The unperturbed oxo-sulfido functional group cis-MoVIOS related to that in the xanthine oxidase family of molybdoenzymes: synthesis, structural characterization, and reactivity aspects

Article abstract of DOI:10.1021/ic990440v

The oxo-sulfido functional group cis-Mo^VIOS is essential to the activity of the xanthine oxidase family of enzymes but has proven elusive to synthesis in molecules containing no other four-electron ligands. A direct route to molecules containing this group has been achieved. The reaction system [MoO₂(OSiPh₃)₂]/L in dichloromethane yields the complexes [Mo_VIO₂(OSiPh₃)₂L] (L = phen (1), Me₄phen (2), 4,4′-Me₂bpy (3), 5,5′-Me₂bpy (4), 2 py (5)) (74-96%), which are shown to have a distorted octahedral structure of crystallographically imposed C₂ symmetry (1, 5) with cis oxo and trans silyloxy ligands. The related reaction system [MoO₃S]^2-/2Ph₃SiCl/L in acetonitrile affords the complexes [Mo^VIOS(OSiPh₃)₂L] (L = phen (6), Me₄phen (7), 4,4′-Me₂bpy (8), 5,5′-Me₂-bpy (9)) (36-69%). From the collective results of elemental analysis, mass spectrometry, ₁H NMR, and X-ray structure determinations (6,7), complexes 6-9 are shown to contain the cis-Mo^VIOS group in molecules with the same overall stereochemistry as dioxo complexes 1-5. The crystal structures of 6 and 7 exhibit O/S disorder, which was modeled in refinements with 50% site occupancies. The Mo=S (1.607(5) (6), 1.645(5) (7) A) and Mo=S (2.257(3) (6), 2.203(2) (7) ?) bond distances obtained in this way are somewhat shorter and longer, respectively, than expected. Distances obtained by molybdenum EXAFS analysis using the GNXAS protocol for 6-9 (Mo=O 1.71-1.72 ?; Mo=S 2.18-2.19 ?) are considered more satisfactory and are in good agreement with EXAFS values for xanthine oxidase. Molybdenum K-edge data for 1 and 6-9 are reported. Reaction of 7 with Ph₃P in dichloromethane results in sulfur abstraction and formation of [Mo^vOCl(OSiPh₃)₂(Me₄phen)] (10), which has a distorted octahedral structure with cis O/C1 and cis silyloxy ligands. Sulfur rather than oxygen abstraction is favored by relative Mo=O/Mo=S bond strengths. Complexes 6-9 should allow exploration of the biologically significant cis-Mo^VIOS group.

Full text of DOI:10.1021/ic990440v

4104
Inorg. Chem. 1999, 38, 4104-4114
The Unperturbed Oxo-Sulfido Functional Group cis-MoVIOS Related to That in the
Xanthine Oxidase Family of Molybdoenzymes: Synthesis, Structural Characterization, and
Reactivity Aspects
Anders Thapper,†,‡ James P. Donahue,† Kristin B. Musgrave,§ Michael W. Willer,†
Ebbe Nordlander,‡ Britt Hedman,*,§ Keith O. Hodgson,*,§ and R. H. Holm*,†
Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138,
Department of Chemistry and Stanford Synchrotron Radiation Laboratory, SLAC,
Stanford University, Stanford, California, and Inorganic Chemistry I, Chemical Center,
Lund University, S-22100 Lund, Sweden
ReceiVed April 23, 1999
The oxo-sulfido functional group cis-MoVIOS is essential to the activity of the xanthine oxidase family of enzymes
but has proven elusive to synthesis in molecules containing no other four-electron ligands. A direct route to
molecules containing this group has been achieved. The reaction system [MoO2(OSiPh3)2]/L in dichloromethane
yields the complexes [MoVIO2(OSiPh3)2L] (L ) phen (1), Me4phen (2), 4,4′-Me2bpy (3), 5,5′-Me2bpy (4), 2 py
(5)) (74-96%), which are shown to have a distorted octahedral structure of crystallographically imposed C2
symmetry (1, 5) with cis oxo and trans silyloxy ligands. The related reaction system [MoO3S]2-/2Ph3SiCl/L in
acetonitrile affords the complexes [MoVIOS(OSiPh3)2L] (L ) phen (6), Me4phen (7), 4,4′-Me2bpy (8), 5,5′-Me2-
1
bpy (9)) (36-69%). From the collective results of elemental analysis, mass spectrometry, H NMR, and X-ray
structure determinations (6, 7), complexes 6-9 are shown to contain the cis-MoVIOS group in molecules with the
same overall stereochemistry as dioxo complexes 1-5. The crystal structures of 6 and 7 exhibit O/S disorder,
which was modeled in refinements with 50% site occupancies. The ModO (1.607(5) (6), 1.645(5) (7) Å) and
ModS (2.257(3) (6), 2.203(2) (7) Å) bond distances obtained in this way are somewhat shorter and longer,
respectively, than expected. Distances obtained by molybdenum EXAFS analysis using the GNXAS protocol for
6-9 (ModO 1.71-1.72 Å; ModS 2.18-2.19 Å) are considered more satisfactory and are in good agreement
with EXAFS values for xanthine oxidase. Molybdenum K-edge data for 1 and 6-9 are reported. Reaction of 7
with Ph3P in dichloromethane results in sulfur abstraction and formation of [MoVOCl(OSiPh3)2(Me4phen)] (10),
which has a distorted octahedral structure with cis O/Cl and cis silyloxy ligands. Sulfur rather than oxygen
abstraction is favored by relative ModO/ModS bond strengths. Complexes 6-9 should allow exploration of the
biologically significant cis-MoVIOS group.
Introduction
oxidase,1,7,11 the sulfido ligand in the oxidized site [MoVIOS-
(S2pd)(OH)] acts as a hydride acceptor in the course of
nucleophilic attack of hydroxide on the substrate, resulting in
the formation of the MoIV(O)SH unit and product. The reduced
site is then oxidized by two electrons to regenerate the oxidized
catalytic site. The aldehyde oxidoreductase from DesulfoVibrio
gigas (Dg) is the only member of the xanthine oxidase family
characterized by crystallography. The structures of the cofactor
ligand and the oxidized and reduced forms of the active Dg
enzyme are shown schematically in Figure 1. The distorted
square pyramidal stereochemistry with the sulfido/hydrosulfido
ligand in the axial position was established by crystallography.7
BonddistancesarefromEXAFSanalysesofxanthineoxidase.1,4-6
An inactive form of Dg aldehyde oxidoreductase contains the
MoVIO2 group in which oxo takes the place of the sulfido
ligand.12
In their fully oxidized forms, members of the xanthine oxidase
family of enzymes contain the oxo-sulfido functional group
cis-MoVIOS chelated by one pterin dithiolene cofactor ligand
(S2pd),1-3 a matter directly verified by EXAFS4-6 and X-ray
crystallography.7 The group is essential for catalytic activity;
the desulfo form is inactive.8,9 Enzyme activity can be recon-
stituted, however, in the presence of sulfide under reducing
conditions.10 In a recent mechanistic proposal for xanthine
† Harvard University.
