Steric effects in the binding of hindered dibenzothiophene ligands in [Cp′Ru(CO)2(η1(S)-DBTh)]+ complexes

Article abstract of DOI:10.1021/om048973l

Structural studies of several complexes of the type [Cp′Ru(CO) ₂(η¹(S)-DBTh)]⁺, where Cp′ = Cp or Cp* and DBTh = DBT, 4-MeDBT, 4,6-Me₂DBT, or 2,8-Me ₂DBT, show that only in [Cp*Ru(CO)₂(η

Full text of DOI:10.1021/om048973l

2168
Organometallics 2005, 24, 2168-2176
Steric Effects in the Binding of Hindered
Dibenzothiophene Ligands in [Cp′Ru(CO)2(η1(S)-DBTh)]+
Complexes
Paul A. Vecchi, Arkady Ellern,† and Robert J. Angelici*
Ames Laboratory and Department of Chemistry, Iowa State University, Ames, Iowa 50011
Received December 29, 2004
Structural studies of several complexes of the type [Cp′Ru(CO)2(η1(S)-DBTh)]+, where Cp′
) Cp or Cp* and DBTh ) DBT, 4-MeDBT, 4,6-Me2DBT, or 2,8-Me2DBT, show that only in
[Cp*Ru(CO)2(η1(S)-4,6-Me2DBT)]+ is there evidence for steric crowding between the Cp′ and
DBTh ligands. However, relative equilibrium constants (K′) for the binding of the DBTh
ligands in both the Cp and Cp* complexes show evidence of steric effects in those that contain
4-MeDBT and 4,6-Me2DBT as the K′ values increase in the order 4,6-Me2DBT < 4-MeDBT
< DBT < 2,8-Me2DBT. Kinetics studies of the substitution of the DBTh in [CpRu(CO)2(η1-
(S)-DBTh)]+ by phosphorus donor ligands establish the following order of DBTh lability,
4,6-Me2DBT > 4-MeDBT > DBT > 2,8-Me2DBT, which is consistent with the trend in K′
values. The most labile DBTh ligand in both series of complexes is 4,6-Me2DBT in [Cp*Ru-
(CO)2(η1(S)-4,6-Me2DBT)]+, where both steric crowding and electron donation by the Cp*
ligand accelerate the rate of 4,6-Me2DBT dissociation.
Introduction
The removal of organosulfur compounds from petro-
leum feedstocks is important for the reduction of
atmospheric pollution caused by sulfur oxides.1 Sulfur
compounds in these fuels also poison catalytic convert-
ers, impairing the vehicle emission control systems
necessary to reduce levels of nitrogen oxides released
into the atmosphere during fuel combustion. In ad-
dressing these issues, the United States government
through the Environmental Protection Agency (EPA)
has gradually established lower sulfur limits in trans-
portation fuels.2
Sulfur is currently removed from gasoline and diesel
fuels using a catalytic process known as hydrodesulfu-
rization (HDS). The most difficult of the sulfur-contain-
ing compounds to be removed using HDS are the
hindered dibenzothiophene derivatives (DBTh) that
contain alkyl groups near the sulfur in the 4- and
6-positions (Figure 1).1a,3 These dibenzothiophene com-
pounds are found in diesel fuel and must be removed
in order to lower its sulfur content. Although it is
unclear exactly how dibenzothiophenes interact with the
catalyst surface during HDS, it has been proposed that
an initial step is binding of the substrate through the
Figure 1. Numbering schemes for the aromatic thiophene
(T), benzothiophene (BT), and dibenzothiophene (DBT)
compounds found in petroleum feedstocks.
sulfur atom to active metal sites.4 The slow rates for
the HDS of 4-methyldibenzothiophene (4-MeDBT) and
4,6-dimethyldibenzothiophene (4,6-Me2DBT) have been
attributed to steric hindrance by methyl groups in the
4,6-positions that interfere with their binding to such
sites. Alkyl groups in these positions have also been
proposed to sterically hinder C-S bond cleavage during
HDS.5 Improvements in the HDS process have not
adequately addressed the difficulties of desulfurizing
hindered dibenzothiophenes.
* To whom correspondence should be addressed. E-mail: angelici@
iastate.edu.
† Molecular Structure Laboratory, Iowa State University, Ames, IA
50011.
(4) (a) Topsøe, H.; Clausen, B. S.; Massoth, F. E. Hydrotreating
Catalysis. Science and Technology; Springer-Verlag: Berlin, 1996. (b)
Chianelli, R. R.; Pecoraro, T. A.; Halbert, T. R.; Pan, W. H.; Stiefel, E.
I. J. Catal. 1984, 86, 226. (c) Pecoraro, T. A.; Chianelli, R. R. J. Catal.
1981, 67, 430.
(5) (a) Mijoin, J.; Perot, G.; Bataille, F.; Lemberton, J.-L.; Breysse,
M.; Kasztelan, S. Catal. Lett. 2001, 71, 139. (b) Bataille, F.; Lemberton,
J.-L.; Michaud, P.; Perot, G.; Vrinat, M.; Lemaire, M.; Schulz, E.;
Breysse, M.; Kasztelan, S. J. Catal. 2000, 191, 409. (c) Meille, V.;
Schulz, E.; Lemaire, M.; Vrinat, M. J. Catal. 1997, 170, 29. (d) Kabe,
T.; Ishihara, A.; Zhang, Q. Appl. Catal., A 1993, 97, L1.
(1) (a) Whitehurst, D. D.; Isoda, T.; Mochida, I. Adv. Catal. 1998,
42, 345. (b) Angelici, R. J. In Encyclopedia of Inorganic Chemistry;
King, R. B., Ed.; Wiley & Sons: New York, 1994; p 1433. (c) Kasztelan,
S.; Des Courieres, T.; Breysse, M. Catal. Today 1991, 10, 433. (d) Ho,
T. C. Catal. Rev. 1988, 30, 117.
(2) (a) Connell, C. Fed. Regist. 2000, 65, 6697. (b) Borushko, M. Fed.
Regist. 2001, 66, 5001.
(3) (a) Ho, T. C. Catal. Today 2004, 98, 3. (b) Song, C.; Ma, X. Appl.
Catal., B 2003, 41, 207. (c) Gates, B. C.; Topsøe, H. Polyhedron 1997,
16, 3213.
10.1021/om048973l CCC: $30.25 © 2005 American Chemical Society
Publication on Web 03/31/2005

Binding of Hindered Dibenzothiophene Ligands
Organometallics, Vol. 24, No. 9, 2005 2169
25.0 °C
Recently, the focus has shifted away from improving
current HDS processes toward developing new tech-
nologies for desulfurization.3,6 One promising new method
is the use of adsorbents or solid phase extractants
(SPEs) to selectively bind and remove sulfur compounds
such as the hindered dibenzothiophenes.3,6-8 Adsorbent
technologies have the potential advantage of being
relatively simple to incorporate into the existing infra-
structure as either a pre- or post-HDS treatment. Such
treatments may also be designed to operate under
relatively mild conditions, unlike HDS, which currently
involves the use of moderately high temperatures (350-
500 °C) and pressures (50-150 psi) of hydrogen. An-
other emerging technology for desulfurization of hydro-
carbon fuels is catalytic oxidation of hindered dibenzothio-
phenes to the sulfones or sulfoxides, which can then be
removed by extraction or adsorption.9 In both of these
new desulfurization approaches, an important step
involves binding of the dibenzothiophene to either the
adsorbent or catalyst.
