Re2(CO)10-mediated carbon-hydrogen and carbon-sulfur bond cleavage of dibenzothiophene and 2,5-dimethylthiophene

Article abstract of DOI:10.1021/om000708m

Ultraviolet photolysis of Re₂(CO)₁₀ and excess dibenzothiophene (DBT) in the noncoordinating solvent hexanes produces the S-bound eq-Re₂(CO)₉(η¹(S)-DBT) (1) and the novel C-H-cleaved DBT complex Re₂(CO)₈(μ-C₁₂ H₇S)(μ-H) (2). Under similar conditions, Re₂(CO)₁₀ reacts with excess 2,5-dimethylthiophene (2,5-Me²T) to give the interesting C-S-cleaved 2,5-Me₂T complex Re₂(CO)₇(μ-2,5-Me₂T) (3). The photolysis reactions of Re₂(CO)₁₀ with DBT and 2,5-Me₂T were inhibited by CO (1 atm) and also by the radical scavenger TEMPO, which suggests that both CO dissociation and homolytic Re-Re bond cleavage are involved. The η¹(S)-bound thiophene complexes 1 and Re₂(CO)₉(η¹(S)-2,5-Me₂T) (5) were prepared from Re₂(CO)₉(THF) (4). The DBT ligand in 1 is labile and rapidly (<2 min) reacts with CO (1 atm) to form Re₂(CO)₁₀ in 1,2-DCE. On the basis of its X-ray structure, complex 1 has one of the smallest tilt angles (θ = 113°) observed for a metal-thiophene complex, which may be understood in terms of π-back-bonding arguments. Complexes 1-3 were characterized by spectroscopic (IR, NMR) methods and by their structures that were determined by X-ray crystallography. Mechanisms for the formation of 1-3 are presented and discussed.

Full text of DOI:10.1021/om000708m

Organometallics 2001, 20, 1071-1078
1071
Re2(CO)10-Med ia ted Ca r bon -Hyd r ogen a n d
Ca r bon -Su lfu r Bon d Clea va ge of Diben zoth iop h en e a n d
2,5-Dim eth ylth iop h en e†
Michael A. Reynolds, Ilia A. Guzei,‡ and Robert J . Angelici*
Ames Laboratory and Department of Chemistry, Iowa State University, Ames, Iowa 50011
Received August 14, 2000
Ultraviolet photolysis of Re2(CO)10 and excess dibenzothiophene (DBT) in the noncoordi-
nating solvent hexanes produces the S-bound eq-Re2(CO)9(η1(S)-DBT) (1) and the novel C-H-
cleaved DBT complex Re2(CO)8(µ-C12H7S)(µ-H) (2). Under similar conditions, Re2(CO)10 reacts
with excess 2,5-dimethylthiophene (2,5-Me2T) to give the interesting C-S-cleaved 2,5-Me2T
complex Re2(CO)7(µ-2,5-Me2T) (3). The photolysis reactions of Re2(CO)10 with DBT and 2,5-
Me2T were inhibited by CO (1 atm) and also by the radical scavenger TEMPO, which suggests
that both CO dissociation and homolytic Re-Re bond cleavage are involved. The η1(S)-bound
thiophene complexes 1 and Re2(CO)9(η1(S)-2,5-Me2T) (5) were prepared from Re2(CO)9(THF)
(4). The DBT ligand in 1 is labile and rapidly (<2 min) reacts with CO (1 atm) to form
Re2(CO)10 in 1,2-DCE. On the basis of its X-ray structure, complex 1 has one of the smallest
tilt angles (θ ) 113°) observed for a metal-thiophene complex, which may be understood in
terms of π-back-bonding arguments. Complexes 1-3 were characterized by spectroscopic
(IR, NMR) methods and by their structures that were determined by X-ray crystallography.
Mechanisms for the formation of 1-3 are presented and discussed.
Ch a r t 1
In tr od u ction
Hydrodesulfurization (HDS), the catalytic hydrotreat-
ing process used for removing sulfur from organosulfur
compounds present in petroleum feedstocks, is impor-
tant for two primary reasons.1-3 First, current U.S.
federal regulations require refineries to cap gasoline
sulfur levels at 300 ppm (by the year 20043) in order to
decrease the amount of sulfur oxides (SOx, x ) 2, 3)
produced during the combustion of petroleum fuels.
These sulfur oxides are known precursors to acid rain
and present a significant threat to the environment.
Second, sulfur poisons precious metal-based re-forming
catalysts, which consequently increases the cost of
petroleum re-forming. The sulfur in crude petroleum,
whose amount ranges from 0.2 to 4%,2,4 is present in
the form of thiols (RSH), thioethers (RSR), disulfides
(RSSR), and thiophenes (T*) such as thiophene (T),
benzothiophene (BT), and dibenzothiophene (DBT) (Chart
1). It is the thiophenic compounds, however, that are
the most difficult to desulfurize under current com-
mercial hydrotreating conditions due to aromatic sta-
bilization of the thiophene and benzothiophene rings.5
The common commercial HDS catalysts are typically
MoS2 and WS2 promoted by Co or Ni on an alumina
support, but several other metal sulfides, MSx (M ) Ru,
Re, Os, Rh, and Ir), have demonstrated higher catalytic
activities for the desulfurization of thiophenes and
benzothiophenes under various hydrotreating condi-
tions.6,7 The most active catalysts are those of the
second-row (such as Ru and Rh) and third-row elements
(such as Re, Os, and Ir), but their high costs prevent
them from being used commercially.2
* To whom correspondence should be addressed. E-mail: angelici@
iastate.edu.
† Our congratulations to Professor Fred Basolo for being awarded
the 2001 Priestley Medal of the American Chemical Society.
‡ Iowa State University Molecular Structure Laboratory.
(1) (a) Schuman, S. C.; Shalit, H. Catal. Rev. 1970, 4, 245. (b) Prins,
R.; De Beer, V. H. J .; Somorjai, G. A. Catal. Rev-Sci. Eng. 1989, 31, 1.
(c) Topsøe, H.; Clausen, B. S.; Massoth, F. E. Hydrotreating Catalysts
in Catalysis: Science and Technology; Andersen, J . R., Boudart, M.,
Eds.; Pergamon Press: New York, 1981; p 9.
(5) (a) Gates, B. C.; Katzer, J . R.; Schuit, G. C. A. Chemistry of
Catalytic Processes; McGraw-Hill: New York, 1979; pp 390-447. (b)
Galpern, G. D. In The Chemistry of Heterocyclic Compounds; Gronow-
itz, S., Ed.; Wiley: New York, 1985; Vol. 44, Part 1, pp 325-351. (c)
Whitehurst, D. D.; Isoda, T.; Mochida, I. Adv. Catal. 1998, 42, 345.
(6) (a) Pecoraro, T. A.; Chianelli, R. R. J . Catal. 1981, 67, 430. (b)
Chianelli, R. R. Catal. Rev. 1984, 26, 361.
(2) Angelici, R. J . In Encyclopedia of Inorganic Chemistry; King, R.
B., Ed.; Wiley: New York, 1994; Vol. 3, pp 1433-1443.
(3) United States Environmental Protection Agency, Air and Radia-
tion, Office of Transportation and Air Quality, EPA420-F-99-051, 1999;
at http://www.epa.gov/oms/regs/ld-hwy/tier-2/frm/f99051.htm.
(4) (a) Geochemistry of Sulfur in Fossil Fuels; Orr, W. L., White, C.
