Bond cleavage reactions in substituted thiophenes by a rhodium complex

Article abstract of DOI:10.1016/j.ica.2007.09.023

The reactions of (C₅Me₅)Rh(PMe₃)(Ph)H with 2-methoxythiophene, 3-methoxythiophene, 2-cyanothiophene, 3-cyanothiophene, 2-trimethylsilylthiophene, ethylenesulfide, trimethylenesulfide, and several polymethylthiophenes have been investigated. These thiophene derivatives give C-S and in one case C-H insertion products. Ethylene sulfide and trimethylene sulfide undergo ring opening or desulfurization.

Full text of DOI:10.1016/j.ica.2007.09.023

Available online at www.sciencedirect.com
Inorganica Chimica Acta 361 (2008) 3263–3270
www.elsevier.com/locate/ica
Bond cleavage reactions in substituted thiophenes
by a rhodium complex
Andrew W. Myers, Lingzhen Dong, Tulay A. Ateßsin, Roger Skugrud,
¨
*
Christine Flaschenriem, William D. Jones
Department of Chemistry, University of Rochester, Rochester, NY 14627, United States
Received 11 September 2007; accepted 17 September 2007
Available online 21 September 2007
Dedicated to Professor Robert J. Angelici.
Abstract
The reactions of (C5Me5)Rh(PMe3)(Ph)H with 2-methoxythiophene, 3-methoxythiophene, 2-cyanothiophene, 3-cyanothiophene, 2-
trimethylsilylthiophene, ethylenesulfide, trimethylenesulfide, and several polymethylthiophenes have been investigated. These thiophene
derivatives give C–S and in one case C–H insertion products. Ethylene sulfide and trimethylene sulfide undergo ring opening or
desulfurization.
Ó 2007 Elsevier B.V. All rights reserved.
Keywords: Hydrodesulfurization; C–S activation; C–H activation
1. Introduction
[5]. Angelici found that benzothiophene coordinates to
[(C5Me5)Re(CO)2] to form an equilibrium between
g2-C,C and g1-S bound complexes, although no C–H or
C–S cleavage was observed [6]. A few reported investiga-
tions with selenophene heterocycles and homogeneous
transition metal complexes have focused on coordination
modes of the selenophene ring. Choi and Angelici investi-
gated substituted selenophenes with [Cp0(CO)2Re]
(Cp0 = C5H5 or C5Me5) and discovered an equilibrium
between g2-C,C and g1-(Se) bound selenophenes, similar
to results found with substituted thiophenes [7,8]. Studies
by Sanger and Angelici investigated hindered rotation of
g5-coordinated thiophenes and selenophenes with chro-
mium and manganese tricarbonyl complexes [9]. A study
by Bianchini focused on the effects of functionalized thi-
ophenes in reactions with [(triphos)RhH] [10].
The hydrodesulfurization of petroleum is an important
part of the processing of petroleum feedstocks to remove
sulfur [1]. Derivatives of thiophenes, benzothiophenes,
and dibenzothiophenes have been identified as some of
the more difficult species to desulfurize [2]. Over the past
15 years, homogeneous transition metal systems have been
used to provide structural and mechanistic models for key
steps that may occur with the related heterogeneous cata-
lysts [3]. The majority of these model studies have been per-
formed with the parent thiophene/benzothiophene/
dibenzothiophene substrates and their alkylated
derivatives.
In these studies a variety of modes of interaction of
small heterocycles with transition metal fragments has been
elucidated. The fragment [(C5Me5)Rh(PMe3)] reacts with
furan to give C–H activation products at the a-carbon
[4], whereas thiophene formed a six-membered metallacycle
As a continuation of our earlier experiments on the reac-
tions of the [(C5Me5)Rh(PMe3)] fragment with thiophenes
[5], several substituted thiophenes were examined which
possess a variety of electronic and steric substituents. Elec-
tron donating and withdrawing groups were examined with
methoxythiophenes and cyanothiophenes, respectively.
*
Corresponding author. Tel.: +1 585 275 5493; fax: +1 585 276 0011.
E-mail address: jones@chem.rochester.edu (W.D. Jones).
0020-1693/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2007.09.023

3264
A.W. Myers et al. / Inorganica Chimica Acta 361 (2008) 3263–3270
The effects of substituents and their position on the thio-
phene ring will be discussed in light of previous regioselec-
tivities. We had also observed an S-bound thiophene when
tetramethylthiophene was reacted with (C5Me5)Rh(P-
Me3)PhH [11]. Further heating did not result in C–S inser-
tion, but instead to decomposition. Insertion into the
carbon sulfur bond had been observed with 2,5-dimethyl-
thiophene, so a ‘‘buttressing effect’’ of methyl substituents
in the 2 and 3 or 3 and 4 positions was proposed. To inves-
tigate further the effect of methyl substitution, several poly-
methyl-thiophenes were synthesized and their reactivity
examined.
