Hydrodesulfurization of thiophene and benzothiophene to butane and ethylbenzene by a homogeneous iridium complex

Article abstract of DOI:10.1021/om9610105

Reaction of the dimer [Cp*IrHCl]₂ (Cp* = η⁵-C₅Me₅) in benzene solution with either thiophene or benzothiophene at 90°C in the presence of H₂ gives the hydrogenolysis products [Cp*IrCl]₂(μ-H)(μ-SC₄H₉) (1) and [Cp*IrCl]₂(μ-H)[μ-S(C₆H₄)CH ₂CH₃] (2), respectively, in high yields. Upon further thermolysis under H₂, the completely desulfurized products, butane and ethylbenzene, can be made. Complexes 1 and 2 were characterized by single-crystal X-ray diffraction. In the absence of H₂, reaction of [Cp*Ir HCl]₂ with thiophene gives an additional trinuclear product [Cp*IrCl]₃(H)(SC₄H₆) (3), which was also structurally characterized.

Full text of DOI:10.1021/om9610105

1912
Organometallics 1997, 16, 1912-1919
Hyd r od esu lfu r iza tion of Th iop h en e a n d Ben zoth iop h en e
to Bu ta n e a n d Eth ylben zen e by a Hom ogen eou s Ir id iu m
Com p lex
David A. Vicic and William D. J ones*
Department of Chemistry, University of Rochester, Rochester, New York 14627
Received December 3, 1996X
Reaction of the dimer [Cp*IrHCl]2 (Cp* ) η5-C5Me5) in benzene solution with either
thiophene or benzothiophene at 90 °C in the presence of H2 gives the hydrogenolysis products
[Cp*IrCl]2(µ-H)(µ-SC4H9) (1) and [Cp*IrCl]2(µ-H)[µ-S(C6H4)CH2CH3] (2), respectively, in high
yields. Upon further thermolysis under H2, the completely desulfurized products, butane
and ethylbenzene, can be made. Complexes 1 and 2 were characterized by single-crystal
X-ray diffraction. In the absence of H2, reaction of [Cp*Ir HCl]2 with thiophene gives an
additional trinuclear product [Cp*IrCl]3(H)(SC4H6) (3), which was also structurally character-
ized.
In tr od u ction
Previously in our lab the dimeric species [Cp*IrH3]2
was observed to cleave both C-S bonds of thiophene in
the presence of a hydrogen acceptor (eq 1).4a Reaction
The hydroprocessing of crude oil is one of the largest
scale chemical processes carried out in industry today.
In this process, heteroatom impurities such as thio-
phenes, mercaptans, and quinolines are removed, mak-
ing the oil amenable to further refining. Removal of the
sulfur compounds, in particular, decreases the contribu-
tions to acid rain production upon fuel combustion and
is also valuable in preventing catalyst poisoning both
in the refinement process and in automobiles.1 Because
of both the environmental and economic rewards that
can be achieved through the hydrodesulfurization (HDS)
process, recent research has focused on trying to better
understand the mechanism of HDS in hopes of design-
ing more efficient catalysts.2
The difficulty in modeling HDS in solution lies greatly
in the ability of a given system to cleave both carbon-
sulfur bonds of various thiophenes. Mimicking the
industrial process would require that both carbon-
sulfur bonds be cleaved in a dihydrogen environment,
affording the hydrogenated organic products plus ad-
sorbed sulfur or H2S. While many ring-opening oxida-
tive additions of one of the C-S bonds of thiophene have
been reported,3 fewer examples of complete desulfur-
ization to butanes, butenes, or metal alkyls have been
reported.4 In addition, there have been few reports of
hydrodesulfurization of benzothiophene to ethylbenzene
by well-defined homogeneous model systems.5-7 In this
work is presented a homogeneous model system which
can desulfurize thiophene and benzothiophene to butane
and ethylbenzene, respectively, in the presence of H2.
of this butadiene complex with hydrogen yielded the
desulfurized product n-butane. The ability of the dimer-
ic species to completely remove a sulfur atom from
thiophene led to our interest in finding a similar
compound which did not require a hydrogen acceptor
to generate the active species and could also desulfurize
the more refractory thiophenes such as benzothiophene.
This led us to explore the reactivity of the bis(µ-hydrido)-
bis[chloro(pentamethylcyclopentadienyl)]iridium com-
plex, first synthesized by Gill and Maitlis,8 toward
thiophenic molecules.
Resu lts a n d Discu ssion
Heating a solution of [Cp*IrHCl]2 in benzene (90 °C,
3 h) with an excess of thiophene in the presence of 1
atm of H2 afforded [Cp*IrCl]2(µ-H)(µ-SC4H9) (1) (eq 2)
(4) Most examples come from polynuclear organometallic systems:
(a) J ones, W. D.; Chin, R. M. J . Am. Chem. Soc. 1994, 116, 198-203.
(b) Arce, A. J .; Arrojo, P.; Deeming, A. J .; De Sanctis, Y. J . Chem. Soc.,
Dalton Trans. 1992, 2423-2424. (c) Riaz, U.; Curnow, O. J .; Curtis,
M. D. J . Am. Chem. Soc. 1991, 113, 1416-1417. (d) Chen, J .; Daniels,
L. M.; Angelici, R. J . J . Am. Chem. Soc. 1991, 113, 2544-2552.
(5) Ogilvy, A. E.; Draganjac, M.; Rauchfuss, T. B.; Wilson, S. R.
Organometallics 1988, 7, 1171-1177.
X Abstract published in Advance ACS Abstracts, April 1, 1997.
(1) Topsøe, H.; Clausen, B. S.; Massoth, F. E. Hydrotreating
Catalysis; Springer-Verlag: Berlin, 1996.
(2) Angelici, R. J . Acc. Chem. Res. 1988, 21, 387-394. Rauchfuss,
T. B. Prog. Inorg. Chem. 1991, 39, 259-329. Bianchini, C.; Meli, A. J .
Chem. Soc., Dalton Trans. 1996, 801-814.