‡ Lund University.
§ Stanford University.
(1) Hille, R. Chem. ReV. 1996, 96, 2757.
(2) Roma˜o, M. J.; Huber, R. Struct. Bonding 1998, 90, 69.
(3) For a useful comparison of molybdenum enzyme structures, see:
Kisker, C.; Schindelin, H.; Baas, D.; Re´tey, J.; Meckenstock, R. U.;
Kroneck, P. M. H. FEMS Microbiol. ReV. 1999, 22, 503.
(4) Cramer, S. P.; Wahl, R.; Rajagopalan, K. V. J. Am. Chem. Soc. 1981,
103, 7721.
While the structural and reactivity features of the related
biologically relevant units MoVIO2, MoIV,VO, and MoIV,VS are
relatively well developed,13,14 the cis-MoVIOS group, as dis-
(5) Cramer, S. P.; Hille, R. J. Am. Chem. Soc. 1985, 107, 8164.
(6) Hille, R.; George, G. N.; Eidsness, M. K.; Cramer, S. P. Inorg. Chem.
1989, 28, 4018.
(7) Huber, R.; Hof, P.; Duarte, R. O.; Moura, J. J. G.; Moura, I.; Liu,
M.-Y.; LeGall, J.; Hille, R.; Archer, M.; Roma˜o, M. J. Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 8846.
(8) Massey, V.; Edmondson, D. J. Biol. Chem. 1970, 245, 6595.
(9) Gutteridge, S.; Tanner, S. J.; Bray, R. C. Biochem. J. 1978, 175, 887.
(10) Wahl, R. C.; Rajagopalan, K. V. J. Biol. Chem. 1982, 257, 1354.
(11) Hille, R.; Re´tey, J.; Bartlewski-Hof, U.; Reichenbecher, W.; Schink,
B. FEMS Microbiol. ReV. 1999, 22, 489.
(12) Roma˜o, M. J.; Archer, M.; Moura, I.; Moura, J. J. G.; LeGall, J.; Engh,
R.; Schneider, M.; Hof, P.; Huber, R. Science 1995, 270, 1170.
10.1021/ic990440v CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/20/1999

The Oxo-Sulfido Functional Group cis-MoVIOS
Inorganic Chemistry, Vol. 38, No. 18, 1999 4105
ecules related to the active sites of molybdo-26,27 and tung-
stoenzymes,26,28 we have sought methods for the preparation
of the cis-MoVIOS group incorporated in such complexes. Our
results, including synthesis, structural characterization by X-ray
diffraction and molybdenum EXAFS analysis, and certain
reactivity features, are reported here. In this initial investigation,
we have not required that the products of synthesis possess the
mono(dithiolene) ligation of the active sites of the xanthine
oxidase family (Figure 1).
Experimental Section
Preparation of Compounds. All reactions and manipulations were
performed under a pure dinitrogen atmosphere using modified Schlenk
techniques or an inert-atmosphere box. Commercial samples of 4,4′-
dimethyl-2,2′-bipyridine and 5,5′-dimethyl-2,2′-bipyridine were sub-
limed (160 °C/0.1 Torr). Other reagents were of commercial origin
and were used as received. Acetonitrile, dichloromethane, pyridine, and
triethylamine were distilled from CaH2 and stored over 4 Å molecular
sieves. In IR spectra, νSiO features were assigned on the basis of position
and intensity; νMoO bands could not always be located or unambiguously
assigned.
Figure 1. Structures of the pterin dithiolene cofactor (M ) Mo, W; R
is absent or is a nucleotide) and the reduced and oxidized active sites
of D. gigas aldehyde oxidoreductase, a member of the xanthine oxidase
family. The square pyramidal stereochemistry with the sulfur atom in
an axial position was established by protein crystallography.7 Bond
distances are from EXAFS of xanthine oxidase.1,4-6
[MoO2(OSiPh3)2(phen)]. A solution of 0.097 g (0.538 mmol) of
1,10-phenanthroline (phen) in 1 mL of dichloromethane was added to
a solution of 0.280 g (0.413 mmol) of [MoO2(OSiPh3)2]29 in 3 mL of
dichloromethane. A white precipitate appeared within seconds. Ether
(10 mL) was added to the mixture to complete the precipitation. The
solid was collected by filtration, washed with 10 mL of acetonitrile
and 10 mL of ether, and dried in vacuo. The product was obtained as
0.342 g (96%) of a white solid. IR (KBr): νSiO 939 cm-1 (br), νMoO
908, 897 cm-1. Absorption data (CH2Cl2): λmax (ꢀM) 272 (24 200), 294
cussed previously,13 has proved to be an elusive synthetic target
in five- or six-coordinate molecules where, as in the enzymes,
the remaining binding sites are occupied by two-electron
ligands.15-20 This group may be present in six-coordinate [MoV-
OS(N2S2)], for which, however, there is no XAS evidence for
a conventional ModS bond (ca. 2.15 Å).21 In a material
formulated as a physical mixture of [MoO2Cl2(OPPh3)2] and
[MoOSCl2(OPPh3)2], the group is indicated as present in one
component with Mo-O ) 1.67(1) Å and Mo-S ) 2.249(7)
Å.22 It is present in [(HB(Me2pz)3)MoOS(η1-S2PR2)] (R ) Pri,
Ph)23,24 and in the organometallic complex [(C5Me5)MoOS(CH2-
SiMe3)], which is disordered in the crystalline state.25 In
[(HB(Me2pz)3)MoOS(η1-S2PPri2)], terminal sulfido ligation is
perturbed (and presumably stabilized) by the intramolecular
1
(sh, 10 600), 328 (sh, 1100) nm. H NMR (CDCl3): δ 9.28 (dd, 2,9-
H), 8.09 (dd, 4,7-H), 7.59 (s, 5,6-H), 7.41 (dd, 3,8-H). Anal. Calcd for
C48H38MoN2O4Si2: C, 67.12; H, 4.46; Mo, 11.17; N, 3.26. Found: C,
66.95; H, 4.41; Mo, 11.28; N, 3.18.