M., Eds.; ACS Symposium Series 429; American Chemical Society:
Washington, DC, 1990.
(7) Ledoux, M. J .; Michaux, O.; Agostini, G.; Panissod, P. J . Catal.
1986, 102, 275.
10.1021/om000708m CCC: $20.00 © 2001 American Chemical Society
Publication on Web 02/13/2001

1072 Organometallics, Vol. 20, No. 6, 2001
Reynolds et al.
Sch em e 1
Organometallic modeling of thiophene binding modes
and ring-opened intermediates formed at transition-
metal centers is one approach to better understanding
how desulfurization might occur in the commercial HDS
catalytic process. Examples of organometallic complexes
that contain thiophenes and benzothiophenes bound to
transition-metal centers through many different binding
modes (i.e.; η1(S), η2, η4, η5, and η6)8-10 are known.
Several organometallic complexes of ring-opened thio-
phenes and benzothiophenes at single metal centers
have also been reported;11-13 however, there are few
reports of such complexes with two or more metals.
Sweigart et al.14-17 described the synthesis of the
C-S-cleaved thiophene complexes Mn2(CO)7(µ-T*) by
reduction of the complexes [(η5-T)Mn(CO)3]+ and [(η6-
BT)Mn(CO)3]+. J ones and Vicic18 recently reported that
the bimetallic Ni complex [(dippe)Ni(H)]2 reacts with T
or BT under mild conditions (22 °C) in hexanes solvent,
with loss of H2, to produce the metal-inserted thiophene-
based metallacycles (dippe)Ni(η2(C,S)-T) and (dippe)Ni-
(η2(C,S)-BT). Both of these Ni-thiophene complexes
reacted further in solution to give the bimetallic species
[(dippe)Ni]2(µ-T) and [(dippe)Ni]2(µ-BT), respectively,
which contain bridging, ring-opened thiophenes.
Building on previous studies,19 Rauchfuss20 reported
both C-S cleavage and desulfurization of thiophenes
and benzothiophene during thermolysis with Fe3(CO)12
to obtain the thiaferroles Fe2(CO)6(µ-T*)19,20 (T* ) T,
BT). In the case of thiophene, ferroles such as Fe2(CO)6-
(C4H4) resulted from metal insertion and desulfurization
of the parent thiophene ligand.
two Re centers of a Re2(CO)7 moiety through both the
S atom and the vinyl group, which behaves as an η1 and
η2 ligand to the two Re centers. Our recent success in
preparing Re2(CO)7(µ-BT) prompted us to investigate
the reactivity of DBT and thiophenes such as 2,5-Me2T
under similar conditions. Here we report our results of
UV-light-promoted reactions of dibenzothiophene (DBT)
and 2,5-dimethylthiophene (2,5,-Me2T) with Re2(CO)10
to afford dinuclear Re complexes with S-bound, C-H-
cleaved, and C-S-cleaved thiophene ligands.
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All reactions were performed
under a nitrogen or argon atmosphere in reagent grade
solvents, using standard Schlenk techniques. Hexanes, THF,
ethyl ether, and methylene chloride were dried and purified
using the Grubbs solvent purification process22 purchased from
Solv-Tek, Inc. Benzene was dried over CaH2 and then distilled
prior to use. Deuteriomethylene chloride (Cambridge) was
stored over 4 Å molecular sieves. Rhenium carbonyl was
purchased from Strem Chemicals Inc. TEMPO (2,2,6,6-tet-
ramethyl-1-piperidinyloxy free radical), dibenzothiophene, and
2,5-dimethylthiophene were purchased from Aldrich Chemical
Co. and used without further purification. Trimethylamine
oxide hydrate (Aldrich) was dried by azeotropic distillation of
the water from benzene. Silica gel (J . T. Baker, 40-140 mesh)
was dried under vacuum at ambient temperature for 14 h and
stored under argon prior to use. Neutral alumina (Aldrich,
Brockmann I) was dried under vacuum at ambient tempera-
ture for 14 h and treated with 10% (by weight) water under
argon with vigorous shaking.
The 1H NMR spectra for all complexes were recorded on
either a Varian VXR-300 MHz or an in-house 400 MHz NMR
spectrometer using the deuterated solvent as both internal lock
and internal reference. Solution infrared spectra were recorded
on a Nicolet-560 spectrophotometer using NaCl cells with 0.1
mm spacers. Elemental analyses were performed on a Perkin-
Elmer 2400 Series II CHNS/O analyzer. All photochemical
reactions were carried out in a 40-60 mL capacity quartz
Schlenk photolysis tube fitted with a coldfinger, which was
immersed into the reaction solution. Irradiation was performed
using a Hanovia 450 W Hg medium-pressure lamp as the light
source (inserted into either a quartz or Pyrex water-cooled
jacket). The reaction temperature was controlled using an
Isotemp 1013P refrigerated circulating bath (Fisher Scientific)
with circulation hoses connected to the coldfinger.
Recently, we reported the preparation of the C-S-
cleaved BT complex Re2(CO)7(µ-BT),21 by UV irradiation
of Re2(CO)10 and benzothiophene in hexanes solvent
(Scheme 1). In this novel bimetallic Re complex, the BT
ligand is opened at the Cvinyl-S bond and bridges the
(8) (a) Angelici, R. J . Polyhedron 1997, 16, 3073. (b) Angelici, R. J .
Bull. Soc. Chim. Belg. 1995, 104, 265. (c) Angelici, R. J . Coord. Chem.
Rev. 1990, 105, 61.
(9) Bianchini, C.; Meli, A. J . Chem. Soc., Dalton Trans. 1996, 801.
(10) Rauchfuss, T. B. Prog. Inorg. Chem. 1991, 21, 259.
(11) (a) J ones, W. D.; Dong, L. J . Am. Chem. Soc. 1991, 113, 559.
(b) Dong, L.; Duckett, S. B.; Ohman, K. F.; J ones, W. D. J . Am. Chem.
Soc. 1992, 114, 151. (c) J ones, W. D.; Chin, R. M.; Crane, T. W.; Baruch,
D. M. Organometallics 1994, 13, 4448.
(12) (a) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F.; Frediani,
P.; Herrera, V.; Sanchez-Delgado, R. A. J . Am. Chem. Soc. 1993, 115,
2731. (b) Bianchini, C.; Herrera, V.; J imenez, M. V.; Laschi, F.; Meli,
A.; Sanchez-Delgado, R. A.; Vizza, F.; Zanello, P. Organometallics 1995,
14, 4390. (c) Bianchini, C.; Frediani, P.; Herrera, V.; J imenez, M. V.;
Meli, A.; Rincon, L.; Sanchez-Delgado, R. A.; Vizza, F. J . Am. Chem.
Soc. 1995, 117, 4333.
(13) (a) Garcia, J . J .; Maitlis, P. M. J . Am. Chem. Soc. 1993, 115,
12200. (b) Garcia, J . J .; Mann, B. E.; Adams, H.; Bailey, N. A.; Maitlis,
P. M. J . Am. Chem. Soc. 1995, 117, 2179. (c) Chen, J .; Angelici, R. J .
Organometallics 1999, 18, 5721.
(14) (a) Dullaghan, C. A.; Sun, S.; Carpenter, G. B.; Weldon, B.;
Sweigart, D. A. Angew. Chem., Int. Ed. Engl. 1996, 35, 212. (b)
Dullaghan, C. A.; Carpenter, G. B.; Sweigart, D. A.; Choi, D. S.; Lee,
S. S.; Chung, Y. K. Organometallics 1997, 16, 5688.
(15) Zhang, X.; Dullaghan, C. A.; Watson, E. J .; Carpenter, G. B.;
Sweigart, D. A. Organometallics 1998, 17, 2067.