Several experimental and theoretical studies on the
binding of thiophenes, benzothiophenes, and dibenzo-
thiophenes to transition metal complexes have been
reported.10 However, there are comparatively few ex-
perimental studies of hindered dibenzothiophenes that
are η1(S)-coordinated in metal complexes,11 presumably
a result of the instability of such complexes. An under-
standing of the relative bond strengths and kinetic
labilities of hindered dibenzothiophenes in metal com-
plexes should be useful for developing new approaches
to petroleum feedstock desulfurization. We recently
reported the synthesis and structural characterization
of the first transition metal complexes of η1(S)-bound
4-MeDBT and 4,6-Me2DBT, [Cp*Ru(CO)2(η1(S)-4-Me-
DBT)]+ and [Cp*Ru(CO)2(η1(S)-4,6-Me2DBT)]+, where
Cp* ) η5-C5Me5.11 Equilibrium studies of the displace-
ment of 4,6-Me2DBT in [Cp*Ru(CO)2(η1(S)-4,6-Me2-
DBT)]+ by the dibenzothiophenes (DBTh) DBT, 4-Me-
DBT, and 2,8-Me2DBT (eq 1) showed that their binding
abilities increase in the following order: 4,6-Me2DBT
(1.00) < 4-MeDBT (20.2(1)) < DBT (62.7(6)) < 2,8-Me2-
DBT (223(3)). In this report, we continue our investiga-
tions of sulfur-bound dibenzothiophenes, with special
attention directed toward understanding the role of
steric and electronic effects in both the [CpRu(CO)2(η1-
[Cp*Ru(CO)2(η1(S)-4,6-Me2DBT)]+ + DBTh y
z
CD2Cl2
[Cp*Ru(CO)2(η1(S)-DBTh)]+ + 4,6-Me2DBT (1)
(S)-DBTh)]+ and [Cp*Ru(CO)2(η1(S)-DBTh)]+ series of
complexes (where Cp ) η5-C5H5, and DBTh ) DBT,
4-MeDBT, 4,6-Me2DBT, and 2,8-Me2DBT). Equilibrium
binding constants and rates of DBTh substitution for
these complexes provide insight into factors that influ-
ence the thermodynamic and kinetic binding of 4- and
4,6-methyl-substituted dibenzothiophenes to {Cp′Ru-
(CO)2}+, where Cp′ ) Cp and Cp*.
Experimental Section
General Considerations. All reactions were performed
under an atmosphere of dry argon using standard Schlenk
techniques. Methylene chloride (CH2Cl2), diethyl ether (Et2O),
and hexanes were purified on alumina using a Solv-Tek
solvent purification system, similar to that described by
Grubbs and co-workers.12 Nitromethane (CH3NO2, 96+%) was
purchased from Aldrich and subjected to three freeze-pump-
thaw cycles before use. Acetone was stirred with calcium
chloride overnight, distilled, subjected to three freeze-pump-
thaw cycles, and stored under argon until use. Methylene
chloride-d2 (CD2Cl2) was refluxed overnight with calcium
hydride, distilled, subjected to three freeze-pump-thaw
cycles, and stored under argon until use. Nitromethane-d3
(CD3NO2) was purchased from Aldrich, subjected to three
freeze-pump-thaw cycles, and stored under argon before use.
Solid DBT and 4-MeDBT were purchased from Aldrich and
sublimed prior to use. Solid 4,6-Me2DBT and 2,8-Me2DBT were
purchased from Acros and TCI, respectively, and used without
further purification. Compounds AgBF4 (99.99+%), P(OPh)3,
PPh3, and PPh2Me were purchased from Aldrich and used
without further purification. Complexes CpRu(CO)2Cl,13 [CpRu-
(CO)2(η1-(S)-DBT)]BF4 (1),14 and [Cp*Ru(CO)2(η1-(S)-DBTh)]+
(DBTh ) DBT (5), 4-MeDBT (6), 4,6-Me2DBT (7), 2,8-Me2DBT
(8))11 were prepared as described previously.
Filtrations were performed with Celite on filter paper.
Solution NMR spectra were recorded on a Bruker DRX-400
spectrometer using either CD2Cl2 (δ ) 5.32 (1H), 54.0 (13C)) or
CD3NO2 (δ ) 4.33 (1H), 62.8 (13C)) as the solvent, internal lock,
and reference. Solution infrared spectra of the compounds in
CH2Cl2 were recorded on a Nicolet-560 spectrometer using
NaCl cells with 0.1 mm spacers. Elemental analyses were
performed on a Perkin-Elmer 2400 series II CHNS/O analyzer.
General Procedure for Preparation of the [CpRu-
(CO)2(η1(S)-DBTh)]BF4 Complexes (2-4). To a solution of
CpRu(CO)2Cl (75 mg, 0.291 mmol) and 0.320 mmol of DBTh
(DBTh ) 4-MeDBT, 4,6-Me2DBT, 2,8-Me2DBT) in 10 mL of
CH2Cl2 was added solid AgBF4 (58.4 mg, 0.300 mmol), and the
solution was stirred at room temperature under argon for 30
min. A solid precipitate formed and the yellow solution color
gradually lightened during the reaction. The reaction solution
was then filtered and transferred by cannula into a flask
containing 40 mL of diethyl ether in an ice water bath, which
resulted in precipitation of the product. The remaining solid
in the reaction flask was washed with a 1 mL portion of CH3-
NO2, and the solution was also transferred by cannula into
the diethyl ether to ensure complete transfer of the product.
The light yellow solid products were isolated by filtration and
washed with three 5 mL portions of diethyl ether to remove
excess DBTh. Isolated yields were typically 70-85%. Due to
(6) Song, C. Catal. Today 2003, 86, 211.
(7) (a) Hernandez-Maldonado, A. J.; Yang, R. T. J. Am. Chem. Soc.
2004, 126, 992. (b) Yang, F. H.; Hernandez-Maldonado, A. J.; Yang,
R. T. Sep. Sci. Technol. 2004, 39, 1717. (c) Haji, S.; Erkey, C. Ind. Eng.
Chem. Res. 2003, 42, 6933. (d) Hernandez-Maldonado, A. J.; Yang, R.
T. Ind. Eng. Chem. Res. 2003, 42, 3103. (e) Yang, R. T.; Hernandez-
Maldonado, A. J.; Yang, F. H. Science 2003, 301, 79. (f) Velu, S.; Ma,
X.; Song, C. Ind. Eng. Chem. Res. 2003, 42, 5293. (g) Ma, X.; Sun, L.;
Song, C. Catal. Today 2002, 77, 107. (h) Kobayashi, M.; Shirai, H.;
Nunokawa, M. Energy Fuels 2002, 16, 1378. (i) McKinley, S. G.;
Angelici, R. J. Energy Fuels 2003, 17, 1480.
(8) McKinley, S. G.; Vecchi, P. A.; Ellern, A.; Angelici, R. J. Dalton
Trans. 2004, 5, 788.
(9) (a) Murata, S.; Murata, K.; Kidena, K.; Nomura, M. Energy Fuels
2004, 18, 116. (b) Wang, D.; Qian, E. W.; Amano, H.; Okata, K.;
Ishihara, A.; Kabe, T. Appl. Catal., A 2003, 253, 91. (c) Wang, Y.; Lente,
G.; Espenson, J. H. Inorg. Chem. 2002, 41, 1272. (d) Te, M.; Fairbridge,
C.; Ring, Z. Appl. Catal., A 2001, 219, 267.
(10) For reviews see: (a) Sa´nchez-Delgado, R. A. Organometallic
Modeling of the Hydrodesulfurization and Hydrodenitrogenation Reac-
tions; Kluwer Academic Publishers: Dordrecht, Netherlands, 2002. (b)
Angelici, R. J. Organometallics 2001, 20, 1259. (c) Chen, Jiabi; Angelici,
R. J. Coord. Chem. Rev. 2000, 206-207, 63. (d) Harris, S. Polyhedron
1997, 16, 3219. (e) Angelici, R. J. Coord. Chem. Rev. 1990, 105, 61.
(11) Vecchi, P. A.; Ellern, A.; Angelici, R. J. J. Am. Chem. Soc. 2003,
125, 2064.
(12) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518.
(13) Eisenstadt, A.; Tannenbaum, R.; Efraty, A. J. Organomet.
Chem. 1981, 221, 317.
(14) Benson, J. W.; Angelici, R. J. Organometallics 1993, 12, 680.

