Direct Conversion of α-Substituted Ketones to Metallo-1,2-enedithiolates

Article abstract of DOI:10.1021/ic9602052

A new synthetic route to metallo-1,2-enedithiolates is presented. The addition of 1 equiv of the α-bromo ketones Ar-C(O)CHXR (X = Br) {Ar = 2-quinoxaline, 2-, 3-, or 4-pyridine, Ph, Cl-Ph, and pyrene (R = H); Ar = 2-quinoxaline (R = Me); and Ar = R = Ph} to Cp₂Mo(SH)₂ followed by the addition of base results in the formation of the corresponding metallo-1,2-enedithiolate Cp₂Mo{η²-SC(Ar)C(R)S}. The α-tosyl ketones quinoxaline-C(O)CHR-tosyl {R = H, Me} and the α-phosphorylated ketone 3-pyridine-C(O)CH₂-O-P(O)(OEt)₂ yield the same products as the corresponding α-bromo ketones upon reaction with Cp₂Mo(SH)₂. The addition of acid to the heterocyclic substituted complexes yields Cp₂Mo{η²-SC(HetH⁺)C(R)S}. Both Cp₂Mo{η²-SC(quinoxaline)C(H)S}and [Cp₂Mo{η²-SC(quinoxalinium)C(H)S}][BR₄] have been crystallographically characterized. Cp₂Mo{η²-SC(quinoxaline)C(H)S} crystalizes in the C2/c space group with a = 21.451(2) A?, b = 15.474 A?, c = 12.2201(13) A?, and β= 107.440(7)°. [Cp₂Mo{η²-SC(quinoxalinium)C(H)S}][BF₄] crystalizes in the P1? space group with a = 7.4009(8) A?, b = 10.1192(13)° A?, c = 15.930(4) A?; α = 81.49(2)°, β = 76.14(2)°, and γ = 85.784°. In the solid state [Cp₂Mo{η²-SC(quinoxalinium)C(H)S}][BF₄] π-stacks the heterocycle of two adjacent molecules with atom-atom distances of ≈ 3.6 A?. The stacks are limited to pairs of molecules, and there is no long-range order. The pK_a values for the quinoxalinium (R = H and Me) and the 2-, 3-, and 4-pyridinium (R = H) complexes have been determined in acetonitrile to be 1 -3 units larger than the free heterocycles. The pK_a of the pyridinium complexes follows the substitution trend 2 ≈ 4 > 3 > free pyridinium and is consistent with resonance stabilization of pyridinium by the metallo-1,2-enedithiolate. Electronic transitions in these complexes have been assigned to a LMCT transition and an ILCT transition by comparison of the various complexes accompanied with solvent sensitivity studies.

Full text of DOI:10.1021/ic9602052

Inorg. Chem. 1996, 35, 4743-4751
4743
Direct Conversion of r-Substituted Ketones to Metallo-1,2-enedithiolates
John K. Hsu, Cecilia J. Bonangelino, Sharada P. Kaiwar, Christine M. Boggs,
James C. Fettinger, and Robert S. Pilato*
Department of Chemistry and Biochemistry, The University of Maryland, College Park, Maryland 20742
ReceiVed February 26, 1996X
A new synthetic route to metallo-1,2-enedithiolates is presented. The addition of 1 equiv of the R-bromo ketones
Ar-C(O)CHXR (X ) Br) {Ar ) 2-quinoxaline, 2-, 3-, or 4-pyridine, Ph, Cl-Ph, and pyrene (R ) H); Ar )
2-quinoxaline (R ) Me); and Ar ) R ) Ph} to Cp2Mo(SH)2 followed by the addition of base results in the
formation of the corresponding metallo-1,2-enedithiolate Cp2Mo{η2-SC(Ar)C(R)S}. The R-tosyl ketones
quinoxaline-C(O)CHR-tosyl {R ) H, Me} and the R-phosphorylated ketone 3-pyridine-C(O)CH2-O-P(O)(OEt)2
yield the same products as the corresponding R-bromo ketones upon reaction with Cp2Mo(SH)2. The addition of
acid to the heterocyclic substituted complexes yields Cp2Mo{η2-SC(HetH+)C(R)S}. Both Cp2Mo{η2-
SC(quinoxaline)C(H)S}and [Cp2Mo{η2-SC(quinoxalinium)C(H)S}][BF4] have been crystallographically character-
ized. Cp2Mo{η2-SC(quinoxaline)C(H)S} crystalizes in the C2/c space group with a ) 21.451(2) Å, b ) 15.474
Å, c ) 12.2201(13) Å, and â ) 107.440(7)°. [Cp2Mo{η2-SC(quinoxalinium)C(H)S}][BF4] crystalizes in the P1h
space group with a ) 7.4009(8) Å, b ) 10.1192(13)° Å, c ) 15.930(4) Å; R ) 81.49(2)°, â ) 76.14(2)°, and
γ ) 85.784°. In the solid state [Cp2Mo{η2-SC(quinoxalinium)C(H)S}][BF4] π-stacks the heterocycle of two
adjacent molecules with atom-atom distances of ≈ 3.6 Å. The stacks are limited to pairs of molecules, and
there is no long-range order. The pKa values for the quinoxalinium (R ) H and Me) and the 2-, 3-, and 4-pyridinium
(R ) H) complexes have been determined in acetonitrile to be 1-3 units larger than the free heterocycles. The
pKa of the pyridinium complexes follows the substitution trend 2 ≈ 4 > 3 > free pyridinium and is consistent
with resonance stabilization of pyridinium by the metallo-1,2-enedithiolate. Electronic transitions in these complexes
have been assigned to a LMCT transition and an ILCT transition by comparison of the various complexes
accompanied with solvent sensitivity studies.
Introduction
Considerable research has focused on the synthesis, reactivity,
and physical properties of metallo-1,2-enedithiolates.1-3 These
complexes are of interest as components in magnetic4,5 and
conducting6,7 materials, as solution lumophores,8-11 and as
models for the molybdenum cofactor (Moco).12-16
As shown in eq 1, Moco contains a 1,2-enedithiolate at the
interface between a molybdenum oxo center (or a tungsten oxo
X Abstract published in AdVance ACS Abstracts, July 1, 1996.
(1) Zuleta, J. A.; Burberry, M. S.; Eisenberg, R. Coord. Chem. ReV. 1990,
97, 47.
(2) Pilato, R. S.; Stiefel, E. I. Catalysis by Molybdenum-Cofactor
Enzymes. In Bioinorganic Catalysis; Reedijk, J., Ed.; Marcel Dekker
Inc.: New York, 1993; pp 133-188.
(3) Cassoux, P.; Valade, L.; Kobayashi, H.; Kobayashi, A.; Clark, R. A.;
Underhill, A. E. Coord. Chem. ReV. 1991, 110, 115.
(4) Manoharan, P. T.; Noordik, J. H.; de Boer, E.; Keijzers, C. P. J. Chem.
Phys. 1980, 74, 1980.
(5) Kuppusamy, P.; Manoharan, P. T. Chem. Phys. Lett. 1985, 118, 159.
(6) Veldhuizen, Y. S. J.; Veldman, N.; Spek, A. L.; Faulmann, C.;
Haasnoot, J. G.; Reedijk, J. Inorg. Chem 1995, 34, 140.
(7) Fourmigue, M.; Lenoir, C.; Coulon, C.; Guyon, F.; Amaudrut, J. Inorg.
Chem. 1995, 34, 4979.
(8) Cummings, D. S.; Eisenberg, R. Inorg. Chem. 1995, 34, 3396.
(9) Cummings, D. S.; Eisenberg, R. Inorg. Chem. 1995, 34, 2007.
(10) Bevilacqua, M. J.; Eisenberg, R. Inorg. Chem. 1994, 33, 2913.
(11) Kaiwar, S. P.; Vodacek, A.; Fettinger, J. A.; Blough, N. V.; Pilato, R.
S. Manuscript in preparation.
(12) Pilato, R. S.; Gea, Y.; Eriksen, K. A.; Greaney, M. A.; Stiefel, E. I.;
Goswami, S.; Kilpatrick, L.; Spiro, T. G.; Taylor, E. C.; Rheingold,
A. L In ACS Symposium Series; Stiefel, E. I.; Coucouvanis, D.;
Newton, W. E., Eds.; American Chemical Society: Washington, DC,
1993; Vol. 635; pp 83-97.