Removal of solvent and excess 3-methoxythiophene fol-
lowed by addition of dry C6D6 gave a dark red solution
of (C5Me5)Rh[j2-C,S–SCH@CH–C(OMe)@CH]. X-ray
quality single crystals were grown by slow evaporation of
1
solvent at ꢀ20 °C in the glove box. H NMR (C6D6): d
1.006 (dd, J = 10.3, 0.7 Hz, 9H), 1.481 (d, J = 2.8 Hz,
15H), 3.647 (s, 3 H), 5.162 (dt, J = 8.3, 3.0 Hz, 1H, H-2),
6.184 (dd, J = 10.0, 2.8 Hz, 1H, H-4), 6.353 (dt, J = 10.0,
3.8 Hz, 1H, H-5). 31P NMR (C6D6):
d 11.41 (d,
J = 155.6 Hz). 13C NMR (C6D6): d 8.88 (s, C5Me5), 14.22
(d, J = 35.4 Hz, PMe3), 50.29 (s, OCH3), 98.0 (s, C5Me5),
122.94 (s,C–H4), 129.00 (s, C–H5), 165.8 (dd, J = 34.9,
25.2 Hz), Rh–C–H2, 166.0 (s, Rh–C–OCH3). Anal. Calc.
for C18H30OSPRh: C, 50.47; H, 7.06. Found: C, 50.67; H
6.92%.
2. Experimental
2.1. General procedures
2.4. Reaction of 1 with 2-cyanothiophene
All manipulations were carried out under an N2 atmo-
sphere or on a high-vacuum line using Schlenk techniques.
All solvents were distilled from dark purple solutions of
sodium benzophenone ketyl under a nitrogen atmosphere.
Reagent grade 2-methoxythiophene, 3-methoxythiophene,
2-cyanothiophene, 3-cyanothiophene, and 2-(trimethyl-
silyl)thiophene were purchased from Aldrich Chemical
Company and were used without further purification,
although each liquid was freeze–pump–thaw degassed
(three cycles) prior to use. 2,3,4-Trimethylthiophene,
2,3,5-trimethylthiophene, and 3,4-dimethylthiophene were
prepared according to literature methods [12]. Methylated
thiophenes were purified on a Varian Aerograph Model
90P GC with an 8 ft prep column packed with a 10% SE-
30 fluid phase on ChromosorbW acid washed 60/80 mesh
support. 1H (400 MHz), 31P (162 MHz) and 13C
(100 MHz) NMR spectra were recorded on a Bruker
AMX-400 and Avance 400 spectrometers.
Thermolysis of 1 (10 mg, 0.021 mmol) with 2-cyanothi-
ophene ( 4.6 lL, 0.05 mmol) in dry C6D12 at 67 °C for
15.5 h gave a single organometallic product. The product
was formulated as insertion into the S–C(CN) bond by a
13C JMOD experiment. X-ray quality single crystals were
grown by slow evaporation of solvent at ꢀ20 °C in the
glove box. 1H NMR (C6D12): d 1.035 (dd, J = 10.6,
0.8 Hz, 9H), 1.507 (d, J = 2.9 Hz, 15H), 6.152 (dd,
J = 5.0, 3.8 Hz, 1H, H-4), 6.434 (dd, J = 5.1, 1.2 Hz, 1H,
H-5), 6.732 (dd, J = 3.8, 1.2 Hz, 1H, H-3). 31P NMR
(C6D12): d 9.014 (d, J = 149.4 Hz). 13C NMR (acetone-
d6): d 8.40 (s, C5Me5), 14.10 (d, J = 35.0 Hz, PMe3),
100.55 (br s, C5Me5), 114.30(s, CN), 122.15 (s, CH),
128.70 (dd, J = 35.0, 24.0 Hz, Rh–C(CN)), 133.40 (s,C–
H), 139.20 (s, C–H). Anal. Calc. for C18H27NSPRh: C,
51.07; H, 6.43; N, 3.31. Found: C, 51.71; H, 5.99; N, 2.83%.
2.5. Reaction of 1 with 3-cyanothiophene
2.2. Reaction of 1 with 2-methoxythiophene
A C6D12 solution of 1 (10 mg, 0.021 mmol) was heated
at 67 °C for 5.5 h with 3-cyanothiophene (5 lL,
0.05 mmol). Two C–S inserted products were observed.
Further heating for 10 additional hours resulted in the dis-
appearance of 1 and increase of these two products, main-
taining the same 1:1 ratio. X-ray quality single crystals
were grown by slow evaporation of solvent at ꢀ20 °C in
the glove box. 1H NMR ((CD3)2CO): d 1.312 (d, J =
10.6, 9H), 1.522 (d, J = 11.4, 9H), 1.636 (d, J = 2.8 Hz,
15H), 1.753 (d, J = 2.7 Hz, 15H), 5.526 (d, J = 9.9 Hz,
1H), 5.613 (s, 1H), 5.783 (s, 1H), 6.182 (dd, J = 10.5,
1.5 Hz, 1H), 6.393 (ddd, J = 9.8, 3.0, 0.8 Hz, 2H). 31P
NMR ((CD3)2CO): d 6.23 (d, J = 151.9 Hz), 5.81 (d,
J = 146.0 Hz). Anal. Calc. for C18H27NSPRh: C, 51.07;
H, 6.43; N, 3.31. Found: C, 50.62; H, 6.33; N, 3.37%.
2-Methoxythiophene (0.010 mL, 0.099 mmol) was added
to an ampule containing 23 mg of 1 (0.059 mmol) in dry
hexane (5 mL). The mixture was stirred at 64 °C for 48 h.
After cooling, evaporation to dryness gave (C5Me5)-
Rh[j2-C,S–SCH@CH–CH@C(OMe)] as a dark red solid.
