudevan, P. T.; Cambra, J . F.; Arias, P. L. Appl. Catal., A 1996,
137, 269. (d) Cooper, B. H.; Stanislaus, A.; Hannerup, P. N. Hydro-
carbon Process. 1993, 83. (e) Sugioka, M.; Sado, F.; Matsumoto, Y.;
Maesaki, N. Catal. Today 1996, 29, 255.
S0276-7333(98)00080-6 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 05/15/1998

HDS Activity of Ruthenium-Based Catalysts
Organometallics, Vol. 17, No. 12, 1998 2637
or organometallic complexes5 point to ruthenium as one
of the most active metals for the activation and/or
degradation of the thiophenes, i.e. the sulfur compounds
in petroleum which are the most difficult to desulfurize
by hydrotreating. The optimistic picture provided by
the model studies, however, contrasts with the fact that
Ru-promoted catalysts are very rarely employed in
refineries due to their quick deactivation. Similarly, the
wealth, in number and variety of the stoichiometric
transformations of thiophenes assisted by soluble ru-
thenium complexes6 contrasts with the scarcity of
homogeneous catalytic reactions.5,7 We therefore de-
cided to examine the catalytic activity of various ruthe-
nium precursors in the hydrogenolysis and hydrogena-
tion reactions of thiophenic molecules. To this end, we
applied concepts and technologies previously and suc-
cessfully developed for homogeneous rhodium and iri-
dium catalysts.8 In agreement with the heterogeneous
studies, we were gratified to find that, under comparable
conditions, ruthenium is catalytically more active than
rhodium or iridium.
Sch em e 1
Reaction 2 involves C-S opening of BT, followed by
hydrogenation of the C-S inserted product, and thus
represents an important segment (steps a and b) of one
of the paths through which the desulfurization of BT to
ethylbenzene and H2S is proposed to take place over
heterogeneous HDS catalysts (Scheme 1).1
Exp er im en ta l Section
The present paper constitutes a detailed account of
the first examples of homogeneous hydrogenolysis of
benzo[b]thiophene (BT) to 2-ethylthiophenol (ETP) (eq
2) effectively catalyzed by ruthenium complexes.
Gen er a l In for m a tion . All reactions and manipulations,
except as stated otherwise, were routinely performed under a
nitrogen atmosphere by using standard Schlenk techniques.
Reactions under controlled pressure of hydrogen were per-
formed with a stainless steel Parr 4565 reactor equipped with
a Parr 4842 temperature and pressure controller. The ruthe-
nium complex [(triphos)Ru(H)BH4 (1) was prepared as previ-
ously described.9 The complex [(triphos)Ru(NCMe)3](BPh4)2
(3) was prepared from the known compound [(triphos)Ru-
(NCMe)3](SO3CF3)210 by a metathetical reaction with NaBPh4
in McCN/ethanol. All the isolated metal complexes were
collected on sintered-glass frits and washed with appropriate
(4) (a) Pecoraro, T. A.; Chianelli, R. R. J . Catal. 1981, 67, 439. (b)
Harris, S.; Chianelli, R. R. In Theoretical Aspects of Heterogeneous
Catalysis; Moffat, J . B., Ed.; van Nostrand Reinhold Catalysis Series;
Van Nostrand Reinhold: New York, 1990; p 206. (c) Chianelli, R. R.
Catal. Rev.-Sci. Eng. 1984, 26, 361. (d) Topsøe, H.; Clausen, B. S.;
Topsøe, N.-J .; Hyldoft, J .; Nørskov, J . K. Prepr.-Am. Chem. Soc., Div.
Pet. Chem. 1993, 38, 638. (e) Topsøe, H.; Clausen, B. S.; Topsøe, N.-J .;
Nørskov, J . K.; Ovesen, C. V.; J acobsen, C. J . H. Bull. Soc. Chim. Belg.
1995, 104, 283. (f) Kasztelan, S. Appl. Catal., A 1992, 83, L1. (g) Xiao,
F.-S.; Xin, Q.; Guo, X.-X. Appl. Catal., A 1993, 95, 21. (h) Mitchell, P.
C. H.; Scott, C. E. Bull. Soc. Chim. Gelg. 1984, 3, 619.
(5) (a) Bianchini, C.; Meli, A. In Applied Homogeneous Catalysis with
Organometallic Compounds; Cornils, B., Herrmann, W. A., Eds.;
VCH: Weinheim, Germany, 1996; Vol. 2, p 969. (b) Bianchini, C.; Meli,
A. J . Chem. Soc., Dalton Trans. 1996, 801. (c) Angelici, R. J . Bull. Soc.
Chim. Belg. 1995, 104, 265. (d) Angelici, R. J . In Encyclopedia of
Inorganic Chemistry; King, R. B., Ed.; Wiley: New York, 1994; Vol. 3,
p 1433. (e) Sa´nchez-Delgado, R. A. J . Mol. Catal. 1994, 86, 287. (f)
Rauchfuss, T. B. Prog. Inorg. Chem. 1991, 39, 259. (g) Angelici, R. J .
Coord. Chem. Rev. 1990, 105, 61. (h) Angelici, R. J . Acc. Chem. Res.
1988, 21, 387.
(6) (a) Bianchini, C.; Casares, J . A.; Osman, R.; Pattison, D. I.;
Peruzzini, M.; Perutz, R. N.; Zanobini, F. Organometallics 1997, 16,
4611. (b) Hachgenei, J . W.; Angelici, R. J . Angew. Chem., Int. Ed. Engl.
1987, 26, 909. (c) Hachgenei, J . W.; Angelici, R. J . J . Organomet. Chem.
1988, 355, 359. (d) Krautscheid, H.; Feng, Q.; Rauchfuss, T. B.
Organometallics 1993, 12, 3273. (e) Feng, Q.; Rauchfuss, T. B.; Wilson,
S. R. Organometallics 1995, 14, 2923. (f) Luo, S.; Rauchfuss, T. B.;
Gan, Z. J . Am. Chem. Soc. 1993, 115, 4943. (g) Dailey, K. M.;
Rauchfuss, T. B.; Rheingold, A. L.; Yap, G. P. A. J . Am. Chem. Soc.
1995, 117, 6396. (h) Arce, J . A.; De Sanctis, Y.; Karam, A.; Deeming,
J . A. Angew. Chem., Int. Ed. Engl. 1994, 33, 1381.
(7) (a) Sa´nchez-Delgado, R. A.; Gonza´les, E. Polyhedron 1989, 8,
1431. (b) Fish, R. H.; Tan, J . L.; Thormodsen, A. D. Organometallics
1985, 4, 1743. (c) INTEVEP SA (Pa´ez, D. E.; Andriollo, A.; Sa´nchez-
Delgado, R.; Valencia, N.; Lo´pez-Linares, F.; Galiasso, R. E.). U.S.
Patent 08/657.960, 1996.
solvents before being dried under
a stream of nitrogen.
Tetrahydrofuran (THF) and THF-d8 were purified by distil-
lation under nitrogen from LiAlH4. MeCN was distilled from
CaH2. Benzo[b]thiophene (99%, Aldrich) was sublimed prior
to use. Potassium tert-butoxide (KOBut, 95%), 18-crown-6
ether (99.5%), LiHBEt3 (1.0 M solution in THF), 2-ethylth-
iophenol (90%), and polyvinylpyrrolidone K25 (average Mw
29 000, PVP) were purchased from Aldrich and used without
further purification. 2,3-Dihydrobenzo[b]thiophene (DHBT)
was prepared by catalytic hydrogenation of BT assisted by 3.11
All the other reagents and chemicals were reagent grade and
were used as received from commerical suppliers. 1H (200.13
MHz) and 31P{1H} (80.01 MHz) NMR spectra were obtained
on a Bruker ACP 200 spectrometer. All chemical shifts are
reported in ppm (δ) relative to tetramethylsilane, referenced
to the chemical shifts of residual solvent resonances (1H) or
85% H3PO4 (31P). The 10 mm sapphire NMR tube was
purchased from Saphikon, Milford, NH, while the titanium
high-pressure charging head was constructed at the ISSECC-
CNR (Firenze, Ital).12 Note! Since high gas pressures are
involved, safety precautions must be taken at all stages of
studies involving high-pressure NMR tubes. The computer
simulation of NMR spectra was carried out with a locally
developed package containing the programs LAOCN313 and
Davins.14 The initial choices of shifts and coupling constants
were refined by iterative least-squares calculations using
(9) Rhodes, L. F.; Venanzi, L. M. Inorg. Chem. 1987, 26, 2692.