(3) See, for example: (a) Selnau, H. E.; Merola, J . S. Organometallics
1993, 12, 1583-1591. (b) Garcia, J . J .; Mann, B. E.; Adams, H.; Bailey,
N. A.; Maitlis, P. M. J . Am. Chem. Soc. 1995, 117, 2179-2186. (c) Buys,
I. E.; Field, L. D.; Hambley, R. W.; McQueen, A. E. D. J . Chem. Soc.,
Chem. Commun. 1994, 557-558. (d) Paneque, M.; Taboada, S.;
Carmona, E. Organometallics 1996, 15, 2678-2679. (e) J ones, W. D.;
Dong, L. J . Am. Chem. Soc. 1991, 113, 559-564. (f) Chen, J .; Angelici,
R. J . J . Am. Chem. Soc. 1990, 112, 199-204. (g) Bianchini, C.; Meli,
A.; Peruzzini, M.; Vizza, F.; Frediani, P.; Herrer, V.; Sanchez-Delgado,
R. J . Am. Chem. Soc. 1993, 115, 2731-2742.
(6) Desulfurization of benzothiophene to ethylbenzene was reported
to occur by Bianchini et al., but was found to be due to the presence of
heterogeneous particles. See: Bianchini, C.; Herrera, V.; J imenez, M.
V.; Meli, A.; Sa´nchez-Delgado, R.; Vizza, F. J . Am. Chem. Soc. 1995,
117, 8567-8575.
(7) Bianchini, C.; J ime´nez, M. V.; Mealli, C.; Meli, A.; Moneti, S.;
Patinec, V.; Vizza, F. Angew. Chem., Int. Ed. Engl. 1996, 35, 1706-
1707.
(8) Gill, D. S.; Maitlis, P. M. J . Organomet. Chem. 1975, 87, 359-
364.
S0276-7333(96)01010-2 CCC: $14.00 © 1997 American Chemical Society

Hydrodesulfurization of Thiophene and Benzothiophene
Organometallics, Vol. 16, No. 9, 1997 1913
F igu r e 1. ORTEP drawing of [Cp*IrCl]2(µ-H)(µ-SC4H9).
Ellipsoids are shown at the 30% probability level. Hydrogen
atoms are omitted from the Cp* ligand for clarity.
F igu r e 2. ORTEP drawing of [Cp*IrCl]2(µ-H)(µ-S(C6H4)-
C2H5). Ellipsoids are shown at the 30% probability level.
Hydrogen atoms on the ligands have been omitted for
clarity.
Ta ble 1. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for 1
S(1)-C(21)
S(1)-Ir(2)
S(1)-Ir(1)
Ir(1)-Cl(1)
Ir(1)-Ir(2)
1.799(12)
2.335(2)
2.337(2)
2.419(2)
2.9041(4)
Ir(2)-Cl(2)
C(21)-C(22)
C(22)-C(23)
C(23)-C(24)
2.404(3)
1.53(2)
1.53(2)
1.55(2)
There are two singlets representing the magnetically
inequivalent Cp* methyl groups. The R-CH2 hydrogen
atoms of the butanethiolate moiety are diastereotopic,
appearing as multiplets at δ 3.36 and 2.05. A COSY
spectrum shows coupling of the R-protons to broad
multiplets in the region δ 1.5-1.9, which in turn couple
to the terminal methyl group at δ 1.02. A hydride
C(21)-S(1)-Ir(2) 112.5(4)
C(21)-S(1)-Ir(1) 111.3(4)
S(1)-Ir(2)-Ir(1)
Cl(2)-Ir(2)-Ir(1)
51.60(6)
90.11(7)
110.0(8)
Ir(2)-S(1)-Ir(1)
S(1)-Ir(1)-Cl(1)
S(1)-Ir(1)-Ir(2)
Cl(1)-Ir(1)-Ir(2)
76.86(7) C(22)-C(21)-S(1)
84.89(9) C(23)-C(22)-C(21) 112.7(10)
51.54(6) C(22)-C(23)-C(24) 111.0(12)
90.51(6)
1
resonance is seen in the H NMR spectrum at δ -15.54
in C6D6, similar to that seen in the spectrum of the
isoelectronic complex [{Cp*Ir}2(µ-H)(µ-Cl)Cl2],9 which
has a hydride resonance at -13.73 in C6D6. Further-
more, the absence of an IR band in the 1800-2100 cm-1
range is consistent with a bridging hydride.8
as a red crystalline solid in 80% isolated yield after
purification. Under relatively mild conditions the
The complex [Cp*IrHCl]2 was found to undergo a
similar reaction with benzothiophene and H2 (90 °C, 2
h), yielding the red crystalline species [Cp*IrCl]2(µ-H)-
(µ-S(C6H4)CH2CH3) (2) in 85% isolated yield after
purification (eq 3). Selective cleavage of the vinylic C-S
thiophene molecule was ring opened and completely
hydrogenated, having been transformed into a bridging
n-butanethiolate ligand. No intermediates were ob-
served while monitoring this reaction by 1H NMR
spectroscopy.
As is evident from the crystal structure of 1 (Figure
1), the Cp* and chlorine ligands are trans and the sulfur
atom is pyramidal, rendering the complex with C1
symmetry. Crystallographic data for 1 are listed in
Tables 1 and 2. Carbon-carbon bond lengths of the
butanethiolate moiety confirm that hydrogenation of
thiophene is complete, with distances of 1.53(2), 1.53-
(2), and 1.55(2) Å for the C21-C22, C22-C23, and C23-
C24 bonds, respectively, thus establishing their single-
bond character. The bridging hydride ligand was
located and refined. The solid state structure is con-
sistent with the 1H NMR data of the complex in solution.
bond is observed. Again, no intermediates were ob-
served while monitoring the reaction by 1H NMR
spectroscopy. The X-ray crystal structure is shown in
Figure 2, and the corresponding crystallographic data
are given in Tables 2 and 3. The bridging hydride
ligand was located and refined. The C1 symmetry of
complex 2 due to sulfur pyramidalization is similar to
(9) White, C.; Oliver, A. J .; Maitlis, P. M. J . Chem. Soc., Dalton
Trans. 1973, 1901-1907.