2170 Organometallics, Vol. 24, No. 9, 2005
Table 1. Crystal Data and Structure Refinement for 2, 3, and 4
Vecchi et al.
[CpRu(CO)2(η1(S)-4-
MeDBT)]BF4 (2)
[CpRu(CO)2(η1(S)-4,6-
Me2DBTh)]BF4 (3)
[CpRu(CO)2(η1(S)-2,8-
Me2DBTh)]BF4 (4)
empirical formula
fw
C
20H15BF4O2RuS
C21H17BF4O2RuS
C21H17BF4O2RuS
507.26
521.29
521.29
temperature
wavelength
cryst syst
173(2) K
293(2) K
173(2) K
0.71073 Å
0.71073 Å
0.71073 Å
monoclinic
triclinic
monoclinic
space group
P2(1)/c
P1h
P2(1)/n
a ) 11.699(4) Å
b ) 12.520(5) Å
c ) 13.149(5) Å
R ) 90.0°
a ) 9.3637(8) Å
b ) 10.4141(9) Å
c ) 11.7381(9) Å
R ) 73.448(1)°
â ) 69.504(1)°
γ ) 88.382(2)°
1024.48(15) Å3
2
a ) 9.394(4) Å
b ) 11.003(5) Å
c ) 19.963(9) Å
R ) 90.000°
unit cell dimens
â ) 94.534(6)°
γ ) 90.0°
â ) 98.660(8)°
γ ) 90.000°
volume
Z
cryst color, habit
density (calcd)
abs coeff
F(000)
cryst size
θ range for data collection
1920.0(12) Å3
4
2039.7(16) Å3
4
yellow plate
yellow prism
1.690 Mg/m3
0.918 mm-1
yellow prism
1.698 Mg/m3
0.922 mm-1
1.755 Mg/m3
0.977 mm-1
1008
520
1040
0.20 × 0.20 × 0.10 mm3
1.75 to 25.00°
-13 e h e 13
-13 e k e 14
-15 e l e 14
13 529
0.30 × 0.28 × 0.21 mm3
1.94 to 26.37°
-11 e h e 11
-13 e k e 8
-14 e l e 14
6757
0.30 × 0.30 × 0.23 mm3
2.06 to 28.29°
-11 e h e 12
-14 e k e 14
-26 e l e 26
19 275
index ranges
no. of reflns collected
no. of ind reflns
completeness to θ )
abs corr
3341 [R(int) ) 0.0422]
25.00° ) 98.8%
semiempirical from equivalents
0.44 and 0.31
full-matrix least-squares on F2
3341/0/262
4112 [R(int) ) 0.0183]
26.37° ) 98.1%
empirical with SADABS
0.86 and 0.76
full-matrix least-squares on F2
4112/0/271
4706 [R(int) ) 0.0394]
28.29° ) 99.0%
semiempirical from equivalents
1.00 and 0.88
full-matrix least-squares on F2
4706/0/271
max. and min. transmn
refinement method
no. of data/restraints/
params
goodness-of-fit on F2
final Ra indices [I > 2σ(I)]
1.100
R1 ) 0.0677
1.013
R1 ) 0.0311
1.221
R1 ) 0.0787
wR2 ) 0.2007
R1 ) 0.0744
wR2 ) 0.0777
R1 ) 0.0380
wR2 ) 0.1936
R1 ) 0.0833
Ra indices (all data)
wR2 ) 0.2085
wR2 ) 0.0805
wR2 ) 0.1956
largest diff peak and hole
4.094 and -0.874 e Å-3
0.785 and -0.782 e Å-3
2.792 and -1.822 e Å-3
2
2
a R1 ) ∑||Fo| - |Fc||/∑|Fo| and wR2 ) {∑[w(Fo - Fc )2]/∑[w(Fo2)2]}1/2
.
the low solubility of the products in methylene chloride, 1H
NMR and 13C NMR spectra were acquired in CD3NO2. Spec-
troscopic data for compound 1 were reported previously.14
Characterization of Compounds 2-4. [CpRu(CO)2(η1-
(S)-4-MeDBT)][BF4] (2). 1H NMR (400 MHz, CD3NO2): δ
8.29-8.26 (m, 1H), 8.18 (d, J ) 7.6 Hz, 1H), 8.05-8.03 (m,
1H), 7.76-7.69 (m, 3H), 7.51 (d, J ) 7.6 Hz, 1H), 2.64 (s, 3H,
CH3), 4-MeDBT; 5.79 (s, 5H), Cp. 13C NMR (100.6 MHz, CD3-
NO2): δ 194.53 (CO); 142.98, 140.22, 138.79, 138.31, 136.44,
131.85, 131.69, 131.13, 131.06, 126.67, 124.80, 122.62, 21.42
(4-MeDBT); 91.37 (Cp). IR (CH2Cl2): ν(CO) (cm-1) 2076(s),
2033(s). Anal. Calcd for C20H15BF4O2RuS: C, 47.35; H, 2.98;
S, 6.32. Found: C, 47.16; H, 2.95; S, 6.28.
single crystal of 2 suitable for an X-ray diffraction study was
obtained by slow cooling of a saturated CH2Cl2 solution of the
complex at -25 °C for 1 week. A yellow single crystal of 3
suitable for an X-ray diffraction study was obtained by layering
a saturated CH2Cl2 solution of the complex with Et2O under
argon and storing at -25 °C for 1 week. Crystals of 4, also
yellow, were obtained by preparing an acetone solution of the
complex, filtering it through Celite, and allowing it to slowly
evaporate in air at room temperature.
The crystals were selected under ambient conditions, coated
in epoxy, and mounted on the end of a glass fiber. Crystal data
collection was performed on a Bruker CCD-1000 diffractometer
with Mo KR (λ ) 0.71073 Å) radiation and a collector-to-crystal
distance of 5.03 cm. Cell constants were determined from a
list of reflections found by an automated search routine. Data
were collected using the full sphere routine and were corrected
for Lorentz and polarization effects. The absorption corrections
were based on fitting a function to the empirical transmission
surface as sampled by multiple equivalent measurements
using SADABS software.15 Positions of the heavy atoms were
found by the Patterson method. The remaining atoms were
located in an alternating series of least-squares cycles and
difference Fourier maps. All non-hydrogen atoms were refined
in full-matrix anisotropic approximation. All hydrogen atoms
were placed in the structure factor calculation at idealized
positions and refined using a riding model. Complete data
collection and reduction information for each compound are
given in Table 1.
[CpRu(CO)2(η1(S)-4,6-Me2DBT)][BF4] (3). 1H NMR (400
MHz, CD3NO2): δ 8.12 (d, J ) 7.6 Hz, 2H), 7.69 (t, J ) 7.6
Hz, 2H), 7.50 (d, J ) 7.6 Hz, 2H), 2.62 (s, 6H, CH3), 4,6-Me2-
DBT; 6.02 (s, 5H), Cp. 13C NMR (100.6 MHz, CD3NO2):
δ
194.15 (CO); 141.23, 138.87, 135.94, 131.89, 131.49, 122.66,
21.39 (4,6-Me2DBT); 91.28 (Cp). IR (CH2Cl2): ν(CO) (cm-1
)
2076(s), 2033(s). Anal. Calcd for C21H17BF4O2RuS: C, 48.38;
H, 3.29; S, 6.15. Found: C, 47.98; H, 3.43; S, 6.04.
[CpRu(CO)2(η1(S)-2,8-Me2DBT)][BF4] (4). 1H NMR (400
MHz, CD3NO2): δ 8.08 (s, 2H), 7.85 (d, J ) 8.4 Hz, 2H), 7.52
(d, J ) 8.4 Hz, 2H), 2.56 (s, 6H, CH3), 2,8-Me2DBT; 5.82 (s,
5H), Cp. 13C NMR (100.6 MHz, CD3NO2): δ 194.63 (CO);
142.17, 139.02, 138.48, 131.74, 126.40, 124.91, 21.65 (2,8-Me2-
DBT); 91.16 (Cp). IR (CH2Cl2): ν(CO) (cm-1) 2076(s), 2032(s).