(13) Armstrong, E. M.; Austerberry, M. S.; Birks, J. H.; Beddoes, R. L.;
Helliwell, M.; Joule, J. A.; Garner, C. D. Heterocycles 1993, 35, 563.
(14) Soricelli, C. L.; Szalai, V. A.; Burgmayer, S. J. N. J. Am. Chem. Soc.
1991, 113, 9877.
center in the analogous tungsten enzymes) and a pterin
(2-amino-4(3H)-pteridinone).17,18 Recent X-ray crystallographic
work suggests that at least two Moco enzymes contain a pterin
fused pyran.18,24-26 Rajagopalan and co-workers have isolated
the precursor to Moco, a pterin six-substituted with a four carbon
side chain.19-26 The side chain is an R-phosphorylated ketone
where the phosphate is contained in a six-membered ring.
(16) Oku, H.; Ueyama, N.; Nakamura, A.; Kai, Y.; Kanehisa, N. Chem.
Lett. 1994, 607.
(17) Rajagopalan, K. V.; Johnson, J. L. J. Biol. Chem. 1992, 267, 10199.
(18) Chan, M. K.; Mukund, S.; Kletzin, A.; Adams, M. W. W.; Rees, D.
C. Science 1995, 267, 1463.
(19) Johnson, J. L.; Wuebbens, M. M.; Rajagopalan, K. V. J. Biol. Chem.
1989, 264, 13440.
(20) Wuebbens, M. M.; Rajagopalan, K. V. J. Biol. Chem 1993, 268, 13493.
(15) Das, S. K.; Chaudhury, P. K.; Biswas, D.; Sarkar, S. J. Am. Chem.
Soc. 1994, 116, 9061.
S0020-1669(96)00205-4 CCC: $12.00 © 1996 American Chemical Society

4744 Inorganic Chemistry, Vol. 35, No. 16, 1996
Hsu et al.
The biosynthesis of Moco suggests that R-substituted ketones
could serve as potential precursors to metallo-1,2-enedithiolates.
Moco synthesis also suggests that sensitive functional groups
(such as a pterin) should be tolerated in a synthetic route
modeled after Moco biosynthesis. A review of the synthetic
methods to metallo-1,2-enedithiolates, shown in eqs 2-6,
in these reactions was limited to R′ ) -C(O)CH2R, a group
that serves to activate the alkyne.11,38 In an attempt to develop
a new synthetic route to 1,2-enedithiolates, we have drawn from
the biological synthesis of Moco and have studied the direct
conversion of R-substituted ketones to metallo-1,2-enedithi-
olates. This paper reports the novel synthesis of new metallo-
1,2-enedithiolates from the reaction of the bis(hydrosulfido)
complex, Cp2Mo(SH)2, with R-bromo, R-tosyl, and R-phos-
phorylated ketones. The Cp2Mo framework was selected for
this initial study since its metallo-1,2-enedithiolates are known
to be stable and easily isolated.7,38,39 Also included in this paper
are spectroscopic studies that have defined the electronic
transitions in a number of heterocyclic substituted 1,2-enedithi-
olates and pKa studies that help to define the effects of the 1,2-
enedithiolate on an appended heterocycle.
Experimental Section
Materials. Cp2Mo(SH)2,39 (bromoacetyl)-2-quinoxaline,40 1-(quin-
oxalin-2-yl)-2-bromopropanone,40 (bromoacetyl)-2-pyridine,41 (bro-
moacetyl)-3-pyridine,41 and (bromoacetyl)-4-pyridine41 were prepared
according to the literature procedures. (Tosylacetyl)-2-quinoxaline and
1-(quinoxalin-2-yl)-2-tosylethanone were prepared from the reaction
of the bromides with Ag(tosylate) in refluxing acetonitrile.42 Pyridinium
tetraphenylborate was isolated as the precipitate from the reaction of
pyridinium chloride and sodium tetraphenylborate in H2O. Densyl
bromide, 2-bromoacetophenone, 2-bromo-4-chloroacetophenone, 1-(bro-
moacetyl)pyrene, ethyldiisopropylamine, HBF4-etherate, pyridine, and
Ag(tosylate) were purchased from Aldrich and used without further
purification. All reactions were performed under an atmosphere of
nitrogen using standard Schlenk line techniques. Workups were
preformed in air unless stated otherwise. Dichloromethane, acetonitrile,
and pentane were dried over calcium hydride and distilled under
nitrogen. Diethyl ether, tetrahydrofuran, and dioxane were dried over
sodium/benzophenone and distilled under nitrogen. Triethylamine was
dried over potassium hydroxide and vacuum distilled.
Physical Measurements. NMR measurements were made on either
a Bru¨ker AF 200 or a Bru¨ker AM 400. ESR measurments were made
on a Bru¨ker ES-D 200. IR spectra were collected on either a Perkin-
Elmer 1600 Series FT-IR spectrometer or a Nicolet 5 DXL FT-IR
spectrometer. Mass spectral data were collected on a Magnetic Sector
VG 7070E. Chemical analyses were preformed by M-H-W Labora-
tories.
Synthetic Procedures. Preparation of Cp2Mo{η2-SC(2-quinoxa-
line)C(H)S} (7). To a dichloromethane solution of Cp2Mo(SH)2 (0.100
g, 0.342 mmol) was added (bromoacetyl)-2-quinoxaline (0.086 g, 0.342
mmol). To this blue solution was added ethyldiisopropylamine,
dropwise, until it became deep red. The red solution was evaporated
to dryness in air. The solid was chromatographed in air on a 2.5 cm
× 20 cm silica gel (60-200) column, and the product was eluted with
2% methanol:dichloromethane. This solution was evaporated to dryness
to give 7 as a red solid in 53% yield (0.081 g, 0.182 mmol).
Crystallization from dichloromethane-pentane gave analytically pure
includes a method where an R-hydroxy ketone is the organic
precursor. This approach, shown in eq 2, requires the sulfiding
agent P4S10 and the generation of a thiophosphate intermedi-
ate.27,28 While this method has allowed a number of complexes
to be prepared, it is limited to the synthesis of metallo-1,2-
enedithiolates that do not include functional groups that would
react with P4S10, a potent sulfiding agent. With some excep-
tions,29,30 most other synthetic methods require electron-
withdrawing functional groups either to stabilize the free 1,2-
enedithiolate dianion or to activate an alkyne for reaction with
a metal polysulfido complex. This accounts for the large
number of metal complexes that contain the 1,2-enedithiolate
ligands [S2C2R2]2- where R ) CN, C(O)Me, C(O)OMe, CF3,
and Ph.31-38 Pterin substituted metallo-1,2-enedithiolates have
been prepared using the method shown in eq 5. The R′ group
(21) Pitterle, D. M.; Rajagopalan, K. V. J. Biol. Chem. 1993, 268, 13499.
(22) Pitterle, D. M.; Johnson, J. L.; Rajagopalan, K. V. J. Biol.Chem.1993,
268, 13506.
(23) Pitterle, D. M.; Rajagopalan, K. V. J. Bacteriol. 1989, 171, 3373.
(24) Romao, M. J.; Archer, M.; Moura, I.; Moura, J. J. G.; LeGall, J.; Engh,
R.; Schneider, M.; Hof, P.; Huber, R. Science 1995, 270, 1170.
(25) Albert, A.; Mizuno, H. J. Chem. Soc. B 1971, 2423.
(26) Albert, A.; Mizuno, H. J. Chem. Soc., Perkin Trans. 1973, 1615.
(27) Schrauzer, G. N.; Mayweg, V. P.; Heinrich, W. Inorg. Chem. 1965,
4, 1615.
(37) Yoshinaga, N.; Ueyama, N.; Okamura, T.; Nakamura, A. Chem. Lett.
1990, 1655.
(28) Graczyk, A.; Bialkowska, E.; Konarzewski, A. Tetrahedron 1982, 38,
2715.
(38) Pilato, R. S.; Eriksen, K. A.; Greaney, M. A.; Stiefel, E. I.; Goswami,
S.; Kilpatrick, L.; Spiro, T. G.; Taylor, E. C.; Rheingold, A. L. J. Am.
Chem. Soc. 1991, 113, 9372.
(39) Green, M. L. H.; Lindsell, W. E. J. Chem. Soc A 1967, 1455.