1H NMR (C6D6): d 1.146 (d, J = 10.5 Hz, 9H), 1.642 (d,
J = 2.6 Hz, 15H), 3.382 (s, 3H), 5.515 (dd, J = 7.4,
3.3 Hz, 1H), 5.865 (td, J = 9.7, 3.5 Hz, 1H), 6.055 (dd,
J = 9.5, 7.3 Hz, 1H). 31P NMR (C6D6): d 11.58 (d,
J = 158.2 Hz). 13C NMR (C6D6): d 9.21 (s, C5Me5), 15.33
(d, J = 33.8 Hz, PMe3), 57.34 (s, OCH3), 97.67 (s, CH),
99.78 (t, J = 3.9 Hz, C5Me5), 115.37 (s, CH), 121.10 (s,
CH), 176.91 (dd, J = 37.5, 20.7 Hz, RhC).
2.3. Reaction of 1 with 3-methoxythiophene
2.6. Reaction of 1 with 2-trimethylsilylthiophene
3-Methoxythiophene (4.2 lL, 0.042 mmol) and
1
(10 mg, 0.021 mmol) were heated in dry C6D12 in a reseal-
2-Trimethylsilylthiophene (0.20 mL, 1.21 mmol) was
able NMR tube with Teflon adapter at 63 °C for 23.5 h.
added to a hexane solution of 1 (30 mg, 0.076 mmol). The

A.W. Myers et al. / Inorganica Chimica Acta 361 (2008) 3263–3270
3265
mixture was stirred at 70 °C for 21 h during which time a
color change to red was observed. The solvent and the
excess 2-trimethylsilylthiophene were removed under vac-
(dd, J = 9.6, 0.8 Hz, 9H), 1.696 (d, J = 2.6 Hz, 15H),
1.890 (s, 3H), 1.980 (s, 3H), 2.080 (s, 3H), 5.830 (br s,
1H). 31P NMR (C6D12): d 5.43 (d, J = 168.0 Hz). Minor
1
1
uum. H and 31P NMR spectroscopic analysis showed a
product: H NMR (C6D12): d 1.285 (dd, J = 7.1, 0.8 Hz,
3:2 mixture of a C–S insertion product and a C–H activa-
tion product. Upon heating the mixed products in C6D12
at 70 °C for 9 days, the ratio of the two complexes begins
to change with the C–H activation product decreasing,
the C–S insertion product increasing and some
(C5Me5)Rh(PMe3)2 decomposition product forming. For
(C5Me5)Rh[j2-C,S–SC(SiMe3)@CH–CH@CH], 1H NMR
(C6D12): d 0.138 (s, 9H), 1.208 (dd, J = 10.2, 0.6 Hz, 9H),
1.607 (d, J = 2.7 Hz, 15H), 6.102 (d, J = 6.7 Hz, 1H),
6.429 (dt, J = 7.6, 3.0 Hz, 1H), 6.744 (ddt, J = 8.6, 3.7,
0.9 Hz, 1H). 31P NMR (C6D12): d 10.32 (d, J = 162.2 Hz).
13C NMR (C6D12): d ꢀ0.65 (s, SiMe3), 8.98 (d, J =
2.7 Hz, C5Me5), 14.91 (d, J = 34.8 Hz, PMe3), 99.62 (t,
J = 3.8 Hz, C5Me5), 125.90 (s, CH), 129.09 (s, CH),
133.69 (s, C), 141.03 (dd, J = 31.0, 22.9 Hz, RhCH). For
9H), 1.679 (d, J = 2.7 Hz, 15H), 2.155 (s, 3H), 2.710 (s,
3H), 3.140 (s, 3H), 5.380 (br s, 1H). 31P NMR (C6D12): d
6.98 (d, J = 167.0 Hz).
2.9. Reaction of (C5Me5)Rh(PMe3)(3,5-xylyl)H with 3,
4-dimethylthiophene
Reaction of a C6D12 solution of (C5Me5)Rh(PMe3)(3,5-
xylyl)H (15 mg, 0.036 mmol) with 3,4-dimethylthiophene
(16 mg, 1.3 mmol) at 47 °C for 6.5 h led to a 1.5:1 ratio
of a C–H activation product to a C–S insertion product.
Further heating for 6 h resulted in a decrease in the C–H
activation product. The C–S insertion product increased,
as did the decomposition product. C–S insertion product:
1H NMR (C6D12): d 1.180 (d, J = 9.0 Hz, 9 H), 1.597 (d,
J = 3.5 Hz, 15H), 1.894 (s, 3H), 1.959 (s, 3H), 5.450 (br s,
1H), the other proton resonance was obscured. 31P NMR
(C6D12): d 10.41 (d, J = 163.0 Hz). C–H activation product:
1H NMR (C6D12): d 1.790 (d, J = 3.4 Hz, 15H), 2.10 (s,
3H). 31P NMR (C6D12): d 9.96 (d, J = 154.0 Hz). The
remaining resonances for the C–H activation product were
obscured by decomposition products.
1
(C5Me5)Rh(PMe3)[5-(2-trimethylsilyl)-thienyl]H, H NMR
(C6D12): d ꢀ12.791 (dd, J = 47.7, 29.2 Hz, 1H), 0.193 (s,
9H), 1.145 (dd, J = 9.9, 0.8 Hz, 9H), 1.817 (dd, J = 2.2,
0.8 Hz, 15H), 6.593 (d, J = 3.1 Hz, 1H), 6.984 (d,
J = 3.0 Hz, 1H). 31P NMR (C6D12): d 8.58 (d, J =
148.9 Hz). 13C NMR (C6D12): d 0.65 (s, SiMe3), 10.08 (d,
J = 3.3 Hz, C5Me5), 19.24 (d, J = 32.6 Hz, PMe3), 98.23
(t, J = 3.3 Hz, C5Me5), 134.54 (s, CH), 134.97 (d, J =
7.0 Hz, CH), 141.40 (s, C), 154.93 (dd, J = 32.1, 20.9 Hz,
RhC).
