Evidence for the existence of a late-metal terminal sulfido complex

Full text of DOI:10.1021/ja984136x

4070
J. Am. Chem. Soc. 1999, 121, 4070-4071
Scheme 1
Evidence for the Existence of a Late-Metal Terminal
Sulfido Complex
David A. Vicic and William D. Jones*
Department of Chemistry, UniVersity of Rochester
Rochester, New York 14620
izations, and to systematically investigate the intermediacy of the
three possible binding modes of a sulfur atom to nickel (Scheme
1). In these efforts we have found kinetic and structural evidence
that support the chemical feasibility of binding mode C, namely
a terminal sulfido complex at nickel.
ReceiVed December 2, 1998
The transition-metal-sulfur bond is an important linkage in
both biological systems1 and industrial catalysts.2 Nickel sulfides,
in particular, are key components of natural hydrogenases3 and
are active promoters in current hydrotreating catalysts.2 Under-
standing the nature of the metal-sulfur linkages at active sites
can offer insight on ways to improve catalysis and provide a better
understanding of cluster formation and cluster interconversion
reactions in general.
Scheme 1 shows three possible binding modes of a sulfur atom
to supported nickel. While numerous examples of binding modes
A4 and B5 can be found throughout the literature, there exists no
sound evidence for a terminal sulfido complex such as C in the
solution phase.6 Additionally, there have been only a few cases
for the group 9 and 10 metals where the intermediacy of a terminal
sulfido complex has been postulated,7 and there has been no
experimental evidence to support the existence of such a species.
Terminal sulfido complexes of the earlier transition-metals are
well-known, and display a wide range of reactivity.8 Late-metal
analogues might then be anticipated to afford similarly rich
chemistry.
Initial studies were directed toward finding an efficient way
to prepare a nickel sulfido complex. Bergman has reported routes
to transient [Cp2TidS] and [Cp2ZrdS] using Cp2Ti(SH)H11 and
Cp2Zr(SH)I,12 respectively. The synthetic strategy that we found
to be successful with nickel was based on two very different but
related findings. It was discovered by Bergman and co-workers
that thermolysis of Cp*2Zr(OH)(Ph) led to the loss of benzene
and formation of the terminal oxo complex, which was subse-
quently trapped by a variety of substrates.12,13 Additionally,
Osakada and co-workers found that thermolysis of trans-Ni(Ar)-
(SH)(PEt3)2 (Ar ) aryl ligand) led to decomposition of the metal
with formation of Ar-H and SdPEt3. Curiously, no Ar-SH was
formed.14 We wondered whether the loss of arene in this instance
was concomitant with the formation of a transient terminal sulfido
complex, mimicking the zirconium-oxo chemistry. We therefore
prepared a nickel complex containing chelating phosphines and
cis-(Ar)(SH) groups to examine if elimination of arene would
readily occur. Chelating phosphines were used since recent reports
have shown that [L2M(µ-S)]2 complexes (where L2 is a chelating
ligand) are thermally quite stable.10b,15 Li2Ni(SH)(Ph) (1a and 1b)
[L2 ) dippe (1,2-bis(diisopropylphosphino)ethane) (1a) and dcpe
(1,2-bis(dicyclohexylphosphino)ethane) (1b)] were prepared and
their thermolysis behavior was studied in solution.
Raney nickel, and homogeneous nickel complexes,9 have been
found to desulfurize a variety of organosulfur substrates. The fate
of the sulfur atom in these reactions is usually nickel sulfide,
thereby preventing catalysis under traditional laboratory condi-
tions. A number of recent reports from our group,10 however,
indicate that sulfur can be extracted from organic compounds
without the formation of nickel sulfide. These results prompted
us to further explore the mechanism of these unique desulfur-
Nickel thiols of type 1 were synthesized by the addition of
NaSH to L2Ni(Cl)(Ph). It was found that mild heating of 1 in
1
THF solutions led to loss of benzene (1 equiv observed by H
NMR spectroscopy) and production of the bridged sulfido dimers
2 in quantitative yields by NMR spectroscopy (eq 1). Complex
(1) (a) Burgess, B. K.; Lowe, D. J. Chem. ReV. 1996, 96, 2983. (b) Howard,
J. B.; Rees, D. C. Chem. ReV. 1996, 96, 2965. (c) Eady, R. R.; Leigh, G. J.
J. Chem. Soc., Dalton Trans. 1994, 2739. (d) Holm, R. H. AdV. Inorg. Chem.
1992, 38, 1.