(40) Rowe, D. J.; Garner, C. D.; Joule, J. A. J. Chem. Soc., Perkin Trans.
1 1985, 1907.
(29) Garner, C. D.; Armstrong, E. M.; Ashcroft, M. J.; Austerberry, M.
S.; Birks, J. H.; Collision, D.; Goodwin, A. J.; Larsen, L.; Rowe, D.
J.; Russell, J. R. In Molybdenum Enzymes, Cofactor, and Model
Systems; Stiefel, I. E.; Coucouvanis, D.; Newton, E. W. Eds.; ACS
Symposium Series 535, American Chemical Society: Washington,
DC, 1993; pp 98-113.
(30) Rakowski DuBois, M.; Haltiwanger, R. C.; Miller, D. J.; Glatzmaier,
G. J. Am. Chem. Soc. 1979, 101, 5245.
(31) Churchill, M. R.; Fennessey, J. P. Inorg. Chem. 1968, 7, 1123.
(32) Kusters, W.; de Mayo, P. J. Am. Chem. Soc. 1974, 96, 3502.
(33) Bolinger, C. M.; Rauchfuss, T. B. Inorg. Chem. 1982, 21, 3947.
(34) Boyde, S.; Garner, C. D.; Joule, J. A.; Rowe, D. J. J. Chem. Soc.;
Chem. Commun. 1987, 800.
(41) Menasse, R. v.; Klein, G.; Erlenmeyer, H. HelV. Chim. Acta 1955,
38, 1289.
1
(42) Unpublished results. Tosylacetyl-2-quinoxaline. H NMR (CDCl3):
δ 9.43 (s, 1H, C8H5N2), 8.3-8.0 (m, 2H, C8H5N2), 7.9 (m, 2H,
C8H5N2) 7.76 (d, 2 H, C6H4 JH-H ) 8 Hz) 7.24 (d, 2 H, C6H4 JH-H
) 8 Hz), 5.75 (s, 2H, CH2), 2.44 (s, 2H, CH3). Mass spectrum (EI):
m/z ) 342 (M+) 1-(Quinoxalin-2-yl)-2-tosylethanone. 1H NMR
(CDCl3): δ 9.37 (s, 1H, C8H5N2), 8.3-8.0 (m, 2H, C8H5N2), 7.9 (m,
2H, C8H5N2) 7.76 (d, 2 H, C6H4 JH-H ) 8 Hz) 7.24 (d, 2 H, C6H4
JH-H ) 8 Hz), 6.50 (q, 1H, CHMe, JH-H ) 7 Hz) 2.33 (s, 2H, CH3),
1.62 (d, 3H, CH3, JH-H ) 7 Hz). Mass spectrum (EI): m/z ) 357
(M+).
(35) King, R. B.; Bisnette, M. B. Inorg. Chem 1967, 6, 469.
(36) Stiefel, E. I.; Bennett, Z. D.; Crawford, T. H.; Simo, C.; Gray, H. B.
Inorg. Chem. 1970, 9, 281.

Direct Conversion of R-substituted Ketones
Inorganic Chemistry, Vol. 35, No. 16, 1996 4745
material. Anal. Calcd for C20H16MoN2S2: C, 54.05; H, 3.63; N, 6.30.
Found: C, 54.13; H, 3.64; N, 6.14. 1H NMR (CDCl3): δ 9.21 (s, 1H,
C8H5N2), 8.18-8.08 (m, 2H, C8H5N2), 7.92-7.80 (m, 1H, C8H5 N2),
7.77 (s, 1H, C2H), 7.60-7.48 (m, 1H, C8H5N2), 5.29 (s, 10H, C5H5).
UV-vis [abs] (CH2Cl2): λmax (ꢀ) ) 320 (10 900), 492 nm (5280). IR
(KBr): 3109 (w), 3050 (w), 1540 (s), 1505 (vs), 1481 (s), 1365 (m),
1309 (m), 1205 (m), 1131 (m), 1016 (m), 996 (m), 903 (m), 838 (s),
812 (s), 763 (s) cm-1. Mass spectrum (EI) m/z ) 446 (M+).
3116 (w), 3077 (w), 3026 (w), 3056 (w), 1595 (s), 1574 (m), 1560
(m), 1490 (m), 1479 (m), 1437 (m), 1243 (m), 1065 (m), 1026 (m),
1016 (m), 1002 (m), 913 (m), 903 (m), 865 (m), 835 (s), 825 (s), 810
(s), 758 (s), 737 (vs), 704 (s), 691 (vs), 640 (m), 630 (m), 609 (m),
516 (m), 506 (m) cm-1. Mass spectrum (EI): m/z ) 470 (M+), 228
(M+ - diphenyl-1,2-enedithiolate).
Preparation of Cp2Mo{η2-SC(phenyl)C(H)S} (13). This was
prepared as described for 7 with 2-bromoacetophenone (0.061 g, 0.306
mmol) in place of (bromoacetyl)-2-quinoxaline. Complex 13 was
isolated as a brown solid in 26% yield (0.032 g, 0.081 mmol). Anal.
Calcd for C18H16MoS2: C, 55.10; H, 4.11. Found: C, 54.96; H, 4.48.
1H NMR (CDCl3): δ 7.63 (m, 2H, C6H5), 7.25-7.15 (m, 4H, C6H5
and C2H), 5.25 (s, 10H, 2C5H5). UV-vis [abs] (CH2Cl2): λmax (ꢀ) )
361 (4470), 565 (1310). IR (KBr): 3136 (w), 3067 (w), 3053 (w),
3011 (w), 1589 (s), 1571 (s), 1528 (vs), 1483 (s), 1437 (s), 1420 (s),
1211 (m), 1072 (m), 1017 (m), 1000 (m), 901 (m), 831 (s), 813 (vs),
750 (vs), 692 (s), 654 (m), 604 (m) cm-1. Mass spectrum (EI): m/z )
394 (M+), 228 (M+ - phenyl(H)-enedithiolate).
Preparation of Cp2Mo{η2-S(2-quinoxaline)C(Me)S} (8). This was
prepared as described for 7 using 1-(quinoxalin-2-yl)-2-bromopropanone
(0.122 g, 0.342 mmol) in place of bromoacetyl-2-quinoxaline. Complex
8 was isolated as a red solid in 49% yield (0.077 g, 0.168 mmol) Anal.
Calcd for C21H18MoN2S2: C, 55.03; H, 4.96; N, 6.11. Found: C, 54.89;
H, 4.62; N, 6.36. 1H NMR (CDCl3): δ 9.05 (s, 1H, C8H5N2), 8.05-
7.95 (m, 2H, C8H5N2), 7.65-7.59 (m, 2H, C8H5 N2), 5.29 (s, 10H,
C5H5), 2.35 (s, 3H, CH3). UV-vis [abs] (CH2Cl2) λmax (ꢀ) ) 321-
(14 900), 518 (3770). IR (KBr): 3110 (w), 3050 (w), 2890 (w), 2910
(w), 2960 (w), 1565 (m), 1528 (s), 1435 (m), 1365 (m), 1332 (m),
1267 (m), 1203 (s), 1135 (m), 1015 (m), 999 (m), 955 (m), 936 (m),
Preparation of Cp2Mo{η2-SC(p-chlorophenyl)C(H)S} (14). Pre-
pared as described for 7 with (0.095 g, 0.407 mmol) 2-bromo-4-
chloroacetophenone in place of (bromoacetyl)-2-quinoxaline. Complex
14 was isolated as a brown solid in 27% yield (0.038 g, 0.089 mmol).
Anal. Calcd for C18H15ClMoS2: C, 50.65; H, 3.54. Found: C, 51.01;
H, 4.01. 1H NMR (CDCl3): δ 7.56 (d, [JH-H ) 8 Hz], 2H, C6H4Cl),
7.16 (d, [JH-H ) 8 Hz], 1H, C6H4Cl), 7.15 (s, 1H,C2H), 5.26 (s, 10H,
2C5H5). UV-vis [abs] (CH2Cl2): λmax (ꢀ) ) 3369 (6710), 556 (1520).
IR (KBr): 3112 (w), 3078 (w), 3054 (w), 3044 (w), 1595 (s), 1477
(m), 1439 (m), 1245 (m), 1065 (m), 1027 (m), 1016 (m), 1000 (m),
836 (s), 825 (s), 811 (s), 757 (m), 738 (vs), 702 (s), 691 (s), 610 (m)
cm-1. Mass spectrum (EI): m/z ) 428 (M+), 227 (M+ - chlorophen-
ylenedithiolate).