(10) Rhodes, L. F.; Sorato, C.; Venanzi, L. M.; Bachechi, F. Inorg.
Chem. 1988, 27, 604.
(11) Bianchini, C.; Meli, A.; Moneti, S.; Vizza, F., to be submitted
for publication.
(12) CNR (Bianchini, C.; Meli, A.; Traversi, A.). Italian Patent FI
A000025, 1997.
(13) (a) Bothner-By A. A.; Castellano, S. QCPE 1967, 11, 111. (b)
Castellano, S.; Bothner-By, A. A. J . Chem. Phys. 1964, 41, 3863.
(14) Stephenson, D. S.; Binsch, G. J . Magn. Reson. 1980, 37, 409.
(8) (a) Bianchini, C.; Meli, A.; Patinec, V.; Sernau, V.; Vizza, F. J .
Am. Chem. Soc. 1997, 119, 4945. (b) Bianchini, C.; Casares, J . A.; Meli,
A.; Sernau, V.; Vizza, F.; Sa´nchez-Delgado, R. A. Polyhedron 1997, 16,
3099. (c) Bianchini, C.; Fabbri, D.; Gladiali, S.; Meli, A.; Pohl, W.; Vizza,
F. Organometallics 1996, 15, 4604. (d) Bianchini, C.; Herrera, V.;
J ime´nez, M. V.; Meli, A.; Sa´nchez-Delgado, R. A.; Vizza, F. J . Am.
Chem. Soc. 1995, 117, 8567. (e) Bianchini, C.; J ime´nez, M. V.; Meli,
A.; Moneti, S.; Vizza, F.; Herrera, V.; Sa´nchez-Delgado, R. A. Orga-
nometallic 1995, 14, 2342.

2638 Organometallics, Vol. 17, No. 12, 1998
Bianchini et al.
experimental digitized spectra. The final parameters gave a
satisfactory fit between experimental and calculated spectra,
the agreement factor R being less than 1% in all cases. GC
analyses were performed on a Shimadzu GC-14 A gas chro-
matograph equipped with a flame ionization detector and a
30 m (0.25 mm i.d., 0.25 µm film thickness) SPB-1 Supelco
fused silica capillary column. GC/MS analyses were performed
on a Shimadzu QP 5000 apparatus equipped with a column
identical with that used for GC analyses. Infrared spectra
were recorded on a Perkin-Elmer 1600 series FT-IR spectro-
photometer using samples mulled in Nujol between KBr
plates. Elemental analyses (C, H, N) were performed using a
Carlo Erba Model 1106 elemental analyzer. Atomic absorption
analyses were performed with a Perkin-Elmer 5000 instru-
ment.
Anal. Calcd (found) for C69H66BN2P3Ru: C, 73.46 (73.12); H,
5.90 (5.80); N, 2.48 (2.31); Ru, 8.96 (8.77). 31P{1H} NMR (THF-
d8, 20 °C): AM2 spin system, δ 48.2 (d, J (PMPA) ) 20.8 Hz,
PM), 6.5 (t, PA). 1H NMR (THF-d8, 20 °C): δ 2.8-2.3 (m, 6H,
CH2P), 1.71 (q, 3H, J (HP) ) 2.3 Hz, CH3), 1.62 (d, 6H, J (HP)
) 1.4 Hz, MeCN), -5.16 (dt, 1H, J (HPA) ) 105.3 Hz, J (HPM)
) 19.3 Hz, Ru-H). IR: ν(Ru-H) 1850 (s) cm-1
.
Rea ction of K[(tr ip h os)Ru H3] (2) w ith BT. A. HP NMR
Exp er im en t u n d er Hyd r ogen . A 10 mm HPNMR tube was
charged under nitrogen first with a solid sample of 1 (18 mg,
0.024 mmol) together with a 3-fold excess of solid KOBut (9
mg, 0.072 mmol) and then with a THF-d8 (2 mL) solution
containing a 5-fold excess of BT (16 mg, 0.12 mmol). The tube
was pressurized with hydrogen to 3 bar at room temperature
and then placed into a NMR probe at room temperature. The
1
reaction was monitored by 31P{1H} and H NMR spectroscopy.
Syn th esis of K[(tr ip h os)Ru H 3] (2). Solid KOBut (85 mg,
0.72 mmol) was added to a stirred suspension of [(triphos)-
Ru(H)BH4] (1; 180 mg, 0.24 mmol) in THF (20 mL) at room
temperature. After 30 min the resulting yellow-orange solu-
tion was pumped to dryness and the beige solid residue was
washed with n-pentane. NMR analysis of this product showed
the formation of a unique Ru-triphos complex with the
following spectral features. 31P{1H} NMR (THF-d8, 20 °C): A3
spin system, δ 44.1 (s). 1H NMR (THF-d8, 20 °C): δ 2.09 (d,
6H, J (HP) ) 5.3 Hz, CH2P), 1.39 (q, 3H, J (HP) ) 2.2 Hz, CH3),
-7.73 (AA′A′′XX′X′′ spin system, second-order doublet of
multiplets, 3H, |J (HPtrans) + 2J (HPcis)| ) 43.1 Hz, Ru-H). All
our attempts to recrystallize 2 from organic solvents to
eliminate the KOBut impurity led to extensive decomposition.
Only when crude 2 was dissolved in THF containing a slight
excess of 18-crown-6 ether did the addition of n-heptane lead
to the precipitation of pure [K(C12H24O6)][(triphos)RuH3] (2a )
in 65% yield. Anal. Calcd (found) for C53H66KO6P3Ru: C,
61.67 (61.09); H, 6.45 (6.38); Ru, 9.79 (9.54). 31P{1H} NMR
(THF-d8, 20 °C): A3 spin system, δ 44.1 (s). 1H NMR (THF-
d8, 20 °C): δ 2.20 (d, 6H, J (HP) ) 5.3 Hz, CH2P), 1.50 (q, 3H,
J (HP) ) 2.1 Hz, CH3), -7.28 (AA′A′′XX′X′′ spin system, second-
order doublet of multiplets, 3H, |J (HPtrans) + 2J (HPcis)| ) 46.9
Within 1 h, 1 completely disappeared; formed in its place was
2. The temperature of the NMR probe was then increased
gradually to 50 and 70 °C. Only at the latter temperature
did a fast reaction occur, which converted all of 2 to the
dihydride thiolate ruthenate complex K[(triphos)Ru(H)2(o-
S(C6H4)C2H5)] (5) in ca. 30 min. After the NMR probe was
cooled to room temperature, 5 was the only metal product
visible by NMR spectroscopy with the following spectral
features. 31{1H} NMR: AM2 spin system, δ 57.2 (t, J (PAPM)
) 24.1 Hz, PA), 25.1 (d, PM). 1H NMR: δ 2.94 (q, 2H, J (HH)
) 7.5 Hz, CH2CH3), 2.6-2.1 (m, 6H, CH2P), 1.48 (q, 3H, J (HP)
) 2.2 Hz, CH3), 1.23 (t, 3H, CH2CH3), -6.88 (AA′XX′Y spin
system, second-order doublet of multiplets, 2H, J (HAHA′) )
9.45 Hz, J (HAPX) ) 85.44 Hz. J (HAPX′) ) -11.56 Hz, J (HPY)
) 23.81 Hz, J (PXPX′) ) -17.22 Hz, Ru-H).