1914 Organometallics, Vol. 16, No. 9, 1997
Ta ble 2. Cr ysta llogr a p h ic Da ta for 1, 2, a n d 3
Vicic and J ones
1
2
3
Crystal Parameters
chemical formula
fw
C24H40Cl2Ir2S
815.92
C28H40Cl2Ir2S
863.96
C34H52Cl3Ir3S‚C6H5CH3
1267.90
cryst syst
space group (No.)
Z
monoclinic
Cc (9)
4
monoclinic
P21/c (14)
4
monoclinic
P21 (4)
2
a, Å
b, Å
c, Å
16.1874(5)
18.2180(6)
9.2952(3)
102.8200(10)
2672.8(2)
2.028
16.9397(7)
10.3584(4)
16.7262(7)
98.8880(10)
2899.7(2)
1.979
12.3172 (1)
12.4859(2)
14.6850(1)
112.098(1)
2092.53(4)
2.012
â, deg
vol, Å3
F
calc, g cm-3
cryst dimens, mm3
temp, °C
0.50 × 0.30 × 0.10
0.24 × 0.30 × 0.08
0.50 × 0.10 × 0.20
-50
-50
-50
Measurement of Intensity Data
diffractometer
radiation
frame range/time, deg/s
2θ range, deg
data collected
Siemens SMART
Mo, 0.710 73 Å
0.3/10
Siemens SMART
Mo, 0.710 73 Å
0.3/10
4.6-56.5
-19 e h e 21,
-7 e k 13,
-21 e l e 20
17 708
Siemens SMART
Mo, 0.710 73 Å
0.3/30
3.4-56.0
3.4-56.0
-21 e h e 21,
-23 e k e 24,
-8 e l e 12
8052
-16 e h e 13,
-10 e k 16,
-17 e l e 19
13 468
no. of data collected
no. of unique data
3948
6851
8283
no. of obs data (Fo > 4σ(Fo))
agreement between equiv data (Rint
no. of params varied
µ, mm-1
3711
0.0346
277
10.235
empirical (SADABS)
0.17-0.92
0.0291, 0.0624
0.0321, 0.0638
1.006
5499
0.0325
313
9.440
empirical (SADABS)
0.29-0.87
0.0323, 0.0585
0.0500, 0.0635
1.046
7629
0.0336
403
9.785
empirical (SADABS)
0.35-0.81
0.0311, 0.0697
0.0359, 0.0714
1.011
)
abs corr
range of trans. factors
R1(Fo), wR2(Fo2), (Fo > 4σ(Fo))
R1(Fo), wR2(Fo2) (all data)
goodness of fit
inversion at sulfur is not rapid on the NMR time scale.
The methyl group appears as a triplet at δ 1.42, and
the hydride resonates at δ -15.62. No IR band in the
1800-2100 cm-1 region was observed, suggesting that
the hydride is bridging.
No reaction took place when [Cp*IrHCl]2 was reacted
with dibenzothiophene under similar reaction conditions
(vide supra). Dibenzothiophene and alkyl-substituted
dibenzothiophenes are representative of compounds that
are most difficult to desulfurize under both heteroge-
neous and homogeneous reaction conditions.10,11
F igu r e 3. Selected region of the 1H NMR spectrum of
[Cp*IrCl]2(µ-H)(µ-S(C6H4)C2H5) showing the methylene
protons of the ethyl fragment.
Ta ble 3. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for 2
Reactions carried out with thiophene and ben-
zothiophene in the absence of hydrogen also led to
products 1 and 2, but in much lower yields (2% and 33%,
respectively, calculated by NMR). Apparently a hydro-
gen-activated species of [Cp*IrHCl]2 is not necessary to
cleave the carbon-sulfur bond of either thiophene or
benzothiophene. The reaction of [Cp*IrHCl]2 with
thiophene in the absence of hydrogen also led to the new
organometallic product [Cp*IrCl]3(H)(SC4H6) (3) as the
major isolated complex (32% crude yield). X-ray struc-
tural characterization of 3 revealed a trinuclear com-
pound. An ORTEP drawing of 3 is provided in Figure
4, and the corresponding crystallographic data are given
in Tables 2 and 4. The Ir2(µ-H)(µ-SR) core of the
molecule is similar to that found in 1. The η3-allyl
fragment of this complex aroused our curiosity in the
mechanism of thiophene hydrogenation and brought
about further investigation of the trinuclear complex.
Heating a benzene solution of 3 at 90 °C for 24 h under
S(1)-C(21)
S(1)-Ir(2)
S(1)-Ir(1)
Ir(1)-Cl(1)
Ir(1)-Ir(2)
Ir(2)-Cl(2)
C(21)-C(22)
1.789(5)
C(21)-C(26)
C(26)-C(25)
C(26)-C(27)
C(22)-C(23)
C(23)-C(24)
C(24)-C(25)
C(27)-C(28)
1.409(7)
1.396(7)
1.518(7)
1.395(8)
1.368(8)
1.382(8)
1.507(8)
2.3449(12)
2.3515(13)
2.4153(13)
2.9139(3)
2.4101(13)
1.383(7)
C(21)-S(1)-Ir(2)
C(21)-S(1)-Ir(1)
Ir(2)-S(1)-Ir(1)
S(1)-Ir(1)-Cl(1)
S(1)-Ir(1)-Ir(2)
Cl(1)-Ir(1)-Ir(2)
S(1)-Ir(2)-Cl(2)
S(1)-Ir(2)-Ir(1)
Cl(2)-Ir(2)-Ir(1)
119.9(2)
112.7(2)
C(22)-C(21)-S(1)
C(26)-C(21)-S(1)
121.6(4)
117.1(4)
76.70(4) C(25)-C(26)-C(21) 116.7(5)
88.51(5) C(25)-C(26)-C(27) 121.9(5)
51.55(3) C(21)-C(26)-C(27) 121.4(5)
90.12(4) C(21)-C(22)-C(23) 119.8(5)
104.67(5) C(24)-C(23)-C(22) 120.1(6)
51.75(3) C(23)-C(24)-C(25) 119.8(5)
90.66(3) C(24)-C(25)-C(26) 122.3(5)
C(22)-C(21)-C(26) 121.2(5)
C(28)-C(27)-C(26) 116.3(5)
1
that observed in 1 and can be observed in the H NMR
spectrum. Figure 3 shows the region of the H NMR
1
spectrum displaying the inequivalent methylene protons
of the ethyl substituent of the phenyl ring. Two
doublets of quartets (seen as apparent sextets due to
overlap, J ) 7.6 Hz) were found to resonate at δ 3.51
and 2.83 ppm, signifying the location of each proton in
quite different environments, and demonstrating that
(10) Myers, A. W.; J ones, W. D. Organometallics 1996, 15, 2905-
2917 and references therein.