834 (s), 815 (s), 763 (s), 738 (m), 562 (m), 413 (m) cm-1
. Mass
spectrum (EI): m/z ) 460 (M+), 228 (M+ - quinoxalin-(methyl)-
1,2-enedithiolate).
Preparation of Cp2Mo{η2-SC(2-pyridine)C(H)S} (9). To an
acetonitrile solution of Cp2Mo(SH)2 (0.100 g, 0.342 mmol) was added
(bromoacetyl)-2-pyridine (0.068 g, 0.342 mmol). The mixture was
allowed to stir for 20 min and became a deep red. To the solution
was added ethyldiisopropylamine, dropwise, until the solution become
brown. The brown solution was evaporated to dryness in air. The
solid was chromatographed in air on a 2.5 cm × 20 cm silica gel (60-
200) column and was eluted with 2% methanol-dichloromethane. This
solution was evaporated to dryness to give 9 as a brown solid in 29%
yield (0.039 g, 0.099 mmol). Anal. Calcd for C17H15MoNS2: C, 51.91;
H, 3.84; N, 3.56. Found: C, 51.91; H, 3.90; N 3.38. 1H NMR
(CDCl3): δ 8.48 (m, 1H, C6H4N), 7.68 (m, 1H, C6H4N), 7.65 (m, 1H,
C6H4N), 7.39 (s, 1H, C2H), 6.92 (m, 1H, C6H4N), 5.28 (s, 10H, 2C5H5).
UV-vis [abs] [CH2Cl2]: λmax (ꢀ) ) 380 (5750), 548 (1460). IR
(KBr): 3107 (w), 1614 (m), 1581 (s), 1519 (m), 1491 (m), 1460 (m),
1439 (m), 1424 (m), 1128 (m), 1021 (m), 1001 (m), 921 (m), 837 (s),
818 (m), 759 (m) cm-1. Mass spectrum (EI): m/z ) 395 (M+).
Preparation of Cp2Mo{η2-SC(3-pyridine)C(H)S} (10). This was
prepared as described for 9 with (bromoacetyl)-3-pyridine (0.068 g,
0.342 mmol) in place of (bromoacetyl)-2-pyridine. Complex 9 was
isolated as a brown solid in 27% yield (0.036 g, 0.092 mmol). Anal.
Calcd for C17H15MoNS2: C, 51.91; H, 3.84; N, 3.56. Found: C, 52.44;
H, 4.33; N, 3.54. 1H NMR (CDCl3): δ 8.86 (s, 1H, C6H4N), 8.28 (d,
[JH-H ) 5 Hz], 1H, C6H4N), 7.90 (d, [JH-H ) 8 Hz], 1H, C6H4N), 7.12
(dd, [JH-H ) 5 Hz] [JH-H ) 8 Hz], 1H, C6H4N), 6.85 (s, 1H, C2H),
5.27 (s, 10H, 2C5H5). UV-vis [abs] (CH2Cl2): λmax (ꢀ) ) 371 (5540),
547 (1460). IR (KBr): 3092 (w), 3050 (w), 2970 (w), 1581 (m), 1532
(s), 1475 (s), 1439 (s), 1404 (s), 1233 (m), 1124 (m), 1021 (s), 1005
(s), 907 (m), 828 (vs), 817 (vs), 801 (vs), 765 (vs), 711 (s), 660 (m),
Preparation of Cp2Mo{η2-SC(1-pyrene)C(H)S} (15). Prepared
as described for 7 with 1-(bromoacetyl)pyrene (0.086 g, 0.266 mmol)
in place of (bromoacetyl)-2-quinoxaline. Complex 15 was isolated as
a brown solid in 51% yield (0.070 g, 0.136 mmol). 1H NMR
(CDCl3): 8.45 (d, [JH-H ) 10 Hz], 1H, C16H9), 8.11-7.93 (m, 8H,
C16H9), 6.50 (s, 1H, C2H), 5.34 (s, 10H, 2C5H5). UV-vis [abs]
(CH2Cl2): λmax (ꢀ) ) 345 (19 500), 351 (20 700), 445 (4120). IR
(KBr): 3119 (w), 3104 (w), 3037 (w), 3008 (w), 1596 (m), 1580 (m),
1559 (m), 1542 (m), 1484 (m), 1430 (m), 1420 (m), 1369 (w), 1242
(m), 1180 (m), 1016 (m), 998 (m), 922 (m), 840 (vs), 811 (vs), 761
(s), 718 (vs), 681 (m) cm-1. High resolution mass spectra: calcd for
C28H2098MoS2, m/z ) 518.00604; found, m/z 518.00876.
Structural Determination for Cp2Mo{η2-SC(2-quinoxaline)-
C(H)S}‚0.5CH2Cl2 (7). Crystallographic data are collected in Table
1. Crystals were photographically characterized and determined to
belong to the monoclinic crystal system. Systematic absences indicated
the centrosymmetric space group C2/c or the noncentrosymmetric space
group Cc. Intensity statistics clearly indicated the centric case. An
absorption correction was applied with transmission factors ranging
from 0.8514 to 0.9952, with an average correction of 0.9415. The
structure was determined with the successful location of the molyb-
denum and two sulfur atoms using direct methods. An initial difference
map located all remaining non-hydrogen atoms. All hydrogen atoms
were directly located from two subsequent difference-Fourier maps.
The CH2Cl2 solvent molecule’s central carbon atom was found to lie
at 1/2, y, 1/4 and resulted in the location of two half-occupancy chlorine
atoms about the local 2-fold axis. Solvent hydrogen atoms were placed
in calculated positions that were dependent upon both local geometry
restraints and temperature (293 K in this case) with the methylene
605 (s), 586 (m), 551 (s), 450 (w), 313 (m), 365 (m) cm-1
. Mass
spectrum (EI): m/z ) 395 (M+).
Preparation of Cp2Mo{η2-SC(4-pyridine)C(H)S} (11). This was
prepared as described for 9 with (bromoacetyl)-4-pyridine (0.068 g,
0.342 mmol) in place of (bromoacetyl)-2-pyridine. Complex 11 was
isolated as a brown solid in 23% yield (0.031 g, 0.079 mmol). Anal.
Calcd for C17H15MoNS2: C, 51.91; H, 3.84. Found: C, 52.20; H, 3.46.
1H NMR (CDCl3): δ 8.36(d, [JH-H ) 6 Hz], 2H, C6H4N), 7.50 (d,
[JH-H ) 6 Hz], 2H, C6H4N), 7.14 (s, 1H, C2H), 5.27 (s, 10H, 2C5H5).
UV-vis [abs] (CH2Cl2): λmax (ꢀ) ) 382 (5350), 537 (1520). IR
(KBr): 3119 (w), 3091 (w), 3037 (w), 1597 (s), 1586 (vs), 1541 (vs),
1515 (vs), 1409 (s), 1211 (s), 989 (vs), 919 (s), 833 (s), 813 (s), 804
(s), 772 (vs), 731 (s), 671 (m), 652 (s), 608 (s), 555 (m), 458 (m), 393
(m). Mass spectrum (EI) m/z ) 395 (M+).
hydrogen’s C-H distance ) 0.970 Å and UH set equal to 1.2Uparent
.
These calculated positions were optimized after each cycle of the
refinement but not refined. The hydrogen atoms on the main molecule
were all full occupancy and allowed to refine freely resulting in C-H
distances ranging from 0.783 to 1.015 Å and U(eq) ranging from 0.041
to 0.072. This clearly supporting refinement of these atoms. All
computations used SHELXTL 93 software.43
Preparation of Cp2Mo{η2-SC(phenyl)C(phenyl)S} (12). This was
prepared as described for 7 with densyl bromide (0.068 g, 0.342 mmol)
in place of (bromoacetyl)-2-quinoxaline. Complex 12 was isolated in
40% yield as a brown solid (0.054 g, 0.137 mmol). Anal. Calcd for
C24H20MoS2: C, 61.53; H, 4.30. Found: C, 61.11; H, 4.61. 1H NMR
(CDCl3): δ 7.18-6.95 (m, 10H, 2C6H5) 5.29 (s, 10H, 2C5H5). UV-
vis [abs] (CH2Cl2): λmax (ꢀ) ) 371 (5030), 570 (1940). IR (KBr):
Structural Determination for [Cp2Mo{η2-SC(2-quinoxalineH)-
C(H)S}][BF4] (2). Crystallographic data are collected in Table 1.