2.10. Competition reaction of 1 with thiophene/2,5-
dimethylthiophene
2.7. Reaction of (C5Me5)Rh(PMe3)(3,5-xylyl)H with
2,3,4-trimethylthiophene
About 0.20 g (0.051 mmol) 1 was dissolved in 4 mL
hexane. 8.0 lL of thiophene (0.100 mmol) and 12 lL of
2,5-dimethylthiophene (0.105 mmol) were added to this
solution separately via syringe. The reaction mixture was
stirred in a glass ampule fitted with a Teflon stopcock for
20 h at 60 °C. After cooling, the solvent was removed
and the dark red solid was dissolved in 0.5 mL C6D6.
The C–S insertion products from thiophene and 2,5-dim-
A dry C6D12 solution of (C5Me5)Rh(PMe3)(3,5-xylyl)H
(10 mg, 0.024 mmol) was heated at 47 °C with 2,3,4-trim-
ethylthiophene (7 mg, 0.05 mmol) for 1 h. Two products
were observed, one from activation of C–H5 and the other
insertion into a C–S bond. Further heating resulted in
decomposition. C–H activation product: 1H NMR
(C6D12): d ꢀ12.722 (dd, J = 48.3, 28.3 Hz, 1H), 1.152 (d,
J = 10.2 Hz, 9H), 1.834 (d, J = 2.5 Hz, 15H), 1.929 (s,
3H), 2.086 (s, 3H), 2.198 (s, 3H). 31P NMR (C6D12): d
1
ethylthiophene were observed in a 2:1 ratio by H NMR
spectroscopy.
2.11. Reaction of 1 with ethylene sulfide
1
10.00 (d, J = 153.1 Hz). C–S insertion product: H NMR
(C6D12):
d
1.323 (d, J = 10.8 Hz, 9H), 1.920 (d,
Ethylene sulfide (0.015 mL, 0.266 mmol) was added by
syringe to a C6D12 (0.5 mL) solution of 1 (20 mg,
0.051 mmol) in a resealable NMR tube equipped with a
Teflon valve. A yellow solid precipitated immediately. Fol-
lowing reaction at room temperature for 24 h, the sample
was examined by NMR spectroscopy. A 31P NMR spec-
trum showed that only a single organometallic complex
was formed at d 6.72 (d, J = 156.2 Hz). The 1H NMR spec-
trum showed a proton resonance at d ꢀ2.876 (d,
J = 3.4 Hz). The compound was identified as (C5Me5)-
Rh(PMe3)(C6H5)(SH) by comparison of its 1H NMR spec-
trum with that of an authentic sample. Free ethylene (d
5.265) and sulfur precipitate were also generated in this
reaction. For (C5Me5)Rh(PMe3)(C6H5)(SH), 1H NMR
J = 2.5 Hz, 15H), 1.920 (s, 3H), 2.095 (s, 3H), 2.297 (s,
3H), 6.875 (s, 1H). 31P NMR (C6D12): d 5.50 (d,
J = 154.8 Hz).
2.8. Reaction of (C5Me5)Rh(PMe3)(3,5-xylyl)H with
2,3,5-trimethylthiophene
Thermolysis of (C5Me5)Rh(PMe3)(3,5-xylyl)H (10 mg,
0.024 mmol)
0.2 mmol) at 47 °C for 3.5 h gave two products in a 4:1
ratio, assigned as C–S insertion isomers. The major isomer
was not identified as both products decomposed upon fur-
and
2,3,5-trimethylthiophene
(4 mg,
1
ther heating. Major product: H NMR (C6D12): d 1.319

3266
A.W. Myers et al. / Inorganica Chimica Acta 361 (2008) 3263–3270
(C6D12):
d
ꢀ2.876 (d, J = 3.4 Hz, 1H), 1.280 (d,
electrophilic carbene complexes, and such a factor could
be invoked in this case [13,14]. A similar methoxy car-
bene resonance form could stabilize the monomer and
lead to cleavage of the S–C(OMe) bond, although this
would formally invoke a 20eꢀ species. Previous experi-
ments with (C5Me5)Rh(C2H4) and 2-methoxythiophene
showed similar selective insertion behavior by cleaving
the C–S bond on the side of the substituent forming a
(C5Me5)Rh[j2-C,S–SCH@CH–CH@C(OMe)] 16 electron
monomer, which was in equilibrium with cis- and trans-
dimers [15].
J = 10.3 Hz, 9H), 1.574 (s, 15 H), 6.709 (m, 1 H), 6.752
(m, 2H), 7.183 (br s, 2H).
2.12. Catalytic reaction of 10 with ethylene sulfide
Ethylene sulfide (0.030 mL, 0.500 mmol) was added by
syringe to a C6D12 (0.5 mL) solution of 10 (21.2 mg,
0.050 mmol) in a resealable NMR tube equipped with a
Teflon valve. A yellow solid precipitated immediately.