(2) (a) Topsøe, H.; Clausen, B. S.; Massoth, F. E. Hydrotreating Catalysis;
Springer-Verlag: Berlin, 1996. (b) Startsev, A. N. Catal. ReV.-Sci. Eng. 1995,
37, 353.
(3) For recent reviews see: (a) Fontecilla-Camps, J. C. Struct. Bonding
1998, 91, 1. (b) Albracht, S. P. J. Biochim. Biophys. Acta 1994, 1188, 167.
(4) See, for example: (a) Sa´nchez, G.; Momblona, F.; Sa´nchez, M.; Pe´rez,
J.; Lo´pez, G.; Casabo´, J.; Molins, E.; Miravitlles, C. Eur. J. Inorg. Chem.
1998, 1199. (b) Sa´nchez, G.; Momblona, F.; Garcia, G.; Lo´pez, G.; Pinilla,
E.; Monge, A. J. Chem. Soc., Dalton Trans. 1994, 2271. (c) Wei, C. H.; Dahl,
L. F. Inorg. Chem. 1970, 9, 1878. (d) Gibson, D.; Lippard, S. J. Inorg. Chem.
1986, 25, 219. (e) Henkel, G.; Kriege, M.; Matsumoto, K. J. Chem. Soc.,
Dalton Trans. 1988, 657.
(5) See, for example: (a) Song, L.-C.; Dong, Y.-B.; Hu, Q.-M.; Li, Y.-K.;
Sun, J. Polyhedron 1998, 17, 1579. (b) Teo, B. K.; Bakirtzis, V.; Snyder-
Robinson, P. A. J. Am. Chem. Soc. 1983, 105, 6330. (c) North, T. E.; Thoden,
J. B.; Spencer, B.; Dahl, L. F. Organometallics 1993, 12, 1299. (d) Braunstein,
P.; Sappa, E.; Tiripicchio, A.; Tiripicchio Camellini, M. Inorg. Chim. Acta
1980, 45, L191. (e) Cecconi, F.; Ghilardi, C. A.; Midollini, S.; Orlandini, A.;
Vacca, A.; Ramirez, J. A. J. Chem. Soc., Dalton Trans. 1990, 773.
(6) Evidence has been provided for the existence of a terminal sulfido
complex of nickel in the gas phase. See: Gregor, I. K.; Gregor, R. C. Inorg.
Chim. Acta 1992, 194, 37.
2a was also synthesized independently by the addition of 2 equiv
of (dippe)Ni(SH)2 to [(dippe)NiH]2. A single-crystal X-ray
structure of 2a shows the “bent”16 Ni2S2 core of the bridged sulfido
dimer (see Supporting Information).
(9) (a) Hillhouse, G. L.; Matsunaga, P. T. Angew. Chem., Int. Ed. Engl.
1994, 33, 1748. (b) Eisch, J. J.; Hallenbeck, L. E.; Han, K. I. J. Org. Chem.
1983, 48, 2963-2968. (c) Becker, S.; Fort, Y.; Vanderesse, R.; Caube`re, P.
J. Org. Chem. 1989, 54, 4848-4853.
(10) (a) Vicic, D. A.; Jones, W. D. Organometallics 1998, 17, 3411. (b)
Vicic, D. A.; Jones, W. D. J. Am. Chem. Soc. 1997, 119, 10857.
(11) Sweeney, Z. K.; Polse, J. L.; Andersen, R. A.; Bergman, R. G.;
Kubinec, M. G. J. Am. Chem. Soc. 1997, 119, 4543-4544.
(12) Carney, M. J.; Walsh, P. J.; Hollander, F. J.; Bergman, R. G.
Organometallics 1992, 11, 761.
(13) Carney, M. J.; Walsh, P. J.; Hollander, F. J.; Bergman, R. G. J. Am.
Chem. Soc. 1989, 111, 8751.