[MoO2(OSiPh3)2(Me4phen)]. A solution of 0.088 g (0.372 mmol)
of 3,4,7,8-tetramethyl-1,10-phenanthroline (Me4phen) in 1 mL of
dichloromethane was added to a solution of 0.233 g (0.343 mmol) of
[MoO2(OSiPh3)2] in 2 mL of dichloromethane. Addition of 15 mL of
ether gave a white precipitate, which was collected and washed as in
the above procedure. The product was obtained as 0.285 g (91%) of a
1
white solid. IR (KBr): νSiO 941 cm-1 (br), νMoO 910, 897 cm-1. H
cyclic interaction ModS‚‚‚SdP(S)R2 with ModS and S‚‚‚S
distances of 2.227(2) and 2.396(3) Å, respectively.23 The
fundamental chemical and structural properties of the cis-MoVI-
OS group could be examined in more detail than presently
feasible if it were available, preferably in an unperturbed
condition, in five- or six-coordinate complexes. As part of our
ongoing investigation of the synthesis and properties of mol-
NMR (CDCl3): δ 9.04 (s, 2,9-H), 7.78 (s, 5,6-H), 2.54, 2.28 (2 s,
3,4,7,8-Me). Anal. Calcd for C52H46MoN2O4Si2: C, 68.25; H, 5.07;
Mo, 10.48; N, 3.06. Found: C, 68.34; H, 5.06; Mo, 10.63; N, 3.03.
[MoOS(OSiPh3)2(phen)]. A solution of 0.540 g (1.83 mmol) of Ph3-
SiCl in 10 mL of acetonitrile was added dropwise to a stirred suspension
of 0.260 g (1.02 mmol) of K2[MoO3S]30 and 0.290 g (1.61 mmol) of
phen in 5 mL of acetonitrile containing 1 mL of triethylamine. The
reaction mixture was stirred for 1 h, during which an orange precipitate
and a light red solution formed. The mixture was reduced to dryness
in vacuo, the residue was dissolved in 200 mL of THF, and the solution
was filtered. The filtrate volume was reduced to ca. 125 mL, at which
point an orange precipitate began to form. Ether (100 mL) was added,
and the mixture was maintained at -20 °C overnight. The solid was
collected by filtration, washed with acetonitrile (2 × 5 mL) and ether
(2 × 5 mL), and dried in vacuo to afford the product as 0.550 g (69%)
of an orange solid. The product can be recrystallized from dichlo-
romethane/ether for an analytical sample. IR (KBr): νSiO 924 cm-1
(br), νMoO 920 or 895 cm-1. Absorption data (CH2Cl2): λmax (ꢀM) 272
(13) Holm, R. H. Coord. Chem. ReV. 1990, 100, 183.
(14) Enemark, J. H.; Young, C. G. AdV. Inorg. Chem. 1994, 40, 1.
2-
(15) This group is commonly encountered in [MoOnS4-n
]
(n ) 1-3),
[(R2NO)2MoOS]2-, and related tetrahedral species with four-electron
ligands, examples of which have been structurally characterized.16-20
(16) Krebs, B.; Mu¨ller, A.; Kindler, E. Z. Naturforsch. 1970, 25B, 222.
(17) Kutzler, F. W.; Scott, R. A.; Berg, J. M.; Hodgson, K. O.; Doniach,
S.; Cramer, S. P.; Chang, C. H. J. Am. Chem. Soc. 1981, 103, 6083.
(18) Wieghardt, K.; Hahn, M.; Weiss, J.; Swiridoff, W. Z. Anorg. Allg.
Chem. 1982, 492, 164.
(19) Bristow, S.; Collison, D.; Garner, C. D.; Clegg, W. J. Chem. Soc.,
Dalton Trans. 1983, 2495.
(20) Coucouvanis, D.; Toupadakis, A.; Lane, J. D.; Koo, S. M.; Kim, C.
G.; Hadjikyriacou, A. J. Am. Chem. Soc. 1991, 113, 5271.
(21) Singh, R.; Spence, J. T.; George, G. N.; Cramer, S. P. Inorg. Chem.
1989, 28, 8. N2S2 is a tetradentate dianionic ligand.
(22) Romanenko, G. V.; Podberezskaya, N. V.; Fedin, V. P.; Geras’ko, O.
A.; Fedorov, V. E.; Bakakin, V. V. J. Struct. Chem. 1988, 29, 79.
(23) Eagle, A. A.; Laughlin, L. J.; Young, C. G.; Tiekink, E. R. T. J. Am.
Chem. Soc. 1992, 114, 9195. HB(Me2pz)3 ) hydrotris(3,5-dimeth-
ylpyrazolyl)borate(1-).
(24) Young, C. G.; Laughlin, L. J.; Colmanet, S.; Scrofani, S. D. B. Inorg.
Chem. 1996, 35, 5368.
(25) Faller, J. W.; Ma, Y. Organometallics 1989, 8, 609.
1
(28 100), 295 (sh, 15 000), 315 (sh, 8840), 426 (370) nm. H NMR
(26) Donahue, J. P.; Lorber, C.; Nordlander, E.; Holm, R. H. J. Am. Chem.
Soc. 1998, 120, 3259.
(27) Donahue, J. P.; Goldsmith, C. R.; Nadiminti, U.; Holm, R. H. J. Am.
Chem. Soc. 1998, 120, 12869.
(28) Lorber, C.; Donahue, J. P.; Goddard, C. A.; Nordlander, E.; Holm, R.
H. J. Am. Chem. Soc. 1998, 120, 8102.
(29) Huang, M.; DeKock, C. W. Inorg. Chem. 1993, 32, 2287.
(30) Mu¨ller, A.; Dornfeld, H.; Schulze, H.; Sharma, R. C. Z. Anorg. Allg.
Chem. 1980, 468, 193.

4106 Inorganic Chemistry, Vol. 38, No. 18, 1999
Thapper et al.
Chart 1. Designation of Molybdenum Complexes
(CDCl3): δ 9.80, 9.10 (2 dd, 2,9-H), 8.14, 8.11 (dd, 4,7-H), 7.66, 7.63
(d, 5,6-H), 7.43 (m, 3,8-H). Anal. Calcd for C48H38MoN2O3SSi2: C,
65.89; H, 4.38; Mo, 10.96; N, 3.20; S, 3.66. Found: C, 65.72; H, 4.43;
Mo, 10.85; N, 3.16; S, 3.60.
[MoOS(OSiPh3)2(Me4phen)]. A solution of 0.293 g (0.994 mmol)
of Ph3SiCl in 4 mL of acetonitrile was added dropwise to a stirred
suspension of 0.134 g (0.527 mmol) of K2[MoO3S] and 0.130 g (0.550
mmol) of Me4phen in 5 mL of acetonitrile containing 1 mL of
triethylamine. An orange precipitate and a red-brown solution formed.
The mixture was stirred for 1 h, and the product was isolated by
following the procedure in the previous preparation. The product was
obtained as 0.202 g (41%) of an orange solid, which can be
recrystallized from dichloromethane/ether. IR (KBr): νSiO 931 cm-1
,
νMoO 920 or 901 cm-1. Absorption data (CH2Cl2): λmax (ꢀM) 283
(29 000), 304 (sh, 10 000), 333 (1200), 434 (340) nm. 1H NMR
(CDCl3): δ 9.58, 8.82 (2 s, 2,9-H), 7.84, 7.82 (2 d, 5,6-H), 2.57, 2.56,
2.34, 2.29 (4 s, 3,4,7,8-Me). Anal. Calcd for C52H46MoN2O3SSi2: C,
67.08; H, 4.98; Mo, 10.30; N, 3.01; S, 3.44. Found: C, 66.92; H, 5.02;
Mo, 10.37; N, 3.08; S, 3.46.