(16) Dullaghan, C. A.; Zhang, X.; Walther, D.; Carpenter, G. B.;
Sweigart, D. W.; Meng, Q. Organometallics 1997, 16, 5604.
(17) Zhang, X.; Dullaghan, C. A.; Carpenter, G. B.; Sweigart, D. A.;
Meng, Q. Chem. Commun. 1998, 93.
(18) Vicic, D. A.; J ones, W. D. J . Am. Chem. Soc. 1999, 121, 7606.
(19) (a) King, R. B.; Stone, F. G. A. J . Am. Chem. Soc. 1960, 82,
4557. (b) Dettlaf, G.; Weiss, E. J . Organomet. Chem. 1976, 108, 213.
(c) Hu¨bener, P.; Weiss, E. J . Organomet. Chem. 1977, 129, 105.
(20) Ogilvy, A. E.; Draganjac, M.; Rauchfuss, T. B.; Wilson, S. R.
Organometallics 1988, 7, 1171.
P r ep a r a tion of eq-Re2(CO)9(η1(S)-DBT) (1) a n d Re2-
(CO)8(µ-C12H7S)(µ-H) (2). A hexanes solution (30 mL) of Re2-
(CO)10 (150 mg, 0.230 mmol) and DBT (82.6 mg, 0.448 mmol)
was prepared in
a quartz photolysis tube, containing a
(21) Reynolds, M. A.; Guzei, I. A.; Angelici, R. J . Chem. Commun.
2000, 513.
(22) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J . Organometallics 1996, 15, 1518.

Re2(CO)10-Mediated C-H and C-S Bond Cleavage
Organometallics, Vol. 20, No. 6, 2001 1073
magnetic stir bar. A coldfinger (15 °C) was immersed into the
reaction solution, and an oil bubbler was then connected to
the tube. The solution was irradiated under nitrogen for 18-
24 h with stirring, during which time a light yellow solution
containing a brown precipitate was produced. More yellow
precipitate was also present above the solution on the sides of
the tube. The solution was filtered into a Schlenk flask, and
the solvent was removed in vacuo to produce a crude lemon
yellow solid residue containing complexes 1 and 2, unreacted
Re2(CO)10, DBT, and the Re cluster HRe3(CO)1423 (<10% based
on Re2(CO)10). The yellow residue on the photolysis tube wall
was washed with hexanes (2 × 5 mL), dried in vacuo, and
dissolved in CH2Cl2. This solution was then filtered and the
filtrate combined with the lemon yellow solid residue from the
previous filtration. More CH2Cl2 was added to the now yellow-
brown solution until all solids were dissolved (3-6 mL). The
resulting solution was layered with hexanes and cooled (-20
°C) overnight. After 1 day, yellow crystals of complex 2 formed
and complex 1 also precipitated from solution as a yellow
powder. The crystals and yellow solid were filtered and dried
in vacuo. Complexes 1 (9.2-28 mg, 5-15%) and 2 (18-36 mg,
10-20%) were separated by hand-picking crystals of 2 from
the mixture. Attempts to purify the crude mixture of products
by chromatography on silica gel or alumina were unsuccessful
due to decomposition. However, HRe3(CO)14 was recovered
during chromatography of the crude mixture on silica gel.
was isolated as an impure orange-yellow oil by removing the
solvent under vacuum. The oil also contained unreacted Re2-
(CO)10 (on the basis of IR spectroscopy) and other unidentified
1
impurities (on the basis of H NMR spectroscopy). Attempts
to purify 4 on silica gel or alumina using THF as the solvent
produced an orange band that did not elute in THF, CH2Cl2,
or benzene solvents. However, crude 4 was sufficiently stable
for spectroscopic analysis. 1H NMR (CD2Cl2, 300 MHz): δ 3.86
(m, 4 H), 1.89 (m, 4 H). IR (THF): 2101 (w), 2038 (m), 1987
(vs), 1980 (sh), 1952 (m), 1912 (m) cm-1
.
P r ep a r a tion of th e Re2(CO)9(η1(S)-T*) Com p lexes 1
a n d 5 fr om 4. Com p lex 1. A solution of Re2(CO)9(THF) (4)
was prepared as described above (using method A) from a THF
(30 mL) solution of Re2(CO)10 (206 mg, 0.316 mmol). The
solution of complex 4 was warmed to room temperature,
followed by the addition of DBT (114 mg, 0.619 mmol) with
stirring. The orange-yellow solution lightened in color after
1-2 h, and stirring was continued for an additional 18 h. The
solvent was then removed in vacuo, and the remaining yellow
residue was extracted with CH2Cl2 (1 mL) and layered with
hexanes (5 mL) followed by cooling to -20 °C until crystals of
1 were produced (1-2 days). The crystals were then filtered,
washed with hexanes (2 × 10 mL), and dried to produce pure
1 (70 mg, 27% based on Re2(CO)10). 1H NMR (CD2Cl2, 400
MHz): δ 8.17 (m, 2 H), 7.84 (m, 2 H), 7.61 (m, 4 H). IR (CH2-
Cl2): νCO 2102 (w), 2042 (m), 1988 (vs), 1963 (m), 1933 (m)
cm-1. Anal. Calcd for C21H8O9Re2S: C, 31.19; H, 1.00. Found:
C, 31.09; H, 0.83.
1
Complex 1 was characterized by IR and H NMR spectroscopy
of the mixture of 1 and 2 and compared to an authentic sample
that was prepared as described below from Re2(CO)9(THF) and
DBT. Data for 2: 1H NMR (CD2Cl2, 300 MHz) δ 8.06 (m, 1 H),
7.87 (d, 1 H, J ) 7.2 Hz), 7.80 (d, 1 H, J ) 7.2 Hz), 7.69 (m, 1
H), 7.56 (m, 2 H), 7.21 (t, 1 H, J ) 7.2 Hz), -14.8 (s, 1 H); IR
(hexanes) νCO 2114 (w), 2087 (w), 2023 (vs), 2013 (s), 1995 (m),
1985 (s), 1958 (m). Anal. Calcd for C20H8O8Re2S: C, 30.77; H,
1.03. Found: C, 30.74; H, 1.03.
Com p lex 1 (fr om Meth od B).24 A THF solution (20 mL)
of Re2(CO)10 (306 mg, 0.469 mmol) was prepared in a 100 mL
Schlenk flask containing a magnetic stir bar. Anhydrous Me3-
NO (35.6 mg, 0.475 mmol) was then added, and the yellow
solution was stirred for an additional 50 min. At this time, an
IR spectrum of the solution showed νCO bands corresponding
to complex 4. The volatiles (THF and NMe3) were then
removed under vacuum, followed by the addition of more THF
(20 mL) and DBT (94.7 mg, 0.514 mmol) with stirring for an
additional 15 h. The solvent was then removed in vacuo, and
the yellow residue was dissolved in CH2Cl2 (1 mL) and layered
with hexanes (6 mL). Yellow crystals of 1 (128 mg, 34% based
on Re2(CO)10) were grown after 1 day (-20 °C) and isolated
after filtration, washing with hexanes (1 × 5 mL), and drying
in vacuo.