B. HP NMR Exp er im en t u n d er Nitr ogen . A 10 mm
HPNMR tube was charged under nitrogen first with a solid
sample of 1 (18 mg, 0.024 mmol) together with a 3-fold excess
of solid KOBut (9 mg, 0.072 mmol) and then with a THF-d8 (2
mL) solution containing a 5-fold excess of BT (16 mg, 0.12
mmol). The reaction was monitored by variable-temperature
1
31P{1H} and H NMR spectroscopy. A sequence of selected 31P-
Hz, Ru-H). IR: ν(Ru-H) 1850 (m), 1828 (m) cm-1
. Com-
{1H} NMR spectra is reported in Figure 1. Within 1 h, 1
completely converted into 2 (trace a). The temperature of the
NMR probe was then increased to 100 °C. The 31P{1H} NMR
spectra acquired during the following 3 h showed the gradual
disappearance of 2; formed in its place was a broad resonance
centered at ca. 24 ppm which was assigned (see below) to the
fluxional vinylthiophenolate complex K[(triphos)Ru(η3-S(C6H4)-
CHdCH2)] (6; trace b). The slow-exchange regime of the latter
complex was attained at room temperature, where 6 was the
only metal product visible in the reaction mixture by NMR
spectroscopy with the following spectral features (trace c). 31P-
{1H} NMR: AMQ spin system, δ 47.3 (t, J (PAPM) ) 33.1 Hz,
J (PAPQ) ) 28.1 Hz, PA), 25.5 (dd, J (PMPQ) ) 13.1 Hz, PM), 12.7
(dd, PQ). 1H NMR: δ 2.79 (m, CHdCH2), 2.3 (masked by
aliphatic resonances of triphos, CHdCHH′), 1.7 (masked by
THF, CHdCHH′). The tube was pressurized with hydrogen
to 30 bar at room temperature and then placed into a NMR
probe. The 31P{1H} NMR spectra of this sample, recorded at
room temperature over the following 2 h, showed the gradual
conversion of 6 into 5 (trace d, after 2 h). Complete conversion
of 6 into 5 was achieved by increasing the temperature to 40
°C; a small amount of the trihydride 2 was also detected. The
reaction between 2 and BT, although slow, occurred even at
70 °C with the same sequence of products.
pounds 2 and 2a are stable in both the solid state and dry
THF solution under a dry nitrogen or hydrogen atmosphere;
both compounds are extremely sensitive to moisture as well
as any sort of Brønsted and Lewis acids.
Rea ction of [(tr ip h os)Ru (NCMe)3](BP h 4)2 (3) w ith H2
in t h e P r esen ce of KOBu t. Sa p p h ir e Tu b e HP NMR
Exp er im en t. A 10 mm sapphire HPNMR tube was charged
with a solution of 3 (36 mg, 0.024 mmol) and a 10-fold excess
of KOBut (28 mg, 0.24 mmol) in THF-d8 (2 mL) under nitrogen.
The tube was pressurized with hydrogen to 30 bar at room
temperature and then placed into a NMR probe. The reaction
was followed by variable-temperature 31P{1H} and 1H NMR
spectroscopy. The transformation of the starting tris(aceto-
nitrile) complex 3 occurred even at room temperature, yielding
[(triphos)Ru(H)(NCMe)2]BPh4 (4; see below) and other uni-
dentified minor species (1 h). Increasing the temperature to
40 °C led to the gradual conversion of all the previously formed
compounds to the ruthenate complex 2 (3 h). After the NMR
probe was cooled to room temperature, 2 was the only metal
product visible by 31P{1H} NMR spectroscopy. The tube was
then removed from the spectrometer and cooled to -20 °C and
the gas phase was analyzed by GC, showing the presence of
NH3. The cool liquid contents of the tube were then trans-
ferred into a Schlenk-type flask maintained at -20 °C. GC
and GC/MS analysis of the solution showed the formation of
NH2Et, NHEt2, and NEt3 in a ratio of 11:4:1. Further NH3
was also detected.
All our attempts to isolate 6 from batch reactions were
unsuccessful. Every treatment of the reaction mixture invari-
ably led to the decomposition of 6 to several unidentified
products.
Syn th esis of [(tr ip h os)Ru (H)(NCMe)2]BP h 4 (4). To a
solution of 3 (150 mg, 0.1 mmol) in MeCN (5 mL) was added
a 4-fold excess of LiHBEt3 (0.4 mL, 0.4 mmol). After 10 min,
addition of ethanol (5 mL) and n-heptane (60 mL) led to the
precipitation of 4 as pale yellow microcrystals in 80% yield.
Rea ction s of K[(tr ip h os)Ru H 3] (2) w ith ETP . A. NMR
Exp er im en t. A Teflon-capped resealable 5 mm NMR tube
was charged first with a solid sample of 1 (18 mg, 0.024 mmol)
together with a 3-fold excess of solid KOBut (9 mg, 0.072 mmol)
and then with THF-d8 (1 mL) under nitrogen. The tube was

HDS Activity of Ruthenium-Based Catalysts
Organometallics, Vol. 17, No. 12, 1998 2639
placed into a NMR probe at room temperature. The reaction
Sch em e 2
1
was monitored by 31P{1H} and H NMR spectroscopy. Within
1 h, compound 1 converted to 2. Afterward, 1 equiv of ETP
was syringed into the tube at room temperature. The 31P{1H}
and 1H NMR spectra of this sample showed the complete
conversion of 2 to 5 and H2 (1H NMR singlet at δ 4.7).
B. Syn th etic Exp er im en t. A suspension of 1 (180 mg,
0.24 mmol) together with a 3-fold excess of solid KOBut (85
mg, 0.72 mmol) in THF (20 mL) was stirred at room temper-
ature. After ca. 30 min, 1 equiv of ETP was syringed into the
solution. After 10 min, a 0.5 mL portion of this solution was
withdrawn and diluted with 0.5 mL of THF-d8 for the 31P{1H}
1
and H NMR analysis. Compound 5 was the only ruthenium
complex detected in solution on the NMR time scale. All our
attempts to isolate 5 from this solution were unsuccessful.
Attempted isolation of 5 by either precipitation with various
solvents under
a nitrogen or dihydrogen atmosphere or
concentration of the reaction mixture to dryness under vacuum
invariably led to the decomposition of the desired complex to
several unidentified products. Solutions of 5 in THF are
extremely sensitive to air and acids.
as polyvinylpyrrolidone (PVP) from 0.1 to 100% with respect
to the catalyst precursor).
B. Sa p p h ir e Tu be HP NMR Exp er im en t w ith th e
Ca t a lyst P r ecu r sor [(t r ip h os)R u (H)BH 4] (1). A 10 mm
sapphire HPNMR tube was charged first with solid samples
of 1 (18 mg, 0.024 mmol) and KOBut (28 mg, 0.24 mmol) and
then with a solution containing a 30-fold excess of BT (96 mg,
0.72 mmol) in THF-d8 (2 mL) under nitrogen. The tube was
pressurized with hydrogen to 30 bar at room temperature and
then placed into a NMR probe at 20 °C. The reaction was
followed by variable-temperature 31P{1H} and 1H NMR spec-
troscopy in the range 20-70 °C. The results of this study are
described in detail in a forthcoming section and illustrated in
Figure 2. After ca. 4 h at 70 °C, the tube was removed from
the spectrometer and depressurized to ambient pressure. the
31P{1H} NMR spectrum was identical with that of a pure
sample of 5. The contents of the tube were then transferred
into a Schlenk-type flask and acidified with HCl(aq). A sample
of the solution, withdrawn and analyzed by GC and GC/MS,
was found to contain ETP (18%), BT (82%), and DHBT (<1%).