Anal. Calcd for C21H17BF4O2RuS: C, 48.38; H, 3.29; S, 6.15.
Found: C, 47.99; H, 3.46; S, 6.04.
Equilibrium Studies. Solutions for the equilibrium studies
(eq 2) were prepared by placing 0.020 mmol of [CpRu(CO)2-
X-ray Structural Determinations of [CpRu(CO)2(η1(S)-
4-MeDBT)][BF4] (2), [CpRu(CO)2(η1(S)-4,6-Me2DBT)][BF4]
(3), and [CpRu(CO)2(η1(S)-2,8-Me2DBT)][BF4] (4). A yellow
(15) Blessing, R. H. Acta Crystallogr. 1995 A51, 33.

Binding of Hindered Dibenzothiophene Ligands
Organometallics, Vol. 24, No. 9, 2005 2171
Table 2. Ru-S Bond Distances (Å) and Twist Angles (deg) for [CpRu(CO)2(η1(S)-DBTh)][BF4] Complexes
1-4 and for the [Cp*Ru(CO)2(η1(S)-DBTh)][BF4] Complexes 5-7
complex
Ru-S distance (Å)
twist angle (deg)
[CpRu(CO)2(η1(S)-DBT)]+ a (1)
2.398(2)
22.8
8.6
[CpRu(CO)2(η1(S)-4-MeDBT)]+ b (2)
[CpRu(CO)2(η1(S)-4,6-Me2DBT)]+ b (3)
[CpRu(CO)2(η1(S)-2,8-Me2DBT)]+ b (4)
[Cp*Ru(CO)2(η1(S)-DBT)]+ c (5)
2.377(2)
2.405(1)
8.9
2.400(1)
25.0
20.2
11.3, 12.3
0.4
2.394(1)
[Cp*Ru(CO)2(η1(S)-4-MeDBT)]+ c (6)
[Cp*Ru(CO)2(η1(S)-4,6-Me2DBT)]+ c (7)
2.399(1), 2.404(1)
2.419(1)
a Reference 8. b This study. c Reference 11.
Table 3. Equilibrium Constants, K, for Reactions
(eq 2) of [CpRu(CO)2(η1(S)-DBTh)]+ with DBTh′ at
25.0 °C in CD3NO2
by their NMR spectra. Relative concentrations of reactants and
kobs
[Cp′Ru(CO)2(η1(S)-DBTh)]+ + PR3 25.0 °C8
DBTh
DBTh′
K
[Cp′Ru(CO)2(PR3)]+ + DBTh (4)
DBT
4-MeDBT
0.358(8)
0.103(2)
3.77(3)
DBT
4,6-Me2DBT
2,8-Me2DBT
4,6-Me2DBT
Cp′ ) η5-C5H5; PR3 ) P(OPh)3, PPh3;
Cp′ ) η5-C5Me5; PR3 ) PPh3, PPh2Me
DBT
4-MeDBT
0.344(15)
(η1(S)-DBTh)][BF4] (DBTh ) DBT, 4-MeDBT, 4,6-Me2DBT,
2,8-Me2DBT) and an equimolar amount of a different diben-
zothiophene (DBTh′) in a 5 mm NMR tube. Approximately 0.7
products were determined by integration of the 1H NMR
signals for either the Cp or DBTh ligands. Rate constants, kobs
,
were obtained from the least-squares slopes of plots of ln(1 +
F) (where F ) [product]/[reactant]) versus time, and the
correlation coefficients of these plots were always greater than
0.995.
K
[CpRu(CO)2(η1(S)-DBTh)]+ + DBTh′ yCD NO , 25.0 °C
z
3
2
[CpRu(CO)2(η1(S)-DBTh′)]+ + DBTh (2)
Solutions of [Cp*Ru(CO)2(η1(S)-DBTh)][BF4] (DBTh ) DBT,
4-MeDBT, 4,6-Me2DBT, 2,8-Me2DBT) for the kinetic studies
were prepared in a slightly different manner. A 0.010 mmol
sample of the complex was added to an NMR tube with an
excess, weighed amount of PPh3. The tube was then evacuated
and flushed with argon. A 0.50 mL aliquot of CD2Cl2 was
added; the tube was capped with a septum and shaken to
dissolve the reactants. The tube was immediately placed into
the probe of a Bruker DRX-400 spectrometer thermostated at
25.0 ( 0.1 °C. The spectrometer was preprogrammed to
acquire 1H NMR spectra at specific time intervals using CD2-
Cl2 as the internal lock and reference (δ ) 5.32); the acquisition
time to take each spectrum was 60 s (16 scans at 3.744 s/scan).
Typically 8-12 spectra were collected over 2-3 half-lives for
each reaction. The products formed during the course of these
kinetic reactions were [Cp*Ru(CO)2(PPh3)]+ 17 and the free
dibenzothiophenes (eq 4), which were identified by their NMR
spectra. Relative concentrations of reactants and products were
determined by integration of methyl groups on the Cp* and/
or DBTh ligands. Rate constants, kobs, were obtained as
described above, and the correlation coefficients of these plots
were always greater than 0.995. Reactions with the liquid
phosphine PPh2Me were undertaken in the same manner,
except that the phosphine was injected by syringe immediately
after the deuterated solvent had been added. The ruthenium
product, [Cp*Ru(CO)2(PPh2Me)]+, gave an 1H NMR spectrum
mL of CD3NO2 was added to dissolve the reactants, and the
reaction solution was frozen in liquid nitrogen. After degassing,
the solution was thawed and flame-sealed under argon at room
temperature. The tube was then placed in a circulating bath
thermostated at 25.0 ( 0.1 °C, and the reaction progress was
monitored periodically by 1H NMR spectroscopy using CD3-
NO2 as the internal lock and reference (δ ) 4.33) with a 10 s
pulse delay between scans to ensure complete relaxation of
proton signals. Equilibrium constants (K) were calculated from
1
the H NMR spectra using eq 3,
(ICp′/5)2 [CpRu(CO)2(η1(S)-DBTh′)]+[DBTh]
(ICp/5)2 [CpRu(CO)2(η1(S)-DBTh)]+[DBTh′]
K )
)
(3)
where ICp′ and ICp are the Cp peak integrals of [CpRu(CO)2-
(η1(S)-DBTh′)]+ and [CpRu(CO)2(η1(S)-DBTh)]+, respectively.
All equilibria were established from both reaction directions
by separate experiments, and the results (Table 3) are the
average of the forward and reverse reactions with the average
deviation given in parentheses. All equilibria were established
within 10 days, and no further changes in the 1H NMR spectra
were observed when examined after 14 days. Equilibrium
constants for analogous reactions of the [Cp*Ru(CO)2(η1(S)-
DBTh)][BF4] complexes (DBTh ) DBT, 4-MeDBT, 4,6-Me2-
DBT, 2,8-Me2DBT) were reported previously.11
3
(for PPh2Me, δ ) 2.27, d, J(P,H) ) 9.3 Hz, 3H; for Cp*, δ )
Kinetic Studies. Solutions of [CpRu(CO)2(η1(S)-DBTh)]-
[BF4] (DBTh ) DBT, 4-MeDBT, 4,6-Me2DBT, 2,8-Me2DBT) for
kinetic studies were prepared according to the following
procedure: A 0.010 mmol sample of the complex was added
to an NMR tube, and the tube was evacuated and flushed with
argon. Deoxygenated CD3NO2 was added, followed by a
measured excess of P(OPh)3 so that the total volume of the
reaction mixture was 0.50 mL. The tube was capped with a
septum, shaken to dissolve the reactants, and placed in a
circulating bath thermostated at 25.0 ( 0.1 °C. At periodic
intervals, the reaction tube was removed and monitored by
1H NMR spectroscopy using CD3NO2 as the internal lock and
reference (δ ) 4.33). Typically 8-12 spectra were collected over
2-3 half-lives for each reaction. The products formed during
the course of the kinetic reactions were [CpRu(CO)2(P(OPh)3)]+ 16
and the free dibenzothiophenes (eq 4), which were identified
1.87, d, J(P,H) ) 2.0 Hz, 15H) that is similar to that of [Cp*Ru-
(CO)2(PPh3)]+. The amount of deuterated solvent in each
reaction was calibrated to ensure that the total volume of each
reaction solution was 0.50 mL.