P r ep a r a tion of Re2(CO)7(2,5-Me2T) (3). A hexanes solu-
tion (30 mL) of Re2(CO)10 (204 mg, 0.313 mmol) and 2,5-Me2T
(0.20 mL, 1.76 mmol) was prepared in a quartz reaction tube
containing a magnetic stir bar. The solution was irradiated
with stirring for 15-20 h, at 10 °C under a constant flow of
nitrogen. During this time, the solution turned yellow-orange
and a brown precipitate formed. The solution was then
transferred to a column of silica gel (1 × 8 cm) packed in
hexanes. A yellow band eluted, using a combination of CH2-
Cl2 and hexanes (1:5), and was collected. The volatiles were
removed in vacuo, leaving a yellow-orange oily residue which
was then dissolved in CH2Cl2 (1 mL) and layered with hexanes
(5 mL) followed by slow cooling (-20 °C) until yellow crystals
of 3 formed (1 day). The solution was then filtered, and the
yellow crystals of 3 (20-30 mg, 10-15% yield based on Re2-
(CO)10) were dried under vacuum. 1H NMR (CD2Cl2, 400
MHz): δ 6.49 (d, 1 H, J ) 6.0 Hz), 5.05 (d, 1 H, J ) 5.6 Hz),
2.71 (s, 3 H, Me), 2.31 (s, 3 H, Me). IR (hexanes): νCO 2093
(w), 2039 (s), 1993 (s), 1990 (s), 1963 (m), 1950 (vs) cm-1. Anal.
Calcd for C13H8O7Re2S: C, 22.94; H, 1.18; S, 4.71. Found: C,
22.80; H, 1.20; S, 4.08.
P r ep a r a tion of Re2(CO)9(THF ) (4). Complex 4 was pre-
pared using the following preparation (method A) and a
previously described procedure (method B).24 In method A, a
THF solution (30 mL) of Re2(CO)10 (200 mg, 0.307 mmol) was
added to a quartz photolysis tube containing a magnetic stir
bar. A coldfinger was then inserted into the solution (10 °C),
and an oil bubbler was connected to the tube. The solution
was then irradiated with stirring under N2 until the νCO bands
for 4 reached a maximum (1-1.5 h). The orange-yellow
solution of 4, which contains some residual Re2(CO)10, was then
ready for use in further experiments (see below). Complex 4
Re2(CO)9(η1(S)-2,5-Me2T) (5). Complex 5 was prepared in
a similar manner as for 1 (method A) from a solution of Re2-
(CO)9(THF) (0.222 mmol based on Re2(CO)10). The freshly
prepared solution of Re2(CO)9(THF) and 2,5-Me2T (0.150 mL,
1.32 mmol) was stirred at room temperature for 23 h, during
which time the solution lightened in color. The solvent and
residual 2,5-Me2T were then removed in vacuo, leaving a
yellow-brown oil. Attempts to purify the crude product by
column chromatography on silica gel or neutral alumina were
unsuccessful due to complete decomposition of 5. 1H NMR
1
spectroscopic analyses of crude 5 indicated free 2,5-Me2T. H
NMR (CD2Cl2, 300 MHz): δ 6.71 (s, 2 H), 2.42 (s, 6 H, Me’s).
IR (hexanes): 2101 (w), 2047 (m), 1991 (s), 1987 (s), 1961 (m),
1928 (m) cm-1
.
Cr ysta llogr a p h ic Str u ctu r a l Deter m in a tion s of 1-3.
The single-crystal X-ray diffraction experiments were per-
formed on a Bruker CCD-1000 diffractometer for 1 and 3 and
on a CAD4 diffractometer for 2. The systematic absences in
the diffraction data were consistent for space groups P1 and
P1h for 1 and 3 and for space groups Cc and C2/c for 2. In all
cases the latter centrosymmetric space groups were chosen,
on the basis of the chemically reasonable and computationally
stable results of refinement.25 The structures were solved using
direct methods, completed by subsequent difference Fourier
(23) Fellmann, W.; Kaesz, H. D. Inorg. Nucl. Chem. Lett. 1966, 2,
63.
(24) Bergamo, M.; Beringhelli, T.; D’Alfonso, G.; Mercandelli, P.;
Moret, M.; Sironi, A. Organometallics 1997, 16, 4129.
(25) All software and sources of the scattering factors are contained
in the SHELXTL (version 5.1) program library (G. Sheldrick, Bruker
Analytical X-ray Systems, Madison, WI, 1997).

1074 Organometallics, Vol. 20, No. 6, 2001
Ta ble 1. Cr ysta llogr a p h ic Da ta for 1-3
Reynolds et al.
Sch em e 2
1
2
3
formula
C
21H8O9Re2S‚ C20H8O8Re2S C13H8O7Re2S
1/2CH2Cl2
fw
851.20
780.72
680.65
space group
a, Å
b, Å
P1h
C2/c
P1h
9.4152(4)
11.2600(5)
11.9289(6)
95.972(1)
109.303(1)
93.279(1)
1181.41(9)
2
18.231(2)
17.490(1)
15.677(2)
8.5886(6)
9.2722(6)
12.4626(8)
69.566(1)
87.392(1)
62.591(1)
817.65(9)
2
c, Å
R, deg
â, deg
γ, deg
V, Å3
Z
122.64(1)
4214.2(7)
8
cryst color, habit yellow block
yellow block
2.461
11.62
293(2)
Enraf-Nonius Bruker
yellow block
2.765
14.95
D(calcd), g cm-3 2.393
-1
µ(Mo KR), mm
temp, K
10.49
163(2)
Bruker
163(2)
diffractometer
CCD-1000
CAD4
empirical
1.00/0.53
4321
CCD-1000
abs cor
empirical
0.32/0.16
10 433
empirical
1.00/0.52
9606
T(max)/T(min)
no. of rflns
collected
(CO)5ReCl29a (<10%) and Re2(CO)8(µ-Cl)2
(>90%)
29b
no. of indep rflns 4795 (R(int) ) 3475 (R(int) ) 3338 (R(int) )
0.020) 0.064) 0.031)
R(F) (I g 2σ(I)), 1.66 3.19 2.32
within 1 h, and no evidence for complexes 1 and 2 was
observed. THF reacts with Re2(CO)10 during photolysis
to give Re2(CO)9(THF); however, no Re-DBT-containing
complexes were observed when Re2(CO)10 and DBT were
irradiated in THF (20 h, 10 °C).
Complex 1 is a moderately air-stable solid that is
soluble in CH2Cl2 and benzene but sparingly soluble in
hydrocarbons such as hexanes. The DBT ligand in 1 is
labile and is completely replaced at room temperature
by CO (1 atm) to give Re2(CO)10 in 1,2-DCE solvent in
less than 2 min.
The 1H NMR spectrum of 1 indicates S-binding of
DBT, because the benzo-ring proton signals are shifted
only 0.05-0.12 ppm downfield in comparison to those
of free DBT (δ 8.21-7.47). This small chemical shift is
common for η1(S)-bound DBT complexes such as CpRe-
(CO)2(η1(S)-DBT),30 [CpFe(CO)2(η1(S)-DBT)+],31 Cp*Ir-
(Cl)2(η1(S)-DBT),32 CpMn(CO)2(η1(S)-DBT),33 (CO)5W-
(η1(S)-DBT),)33 and (CpSiMe2Cp)Mo(η1(S)-DBT)34 (Cp )
η5-C5H5, Cp* ) C5Me5).