The reaction went to completion at room temperature in
several days.
S
OMe
OMe
2.13. Reaction of 1 with trimethylene sulfide
OMe
Rh
Rh
Rh
H
Me3P
Me3P
ð1Þ
Me3P
60ºC
S
S
2
1
A cyclohexane solution of 1 (30 mg, 0.076 mmol) was
treated with trimethylene sulfide (0.011 mL, 0.150 mmol).
The solution was stirred for 3 h at room temperature. After
removal of the solvent the residue was analyzed by NMR
spectroscopy. A 31P NMR spectrum revealed the presence
of a single organometallic compound with a resonance at d
8.45 (d, J = 160.3 Hz). This product was assigned as
Thermolysis of 1 with 3-methoxythiophene also gave a sin-
gle C–S insertion product 3 identified by 1H and 31P NMR
spectroscopies. A 1H{31P} experiment which revealed large
(JP–H = 8.4 Hz) and small (JP–H = 3.8 Hz) phosphorus
couplings to H-2 and H-5, respectively. Proton resonances
of the inserted thiophene were identified through a series of
homonuclear decoupling experiments. A large cis- coupling
(JH–H = 10.0 Hz) was observed between H-5 (d 6.4) and H-
4 (d 6.2) while only a small w-coupling (JH–H = 3.0 Hz) was
seen between H-2 (d 5.2) and H-4. A 13C NMR spectrum
showed a doublet of doublets resonance characteristic of
the carbon bound to rhodium (d 165.8, dd, J = 34.9,
(C5Me5)Rh(Ph)(SCH2CH2CH3). 1H NMR (C6D6):
d
0.987 (t, J = 7.3 Hz, 3H), 1.314 (d, J = 10.0 Hz, 9H),
1.577 (d, J = 2.7 Hz, 15H), 1.616 (m, 2H), 2.168 (t,
J = 14.6 Hz, 2H), 6.751 (m, 1H), 6.805 (t, J = 6.8 Hz,
2H), 7.247 (d, J = 7.6 Hz, 2H). Further stirring the reac-
tion mixture at room temperature for several days resulted
in the formation of 10.
25.2 Hz), and a H–13C heteronuclear correlation NMR
1
3. Results and discussion
experiment showed this carbon to be attached to H-2. This
data is most consistent with insertion toward the methoxy
substituent, (C5Me5)Rh(PMe3)[j2-C,S–SCH@CH–C(OMe)@
CH] (Eq. (2)). As proof of this geometry, a single crystal X-
ray structure determination was made of 3. As shown in
Fig. 1, the indicated position of the methoxy group is
confirmed.
3.1. Reaction of (C5Me5)Rh(PMe3)(Ph)H with
methoxythiophenes
The thermal reaction (60 °C) of (C5Me5)Rh(P-
Me3)(Ph)H, 1, with 2-methoxythiophene in hexane solution
leads to the selective formation of a single C–S insertion
product
assigned
as
(C5Me5)Rh(PMe3)[j2-C,S–
SCH@CH–CH@C(OMe)], 2. Evidence for this formula-
tion in which the metal has inserted into the C–S bond
on the more hindered side of the ring comes from examina-
tion of a 13C JMOD (attached proton test) experiment. The
carbon bound to rhodium is easily identified by its charac-
1
teristic JRh–C and JP–C couplings (shown by 31P and H
decoupling). The 13C JMOD experiment revealed the reso-
nance at d 176.91 (dd, J = 37.5, 20.7 Hz) as having oppo-
site phase as the CH3- and C–H resonances. This
information identifies the carbon bound to rhodium as
being C2–OMe, i.e., with no attached hydrogens.1
The insertion into the substituted side of 2-methoxythi-
ophene in 2 can be rationalized in terms of a Fischer-car-
bene stabilized resonance form, shown in Eq. (1).
Heteroatom substituents are well known to stabilize
Rh
S
C2
C3
Fig. 1. ORTEP drawing of 3. Ellipsoids are shown at the 30% probability
level.
1
C5
The numbering scheme in the inserted complexes follows:
.
C4

A.W. Myers et al. / Inorganica Chimica Acta 361 (2008) 3263–3270
3267
mediate has been envisioned as a nucleophilic attack of the
electron rich rhodium on an a-carbon. This insertion could
be driven to the substituted side by the removal of electron
density by the cyano group, as was proposed in reactions of
[(C5Me5)Rh(PMe3)] with 2-cyanodibenzothiophene and 2-
trifluoromethyldibenzothiophene [16].
S
Rh
Rh
OMe
60ºC
Me3P
H
Me3P
ð2Þ
OMe
S
3
1
Previous results with the initial product from the reac-
tion of 2-methylbenzothiophene with [(C5Me5)Rh(PMe3)]
in which rhodium had inserted into the S-vinyl bond
showed phosphorus coupling to the methyl hydrogens,
i.e. the group bound to the rhodium a-carbon. The larger
phosphorus coupling to H-2 in the 3-methoxythiophene
complex suggests that rhodium has inserted between S
and C-2. This insertion selectivity was confirmed by the
heteronuclear correlation experiment.
Contrary to our observations for both 2-methoxy- and
3-methoxythiophene is a report by Bianchini and co-work-
ers who reported C–S insertion toward the unsubstituted
side in both molecules with [(triphos)RhH] [10]. These
products, however, are hindered by the bulky triphos
ligand, and selectivities reflect steric constraints which are
absent in our [(C5Me5)Rh(PMe3)] system. In addition, the
products eliminate a Rh–vinyl bond to generate a p-vinyl
ligand, which may affect the thermodynamics of the cleav-
age reaction.