(11) Bianchini, C.; J imenez, M. V.; Meli, A.; Moneti, S.; Vizza, F.;
Herrera, V.; Sa´nchez-Delgado, R. Organometallics 1995, 14, 2342-
2352 and references therein.

Hydrodesulfurization of Thiophene and Benzothiophene
Organometallics, Vol. 16, No. 9, 1997 1915
Sch em e 1
presence of thiophene and hydrogen, all possible per-
mutations of product 1 are seen in a 1:1:1:1 ratio (eq
4).
F igu r e 4. ORTEP drawing of the trinuclear species 3.
Ellipsoids are shown at the 30% probability level. Hydrogen
atoms on the ligands are omitted for clarity.
Ta ble 4. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for 3
Ir(1)-C(32)
Ir(1)-C(31)
Ir(1)-C(33)
Ir(1)-Cl(1)
Ir(2)-S(1)
Ir(2)-Cl(2)
Ir(2)-Ir(3)
2.134(8)
2.159(8)
2.207(8)
2.380(2)
2.338(2)
2.392(2)
2.8859(4)
Ir(3)-S(1)
2.340(2)
2.415(2)
1.790(9)
1.429(12)
1.425(11)
1.497(12)
Ir(3)-Cl(3)
S(1)-C(31)
C(31)-C(32)
C(32)-C(33)
C(33)-C(34)
[Cp*IrHCl]2 also reacts with 1-butanethiol in a fash-
ion which we believe involves cleavage of the dimer.
Reaction in benzene (90 °C, 30 h) affords 1 in quantita-
tive yield (NMR yield) with the production of hydrogen
gas. The proposed steps in this reaction pathway are
given in Scheme 1. The dihydride thiolate chloride
C(32)-Ir(1)-C(31)
C(32)-Ir(1)-C(33)
C(31)-Ir(1)-C(33)
38.9(3)
38.3(3)
67.7(3)
C(31)-S(1)-Ir(2)
C(31)-S(1)-Ir(3)
Ir(2)-S(1)-Ir(3)
C(32)-C(31)-S(1)
C(32)-C(31)-Ir(1)
S(1)-C(31)-Ir(1)
113.6(3)
114.6(3)
76.18(6)
115.3(7)
69.6(5)
C(32)-Ir(1)-Cl(1) 104.9(3)
C(31)-Ir(1)-Cl(1)
C(33)-Ir(1)-Cl(1)
S(1)-Ir(2)-Cl(2)
S(1)-Ir(2)-Ir(3)
Cl(2)-Ir(2)-Ir(3)
S(1)-Ir(3)-Cl(3)
S(1)-Ir(3)-Ir(2)
Cl(3)-Ir(3)-Ir(2)
87.4(2)
84.6(3)
117.3(4)
shown in Scheme 1 can be viewed as structurally similar
92.45(8) C(33)-C(32)-C(31) 117.0(9)
12
51.94(5) C(33)-C(32)-Ir(1)
89.86(6) C(31)-C(32)-Ir(1)
73.6(5)
71.5(5)
to the known stable Ir(V) species Cp*IrH4
and
Cp*IrMe4.13 Additionally, [Cp*IrHCl]2 has been previ-
ously found to undergo oxidative additions producing
monomeric Ir(V) complexes such as Cp*IrH2Cl(SiR3).14
Further mention of related Ir(V) chemistry can be found
in a recent theoretical treatment of a Cp* iridium
complex, where oxidative addition producing Ir(V) over
Ir(III) intermediates was favored in substrate activa-
tion.15
83.72(8) C(32)-C(33)-C(34) 118.3(9)
51.87(5) C(32)-C(33)-Ir(1)
89.98(6) C(34)-C(33)-Ir(1)
68.1(5)
122.2(6)
1 atm of H2 did not lead to any new products, nor did
heating a benzene solution of excess [Cp*IrHCl]2 with
3 at 90 °C for 20 h. Therefore it can be concluded that
the trinuclear species 3 is not an intermediate in the
formation of 1.
The fact that the trinuclear species 3 does not produce
the hydrogenated complex 1 when heated under 1 atm
of H2 suggests that the hydrogenation of thiophene may
be an intramolecular process rather than a process of
hydrogenation of a butadiene thiolate ligand by an
external metal complex. A proposed mechanism for the
production of 1 is given in Scheme 2. Plots of product
growth versus time for the reaction shown in eq 2 follow
first-order kinetics.16 Nearly identical rates were ob-
served for runs using 0.5 and 1 atm of H2, indicating
The dimeric complex [Cp*IrHCl]2 does not readily
dissociate into monomeric fragments. This was shown
by heating a 1:1 mixture of [Cp*IrHCl]2 and [(C5Me4-
Et)IrHCl]2 at 90 °C in C6D6 with no added substrates.
Only 15% of the mixed species [(C5Me4Et)(Cp*)Ir2H2-
Cl2] was observed after 3 h, a time scale comparable to
what is required for reaction with thiophene and H2.
However, when a mixture of the Cp* and (C5Me4Et)
starting materials was heated in the presence of
thiophene at 90 °C, scrambling of the starting materials
to a 1:2:1 ratio of pure Cp*, mixed Cp*(C5Me4Et), and
pure (C5Me4Et) complexes was observed after 1.5 h.
Consequently, cleavage of the dimers appears to be
promoted by a reversible reaction with thiophene.
Involvement of mononuclear iridium species in these
reactions is further supported by the following experi-
ment. When a mixture of [(C5Me4Et)IrHCl]2 and
[Cp*IrHCl]2 is heated in benzene at 90 °C in the
(12) Gilbert, T. G.; Hollander, F. J .; Bergman, R. G. J . Am. Chem.
Soc. 1985, 107, 3508-3516.
(13) Isobe, K.; Va`zquez de Miguel, A.; Nutton, A.; Maitlis, P. M. J .