[MoOCl(OSiPh3)2(Me4phen)]. A solution of 34.5 mg (0.037 mmol)
of [MoOS(OSiPh3)2(Me4phen)] in 2 mL of dichloromethane was treated
with 9.70 mg (0.037 mmol) of PPh3 in 0.5 mL of dichloromethane.
The orange solution turned red within minutes; ether (40 mL) was added
until precipitation occurred. The solid was collected by filtration and
dried in vacuo, affording the product as 21.7 mg (63%) of a red solid.
The compound may be further purified by recrystallization from THF/
h. The solid was collected by filtration, washed with acetonitrile (4 ×
5 mL), and redissolved in 5 mL of dichloromethane, and the mixture
was filtered to remove KCl. The product was isolated by immediate
volume reduction of the filtrate in vacuo to minimize decomposition
that slowly occurs in solution.
(a) 4,4′-Me2bpy. The product was collected as 0.247 g (36%) of an
orange solid. Absorption data (CH2Cl2): λmax (ꢀM) 250 (sh, 27 900),
303 (sh, 20 200), 310 (21 200), 424 (486) nm. IR (KBr): 924 (vs, νSiO),
908 (sh, νMoO) cm-1. FAB-MS: m/z 880 (M+), 864 (M+ - O), 848
(M+ - S). 1H NMR (CDCl3): δ 9.49 (d, 1), 8.74 (d, 1), 7.23 (m, 18),
7.17 (s, 2), 7.11 (m, 12), 6.93 (dd, 2), 2.36 (s, 6). Anal. Calcd for
C48H42MoN2O3SSi2: C, 65.58; H, 4.82; Mo, 10.91; N, 3.19; S, 3.65.
Found: C, 65.75; H, 4.82; Mo, 10.73; N, 3.25; S, 3.85.
hexanes and obtained as red crystals. IR (KBr): νSiO 927 (s) cm-1
.
Absorption data (CH2Cl2): λmax (ꢀM) 269 (sh, 23 900), 284 (23 900),
308 (sh, 11 100), 329 (sh, 4020), 382 (sh, 973), 407 (sh, 709), 504
(524) nm. FAB-MS: m/z 935 (M+), 900 (M+ - Cl). Anal. Calcd for
C52H46ClMoN2O3Si2: C, 66.83; H, 4.96; Cl, 3.79; Mo, 10.27; N, 3.00.
Found: C, 67.04; H, 5.04; Cl, 3.85; Mo, 10.24; N, 2.88.
[MoO2(OSiPh3)2(Me2bpy)]. The same procedure was used for two
compounds. A solution of 0.150 g (0.221 mmol) of [MoO2(OSiPh3)2]
in 4 mL of dichloromethane was treated with a solution of 0.049 g
(0.265 mmol) of Me2bpy (4,4′- or 5,5′-dimethyl-2,2′-bipyridine). After
10 min, the colorless solution was filtered and the filtrate was layered
with 15 mL of ether. The product separated as colorless crystals, which
were collected by filtration, washed with acetonitrile (4 × 5 mL) and
ether (2 × 5 mL), and dried in vacuo.
(b) 5,5′-Me2bpy. The product was collected as 0.345 g (50%) of an
orange solid. Absorption data (CH2Cl2): λmax (ꢀM) 252 (16 400), 318
(12 500), 424 (464) nm. IR (KBr): 924 (vs, νSiO), 908, (sh, νMoO) cm-1
.
1
FAB-MS: m/z 880 (M+), 864 (M+ - O), 848 (M+ - S). H NMR
(CDCl3): δ 9.33 (s, 1), 8.64 (s, 1), 7.43 (m, 4), 7.23 (m, 18), 7.11 (m,
12), 2.23 (s, 3), 2.21 (s, 3). Anal. Found: C, 65.39; H, 4.93; Mo, 10.97;
N, 3.26; S, 3.54.
In the sections that follow, complexes are designated according to
Chart 1.
(a) 4,4′-Me2bpy. The product was collected as 0.162 g (85%). IR
(KBr): 933 (vs, νSiO), 912, 901 (s, νMoO) cm-1. Absorption spectrum
(CH2Cl2): λmax (ꢀM) 254 (14100), 298 (sh, 9420), 308 (10300) nm.
X-ray Structure Determinations. Crystallizations were conducted
anaerobically. Crystals of 1 (clear plates) and of 6 and 7 (orange blocks)
were grown by vapor diffusion of ether into dichloromethane solutions.
For the last two compounds, rigorously dry conditions were used.
Crystals of 5 (clear blocks) were grown by slow evaporation of an
acetonitrile solution. Crystals of 10 (red needles) were obtained by slow
diffusion of cyclohexane into a THF solution. Crystals were coated in
oil and mounted on a Siemens (Bruker) CCD area detector instrument
operated by the SMART software package. Data were collected at 213
K and were measured by using ω scans of 0.3° per frame, with 30 or
60 s frames, such that 1271 frames were collected for a hemisphere of
data. The first 50 frames were recollected at the end of the data
collection to monitor for decay; no appreciable decay was detected for
any compound. For 1, data were used only out to 2θ of 45° owing to
the weakness of higher angle reflections, but for 6 and 7, all data out
to 2θ of 56° were used, allowing for a final resolution of 0.76 Å. For
10, data extended to 2θ of 50°. Cell parameters were retrieved using
SMART software and refined using SAINT software on all observed
reflections between 2θ of 3° and the indicated upper thresholds. Data
reduction was performed with SAINT, which corrects for Lorentz-
polarization and decay effects. Absorption corrections were applied
using SADABS, as described by Blessing.31 For compounds 1, 5, 6,
and 7, analysis of systematic absences by the program XPREP identified
the space group as either Cc or C2/c. The latter was chosen over the
former in all four cases on the basis of successful structure solution
and refinement. The space group of 10 was unambiguously identified
as P21/c by systematic absences using XPREP. Crystal parameters are
collected in Tables 1 and 2.
1
FAB-MS: m/z 864 (M+), 848 (M+-O). H NMR (CDCl3): δ 8.94 (d,
1), 7.22 (m, 9), 7.09 (m, 6), 7.03 (s, 1), 6.93 (d, 1), 2.32 (s, 3). Anal.
Calcd for C48H42MoN2O4Si2: C, 66.81; H, 4.91; Mo, 11.12; N, 3.25.
Found: C, 66.67; H, 4.91; Mo, 11.06; N, 3.32.
(b) 5,5′-Me2bpy. The product was collected as 0.146 g (76%).
Absorption data (CH2Cl2): λmax (ꢀM) 258 (12 600), 319 (8950) nm.