The X-ray structure of 1 (Figure 1) shows an S-bound
DBT ligand coordinated in the equatorial position to the
Re2(CO)9 unit. The Re(2)-S distance (2.5375(8) Å) is
similar to those observed in the structures of Re2(CO)9-
a
%
Rw(F2), %a
4.24
7.77
5.99
a
Quantity minimized ) Rw(F2) ) ∑[w(Fo2 - Fc2)2]/∑[(wFo2)2]1/2
;
R ) ∑∆/∑(Fo), ∆ ) |(Fo - Fc)|.
synthesis, and refined by full-matrix least-squares procedures.
The empirical absorption corrections for 1 and 3 were applied
by using the program SADABS26 and by using the program
DIFABS27 for 2. All non-hydrogen atoms were refined with
anisotropic displacement coefficients. All hydrogen atoms were
treated as idealized contributions. In the case of 1, there was
a severely disordered solvent molecule also present in the
asymmetric unit. Attempts to identify and refine this molecule
gave models that suggested the molecule was mobile. In
addition, the refinement was computationally unstable. The
option SQUEEZE of the program PLATON28 was used to
correct the diffraction data for diffuse scattering effects and
to identify the solvate molecules. PLATON calculated the
upper limit of volume that can be occupied by the solvent to
be 110.6 Å3, or 9.4% of the unit cell volume. The program
calculated 40 electrons in the unit cell for the diffuse species.
This approximately corresponds to one dichloromethane mol-
ecule in the unit cell (42 electrons). All data in Table 1 reflect
the presence of a half-molecule of dichloromethane per mol-
ecule of complex in the lattice of 1. In the case of 2, the hydride
ligand was restrained to be equidistant from the Re atoms.
(SCH2CMe2CH2) (2.485(4) Å)35 and Re2(CO)9[(SCH2CH2-
CH2)3] (2.498(3) Å)36 but longer than those in Cp*Re-
(CO)2(η1(S)-T)30 (2.360(3) Å) and Cp*Re(CO)2(η1(S)-3-
MeBT)37 (2.356(4) Å) (3-MeBT ) 3-methylbenzothio-
phene), both of which contain the strongly electron-
donating Cp* ligand. The geometry about the bound
DBT sulfur is trigonal pyramidal, as suggested by the
Resu lts a n d Discu ssion
Rea ction of Re2(CO)10 w ith DBT. Ultraviolet ir-
radiation of a stirred hexanes solution (10 °C) containing
Re2(CO)10 and 2 equiv of DBT produced the bi- and
trimetallic Re complexes eq-Re2(CO)9(η1(S)-DBT) (1),
Re2(CO)8(µ-C12H7S)(µ-H) (2), and HRe3(CO)14 after 18-
24 h (Scheme 2). Hexanes was chosen as the reaction
solvent because it is noncoordinating, inert during UV
photolysis, and easy to remove under reduced pressure.
No reaction occurred between Re2(CO)10 and DBT under
similar photolytic conditions in either benzene or diethyl
ether solvents. The photolysis reaction of DBT (2-4-
fold excess) and Re2(CO)10 in CH2Cl2 produced only
(29) (a) Dolcetti, G.; Norton, J . R. Inorg. Synth. 1976, 16, 35. (b)
Wrighton, M. S.; Bredesen, D. J . J . Organomet. Chem. 1973, 50, C35.
(30) Choi, M.-G.; Angelici, R. J . Organometallics 1991, 10, 2436.
(31) Goodrich, J . D.; Nickias, P. N.; Selegue, J . P. Inorg. Chem. 1987,
26, 3426.
(32) Rao, K. M.; Day, C. L.; J acobson, R. A.; Angelici, R. J . Inorg.
Chem. 1991, 30, 5046.
(33) Reynolds, M. A.; Guzei, I. A.; Logsdon, B. C.; Thomas, L. M.;
J acobson, R. A.; Angelici, R. J . Organometallics 1999, 18, 4075.
(34) Churchill, D. G.; Bridgewater, B. M.; Parkin, G. J . Am. Chem.
Soc. 2000, 122, 178.
(35) Adams, R. D.; Perrin, J . L.; Queisser, J . A.; Wolfe, J . B.
Organometallics 1997, 16, 2612.
(36) Adams, R. D.; Falloon, S. B. Organometallics 1995, 14, 1748.
(37) Choi, M.-G.; Angelici, R. J . Organometallics 1992, 11, 3328.
(26) Blessing, R. H. Acta Crystallogr. 1995, A51, 33-38.
(27) Walker, N.; Stuart, D. Acta Crystaogr. 1983, A39, 158.
(28) Spek, A. L. Acta Crystallogr. 1990, A46, C-34.

Re2(CO)10-Mediated C-H and C-S Bond Cleavage
Organometallics, Vol. 20, No. 6, 2001 1075
F igu r e 1. Molecular structure of Re2(CO)9(η1(S)-DBT) (1)
with 50% probability thermal ellipsoids. Selected bond
distances (Å) and angles (deg): Re(1)-Re(2), 3.0389(2);
Re(2)-S, 2.5375(8); S-C(10), 1.781(3); S-C(21), 1.782(3);
Re(1)-C(1), 1.933(4); Re(2)-C(9), 1.926(3); C(9)-Re(2)-S,
174.69(10); Re(2)-S-C(10), 107.08(11); Re(2)-S-C(21),
106.60(11); C(10)-S-C(21), 90.98(16); Re(2)-S-Midpt,
113.6.
F igu r e 2. Molecular structure of Re2(CO)8(µ-C12H7S)-
(µ-H) (2) with 50% probability thermal ellipsoids. Selected
bond distances (Å) and angles (deg): Re(1)-S, 2.475(2);
Re(2)-C(19), 2.230(9); Re(1)- - -Re(2), 3.345; S-C(9), 1.765-
(8); S-C(20), 1.768(7); Re(1)-C(2), 1.996(10); Re(1)-C(3),
1.961(10); Re(1)-C(4), 2.002(10); Re(2)-C(5), 1.981(11);
Re(2)-C(6), 1.962(10); Re(2)-C(7), 1.897(11); Re(2)-C(8),
2.009(12); C(1)-Re(1)-S, 173.5(3); C(3)-Re(1)-S, 94.5(3);
C(6)-Re(2)-C(19), 172.6(4); C(7)-Re(2)-C(19), 87.8(4);
Re(1)-S-C(9), 116.0(3); C(9)-S-C(20), 92.7(4); Re(1)-
S-C(20), 107.4(3); Re(1)-S-Midpt, 124.8.
sum of the angles around the sulfur atom (304.7°). A
second parameter used for determining the sulfur
geometry in metal-thiophene complexes is the tilt angle
(θ). In complex 1, θ is defined as the angle between the
Re-S bond and the vector from the sulfur to the
midpoint of the C(15) and C(16) ring carbon atoms of
the DBT ligand. The tilt angle for 1 (θ ) 113.6°) is
similar to that reported for other η1(S)-bound DBT
complexes which contain relatively electron-deficient
metal centers, including (CO)5W(η1(S)-DBT)33 (118.8°),
[CpFe(CO)2(η1(S)-DBT)+]31 (119.4°), and (CO)5Cr(η1(S)-
DBT)33 (121.8°). Notably, θ is much larger in the Re-
thiophene complexes Cp*Re(CO)2(η1(S)-T) (140.4°)33 and
Cp*Re(CO)2(η1(S)-3-MeBT)37 (131.0°), which contain Re
centers that are more electron rich than in 1.