Results and Discussion
Synthesis of the [CpRu(CO)2(η1(S)-DBTh)][BF4]
Complexes. The complexes [CpRu(CO)2(η1(S)-DBTh)]-
[BF4], where DBTh ) DBT (1), 4-MeDBT (2), 4,6-Me2-
DBT (3), and 2,8-Me2DBT (4), are prepared by Cl-
abstraction from CpRu(CO)2Cl with AgBF4 in the pres-
ence of excess dibenzothiophene ligand in CH2Cl2 (eq
5). Complexes 1-4 are isolated as air-stable, pale yellow
powders that are insoluble in diethyl ether and hexanes,

2172 Organometallics, Vol. 24, No. 9, 2005
Vecchi et al.
AgBF4, CH2Cl2
CpRu(CO)2Cl + DBTh RT, 30 min, -AgCl8
[CpRu(CO)2(η1(S)-DBTh)][BF4] (5)
minimally soluble in CH2Cl2 and acetone, and soluble
in CH3NO2. 1H NMR data show that the signals of the
DBTh ligands in these complexes are shifted slightly
downfield of the free ligand signals, as is observed in
other η1(S) complexes of dibenzothiophene.8,11,18 If the
DBTh ligand were bound η2 through a CdC bond or η6
to one of the arene rings, one would expect to see
significant upfield 1H and 13C NMR shifts for signals of
the coordinated groups, as are observed for η2-olefin19
and η6-arene complexes of ruthenium.20 The IR spectra
of these complexes in CH2Cl2 exhibit two ν(CO) signals
(2076, 2033 cm-1) that are essentially the same for all
four compounds and are on average 25 cm-1 higher than
those (ν(CO) ) 2056, 2004 cm-1 in CH2Cl2) of the
CpRu(CO)2Cl starting material. The 13C NMR spectrum
for [CpRu(CO)2(η1(S)-4-MeDBT)]+ (2) has only one
signal for the two diastereotopic carbonyl ligands at
room temperature, indicating that there is a dynamic
process at room temperature that makes the CO groups
equivalent on the 13C NMR time scale. This process is
likely to involve rapid inversion at the sulfur atom. Due
to the poor solubility of 2 in CD2Cl2, THF-d6, and (CD3)2-
CO, NMR studies of the complex were limited to
temperatures above the freezing point of CD3NO2 (244
K), where inversion remained rapid. In previous studies
of related complexes, the ∆Gq barriers to inversion were
determined for [CpRu(CO)2(η1(S)-BT)]+ (43 kJ/mol at
205 K; BT ) benzothiophene),14 [CpRu(CO)(PPh3)(η1-
(S)-Th)]+ (40 kJ/mol at 213 K; Th ) 2,5-Me2T and
Me4T),18a and [CpFe(CO)2(η1(S)-BT)]+ (39 kJ/mol at 190
K).18b
Structures of Compounds 2, 3, and 4. In all of the
structures (Figure 2) of the [CpRu(CO)2(η1(S)-DBTh)]-
[BF4] (2-4) complexes, the DBTh ligands are η1-
coordinated through a pyramidal sulfur atom, and the
DBTh ligand is oriented away from the Cp ligand, as
has been observed in other cationic complexes contain-
ing η1(S)-bound dibenzothiophene ligands: [CpRu(CO)2-
(η1(S)-DBT)[BF4],8 [Cp*Ru(CO)2(η1(S)-DBTh][BF4] (where
DBTh ) DBT, 4-MeDBT, 4,6-Me2DBT),11 and [CpFe-
(CO)2(η1(S)-DBT)][BF4].18b The Ru-S distances (Table
2) for complexes 1,8 2, 3, and 4 are 2.398(2), 2.377(2),
2.405(1), and 2.400(2) Å, respectively. All of these values
are very similar to each other except in complex 2, which
has a shorter Ru-S bond than the other complexes. The
orientation of the DBTh ligand around the Ru-S bond
Figure 2. Thermal ellipsoid drawings of the cations in
[CpRu(CO)2(η1(S)-4-MeDBT)][BF4] (2), [CpRu(CO)2(η1(S)-
4,6-Me2DBT)][BF4] (3), and [CpRu(CO)2(η1(S)-2,8-Me2DBT)]-
[BF4] (4). Ellipsoids are shown at the 50% probability level;
hydrogen atoms are omitted for clarity.
may be defined by the dihedral angle Cp(centroid)-Ru-
S-midpoint between C(10) and C(11). For a sym-
metrical orientation, this angle would be 180°, and the
deviation from 180° is defined as the twist angle. In
compounds 1, 2, 3, and 4, this angle is 22.8°, 8.6°, 8.9°,
and 25.0°, respectively (Table 2). These angles indicate
that there is some degree of rotational freedom around
the Ru-S bond in all of the complexes, even in the 4,6-
Me2DBT complex (3).
Thus, there is no structural evidence for steric crowd-
ing caused by the 4,6-methyl groups in the [CpRu(CO)2-
(η1(S)-DBTh)]+ complexes 1-4. This is in contrast to a
previous structural investigation of the series of [Cp*Ru-
(CO)2(η1(S)-DBTh)]+ complexes,11 where DBTh ) DBT,
4-MeDBT, and 4,6-Me2DBT, in which methyl groups in
both the 4- and 4,6-positions affect significantly the
structures of these complexes. In that study, the Ru-S
distance (Table 2) increased with the number of 4,6-
methyl groups as follows: DBT (2.394(1)) e 4-MeDBT
(2.399(1), 2.404(1))21 < 4,6-Me2DBT (2.419(1)). The
lengthening of the Ru-S bond, especially for the 4,6-
Me2DBT complex, was attributed to steric repulsions
between methyl groups on the DBTh ligand and Cp*
methyl groups. The twist angles for the [Cp*Ru(CO)2-
(η1(S)-DBTh)]+ complexes (Table 2) decrease with an
increasing number of 4,6-methyl groups: DBT (20.2°)
> 4-MeDBT (11.3°, 12.3°) > 4,6-Me2DBT (0.4°). Espe-
cially in [Cp*Ru(CO)2(η1(S)-4,6-Me2DBT)]+, rotation
around the Ru-S bond is prevented by the close
approach (3.076, 3.111 Å) of the 4,6-methyl groups to a
(16) Nakazawa, H.; Kawamura, K.; Kubo, K.; Miyoshi, K. Organo-
metallics 1999, 18, 2961.
(17) Bruce, M. I.; Zaitseva, N. N.; Skelton, B. W.; White, A. H. Aust.
J. Chem. 1998, 51, 433. The 1H NMR spectrum for the ruthenium
product observed was similar to that reported for the cation in [Cp*Ru-
(CO)2(PPh3)][Fe3(µ3-C2But)(CO)9].
(18) (a) Benson, J. W.; Angelici, R. J. Organometallics 1992, 11, 922.
(b) Goodrich, J. D.; Nickias, P. N.; Selegue, J. P. Inorg. Chem. 1987,
26, 3424..
(19) (a) Harman, W. D.; Schaefer, W. P.; Taube, H. J. Am. Chem.
Soc. 1990, 112, 2682. (b) Burns, C. J.; Andersen, R. A. J. Am. Chem.
Soc. 1987, 109, 915.
(20) (a) Fagan, P. J.; Ward, M. D.; Calabrese, J. C. J. Am. Chem.
Soc. 1989, 111, 1698. (b) Polam, J. R.; Porter, L. C. Inorg. Chim. Acta
1993, 205, 119. (c) Wang, C.-M. J.; Angelici, R. J. Organometallics 1990,
9, 1770. (d) Xia, A.; Selegue, J. P.; Carillo, A.; Brock, C. P. J. Am. Chem.