FAB-MS: m/z 864 (M+), 848 (M+ - O). IR (KBr): 945 (vs, νSiO),
908 (s, νMoO) cm-1. 1H NMR (CDCl3): δ 8.83 (s, 1), 7.41 (d, 1), 7.30
(d, 1), 7.22 (m, 9), 7.10 (m, 6), 2.18 (s, 3). Anal. Found: C, 66.90; H,
4.86; Mo, 11.22; N, 3.21.
[MoO2(OSiPh3)2(py)2]. A solution of 0.050 g (0.074 mmol) of
[MoO2(OSiPh3)2] in 1 mL of dichloromethane was treated with 10 drops
of pyridine. After 10 min of stirring, the colorless solution was filtered
and the filtrate was layered with 4 mL of ether, causing precipitation
of a colorless crystalline solid. This material was collected by filtration,
washed with 2 × 5 mL of ether, and dried in vacuo to afford the product
as 0.066 g (89%) of colorless crystals. Absorption data (CH2Cl2): λmax
(ꢀM) 257 (7260) nm. IR (KBr): 903 (br, s). 1H NMR (CDCl3): δ 8.68
(d, 2), 8.03 (t, 1), 7.64 (d, 6), 7.59 (t, 2), 7.38-7.44 (m, 9). Anal. Calcd
for C46H40MoN2O4Si2: C, 66.01; H, 4.82; Mo, 11.46; N, 3.35. Found:
C, 65.88; H, 4.76; Mo, 11.42; N, 3.28.
[MoOS(OSiPh3)2(Me2bpy)]. The same procedure was used for two
compounds. A suspension of 0.200 g (0.787 mmol) of K2[MoO3S] in
5 mL of acetonitrile containing 0.5 mL of triethylamine was treated
with 0.261 g (1.42 mmol) of Me2bpy. To this mixture was added a
solution of 0.441 g (1.50 mmol) of Ph3SiCl in 5 mL of acetonitrile,
resulting in the appearance of a red-brown solution within 10 s and an
orange precipitate within 1 min. The mixture was stirred for at least 2
(31) Blessing, R. H. Acta Crystallogr. 1995, A51, 33.

The Oxo-Sulfido Functional Group cis-MoVIOS
Inorganic Chemistry, Vol. 38, No. 18, 1999 4107
Table 1. Crystallographic Dataa for [MoO2(OSiPh3)2(phen)] (1), [MoO2(OSiPh3)2(py)2] (5), [MoOS(OSiPh3)2(phen)] (6),
[MoOS(OSiPh3)2(Me4phen)] (7), and [MoOCl(OSiPh3)2(Me4phen)]‚C6H12 (10‚C6H12)
1
5
6
7
10‚C6H12
formula
fw
crystal system
space group
Z
C48H38MoN2O4Si2
858.92
monoclinic
C2/c
C46H40MoN2O4Si2
836.92
monoclinic
C2/c
C48H38MoN2O3SSi2
874.98
monoclinic
C2/c
C52H46MoN2O3SSi2
931.09
monoclinic
C2/c
C58H58ClMoN2O3Si2
1018.63
monoclinic
P21/c
4
4
4
4
4
a, Å
b, Å
c, Å
15.686(2)
9.758(1)
27.442(3)
105.033(3)
4056.7(9)
0.0503, 0.0755
10.0939(2)
15.8957(4)
25.8051(7)
98.056(1)
4099.6(2)
0.0339, 0.0755
15.7695(6)
9.7836(3)
27.7370(2)
103.714(2)
4157.3(2)
0.0411, 0.1028
16.7020(8)
10.3747(5)
27.039(1)
103.472(1)
4556.3(4)
0.0430, 0.0884
19.0483(9)
14.4601(7)
18.1676(8)
90.576(2)
5003.8(4)
0.0658, 0.1148
â, deg
V, Å3
R1,b wR2
c
a Obtained with graphite-monochromatized Mo KR (λ ) 0.710 73 Å) radiation at T ) 213 K. b R1 ) ∑|Fo| - |Fc|/ ∑|Fo|. c wR2 ) {∑[w(Fo
-
2
Fc2)2]/∑[w(Fo )2]}1/2
.
2
Table 2. Selected Interatomic Distances (Å) and Angles (deg) for
MoVIO2 (1, 5) and MoVI/VOXa (6, 7, 10) Complexes
reported. Use of the checking program PLATON did not identify any
missing or higher symmetry. Final agreement factors are given in Table
1.32
X-ray Absorption Spectroscopy. Solid samples of complexes 1 and
6-9 were ground into fine powders in a dinitrogen atmosphere and
diluted with boron nitride. The mixtures were pressed into pellets and
sealed between 63.5 µm Mylar tape windows in a 1 mm aluminum
spacer. The samples were frozen in liquid nitrogen immediately upon
removal from the glovebox and maintained at this or a lower
temperature throughout storage and data collection.
1
5
6g
7g
10‚C6H12
b
Mo-Ooxo
1.70(1) 1.695(2) 1.611(5) 1.645(5) 1.703(4)
2.257(3) 2.203(2) 2.424(2)
Mo-Xa
1.92(1) 1.931(1) 1.923(2) 1.932(2) 1.935(8)f
2.34(1) 2.404(4) 2.445(4) 2.41(2) 2.297(4)
2.194(3) 2.28(2)
c
Mo-OSi
d
Mo-NO
d
Mo-NS
Si-O
1.61(1) 1.601(1) 1.595(2) 1.618(2) 1.622(4)
106.5(9) 104.5(1)
b
Ooxo-Mo-Ooxo
Ooxo-Mo-Xa
Ooxo-Mo-OSi
′
(a) Data Collection. XAS data were measured at the Stanford
Synchrotron Radiation Laboratory (SSRL) on unfocused 8-pole wiggler
beamline 7-3 under ring conditions of 3.0 GeV and 70-100 mA. A
Si(220) double-crystal monochromator was used and detuned 50% at
21549 eV to minimize contamination from higher harmonics. An
Oxford Instruments CF1208 continuous-flow liquid helium cryostat
maintained a constant sample temperature of 10 K. Data were measured
in the transmission mode to k ) 20 Å-1 with argon as the absorbing
gas. Internal calibration was performed by simultaneous measurement
of the absorption edge of a molybdenum foil placed between a second
and a third ionization chamber. The first inflection point of the foil
spectrum was assigned to 20 003.9 eV. The data represent averages of
two or three scans for each sample with data reduction as described
previously in detail.33
(b) Data Analysis. The data analysis was performed using the ab
initio GNXAS method. The theoretical basis for the GNXAS approach
and its fitting methodologies have been described in detail elsewhere.34-36
The program code generates theoretical EXAFS signals on the basis
of an initial structural model. For the analysis reported here, the
crystallographic coordinates for complexes 1, 6, and 7 were used to
generate an initial structural model up to a distance cutoff of 5.0 Å.