When isolated 2 was employed instead of 1, the reaction
followed an identical path.
C. Sa p p h ir e Tu be HP NMR Exp er im en t w ith th e
Ca ta lyst P r ecu r sor [(tr ip h os)Ru (NCMe)3](BP h 4)2 (3). A
10 mm sapphire HPNMR tube was charged with a solution of
3 (36 mg, 0.024 mmol), a 10-fold excess of KOBut (28 mg, 0.24
mmol), and a 30-fold excess of BT (96 mg, 0.72 mol) in THF-
d8 (2 mL) under nitrogen. The tube was pressurized with
hydrogen to 30 bar at room temperature and then placed into
a NMR probe at 20 °C. The reaction was followed by variable-
temperature 31P{1H} and 1H NMR spectroscopy in the tem-
perature range 20-70 °C for an overall time of ca. 4 h. The
results of this study are described in detail in a forthcoming
section and illustrated in Figure 3. After the catalytic reaction
was quenched by cooling to room temperature, 5 was the only
metal product visible by 31P{1H} NMR spectroscopy. After the
tube was depressurized, its contents were transferred into a
Schlenk-type flask. A portion of this solution, after acidifica-
tion with HCl(aq), was analyzed by GC and GC/MS: ETP
(11%), BT (84%), and DHBT (5%). GC and GC/MS analysis
of the rest of the solution led to the identification of NH2Et,
NHEt2, and NEt3 in a ratio of ca. 9:3:1.
Reaction of K[(tr iph os)Ru (H)2(o-S(C6H4)C2H5)] (5) with
H2 (30 ba r ) in th e P r esen ce of Excess K[o-S(C6H4)C2H5].
HP NMR Exp er im en t. ETP (5 equiv) was syringed into a
10 mm sapphire HPNMR tube containing a solution of 2
(prepared by reacting 1 (18 mg, 0.024 mmol) with a 10-fold
excess of KOBut (28 mg, 0.24 mmol) in THF-d8 (2 mL) under
nitrogen). The tube was placed into a NMR probe at room
temperature and heated to 70 °C. After ca. 1 h, the 31P{1H}
and 1H NMR spectra of this sample showed the complete
conversion of 2 to 5. At this point, the tube was cooled to room
temperature and pressurized with hydrogen to 30 atm; then
it was placed into the NMR probe again and heated to 70 °C.
Even the first 31P{1H} NMR spectrum showed the character-
istic resonances of 2. Spectra acquired after 1 and 2 h were
practically identical with each other, showing a 2:5 ratio of
ca. 10. No other signal was visible by NMR. After the tube
was cooled to room temperature, 2 and 5 were still the only
detectable metal products. The tube was depressurized in
order to add further KOBut (0.24 mmol). After the tube was
repressurized to 30 bar of H2, it was introduced into the probe
head and heated to 70 °C. Under these conditions, the
concentration of the trihybride 2 increased remarkably at the
expense of that of 5 (2:5 ratio of 50) as a result of the delivery
of the thiolate ligand into the solution as the potassium salt.
Hyd r ogen olysis of BT of 2-Eth ylth iop h en ola te. A.
P a r r Rea ctor Exp er im en ts. The reaction conditions and
the results of these experiments have been collected in Table
1. In a typical experiment, a solution of either the tetrahy-
droborate complex 1 (32 mg, 0.043 mmol) or the tris(acetoni-
trile) derivative 3 (65 mg, 0.043 mmol) in THF (30 mL) and
100-fold excesses of both BT (579 mg, 4.3 mmol) and KOBut
(510 mg, 4.3 mmol) were placed into the Parr reactor under a
nitrogen atmosphere. After pressurizing with hydrogen to 30
bar at room temperature, the mixture was heated to the
appropriate temperature and then immediately stirred (750
rpm). After the desired time, the reactor was cooled to room
temperature and slowly depressurized. The contents of the
reactor were transferred into a Schlenk-type flask and acidified
with aqueous HCl to ca. pH 5 to convert 2-ethylthiophenolate
sodium salt to 2-ethylthiphenol (ETP).8a This procedure is
necessary to have reliable GC analyses as well as to prevent
the base-promoted oxidation of the thiophenolate product to
bis(2-ethylphenyl) sulfide.8a A sample of the solution was
withdrawn and analyzed by GC and GC/MS. Several catalytic
reactions were carried out in the presence of excess elemental
Hg (2000:1) to test the homogeneous character of the reactions.
In all cases, no change in both activity and chemoselectivity
was observed. Similarly, the catalytic activity did not signifi-
cantly change when the reactions were carried out in the
presence of variable amounts of a colloid-protecting agent such
Resu lts a n d Discu ssion
Syn th esis a n d Ch a r a cter iza tion of th e Ca ta lyst
P r ecu r sor K[(tr ip h os)Ru H3]. The novel Ru(II) tri-
hydride complex K[(triphos)RuH3] (2) is quantitatively
formed upon treatment of [(triphos)RuH(BH4)]9 (1) in
THF with a 3-fold excess of KOBut at room temperature
(Scheme 2a). The elimination of the BH2 moiety from
1 by reaction with Lewis bases is a known process,
previously employed by Venanzi to generate in situ

2640 Organometallics, Vol. 17, No. 12, 1998
Bianchini et al.
heteropolymetallic complexes with bridging hydride
ligands.15 The mononuclear compound 2 is isolated in
the solid state as a beige powder by solvent elimination
under reduced pressure. Once isolated, 2 can be handled
and weighed safely under a dry nitrogen atmosphere
to prepare solutions in anhydrous THF. The compound
is extremely sensitive to moisture and protic solvents
(vide infra); in the presence of a slight excess of KOBut,
however, THF solutions of 2 do not show appreciable
decomposition for days, given the rigorous absence of
water. All our attempts to recrystallize 2 from organic
solvents resulted in its extensive decomposition unless
a crown ether such as 18-crown-6 ether was added. In
this case, the complex [K(C12H24O6)][triphos)RuH3] (2a )
was obtained in fairly good yields from THF/n-heptane.
An alternative, very efficient synthesis of 2 in THF
solution is provided by the hydrogenation (30 bar of H2)
of the tris(acetonitrile) complex [(triphos)Ru(MeCN)3]-
(BPh4)2 (3) at 40 °C in the presence of a 10-fold excess
of KOBut (Scheme 2b). Water must be rigorously
eliminated from the reaction mixture to avoid the
irreversible formation of the dimer [(triphos)Ru(µ-
OH)3Ru(triphos)]BPh4 (7).16
has been suggested to proceed via the reaction sequence
summarized in eq 4.17a
H2
H2
RCN 8 RCHdNH 8 RCH2NH2 RCN8
H2
-NH3
RCH2NHC(R)dNH 8 RCH2NHCH(R)NH2
8
H2
H2 -NH3 H2
RCH2NdCHR 8 (RCH2)2NH RCN8
8
8
8
(RCH2)3N (4)
The fact that the first metal product to form is the
monohydrido complex 4 is a clear indication that a
heterolytic splitting of H2 assisted by the strong base
occurs initially at the metal center. Most likely, it is a
Ru(II) hydride complex, similar to 4, the species which
catalyzes the acetonitrile hydrogenation/amine redis-
tribution reactions, in the course or at the end of which
2 is formed upon two further heterolytic splittings of
H2.