(43) (a) Gabe, E. J.; Le Page, Y; Charland, J. P.; Lee, F. L.; White, P. S.
J. Appl. Crystallogr. 1989, 22, 384. (b) Sheldrick, G. M. Acta
Crystallogr. 1990, A46, 467.

4746 Inorganic Chemistry, Vol. 35, No. 16, 1996
Hsu et al.
Scheme 1
Table 1. Summary of X-ray Crystal Structure Data for Complexes
7 and 2a
spectrometer in absorbance mode. Solvent sensitivity plots were
generated by graphing absorption λmax vs either E*MLCT or π*.44
The pKa measurements for complexes 2, 3, and 5 were determined
by the progressive addition of [pyridinium][BPh4] to acetonitrile
solutions of 7, 8, and 10, respectively. The concentrations of the metal
complexes were determined by single point UV-visible measurements.
The pyridinium and pyridine concentrations were determined from the
known amount of [pyridinium][BPh4] added and the relative concentra-
tion of the protonated, [XH+], and neutral, [X], metal complexes. The
pKa values were calculated from the Keq values {Keq) [XH+][pyridine]/
[pyridinium][X] ) [XH+]2/([pyridinium]added - [XH+])[X] at five points
between 20 and 80% of the initial concentration of the neutral
complexes; pKa ) pKa′ + log Keq, where pKa′ is 12.3 for [pyridinium]-
[BPh4] in acetonitrile.45 It was not necessary to consider the formation
of [Py-H-Py]+ since its concentration would impart less than a 1%
error in the pKa calculations.46 The experiments were repeated three
times and the results averaged.
The pKa values for complexes 4 and 6 were determined by the
addition of 1 equiv of [pyridinium][BPh4] to 9 and 11 followed by
titration with a standardized solution of pyridine. The pKa values were
calculated at five points from 20 to 80% of the initial concentration of
the neutral complexes. The concentrations of the metal complexes were
determined from single point UV-visible measurements. The pyri-
dinium and pyridine concentrations were determined from the added
amounts and the relative concentration of the protonated and neutral
metal complexes. The pKa values were determined as described above
where Keq ) [XH+]([pyridine]added + [pyridinium]added - [X])/[X]2. The
experiments were repeated three times and the results averaged. The
reported uncertainties are the standard deviations of the average pKa
values.
7
2
formula
fw
cryst syst
space group:
a (Å)
b (Å)
c (Å)
C20.50H17Cl MoN2S2
486.87
monoclinic
C2/c
21.451(2)
15.474(2)
12.2201(13)
C21H18BCl2 F4MoN2S2
616.14
triclinic
P1h
7.4009(8)
10.1192(13)
15.930(4)
81.49(2)
76.14(2)
85.784(10)
1144.6(4)
2
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
107.440(7)
3869.8(7)
8
1.671
293(2)
Mo KR, 0.710 73
1.039
F
calc (g/cm-3
T (K)
)
1.788
153(2)
Mo KR, 0.710 73
1.035
radiation, λ (Å)
abs coeff (mm-1
)
wR2
R1
GOF
0.0698
0.0364
1.020
0.1915
0.0764
1.113
2
a Residuals R1 ) Σ[|Fo| - |Fc|]/Σ|Fo| wR2) {[Σw(Fo - Fc2)2}/
2
2
Σ[w(Fo )2]}1/2 where w-1 ) [σ2(Fo ) + (0.0422P)2 + 3.865P] for 7
2
and w-1 ) [σ2(Fo ) + (0.1215P)2 + 0.49698P] for 2 and P )
(max(Fo ,0) + 2Fc2)/3. GOF ) δ ) [Σ[w((Fo - Fc2)2]/(n - p)]1/2
2
2
where n ) number of reflections and p ) number of parameters.
Crystals were photographically characterized and determined to belong
to the triclinic crystal system P1h. An absorption correction was applied
with transmission factors ranging from 0.586 to 0.997 with an average
correction of 0.773. The structure was determined with the successful
location of the molybdenum and two sulfur atoms. Subsequent
difference-Fourier maps revealed the location of all of the remaining
non-hydrogen atoms. Hydrogen atoms were placed in calculated
positions that were dependent on both the type of bonding at the central
atom and the temperature (153 K in this case) resulting in C-H
distances ) 0.950 Å, UH being set equal to 1.2Uparent, and C-H2
distances ) 0.990 Å, UH being set equal to 1.2Uparent. All computations
used SHELXTL 93 software.43
Electrochemical Experiments. All electrochemical experiments
where conducted with a BAS CV50 in either CV or OSW modes using
a glassy carbon working, platinum auxiliary, and Ag/AgCl reference
electrode with [NBu4][PF6] as the supporting electrolyte. The values
reported in Table 4 were recorded at 100 mV/s under condition where
the ferrocene/ferrocenium couple was found at 452 mV. Complexes
2-6 were generated in the electrochemical cell by the addition of either
[pyridinium][BPh4] or HBF4 etherate to solutions of 7-11. Small
samples of the protonated species were removed from the electrochemi-
cal cell and diluted and UV-visible spectra recorded to insure that
only the monoprotonated species were present. The samples prepared
by the addition of HBF4 and pyridinium were compared by their ∆E1/2
values {∆E1/2 ) E1/2(protonated) - E1/2(neutral)}.
Results and Discussion
Metallo-1,2-enedithiolate Synthesis. The reaction of com-
plex 1 with a range of R-bromo ketones yields the corresponding
metallo-1,2-enedithiolate complexes with accompanying loss of
H2O as shown in eq 7, Scheme 1. Complexes 7-15 are isolated
upon the addition of base and subsequent silica gel column
chromatography. Complexes 2-6 can be regenerated quanti-
tatively by the addition of a proton source to 7-11, respectively.
Both the neutral and protonated complexes are air stable in
solution for several days. Replacing the R-bromo ketones
Quin-C(O)CHRBr with the R-tosyl ketones Quin-C(O)-
CHRTosyl (R ) H, Me) did slow the reactions but had no effect
on the yield of either complex 7 or 8, respectively.
As Moco biosynthesis would suggest, the R-phosphorylated
ketone 1647 is also a viable precursor to a metallo-1,2-
enedithiolate. As shown in eq 8, 16 reacts with 1 to yield
(44) Manuta, D. M.; Lees, A. J. Inorg. Chem. 1983, 22, 3825.
(45) Coetzee, J. F.; Padmanabhan, G. R. J. Am. Chem. Soc 1965, 87, 5005.
(46) Moore, E. J.; Sullivan, J. M.; Norton, J. R. J. Am. Chem Soc. 1986,
108, 2257.
UV-Visible and pKa Measurements for Complexes 2-15. All
UV-visible experiments were conducted using a Perkin-Elmer (λ-2)

Direct Conversion of R-substituted Ketones
Inorganic Chemistry, Vol. 35, No. 16, 1996 4747
Figure 1. ORTEP drawing of 7 with the thermal ellipsoids drawn at
50% probability. Selected bond lengths (Å) and angles (deg) are as
follows: Mo-S(1), 2.4412(14); Mo-S(2), 2.4405(12); S(1)-C(1),
1.725(5); S(2)-C(2), 1.769(5); C(1)-C(2), 1.339(7); C(2)-C(3),
1.463(7); C(3)-C(4), 1.426 (7); C(3)-N(2), 1.315 (7); C(4)-N(1),
1.298(7): S(1)-Mo-S(2), 82.25(4); Mo-S(1)-C(1), 105.68(18); Mo-
S(2)-C(2), 106.62(16); S(1)-C(1)-C(2), 125.2(4); S(2)-C(2)-C(1),
119.6 (4); S(2)-C(2)-C(3), 117.7 (4); C(2)-C(3)-N(2); 119.3(4).
complex 5 with accompanying loss of diethyl phosphate. The
addition of triethylamine to the reaction mixture yields complex
10.
X-ray Crystallographic Results for Complexes 2 and 7.
Both complexes 2 (as the BF4- salt) and 7 were the subject of
single crystal X-ray studies, the results of which are presented
in Tables 1 and shown in Figures 1 and 2, respectively.