Chem. Soc., Dalton Trans. 1984, 929-933.
(14) Fernandez, M. J .; Maitlis, P. M. J . Chem. Soc., Dalton Trans.
1984, 2063-2066.
(15) Strout, D. L.; Zaric, S.; Niu, S.; Hall, M. B. J . Am. Chem. Soc.
1996, 118, 6068-6069.
(16) Data were fitted to the equation C(t) ) (C0 - C∞)e-kt + C∞. At
1 atm of H2, kobs ) 0.004 96 min-1, and at 0.5 atm of H2, kobs ) 0.004 15
min-1
.

1916 Organometallics, Vol. 16, No. 9, 1997
Sch em e 2. P r op osed Mech a n ism for th e F or m a tion of 1
Vicic and J ones
that the reaction is first-order in [Cp*IrHCl]2 and zero-
order in hydrogen concentration. This data implies that
hydrogenation of thiophene is occurring after the rate-
determining step, which is assumed to be carbon-sulfur
bond cleavage. The hydrogenation of the double bonds
of thiophene might then occur by a concerted four-center
pathway, or by way of a coordination-insertion pathway
where the vacant site would form by slippage of the Cp*
ring to η3-coordination or by migration of a hydride to
the Cp* ring.17
in homogeneous model systems, and all contain such
thiolato bridges. Complexes 2 and 6 also have strikingly
similar structural motifs with both complexes displaying
an additional bridging hydride ligand.
Hyd r ogen a tion Exp er im en ts w ith 1 a n d 2
The desulfurization of thiophene itself has been more
successful with polynuclear model systems.4b-d When
complex 1 is dissolved in THF-d8 and heated to 145 °C
under 600 psi of H2, butane is produced in 30% yield,
relative to starting material. Lower yields of the
hydrogenolysis product of 1 relative to 2 may be due to
the difficulty in quantifying the volatile gas produced.
The NMR yield was determined by comparison to a
known amount of standard added before the reaction
(see experimental procedure for details). The production
of butane was not suppressed when the reaction was
repeated in the presence of a 2000-fold excess of
elemental Hg, providing evidence that the desulfuriza-
tion reaction is not due to the presence of heterogeneous
particles in solution.20 Higher temperatures are needed
for the desulfurization reactions to take place, indicating
that complexes 1 and 2 are thermally quite stable.
NMR runs using 1 atm of H2 only yielded the desulfu-
rized products when temperatures near 150 °C were
obtained. This work provides the first example of HDS
of thiophene to n-butane via a well-defined 1-butane-
thiolate intermediate (eq 6). In contrast, earlier studies
with [Cp*IrH3]2 show that thiophene can be desulfurized
to afford butane via a µ2-η4-butadiene complex (eq 1).4a
Removal of the butanethiolate moiety in 1 can be
realized under mild conditions by the addition of H2S.
Heating a benzene solution of 1 at 90 °C for 15 h with
excess H2S (1 atm) yielded [Cp*IrCl]2(µ-SH)2 (7),21 in
A benzene solution of complex 2 was heated at 150
°C for 16 h under 600 psi of H2, yielding a dark red
solution. Analysis of the solution by GC/MS and 1H
NMR spectroscopy reveals that the main organic prod-
uct is ethylbenzene, formed in ∼60% yield (eq 5).
2-Ethylthiophenol and bis(2-ethylphenyl) disulfide were
not detected in the reaction mixture. Lower yields of
ethylbenzene were found when the reaction was carried
out under 1 atm of H2, although identification of an
18,19
organometallic product, [(Cp*Ir)3(µ3-S)2](Cl)2
(18%
crude yield), was possible.
The notion that the formation of a µ-thiolato bridge
may be important in cleaving both carbon-sulfur bonds
in thiophenes has been previously suggested,4a and this
may be particularly important for the benzothiophene
derivatives. Complexes 2, 5,5 and 67 have successfully
yielded ethylbenzene under a dihydrogen atmosphere
(17) J ones, W. D.; Kuykendall, V. L.; Selmeczy, A. D. Organome-
tallics 1991, 10, 1577-1586.
(18) Venturelli, A.; Rauchfuss, T. B. J . Am. Chem. Soc. 1994, 116,
4824-4831.
(20) Lin, Y.; Finke, R. G. Inorg. Chem. 1994, 33, 4891-4910.
(21) Tang, Z.; Nomura, Y.; Ishii, Y.; Mizobe, Y.; Hidai, M. Organo-
metallics 1997, 16, 151-154.
(19) Nishioka, T.; Isobe, K. Chem. Lett. 1994, 1661-1664.

Hydrodesulfurization of Thiophene and Benzothiophene
Organometallics, Vol. 16, No. 9, 1997 1917
were recorded on a Bruker AMX400 NMR spectrometer, and
all 1H chemical shifts are reported relative to the residual
proton resonance in the deuterated solvent. High-pressure
reactions were performed with a Parr 4702 bomb equipped
with a Teflon liner.
P r epar ation of [Cp*Ir Cl]2(µ-H)(µ-SC4H9) (1). [Cp*IrHCl]2
(112 mg, 0.154 mmol) was dissolved in 3 mL of thiophene and
5 mL of benzene and placed in an ampule fitted with a Teflon
valve. H2 (1 atm) was added, and the solution was stirred at
90 °C for 3 h. All volatiles were then removed under vacuum,
and the red residue was passed through a neutral alumina
column using THF/benzene (1:1) as an eluant. The solvents
were removed under vacuum, leaving 1 as a red solid (0.10 g,
80%). 1H NMR (400 MHz, C6D6, 25 °C): δ 3.36 (ddd, 1 H,
R-CH), 2.10-2.01 (m, 1 H, R-CH), 1.89-1.70 (m, 2 H, â-CH2),
1.75-1.50 (m, 2 H, γ-CH2), 1.81 (s, 15 H, C5Me5), 1.73 (s, 15
H, C5Me5), 1.02 (t, 3 H, J ) 7.6 Hz, Me), -15.54 (s, 1 H, IrH).