Compound 2 belongs to a relatively numerous family
of fac triphosphine trihydride metal complexes, among
which [(triphos)RhH3] and [(triphos)IrH3] are those
sharing with 2 the greatest number of structural and
spectroscopic analogies.18 As for the Rh and Ir deriva-
tives, the 31P{1H} NMR spectrum of 2 consists of a
temperature-invariant A3 spin system, while the three
chemically but not magnetically equivalent hydride
ligands give rise to a highly perturbed AA′A′′XX′X′′
pattern. From a chemical viewpoint, however, the
closest parent of 2 is the ruthenate K[fac-RuH3-
(PPh3)3].19 Indeed, like the PPh3 analog, 2 reacts with
proton donors of different strength (water, alcohols,
protic acids), yielding equilibrium concentrations of the
tetrahydrido complex [(triphos)RuH4] (8).16 In the
absence of a positive H2 pressure, the latter complex is
not stable in solution, where it readily forms the 16e-
Ru(II) fragment [(triphos)Ru(H)2], which either dimer-
izes to give the red complex [(triphos)RuH(µ-H)2HRu-
(triphos)] (9) or is intercepted by nucleophiles to give
mononuclear adducts. For example, the novel dihydride
[(triphos)Ru(H)2(CO)] (10) is quantitatively obtained by
bubbling CO into a THF/ ethanol solution of 2.16 Details
of these transformations are not given here, as the
reactions which are of peculiar interest to this work are
just those with either BT or ETP.
Rea ction s of K[(tr ip h os)Ru H3] w ith Ben zo[b]-
th iop h en e or 2-Eth ylth iop h en ol. NMR shows that
2 in THF is capable of cleaving BT, yielding the
vinylthiophenolate complex K[(triphos)Ru(η3-S(C6H4)-
CHdCH2)] (6) even at 70 °C. At the same temperature
but in the presence of a positive pressure of H2, the
dihydride 2-ethylthiophenolate complex K[(triphos)Ru-
(H)2(o-S(C6H4)C2H5)] (5) is quantitatively obtained
(Scheme 3a,b; Figure 1). Since 6 is a precursor to 5 via
hydrogenation, the overall hydrogenolysis of BT defi-
nitely proceeds via C-S insertion/hydrogenation and not
by the alternative hydrogenation/C-S insertion se-
qeuence.1,5 Consistently, the hydrogenated product 2,3-
dihydrobenzo[b]thiophene (DHBT) does not react with
2 under identical reaction conditions.
The reaction between 3 and H2 in the presence of
KOBut was followed by NMR spectroscopy in a high-
pressure sapphire tube sealed with a Ti-alloy valve. As
shown by an in situ HPNMR experiment at 40 °C, the
formation of 2 is preceded by that of several ruthenium
complexes, among which only the hydride complex
[(triphos)RuH(MeCN)2]BPh4 (4) was unambiguously
identified by its independent synthesis from 3 and
LiHBEt3 in MeCN (eq 3). All the other ruthenium
complexes have a fleeting existence and are rapidly
converted to 2.
After all 3 has been converted to 2, no trace of MeCN
is detected in solution by GC. Formed in its place the
mixtures of NH2Et, NHEt2, NEt3, and NH3. These
amines are most likely produced upon hydrogenation
of MeCN to the primary amine (via an imine intermedi-
ate), followed by amine redistribution reactions which
first give the secondary amine and ammonia and then
the tertiary amine and further ammonia.17 As expected,
the NH2Et:NHEt2:NEt3 ratios decrease with the tem-
perature, while the concentration of NH3 increases. In
an in situ experiment at 40 °C for 3 h, a 11:4:1 ratio
between NH2Et, NHEt2, and NEt3 was determined by
GC/MS. The overall transformation pattern undergone
by the acetonitrile ligands in 3 is well-known for Ru-
catalyzed homogeneous hydrogenations of nitriles and
(15) Albinati, A.; Venanzi, L. M.; Wang, G. Inorg. Chem. 1993, 32,
3660.
(16) A detailed account of the reactivity of 2 toward proton donors
alone or in combination with nucleophiles or strong bases will be given
elsewhere. This kind of information is not relevant to the main subject
of this work. Spectral and analytical data for compounds 7-10 are
given as Supporting Information.
(17) (a) Holy, N. L. J . Org. Chem. 1979, 44, 239. (b) J ung, C. W.;
Fellmann, J . D.; Garrou, P. E. Organometallics 1983, 2, 1042. (c) J oshi,
A. M.; MacFarlane, K. S.; J ames, B. R.; Frediani, P. Chem. Ind. 1994,
53, 497.
(18) (a) Ott, J .; Venanzi, L. M.; Ghilardi, C.A.; Midollini, S.;
Orlandini, A. J . Organomet. Chem. 1985, 291, 89. (b) J anser, P.;
Venanzi, L. M.; Bachechi, F. J . Organomet. Chem. 1985, 296, 229.
(19) Linn, D. E., J r.; Halpern, J . J . Am. Chem. Soc. 1987, 109, 2969.

HDS Activity of Ruthenium-Based Catalysts
Organometallics, Vol. 17, No. 12, 1998 2641
Sch em e 3
even at room temperature unless a protective atmo-
sphere of this gas is employed. As the H2 pressure is
increased to 30 bar, however, 5 partially reconverts to
2 through the elimination of ETP (Scheme 3b). HPNMR
also shows that, under 30 bar of H2, the equilibrium (5)
is shifted to the right by addition of increasing amounts
of KOBut.
[(triphos)Rh(H)2(o-S(C6H4)C2H5)]- + H2 + ButO- a
[(triphos)RuH3]- + [o-S(C6H4)C2H5]- + ButOH (5)
Unlike 5, we were unable to independently prepare
6. The identification of this C-S insertion product was
therefore based on its spectroscopic characteristics,
which are quite similar to those of the related Rh(I) and
Ir(I) 2-vinylthiophenolate neutral complexes [(triph-
os)M(η3-S(C6H4)CHdCH2)] (M ) Ir,20 Rh) (see Scheme
4).21d In particular, the 31P NMR spectrum shows a
canonical AMQ pattern, while signals at 2.79 (CH2)
1
and 2.3 and 1.7 ppm (CHH′) in the H NMR spectrum
can safely be attributed to a vinyl group bound to
ruthenium.20,21d Further support tothe proposed struc-
ture for 6 is provided by its quantitative conversion to
5 by treatment with H2.
Due to the availability of filled metal dπ orbitals of
appropriate energy and symmetry to interact withthe
π* C2-S orbital,5 the cleavage of the C2-S bond of BT
by 16e- metal fragments of the type [(triphos)MH] (M
) Rh, Ir) is a low-energy process indeed.8,20,21 In
particular, it is lower in energy than the thermal
processes which are necessary to generate the 16e-
fragments from their 18e- precursors (see Scheme 4).
The Ru(0) system [(triphos)RuH]-, generated by reduc-
tive elimination of H2 from 2, is a full member of this
family, as it regioselectively cleaves BT even at 40 °C
(see below).
Ca ta lytic Hyd r ogen olysis of Ben zo[b]th iop h en e
to 2-Eth ylth iop h en ol. A. Au tocla ve Rea ction s.
Irrespective of the catalyst precursor (either 1 or 3), BT
is selectively transformed into the 2-ethylthiophenolate
anion; traces of the hydrogenation product DHBT are
often formed (e1%). The results obtained are reported
in Table 1.
The production of DHBT decreases by increasing the
base to catalyst ratio. The hydrogenolysis product may
be recovered as the potassium salt K[o-S(C2H4)C2H5];
for product quantification, however, it is convenient to
transform the thiolate salt into ETP by treatment with
aqueous HCl. If the final reaction mixtures are not
acidified, spontaneous oxidation of the thiophenolate
product to bis(2-ethylphenyl) sulfide takes place upon
F igu r e 1. 31P{1H} NMR study (sapphire tube, THF-d8,
81.01 MHz) of the reaction of 2 with BT: (a) at room
temperature under nitrogen; (b) at 100 °C after 3 h; (c) after
the tube was cooled to room temperature; (d) at room
temperature after the tube was pressurized with 30 bar of
H2 for 2 h.