Complex 7 is structurally similar to other molybdenum 1,2-
enedithiolate complexes in bond lengths and angles.36,48,49 The
molybdenum 1,2-enedithiolate is a planar five-membered ring.
The plane bisects the angle formed by the intersection of the
planes of the two cyclopentadienyl ligands. The quinoxaline
ring is 26.67°(7) from being coplanar with the five-membered
metallo-1,2-enedithiolate. The Mo-S and C-S bonds are best
described as single bonds with lengths of 2.4412(14) and
2.4405(12) Å and 1.725(5) and 1.769(5) Å, respectively, and
the C(1)-C(2) bond at 1.339(7) Å is best described as a double
bond.
Protonation causes substantial changes in the solid state
structure of the metallo-1,2-enedithiolate complex as shown in
Figure 2. While the 1,2-enedithiolate ligand of 2 is essentially
planar {S(1)sC(1)dC(2)sS(2) deviates less than 0.01 Å from
the plane}, the molybdenum is 0.344(10) Å out of the plane.
The solid state structure of complex 2 is similar to that of
Cp2MoV{S2C2(CN)2}7 and Cp2Ti{S2C2(C(O)Me)2}31 in that
while the 1,2-enedithiolate ligand is planar, the metallo-1,2-
enedithiolate is not. This is in contrast with the solid state
structure of 7 where the the metallo-1,2-enedithiolate is es-
sentially planar.
Figure 2. ORTEP drawing of 2 with the thermal ellipsoids drawn at
50% probability. Selected bond lengths (Å) and angles (deg) are as
follows: Mo-S(1), 2.458(3); Mo-S(2), 2.437(3); S(1)-C(1), 1.695(11);
S(2)-C(2), 1.770(10); C(1)-C(2), 1.33(2); C(2)-C(3), 1.427(14);
C(3)-C(4), 1.45(2); C(3)-N(2), 1.336(14); C(4)-N(1), 1.282(14):
S(1)-Mo-S(2), 81.30(10); Mo-S(1)-C(1), 106.7(4); Mo-S(2)-C(2),
105.8(3); S(1)-C(1)-C(2), 123.5(8); S(2)-C(2)-C(1), 121.3(8); S(2)-
C(2)-C(3), 116.3(4); C(2)-C(3)-N(2); 119.1(9).
While the metallocycle of 2 is no longer planar in the solid
state, the quinoxaline ring is approaching coplanarity with the
1,2-enedithiolate ligand as is seen in the S(2)-C(2)-C(3)-
N(2) dihedral angle of 1.6(13)°. The coplanarity of the
quinoxaline and the 1,2-enedithiolate forces the 1,2-enedithiolate
proton at the quinoxaline C(3) proton in a configuration that is
similar to eclipsed cis-butadiene.
Figure 3. Ball and stick drawing of 2 showing the head to tail
π-stacking of the quinoxalinium in the solid state.
anions are well removed from both the metal and ligands. The
plane of one quinoxaline ring is an average distance of 3.4 Å
from the atoms within the quinoxaline ring of its partner. The
atom-atom distances are an average of 3.6 Å, showing that
the stack is only slightly staggered.
Electronic Transitions. The electronic transitions for all of
the complexes are shown in Table 2. All of the 1,2-enedithiolate
complexes exhibit broad weak (ꢀ ≈ 1500) absorbance bands at
≈550 nm in CH2Cl2.
In addition to the coplanarity of the quinoxalinium and 1,2-
enedithiolate, in the solid state two adjacent molecules of 2 head
to tail π-stack as shown in Figure 3. The head to tail stacking
places the electron-deficient protonated pyrazine above the
benzenoid ring of the quinoxaline of its stacking partner.
-
Solvent molecules separate the stacking pairs, and the BF4
In the quinoxaline complexes this absorbance is overshad-
owed by a more intense transition (ꢀ g 3700). However, when
the absorbance spectra is taken in nonpolar solvents, such as
CCl4 and toluene, the low intensity band is discernible. This
transition is solvent sensitive and shifts to higher energy in more
polar solvents. A plot of absorbance energy vs E*MLCT, a
solvent polarity constant for metal charge transfer transitions,44
as shown for complex 12, is linear for a broad range of solvents
(Figure 4). In alcohols the energy of the absorbance for 12
does not fall on the line with the other solvents. However, as
shown, a plot of energy vs E*MLCT for five alcohols is also linear
(47) Unpublished results. 3-PyC(O)CH2OP(O)(OEt)2 (16). 1H NMR
(CDCl3): δ 9.05 (d, 1H, C5H4N, J ) 10 Hz), 8.76 (m, 1H, C5H4N),
8.15 (m, 1H, C5H4N), 7.43 (m, 1H, C5H4N), 5.23 (d, 2H,C(O)CH2,
JP-H ) 10 Hz), 4.16 (dq, 4H, P-OCH2CH3, JP-H ) 10 Hz, JH-H
)
6 Hz), 1.31 (t, 6H, P-OCH2CH3, JH-H ) 6 Hz). 13C NMR (CDCl3):
δ 191.6 (d, CO, JP-C ) 5 Hz), 154, 149.1, 135.2, 129.6, 123.8, 68.6
(d, CH2, JP-C ) 6 Hz) 64.4, (d, CH2, JP-C ) 6 Hz), 16.0 (d, CH3,
JP-C ) 7 Hz). 31P NMR (CDCl3): δ -0.29. Mass spectrum (EI): m/z
) 273 M+).
(48) Coucouvanis, D.; Toupadakis, A.; Koo, S.-M.; Hadjikyriacou, A.
Polyhedron 1989, 8, 1705.
(49) Brown, G. F.; Stiefel, E. I. J. Chem. Soc., Chem. Commun. 1970, 728.

4748 Inorganic Chemistry, Vol. 35, No. 16, 1996
Hsu et al.
Table 2. UV-Visible Data for Complexes 2-15 Recorded in
CH2Cl2
λ
max(ꢀ)
(protonated)
complex
R, R′
quin, H
(neutral)
7 and 2
320 (10 900)
492 (5280)
323 (5500)
378 (7830)
648 (12 700)
325 (11 100)
389 (9660)
729 (6400)
376 (3810)
525 (6170)
392 (5660)
457 (3930)
547 sh (3250)
329 (9540)
538 (8550)
8 and 3
quin, Me
321 (14800)
538 (3770)
9 and 4
2-pyridine, H
3-pyridine, H
380 (5750)
528 (1460)
371 (5540)
547 (1460)
10 and 5
11 and 6
4-pyridine, H
phenyl, phenyl
phenyl, H
382 (5350)
537 (1520)
371 (5030)
570 (1940)
361 (4470)
565 (1310)
369 (6710)
556 (1520)
345 (19 500)
351 (20 700)
445 (4120)
12
13
14
15
Figure 4. Plots of absorbance energy vs E*MLCT for complex 12.
p-Cl-phenyl, H
pyrene, H
but shifted to higher energy at a given solvent polarity. Plots
of energy vs E*MLCT are commonly grouped by solvent family,
and these observations are generally attributed to association
of certain solvents with the molecule in question.44 The
transition is not particularly sensitive to R group substitution,
and a subtle change such as replacing a p-H with a p-Cl in 13
and 14, respectively, is barely discernible. Given the solvent
sensitivity as well as the assignment of a similar transition in
other Cp2Mo sulfur ligated complexes,50 this transition has been
assigned to a weak 1,2-enedithiolate to metal charge transfer
transition (LMCT). The lack of sensitivity to R-group changes
on the 1,2-enedithiolate suggests that the donor orbital is
localized on the 1,2-enedithiolate and interacts poorly with
appended groups. This is consistent with a metallo-1,2-
enedithiolate orbital that is localized primarily on the sulfur
atoms. The acceptor orbital is thought to be primarily the metal
d(xz) orbital.50
Additional low energy electronic transition are observed with
the strongest bands found at 360-380 nm in the pyridine- and
phenyl-containing complexes, 9-13, at 445 nm in the pyrene
complex 15, and at 492 and 538 nm in the quinoxaline
complexes 7 and 8. This transition is sensitive to the substitution
of the 1,2-enedithiolate as is evident by the ≈1740 cm-1 red
shift upon replacing the H in 7 with Me in 8. The red shift
upon methyl substitution rules out assignment of this transition
to a MLCT. In complexes 7-11, all containing a basic nitrogen,
the absorbance is substantially red shifted upon protonation of
the heterocycle. A lack of such a red shift upon the addition
of acid to complexes 12-15, supports protonation occurring at
the heterocyclic nitrogen in 7-11. The red shift upon proto-
nation is least pronounced in the 3-pyridine-containing complex
where it is ≈1450 cm-1, as shown in Figure 5. Protonation of
the 2-pyridine- and 4-pyridine-containing complexes results in
red shifts of 7270 cm-1 (380 to 525 nm) and 7590 cm-1 (382
to 538 nm), respectively. These shifts are accompanied by an
increase in the extinction coefficient that results in an over-
shadowing of the LMCT band in 4 and 6. With a small red
shift upon protonation of the 3-pyridine-containing complex,
the lowest energy band in complex 5 is still the LMCT, which
is clearly visible in Figure 5, insert B.