Thermolysis of 1 with 3-cyanothiophene gave two C–S
inserted products in a 1:1 ratio, 5a and 5b (Eq. (4)). The
reaction was followed periodically over 15 h at 67 °C and
no change was observed in the 1:1 ratio. The electron with-
drawing group in the 3 position does not appear to influ-
ence selectivities.
S
+
Rh
CN
67ºC
Rh
Rh
Me3P
Me3P
H
Me3P
CN
S
S
5b
5a
1
CN
1 : 1
ð4Þ
The proposed structures for 4 and 5a were confirmed by
single crystal X-ray diffraction (Fig. 2a and b). In both
molecules, the thiophene ring is nearly planar. In 4, the
S1-Rh1-C1 plane is at an angle of 14.3° to the S1–C4–
C3–C2–C1 plane. In contrast, the related dimethyl substi-
tuted derivative (C5Me5)Rh(PMe3)(j2-C,S–SCMe@CH–
CH@CMe) is more strongly puckered (26°) due to steric
interactions [5]. The structure of 5a shows disorder in the
possible orientation of the ring-opened thiophene, but a
reasonable model can be obtained using the Shelx SAME
instruction to constrain the two rings to have similar
geometries.
3.2. Reaction of (C5Me5)Rh(PMe3)(Ph)H with
cyanothiophenes
The reactions with 2- and 3-methoxythiophene sug-
gested that p-donating groups on the thiophene ring direct
insertion toward the side of the substituent. To examine an
electron withdrawing substituent, 2- and 3-cyanothiophene
were reacted with 1. A C6D12 solution of 1 was heated with
2-cyanothiophene at 67 °C for 15.5 h. A C–H activation
product, present at early reaction times, was observed to
go away with formation of a single C–S insertion product
4. A 13C JMOD experiment in acetone-d6 showed the dou-
blet of doublets resonance for the Rh–C with the same
phase as the C5Me5 ring carbons, indicating that rhodium
had inserted into the substituted C–S bond (Eq. (3)). This
assignment was confirmed by the absence of any phospho-
rus coupling to thiophene ring protons. No change was
seen in the coupling patterns of thiophene ring protons of
3.3. Reaction of (C5Me5)Rh(PMe3)(Ph)H with
trimethylsilylthiophene
Thermolysis of 1 with 2-(trimethylsilyl)thiophene at
70 °C for 21 h in dry C6D12 gave two organometallic com-
plexes 6a and 6b in a 3:2 ratio. The major product (d 10.32)
had a coupling constant characteristic of a C–S insertion
1
the product during a H{31P} experiment, consistent with
no hydrogens attached to a-carbon atoms.
S
CN
CN
Rh
Rh
Me3P
H
Me3P
ð3Þ
67ºC
S
4
1
The electron withdrawing cyano substituent would not
stabilize a Fischer-carbene resonance form as in the case
of 2-methoxythiophene. A strongly electron-withdrawing
substituent could serve to reduce electron density at the
a-carbon. Insertion of the rhodium from an S-bound inter-
Fig. 2. (a) ORTEP drawing of 4. Ellipsoids are shown at the 30%
probability level. (b) Ball and stick drawing of 5a. The two components
(50/50) of the disorder model are shown.

3268
A.W. Myers et al. / Inorganica Chimica Acta 361 (2008) 3263–3270
product with JRh–P = 162.2 Hz while the minor product
displayed a coupling constant consistent with a C–H acti-
vation product, d 8.58 (J = 148.9 Hz). A 13C-JMOD
NMR experiment revealed that the major product dis-
played a doublet of doublets at d 141.03 (J = 31.0,
22.9 Hz), in phase with C–H and CH3-resonances. This
result identified the major product as C–S insertion away
from the substituted side. A doublet of doublets at d
154.93 (J = 32.1, 20.9 Hz) of opposite phase was also
observed for the minor product, characteristic of activation
of the products unrealizable. Earlier work with 2,5-dimeth-
ylthiophene, however, showed that insertion into the C–S
bond was possible [5], although no insertion occurred with
tetramethylthiophene due to the ‘‘buttressing effect’’ [11].
Consequently the major product is likely assigned structure
8a due to the absence of a methyl group on C3.
S
Rh
Rh
Rh
Me3P
Me3P
+
ð7Þ
H
Me3P
S
S
47ºC
1
of a thiophene ring C–H bond. A H–1H COSY NMR
8a
8b
experiment allowed assignment of the minor product as
the C–H activation of the C–H5 bond (Eq. (5)). Activation
of the a-C–H bonds were previously observed as a kinetic
product when (C5Me5)Rh(PMe3)H2 was photolyzed with
thiophene [11]. Further heating at 70 °C for nine days
resulted in a decrease in the C–H activation product with
an increase in the C–S insertion product and the
(C5Me5)Rh(PMe3)2 decomposition product.
