Mimicking the HDS activity of promoted tungsten catalysts. A homogeneous modeling study using a two-component tungsten/rhodium system

Article abstract of DOI:10.1021/om970711e

Reaction of W(CO)₅THF with (triphos)Rh[η³-S(C₆H₄)CH=CH₂] (1), obtained by insertion of the 16e^- fragment [(triphos)RhH] into the C₂-S bond of benzo[b]thiophene (BT), gives the dimer (triphos)Rh[η³-(CO)₅WS(C₆H ₄)CH=CH₂] (2; triphos = MeC(CH₂PPh₂)₃). Unlike 1, the heterometal dimer 2 reacts with H₂ (30 atm) above 70°C in THF, undergoing the desulfurization of the C-S-inserted BT. As a result, a mixture of the hydrido carbonyl species (triphos)RhH(CO), ethylbenzene, and WS_x (x_av = 1.5) is obtained. High-pressure NMR spectroscopy in the temperature range from 20 to 70°C shows that the desulfurization step is preceded by the formation by the dimer (triphos)RhH(μ-H)[μ-o-S(C₆H₄)C₂H ₅]W(CO)₄ (5), in which the Rh and W centers are held together by bridging 2-ethylthiophenolate and hydride ligands. Complex 5 has been characterized in both the solid state (single-crystal X-ray analysis) and solution (multinuclear NMR spectroscopy). The desulfurization of 5 occurs also by thermolysis in THF at 120°C under a nitrogen atmosphere. Reaction of 5 with CO (30 atm, 40°C) gives the complex [(triphos)Rh(CO)₂][(CO)₅W(o-S(C₆H ₄)C₂H₅)] (8), in which the thiolate ligand is η¹-S bound to the tungsten atom in the complex anion [(CO)₅W-(o-S(C₆H₄)C₂H ₅)]^-. The hydrogenation of 8 (30 atm of H₂, > 70°C) gives exclusively free 2-ethylthiophenol. The carbonylation of 2 (30 atm of CO, room temperature) results in the formation of [(triphos)Rh(CO)₂][(CO)₅W(o-S(C₆H ₄)CH=CH₂)] (3), in which the 2-vinylthiophenolate ligand is η¹-S bound to the tungsten atom. The possible similarity in the C-S bond cleavage mechanism in the desulfurization of 5 to those occurring in the HDS over promoted heterogeneous catalysts is discussed.

Full text of DOI:10.1021/om970711e

5696
Organometallics 1997, 16, 5696-5705
Mim ick in g th e HDS Activity of P r om oted Tu n gsten
Ca ta lysts. A Hom ogen eou s Mod elin g Stu d y Usin g a
Tw o-Com p on en t Tu n gsten /Rh od iu m System
Claudio Bianchini,* M. Victoria J ime´nez, Andrea Meli, Simonetta Moneti,
Ve´ronique Patinec, 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 August 12, 1997X
Reaction of W(CO)5THF with (triphos)Rh[η3-S(C6H4)CHdCH2] (1), obtained by insertion
of the 16e- fragment [(triphos)RhH] into the C2-S bond of benzo[b]thiophene (BT), gives
the dimer (triphos)Rh[η3-(CO)5WS(C6H4)CHdCH2] (2; triphos ) MeC(CH2PPh2)3). Unlike
1, the heterometal dimer 2 reacts with H2 (30 atm) above 70 °C in THF, undergoing the
desulfurization of the C-S-inserted BT. As a result, a mixture of the hydrido carbonyl species
(triphos)RhH(CO), ethylbenzene, and WSx (xav ) 1.5) is obtained. High-pressure NMR
spectroscopy in the temperature range from 20 to 70 °C shows that the desulfurization step
is preceded by the formation by the dimer (triphos)RhH(µ-H)[µ-o-S(C6H4)C2H5]W(CO)4 (5),
in which the Rh and W centers are held together by bridging 2-ethylthiophenolate and
hydride ligands. Complex 5 has been characterized in both the solid state (single-crystal
X-ray analysis) and solution (multinuclear NMR spectroscopy). The desulfurization of 5
occurs also by thermolysis in THF at 120 °C under a nitrogen atmosphere. Reaction of 5
with CO (30 atm, 40 °C) gives the complex [(triphos)Rh(CO)2][(CO)5W(o-S(C6H4)C2H5)] (8),
in which the thiolate ligand is η1-S bound to the tungsten atom in the complex anion [(CO)5W-
(o-S(C6H4)C2H5)]-. The hydrogenation of 8 (30 atm of H2, >70 °C) gives exclusively free
2-ethylthiophenol. The carbonylation of 2 (30 atm of CO, room temperature) results in the
formation of [(triphos)Rh(CO)2][(CO)5W(o-S(C6H4)CHdCH2)] (3), in which the 2-vinylthio-
phenolate ligand is η1-S bound to the tungsten atom. The possible similarity in the C-S
bond cleavage mechanism in the desulfurization of 5 to those occurring in the HDS over
promoted heterogeneous catalysts is discussed.
In tr od u ction
also comprises late transition metals.3,4 These are
termed promoters and almost exclusively belong to
group 8 (Co, Rh, Ir, Ru, Os, Ni, Pd, Pt).
Due to environmental concerns, increasing research
attention is being paid to the development of more
efficient technologies for sulfur removal in crude oil. The
existing processes can hardly reduce the sulfur contents
in fuels to the internationally agreed level of 100 ppm
and most likely will be unable to meet new standards
dictated by stricter pollution regulations.1 In refineries,
sulfur, in the form of various organosulfur compounds
among which the thiophenes are the most difficult to
degrade, is eliminated from crude oil as H2S (later dis-
posed of via the Claus process)2 by treatment with a
high pressure of H2 over hot heterogeneous catalysts.3
This process is referred to as hydrodesulfurization
(HDS) and is typically catalyzed by metal sulfides de-
posited on a support; Mo and W are essential compo-
nents, but increased catalytic activity, particularly
toward the thiophenes, is observed when the catalyst
Despite the huge amount of studies dealing with the
structure and activity of heterogeneous HDS catalysts,
particularly those classified as “Co/Mo/S” phase,3a little
is known about the chemical role played by the promoter
metal atom. It is generally agreed that the promotion
may be due to the creation of new sites with chemical-
physical properties different from those responsible for
the HDS by unpromoted catalysts or single-component
catalysts containing only metal promoter sulfides.3a For
the specific case of the thiophenes, it has been suggested
that the active centers for the activation of the substrate
are the promoter atoms located at the edge plane of a
WS2 (MoS2) single slab, whereas H2 activation occurs
on WS2 (MoS2) to give hydrosulfyl species.4a Consis-
(3) (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; 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) Friend, C. M.; Roberts, J . T. Acc. Chem. Res. 1988,
21, 394. (f) Leffler, W. L. Petroleum Refining; PennWell: Tulsa, OK,
1985. (g) Topsøe, H.; Clausen, B. S. Catal. Rev.-Sci. Eng. 1984, 26,
395. (h) Chianelli, R. R. Catal. Rev.-Sci. Eng. 1984, 26, 361.
(4) (a) Startsev, A. N. Catal. Rev.-Sci. Eng. 1995, 37, 353. (b) Benson,
J . W.; Schrader, G. L.; Angelici, R. J . J . Mol. Catal. A 1995, 96, 283.
(c) Wiegand, B. C.; Friend, C. M. Chem. Rev. 1992, 92, 491. (d) Girgis,
M. J .; Gates, B. C. Ind. Eng. Chem. Res. 1991, 30, 2021. (e) Harris, S.;
Chianelli, R. R. J . Catal. 1984, 86, 400. (f) Pecoraro, T. A.; Chianelli,
R. R. J . Catal. 1981, 67, 430.
X Abstract published in Advance ACS Abstracts, November 15, 1997.
(1) (a) Hydrotreating Technology for Pollution Control; Occelli, M.
L., Chianelli, R., Eds.; Dekker: New York, 1996. (b) Navarro, R.;
Pawelec, B.; Fierro, J . L. G.; Vasudevan, P. T.; Cembra, J . F.; Arias,
P. L. Appl. Catal. A 1996, 137, 269. (c) Vasudevan, P. T.; Fierro, J . L.
G. Catal. Rev.-Sci. Eng. 1996, 38, 161. (d) Shih, S. S.; Mizrahi, S.;
Green, L. A.; Sarli, M. S. Ind. Eng. Chem. Res. 1992, 31, 1232. (e)
Takatsuka, T.; Wada, Y.; Suzuki, H.; Komatsu, S.; Morimura, Y. J .