13C{1H} NMR (100 MHz, C6D6, 25 °C): 90.58 (s, C5Me5), 89.79
(s, C5Me5), 34.13 (s, CH2), 31.29 (s, CH2), 22.64 (s, CH2), 14.25
(s, CH3), 9.95 (s, C5Me5), 9.76 (s, C5Me5). Anal. Calcd (found)
for C24H40Cl2Ir2S: C, 35.33 (35.79); H, 4.94 (4.93).
64% yield, along with n-butanethiol (eq 7). It is
important to investigate the effects of H2S gas on HDS
P r ep a r a tion of [Cp *Ir Cl]2(µ-H)(µ-S(C6H4)CH2CH3) (2).
[Cp*IrHCl]2 (100 mg, 0.14 mmol) and benzothiophene (290 mg,
2.16 mmol) were dissolved in benzene (13 mL) and placed in
an ampule fitted with a Teflon valve. H2 (1 atm) was added,
and the solution was stirred at 90 °C for 2 h. All volatiles
were removed under vacuum and the red residue was passed
through a neutral alumina column using THF/benzene (9:1)
as an eluant. The solvents were removed under vacuum, and
excess benzothiophene was removed by sublimation under
reduced pressure at room temperature, leaving 2 as a red solid
(0.10 g, 85%). 1H NMR (400 MHz, C6D6, 25 °C): δ 8.89 (d, J
) 7.2 Hz, 1 H), 7.12-7.01 (m, 3 H), 3.51 (sextet, J ) 7.6 Hz,
1 H, R-CH), 2.83 (sextet, J ) 7.6 Hz, 1 H, R-CH), 1.75 (s, 15
H, C5Me5), 1.48 (s, 15 H, C5Me5), 1.42 (t, J ) 7.6 Hz, 3 H, Me),
-15.62 (s, 1 H, IrH). 13C{1H} NMR (100 MHz, C6D6, 25 °C):
δ ) 146.85 (s, C), 137.56 (s, C), 134.40 (br s, CH), 127.44 (s,
CH), 127.33 (s, CH), 126.26 (s, CH), 91.45 (s, C5Me5), 89.94 (s,
C5Me5), 28.07 (s, CH2), 15.30 (s, Me), 9.61 (s, C5Me5), 8.85 (s,
C5Me5). Anal. Calcd (found) for C28H40Cl2Ir2S: C, 38.92
(39.09); H, 4.67 (4.44).
intermediates because H2S is a noninnocent byproduct
in the HDS process, and also because many industrial
catalysts are based on sulfided phases whose structure
and catalytic properties are modified by the presence
of H2S.1,22 Here we report that a bridging butane-
thiolate moiety bound to two iridium centers can be
removed by treatment with H2S, producing 1-butane-
thiol, H2, and complex 7.
Con clu sion
It was found that [Cp*IrHCl]2, in the presence of H2,
was able to completely desulfurize thiophene and ben-
zothiophene to butane and ethylbenzene, respectively.
Both hydrodesulfurization reactions proceed via orga-
nometallic intermediates containing bridging thiolate
moieties, providing further evidence that cleavage of
both carbon-sulfur bonds in various thiophenes may
require the participation of more than one metal center
with the capability to form a bridging thiolate interme-
diate. Topsøe, et al. have provided recent models for
the heterogeneous reactive site in MoS2 catalysts that
are consistent with such a hypothesis.1
P r ep a r a tion of th e Tr in u clea r Sp ecies 3. [Cp*IrHCl]2
(0.20 g, 0.27 mmol) was dissolved in 8 mL of thiophene and 8
mL of benzene and placed in an ampule fitted with a Teflon
valve. The solution was stirred at 90 °C for 10 h. The volatiles
were removed under vacuum, and the residue was passed
through an alumina column using THF/benzene (1:1) as an
eluant. The solvents were removed under vacuum, leaving
crude product (32%). Trimer 3 was purified by dissolving the
residue in the minimum amount of toluene and cooling to -30
°C overnight. The precipitate was collected and dried under
vacuum, leaving a bright orange solid (25 mg, 8.1%). 1H NMR
(400 MHz, C6D6, 25 °C): δ 4.48 (t, J ) 8.3 Hz, 1 H, â-CH),
4.24 (d, J ) 8.0 Hz, 1 H, R-CH), 3.82 (m, 1 H, γ-CH), 1.96 (s,
15 H, C5Me5), 1.85 (s, 15 H, C5Me5), 1.63 (s, 15 H, C5Me5), 1.28
(d, J ) 5.9 Hz, 3 H, Me), -15.29 (s, 1 H, IrH). 13C {1H}NMR
(100 MHz, C6D6, 25 °C): δ 92.91 (s, C5Me5), 90.53 (s, C5Me5),
90.10 (s, C5Me5), 85.36 (s, CH), 57.66 (s, CH), 53.30 (s, CH),
18.08 (s, Me), 10.22 (s, C5Me5), 10.04 (s, C5Me5), 8.88 (s, C5Me5).
Anal. Calcd (found) for C34H52Cl3Ir3S: C, 35.70 (36.38); H, 4.58
(4.61).
[(C5Me4Et)Ir HCl]2. 1H NMR (400 MHz, C6D6, 25 °C): δ
2.16 (q, J ) 7.8 Hz, 4 H, CH2), 1.66 (s, 12 H, CH3), 1.52 (s, 12
H, CH3), 0.82 (t, J ) 7.5 Hz, 6 H, CH3), -13.43 (s, 2 H, IrH).
[(C5Me4Et)Ir Cl]2(µ-H)(µ-SC4H9) (1′). 1H NMR (400 MHz,
C6D6, 25 °C): δ 3.40 (ddd, 1 H), 2.43-2.31 (m, 4 H), 2.14-
2.07 (m, 1 H), 2.02-1.95 (m, 2 H), 1.90-1.86 (m, 2 H) 1.85 (s,
3 H), 1.84 (s, 3 H), 1.82 (s, 3 H), 1.79 (s, 6 H), 1.78 (s, 3 H),
1.77 (s, 3H), 1.70 (s, 3 H), 1.63-1.51 (m, 2 H), 1.03 (t, J ) 7.44
Hz, 3 H), 0.94 (t, J ) 7.44 Hz, 3 H), 0.88 (t, J ) 7.82 Hz, 3H),
-15.51 (s, 1 H).