3,4-Dimethylthiophene gave a 1:1.5 ratio of C–S to C–H
activation products (9a and 9b) when heated for 6.5 h at
47 °C with (C5Me5)Rh(PMe3)(3,5-xylyl)H (Eq. (8)). The
C–H activation product began to decrease upon heating
for an additional 6 h with an increase in the amounts of
the C–S inserted product and decomposition product
(C5Me5)Rh(PMe3)2. Only one isomer is available for C–S
insertion, and activation of the a-C–H bond is the assigned
C–H activation product based on earlier thiophene kinetic
studies [11]. 3,4-Dimethylthiophene gave more stable prod-
ucts than those of the two trimethylthiophenes. a-Methyl
substituents have been known to destabilize C–S inserted
products relative to thiophene [5] or benzothiophene [17],
and are presumably contributing to the thermal instabili-
ties seen in the reaction with 2,3,5-trimethylthiophene.
S
SiMe3
Rh
Rh
Rh
H
Me3P
Me3P
+
H
Me3P
ð5Þ
70ºC
S
S
6b
6a
1
SiMe3
SiMe3
3 : 2
3.4. Reaction of (C5Me5)Rh(PMe3)(Ph)H with
polymethylthiophenes
S
Treatment of a hexane solution of the more thermally
labile (C5Me5)Rh(PMe3)(3,5-xylyl)H with 2,3,4-trimethyl-
thiophene at 47 °C for 3 h resulted in the formation of
Rh
Rh
Rh
H
Me3P
+
Me3P
H
Me3P
ð8Þ
47ºC
S
S
9a
9b
1
two products in a 1:1 ratio as observed by H and 31P
1 : 1.5
NMR spectroscopies. One product (7b) was assigned as
activation of the a-C–H bond by observation of a hydride
resonance (d ꢀ12.722, dd, J = 48.3, 28.3 Hz) in the 1H
NMR spectrum and typical 31P NMR data (d 10.00, dd,
J = 153.1 Hz). The other product (7a) was assigned as aris-
ing from insertion into the S–C5 bond, away from the ste-
rically hindered 2-methyl group based on previous evidence
with 2-methylthiophene (Eq. (6)). (Reaction of 1 with 2-
methylthiophene gave exclusive insertion into the unhin-
dered C–S bond [5]). Unfortunately, further heating led
to decomposition and full characterization was not
possible.
The reactivity patterns arising from study of these
polymethylthiophenes shows a lower stability with increas-
ing methyl substitution. The ‘‘buttressing effect’’ which
may have prevented [(C5Me5)Rh(PMe3)] from inserting
into the C–S bond of tetramethylthiophene may not only
decrease insertion rates but also destabilize the C–S
inserted products of di- or tri-methylthiophenes. In addi-
tion a competition between thiophene and 2,5-dimethylthi-
ophene (1:1) in the thermal reaction with 1 at 60 °C showed
a 2:1 preference for C–S activation of the less hindered sub-
strate. Under these conditions, the C–S addition is irrevers-
ible. Consequently, a-methyl substitution has an effect on
the kinetic selectivity as well.
S
Rh
Rh
Rh
3.5. Reaction of (C5Me5)Rh(PMe3)(Ph)H with ethylene
sulfide and trimethylene sulfide
H
Me3P
Me3P
+
H
Me3P
ð6Þ
47ºC
S
S
7a
7b
1 : 1
The previous experiments with a wide range of sub-
strates have demonstrated that the rhodium complex is
capable of insertion into C(sp2)–S bonds of thiophenes in
thermal reactions. The ability of the fragment
[(C5Me5)Rh(PMe3)] to insert into these strong C–S bonds
suggests that it might also be energetically capable of
Thermolysis of (C5Me5)Rh(PMe3)(3,5-xylyl)H with
2,3,5-trimethylthiophene in dry C6D12 at 47 °C for 3.5 h
resulted in the formation of two C–S inserted products in
a 4:1 ratio (Eq. (7)). Further heating led to decomposition.
Problems with thermal stability made thorough assignment

A.W. Myers et al. / Inorganica Chimica Acta 361 (2008) 3263–3270
3269
insertion into C(sp3)–S bonds. Consequently, the thermal
reactions of 1 with ethylene sulfide (C2H4S) and trimethyl-
ene sulfide (C3H6S) were examined.
Results with polymethylthiophenes indicate that pres-
ence of two or three methyl substituents on the thiophene
ring do not prevent C–S insertion, but may destabilize
the C–S inserted products. A single C–S insertion product
was seen with 2,3,4-trimethylthiophene, presumably direc-
ted away from the a-methyl substituent. One C–H activa-
tion product was observed as well, likely activation at the
a-carbon. Two C–S inserted products were seen with
2,3,5-trimethylthiophene in a 4:1 ratio. 3,4-Dimethylthi-
ophenes shows single C–S inserted and C–H activation
products. Small ring thiatanes lead to ring-opening addi-
tions and extrusion of sulfur by what may be radical pro-
cesses. As all of the reactions that give mixtures of C–S
insertion products do not change over time, the product
formation can be either under kinetic control (i.e., C–S
cleavage is irreversible) or under thermodynamic control,
with fast equilibration of C–S bond cleavage products
under the reaction conditions.
Reaction of 1 with C2H4S in C6D12 at room temperature
gave an orange solution with concomitant formation of
yellow precipitate, elemental sulfur. The 31P NMR spec-
trum indicated that a single Rh(III) organometallic com-
1
pound (d 6.72, d, J = 156 Hz) was formed. A H NMR
spectrum showed resonances for free ethylene (d 5.265)
and for a Rh–SH proton at d ꢀ2.876 (d, J = 3.4 Hz).