spectrum differs from that of the DBT ligand in complex
1 (δ 8.20-7.55) and free DBT (8.21-7.47 ppm), both of
which show only three signals (multiplets) for the DBT
ring protons in the aromatic region due to the two
equivalent DBT benzo-rings. A signal for the hydride
ligand was also observed in the spectrum of 2 (δ -14.8)
in a region which is characteristic of a hydride residing
between two Re metal centers, as reported for other
Re-H-Re-containing complexes.39,40
The smaller θ and longer Re-S distance in 1 com-
pared to those parameters in the Cp*Re(CO)2(η1(S)-
T*)33,37 complexes are in agreement with Harris’ pro-
posal,38 which suggests that the more electron-rich the
metal center, the more π-back-donation from the metal
to the thiophene ligand and, hence, the larger the tilt
angle. Related to this proposal is the trend in the Re-S
distances in Cp*Re(CO)2(η1(S)-T)33 (2.360(3) Å) and
Cp*Re(CO)2(η1(S)-3-MeBT)37 (2.356(4) Å), which are
shorter than in 1 (2.5375(8) Å). We previously used
Harris’ argument to explain the smaller tilt angle and
longer metal-sulfur distance observed in [CpFe(CO)2-
(η1(S)-DBT)+] (119.4°, 2.289(1) Å) compared with those
parameters for the more electron-rich CpMn(CO)2-
(η1(S)-DBT) (125.6°, 2.255(1) Å).33
In the molecular structure of 2 (Figure 2) the DBT
ligand is bridging the two Re(CO)4 units through both
the sulfur and the C(19) carbon. Although the hydride
ligand was not observed in this structure, it was
observed in the 1H NMR spectrum. The Re(1)-S dis-
tance (2.475(2) Å) is shorter than that in 1 (2.5375(8)
Å), while the Re-Re distance (3.345 Å) is longer
compared to that in both 1 (3.0389(7) Å) and Re2(CO)10
(3.0413(11) Å),41 possibly due to the bridging nature of
the DBT and hydride ligands. The DBT sulfur is trigonal
pyramidal, on the basis of both the tilt angle (θ ) 124.8°)
and the sum of the angles around the sulfur (316.1°).
Complex 2 is a rare structural example of selective
C-H bond cleavage in DBT at a transition-metal center.
In related studies, J ones et al.42 reported that Cp*Rh-
(PMe3) reacts with alkyl-substituted dibenzothiophenes
to afford complexes such as Cp*Rh(PMe3)(C12H7S)(H),
which exist as several isomers resulting from the
cleavage of a DBT C-H bond. More recently,43 Os3(CO)10-
Complex 2 is a moderately air-stable yellow solid that
is soluble in benzene and CH2Cl2 but sparingly soluble
in hexanes. Crystals of 2, hand-picked from the mixture
of 1 and 2, were characterized by spectroscopy (1H NMR
1
and IR) and by an X-ray diffraction study. The H NMR
spectrum for 2 shows six signals in the aromatic region
with chemical shifts in the range δ 8.06-7.21 for the
ring protons of the C-H-cleaved DBT ligand. This
(39) (a) Egold, H.; Schwarze, D.; Florke, U. J . Chem. Soc., Dalton
Trans. 1999, 3203. (b) Beringhelli, T.; D’Alfonso, G.; Ciani, G.; Moret,
M.; Sironi, A. J . Chem. Soc., Dalton Trans. 1993, 1101.
(40) Adams, R. D.; Chen, L.; Wu, W. Organometallics 1993, 12, 4962.
(41) Churchill, M. R.; Amoh, K. N.; Wasserman, H. J . Inorg. Chem.
1981, 20, 1609.
(38) (a) Harris, S. Polyhedron 1997, 16, 3219. (b) Harris, S. Orga-
nometallics 1994, 13, 2628.
(42) Myers, A. W.; J ones, W. D. Organometallics 1996, 15, 2905.

1076 Organometallics, Vol. 20, No. 6, 2001
Reynolds et al.
Sch em e 3
(CH3CN)2 was reported to react with DBT to give the
triosmium benzyne cluster Os3(CO)9(µ-C12H6S)(µ-H)2, in
which the C-H bonds of the DBT ligand were cleaved
in both the 2- and 3-positions. This complex was
proposed to form by initial DBT C-H bond cleavage in
the 2-position to give the triosmium intermediate Os3-
(CO)10(µ-C12H7S)(µ-H). Although this intermediate was
not detected, its formation is supported by the isolation
and characterization of complex 2 described here.
Rea ction s of Re2(CO)10 w ith Th iop h en es. UV
photolysis of a hexanes solution (10 °C) containing Re2-
(CO)10 and 3-5 equiv of 2,5-Me2T in a quartz photolysis
tube produced the ring-cleaved thiophene complex Re2-
(CO)7(µ-2,5-Me2T) (3) in 15-20% yield after 24 h
(Scheme 3). Complex 3, which is soluble in most organic
solvents such as benzene, CH2Cl2, and hexanes, is stable
for months in air in the solid state but will decompose
in solution unless kept under an inert atmosphere.
F igu r e 3. Molecular structure of Re2(CO)7(2,5-Me2T) (3)
with 30% probability thermal ellipsoids. Selected bond
distances (Å) and angles (deg): Re(1)-S, 2.5048(13);
Re(2)-S, 2.4722(14); Re(1)-C(12), 2.554(5); Re(2)-C(12),
2.206(6); S-C(9), 1.773(5); C(9)-C(10), 1.382(8); C(10)-
C(11), 1.473(7); C(11)-C(12), 1.370(8); Re(1)-S-Re(2),
97.74(5); Re(1)-C(12)-Re(2), 103.7(2); S-Re(2)-C(12),
80.73(14); C(9)-S-Re(1) 62.77(18); C(9)-S-Re(2),
108.40(18).
2-methylthiophene (2-MeT), and 3-methylthiophene (3-
MeT) ligands are also known.14b We have not, however,
been successful in preparing the analogous Re com-
plexes by photolysis of T, 2-MeT, and 3-MeT with Re2-
(CO)10 in hexanes solvent and under conditions similar
to those used in the preparation of 3. This lack of
reactivity may be attributed to the fact that T, 2-MeT,
and 3-MeT are weaker S-donor ligands44 than 2,5-Me2T
and may not form the S-bonded Re2(CO)9(η1(S)-T*)
(T* ) T, 2-MeT, or 3-MeT) complex which are possible
precursors to the C-S-cleaved products. Complex 3 does
not react with electrophiles such as triflic acid and
methyl triflate (MeOSO2CF3). The sulfur in the Mn
analogue is similarly unreactive with these same
electrophiles.14a
P r ep a r a tion a n d Ch a r a cter iza tion of Re2(CO)9-
(THF ) (4). Complex 4 was prepared using two separate
methods (A and B). The first, method A, involves
irradiation of a dry THF solution containing Re2(CO)10
at 10 °C with UV light (1-1.5 h), which produces a
yellow-orange solution characterized as predominately
4 (>50%). The second method (B) was previously
reported24 and involves the chemically induced removal
of a CO ligand from Re2(CO)10 using anhydrous Me3-
NO. The IR spectrum for samples taken from solutions
produced by both methods A and B exhibited νCO bands
at 2101 (w), 2038 (m), 1987 (vs), 1980 (sh), 1952 (m)
and 1912 (m) cm-1, which correspond well with those
previously reported for 4.24 Solvent removal under
reduced pressure from solutions of 4 generated by either
method A or B gave an oily yellow-orange residue that
was characterized by 1H NMR spectroscopy. The 1H
1
Complex 3 has been characterized by IR and H NMR
spectroscopy and elemental analysis and also by an
X-ray diffraction study. Its IR spectrum (in hexanes)
shows six νCO bands at 2093 (w), 2039 (s), 1993 (s), 1990
(s), 1963 (m), and 1950 (vs) cm-1. This spectrum is
similar to that of its Mn analogue Mn2(CO)7(µ-2,5-
Me2T)14 (in hexanes: 2079 (m), 2033 (vs), 1995 (s), 1991
(vs), 1969 (m) and 1958 (vs) cm-1), which was prepared
1
using a very different synthetic method. The H NMR
spectrum of 3 (in CD2Cl2) consists of two doublets at δ
6.49 and 5.05 which correspond to the two inequivalent
protons H(3) and H(4) and two singlets for the inequiva-
lent methyl groups at δ 2.71 and 2.31; these chemical
shifts are similar to those observed in the Mn analogue
Mn2(CO)9(µ-2,5-Me2T).14
In the structure of 3 (Figure 3), a Re(CO)4 fragment
is inserted into a C-S bond of the 2,5-Me2T ring and
bent out of the thiophene plane; the resulting ring-
opened thiophene is also η5-coordinated to the Re(CO)3
unit. The structure of 3 is similar to that reported by
Sweigart14 and co-workers for the 2-MeT derivative Mn2-
(CO)7(µ-2-MeT). The distance between Re atoms (3.749
Å) is out of bonding range compared to those reported
for both Re2(CO)10 (3.0413(11) Å)41 and complex 1
(3.0389(2) Å). The Re(1)-S distance of 2.5048(13) Å and
the Re(2)-S distance of 2.4722(14) Å are nearly the
same.