Soc. 2000, 122, 3973. (e) Chaudret, B.; Jalon, F. A. J. Chem. Soc.,
Chem. Commun. 1988, 11, 711.
(21) The X-ray data showed two structurally independent molecules
in the asymmetric unit cell of [Cp*Ru(CO)2(η1(S)-4-MeDBT)]BF4. See
ref 11.

Binding of Hindered Dibenzothiophene Ligands
Organometallics, Vol. 24, No. 9, 2005 2173
Table 4. Relative Equilibrium Constants, K′, for Reactions of [Cp′Ru(CO)2(η1(S)-4,6-Me2DBT)]+ with DBTh
in CD3NO2 or CD2Cl2 at 25.0 °C
DBTh
[CpRu(CO)2(η1(S)-DBTh)]+ a
[Cp*Ru(CO)2(η1(S)-DBTh)]+ b
4,6-Me2DBT
4-MeDBT
DBT
1.00
1.00
20.2(1)
62.7(6)
223(3)
2.79(7)
9.71(9)
36.5(3)
2,8-Me2DBT
a This study. CD3NO2 solvent. b Reference 11. CD2Cl2 solvent.
Table 5. Relative Equilibrium Constants (K′) for
the Reactions of [CpRu(CO)2(η1(S)-T)]+ with Th at
25.0 °C
plane defined by the Cp* methyl carbon atoms, which
restricts the twist angle to only 0.4°. In [Cp*Ru(CO)2-
(η1(S)-4-MeDBT)]+, the twist angle is larger (11.3°,
12.3°) and the 4-MeDBT ligand is oriented along the
Ru-S bond such that the 4-methyl group is twisted
away from the Cp* ligand, reducing the steric interac-
tion between the 4-methyl group and the Cp* methyl
groups; distances between the 4-methyl carbon atoms
and the Cp* planes are 3.417 and 3.508 Å. An even
larger twist angle (22.3°) was observed in [Cp*Ru(CO)2-
(η1(S)-DBT)]+, where the distance of closest approach
was 3.463 Å; the large twist angle in this complex is
presumably controlled by crystal packing forces. The
similar values for the distances of closest approach of
DBTh carbon atoms in [Cp*Ru(CO)2(η1(S)-4-MeDBT)]+
and [Cp*Ru(CO)2(η1(S)-DBT)]+ to the Cp* planes (3.42-
3.51 Å) suggest that those distances are sterically
noncrowding. The significantly shorter values in [Cp*Ru-
(CO)2(η1(S)-4,6-Me2DBT)]+ (3.076 and 3.111 Å) indicate
crowding of the 4,6-Me2DBT methyl groups and the Cp*
ligand. Thus, these structural studies show that [Cp*Ru-
(CO)2(η1(S)-4,6-Me2DBT)]+ (7) is the only complex in the
series of either Cp or Cp* [Cp′Ru(CO)2(η1(S)-DBTh)]+
complexes for which there is clear evidence of steric
crowding between the 4- or 6-methyl groups on the
DBTh and the Cp′ ligands.
Equilibrium Studies. Equilibrium constants (K) for
the exchange of one dibenzothiophene by another in
[CpRu(CO)2(η1(S)-DBTh)]+ were calculated using eq 2,
and the values obtained are listed in Table 3. Relative
equilibrium constants, K′ (given in parentheses), for the
displacement of 4,6-Me2DBT from [CpRu(CO)2(η1(S)-4,6-
Me2DBT)]+ by 4-MeDBT, DBT, and 2,8-Me2DBT are
calculated from the experimental K values and are listed
in Table 4. The K′ values increase in the order 4,6-Me2-
DBT (1.00) < 4-MeDBT (2.79 (7)) < DBT (9.71(9)) < 2,8-
Me2DBT (36.5(3)). The larger K′ value for 2,8-Me2DBT
(36.5), as compared to DBT, demonstrates that electron-
donating methyl groups in the nonhindering 2,8-posi-
tions increase the binding ability of DBT by a factor of
3.8, presumably by making the sulfur atom on DBT a
better donor to ruthenium. On the other hand, the
4-MeDBT and 4,6-Me2DBT ligands are only 0.29 and
0.10 times as strongly binding, respectively, as DBT.
The first methyl group in the 4-position decreases
binding by a factor of 0.29, while a second methyl group
in the 6-position reduces the binding even further by a
factor of 0.36. Although there is no direct evidence for
steric crowding by the 4,6-methyl groups in structural
studies of the [CpRu(CO)2(η1(S)-DBTh)]+ complexes,
there is clearly an effect on the equilibrium constants.
These constants show that the steric effect of the 4,6-
methyl groups overrides their increased electron-donat-
ing ability.
Th
K′
Ta
1.00
2-MeTa
3.30
4.76
3-MeTa
2,5-Me2Ta
BTa,b
20.7
47.6
80.9
4,6-Me2DBTb
4-MeDBTb
DBTb
233
800
887
Me4Ta
2,8-Me2DBTb
THTa
3016
>105
a Reference 14. CD2Cl2 solvent. b This study. CD3NO2 solvent.
as that reported previously for the [Cp*Ru(CO)2(η1(S)-
DBTh)]+ complexes11 that contain the bulkier η5-C5Me5
ligand (Table 4). As mentioned above, K′ values for the
2,8-Me2DBT and 4-MeDBT ligands are factors of 3.8
times and 0.29 times, respectively, of that for DBT in
the [CpRu(CO)2(η1(S)-DBTh)]+ complexes. These values
are very similar to factors of 3.6 times and 0.32 times
for 2,8-Me2DBT and 4-MeDBT, respectively, as com-
pared with DBT in the [Cp*Ru(CO)2(η1(S)-DBTh)]+
complexes. The major difference between K′ values for
the Cp and Cp* series of complexes is that between 4,6-
Me2DBT and 4-MeDBT. For the Cp complexes (2 and
3), 4-MeDBT is 2.8 times more strongly binding than
4,6-Me2DBT. On the other hand, for the Cp* complexes
(6 and 7), 4-Me2DBT binds 20 times more strongly than
4,6-Me2DBT. This much larger decrease in the binding
ability of 4,6-Me2DBT as compared with 4-MeDBT in
the Cp* complexes undoubtedly reflects the substantial
steric crowding between the 4,6-Me2DBT methyl groups
and the Cp* ligand that was evident in the structural
studies.
Previous work showed that in the series of [CpRu-
(CO)(PPh3)(η1(S)-Th)]+ complexes, where Th denotes the
thiophene (T), benzothiophene (BT), or dibenzothiophene
ligands, relative equilibrium constants (K′) follow the
order T (1.00) < BT (29.9) < DBT (74.1) < 2,8-Me2DBT
(358).22 A similar study for the [CpRu(CO)2(η1-(S)-Th)]+
complexes revealed the same relative trend for T (1.00)
and BT (47.6), but DBT was not examined due to its
poor solubility in the chosen CH2Cl2 solvent system.14
To directly compare our results to the previous study
undertaken in methylene chloride, three independent
reactions of [CpRu(CO)2(η1(S)-BT)]+ and DBT were
performed in CD3NO2; the K was determined to be
0.0597(9). This K value allows for the calculation of the
relative equilibrium constants listed in Table 5. The
weakest ligand in this series is thiophene and the
strongest is tetrahydrothiophene (THT). It is clear that
The relative trend in K′ values for the DBTh ligands
(22) White, C. J.; Wang, T.; Jacobson, R. A.; Angelici, R. J. Orga-
nometallics 1994, 13, 4474.
in the [CpRu(CO)2(η1(S)-DBTh)]+ complexes is the same

2174 Organometallics, Vol. 24, No. 9, 2005
Vecchi et al.