The complex produced was assigned as (C5Me5)Rh(P-
Me3)(Ph)(SH) (10) by comparison with an authentic sam-
ple (Eq. (9)) [18].
S
Rh
H
Rh
Ph
Me3P
+
C2H4 + S8
ð9Þ
SH
Me3P
23ºC
1
10
Room temperature reaction of 1 with trimethylene sul-
fide (C3H6S) in cyclohexane solution resulted in initially
only the formation of an organometallic complex identified
as (C5Me5)Rh(PMe3)(Ph) (SCH2CH2CH3) (Eq. (10)). The
assignment of the C–S insertion product was based on
the 31P NMR data, showing a doublet at d 8.45 (d,
Acknowledgement
W.D.J. acknowledges support from the National Sci-
ence Foundation (Grant CHE-0717040) for this work.
Appendix A. Supplementary material
1
J = 160 Hz). The H NMR data showed resonances at d
0.987 (t, J = 7.3 Hz, 3H), 1.616 (m, 2H) and 2.168 (t,
J = 14.6 Hz, 2H) for the distinct opened chain hydrogens
as well as doublets at d 1.577 (d, J = 2.7 Hz, 15H) and
1.314 (d, J = 10.0 Hz, 9H) for the C5Me5 and PMe3 pro-
tons. Further stirring the reaction mixture at room temper-
ature for several days resulted in the formation of 10.
CCDC 659675, 658453, and 658454 contains the supple-
mentary crystallographic data for 3, 4 and 5a. These data
can be obtained free of charge from The Cambridge Crys-
tallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif. Supplementary data associated with this
article can be found, in the online version, at
doi:10.1016/j.ica.2007.09.023.
slow
S
Rh
H
Rh
Ph
Me3P
ð10Þ
+ 10
CH2CH2CH3
References
S
Me3P
23ºC
1
11
[1] H. Topsøe, B.S. Clausen, F.E. Massoth, Hydrotreating Catalysis,
Springer-Verlag, Berlin, 1996.
[2] T. Kabe, A. Ishihara, H. Tajima, Ind. Eng. Chem. Res. 31 (1992)
1577.
[3] R.A. Sanchez-Delgado, Organometallic Modeling of the Hydrode-
sulfurization and Hydronitrogenation Reactions, Kluwer Academic
Publishers, Dordrecht, 2002.
[4] W.D. Jones, L. Dong, A.W. Myers, Organometallics 14 (1995) 855.
[5] W.D. Jones, L. Dong, J. Am. Chem. Soc. 113 (1991) 559.
[6] M.-G. Choi, R.J. Angelici, Organometallics 11 (1992) 3328.
[7] M.-G. Choi, R.J. Angelici, J. Am. Chem. Soc. 113 (1991) 5651, and
references therein.
Reaction with 1-d6 leads to formation of the prod-
uct (C5Me5)Rh(PMe3) (C6D5)(SCH2CH2CH2D). Reaction
with a 60/40 mixture of 1 and 1-d6 produces a mixture of
d0, d1, d5, and d6 products, indicating that crossover occurs
during the addition of the C–S bond across the Rh–H(D)
bond. While the mechanism was not examined in detail,
a radical-based process could account for these results.
[8] R.J. Angelici, Coord. Chem. Rev. 105 (1990) 61.
[9] M.J. Sanger, R.J. Angelici, Organometallics 13 (1994) 1821, and
references therein.
4. Summary
Insertion into the C–S bond of thiophene is highly influ-
enced by both steric and electronic effects. Regiospecific
insertion toward the substituted side of thiophene was seen
with 2-methoxythiophene, 3-methoxythiophene, and 2-cya-
nothiophene. 3-Cyanothiophene showed the formation of
two C–S insertion isomers in a 1:1 ratio while 2-trimethyl-
silylthiophene showed selective insertion away from the
substituent as well as activation of the a-C–H bond.
´
[10] C. Bianchini, M.V. Jimenez, A. Meli, F. Vizza, Organometallics 14
(1995) 3196.
[11] L. Dong, S.B. Duckett, K.F. Ohman, W.D. Jones, J. Am. Chem. Soc.
114 (1992) 151.
ˇ
[12] M. Janda, J. Srogl, I. Stibor, M. Nemec, P. Vopatrna´, Synthesis
(1972) 545.
[13] K.H. Do¨tz, H. Fischer, P. Hoffman, F.R. Kreissl, U. Schubert, K.
Weiss, Transition Metal Carbene Complexes, Verlag Chemie, Deer-
field Beach, FL, 1983.

3270
A.W. Myers et al. / Inorganica Chimica Acta 361 (2008) 3263–3270
[14] (a) C.P. Casey, in: M. Jones Jr., R.A. Moss (Eds.), Reactive
Intermediates, vol. 2, Wiley, New York, 1981, p. 135;
[16] A.W. Myers, W.D. Jones, Organometallics 15 (1996) 2905.
[17] A.W. Myers, W.D. Jones, S.M. McClements, J. Am. Chem. Soc. 117
(b) C.P. Casey, in: M. Jones Jr., R.A. Moss (Eds.), Reactive
Intermediates, vol. 2, Wiley, New York, 1985, p. 150.
[15] W.D. Jones, R.M. Chin, J. Am. Chem. Soc. 114 (1992) 9851.
(1995) 11704.
[18] W.D. Jones, V.L. Chandler, F.J. Feher, Organometallics 9 (1990)
164.