Figure 5. UV-vis absorption spectra: (a) 9 f 4; (b) 10 f 5; (c) 11
f 6. The arrows V denote the position of the acid-sensitive ILCT band
in the neutral complexes. The arrows v denote the position of the acid-
sensitive ILCT band in the protonated complexes. The values shown
are the energy differences between this transition in the neutral and
protonated complexes.
Protonation of the 2-quinoxaline complexes 7 and 8 also red
shifts this absorbance by 4890 cm-1 (492 to 648 nm) and 4870
cm-1 (538 to 729 nm), respectively. Methyl substitution of the
1,2-enedithiolate again red shifts the transition (≈1715 cm-1).
As stated above, methyl substitution of the 1,2-enedithiolate in
the neutral complexes red shifts the transition by ≈ 1740 cm-1
.
The similarity in the red shifts of the absorbances upon methyl
substitution suggests a common origin for the lowest energy
bands in the neutral and protonated complexes. The red shift
upon protonation and further red shift upon methyl substitution
rules out assignment of this band to either a d f d, a MLCT,
or a LMCT transition.
The acid-sensitive transition of 7 shifts to lower energy in
more polar solvents. This bathochromic shift with increasing
polarity is the opposite of what is observed for the LMCT
transition. In the protonated complex 2, the acid-sensitive
(50) Bruce, A. E.; Bruce, M. R. M.; Sclafani, A.; Tyler, D. R. Organo-
metallics 1984, 3, 1610.

Direct Conversion of R-substituted Ketones
Inorganic Chemistry, Vol. 35, No. 16, 1996 4749
Figure 6. Titration of 7 with HBF4 to generate 2 as monitored by
UV-visible spectroscopy. The arrows V denote the absorbances
decreasing upon the addition of acid. The arrows v denote the
absorbances increasing upon the addition of acid.
transitions shifts to higher energy in more polar solvents. While
the energy of the acid-sensitive transitions is effected by solvent,
attempts to rectify these energy shifts with the polarity constants
E*MLCT and π* for 15 solvents gave rather poor fits to a single
line. The graphs vs π* gave marginally better fits than those
44
vs E*MLCT
.
Plots of energy vs either E*MLCT and π* for just
the chlorinated solvents or the alcohols did not improve the
correlations.
The assignment most consistent with the red shift upon
heterocycle protonation, continued red shift upon methyl
substitution of the 1,2-enedithiolate, and the solvent sensitivity
is an intraligand charge transfer transition (ILCT). The donor
orbital is thought to be on the 1,2-enedithiolate, and the acceptor
orbital is near the site of protonation. This is best described as
a 1,2-enedithiolate π f aromatic π* charge transfer transition.
Since neither E*MLCT or π* were derived for an ILCT transition,
it is not surprizing that the energy shifts observed with solvent
polarity were not modeled by either of these polarity constants.
Further support for this assignment comes from related work
on a similar family of complexes (dppe)M{η2-SC(2-quinoxa-
line)C(R)S} (M ) Ni, Pd, Pt; R ) H, Me).11 These complexes
contain acid-sensitive electronic transitions similar to those of
7 and 8. These transitions are nearly equivalent for the
corresponding Ni, Pd, and Pt complexes (differing by e5 nm).
It is expected that if these transitions had considerable MLCT
(or LMCT) character, they would be substantially more sensitive
to the metal center.51-55 While plots of the absorption energy
vs either E*MLCT and π* for 7 were not linear, a plot of the
absorption energies of 7 vs the absorption energies of (dppe)-
Pd{η2-SC(2-quinoxaline)C(R)S}in a given solvent was linear.
This suggests that this absorbance can be assigned to the same
electronic transition in both the group VI and group VIII metal
complexes and supports the ligand nature of this transition.
The relatively small energy shift of the ILCT upon protonation
in the 3-pyridine complex, 10, relative to the 2- or 4-pyridine
complexes, 9 and 11, is expected given the resonance structures
shown for the pyridine complexes in eqs 9 and 10.13 The 1,2-
enedithiolate ligand stabilizes the protonated 2- and 4-substituted
heterocycles through resonance forms B and C. As will be
discussed, this also has a profound effect on the pKa of the 2-
and 4-substituted heterocycles.
The 3-pyridine complex 5 only has resonance form A
available. Protonation of the 3-pyridine complex (10 f 5) will
lower the π* orbital of the heterocycle but will have little affect
upon the donating 1,2-enedithiolate π orbital. This results in a
relatively small energy shift of the ILCT band upon protonation.
Not only does protonation of the 2- and 4-substituted complexes
lower the π* orbital of the heterocycle, but resonance forms B
and C would raise the energy of the 1,2-enedithiolate π orbital,
thereby, adding to the red shift of the ILCT band upon
protonation.
As shown for the conversion of 7 f 2, in Figure 6,
protonation appears to be a clean reaction; however, a trace
paramagnetic impurity is generated. The paramagnetic impurity
accounts for less than 5% of the sample as measured by ESR
against a MoV standard. Addition of base eliminated the
presence of the MoV complex and converts the protonated
complexes to their neutral analogs. The identity of this MoV
complex has yet to be determined.
The addition of excess HBF4 to the quinoxaline complexes
2 and 3 in nonpolar solvents leads to an additional red shift
(≈2300 cm-1) of the band assigned to the ILCT transition. This
has been assigned to a second protonation of the quinoxaline
ligand and the formation of a dicationic complex. These
complexes are extremely acidic and readily lose the second
proton in the presence of even weakly basic substrates, reverting
to the monoprotonated species.
1H NMR and IR. In the 1H NMR all of the neutral
complexes have cyclopentadienyl resonances at 5.3 ((0.05)
ppm. The protonation of 7-11 results in downfield shifts of
≈ 0.1 ppm for the cyclopentadienyl resonances, while the other
resonances associated with 2-6 are shifted downfield by
between 0.2 and 0.3 ppm. The resonances of the protonated
complexes are broadened slightly, presumably due to the MoV
impurity.
(51) Gray, H. B.; Ballhausen, C. J. J. Am. Chem. Soc. 1963, 85, 260.
(52) Shupack, S. I.; Billig, E. C.; Williams, R.; Gray, H. B. J. Am. Chem.
Soc. 1964, 86, 4594.
(53) Werden, B. G.; Billig, E.; Gray, H. B. Inorg. Chem. 1966, 5, 78.
(54) Fackler, J. P.; Coucouvanis, D. J. J. Am. Chem. Soc. 1966, 88, 3913.
(55) Bowmaker, G. A.; Boyd, P. D. W.; Campbell, G. K. Inorg. Chem.
1983, 22, 1208.

4750 Inorganic Chemistry, Vol. 35, No. 16, 1996
Table 3. pKa Values for Complexes 2-6 in Acetonitrile
Hsu et al.
Table 4. E1/2 Values for Complexes 7-15a
complex
R, R′
pKa
complex
R R′
E1/2 (mV)
2
3
4
5
6
2-quinoxaline, H
2-quinoxaline, Me
2-pyridine, H
3-pyridine, H
4-pyridine, H
12.1 ( 0.1
11.1 ( 0.1
15.4 ( 0.2
13.5 ( 0.2
15.2 ( 0.1
7
8
9
10
11
12
13
14
15
2-quinoxaline (H)
2-quinoxaline (Me)
2-pyridine (H)
3-pyridine (H)
4-pyridine (H)
phenyl (phenyl)
phenyl (H)
256
200
173
200
171
118
119
158
121
In addition to infrared bands associated with the cyclopen-
tadienyl groups, complexes 7-11 have bands associated with
the CdN and CdC stretching modes of the heterocycles and
1,2-enedithiolate. In the quinoxaline containing complex 7, two
of these bands, at 1540 and 1505 cm-1, shift to 1595 and 1574
cm-1 upon protonation. Protonation of the pyridine complexes,
however, led to only subtle changes in the energy of these
stretching modes (e10 cm-1).
p-Cl-phenyl (H)
pyrene (H)
a The values are reported in CH2Cl2 vs Ag/AgCl where the ferrocene/
ferrocenium is found at 452 mV.