Compound 5 has been identified by NMR spectroscopy
as well as its independent preparation by the reaction
of 2 in THF with pure ETP (Scheme 3b). In particular,
the AM2 pattern of the phosphorus nuclei and the
AA′XX′Y (X, Y, ) P) pattern of the two terminal
hydrides are quite comparable to those of the congeners
[(triphos)M(H)2(o-S(C6H4)C2H5)] (M ) Rh, Ir) similarly
prepared by C-S insertion/hydrogenation of BT (Scheme
4).8d,20
(20) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F.; Moneti, S.;
Herrera, V.; Sa´nchez-Delgado, R. A. J . Am. Chem. Soc. 1994, 116, 4370.
(21) (a) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F.; Frediani,
P.; Herrera, V.; Sa´nchez-Delgado, R. A. J . Am. Chem. Soc. 1993, 115,
2731. (b) Bianchini, C.; J ime´nez, M. V.; Meli, A.; Moneti, S.; Vizza, F.
J . Organomet. Chem. 1995, 504, 27. (c) Bianchini, C.; Casares, J . A.;
J ime´nez, M. V.; Meli, A.; Moneti, S.; Vizza, F.; Herrera, V.; Sa´nchez-
Delgado, R. A. Organometallics 1995, 14, 4850. (d) Bianchini, C.;
Frediani, P.; Herrera, V.; J ime´nez, M. V.; Meli, A.; Rinco´n, L.; Sa´nchez-
Delgado, R. A.; Vizza, F. J . Am. Chem. Soc. 1995, 117, 4333. (e)
Bianchini, C.; J ime´nez, M. V.; Meli, A.; Vizza, F. Organometallics 1995,
14, 3196. (f) Bianchini, C.; Meli, A.; Pohl, W.; Vizza, F.; Barbarella, G.
Organometallics 1997, 16, 1517. (g) Bianchini, C.; Casares, J . A.; Masi,
D.; Meli, A.; Pohl, W.; Vizza, F. J . Organomet. Chem. 1997, 541, 143.
Like the rhodium analog [(triphos)Rh(H)2(o-S(C6H4)-
C2H5)],8d 5 is thermally unstable in solution, losing H2

2642 Organometallics, Vol. 17, No. 12, 1998
Bianchini et al.
Sch em e 4
Ta ble 1. Hyd r ogen olysis of Ben zo[b]th iop h en e
Ca ta lyzed by [(tr ip h os)Ru (H)BH4] (1) a n d
[(tr ip h os)Ru (NCMe)3](BP h 4)2 (3) in th e P r esen ce of
tBu OKa
tration (compare entries 8 and 14, 15). The rate also
increases with the temperature (entries 1, 2, 5 and 8,
17, 18). At the highest temperature investigated (160
°C), however, some decomposition of the catalyst occurs,
as shown by the formation of some black precipitate in
the reactor (entry 1). Atomic absorption analysis showed
it to be metallic ruthenium. A sample of this solid
precipitate was used as the catalyst precursor for the
hydrogenation of BT, but no catalytic transformation
was observed under comparable experimental condi-
tions.
The presence of a substantial amount of water in the
reaction mixture inhibits the hydrogenolysis of BT
(entries 6 and 16) as a consequence of the formation of
the dimer [(triphos)Ru(µ-OH)3Ru(triphos)]+ (7),16 which
independent reactions show to be catalytically inactive.
The addition of either elemental mercury or PVP to
the catalytic mixtures did not significantly affect the
overall conversion of BT (entries 3 and 9). These two
experiments are thus consistent with a truly homoge-
neous process.23,24
reacn mixture
complex
(amt,
mmol)
amt of
amt of
compositn, %b
entry
no.
substrate, base,
mmol
mmol T, °C BT ETP DHBT ratec
1
2
1 (0.043)
1 (0.043)
4.3
4.3
4.3
4.3
4.3
4.3
4.3
4.3
4.3
4.3
4.3
8.6
8.6
4.3
4.3
4.3
4.3
4.3
4.3
4.3
4.3
0.3
4.3
4.3
8.6
4.3
4.3
2.2
1.1
8.6
4.3
4.3
4.3
4.3
4.3
4.3
160 32.5 67.1
100 52.7 46.8
100 53.7 45.9
100 95.3 3.6
70 61.4 38.1
100 96.5 2.3
100 30.0 69.8
100 55.6 43.9
100 56.8 42.8
100 79.2 19.9
100 82.6 16.8
100 46.6 53.1
100 75.6 23.9
100 38.2 60.6
100 68.2 31.4
100 99.1 0.2
70 68.2 31.2
40 98.7 1.2
0.4
0.5
0.4
1.1
0.5
1.2
0.2
0.5
0.4
0.9
0.6
0.3
0.5
1.2
0.4
0.7
0.6
0.1
0.58
0.40
0.39
0.03
0.33
3d 1 (0.043)
4
5
1 (0.043)
1 (0.043)
6e 1 (0.043)
7
8
3 (0.043)
3 (0.043)
0.60
0.38
0.37
0.17
0.14
0.91
0.41
0.52
0.27
9f 3 (0.043)
10
3 (0.043)
3 (0.043)
3 (0.043)
3 (0.043)
3 (0.086)
3 (0.022)
11
12
13
14
15
16e 3 (0.043)
17
18
3 (0.043)
3 (0.043)
0.27
a
The overall picture of the present hydrogenolysis of
BT is quite similar to that of other reactions of this type
catalyzed by the [(triphos)RhH] fragment in different
phase-variation systems.8 Linear correlations between
the hydrogenolysis rate and the concentration of cata-
lyst, substrate, or base are generally observed. The
accelerating effect of the base has been attributed to the
positive influence exerted by the base on the rate-
determining step, i.e. the reductive elimination of the
thiol from the metal center (vide infra).8 In the case at
hand, the influence of the base concentration on the
hydrogenolysis rate is even more pronounced, as shown
by the fact that the rate does not increase with the
concentration of BT (entry 13) unless the concentration
of KOBut is proportionally increased (entry 12). We
ascribe this unusual effect to the great basicity of the
trihydride 2 and hence of [(triphos)RuH]-. Both these
species rapidly react with water or alcohols to give the
neutral dihydride fragment [(triphos)Ru(H)2] (see above),
which current studies from this laboratory show to be
unable to catalyze the hydrogenolysis of BT.16 While
the presence of water in the catalytic mixture can be
extensively reduced and eventually suppressed by using
Reaction conditions: THF, 30 mL; H2 pressure, 30 bar; time,
b
5 h. Key: benzo[b]thiophene (BT), 2,3-dihydrobenzo]b]thiophene
(DHBT), 2-ethylthiophenol (ETP). c Average rate expressed as
(mmol of ETP) h-1
.
The reaction was carried out in the presence
d
of elemental mercury. e THF/H2O, 28:2 (v:v). f The reaction was
carried out in the presence of PVP.
Sch em e 5
exposure to air (Scheme 5).22 For sake of clarity, the
hydrogenolysis product, from now on, is referred to as
ETP.
Effective catalytic transformation of BT occurs at 70
°C (entries 5 and 17), but even at 40 °C, ETP is formed
in a concentration higher than the stoichiometric one
(entry 18). At 100 °C and a 1:1 ratio between substrate
and base, the hydrogenolysis rate, expressed as (mmol
of ETP) h-1, is ca. 0.4 (entries 2 and 8). This run has
been considered as the standard reaction and has been
studied in some detail in order to gain insight into the
kinetics of the hydrogenolysis process.