Table 6. Rate Constants, kobs (s-1), for Reactions
(eq 4) of 0.020 M [Cp′Ru(CO)2(η1(S)-DBTh)]+ with
PR3 at 25.0 °C
although methyl groups in the 4- and 6-positions reduce
binding of DBT to {CpRu(CO)2}+, 4,6-Me2DBT is still a
better ligand than benzothiophene in this series. This
suggests that it should be possible to use metal com-
plexes to remove hindered dibenzothiophenes from
diesel fuels by solution phase or solid extraction.
DBTh
PR3
[PR3], M
106kobs, s-1
[CpRu(CO)2(η1(S)-DBTh)]+ a
4,6-Me2DBT
PPh3
0.10
0.20
0.30
0.40
0.20
0.60
1.00
0.20
0.60
0.20
0.20
0.60
0.20
0.20
1.00
5.21
5.30
5.28
5.25
5.21
5.07
5.25
4.47
4.54
3.97
4.03
4.02
1.32
1.28
1.38
The DBTh binding studies discussed above were the
basis for the use of {CpRu(CO)2BF4} adsorbed on silica-
based supports for the removal of dibenzothiophenes
from a simulated gasoline/diesel fuel.8 The adsorbent
was created by reacting CpRu(CO)2Cl with AgBF4 in
CH2Cl2, filtering the solution onto a solid silica-based
support, and evaporating the solution to dryness. Simu-
lated petroleum solutions of DBT and of 4,6-Me2DBT
were then stirred with the solid phase adsorbent. Within
30 min, the sulfur levels were reduced from 400 to 4
ppm for a DBT solution and from 400 to 113 ppm for a
4,6-Me2DBT solution, corresponding to decreases in the
sulfur levels of 96% and 72%, respectively. The less
efficient removal of 4,6-Me2DBT is a direct consequence
of methyl groups in the 4,6-positions sterically hindering
sulfur coordination to the adsorbent. These results
indicate that solution binding studies can be useful in
developing new strategies for sulfur removal.
Kinetic Studies of Substitution Reactions of
[Cp′Ru(CO)2(η1(S)-DBTh)]+. Kinetic studies of the
reactions of [Cp′Ru(CO)2(η1(S)-DBTh)]+ with either
P(OPh)3, PPh3, or PPh2Me according to eq 4 were
performed under pseudo-first-order conditions with a
minimum of 10-fold excess of PR3. The pseudo-first-
order rate constants, kobs, are listed in Table 6. Plots of
kobs versus [PR3] for the reactions of [CpRu(CO)2(η1(S)-
DBTh)]+ are independent of the PR3 concentration so
that kobs ) k1. However, for the [Cp*Ru(CO)2(η1(S)-
DBTh)]+ complexes, kobs depends partially on the con-
P(OPh)3
4-MeDBT
DBT
P(OPh)3
P(OPh)3
2,8-Me2DBT
P(OPh)3
[Cp*Ru(CO)2(η1(S)-DBTh)]+ b
4,6-Me2DBT
DBT
PPh3
0.20
0.40
0.60
1.00
0.20
0.40
0.60
0.80
1.00
0.20
0.60
0.80
1.00
0.20
0.40
0.60
0.80
1.00
0.20
0.40
0.60
0.80
1.00
0.20
0.40
0.60
0.80
1.00
475
517
548
600
PPh3
131
150
169
183
201
PPh2Me
PPh3
139
175
194
214
4-MeDBT
71.0(5)
77.4
84.1
91.4
98.2
71.1
78.0
85.9
97.8
101
PPh2Me
PPh3
centration of the phosphine such that kobs ) k1
+
2,8-Me2DBT
29.2
35.1
38.9
42.5
44.0
k2[PR3]. The k1 values for both series of complexes are
given in Table 7.
The first-order rate law indicates that the reaction of
[CpRu(CO)2(η1(S)-DBTh)]+ with P(OPh)3 proceeds by a
mechanism involving slow dissociation of the DBTh
ligand followed by rapid reaction of the unsaturated
intermediate with P(OPh)3 to give the final product. The
k1 values for these reactions, listed in Table 7, decrease
with the DBTh ligand in the following order: 4,6-Me2-
DBT (5.18(5) × 10-6 s-1) > 4-MeDBT (4.50(3) × 10-6
s-1) > DBT (4.01(3) × 10-6 s-1) > 2,8-Me2DBT (1.33(5)
× 10-6 s-1). This trend parallels the trend observed in
the equilibrium studies (Table 4), which shows that the
more strongly binding the DBT ligand thermodynami-
cally, the slower its rate of dissociation from [CpRu-
(CO)2(η1(S)-DBTh)]+. The rate of dissociation of the 2,8-
Me2DBT ligand is a factor of 0.33 as fast as that of DBT.
Thus, electron-donating methyl groups located away
from the sulfur atom in 2,8-Me2DBT not only increase
the thermodynamic stability of the Ru-S bond, as the
equilibrium studies indicate, but also decrease the
kinetic lability of the Ru-S bond. On the other hand,
the rates of dissociation of the 4-MeDBT and 4,6-Me2-
DBT ligands are 1.13 and 1.30 times faster, respectively,
than that of DBT. The steric effects of methyl groups
in the 4,6-positions of DBT outweigh their electron-
donating effects, which results in lower thermodynamic
a CD3NO2 solvent. b CD2Cl2 solvent.
stabilities and greater kinetic labilities of the Ru-S
bond in the 4-MeDBT and 4,6-Me2DBT complexes.
When PPh3 was used as the incoming ligand (eq 4),
[CpRu(CO)2(η1(S)-4,6-Me2DBT)]+ reacted cleanly to give
[CpRu(CO)2(PPh3)]+, whose 1H NMR spectrum was
nearly identical to that reported previously for this
compound.23 Due to the relative insolubility of PPh3 in
CD3NO2, reaction solutions containing only 10-, 15-, and
20-fold excess of PPh3 were possible. The rates of these
reactions were independent of the PPh3 concentration
(Table 6), and the k1 rate constant was the same within
experimental error as that with P(OPh)3 (Table 7).
However, reactions of the DBT (1) and 4-MeDBT (2)
complexes with PPh3 resulted in the formation of two
phosphine products within 12 h: [CpRu(CO)2(PPh3)]+
and a product whose 1H NMR spectrum was similar to
that reported for [CpRu(CO)(PPh3)2]+.24 Qualitatively,
(23) (a) Davies, S. G.; Simpson, S. J. Dalton Trans. 1984, 993. (b)
Humphries, A. P.; Knox, S. A. R. Dalton Trans. 1975, 1710.
(24) Stone, F. G. A.; Blackmore, T.; Bruce, M. I. J. Chem. Soc., A
1971, 2376.

Binding of Hindered Dibenzothiophene Ligands
Organometallics, Vol. 24, No. 9, 2005 2175
Table 7. First- and Second-Order Rate Constants, k1 and k2, for the Substitution of DBTh in
[Cp′Ru(CO)2(η1(S)-DBTh)]+ by PR3 (eq 4) at 25.0 °C
[CpRu(CO)2(η1(S)-DBTh)]+,a
[Cp*Ru(CO)2(η1(S)-DBTh)]+,b
DBTh
106 k1, s-1
106k1, s-1 106k2, M-1 s-1
4,6-Me2DBT
4-MeDBT
DBT
5.18(10),c 5.26(4)d
4.50(3)c
451((8)d
152((13)d
63.9((0.3),d 62.7((2.1)e
115((2),d 120((2)e
26.8((1.5)d
34.2((0.4),d 40.2((3.2)e
86.5((2.8),d 93.0((1.6)e
18.5((2.2)d
4.01(3)c
2,8-Me2DBT
1.33(5)c
a In CD3NO2. b In CD2Cl2. c For PR3 ) P(OPh)3. d For PR3 ) PPh3. e For PR3 ) PPh2Me.
this reaction occurred more rapidly and more of the
second phosphine product was formed in the reactions
of the DBT complex (1) than the 4-MeDBT complex (2).