(14) Osakada, K.; Hayashi, H.; Maeda, M.; Yamamoto, T.; Yamamoto,
A. Chem. Lett. 1986, 597.
(15) Capdevila, M.; Carrasco, Y.; Clegg, W.; Coxall, R.; Gonza´lex-Duarte,
P.; Lledo´s, A.; Sola, J.; Ujaque, G. Chem. Commun. 1998, 597.
(16) Aullo´n, G.; Ujaque, G.; Lledo´s, A.; Alvarez, S.; Alemany, P. Inorg.
Chem. 1998, 37, 804.
(7) (a) Wachter, J. Angew. Chem., Int. Ed. Engl. 1989, 28, 1613. (b) Garcia,
J. J.; Mann, B. E.; Adams, H.; Bailey, N. A.; Maitlis, P. M. J. Am. Chem.
Soc. 1995, 117, 2179.
(8) For an in depth review see: Parkin, G. Prog. Inorg. Chem. 1998, 47,
1. For recent examples, see: (a) Sweeney, Z. K.; Polse, J. L.; Andersen, R.
A.; Bergman, R. G. J. Am. Chem. Soc. 1998, 120, 7825. (b) Kuchta, M. C.;
Parkin, G. J. Chem. Soc., Dalton Trans. 1998, 2279. (c) Kuchta, M. C.; Hascall,
T.; Parkin, G. Chem. Commun. 1998, 751. (d) Hagadorn, J. R.; Arnold, J.
Organometallics 1998, 17, 1355. (e) Firth, A. V.; Witt, E.; Stephan, D. W.
Organometallics 1998, 17, 3716. (f) Hagadorn, J. R.; Arnold, J. Inorg. Chem.
1997, 36, 2928.
10.1021/ja984136x CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/14/1999

Communications to the Editor
J. Am. Chem. Soc., Vol. 121, No. 16, 1999 4071
Table 1. Kinetic Data for Benzene Loss from (dcpe)Ni(Ph)(SH)
(1b) in THF Solution
[1b], M
T, °C
k, s-1 a
0.00030
0.00030
0.00030
0.00030
0.00030
0.00030
0.00030b
0.00030c
0.012d
30.4
37.9
48.0
56.1
64.8
60.2
60.2
60.2
22
1.94(2) × 10-5
4.66(2) × 10-5
1.56(1) × 10-4
4.06(2) × 10-4
1.03(1) × 10-3
5.14(4) × 10-4
3.57(4) × 10-4
4.94(6) × 10-4
5.0(8) × 10-6
4.8(10) × 10-6
5.0(4) × 10-6
4.5(2) × 10-6
0.037d
22
22
22
0.034d,e
0.034d,f
a Errors are listed at the 95% confidence level, k ( tσ. Rate
measurements by UV-vis unless otherwise noted. b (dcpe)Ni(Ph)(SD)
used. c NaOH (s) added. d Rate measurements by NMR spectroscopy.
e 0.34 M 6 added. f 1.0 M 6 added.
Figure 1. ORTEP drawing of 7. Ellipsoids shown at 30% probability
level. Hydrogen atoms on the phosphine ligand are omitted for clarity.
The formation of dimer 2 from 1 presumably proceeds by way
of the terminal sulfido complex A. The kinetics of eq 1 was
examined to establish the order of the reaction. Thermolysis of
solutions of 1b in THF to form 2b was examined in the
concentration range 3-37 mM and 22-65 °C. The disappearance
of 1b was found to follow first-order kinetics for 3-5 half-liVes
(Table 1). Added base (NaOH) had no effect upon the observed
rate constant. The monodeuterated complex (dcpe)Ni(Ph)(SD)
reacted more slowly than its protio analogue, with kH/kD ) 1.44
at 60 °C. An Eyring plot of the rate data determined over a 43
°C range provides activation parameters of ∆Hq ) 23.1 ( 0.8
kcal/mol and ∆Sq ) -4.2 ( 2.5 eu. These results are consistent
with a unimolecular rate determining step inVolVing the loss of
benzene and formation of a terminal sulfido intermediate.