The rate of conversion of BT dramatically depends on
the base concentration, as shown by entries 2, 4, 7, 8,
and 10-13. At a constant concentration of substrate
and base, the rate increases with the catalyst concen-
t
freshly distilled THF over LiAlH4, BuOH cannot be
avoided, as it is the conjugate acid of the base employed
to generate the catalyst (any other base would have the
same effect). A similar situation has been reported to
occur for the two catalysts [fac-RuH3(PPh3)3]-/[RuH4-
(23) (a) Lewis, L. N.; Lewis, N. J . Am. Chem. Soc. 1986, 108, 7228.
(b) Lewis, L. N. J . Am. Chem. Soc. 1986, 108, 743. (c) Anton, D. R.;
Crabree, R. H. Organometallics 1983, 2, 855.
(24) (a) Wang, Q.; Liu, H.; Han, M.; Li, X.; J iang, D. J . Mol. Catal.,
A 1997, 118, 145. (b) Hirai, H.; Chawanya, H.; Toshima, H. Bull. Chem.
Soc. J pn. 1985, 58, 682.
(22) March, J . Advanced Organic Chemistry, 4th ed.; Wiley-
Interscience: New York, 1992; pp 1199-1205.

HDS Activity of Ruthenium-Based Catalysts
Organometallics, Vol. 17, No. 12, 1998 2643
(PPh3)3], which have been found to coexist through the
equilibrium (6) in the course of the hydrogenation of
cyclohexanone to cyclohexanol.19
[RuH3(PPh3)3]- + c-C6H11OH a
[RuH4(PPh3)3] + c-C6H11O- (6)
Only the addition of a very large excess of KOBut as
in entry 12 shifts to the left the analogous equilibrium
involving 2 so as to cancel the presence of the tetrahy-
dride complex 8 in the catalytic mixture (eq 7). For this
[(triphos)RuH3]- + tBuOH a
[(triphos)RuH4] + tBuO- (7)
reason, besides accelerating the hydrogenolysis rate, a
large excess of base in the catalytic mixtures also
improves the chemoselectivity of the reactions. The
production of DHBT decreases, in fact, by decreasing
the concentration of the tetrahydro complex 8, which is
an effective catalyst for the hydrogenation of BT.11
B. HP NMR Stu d ies. The hydrogenolysis of BT to
ETP catalyzed by either 1 or 3 has been studied in an
HPNMR tube under experimental conditions that are
as close as possible to those employed in the standard
batch reaction. The only significant differences are a
higher concentration of the catalyst for a better resolu-
tion and acquisition of the NMR spectra and, obviously,
a lower stirring rate. The head space of the 10 mm
tubes is large enough to maintain a high concentration
of H2 into the solution (1H NMR evidence). Indeed, we
generally observe that the mass transfer of H2 from the
head space of the NMR tube is efficient enough to
replenish the solution which is being depleted of this
reagent by the catalyst.
F igu r e 2. 31P{1H} NMR study (sapphire tube, THF-d8,
81.01 MHz) of the catalytic hydrogenolysis of BT in the
presence of 1 (30 bar of H2, substrate:catalyst ratio 30,
substrate:KOBut ratio 10): (a) pure sample of 1; (b) at 20
°C for 1 h; (c) at 70 °C for 30 min; (d) at 70 °C for 1 h; (e)
after the tube was cooled to room temperature. In b′ and
e′ are reported the 1H HPNMR spectra (200.13 MHz) in
the hydride region recorded immediately after the 31P{1H}
NMR spectra b and e, respectively.
spectroscopy (trace e). The tube was then removed from
the spectrometer and depressurized to ambient pres-
sure. The 31P{1H} NMR spectrum of this sample was
identical with that of trace e. GC/MS analysis of the
reaction mixture gave an 18% conversion of BT to ETP.
A sequence of selected 31P{1H} NMR spectra is
reported in Figure 3 for the hydrogenolysis of BT
catalyzed by the tris(acetonitrile) complex 3 under the
same conditions of the reaction assisted by 1. The
transformation of the latter complex (trace a; this trace
represents the spectrum of a pure sample of 3 in THF-
d8 under nitrogen) occurred even at room temperature,
yielding 4 and three other unidentified minor species
(trace b, after ca. 30 min at 20 °C). Increasing the
temperature to 40 °C led to the gradual conversion of
all the previously formed compounds to the ruthenate
complex 5 (traces c-e, after 30 min, 90 min, and 3 h at
40 °C, respectively) via other species which are still
unidentified. (Both 4 and all of these unidentified
species were also seen in analogous experiments carried
out at the same temperature and hydrogen pressure but
in the absence of BT. Their rerlevance to catalysis is
thus negligible. The final product of these blank experi-
ments was the trihydride 2.) During the course of the
transformation of 3 into 5, minor hydrogenation of BT
to DHBT also occurred.25 Only when the reaction mix-
ture was heated to 70 °C was the formation of the tri-
hydride 2 observed (trace f, after 30 min at 70 °C), and
concomitantly, free ETP began to be produced. Spectra
A sequence of selected 31P{1H} NMR spectra is
reported in Figure 2 for a reaction catalyzed by the
hydride tetrahydroborate complex 1 (trace a; this trace
represents the spectrum of a pure sample of 1 in THF-
d8 under nitrogen) in THF-d8 with a BT:KOBut:1 ratio
of 30:10:1 under 30 bar of H2. Even at room tempera-
ture, 1 transformed into 2 after ca. 1 h at 20 °C (trace
b). In the course of this transformation, no reaction
1
involving BT was observed by H NMR spectroscopy.
Increasing the temperature to 50 °C induced no change
1
in the 31P{1H} NMR spectrum, whereas the H NMR
spectrum showed the appearance of resonances due to
free 2-ethylthiophenolate (δ 3.04 (q, J (HH) ) 7.4 Hz,
CH2CH3), 1.31 (t, CH2CH3)). Only when the reaction
mixture was heated to 70 °C did appreciable formation
of the dihydride thiolate ruthenate 5 occur (trace c, after
30 min at 70 °C). No trace of the C-S insertion product
6 was detected spectroscopically, which is consistent
with the fast hydrogenation of this complex to give 5.
With time 5 became the predominant ruthenium prod-
uct (trace d, after 1 h at 70 °C). Spectra acquired over
the following 3 h were practically identical with that of
trace d, which indicated the attainment of a thermos-
tationary state. 1H NMR spectra acquired during the
entire experiment at 70 °C (ca. 4 h) showed a gradual
increase in the concentration of the 2-ethylthiophenolate
product at the expense of that of BT. After the catalytic
reaction was quenched by cooling to room temperature,
5 was the only metal product visible by 31P{1H} NMR
(25) Complex 4, which is an intermediate of 5, is an excellent
catalyst for the chemoselective hydrogenation of BT to DHBT.11

2644 Organometallics, Vol. 17, No. 12, 1998
Bianchini et al.
F igu r e 3. 31P{1H} NMR study (sapphire tube, THF-d8, 81.01 MHz) of the catalytic hydrogenolysis of BT in the presence
of 3 (30 bar of H2, substrate/catalyst ratio 30, substrate/KOBut ratio 10): (a) pure sample of 3; (b) at 20 °C for 30 min; (c)
at 40 °C for 30 min; (d) at 40 °C for 1.5 h; (e) at 40 °C for 3 h; (f) at 70 °C for 30 min; (g) after the tube was cooled to 40
°C; (h) after the tube was heated to 70 °C for 30 min; (i) after 1 day at 20 °C.
acquired in the subsequent 1 h were practically identical
with that of trace f. The NMR probe was cooled to 40
°C and heated again to 70 °C. 31P{1H} NMR spectra
recorded at these two temperatures showed the disap-
pearance (trace g, at 40 °C) and the re-formation of 2
°C. As shown in Scheme 4 for the trihydride complex
[(triphos)RhH3], the reductive elimination of H2 is a
necessary step to generate the 16e- system [(triphos)-
RhH], which ultimately inserts into the C2-S bond of
BT.21 One may thus conclude that a temperature of 70
°C is required to promote the reductive elimination of
H2 from 2 and generate the fragment [(triphos)RuH]-,
which is electronically and sterically tailored for the
cleavage of BT according to the reaction sequence
summarized in Scheme 4.21,26 HPNMR spectroscopy
shows that the energy barrier to C-S insertion is much
lower (40 °C) when 3 is employed as catalyst precursor.