Exp er im en ta l Section
Gen er a l P r oced u r es. All operations were performed
under a nitrogen atmosphere unless otherwise stated. Ben-
zene and THF were distilled from dark purple solutions of
benzophenone ketyl. Neutral alumina was heated to 200 °C
under vacuum for 2 days and stored under nitrogen. Thiophene
(G99%) was purchased from Aldrich Chemical Co. and purified
as previously reported.23 Benzothiophene (99%) and hydrogen
sulfide were purchased from Aldrich Chemical Co. and used
without further purification. Ethylbenzene was purchased
from Kodak Chemical Co. [Cp*IrHCl]2 was prepared as previ-
ously reported.8 All C5Me4Et complexes were synthesized by
procedures analogous to the preparation of the Cp* complexes.
GC analyses were carried out on a Hewlett-Packard 5880
Series gas chromatograph, and GC/MS spectra were recorded
on a Hewlett-Packard 5970 Series mass selective detector. A
Siemens-SMART three-circle CCD diffractometer was used for
X-ray crystal structure determination. Elemental analyses
1
were obtained from Desert Analytics. All H and 13C spectra
(22) Callant, M.; Holder, K. A.; Grange, P.; Delmon, B. Bull. Soc.
Chim. Belg. 1995, 104, 245-251.
(23) Spies, G. H.; Angelici, R. J . Organometallics 1987, 6, 1897-
1903.

1918 Organometallics, Vol. 16, No. 9, 1997
Vicic and J ones
Mon itor in g Exch a n ge betw een [(C5Me4Et)Ir HCl]2 a n d
[Cp *Ir HCl]2. [(C5Me4Et)IrHCl]2 (7.3 mg, 0.01 mmol) and
[Cp*IrHCl]2 (7 mg, 0.01 mmol) were dissolved in 0.7 mL of
C6D6 and heated at 90 °C. In the absence of thiophene, 15%
of mixed species [(C5Me4Et)(Cp*)Ir2H2Cl2] (hydride resonance
at δ -13.46) was found to be present after 3 h. In the presence
of 36 µL of thiophene, a 1:2:1 ratio (pure Cp*:mixed species:
pure (C5Me4Et)) of products was found after 1.5 h.
Hz, 2 H; 1.13, sext, J ) 7.5 Hz, 2 H; 1.04, t, J ) 7.7 Hz, 1 H;
0.71, t, J ) 7.4 Hz, 3 H).
X-r a y Str u ctu r a l Deter m in a tion of [Cp *Ir Cl]2(µ-H)(µ-
SC4H9) (1). Slow evaporation of an acetone solution of 1
produced orange plates. A single crystal of dimensions 0.50
× 0.30 × 0.10 mm3 was mounted on a glass fiber with epoxy.
Data were collected at -50 °C on a Siemens SMART CCD area
detector system employing a 3 kW sealed-tube X-ray source
operating at 1.5 kW. Over a period of 7 h, 1.3 hemispheres of
data were collected, yielding 8052 total data after integration
using SAINT (see Table 2). Laue symmetry revealed a
monoclinic crystal system, and cell parameters were deter-
mined from 7095 reflections.24 The space group was assigned
as Cc on the basis of systematic absences and intensity
statistics using XPREP, and the structure was solved using
direct methods included in the SHELXTL 5.04 package. For
a Z value of 4 there is one independent molecule within the
asymmetric unit. In the final model, non-hydrogen atoms were
refined anisotropically (full matrix on F2), with hydrogens
included in idealized locations. A bridging hydride ligand,
located in a difference Fourier map, was included and refined
isotropically. The Flack parameter value of 0.011 indicated
the selection of the correct enantiomorph. The structure was
refined to R1 ) 0.029 and wR2 ) 0.062.25 Fractional coordi-
nates and thermal parameters are given in the Supporting
Information.
X-r a y Str u ctu r a l Deter m in a tion of [Cp *Ir Cl]2(µ-H)[µ-
S(C6H4CH2CH3)] (2). Slow evaporation of an acetone solution
of 2 produced small, orange plates. A single crystal of
dimensions 0.24 × 0.30 × 0.08 mm3 was mounted on a glass
fiber with epoxy. Data were collected at -50 °C on a Siemens
SMART CCD area detector system employing a 3 kW sealed-
tube X-ray source operating at 1.5 kW. Over a period of 7 h,
1.3 hemispheres of data were collected, yielding 17 708 total
data after integration using SAINT (see Table 2). Laue
symmetry revealed a monoclinic crystal system, and cell
parameters were determined from 8192 unique reflections.24
The space group was assigned as P21/ c on the basis of
systematic absences using XPREP, and the structure was
solved using direct methods included in the SHELXTL 5.04
package. For a Z value of 4 there is one independent molecule
within the asymmetric unit. In the final model, non-hydrogen
atoms were refined anisotropically (full matrix on F2), with
hydrogens included in idealized locations. A bridging hydride
ligand, located in a difference Fourier map, was included and
refined isotropically. The structure was refined with final
residuals of R1 ) 0.032 and wR2 ) 0.059.25 Fractional
coordinates and thermal parameters are given in the Sup-
porting Information.
Rea ction of [(C5Me4Et)Ir HCl]2 a n d [Cp *Ir HCl]2 w ith
Th iop h en e a n d H2. [(C5Me4Et)IrHCl]2 (41 mg, 0.05 mmol)
and [Cp*IrHCl]2 (39 mg, 0.05 mmol) were dissolved in 11 mL
of C6H6 and 4.5 mL of thiophene. The solution was heated at
90 °C for 3 h under 1 atm of H2. The solvents were then
removed on a vacuum line, and the residue was analyzed for
the different derivatives of 1 that were present. The exchange
experiments can easily be analyzed by observing the hydride
1
resonances in the H NMR spectrum, with the mixed species
showing hydride resonances that are intermediate between
those of the two pure species (see eq 4). The pure Cp* complex
shows a hydride resonance at δ -15.54, the two mixed Cp*Cp′
species show resonances at -15.53 and -15.52, and the
resonance for the pure Cp′ complex appears at -15.51. All
four products were found to be present in a 1:1:1:1 ratio.