Attempts at trapping the terminal sulfido with traditional
reagents such as olefins or nitriles failed, but this could be due
to the instability of the resulting metallacycles. To test this
hypothesis, we attempted to prepare a four-membered thia-
metallacycle containing the bisphosphine ligand systems. Reaction
of [(dippe)NiH]2 with cyclohexene sulfide did not lead to the
resulting metallacycle but instead to loss of alkene and formation
of sulfur bridged dimers 2 and 310b (eq 2). Similar reactivity has
3
2b and 5 in a 5:1 ratio, whereas reaction with a 1 M solution of
4 afforded the same products in a 1:1 ratio. These observations
are consistent with a rate-determining step for the formation of
A, followed by a competition between dimerization or reaction
with the nitrone trap.19 The relatively low yields obtained for 5
along with similarities in solubility of 4, 5, and 2b prevented the
structural characterization of the nitrone adduct. However, reaction
of 1b with 20 equiv (0.025 M) of the more reactive nitrone, 3,4-
dihydroisoquinoline N-oxide (6),20,21 afforded the trapped sulfido
product 7 in much higher yields (17:1 ratio of 7 to 2b). The
31P{1H} NMR spectrum of 7 is almost identical to that of 5
(THF-d8, δ 79.77 (d, J ) 30.9 Hz), 74.50 (d, J ) 30.9 Hz)), and
the higher yields permitted the isolation of X-ray quality crystals.
An ORTEP drawing of the single-crystal X-ray structure of 7 is
shown in Figure 1, and confirms the mode of addition as a formal
1,3-dipolar cycloaddition.22
In summary, we have provided both kinetic and structural
evidence that supports the existence of a late-metal terminal
sulfido complex. The chemical feasibility of such an intermediate
may have important mechanistic implications in homogeneous
and heterogeneous desulfurization reactions, as well as in the
reaction chemistry of sulfided clusters in general.
been seen by Hillhouse with (bipy)Ni(COD) and ethylene sulfide,
where at 20 °C the majority of the nickel is precipitated as nickel
sulfide with release of ethylene as the organic product.9a
Nitrones are known to be effective traps for thioketones17 and
proved to be efficient traps of the nickel sulfido A as well. 5,5′-
Dimethyl-1-pyrroline N-oxide (4) reacted with 1b to form a new,
asymmetric product (5) as determined by 31P{1H} NMR spec-
troscopy (eq 3).18 Importantly, the rate of disappearance of 1b
was found to be independent of the presence of nitrone trap or
its concentration, a result consistent with a nonassociative reaction.
Furthermore, the trapping reaction with nitrone was found to be
competitive with the dimerization process, for the product
distribution was found to depend on nitrone concentration.
Reaction of 1b with a 0.34 M THF solution of 4 afforded products
Acknowledgment is made to the National Science Foundation (Grant
CHE-9816365) for their support of this work.
Supporting Information Available: A listing of experimental
procedures, plots of kinetic data, and tables of data collection parameters,
bond lengths, bond angles, fractional atomic coordinates, and anisotropic
thermal parameters (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
JA984136X
(17) (a) Huisgen, R.; Fisera, L.; Giera, H.; Sustmann, R. J. Am. Chem.
Soc. 1995, 117, 9671. (b) Fisera, L.; Huisgen, R.; Kalwinsch, I.; Langhals,
E.; Li, X.; Mloston, G.; Polborn, K.; Rapp, J.; Sicking, W.; Sustmann, R.
Pure Appl. Chem. 1996, 68, 789. (c) Giera, H.; Huisgen, R. Liebigs Ann./
Recueil 1997, 1685. (d) Black, D. St. C.; Watson, K. G. Aust. J. Chem. 1973,
26, 2491.
(19) Alternatively, A could react with 1b to generate 2b, although this
pathway is more complicated than a simple dimerization of transient A as
loss of a second molecule of benzene by a different mechanism must be
involved.
(20) Huisgen, R.; Seidl, H.; Bru¨ning, I. Chem. Ber. 1969, 102, 1102.
(21) Murahashi, S.-I.; Shiota, T. Tetrahedron Lett. 1987, 28, 2383.
(22) In control experiments, the sulfido dimers 2 do not react with 4 or 6
at room temperature to yield any observable products by NMR spectroscopy.
(18) 31P{1H} NMR (THF-d8): δ 80.37 (d, J ) 30.5 Hz), 70.61 (d, J )
30.5 Hz).