J pn. Petrol. Inst. 1992, 35, 179.
(2) (a) Shaver, A.; El-khateeb, M.; Lebuis, A.-M. Angew. Chem., Int.
Ed. Engl. 1996, 35, 2362. (b) Grancer, P. Hydrocarbon Process. 1978,
57, 155, 257.
S0276-7333(97)00711-5 CCC: $14.00 © 1997 American Chemical Society

Mimicking the HDS Activity of Promoted W Catalysts
Organometallics, Vol. 16, No. 26, 1997 5697
Sch em e 1
tently, both the binding of the thiophene directly to the
promoter and the presence of Mo-SH groups on the
surface of Co/Mo/S phase catalysts have been shown by
EXAFS measurements.3a Many other salient features
of the HDS process regarding the specific and/or syn-
ergic role of the component and of the promoter in
crucial steps of the HDS process are still rather obscure,
however (e.g., C-S insertion, C-S cleavage, removal of
surface sulfur, hydrogen transfer from M-SH groups
to activated thiophene).
such as aqueous biphase catalysis, for sulfur removal
in distillates.6a,8
From a perusal of the homogeneous studies recently
reviewed,5 one may readily infer that C-S bond scission
in thiophenes is almost exclusively brought about by
promoter metal complexes. In analogy to heterogeneous
catalysts, however, the desulfurization step to H2S and
hydrocarbons has been observed to occur in solution only
using either polymetallic species or external sources of
“activated” hydrogen.5b,9-13
In order to gain insight into the mechanism of the
many steps occurring in the heterogenous HDS and
ultimately assist in the design of improved catalysts,
considerable effort is being directed toward the study
of model systems involving soluble metal complexes and
thiophenes. Indeed, many types of reactions between
discrete organometallic complexes and thiophenes occur
also on the surface of heterogeneous catalysts, and the
mechanistic understanding obtained in solution has
been applied to elucidate surface phenomena.5 With the
advent of sophisticated physical-chemical techniques in
solution, e.g., high-pressure NMR spectroscopy6 and
transient UV flash kinetic spectroscopy,7 several break-
throughs have recently been obtained that, inter alia,
have largely contributed to design alternative strategies,
Selected examples of HDS of thiophenes by soluble
organometallic complexes are shown in Scheme 1.
Among these, the Co/Mo/S cluster described by Curtis
and co-workers is the model that more closely resembles
the heterogeneous counterparts in terms of both struc-
tural motif and HDS activity (the hydrogen atoms
necessary for the hydrogenolysis step are externally
added and not already incorporated into the organome-
tallic precursor). More recently, the Co/Mo/S cluster has
successfully been employed to show that the C-S bond
scission in the desulfurization of aromatic and aliphatic
thiols may occur in a homolytic fashion and that thiolate
and sulfido groups can move over the face of the cluster
as they are supposed to do over the surface of Co/Mo/S
phase catalysts.14 Both of these aspects, e.g., the
cooperation of component and promoter metals in the
(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.
(8) (a) Bianchini, C.; Meli, A.; Ital. Pat In Aqueous-Phase Organo-
metallic Catalysis: Concepts and Applications; Cornils, B., Herrmann,
W. A., Eds.; VCH: Weinheim, Germany, in press. (b) 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. (c) Southern,
T. G. Polyhedron 1989, 8, 407.
(9) Riaz, U.; Curnow, O. J .; Curtis, M. D. J . Am. Chem. Soc 1994,
116, 4357.
(6) (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.
(10) (a) Vicic, D. A.; J ones, W. D. Organometallics 1997, 16, 1912.
(b) J ones, W. D.; Chin, R. M. J . Am. Chem. Soc. 1994, 116, 198.
(11) Ogilvy, A. E.; Draganjac, M.; Rauchfuss, T. B.; Wilson, S. R.
Organometallics 1988, 7, 1171.
(12) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F.; Frediani, P.;
Herrera, V.; Sa´nchez-Delgado, R. A. J . Am. Chem. Soc. 1993, 115, 2731.
(13) Garcia, J . J .; Mann, B. E.; Adams, H.; Bailey, N. A.; Maitlis, P.
M. J . Am. Chem. Soc. 1995, 117, 2179.
(7) Bianchini, C.; Casares, J . A.; Osman, R.; Pattison, D. I.; Pe-
ruzzini, M.; Perutz, R. N.; Zanobini, F. Organometallics 1997, 16, 4611.
(14) Curtis, M. D.; Druker, S. H. J . Am. Chem. Soc. 1997, 119, 1027.

5698 Organometallics, Vol. 16, No. 26, 1997
Bianchini et al.
200 instrument equipped with a 5 mm inverse probe and a
BFX-5 amplifier device. 13C-DEPT, 1H,13C 2D-HETCOR, and
1H,1H 2D-COSY NMR experiments were conducted on the
Bruker ACP 200 spectrometer. 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, Italy).18 Note: Since high gas pres-
sures 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 LAOCN319
and Davins20 running on a Compaq Deskpro 386/25 personal
computer. The initial choices of shifts and coupling constants
were refined by iterative least-squares calculations using
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 FT) SPB-1 Supelco fused silica
capillary column. GC/MS analyses were performed on a
Shimadzu QP 5000 apparatus equipped with a column identi-
cal with that used for GC analyses.
desulfurization step and the walking of the thiolate from
one metal to another, have attracted our attention
within a long-lasting research project on HDS modeling.
We have, for example, demonstrated that the 16e-
fragment [(triphos)RhH] cleaves almost any kind of
thiophenic substrate to give C-S insertion products15
that behave as catalyst precursors for the hydrogenoly-
sis of thiophene,6b benzo[b]thiophene,6b,d dibenzo[b,d]-
thiophene,6b and dinaphtho[2,1-b:1′,2′-d]thiophene6c to
the corresponding thiols (triphos ) MeC(CH2PPh2)3). In
no case was the desulfurization of the thiol product
observed in a clear homogeneous process even under
drastic reaction conditions, however. We therefore
decided to examine the chemistry of a typical rhodium
C-S insertion product appropriately modified so as to
contain tungsten bound to the sulfur atom. To this end,
we synthesized the benzo[b]thiophene-derived hetero-
metal complex (triphos)Rh[η3-(CO)5WS(C6H4)CHdCH2]
and were gratified to find that the hydrogenation of this
dimer indeed results in the desulfurization of the
cleaved thiophene though a well-defined intermediate
species. Also, we discovered that the slippage of the
thiolate fragment from the Rh promoter, responsible for
the C-S insertion, to the W component can occur by
addition of a nucleophile to the heterometal dimer.
The present paper constitutes a detailed account of
this study, part of which was previously communi-
cated.16
Syn th esis of (tr ip h os)Rh [η3-(CO)5WS(C6H4)CHdCH 2]
(2). A sample of W(CO)6 (0.12 g, 0.33 mmol), dissolved in THF
(30 mL), was placed in a quartz reactor and converted to
W(CO)5(THF) by photolysis at -10 to +10 °C under a flow of
nitrogen. After 4 h, this solution was cannulated into a flask
containing (triphos)Rh[η3-S(C6H4)CHdCH2] (1; 0.28 g, 0.33
mmol) in THF (20 mL) and the resulting mixture stirred for 1
h at room temperature. The volume of the solvent was reduced
to ca. 15 mL and ethanol (40 mL) was added. Orange-yellow
microcrystals of 2 were obtained in 85% yield after filtering,
washing (ethanol and n-pentane), and drying. Anal. Calcd
(found) for C54H46O5P3RhSW: C, 54.65 (54.55); H, 3.91 (3.81);
Rh, 8.67 (8.49). IR: ν(CO) 2060 (m), 1966 (m), 1917 (vs), 1872
Exp er im en ta l Section
Gen er a l In for m a tion . All reactions, manipulations, and
chromatographic separations (except as stated otherwise) were
routinely performed under a nitrogen atmosphere by using
standard Schlenk techniques. Tetrahydrofuran (THF) was
distilled from LiAlH4, stored over molecular sieves, and purged
with nitrogen prior to use. Benzo[b]thiophene (BT, Aldrich)
was sublimed prior to use. All other chemicals were com-
mercial products and were used as received without further
purification. (triphos)Rh[η3-S(C6H4)CHdCH2]15b and (triphos)-
(s) cm-1 31P{1H} NMR (CD2Cl2, 20 °C): AMQX spin system,
.