Because these reactions yielded two products, their
kinetics were not investigated further.
complex presumably is a result of the greater electron-
donating ability of the Cp* ligand. The added electron
density reduces the Lewis acidity of the Ru atom, which
weakens the Ru-S bond and makes the DBT ligand
more labile in [Cp*Ru(CO)2(η1(S)-DBT)]+ (5) as com-
pared to the Cp complex 1. The dissociation rate of
4-MeDBT from [Cp*Ru(CO)2(η1(S)-4-MeDBT)]+ is 14
times faster than that from its Cp analogue, and the
rate for 2,8-Me2DBT dissociation from [Cp*Ru(CO)2-
(η1(S)-2,8-Me2DBT)]+ is 20 times faster than that from
its Cp analogue. The increases in the kinetic labilities
of the 4-MeDBT and 2,8-Me2DBT ligands in the Cp*
complexes as compared with the Cp complexes are also
likely a result of the greater electron-donating ability
of the Cp* ligand as described for DBT above. On the
other hand, the dissociation rate of 4,6-Me2DBT from
[Cp*Ru(CO)2(η1(S)-4,6-Me2DBT)]+ (7) is 86 times its rate
of dissociation from [CpRu(CO)2(η1(S)-4,6-Me2DBT)]+.
This much larger rate of dissociation of 4,6-Me2DBT in
the Cp* complex (7) than in its Cp analogue (3) is due
not only to the higher electron-donating ability of Cp*
but also to the steric effect of the more bulky Cp* ligand.
On the basis of the above comparisons, the electronic
effect of the Cp* ligand increases the rate of 4,6-Me2-
DBT dissociation from 7 by a factor of approximately
20, which means that the rate enhancement caused by
steric erepulsion is a factor of approximately 4.
The k2 values obtained from the plots of kobs versus
phosphine concentration for the reactions of the [Cp*Ru-
(CO)2(η1(S)-DBTh)]+ complexes with PPh3 and PPh2Me
are listed in Table 7. The k2 values for the reaction with
PPh3 decrease in the same order as the k1 values: 4,6-
Me2DBT ((152 ( 13) × 10-6 s-1) > DBT ((86.5 ( 2.8) ×
10-6 s-1) > 4-MeDBT ((34.2 ( 0.4) × 10-6 s-1) > 2,8-
Me2DBT ((18.6 ( 2.2) × 10-6 s-1). The similarity of the
trends in k1 and k2 values suggests that Ru-S bond-
breaking primarily determines the overall rate of the
second-order reaction. That bond-making with the
incoming PR3 ligand is less important is indicated by
the fact that the k2 values for PPh3 and PPh2Me in their
reactions with 5 and 6 are nearly the same (Table 7). If
the mechanism were to involve nucleophilic attack of
the phosphine on the Ru, one would expect a much
larger difference in their rates.25 Thus, the mechanism
for the k2 pathway is best described as a dissociative
interchange (Id) where the nature of the [Cp*Ru(CO)2-
(η1(S)-DBTh)][BF4]‚PR3 intermediate is not clear. It
should be noted that this Id pathway contributes to the
total rate only at relatively high concentrations of the
phosphine.
The kobs rate constants for reactions of the [Cp*Ru-
(CO)2(η1(S)-DBTh)]+ complexes with PPh3 and PPh2Me
show (Table 6) a small but discernible contribution from
a second-order (k2) pathway in addition to the k1
pathway. The k1 values for the reactions of [Cp*Ru(CO)2-
(η1(S)-DBTh)]+ with PPh3 decrease in the following
order: 4,6-Me2DBT ((451 ( 8) × 10-6 s-1) > DBT ((115
( 2) × 10-6 s-1) > 4-MeDBT ((63.9 ( 0.3) × 10-6 s-1) >
2,8-Me2DBT ((26.8 ( 1.5) × 10-6 s-1). The reactions of
[Cp*Ru(CO)2(η1(S)-DBT)]+ and [Cp*Ru(CO)2(η1(S)-4-
MeDBT)]+ with PPh2Me yielded k1 values ((120 ( 2) ×
10-6 s-1 for DBT and (62.7 ( 2.1) × 10-6 s-1 for
4-MeDBT) that were the same within experimental
error as those determined using PPh3. The rate-
determining step for the k1 pathway in the reactions of
the [Cp*Ru(CO)2(η1(S)-DBTh)]+ complexes with both
PPh3 and PPh2Me is likely dissociation of the DBTh
ligand. As for the [CpRu(CO)2(η1(S)-DBTh)]+ complexes,
the rates for three of the four [Cp*Ru(CO)2(η1(S)-
DBTh)]+ complexes decrease as follows: 4,6-Me2DBT >
DBT > 2,8-Me2DBT. Steric and electronic factors ac-
count for this trend, as discussed above. On the other
hand, the 4-MeDBT ligand in [Cp*Ru(CO)2(η1(S)-4-
MeDBT)]+ (6) dissociates more slowly than DBT from
[Cp*Ru(CO)2(η1(S)-DBT)]+, which is different than the
trend in rates for the [CpRu(CO)2(η1(S)-DBTh)]+ com-
plexes and different than the trend in equilibrium
binding constants (Table 4) for both series of [Cp′Ru-
(CO)2(η1(S)-DBTh)]+ complexes. To explain the anoma-
lously slow rate of 4-MeDBT dissociation from 6, one
can propose that the low-energy pathway for 4-MeDBT
dissociation is one in which the DBTh ligand is oriented
so that the complex has a plane of symmetry; that is,
the complex has a 0° twist angle. The 4,6-Me2DBT
complex is forced by steric factors to have this orienta-
tion, and the DBT and 2,8-Me2DBT ligands can easily
achieve this orientation by rotation around the Ru-S
bond. However, there will be a barrier to achieving this
orientation with the 4-MeDBT ligand because of steric
repulsion between the 4-methyl group and the Cp*
methyl groups, which may account for the unexpectedly
slow rate of 4-MeDBT dissociation from complex 6.
The k1 values for the [Cp*Ru(CO)2(η1(S)-DBTh)]+
complexes are all significantly greater than those
determined for the [CpRu(CO)2(η1(S)-DBTh)]+ com-
plexes. For example, the rate of dissociation of DBT is
29 times faster in the Cp* complex (115 × 10-6 s-1) as
compared with that in the Cp complex (4.01 × 10-6 s-1).
Since there is little steric influence in both the Cp and
Cp* complexes of DBT, the faster rate for the Cp*
Conclusions
Investigations of the two series of dibenzothiophene
complexes [Cp′Ru(CO)2(η1(S)-DBTh)]+, where Cp′ ) Cp
(25) Poe, A. J. Pure Appl. Chem. 1988, 60, 1209.

2176 Organometallics, Vol. 24, No. 9, 2005
Vecchi et al.
or Cp*, show that only in the structure of [Cp*Ru(CO)2-
(η1(S)-4,6-Me2DBT)]+ is there evidence for steric crowd-
ing between methyl groups in the 4,6-Me2DBT and Cp*
ligands. This effect is also evident in the equilibrium
binding constants (K′) for the coordination of 4,6-Me2-
DBT to {Cp*Ru(CO)2}+, which are much smaller than
those for 4-MeDBT. In both series of complexes, both
steric and electronic factors affect the K′ values, which
increase in the order 4,6-Me2DBT < 4-MeDBT < DBT
< 2,8-Me2DBT. The rates of DBTh substitution by PR3,
following a dissociative mechanism, are especially fast
for the sterically crowded [Cp*Ru(CO)2(η1(S)-4,6-Me2-
DBT)]+, but steric and electronic effects also contribute
to the rates of substitution of the other DBTh ligands.
Acknowledgment. This work was supported by the
U.S. Department of Energy, Office of Science, Office of
Basic Energy Sciences, Chemical Sciences Division,
under contract W-7405-Eng-82 with Iowa State Uni-
versity.
Supporting Information Available: Crystallographic
files (CIF) and tables giving crystallographic data for com-
plexes 2, 3, and 4, including atomic coordinates, bond lengths
and angles, and anisotropic displacement parameters. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OM048973L