Rea ction of [Cp *Ir HCl]2 w ith 1-Bu ta n eth iol. Butane-
thiol (3.7 µL, 0.03 mmol) was added to a solution of [Cp*IrHCl]2
(25 mg, 0.03 mmol) in C6D6 (0.7 mL) and heated at 90 °C for
30 h under 1 atm of H2 to produce 1 in quantitative yield (NMR
yield based on internal standard).
Hyd r ogen a tion of [Cp *Ir Cl]2(µ-H)(µ-S(C6H4)CH2CH3).
[Cp*IrCl]2(µ-H)(µ-S(C6H4)CH2CH3) (39 mg, 0.045 mmol), ben-
zene (0.9 mL), and undecane (2.96 mg) were placed in a Parr
bomb lined with Teflon. The bomb was pressurized with 600
psi of H2, and the solution was heated at 150 °C for 16 h. The
flask was then depressurized, and the dark red solution was
analyzed by GC, giving a yield of 57% for ethylbenzene
formation. Ethylbenzene was also identified by GC/MS (m/z
1
) 106, 91). The H NMR spectrum of the above reaction run
in C6D6 showed the resonances of ethylbenzene as compared
to an authentic sample [δ 7.18-7.04 (m, 5 H), 2.44 (q, J ) 7.6
Hz, 2 H), 1.07 (t, J ) 7.6 Hz, 3 H)]. The production of
ethylbenzene is also seen when the reaction is run at 1 atm of
H2: [Cp*IrCl]2(µ-H)(µ-S(C6H4)CH2CH3) (17 mg, 0.02 mmol) was
dissolved in 0.7 mL of benzene with added internal standard
and heated at 150 °C under 1 atm of H2 for 19.5 h. The
solution was then decanted and analyzed by GC (25% yield of
ethylbenzene). The solid which had precipitated ([(Cp*Ir)3-
(µ3-S)2](Cl)2) was rinsed with benzene and dried under vacuum
(yield 4 mg, 18%)
Hyd r ogen a tion of [Cp *Ir Cl]2(µ-H)(µ-SC4H9). [Cp*IrCl]2-
(µ-H)(µ-SC4H9) (49 mg, 0.06 mmol), cyclohexane (2.73 mg), and
THF-d8 (0.9 mL) were placed in a Parr bomb lined with Teflon,
and the bomb was charged with 600 psi of H2. The vessel was
heated at 145 °C for 7.5 h. The bomb was then cooled to -78
°C and the vessel depressurized. The solution was transferred
to a prechilled flask, where it was then freeze-pump-thawed
on a high vacuum line. The volatiles were vacuum transferred
to an NMR tube, and the solution was analyzed for butane
relative to cyclohexane. NMR yield of butane: 30%. When
the reaction was repeated in the presence of a 2000-fold excess
of elemental Hg, the production of butane was not suppressed.
X-r a y St r u ct u r a l Det er m in a t ion of [Cp *Ir Cl]3SC4H 6
(3). After recrystallization from toluene, the filtered solid was
dissolved in acetone and orange plates were grown by slow
evaporation. A single crystal of dimensions 0.50 × 0.10 × 0.20
mm3 was mounted on a glass fiber with epoxy. Data were
collected at -50 °C on a Siemens SMART CCD area detector
system employing a 3 kW sealed-tube X-ray source operating
at 1.5 kW. Over a period of 14 h, 1.3 hemispheres of data were
collected, yielding 13 468 total data after integration using
SAINT (see Table 2). Laue symmetry revealed a monoclinic
crystal system, and cell parameters were determined from
8283 reflections.24 The space group was assigned as P21 on
the basis of systematic absences and intensity statistics using
XPREP, and the structure was solved using direct methods
included in the SHELXTL 5.04 package. For a Z value of 2
there is one independent molecule within the asymmetric unit,
Rea ction of [Cp *Ir Cl]2(µ-H)(µ-SC4H9) (1) w ith H2S. 1
(50 mg, 0.06 mmol) and C6Me6 (9.1 mg, 0.056 mmol) were
placed in an ampule fitted with a Teflon stopcock and dissolved
in 20 mL of benzene. H2S (1 atm) (CAUTION! Poisonous
gas!) was added to the flask and the solution stirred at 90 °C
for 15 h. The solvent was removed under vacuum and the
yellow residue dissolved in CDCl3. The NMR yield of [Cp*Ir-
(Cl)(SH)]2 (7) was found to be 64% by comparison to the
internal standard. n-Butanethiol was also observed in 96%
yield by 1H NMR spectroscopy when the reaction was moni-
tored in C6D6 (δ 2.12, q, J ) 7.5 Hz, 2 H; 1.27, quint, J ) 6.3
(24) It has been noted that the integration program SAINT produces
cell constant errors that are unreasonably small, since systematic error
is not included. More reasonable errors might be estimated at 10×
the listed values.
(25) Using the SHELXTL 5.04 package, R1 ) (∑||Fo| - |Fc||)/∑|Fo|,
wR2 ) [∑[w(Fo - Fc2)2]/∑[w(Fo2)2]}1/2, where w ) 1/[σ2(Fo2) + (aP)2
bP] and P ) [f(maximum of 0 or Fo2) + (1 - f)Fc2].
+
2

Hydrodesulfurization of Thiophene and Benzothiophene
Organometallics, Vol. 16, No. 9, 1997 1919
and one toluene molecule was also located in the structure.
In the final model, non-hydrogen atoms other than those found
in toluene were refined anisotropically (full matrix on F2), with
hydrogens included in idealized locations. A bridging hydride
ligand, located in a difference Fourier map, was included and
refined isotropically. The Flack parameter value of 0.011
indicated the selection of the correct enantiomorph. The
structure was refined with final residuals of R1 ) 0.031 and
wR2 ) 0.070.25 Fractional coordinates and thermal parameters
are given in the Supporting Information.
this work. We also thank Dr. Rene Lachicotte for
assistance with the crystal structure determinations.
Su p p or tin g In for m a tion Ava ila ble: Tables of bond
distances, angles, atomic coordinates, and thermal parameters
for 1, 2, and 3 (18 pages). Ordering information is given on
any current masthead page.
Ack n ow led gm en t is made to the National Science
Foundation (Grant CHE-9421727) for their support of
OM9610105