δ 30.9 (ddd, J (PAPM) ) 36.0 Hz, J (PAPQ) ) 32.0 Hz, J (PARh)
) 115.8 Hz, PA), -7.1 (ddd, J (PMPQ) ) 42.6 Hz, J (PMRh) )
112.4 Hz, PM), -10.0 (ddd, J (PQRh) ) 104.6 Hz, PQ). 1H NMR
(CD2Cl2, 20 °C): δ 3.84 (m, J (H2'H3) ) 9.5 Hz, J (H2'H2) ) 1.7
Hz, J (H2'Rh) ) 1.8 Hz, H2'), 2.86 (m, J (H3H2) ) 8.3 Hz, J (H3-
Rh) ) 0.9 Hz, H3), 2.70 (m, J (H2Rh) ) 1.5 Hz, H2). The J (HH)
17
1
and J (HRh) values were determined on the basis of H{31P}
WH6 were prepared as previously described. All metal
NMR experiments. 13C{1H} NMR (CD2Cl2, 20 °C): δ 42.5 (dm,
J (CP) 35.9, C2), 65.5 (dm, J (CP) 28.7, C3), 200.1 (s, 183W
satellites with J (CW) ) 128.7 Hz, COeq), 202.6 (s, COax). The
remaining 13C resonances were obscured by those of the triphos
carbon atoms.
complexes were collected on sintered-glass frits and washed
with appropriate solvents before being dried under a stream
of nitrogen. Photochemical reactions were performed by using
a Helios Italquartz UV 13F apparatus. The photolysis source
was a 135 W (principal emission wavelength 366 nm) high-
pressure mercury vapor immersion lamp equipped with a
water filter to remove excess heat. Reactions under controlled
pressure of hydrogen or carbon monoxide were performed with
a Parr 4565 reactor equipped with a Parr 4842 temperature
and pressure controller. Infrared spectra were recorded on a
Perkin-Elmer 1600 Series FT-IR spectrophotometer using
either samples mulled in Nujol between KBr plates or THF
and CH2Cl2 solutions. Deuterated solvents for NMR measure-
ments were dried over molecular sieves. 1H (200.13 MHz), 13C-
{1H} (50.32 MHz), and 31P{1H} (81.01 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, 13C) or 85% H3PO4 (31P). Broad-band and selective 1H-
{31P} NMR experiments were carried out on the Bruker ACP
Rea ction of 2 w ith Ca r bon Mon oxid e. A THF (50 mL)
solution of 2 (0.24 g, 0.2 mmol) was reacted with CO (30 atm)
at room temperature for 8 h in a Parr reactor. After the bomb
was depressurized and vented under a nitrogen stream, the
contents were transferred into a Schlenk-type flask. The
volatiles were then removed in vacuo at room temperature,
and a portion of the residue dissolved in CD2Cl2 was analyzed
by NMR spectroscopy. In the 31P{1H} NMR spectrum only the
resonances due to the vinylthiophenolate complex 1 and the
dicarbonyl [(triphos)Rh(CO)2]+ (3:1 ratio) were observed.15b,21
Besides the resonances due to 1 and to the dicarbonylrhodium
cation, the 1H NMR spectrum showed a duo of doublets of
doublets (δ 5.61, J (HH) ) 17.6, 1.7 Hz; δ 5.26, J (HH) ) 11.1,
1.7 Hz) characteristic of the cis and trans protons of a vinyl
group. The rest of the residue was chromatographed on a silica
column by using as eluants n-hexane to separate W(CO)6, THF/
n-hexane (1:1 ratio) to separate 1, and THF to separate
(15) (a) Bianchini, C.; Meli, A.; Pohl, W.; Vizza, F.; Barbarella, G.
Organometallics 1997, 16, 1517. (b) 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. (c) Bianchini, C.;
J ime´nez, M. V.; Meli, A.; Vizza, F. Organometallics 1995, 14, 3196.
(16) Bianchini, C.; J ime´nez, M. V.; Mealli, C.; Meli, A.; Moneti, S.;
Patinec, V.; Vizza, F. Angew. Chem., Int. Ed. Engl. 1996, 35, 1706.
(17) Michos, D.; Luo, X.-L.; Faller, J . W.; Crabtree, R. H. Inorg.
Chem. 1993, 32, 1370.
(18) CNR (Bianchini, C.; Meli, A.; Traversi, A.), Ital. Pat. FI
A000025, 1997.
(19) (a) Bothner-By, A. A.; Castellano, S. QCPE 1967, 11, 111. (b)
Castellano, S.; Bothner-By, A. A. J . Chem. Phys. 1964, 41, 3863.
(20) Stephenson, D. S.; Binsch, G. J . Magn. Reson. 1980, 37, 409.
(21) Ott, J .; Venanzi, L. M.; Ghilardi, C. A.; Midollini, S.; Orlandini,
A. J . Organomet. Chem. 1985, 291, 89.

Mimicking the HDS Activity of Promoted W Catalysts
Organometallics, Vol. 16, No. 26, 1997 5699
[(triphos)Rh(CO)2][(CO)5W(o-S(C6H4)CHdCH2)] (3). This prod-
uct was characterized on the basis of IR and 1H and 31P{1H}
NMR spectroscopy. IR (THF): ν(CO) 2060 (s), 1990 (s), 1916
Rea ction of 2 w ith Hyd r ogen . Syn th esis of (tr ip h os)-
Rh H(µ-H)[µ-o-S(C6H4)C2H5]W(CO)4 (5). A THF (50 mL)
solution of 2 (0.24 g, 0.2 mmol) was reacted with H2 (30 atm)
at 70 °C for 3 h in a Parr reactor. After the bomb was
(s), 1852 (m) cm-1 1H NMR (CD2Cl2, 20 °C): cation, δ 7.8-
.
6.8 (m, Ph triphos), 2.61 (m, CH2 triphos), 1.83 (q, J (HP) )
3.8 Hz, CH3 triphos); anion, δ 7.1 (H3, masked by the aromatic
depressurized and vented under a nitrogen stream, the
contents were transferred into a Schlenk-type flask. The
solution was concentrated to ca. 20 mL under vacuum.
Portionwise addition of n-heptane (10 mL) led to the precipita-
tion of 5 as brick red crystals which were washed with
n-pentane; yield 75%. Anal. Calcd (found) for C53H50O4P3-
RhSW: C, 54.75 (54.68); H, 4.33 (4.25); Rh, 8.85 (8.69). IR:
Nujol mull, ν(Rh-H) 1995 (s), ν(CO) 2057 (w), 1868 (s), 1852
(s), 1811 (s) cm-1; CH2Cl2, ν(Rh-H) 1996 (s), ν(CO) 2058 (w),
1
protons of triphos; the chemical shift was determined by a H-
1H 2D-COSY experiment), 5.61 (dd, J (H2transH3) ) 17.6 Hz,
J (H2transH2cis) ) 1.7 Hz, H2trans), 5.26 (dd, J (H2cisH3) ) 11.1,
H
2cis), remaining resonances were obscured by those of the
aromatic protons of triphos. 31P{1H} NMR (CD2Cl2, 20 °C): δ
8.7 (d, J (PRh) ) 98.6 Hz).
Rea ction of 3 w ith [NBu 4]I. A THF (30 mL) solution of
3 (0.15 g, 0.12 mmol) was treated with an equimolar amount
of [NBu4]I (0.05 g, 0.14 mmol) at room temperature for 10 h.
Chromatography on a silica column, with first THF/n-hexane
(1:1 ratio) and then pure THF as eluants, gave (triphos)Rh-
(CO)I21 and [NBu4][(CO)5W(o-S(C6H4)CHdCH2)] (4; yield 80%),
respectively. Anal. Calcd (found) for C29H43NO5SW: C, 49.65
(49.06); H, 6.18 (6.19); N, 2.00 (1.89); S, 4.57 (4.32). IR
1872 (vs), 1860 (sh), 1822 (s) cm-1 31P{1H} NMR (CD2Cl2, 20
.