These findings suggest that the metallo-1,2-enedithiolate is as
good at resonance stabilizing pyridinium as an amino functional
group.
The CdC and MsS stretching modes of the 1,2-enedithiolate
are difficult to properly assign. These bands have only been
reliable assigned in a small number of 1,2-enedithiolate
complexes using linear coordinate analysis.56-59 Attempts to
identify the N-H+ stretching modes in the protonated com-
Resonance forms A, B, and C are also available to the
2-quinoxalinium complexes, assuming protonation of N(1), as
shown in eq 12. This accounts for the increase in basicity of
plexes were obscured by C-H stretching bands at ≈3000 cm-1
.
pKa Measurements. In an attempt to better understand the
fundamental properties of heterocyclic substituted metallo-1,2-
enedithiolates, the pKa values for 2-6 were measured. The pKa
measurements where made in accord with the equilibrium shown
in eq 11, using [pyridinium][BPh4] (pKa ) 12.3 acetonitrile)45
as the acid. The reactions were easily followed by UV-visible
spectroscopy and a summary of the results are shown in Table
3. In acetonitrile the pKa values for the pyridine-containing
metal complexes are 1-3 units higher than that of pyridinium.
The pKa value for quinoxalinium is not known in acetonitrile,
but from the aqueous value and the observed pKa shifts in
CH3CN of other aromatic heterocycles, it is assumed that the
values would be in the 9-10 range. As such, the quinoxaline
complexes, like their pyridine analogs, are more basic than the
free heterocycle by 2-3 pKa units.
The pKa values for the substituted pyridinium complexes
follows the pyridine substitution trend 2 ≈ 4 > 3 > free
pyridinium. The increase in the pKa of all the pyridines along
with the pyridine trend is consistent with resonance stabilization
of the protonated heterocycle by the 1,2-enedithiolate. As
shown in eqs 9 and 10, resonance forms B and C localize the
positive charge on the 1,2-enedithiolate and the metal center,
respectively. The absence of this resonance stabilization in 5
accounts for the relatively lower pKa of this complex.
the quinoxalines in 7 and 8 relative to the parent heterocycle.
The pKa value for 2 is close to that predicted for 2-aminoquin-
oxaline assuming an eight unit shift in acetonitrile from the value
in H2O. The more electron rich 8, however, is less basic than
complex 7. The pKa drop suggests that resonance forms B and
C are less important in the methyl-substituted complex. In these
resonance forms, maximum conjugation occurs when the 1,2-
enedithiolate is coplanar with the protonated heterocycle. The
methyl group would be expected to hinder this coplanarity,
decreasing the importance of these resonance forms and lower
the pKa of 3 relative to 2.
Electrochemistry. As shown in Table 4 (see Figure 7), all
of the complexes prepared in this study possess reversible MoIV/
MoV oxidation waves. The E1/2 values are sensitive to the
R-groups as expected. Replacing a p-H with a p-Cl shifts the
E1/2 value by ≈40 mV, making 14 more difficult to oxidize
than 13 and methyl substitution of the 1,2-enedithiolate makes
complex 8 easier to oxidize than 7 by 56 mV. While the MoV
species where stable on the electrochemical time scale, a stable
MoV ESR signal was not observed in attempts to prepare these
complexes by chemical oxidation.
The increase in the pKa of 4-6 relative to the parent
heterocycle is similar to that of 2-, 3-, and 4-aminopyridiniums.
Protonation of complexes 7-11 shifts the reversible MoIV/
MoV redox potentials to more positive values as expected. The
E1/2 for complexes that do not possess an appended basic site
are shifted e10 mV upon the addition of acid. The redox
potential shifts ∆E1/2 {∆E1/2 ) E1/2(protonated) - E1/2(neutral)} for
7-11 are dependent upon the acid used. In general, a stronger
(56) Kilpatrick, L.; Rajagopalan, K. V.; Hilton, J.; Bastian, N. R.; Stiefel,
E. I.; Pilato, R. S.; Spiro, T. G. Biochemistry 1995, 34, 3032.
(57) Subramanian, P.; Burgmayer, S.; Richards, S.; Szalai, V.; Spiro, T.
G. Inorg. Chem. 1990, 29, 3849.
(58) Clark, R. J. H; Turtle, P. C. J. Chem. Soc., Dalton Trans. 1978, 1714.
(59) Schlapfer, C. W.; Nakamota, K. Inorg. Chem. 1975, 14, 1338.

Direct Conversion of R-substituted Ketones
Inorganic Chemistry, Vol. 35, No. 16, 1996 4751
Metallo bis(hydrosulfido) complexes, such as Cp2Mo(SH)2,
directly convert R-bromo, R-tosyl, and R-phosphorylated ketones
to metallo-1,2-enedithiolates. The Cp2Mo complexes are easily
prepared from convenient air stable metallic and organic starting
materials, and this method is amenable to a broad range of
functional groups. While this paper reports the synthesis of
complexes of Cp2Mo, future reports will include studies of the
group VIII metal hydrosulfido complexes,11 as well as reactions
of the thiomolybdates with similar organic substrates. All of
these reactions lead to the formation of metallo-1,2-enedithi-
olates.
The complexes prepared in this study have allowed the
general properties of heterocyclic substituted metallo-1,2-
enedithiolates to be explored. While it is expected that d2
metallo-1,2-enedithiolates possess a LMCT transition, the
heterocyclic-substituted complexes prepared in this study have
an additional transition with energies that are extremely acid
sensitive. This transition has been assigned to an intraligand
charge transfer transition (ILCT), that is 1,2-enedithiolate π f
aromatic π* in character.
It has also been demonstrated that the 1,2-enedithiolate ligand
when appended to a heterocycle greatly affects the chemistry
of the heterocycle. In particular, a metallo-1,2-enedithiolate can
increases the basicity of an appended aromatic heterocycles by
as much as 3 pKa units through resonance stabilization of the
protonated species. This increase is equivalent to that imposed
by a similar amino group substitution.
Figure 7. Cyclic voltammogram of complex 7 in CH2Cl2, with [NBu4]-
[PF6] as the supporting electrolyte. Sweep rate ) 100 mV/s. The
ferrocene/ferrocenium couple is found at 452 mV but is not shown.
acid results in a larger ∆E1/2. The positive shifts are small,
e25 mV, for complexes 7, 8, and 10 when pyridinium is used
as the proton source. The addition of 1 equiv of pyridinium to
complexes 9 and 11 shift the potentials by ≈180 mV. The ∆E1/2
values, generated using pyridinium, reflect the pKa’s of the MoV
protonated complexes more than the E1/2 values for the
protonated species. The protonated MoV species are expected
to be more acidic than the corresponding MoIV complexes and
would readily loss a proton (to pyridine) once generated under
conditions where the protonated MoIV complexes are the
dominate species in solution. This results in relatively small
changes in the E1/2 values that roughly follow the pKa trend
established for the MoIV complexes.
When excess HBF4-etherate is used to protonate complexes
7-11, the ∆E1/2 values are much larger. The ∆E1/2 values for
complexes 7-9 are between 200 and 220 mV, while for
complex 10 and 11 they are 375 and 310 mV, respectively.
These values are consistent with those obtained using the
isolated complexes.
Acknowledgment. We are indebted to the American Chemi-
cal Society, Petroleum Research Fund (Grant No. 28499-G3)
and the American Heart Association, Maryland Chapter (Grant
No. MDBG0695) for funding We are also indebted to the
National Science Foundation for an REU grant. (Grant No.
CHE-9300336) for the support of C.J.B. in the summer of 1994.
Supporting Information Available: Tables listing details crystal-
lographic data, positional parameters, anisotropic thermal parameters,
and bond lengths and angles, (18 pages). Ordering information is given
on any current masthead page.
Conclusion
This work demostrates that R-substituted ketones are ideal
starting materials for the synthesis of metallo-1,2-enedithiolates.
IC9602052