For complexes 8 and 9, modifications were made to the crystal structure
of 6 (utilizing Chem3D Pro) to replace the phenanthroline ligand with
that of a known bipyridine structure and generate new coordinates.
Phase shifts were obtained using the standard muffin-tin approximation
to calculate the individual two-body and three-body EXAFS signals.
An initial model EXAFS spectrum was constructed by combining the
individual component signals and an appropriate background. This
model was then fit to the averaged raw absorption data by a least-
squares minimization program as described previously.36 If any of the
distances determined in the fits differed by more than 10% from those
in the initial starting model, the phase and amplitude parameters were
recalculated after modification of the model bond lengths using
103.3(2) 104.6(2) 95.4(1)
98.0(5) 98.65(7) 101.8(2) 100.7(3) 105.9(4)f
a
X-Mo-OSi
95.84(9) 99.56(9) 86.9(1)h
OSi-Mo-OSi′ c
Ooxo-Mo-Ne
Ooxo-Mo-N′ e
X-Mo-Na,e
X-Mo-N′ a,e
OSi-Mo-Ne
OSi-Mo-N′ e
N-Mo-N′ e
Mo-O-Si
153.1(6) 151.7(9) 153.0(1) 154.1(1) 90.8(2)
91.8(6) 90.2(1) 97.0(2) 96.3(7) 88.3(2)
161.7(6) 165.3(1) 167.1(2) 168.2(7) 159.5(2)
159.7(1) 158.5(6) 87.2(1)
89.6(1) 87.2(6) 80.6(1)
80.0(4) 78.8(1) 80.4(1) 81.2(7) 89.4(2)h
78.0(4) 78.9(1) 77.7(1) 77.8(7) 86.0(4)f
70.0(9) 75.1(2) 70.1(2) 72(1)
71.5(2)
160.4(6) 146.3(1) 157.8(1) 153.1(1) 156.6(5)f
b
c
a X ) S, Cl. Ooxo ) O(1), Ooxo′ ) O(1′). OSi ) O(2, 3), OSi′ )
O(2′, 4). d NO, NS ) trans to Ooxo, S. N ) N(1), N′ ) N(1′, 2). f Mean
e
values. g Disordered structures; see text. h Trans Cl-Mo-OSi ) 161.8(1)°,
trans OSi-Mo-N ) 161.9(2)°.
All structures were solved by direct methods with SHELXS and
subsequently refined against all data in the 2θ ranges by full-matrix
least-squares calculations based on F2. In the structures of 6 and 7, the
oxo and sulfido ligands as well as the phen and Me4phen ligands were
disordered and refined with site occupancy factors of 0.5 for all atoms.
The Ph3SiO ligands in these two compounds were also disordered
between two rotational configurations, which were modeled with site
occupancy factors determined as optimal fits by SHELXL. In the
structure of 5, a pyridine and three phenyl rings were disordered equally
over two positions. All non-hydrogen atoms were refined anisotropi-
cally, including those which are disordered. The carbon atoms of the
phen and Me4phen ligands in 6 and 7 and a pyridine ligand in 5,
although refined anisotropically, required the use of averaged thermal
parameters to maintain reasonably sized ellipsoids in structure presenta-
tions. Hydrogen atoms were attached at idealized positions on carbon
atoms and were refined as riding atoms with uniform values of Uiso
.
All structures converged in the final stages of refinement, showing no
movement in atom positions and no residual electron density >1 e/Å3.
Bond distances and angles were not constrained in refinements, except
for the two N-containing rings of the phen ligand in 6 and the pyridine
ring of 5 which were constrained to regular hexagons. Because of
disorder, data from two crystals of 6 obtained from independent
preparations were collected and fully refined. Values of all bond
distances and angles involving the molybdenum atom agreed to within
4σ. Results for the refinement with the marginally better R values are
(32) See also paragraph at the end of paper regarding Supporting Informa-
tion.
(33) DeWitt, J. G.; Bentsen, J. G.; Rosenzweig, A. C.; Hedman, B.; Green,
J.; Pilkington, S.; Papaefthymiou, G. C.; Dalton, H.; Hodgson, K. O.;
Lippard, S. J. J. Am. Chem. Soc. 1991, 113, 9219.
(34) Filipponi, A.; Di Cicco, A.; Tyson, T. A.; Natoli, C. R. Solid State
Commun. 1991, 78, 265.
(35) Westre, T. E.; Di Cicco, A.; Filipponi, A.; Natoli, C. R.; Hedman, B.;
Solomon, E. I.; Hodgson, K. O. J. Am. Chem. Soc. 1995, 117, 1566.
(36) Filipponi, A.; Di Cicco, A.; Natoli, C. R. Phys. ReV. B 1995, 52, 15122,
15135.

4108 Inorganic Chemistry, Vol. 38, No. 18, 1999
Thapper et al.
Chem3D Pro. Background subtraction was performed by applying a
flexible three-segment spline which was refined in the GNXAS fits.
The quality of the fit was calculated by the least-squares residual R
and monitored through visual inspection of the fits to the data and
Fourier transform (FT), the residual EXAFS signal, and its Fourier
transform. The structural parameters varied during the refinements were
2
36
the bond distance (R) and the bond variance (σR
)
in the case of a
two-body signal, the two shorter bond distances R1 and R2, the
intervening angle (θ), and the six covariance matrix elements for a
three-body signal. The nonstructural parameters Eo and So2 were varied,
whereas the ΓC (core hole lifetime) and Er (experimental resolution)
were kept fixed to physically reasonable values throughout the analysis.
The coordination numbers were initially set to the values determined
by X-ray crystallography for structures 1, 6, and 7 (complexes 8 and
9 were assumed to have the same coordination numbers as 6 and 7)
and were then systematically stepped through during the analysis as
necessary. All parameters were varied within a preset range, and all
results were checked to ensure that values obtained did not reach the
high or low point of these fitting ranges.
Results and Discussion
At least three synthetic routes to five- or six-coordinate
complexes containing the cis-MoVIOS group are evident: (i)
replacement of oxo by sulfido in a MoVIO2 complex; (ii)
oxidative addition of S(0) to an MoIVO complex and/or O(0)
to an MoIVS complex; (iii) derivatization of a species containing
an MoVIOS fragment. While route i has been accomplished in
the reaction [(R2NO)2MoO2]2- f [(R2NO)2MoOS]2- with H2S18
or B2S3,19 a similarly direct conversion has not been successfully
applied to molecules having only two four-electron ligands. In
a variation of this route, [(HB(Me2pz)3)MoVIO2(η1-S2PR2)] was
reduced to the MoVO2 state, one oxo group was substituted by
reaction with hydrosulfide, and the resultant species was
oxidized to [(HB(Me2pz)3)MoVIOS(η1-S2PR2)].37 Both proce-
dures of route ii have yielded the latter complexes.23,24 In our
approach, we utilized route iii in a manner suggested by two
prior observations. Treatment of Ag2MoO4 with Ph3SiCl affords
tetrahedral [MoO2(OSiPh3)2 ] in >90% yield.29 The coordinative
unsaturation of this species is made evident by the reaction with
Ph3P to give [MoO2(OSiPh3)2(PPh3)].29 These findings are
capable of extension in the form of the reactions in Figure 2.