°C): AMQX spin system, δ 23.1 (ddd, J (PAPM) ) 34.1 Hz,
J (PAPQ) ) 26.9 Hz, J (PARh) ) 109.7 Hz, PA), 13.6 (ddd, J (PMPQ)
) 21.7 Hz, J (PMRh) ) 105.5 Hz, PM), -9.2 (ddd, J (PQRh) )
74.5 Hz, PQ). 31P NMR (CD2Cl2, 20 °C): δ 23.1 (d, J (PARh) )
110 Hz, PA), 13.6 (t, J (PMH4) ) J (PMRh) ) 100 Hz, PM), -9.2
(dd, J (PQRh) ) 75 Hz, J (PQH5) ) 180 Hz, PQ). Variable-
temperature 31P{1H} NMR spectra (THF-d8, 20-70 °C, sap-
phire tube) are reported in Figure 3. 1H NMR (CD2Cl2, 20
°C): δ 2.63 (dq, J (H3′H3) ) 14.0 Hz, J (H3′H2) ) 7.4 Hz, H3′),
2.18 (dq, J (H3H2) ) 7.4 Hz, H3), 1.10 (t, H2), -9.88 (dm, J (H4-
(THF): ν(CO) 2055 (w), 1914 (s), 1851 (m) cm-1
.
1H NMR
(THF-d8, 20 °C): cation, δ 3.51 (m), 1.88 (m), 1.57 (m), 1.15
(t); anion, δ 7.4-6.9 (m, C6H4), 7.13 (dd, J (H3H2trans) ) 17.5
Hz, J (H3H2cis) ) 11.0 Hz, H3), 5.69 (dd, J (H2transH2cis) ) 1.8
Hz, H2trans), 5.30 (dd, H2cis). 13C{1H} NMR (THF-d8, 20 °C):
cation, δ 60.7 (NCH2), 26.1 (CH2CH2), 21.8 (CH2CH3), 15.3
(CH3); anion, δ 152.3 (C-S), 138.2 (C-C3), 136.4 (C3), 134.1
(C-H), 128.3 (C-H), 127.6 (C-H), 122.8 (C-H), 112.8 (C2);
carbonyls, δ 203.5 (ax), 199.5 (183W satellites with J (CW) )
126.8 Hz, eq).
Rea ction of 4 w ith MeI. A 3-fold excess of neat MeI (26
µL, 0.42 mmol) was syringed into a stirred solution of 4 (0.10
g, 0.14 mmol) in THF (30 mL) at room temperature. After 1
h, the solution was pumped to dryness at room temperature
and the residue was chromatographed on a silica column (n-
hexane as eluant). The organic phase was concentrated to
dryness in vacuo, and the residue was characterized by 1H
NMR and GC/MS spectroscopy as o-(methylthio)styrene by
comparison to an authentic specimen.15b,22
Rea ction s of 2 w ith Hyd r ogen in a Sa p p h ir e HP NMR
Tu be. In a typical experiment, a 10-mm sapphire HPNMR
tube was charged with a THF-d8 (2 mL) solution of 2 (0.03 g,
0.025 mmol) under nitrogen, pressurized with hydrogen to 30
atm at room temperature and then placed into the NMR probe
preheated at 70 °C. The reaction was monitored by 31P{1H}
and 1H NMR spectroscopy. A sequence of spectra recorded
during this experiment is reported in Figure 1. In the 31P-
{1H} NMR spectrum recorded after ca. 1 h, the AMQX spin
system of 2 (50%) was accompanied by that of (triphos)RhH-
(µ-H)[µ-o-S(C6H4)C2H5]W(CO)4 (5) (see below). After a further
2 h, 5 was the only species detected in solution (Figure 1b).
The probe was then heated to 80 °C and the 31P{1H} NMR
spectra, recorded at this temperature every 30 min, showed
the slow but gradual conversion of 5 to the known hydrido
carbonyl species (triphos)RhH(CO)21 (6). Increasing the tem-
perature to 100 °C (Figure 1c) and 120 °C (Figure 1d) increased
the conversion rate of 5 to 6. The highest temperature
investigated represents the technical limit of the HPNMR tube.
At 120 °C complete conversion of 5 to 6 was achieved in ca. 1
h. Small amounts (5-10%) of the known (triphos)Rh(CO)[o-
S(C6H4)C2H5]6d (7) were also detected in solution at the end of
the reaction. The probe was cooled to room temperature
(Figure 1e), and the NMR tube was removed from the
spectrometer. An insoluble material was found to be deposited
on the wall of the tube. A sample of the solution, withdrawn
and analyzed by GC and GC/MS, was found to contain almost
a quantitative amount of ethylbenzene (based on 2, n-octane
as internal standard) and traces of W(CO)6.
Rh) ) 19.02 Hz, J (H4H5) ) 5.45 Hz, J (H4PM) ) 100.6 Hz, 183
W
satellites with J (H4W) ) 34.2 Hz, H4), -10.07 (dm, J (H5Rh)
) 6.76 Hz, J (H5PQ) ) 181.5 Hz, H5). The J (HH), J (HRh), and
1
J (HW) values were determined on the basis of H{31P} NMR
experiments. 13C{1H} NMR (CD2Cl2, 20 °C): δ 213.9 (s, CO),
210.6 (d, J (CP) ) 5.3 Hz, CO), 205.5 (d, J (CP) ) 3.8 Hz, CO),
205.2 (d, J (CP) ) 6.0 Hz, CO), 29.1 (s, C2), 14.9 (s, C3). Well-
shaped crystals of (triphos)RhH(µ-H)[µ-o-S(C6H4)C2H5]W(CO)4‚
CH2Cl2 (5‚CH2Cl2) were obtained by slow crystallization of 5
from CH2Cl2 and ethanol under nitrogen at room temperature.
Anal. Calcd (found) for C54H52Cl2O4P3RhSW: C, 51.99 (51.21);
H, 4.20 (4.22); Rh, 8.25 (8.12).
Rea ction of 5 w ith Hyd r ogen in a Sa p p h ir e HP NMR
Tu be. A 10 mm sapphire HPNMR tube was charged with a
THF-d8 (2 mL) solution of 5 (0.03 g, 0.026 mmol) under
nitrogen, pressurized with hydrogen to 30 atm at room
temperature and then placed into the NMR probe preheated
to 120 °C. The reaction was monitored by 31P{1H} and 1H
NMR spectroscopy. Compound 5 was found to convert to 6
(quantitative conversion in ca. 3 h) together with both ethyl-
benzene and a black insoluble material containing W and S
(vide infra).
Th er m a l Rea ction of 5 u n d er Nitr ogen in a Sa p p h ir e
HP NMR Tu be. A 10 mm sapphire HPNMR tube was charged
with a THF-d8 (2 mL) solution of 5 (0.03 g, 0.026 mmol) under
nitrogen at room temperature and then placed into the NMR
probe preheated at 120 °C. The reaction was monitored by
1
31P{1H} and H NMR spectroscopy. Although accompanied by
extensive decomposition to some unidentified species, both 6
and 7 were formed in a ca. 3:1 ratio.
Rea ction of 2 (or 5) w ith Hyd r ogen a t 120 °C in a n
Au tocla ve. A solution of 2 (or 5, 0.15 mmol) in THF (50 mL)
was placed into a Parr reactor, pressurized with hydrogen to
30 atm at room temperature, and heated to 120 °C with
stirring. After ca. 4 h, the reactor was cooled to room temper-
ature and slowly depressurized. The contents of the bomb
were transferred into a Schlenk-type flask. An insoluble black
material, formed during the reaction, was filtered off (ca. 30
mg) and analyzed by elemental analysis (C, H, N), atomic
absorption (Rh, W), gravimetric methods (P, S), and IR
spectroscopy; sulfur to tungsten ratios ranging from 1.3 to 1.8
were found over five preparations (average value of 1.5). A
sample of the filtrate, analyzed by GC and GC/MS, showed
the presence of a quantitative amount of ethylbenzene (n-
octane as internal standard) and traces of W(CO)6. The
remainder of the solution was concentrated to dryness in
vacuo, and the residue, dissolved in CD2Cl2, was studied by
1H and 31P{1H} NMR spectroscopy, which showed the complete
The hydrogenation reaction of 2 to give 5 was found to occur,
although slowly, even at room temperature (20% in 3 days).