Moreover, the formation of the trihydride 2 does not
precede the C-S bond cleavage, although 3 has been
independently shown to react with H2 and KOBut to give
2. The different behavior of 3 may be interpreted
assuming that the generation of the active Ru(0)
fragment [(triphos)RuH]- proceeds via an alternative,
lower energy reductive elimination reaction. For ex-
ample, amine elimination from ruthenate complexes of
the formula [(triphos)Ru(H)2(NR2)]- (R ) H, Et; com-
pounds of this type are common intermediates in
transition-metal-catalyzed hydrogenation of nitriles17
and imines27) may well account for the generation of
[(triphos)RuH]-. On the basis of this reasoning as well
as previous reports,20,21 one may readily conclude that
also for ruthenium the energy barrier to C-S insertion
by [(triphos)RuH]- is lower than that required to
promote the elimination of H2 from the corresponding
trihydrido complex.
1
(trace h, after 30 min at 70 °C). The H NMR spectra
acquired during the entire experiment at 70 °C (ca. 3
h) showed a gradual increase of the concentration of the
2-ethylthiophenolate product at the expense of that of
BT, while the amount of DHBT formed in the first stage
of the experiment did not increase during the entire
experiment at 70 °C. After the catalytic reaction was
quenched by cooling to room temperature, 5 was the
only metal product visible by 31P{1H} NMR spectroscopy
(trace i). GC/MS analysis of the contents of the tube
gave the following product distribution: ETP (11%), BT
(84%), DHBT (5%). The amines NH2Et, NHEt2, and
NEt3 in a ratio of 9:3:1 were also detected in the reaction
mixture. In an identical experiment, after all 3 was
transformed into 5 at the stage of trace i, the tube was
removed from the spectrometer and set aside. The 31P-
{1H} NMR spectrum acquired after 1 day showed the
formation (20%) of 2 at the expense of 5 (trace j).
In conclusion, the HPNMR studies suggest that 1 and
3 are precursors to the same catalytically active species.
Indeed, the presence of the amine side products in the
reactions assisted by 3 does not seem to have any
mechanistic influence on the catalytic hydrogenolysis
of BT. An interesting difference between the two
catalyst precursors may be envisaged, however, in the
mechanism of generation of the catalyst, i.e. the 16e-
fragment [(triphos)RuH]-. In the reactions assisted by
1, the first metal complex to form is the trihydride
ruthenate 2, which cleaves the C2-S bond of BT at 70
(26) Dong, L.; Duckett, S. B.; Ohman, K. F.; J ones, W. D. J . Am.
Chem. Soc. 1992, 114, 151.
(27) Willoughby, C. A.; Buchwald, S. I. J . Am. Chem. Soc. 1994, 116,
8952.

HDS Activity of Ruthenium-Based Catalysts
Organometallics, Vol. 17, No. 12, 1998 2645
Sch em e 6
Con clu sion s
In combination with a strong Brønsted base such as
KOBut, the Ru(II) complexes [(triphos)RuH(BH4)] (1)
and [(triphos)Ru(NCMe)3](BPh4)2 (3) give rise to an
efficient and selective homogeneous catalyst for the
hydrogenolysis of BT to ETP in THF under mild reaction
conditions (g70 °C, 30 bar of H2). Both precursors
generate the same catalytically active species, i.e. the
16e- fragment [(triphos)RuH]-. A clean precursor to
this Ru(0) fragment is the ruthenate complex K[(triph-
os)RuH3 (2), obtained by the reaction of KOBut with 1.
In model studies at 70 °C, it has been shown that the
trihydride 2 reacts with BT, yielding the 2-vinylthiophe-
nolate complex K[(triphos)Ru(η3-S(C6H4)CHdCH2)] (6),
which transforms into the 2-ethylthiophenolate product
K[(triphos)Ru(H)2(o-S(C6H4)C2H5)] (5) by treatment with
H2. Under 30 bar of H2, 5 is in equilibrium with the
trihydride 2 and free ETP. At 70 °C, the equilibrium
concentration of 2 increases with the concentration of
KOBut as a consequence of the accelerated elimination
of ETP as K[o-S(C6H4)C2H5].
Irrespective of the catalyst precursor, 31P HPNMR
spectroscopy shows that the only detectable ruthenium
complexes under catalytic conditions are 5 and 2 in
equilibrium concentrations that depend on both the
temperature and the amount of added base.
The intermediacy of DHBT in the hydrogenolysis of
BT has been ruled out by several experiments. These
include in situ NMR spectroscopy, which shows that 2
reacts with BT at 70 °C to yield the C-S insertion
product 5, whereas at this temperature DHBT is not
transformed.
On the basis of all of this experimental evidence, one
may conclude that the conversion of BT to ETP de-
scribed in this paper takes place through the mechanism
already established for the base-assisted hydrogenolysis
of thiophenes catalyzed by [(triphos)MH] fragments (M
) Rh, Ir).8 The mechanism, illustrated in Scheme 6 for
the [(triphos)RuH]- catalyst, involves the usual steps
of C-S insertion, hydrogenation of the C-S-inserted
thiophene to the corresponding thiolate, and base-
assisted reductive elimination of the thiol to complete
the cycle. Also in the case of ruthenium (HPNMR
evidence, rate-accelerating effect of the base), the re-
moval of the thiol product from the dihydride thiolate
intermediate is proposed to be the rate-determining
step.
mixtures. Complex 8 is a selective catalyst for the
hydrogenation of BT to DHBT in fact. Accordingly, the
Brønsted base can play three distinct roles in the
present hydrogenolysis reactions: it promotes the for-
mation of the catalyst [(triphos)RuH]- by heterolytic
splitting of H2, speeds up the reaction rates, and
improves the chemoselectivity.
As a final consideration, we wish to stress that a novel
trihydride Ru(II) complex with great potential in ho-
mogeneous catalysis has been synthesized. The ruth-
enate 2 is a very electron-rich complex which can
generate highly reactive unsaturated fragments with
different chemical properties by either thermolysis (i.e.
the Ru(0) monohydride [(triphos)RuH]- or reaction with
proton donors (i.e. the Ru(II) dihydride [(triphos)Ru-
(H)2]). Either fragment has already been found capable
of effectively catalyzing the hydrogenation of various
substrates such as thiophenes, ketones, nitriles, imines,
olefins, and nitrogen-containing heterocycles (quino-
lines, pyrroles) with distinct mechansims and selectivi-
ties.11
Ack n ow led gm en t. A contract (PR1/C) from the
Ministero dell’Ambiente of Italy and an international
cooperation agreement between the CNR (Italy) and
CONICIT (Venezuela) are gratefully acknowledged for
financial support.
In the course of the catalytic reactions, small quanti-
ties (e1%) of the hydrogenation product DHBT are often
produced, particularly in the reactions with base to
substrate ratios lower than 1. The formation of DHBT
can further be reduced by increasing the concentration
of the base so as to minimize the presence of the
tetrahydride complex [(triphos)RuH4] (8) in the catalytic
Su p p or tin g In for m a tion Ava ila ble: Text giving relevant
physicochemical characteristics of the complexes [(triphos)Ru-
(µ-OH)3Ru(triphos)]BPh4, [(triphos)RuH4], [(triphos)RuH(µ-H)2-
HRu(triphos)], and [(triphos)Ru(H)2(CO)] (1 page). Ordering
information is given on any current masthead page.
OM9800804