MoVIO2 Complexes. (a) Synthesis and Properties. Assess-
ment of the properties of MoVIOS species requires the corre-
sponding dioxo complexes as references, particularly because
they are not subject to the same potential disorder problem in
the crystalline state. Initially, [MoO2(OSiPh3)2]29 was reacted
with phen, bpy, or py ligands to form the colorless dioxo
complexes 1-5, which are readily isolated in 74-96% yields.
These species show intense νSiO and νMoO stretches (when
observable) in the 890-945 cm-1 region and prominent parent
ion peaks in the FAB mass spectra (3, 4). They display 1H NMR
spectra indicative of C2 symmetry with deshielded 2,9-H (1, 2)
and 6,6′-H (3, 4) resonances at δ 8.8-9.3; the related pyridine
R-H signal of 5 occurs at δ 8.38. This effect is illustrated by
the spectra of 1 and 4 in Figure 3, where the 2,9-H (doublet of
doublets) and 6,6′-H (broadened singlet) signals appear at δ 9.28
and 8.83, respectively.
Figure 2. Preparation of Me4phen complexes 2, 7, and 10. Reactions
affording MoVIO2 complexes 1 and 3-5 and MoVIOS complexes 6, 8,
and 9 are analogous. The MoIVO species is a proposed intermediate in
the reduction of 7 by oxo transfer.
Figure 3. 1H NMR spectra of [MoO2(OSiPh3)2(phen)], [MoOS(OSiPh3)2-
(phen)], [MoO2(OSiPh3)2(5,5′-Me2bpy)], and [MoOS(OSiPh3)2(5,5′-
Me2bpy)] in CDCl3 solutions, emphasizing the absence of C2 symmetry
in the MoVIOS complexes. Chemical shifts are indicated; arrows in
two spectra mark trace impurities of dioxo complexes.
(b) X-ray Structures. The X-ray structures of 1 and 5 are
presented in Figure 4; selected metric data are compiled in Table
2. These compounds crystallize in monoclinic space group
C2/c; the asymmetric unit consists of one half the molecule,
with the other half generated by a C2 axis that bisects the
O-Mo-O and N-Mo-N angles. Given the structure of 5, the
chelate ring plays no role in setting the overall stereochemistry.
Figure 4. Structures of [MoO2(OSiPh3)2(phen)] (1) and [MoO2(OSiPh3)2-
(py)2] (5), showing 50% probability ellipsoids and atom-labeling
schemes. Primed and unprimed atoms are related by a 2-fold axis.
Indeed, these two complexes (and, by implication, 2-4) exhibit
the conventional stereochemistry of [MoVIO2L2L′2] molecules,
in which anionic ligands L′ are mutually trans and cis to oxo
(37) Laughlin, L. J.; Young, C. G. Inorg. Chem. 1996, 35, 1050.

The Oxo-Sulfido Functional Group cis-MoVIOS
Inorganic Chemistry, Vol. 38, No. 18, 1999 4109
Figure 5. FAB+ mass spectrum of [MoOS(OSiPh3)2(4,4′-Me2bpy)]
and the calculated spectrum in the parent ion region.
and neutral ligands L. Dimensions of the MoO2 group are
essentially identical and are normal. Owing to charge repulsion,
the silyloxy groups are bent back from the oxo ligands, causing
O-Mo-O bond angles of 152-154°. The Mo-O-Si fragments
are also nonlinear (146-160°), as in [MoO2(OSiPh3)2]29 and
other molybdenum complexes.27 The Mo-O-Si angles open
toward the MoO2 group, and in 1 and 5, the Mo-O-Si plane
exactly bisects the MoO2 angle. Because in 5 the pyridine rings
are rotated out of the Mo(O1)(O1′) plane (dihedral angle 68.7°),
the smaller Mo-O-Si angle (146.3(1)°) serves to reduce steric
interactions between the pyridine rings and the SiPh3 groups.
MoVIOS Complexes. (a) Synthesis and Properties. Het-
erogeneous reaction of K2[MoO3S], Ph3SiCl, and phen or bpy
in acetonitrile results in the formation of the desired complexes
6-9 in 36-69% yields as orange solids (Figure 2). Compound
identification is based on elemental analyses, mass spectra (8,
9), 1H NMR spectra, X-ray structure determinations (6, 7), and
molybdenum K-edge EXAFS analyses. All compounds show
the correct sulfur content, indicating no appreciable contamina-
tion with dioxo species. The mass spectra of 8 in Figure 5 and
of 9 (not shown) reveal in the high m/z region the parent ion
and two fragments resulting from the loss of one oxygen or
sulfur atom. The NMR spectra of 6 and 9 in Figure 3 disclose
unsymmetrically coordinated phen and 5,5′-Me2bpy ligands,
indicative of the absence of C2 symmetry present in dioxo
counterparts 1 and 4. Complexes 7 (0.76) and 8 (0.75) show
the same behavior with the indicated chemical shift differences
(ppm) between the 2,9-H and 6,6′-H signals, respectively. The
orange color of the four complexes arises from absorption bands
at 424-434 nm shown in Figure 6. These features are absent
in the dioxo complexes and arise, therefore, from S f Mo
charge transfer. The intensity of these bands (ꢀM ≈ 400-500)
is sufficiently low that they would be obscured by other
chromophores (flavins, iron-sulfur clusters) that occur in the
xanthine oxidase enzyme family.
Figure 6. Absorption spectra of MoVIO2 and MoVIOS complexes in
dichloromethane solutions. Band maxima are indicated.
Figure 7. Structures of [MoOS(OSiPh3)2(phen)] (6) and [MoOS-
(OSiPh3)2(Me4phen)] (7) showing 50% probability ellipsoids and atom-
labeling schemes. Primed and unprimed atoms are related by a 2-fold
axis.
oxo and sulfido ligands are not symmetrically disposed on either
side of the C2 axis, such that when the symmetry-equivalent
atoms (O′, S′) are generated, they do not overlap the original
atoms. Thus the atoms O, S, O′, S′ are four separate resolved
peaks in the electron density maps, which were refined
anisotropically as atoms of half-site occupancy. The presence
of the cis-MoVIOS group requires that the phen and Me4phen
ligands be unsymmetrically coordinated. In both cases, the
Mo-N distance trans to the oxo ligand (Mo-NO) is apparently
lengthened by ca. 0.2 Å compared to the Mo-N distance trans
to the sulfido ligand (Mo-NS) (Table 2). This skewed ligand
binding coupled with the presence of the C2 axis necessitates
that the phen and Me4phen ligands be disordered between two
highly overlapping positional variants. All carbon atoms of these
disordered ligands are discernible in electron density maps. They
(b) X-ray Structures. Complexes 6 and 7 also crystallize in
space group C2/c; the unit cells of 1 and 6 are nearly isometric.