(22) Crow, W. D.; McNab, H. Aust. J . Chem. 1979, 32, 123.

5700 Organometallics, Vol. 16, No. 26, 1997
Bianchini et al.
Ta ble 1. Su m m a r y of Cr ysta llogr a p h ic Da ta for
transformation of 2 (or 5) into the hydrido carbonyl species
(triphos)RhH(CO) (6). The known compound (triphos)Rh(o-
S(C6H4)C2H5)(CO) (7) was also detected in solution (<8%).
Rea ction of 5 w ith Ca r bon Mon oxid e in a n HP NMR
Tu be. A 10 mm sapphire HPNMR tube was charged with a
THF-d8 (1.5 mL) solution of 5 (0.03 g, 0.026 mmol) under
nitrogen and then pressurized with CO to 30 atm at room
temperature. The reactions were followed by variable-tem-
perature 31P{1H} and 1H NMR spectroscopy. No reaction
occurred at room temperature within 1 h. At 40 °C, quantita-
tive conversion of 5 to [(triphos)Rh(CO)2][(CO)5W(o-S(C6H4)-
C2H5)] (8) and H2 (1H NMR singlet at 4.6 ppm) occurred in ca.
2 h. Once formed, 8 is stable in THF solution under a nitrogen
atmosphere. 1H NMR (THF-d8, 20 °C): cation, δ 7.8-6.8 (m,
Ph triphos), 2.62 (m, CH2 triphos), 1.89 (q, J (HP) ) 3.8 Hz,
CH3 triphos); anion, δ 2.74 (q, J (H3H2) ) 7.4 Hz, H3), 1.18 (t,
H2), remaining resonances were obscured by those of the
aromatic protons of triphos. 31P{1H} NMR (THF-d8, 20 °C):
δ 9.5 (d, J (PRh) ) 98.1 Hz).
Reaction of [(tr iph os)Rh (CO)2][(CO)5W(o-S(C6H4)C2H5)]
(8) w ith Hyd r ogen in a n HP NMR Tu be. An HPNMR tube
containing a sample of 8, prepared as described above, was
pressurized with H2 to 30 atm at room temperature and was
then placed into the NMR probe. The reaction was monitored
by variable-temperature 1H NMR spectroscopy. Even at 70
°C, free 2-ethylthiophenol (ETP) started appearing in the
solution. After 1 h at 70 °C, the quantitative formation of ETP
was shown by both 1H NMR and GC/MS analysis of the
contents of the tube after this was depressurized. W(CO)6 was
the only tungsten carbonyl complex detected, while no produc-
tion of EB was observed.
Attem p ted Hyd r od esu lfu r iza tion Rea ction s of BT in
th e P r esen ce of Ca ta lytic Am ou n ts of Sin gle-Com p on en t
Com p lexes. In a typical reaction, a solution of (triphos)WH6
(9),17 (triphos)Mo(η6-C6H6) (10),23 [(triphos)MoH(η6-C6H6)]BF4
(11),23 [Cp*WS3](PPh4) (12),24 W(CO)6, or W(CO)5THF (0.025
mmol) and a 100-fold excess of BT (330 mg, 2.5 mmol) in THF
(30 mL) were placed into a Parr reactor, pressurized with
hydrogen to 30 atm at room temperature, heated to the desired
temperature (50-200 °C), and then immediately stirred. After
3 h, the reactor was cooled to room temperature and slowly
depressurized by bubbling the gaseous phase into an aqueous
solution of Pb(II) acetate. No formation of PbS was observed.
The contents of the reactor were transferred into a Schlenk-
type flask. A sample of the solution was withdrawn and
analyzed by GC and GC/MS. With all catalyst precursors
investigated, no catalytic transformation of BT was observed.
The tungsten hexahydrido complex transformed less than the
stoichiometric amount of BT into a mixture of EB and DHBT,
while no production of EB, DHBT, or ETP was found using
10-12. No catalytic activity was observed for reactions
performed in the presence of a large excess of either strong
Brønsted bases (KOBut) or acids (HBF4). In acidic media, the
hydrogenation of the tungsten trisulfide 12 led to the evolution
of some H2S.
5‚CH2Cl2
Crystal Parameters
formula
mol wt
cryst size, mm
cryst syst
space group
a, Å
C54H52Cl2O4P3RhSW
1247.65
0.4 × 0.05 × 0.12
monoclinic
P21/n (No. 14)
15.626(5)
19.129(5)
17.270(5)
92.33(1)
5158(3)
b, Å
c, Å
â, deg
V, Å3
Z
D
4
calcd, g cm-3
1.607
F(000)
2488
Measurement of Intensity Data
diffractometer
Philips PW1100
radiation (monochrom)
scan rate, deg min-1
scan width, deg
2θ range, deg
Cu KR, 1.5418 Å (graphite)
3-5.4
1.2 + 0.15 tan θ
5-110
data collected
-16 e h e 16, 0 e k e 20,
0 e l e 18
6802
6486
no. of data collected
no. of unique data
no. of params varied
µ(Cu KR), mm-1
R1 [I > 2σ(I)]
374
9.213
0.081
0.19
wR2
goodness of fit on F2
1.072
tion correction was applied via ψ scan with transmission
factors in the range 1.001-1.46.27 The computational work
was carried out by intensively using the program SHELXL93.28
Crystallographic details are reported in Table 1. The structure
was solved by direct methods using the SIR92 program,29 and
all of the non-hydrogen atoms were found through a series of
Fo Fourier maps. A CH2Cl2 solvent molecule was also intro-
duced, although it was affected by some disorder. Non-metal-
bound hydrogen atoms were introduced at calculated positions
at a late stage of refinement. The latter was carried out by
full-matrix least-squares calculations, initially with isotropic
thermal parameters. In the last least-squares cycles aniso-
tropic thermal parameters were used for the Rh, W, S, and P
atoms and also for the C atoms, except for those of the P phenyl
substituents and the CO molecules. In the last stage of
refinement a peak of 0.9 e/Å3 was detected in the ∆F map
corresponding to the expected bridging hydride ligand, and it
was succesfully refined.
No evidence whatsoever was found for the terminal H
ligand. In the final ∆F maps significant peaks were found
close to the heavy metal (largest value ca. 1.2 e/Å3), and were
considered as ripples of no chemical relevance.
Resu lts
X-r a y Da ta Collection a n d Str u ctu r e Deter m in a tion
of 5‚CH2Cl2. Intensities of a brick red crystal of 5‚CH2Cl2 were
collected on a Philips PW1100 FEBO diffractometer. A set of
25 carefully centered reflections having 10 < θ < 16.5° was
used to determine the cell constants. Three standard reflec-
tions were measured every 2 h for the orientation and the
intensity control. During data collection no decay for the
specimen was noticed. Intensity data were corrected for
Lorentz-polarization effects. Atomic scattering factors were
those reported by Cromer and Waber,25 with anomalous
dispersion correction taken from ref 26. An empirical absorp-
Syn th esis a n d Ch a r a cter iza tion of th e Heter o-
bim etallic Com plex (tr iph os)Rh [η3-(CO)5WS(C6H4)-
CHdCH2] (2). Complex 2 was straightforwardly pre-
pared by electrophilic addition of the unsaturated frag-
ment W(CO)5 to the sulfur atom of the 2-vinylthiophe-
nolate complex (triphos)Rh[η3-S(C6H4)CHdCH2] (1),
obtained by regioselective insertion of rhodium into the
C2-S bond of benzo[b]thiophene (BT) (Scheme 2).15b
Compound 2 is stable in both the solid state and
deaerated THF solutions, in which it does not decom-
(23) Kowalski, A. S.; Ashby, M. T. J . Am. Chem. Soc. 1995, 117,
12639.
(24) Kawaguchi, H.; Tatsumi, K. J . Am. Chem. Soc. 1995, 117, 3885.
(25) Cromer, D. T.; Waber, J . T. Acta Crystallogr. 1965, 18, 104.