In addition to the Mn analogue of 3, Mn2(CO)7(µ-2,5-
Me2T), similar complexes with bridging thiophene (T),
(43) Arce, A. J .; Karam, A.; De Sanctis, Y.; Capparelli, M. V.;
Deeming, A. J . Inorg. Chim. Acta 1999, 285, 277.
(44) (a) Benson, J . W.; Angelici, R. J . Organometallics 1992, 11, 922.
(b) Benson, J . W.; Angelici, R. J . Organometallics 1992, 12, 680.

Re2(CO)10-Mediated C-H and C-S Bond Cleavage
Organometallics, Vol. 20, No. 6, 2001 1077
hv
NMR spectrum of this solution revealed two multiplets
(δ 3.86 and 1.89) that were integrated in a 1:1 ratio and
were assigned as the THF adduct complex 4. Addition
of a 5-10-fold excess of THF to the NMR sample,
followed by collection of another 1H NMR spectrum,
revealed a second set of two signals (δ 3.69 and 1.86)
corresponding to free THF adjacent to those observed
for the proposed complex 4. This result indicates that
the signals at δ 3.86 and 1.89 are due to the complexed
THF and not free or rapidly exchanging THF. Further
evidence for the existence of species 4 is given by its
reactivity with thiophenes to produce the S-bound
complexes of type Re2(CO)9(η1(S)-T*) (1 and 5, see
Re2(CO)10
y\
z Re2(CO)9 + CO
(1)
(2)
Re2(CO)9 + L h Re2(CO)9(L)
hv
Re2(CO)10
y\
z 2Re(CO)5
(3)
(4)
(5)
Re(CO)5 + L h Re(CO)4(L) + CO
Re(CO)4(L) + Re(CO)5 h Re2(CO)9(L)
formation of Re2(CO)9(L) and Re2(CO)8(L)2 complexes (L
) phosphines, pyridine)46,47 during UV photolysis of M2-
(CO)10 (M ) Mn, Re) with L donor ligands (eqs 1-5).
Complex 1 is also a Re2(CO)9(L)-type complex (L ) DBT)
and could be formed by either pathway 1 (eqs 1 and 2)
or pathway 2 (eqs 3-5) during UV photolysis of Re2-
(CO)10 and DBT.
1
below). The H NMR spectrum (in CD2Cl2) of crude 4,
prepared using method B, also exhibits signals for 4 as
well as other signals which could be attributed to the
previously reported Re2(CO)9(NMe3)24 (δ 2.89) and un-
reacted Me3NO (δ 3.04).
Syn th esis a n d Ch a r a cter iza tion of Re2(CO)9(η1-
(S)-T*) Com p lexes (1 a n d 5). The S-bound thiophene
complexes Re2(CO)9(η1(S)-T*) (T* ) DBT (1), 2,5-Me2T
(5)) were prepared in low to moderate yields (20-35%)
by addition of DBT (2 equiv) or 2,5-Me2T (3-5 equiv)
to a THF solution of 4 with stirring for 15-20 h.
Unreacted Re2(CO)10, Re2(CO)9(NMe3) (method B only),
and other unidentified species were observed in the
reaction mixtures when the crude products were ana-
lyzed spectroscopically (1H NMR and IR). No reaction
was observed under similar conditions when a 3-5-fold
excess of T, 2-MeT, or 4,6-Me2DBT was added to 4 in
THF solution. These results suggest, as mentioned
earlier, that these thiophenes are poor donor ligands44
and cannot compete with other σ-donors such as NMe3
and THF that are also present in the reaction solutions.
The yields of 1 were similar (27-34%) using both
methods A and B, which implies that DBT is a better
σ-donor ligand under these reaction conditions than
THF and NMe3.
In an effort to understand the mechanism of the
photolysis reaction between Re2(CO)10 and DBT (Scheme
2), a series of experiments were undertaken. When the
UV-light-promoted reaction of Re2(CO)10 and 2-3 equiv
of DBT was carried out (in hexanes, 24 h, 15 °C) under
a CO atmosphere instead of N2, the formation of
complexes 1 and 2 did not not occur and the Re2(CO)10
remained unreacted. This result is consistent with the
CO dissociation mechanism (eqs 1 and 2). It is known48
that the radical scavenger TEMPO (2,2,6,6-tetramethyl-
1-piperidinyloxy free radical) reacts with •M(CO)5
(M ) Mn, Re) radicals to form neutral M-N-O cyclic
complexes of the type (TEMPO)M(CO)3. Therefore, if
Re(CO)5 radicals are intermediates in the photolytic
reaction of Re2(CO)10 and DBT, the formation of complex
1 should be inhibited by TEMPO. When a hexanes
solution of Re2(CO)10 and 3 equiv of DBT was photolyzed
under N2 in the presence of TEMPO (1 equiv based on
Re2(CO)10), complex 2 was afforded after 8 h, as detected
by IR spectroscopy, with no indication for the formation
of 1. However, when the ratio of Re2(CO)10 to TEMPO
was changed to 1:2, which is stoichiometric with respect
to the formation of Re(CO)5 radicals, neither complex 1
nor 2 was observed after 10 h (10-15 °C). On the basis
of the results of these two experiments, it is likely that
Re radical species are also involved in the formation of
1. Thus, a mechanism for the formation of 1 is probably
best represented by pathway 2 (eqs 3-5) in which eq 4
is the step that is inhibited by CO.
If 1 were an intermediate in the formation of 2, then
the formation of 2 would be inhibited as well because
of the inhibiting effect of CO and TEMPO on the
formation of 1. Unfortunately, it was not possible to
determine whether 1 converts into 2 under the pho-
tolytic conditions of the reaction (Scheme 2) because 1
was too insoluble in hexanes to perform the experiment.