(26) International Tables of X-Ray Crystallography; Kynoch: Bir-
mingham, England, 1974; Vol. 4.
(27) Walker, N.; Stuart, D. Acta Crystallogr. 1983, A39, 158.
(28) Sheldrick, G. M. SHELXL93 Program for Structure Refinement;
University of Go¨ttingen, Go¨ttingen, Germany, 1994.
(29) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.;
Burla, M. C.; Polidori, G.; Camalli, M. J . Appl. Crystallogr. 1994, 27,
435.

Mimicking the HDS Activity of Promoted W Catalysts
Organometallics, Vol. 16, No. 26, 1997 5701
Sch em e 2
Rh(CO)2][(CO)5W(o-S(C6H4)CHdCH2)] (3), in which the
Rh and W centers belong to two distinct mononuclear
complex species (Scheme 3).
Compound 3 was successfully separated from 1 and
W(CO)6 by column chromatography and authenticated
by NMR and IR spectroscopic techniques as well as by
a sequence of reactions that unambiguously demon-
strated the presence of a 2-vinylthiophenolate ligand
exclusively bound to tungsten. Of crucial importance
was the reaction of isolated 3 with [NBu4]I in THF,
yielding the neutral complex (triphos)Rh(CO)I21 and
[NBu4][(CO)5W(o-S(C6H4)CHdCH2)] (4), which were
separated from each other by chromatography. Finally,
a pure sample of 4, dissolved in THF, was treated with
MeI to give [W(CO)5I][NBu4]32 and free o-methylthio-
styrene.15b,22
pose even at reflux temperature. In line with the
structural formulation given in Scheme 2, the IR
The complex anion in 3 and 4 belongs to a family of
numerous group 6 metal thiolates with the general
formula [M(CO)5SR]-.30 Thus, a detailed description of
the chemical-physical properties of the anion [(CO)5W-
(o-S(C6H4)CHdCH2)]- is not warranted here. The only
unprecedented NMR features pertain to the η1-thiolate
spectrum of 2 (Nujol mull or CH2Cl2 solution) in the ν-
(2)
(C-O) region is characterized by four bands (A1 B1,
,
E, A1(1)), a pattern which is typical of W(CO)5L com-
plexes where L may be a unidentate thiolate ligand.30
The 31P{1H} NMR spectrum features an AMQX pattern
and is thus consistent with the magnetic inequivalence
of the three phosphorus atoms of triphos. On the basis
of NOESY studies on the related complex exo-[(triphos)-
Ir(η3(S,C,C)-S(C6H4)CHdC(H)Me)], obtained by inser-
tion of the [(triphos)IrH] fragment into 2-methylben-
zothiophene,31 the lowest field resonance (30.9 ppm) can
be assigned to the P nucleus trans to the sulfur atom.
The phosphorus chemical shifts do not significantly
diverge from those of the thiolate precursor 1, though
one may notice a small high-field shift of the resonances
pertaining to PM and PQ. This shielding effect may be
due to the decreased electron density at the metal center
in consequence of the electrophilic attack by the W(CO)5
group at the sulfur atom. A slight decrease of the metal
basicity in the heterobimetallic complex (with conse-
quent decrease of the metal π-back-bonding contribution
to the M-(CHdCH2) bond) may be inferred also from
1
ligand bound to tungsten, particularly the H and 13C-
{1H} NMR resonances of the hydrogen and carbon
nuclei of the vinyl moiety. The chemical shifts and
coupling constants are typical of such groups, however
(e.g. the proton spectrum of 4 shows an upfield reso-
nance at 7.13 due to H3, which is trans (J (H3H2trans) )
17.5 Hz) and cis (J (H3H2cis) ) 11.0 Hz) to the gem
hydrogens of C2 at 5.69 and 5.30 ppm, respectively).
Hyd r ogen a tion Rea ction s of 2. In Situ HP NMR
Stu d y. When a THF-d8 solution of 2 was pressurized
with 30 atm of H2 in a sapphire HPNMR tube, 31P{1H}
NMR spectroscopy showed, even at room temperature,
the slow transformation of the starting W/Rh complex
into a different dimeric species with the formula (tri-
phos)RhH(µ-H)[µ-o-S(C6H4)C2H5]W(CO)4 (5). Since the
conversion is very slow at room temperature (20% in 3
days), the probe head of the spectrometer was heated
to 70 °C. At this temperature, all 2 disappeared in 2 h,
yielding 5 (Scheme 4).
As shown by a variable-temperature 31P{1H} NMR
study, the conversion of 2 to 5 proceeds with no detect-
able intermediates. Even at 80 °C, however, 5 started
disappearing. Formed in its place was the rhodium
carbonyl complex (triphos)RhH(CO) (6). Increasing the
temperature to 120 °C led to complete transformation
of 5 into 6 in ca. 1 h, but the reaction was less selective,
yielding a minor amount of the rhodium byproduct
(triphos)Rh(CO)[o-S(C6H4)C2H5] (7).6d A sequence of
31P{1H} NMR spectra illustrating these experiments is
reported in Figure 1. After the HPNMR tube was de-
pressurized, the contents were analyzed by GC/MS,
showing the almost quantitative formation of ethylben-
zene (EB) and traces of W(CO)6. A black, insoluble
material analyzed as WSx (xav ) 1.5 over five prepara-
tive reactions; see below) was also formed.
1
the H and 13C{1H} NMR spectra, in which the H2 and
H2′ protons and the C2 atom are shifted to lower field
relative to the precursor 1. In the latter compound both
theoretical and reactivity studies agree that C2 is a
strong nucleophilic site in competition with the sulfur
atom for binding electrophiles.15b Accordingly, the C2
carbon and its hydrogen atoms would be particularly
sensitive to any alteration of the metal basicity, as is
the case when a thiolate sulfur is replaced by a thioether
sulfur. In conclusion, the NMR studies suggest that the
bonding structure of the Rh-(CHdCH2) moiety acquires
a less pronounced metallacyclopropane character in 2
as compared to 1.
Ca r bon yla tion Rea ction of 2. Besides its affects
on the NMR parameters, the attack of W(CO)5 at the
sulfur atom of the C-S-inserted BT in 2 has remarkable
chemical implications due to the weakening of the
bonding interaction between the metal center and the
2-vinylthiophenolate ligand. In fact, unlike 1, the het-
erobimetallic complex 2 reacts with a relatively high
pressure of CO (30 atm) even at room temperature,
yielding a 3:3:1 mixture of 1, W(CO)6, and [(triphos)-
Isola tion a n d Ch a r a cter iza tion of th e Hyd r oge-
n a tion In ter m ed ia te 5. Having observed that the
intermediate species 5 is stable in THF solution below
80 °C, a preparative-scale hydrogenation reaction of 2
was carried out using 30 atm of H2 and a constant
temperature of 70 °C. Workup of the reaction mixture
(30) (a) Sa´nchez-Pela´ez, A. E.; Perpin˜an, M. F.; Gutierrez-Puebla,
E.; Monge, A.; Ruiz-Valero, C. J . Organomet. Chem. 1990, 384, 79. (b)
Darensbourg, D. J .; Sanchez, K. M.; Reibenspies, J . Inorg. Chem. 1988,
27, 3636.
(31) Bianchini, C.; Casares, J . A.; Masi, D.; Meli, A.; Pohl, W.; Vizza,
F. J . Organomet. Chem. 1997, 541, 143.
(32) Palitzsch, W.; Bo¨hme, U.; Roewer, C. Chem. Commun. 1997,
803.

5702 Organometallics, Vol. 16, No. 26, 1997
Bianchini et al.
Sch em e 3
Sch em e 4
gave brick red crystals of 5, which were recrystallized
from dichloromethane/ethanol to give 5‚CH2Cl2. An
X-ray analysis was carried out on a single crystal (Fig-
ure 2). Crystallographic data and selected bond dis-
tances and angles for 5‚CH2Cl2 are listed in Tables 1
and 2, respectively.
hydride Hb, the terminal hydride Ht was not located,
although its position trans to P1 may be inferred from
the long P1-Rh distance (2.414(4) vs 2.293(4) Å (aver-
age) of Rh-P2,3) that correlates with the great trans
influence of the hydride ligand. The presence of a
terminal hydride ligand bound to rhodium and the
overall solid-state structure of 5 are totally consistent
with the IR (ν(Rh-H) 1996 cm-1) and NMR data for
the complex in solution. At room temperature, the 31P-
{1H} NMR spectrum consists of a canonical AMQX
pattern, the PM and PQ components of which signifi-
cantly broaden, losing their multiplicity at higher tem-
perature (Figure 3). A slow exchange of the terminal
and bridging hydride ligands (Hb, δ -9.88, J (HbRh) )
19.02 Hz, J (HbHt) ) 5.45 Hz, J (HbPM) ) 100.6 Hz,
J (HbW) ) 34.2 Hz; Ht, δ -10.07, J (HtRh) ) 6.76 Hz,
J (HtPQ) ) 181.5 Hz) may well account for the slight
fluxionality exhibited by their trans phosphorus atoms.
As is evident from the Zortep drawing shown in
Figure 2, the two formal d6 metal centers (W0 and RhIII
)
appear as the central atoms of two octahedra sharing
two basal sites, namely the bridging 2-ethylthiopheno-
late and hydride ligands. The hydride ligand with its
two electrons cannot saturate the metals, however. In
terms of the effective atomic number (EAN) rule, the
total electron count of 34 implies the presence of a
metal-metal bond, which in fact is found experimen-
tally (e.g. the Rh-W distance is 3.06 Å, while the
W-Hb-Rh angle is 134°). The electronic situation
closely resembles that of the [M(CO)5(µ-SH)]- anions (M
) Cr, W), comprising two isoelectronic terminal ML5
fragments and a bent M-Hb-M bridge where the three-
center-two-electron bond involves the formally empty
σ-hybrid orbitals of the two ML5 fragments and the filled
H1s orbital.33 The C-C bond lengths in the organic
substituent of the thiophenolate ligand confirm that
hydrogenation of the vinyl double bond has occurred
with formation of an ethyl group. Unlike the bridging
Desu lfu r iza t ion of t h e 2-E t h ylt h iop h en ola t e
Liga n d . In Situ a n d Ba tch Rea ction s. Heating to
120 °C for 3 h a HPNMR tube containing a THF-d8
solution of pure 5 under 30 atm of H2 resulted in the
transformation of the heterobimetallic dimer into the
carbonyl hydride complex 6, EB, and “WSx”. No inter-
1
mediate species was seen by variable-temperature H
and 31P NMR spectroscopy along this transformation,
which occurred even above 70 °C. In the absence of
added H2, the desulfurization of the bridging thiolate
to give 6, EB, and WSx still occurred by thermolysis of
(33) (a) Petersen, J . L.; Brown, R. K.; Williams, J . M. Inorg. Chem.
1981, 20, 158. (b) Roziere, J .; Williams, J . M.; Stewart, R. P., J r.;
Petersen, J . L.; Dahl, L. F. J . Am. Chem. Soc. 1977, 99, 4497.

Mimicking the HDS Activity of Promoted W Catalysts
Organometallics, Vol. 16, No. 26, 1997 5703
F igu r e 1. 31P{1H} HPNMR study (sapphire tube, THF-d8, 81.01 MHz) of the reaction of 2 with hydrogen (30 atm): (a) at
room temperature; (b) after 2 h at 70 °C; (c) after the probe was heated to 100 °C for 1 h; (d) after further heating to 120
°C for 1 h; (e) after the NMR probe was cooled to room temperature.
Ta ble 2. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for 5‚CH2Cl2
Distances
W(1)-C(8)
W(1)-C(9)
W(1)-C(7)
W(1)-C(6)
W(1)-S(1)
W(1)-Rh(1)
1.88(2)
1.95(2)
1.94(2)
1.96(2)
2.493(4)
3.0554(14)
W(1)-Hb
1.75(16)
2.282(4)
2.314(4)
2.407(4)
2.414(4)
1.57(16)
Rh(1)-P(3)
Rh(1)-P(2)
Rh(1)-S(1)
Rh(1)-P(1)
Rh(1)-Hb
Angles
C(8)-W(1)-C(9)
C(8)-W(1)-C(7)
C(9)-W(1)-C(7)
C(8)-W(1)-C(6)
C(9)-W(1)-C(6)
C(7)-W(1)-C(6)
C(8)-W(1)-S(1)
C(9)-W(1)-S(1)
C(7)-W(1)-S(1)
C(6)-W(1)-S(1)
86.5(7)
91.4(7)
88.4(7)
84.4(7)
170.8(7)
90.3(7)
107.5(6)
85.7(5)
159.8(5)
98.5(5)
C(8)-W(1)-Hb
172(5)
P(3)-Rh(1)-P(2)
P(3)-Rh(1)-S(1)
P(2)-Rh(1)-S(1)
P(3)-Rh(1)-P(1)
P(2)-Rh(1)-P(1)
S(1)-Rh(1)-P(1)
89.11(14)
169.89(14)
97.50(14)
88.63(14)
92.34(14)
98.68(13)
P(3)-Rh(1)-W(1) 118.65(11)
P(2)-Rh(1)-W(1) 146.62(11)
F igu r e 2. Zortep drawing of the binuclear compound
[(triphos)RhH(µ-H){µ-o-S(C6H4)C2H5}W(CO)4] in 5‚CH2Cl2.
Phenyl rings of triphos are omitted for clarity. The atom
Ht could not be located by normal X-ray procedures, and it
is shown at the vacant octahedral site about the rhodium
atom.
S(1)-Rh(1)-W(1)
52.70(9)
C(8)-W(1)-Rh(1) 156.4(5)
C(9)-W(1)-Rh(1) 97.8(5)
P(1)-Rh(1)-W(1) 105.59(10)
P(2)-Rh(1)-Hb
Rh(1)-S(1)-W(1)
W(1)-Hb-Rh(1)
170(5)
77.14(11)
134(3)
C(7)-W(1)-Rh(1) 111.9(5)
C(6)-W(1)-Rh(1)
S(1)-W(1)-Rh(1)
91.1(5)
50.16(9)
5 in THF. The reaction, however, was less selective as
appreciable amounts of the carbonyl thiolate complex
7 and of other unidentified species were also formed.
Preparative-scale hydrogenations of 5, carried out in an
autoclave at 120 °C and with 30 atm of H2, gave
sufficient amounts of WSx for analytical purposes.
In Situ P r ep a r a tion of [W(CO)5(o-S(C6H4)C2H5)]-
a n d Its Rea ction w ith H2. A 10 mm sapphire HP-
NMR tube containing a THF-d8 solution of 5 was
pressurized with 30 atm of CO at room temperature
(Scheme 5). Even already at 40 °C, a reaction occurred
that converted all the heterobimetallic Rh/W complex
into two mononuclear species: the cation [(triphos)Rh-
(CO)2]+ and the anion [W(CO)5(o-S(C6H4)C2H5)]-. The
evolution of free H2 was also observed by 1H NMR
spectroscopy. After CO was replaced by H2 (30 atm),
no appreciable reaction occurred below 70 °C. At this
temperature, however, ETP began to form slowly with
concomitant degradation of [W(CO)5(o-S(C6H4)C2H5)]-,
while the dicarbonyl complex [(triphos)Rh(CO)2]+ was
apparently stable. After 1 h at 70 °C, the production of
free ETP was practically quantitative, as shown by GC/
MS. W(CO)6 was the only tungsten complex detected,
while no production of EB was observed.
Attem p ted HDS of BT by Molybd en u m or Tu n g-
sten Mon on u clea r Com p lexes. Various mononuclear
complexes containing metals belonging to the class of
HDS components have been employed as precursors for
the hydrogenation (30 atm of H2) of BT in THF. All of
the compounds investigated, (triphos)WH6 (9),17 (tri-
phos)Mo(η6-C6H6) (10),23 [(triphos)MoH(η6-C6H6)]BF4

5704 Organometallics, Vol. 16, No. 26, 1997
Bianchini et al.
F igu r e 3. Variable-temperature (°C) 31P{1H} NMR spectra (sapphire tube, THF-d8, 81.01 MHz) of 5.
Sch em e 5
Sch em e 6. P r op osed Rea ction P a th w a ys for th e
HDS Rea ction of Ben zo[b]th iop h en e
considered as the initial steps of alternative pathways
leading to EB and H2S (Scheme 6).