However, it seems reasonable that 1 is the precursor to
2 and that this conversion is promoted by UV light, as
1 does not transform to 2 under thermal conditions (in
refluxing toluene). A possible mechanism for the pho-
tolytic conversion of 1 to 2 is shown in Scheme 4. It
involves photolytic loss of CO from the Re(CO)5 unit in
Complex 5 was prepared using method A, from a
stirred THF solution of 4 and 3-5 equiv of 2,5-Me2T.
1
The H NMR spectrum of 5 in CD2Cl2 shows a singlet
(δ 6.71) for the equivalent H(3) and H(4) protons of the
2,5-Me2T ligand; this signal is shifted 0.19 ppm down-
field compared to that in the free ligand (δ 6.52 in CD2-
Cl2). The methyl signal for the two chemically equiva-
lent Me groups in 5 is also a singlet (δ 2.42), which is
shifted only 0.01 ppm from that in the free ligand
(δ 2.41 ppm). Chemical shifts in the range of (0.02-
0.20 ppm are common for the methyl groups in η1(S)-
bound 2,5-Me2T ligands such as those observed in
CpRe(CO)2(η1(S)-2,5-Me2T)30 and (CO)5M(η1(S)-2,5-
Me2T) (M ) Cr, W).33 The S-bound thiophene ligand in
5 is labile and partially dissociates in solution (CD2Cl2,
1
20 °C) at room temperature, on the basis of H NMR
studies.
Mech a n ism s for th e Rea ction s in Sch em es 2 a n d
3. It has been established that Re2(CO)10 is activated
by UV light (1) to lose a CO ligand to generate Re2(CO)9
species (eq 1) or (2) to undergo metal-metal bond
•
homolysis to produce 17-electron Re(CO)5 radical spe-
cies45 (eq 3). Such processes have been proposed for the
(46) Kidd, D. R.; Brown, T. L. J . Am. Chem. Soc. 1978, 100, 4095.
(47) (a) McCullen, S. B.; Brown, T. L. Inorg. Chem. 1981, 20, 3528.
(b) Stiegman, A. E.; Tyler, D. R. Inorg. Chem. 1984, 23, 527.
(48) J aitner, P.; Huber, W.; Huttner, G.; Scheidsteger, O. J . Orga-
nomet. Chem. 1983, 259, C1.
(45) (a) Meyer, T. J .; Caspar, J . V. Chem. Rev., 1985, 85, 187-218.
(b) Geoffroy, G. L.; Wrighton, M. S. Organometallic Photochemistry;
Academic Press: New York, 1979; pp 136-141. (c) Fox, A.; Poe, A. J .
Am. Chem. Soc. 1980, 102, 2497.

1078 Organometallics, Vol. 20, No. 6, 2001
Reynolds et al.
Sch em e 4
1 to give an intermediate in which there is an agostic
interaction between the C-H bond in the 4-position of
the DBT ligand with the Re(CO)4 unit (step 2). In step
3, oxidative addition of the DBT C-H bond leads to the
product 2. Although the mechanism proposed in Scheme
4 does account for the products (1 and 2) observed
during the photolytic reaction, it is not possible to
exclude mechanisms in which Re2(CO)10 is converted to
2 without involving 1 as an intermediate.
The formation (Scheme 3) of Re2(CO)7(µ-2,5-Me2T) (3)
could occur in a fashion similar to that proposed for 2,
from the S-bound complex Re2(CO)9(η1(S)-2,5-Me2T) (5).
In this mechanism, complex 5 undergoes photolytic loss
of CO from the Re(CO)5 unit, followed by η2-binding of
a CdC double bond of the 2,5-Me2T ligand to give an
intermediate of the type (CO)4Re(µ-η1(S),η2-2,5-Me2T)-
Re(CO)4 (A). This intermediate (A) then undergoes
scission of the Re-Re bond, oxidative addition of a C-S
bond, and loss of a CO to give product 3. Attempts to
observe the photolytic conversion of 5 to 3 were unsuc-
cessful due to the instability of 5.
conditions the reaction of Re2(CO)10 with benzothiophene
(BT) did not produce S-bound or C-H-cleaved complexes
but, rather, Re2(CO)7(µ-BT) in which the Cvinyl-S bond
was cleaved (Scheme 1).21 This difference in reactivity
between BT and DBT suggests that Cvinyl-S cleavage
is more facile than Caryl-S cleavage; moreover, the DBT
reaction indicates that C-H cleavage is more favorable
than Caryl-S cleavage. Irradiation of Re2(CO)10 and
excess 2,5-Me2T with UV light produces Re2(CO)7(µ-2,5-
Me2T) (3), in which the C-S bond of the 2,5-Me2T ligand
is cleaved. However, unlike Re2(CO)7(µ-BT), there is no
Re-Re bond in 3. It is apparently the coordination of
the additional olefin in 2,5-Me2T, as compared with BT,
that prevents the formation of a Re-Re bond in 3.
Although mechanistic pathways for the photolytic reac-
tions between Re2(CO)10 and the thiophenes are not
clear, the formation of the C-H-cleaved DBT product
2 and the C-S-cleaved 2,5-Me2T product 3 represents
a new type of reactivity not previously reported for
dinuclear metal complexes. The S-bound complexes Re2-
(CO)9(η1(S)-T*) were prepared for T* ) DBT (1), 2,5-
Me2T (5) by substituting the THF ligand in the reactive
species Re2(CO)9(THF) (4). On the basis of the smaller
tilt angle (θ) and longer Re-S distance in 1 (113.6°;
2.5375(8) Å) compared to those parameters in Cp*Re-
(CO)2(η1(S)-T) (140.4°; 2.360(3) Å)30 and Cp*Re(CO)2-
(η1(S)-3-MeBT)37 (131.0°; 2.356(4) Å), we conclude that
π-back-bonding from Re to the DBT ligand in 1 is much
less important than in the more electron rich Re
complexes Cp*Re(CO)2(η1(S)-T)30 and Cp*Re(CO)2(η1(S)-
3-MeBT).37
Because of the very different route used for the
synthesis of the Mn analogue of 3, Mn2(CO)7(µ-2,5-
Me2T), it is not surprising that the proposed mechanism
for the formation of 3 differs significantly from that
proposed14,15 by Sweigart for the formation of Mn2(CO)7-
(µ-2,5-Me2T). This complex was prepared by the chemi-
cal reduction of [(η5-2,5-Me2T)Mn(CO)3]+ with cobal-
tocene. The proposed mechanism involves electron
transfer to [(η5-2,5-Me2T)Mn(CO)3]+ to generate a (η4-
-
2,5-Me2T)Mn(CO)3 species which then attacks the
thiophene of a second molecule of [(η5-2,5-Me2T)Mn-
(CO)3]+ with subsequent formation of a bimetallic
species which then converts into Mn2(CO)7(µ-2,5-Me2T).
Ack n ow led gm en t. 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.
Con clu sion
In summary, UV photolysis is a useful method for
cleaving C-H or C-S bonds in thiophenes in the
presence of Re2(CO)10. UV photolysis of Re2(CO)10 with
excess DBT in hexanes solvent produces both the
S-bound DBT complex Re2(CO)9(η1(S)-DBT) (1) and the
C-H-cleaved DBT complex Re2(CO)8(µ-C12H7S)(µ-H) (2).
These products were not expected, since under similar
Su p p or tin g In for m a tion Ava ila ble: Tables giving crys-
tallographic data for 1-3, including atomic coordinates, bond
lengths and angles, anisotropic displacement parameters, and
hydrogen coordinates. This material is available free of charge
via the Internet at http://pubs.acs.org.
OM000708M