(11),23 [Cp*WS3](PPh4) (12),24 W(CO)6, and W(CO)5THF,
did not catalyze any transformation of BT even under
forcing conditions (160-200 °C, 30 atm of H2).
In spite of extensive research on promoted Mo or W
catalysts, the chemical role of the promoter metal atom
in the HDS of BT is practically unknown. In analogy
to the model proposed by Topsøe3a and Startsev4a for
thiophene HDS on Co-Mo-S catalysts, it is possible
that BT is preferentially adsorbed on coordination
vacancies associated with the promoter atoms, while
hydrogen is activated on MoS2 (WS2) edge sites. In this
eventuality, the following transfer of hydrogen atoms
(most likely from Mo(W)-SH groups) to the adsorbed
BT would result in either hydrogenolysis or hydrogena-
tion reactions, depending on the type of activation
undergone by the substrate, i.e. C-S insertion (hydro-
genolysis to thiol) or coordination in the intact form
(hydrogenation to DHBT).
Discu ssion
From a mechanistic viewpoint, the heterogeneous
HDS of thiophenes is a very complicated process that
depends critically on many factors, among which im-
portant roles are played by the structure and the
molecular size of the substrate and by the structure and
the composition of the catalyst.3,4 For BT, either hy-
drogenolysis or hydrogenation reactions have been
The results presented in this work show that the C-S
opening of BT readily occurs at a promoter metal (Rh)
but the desulfurization of the hydrogenolysis thiolate
product requires the aid of a component metal (W) to
take place under very mild reaction conditions. It is
just the component metal that ultimately traps the ex-

Mimicking the HDS Activity of Promoted W Catalysts
Organometallics, Vol. 16, No. 26, 1997 5705
truded sulfur atom. The desulfurization of the thiolate
to EB proceeds through a Rh/W dimeric intermediate
in which the metal centers are held together by bridging
2-ethylthiophenolate and hydride ligands and occurs
thermally (>70 °C) with no need of external hydrogen
(an excess of the latter reagent helps to minimize the
side reaction leading to the formation of the thiolate
carbonyl complex 7, however).
It is generally observed that both the C-S opening
of thiophenes by metal promoters5 and the desulfuriza-
tion of thiols by metal components14,34 are low-energy
processes in the homogeneous phase (as shown in this
paper, both reactions take place even at 70 °C). In
contrast, the removal of sulfur from complexes contain-
ing metal-sulfur multiple bonds by reaction with H2 is
a very difficult process, uniquely observed for Cp*2Ti-
(S)py (py ) pyridine).35 For example, Cp*WCl4 de-
sulfurizes at low temperature the dithiolate SCH2CH2S2-
to give [Cp*WS3]-,24 which is stable in THF at 160 °C
under 30 atm of H2. Only by addition of a strong acid
does the hydrogenation of [Cp*WS3]- cause the evolu-
tion of H2S. Once again, results of homogeneous model-
ing studies have implications for the heterogeneous
HDS of thiophenes. It has been proposed, in fact, that
the rate-determining step on real catalysts is the
removal of surface sulfur (as H2S) and not substrate
adsorption, C-S insertion, or C-S cleavage.3,4
The homogeneous HDS of BT has recently been
observed by other groups.10a,11 Among the known exam-
ples, that described by Vicic and J ones shows differences
but also interesting analogies to ours (Scheme 1). The
reaction assisted by the homometallic dimer [Cp*IrHCl]2
is apparently intermolecular, as it requires the addition
of H2 and also needs a high temperature (150 °C) to take
place. Moreover, the fate of the sulfur was not deter-
mined. Nevertheless, like 5, the intermediate that pre-
cedes the S-C(aryl) bond cleavage contains a µ-thiolato
bridge. This leads us to believe that the multipoint
coordination of the RS- groups (particularly when they
simultaneously bridge promoter and component metals)
may be very important in bringing about the C-S bond
scission. Indeed, as recently demonstrated by Curtis
and Druker,14 arene thiols (the products of the hydro-
genolysis of the benzothiophenes) are easily desulfurized
by the cluster Cp′2Mo2Co2S3(CO)4 with a mechanism in
which the C-S bond of µ3-SR groups is homolytically
cleaved. The bridging bonding mode of the thiolato
ligands was proposed to be the factor that drastically
reduces the C-S bond dissociation energy.
The increase in HDS activity of promoted heterog-
enous catalysts has also been correlated to the higher
mobility of the surface S atoms.14 Curtis and Druker
have recently interpreted the fluxionality of certain
intermediates that precede the desulfurization of thiols
with Cp′2Mo2Co2S3(CO)4 in terms of a facile movement
of RS- groups over the face of the cluster. The concept
of latent vacancies was applied to define the Co atoms
from which the thiolate groups move to bind neighboring
Mo atoms. In this work, we have shown that 2-vinyl-
thiophenolate and 2-ethylthiophenolate groups can eas-
ily move from the promoter (responsible for the hydro-
genolysis of BT) to the component (responsible for the
following desulfurization step) (Schemes 3 and 5).
When, however, the thiolate is exclusively bound to W
as in 8, its desulfurization by action of H2 does not occur
in the conditions employed for the desulfurization of the
heterobimetallic complex 5. The concomitant presence
of promoter and component metals thus seems impor-
tant in order to have the HDS of BT through the
mechanism that begins with the hydrogenolysis reaction
(Scheme 6, path a ). On the other hand, the key role of
the metal promoter in the hydrogenolysis of thiophenes
is supported by several homogeneous modeling studies,
which clearly show that C-S bond scission is almost
exclusively brought about by promoter metal com-
plexes.5 Consistently, we have found that various W
or Mo complexes do not catalyze any transformation of
BT under the reaction conditions that are currently
employed with hydrogenation or hydrogenolysis cata-
lysts containing promoter metal complexes.5b
Con clu sion s
The aim of this work was to tackle some questions
regarding the HDS mechanism of thiophenes, particu-
larly the specific role played by metals belonging to the
classes of the promoters (Rh) and of the components (W)
in the degradation of benzo[b]thiophene to ethylbenzene
and “S” or H2S. It has been shown that the hydro-
genolysis of benzo[b]thiophene to 2-vinylthiophenol or
2-ethylthiophenol is a facile process for the promoter,
but the desulfurization step needs the assistance of a
component metal to take place. It has also been found
that the desulfurization of the 2-ethylthiophenolate
ligand is favored when it bridges the Rh and W centers.
Given the complexity of the heterogeneous HDS
reaction and the many possible mechanisms that may
be operative even on the surfarce of the same catalyst,
our results must be considered with extreme caution.
Many of them, however, show a surprising consistency
with related reactions occurring over the surface of Co/
Mo/S phase catalysts.
Ack n ow led gm en t. Thanks are due to Professors M.
T. Ashby and K. Tatsumi for loans of (triphos)Mo(η6-
C6H6) and [Cp*WS3](PPh4), respectively. The contribu-
tions of the CNR (Italy) and CONICIT (Venezuela)
research programs are also acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Tables of bond
distances and angles (Table S1), final coordinates with equiva-
lent isotropic thermal parameters (Table S2), anisotropic
thermal parameters (Table S3), and atomic coordinates of the
hydrogen atoms (Table S4) for 5‚CH2Cl2 (8 pages). Ordering
information is given on any current masthead page.
OM970711E
(34) (a) Burrow, T. E.; Hills, A.; Hughes, D. L.; Lane, J . D.;
Lazarowych, N. J .; Maguire, M. J .; Morris, R. H.; Richards, R. L. J .
Chem. Soc., Chem. Commun. 1990, 1757. (b) J ang, S.; Atagi, L. M.;
Mayer, J . M. J . Am. Chem. Soc. 1990, 112, 6413. (c) Listemann, M.
L.; Schrock, R. R.; Dewan, J . C.; Kolodziej, R. M. Inorg. Chem. 1988,
27, 264. (d) Adams, R. D.; Yang, L.-W. J . Am. Chem. Soc. 1982, 104,
4115.
(35) Sweeney, Z. K.; Polse, J . L.; Andersen, R. A.; Bergman, R. G.;
Kubinec, M. G. J . Am. Chem. Soc. 1997, 119, 4543.