Structures are presented in Figure 7, from which it is evident
that these molecules possess the same overall stereochemistry
as the dioxo species. The imposed C2 symmetry at the
molybdenum special position requires that the crystal contain
a 1:1 mixture of identical molecules distinguished only by the
relative orientation of the cis-MoVIOS group. Fortuitously, the

4110 Inorganic Chemistry, Vol. 38, No. 18, 1999
Thapper et al.
were refined anisotropically as atoms with half-site occupancies
and averaged thermal parameters. Bond distances and angles
for 6 and 7 refined as described are contained in Table 2.
The foregoing evidence collectively demonstrates the presence
of the cis-MoVIOS group in complexes 6-9, albeit with apparent
differences in certain metric parameters as obtained by the
refinement procedure: ModO 1.607(5) (6), 1.645(5) (7) Å;
ModS 2.257(3) (6), 2.203(2) (7) Å. The ModO distances are
0.05-0.09 Å shorter than the values in the dioxo complexes
(Table 2). The ModS bond lengths are also somewhat long
when compared to tetrahedral MoVIdS bonds (2.10-2.18
Å),16-20,38 consistent with the few six-coordinate values avail-
able, viz. for [MoOSCl2(OPPh3)2]22 and [(HB(Me2pz)3)MoOS-
(η1-S2PR2)].23 In both structures, the closest intermolecular and
nonbonded intramolecular distances are not less than 2.70 Å
and involve hydrogen atoms; consequently, the MoVIOS group
is unperturbed. The Mo-NO/NS bond distances differ by 0.27
Å in 5 and 0.14 Å in 6. The longer distances are for bonds
trans to the oxo ligands. The formation of a relatively short
and strong Mo-NS bond leads to a longer and weaker Mo-NO
bond. The effect arises at least in part because of the rigidity
and small bite angle (ca. 70°) of the phen ligands, which prevent
expression of the intrinsic trans influence of oxo and sulfido
ligands in bonds disposed at or near 90°.39 Note that, in [(HB-
(Me2pz)3)MoOS(η1-S2PR2)] (0.10 Å)23 and [(HB(Me2pz)3)WVI-
OS((-)-mentholate)] (0.05 Å),40 where the tridentate ligand is
more flexible, and in [WVIOS(NCS)4]2- (0.02 Å),41 the dif-
ferential O/S trans influence on M-N bonds, as expressed by
bond length differences, is much less. The distorted square
pyramidal [MoVIOS(S2pd)(OH2)] site of D. gigas aldehyde
oxidoreductase places the sulfido ligand in the axial position
and the remaining ligands in the equatorial plane.7 Structural
resolution is insufficient to detect any oxo trans influence on
dithiolene binding.
Terminal oxo-sulfido ligand disorder is well documented
and has been manifested in different modes: within nonlin-
ear25,42,43 and linear44 MOS groups, over the sites of two different
metals in the same molecule,45,46 and in the formation of solid
solutions47,48 (as with (Me4N)2[MoS3Q] (Q ) O, S)48). Potential
disorder is sometimes circumvented by differentiating interac-
tions, such as the intramolecular effect ModS‚‚‚SdP in the
molecular lattice of [(HB(Me2pz)3)MoOS(η1-S2PPri2)]23 and
cation-anion interactions in the ionic lattice of (NH4)2-
[WO2S2].49 Relatively large molecules with pseudo-2-fold
symmetry, in which the MOS group is shielded from the
crystalline environment and any attendant differentiating inter-
actions, are favorable cases for disorder. That is the condition
of 6 and 7, in which protruding phenyl rings provide frontside
hindrance to the MoVIOS group. While the existence of this
group in 6-9 is not in question, the accuracy of its bond lengths
is uncertain, especially when the differences in the ModO and
ModS distances are comparable to the resolution of the X-ray
experiment (λ/(2sinθmax) ≈ 0.5 Å). Any uncertainties in these
distances would not appear to arise from dioxo contamination,
which would have the effect of moving together overlapping
electron densities. If ModO ) 1.70 Å from dioxo structures is
adopted, the presence of the MoVIO2 group would lengthen the
ModO and contract the ModS distance. To obtain metric
parameters of the MoVIOS group uncomplicated by disorder,
we utilized X-ray absorption spectroscopy.
XAS Spectroscopy. Five complexes were examined, one of
the MoVIO2 type (1) and the four MoVIOS species (6-9). Fit
parameters and derived bond distances and angles are compiled
in Table 3. Relevant spectra are presented in Figures 8-11.
(a) EXAFS Analysis. The EXAFS data and Fourier trans-
forms of complexes 6 and 7 are essentially superimposable to
the end of the high-k range (20 Å-1), as are those of 8 and 9.
It therefore suffices to focus on the results for MoVIO2 complex
1 and MoVIOS complexes 7 and 9. As shown in Figure 8, there
are significant differences in the EXAFS beat patterns, particu-
larly in the k ∼ 7 and k ∼ 13 Å-1 regions for 1 as compared to
7 and 9. This is also reflected in the Fourier transforms in Figure
9, where it is seen that the intensities of all peaks of 1 differ
from those of 7 and 9, which themselves are quite similar. The
intensity of the first peak of 1, which corresponds to ModO
interactions, has increased significantly relative to those of the
other two complexes, in agreement with the presence of two
oxo ligands. The second and third peaks have lower intensities
for 1, an effect attributable to the absence of ModS ligation.
For complex 1, the second and third peaks represent Mo-O
and Mo-N interactions, respectively. In addition to these
contributions, the ModS interaction present in 7 and 9
contributes substantially to the intensities of both the second
and third peaks. Sulfur is a very strong backscatterer, and its
presence often dominates, especially in molecules with weak
scattering ligands such as oxygen and nitrogen.
For complex 1, the EXAFS fit results are in outstanding
agreement with the crystal structure (Table 2), with single-
scattering contributions from two ModO, two Mo-O, and two
Mo-N distances at 1.72, 1.94, and 2.36 Å, respectively.
Excellent fits to the data for 6 and 7 were achieved by including
only single-scattering interactions. The results indicate the
presence of one short ModO distance of 1.71-1.72 Å, two
longer Mo-O distances of 1.95 Å, a short ModS distance of
2.19 Å, and two longer Mo-N interactions at a distance of 2.37
Å. The possibility of a split outer Mo-N shell, as indicated in
the X-ray work, was tested in further refinements but did not
lead to significant improvements in R values. The two distances
determined (2.34 and 2.41 Å for 6 and 2.33 and 2.42 Å for 7)
are not quite separable within the resolution of the EXAFS
technique, which is approximated by the relation50 ∆R ) π/2∆k
for scatterers of the same atomic number and which for the
fitting range 3.5-19.5 Å-1 would correspond to a ∆R of 0.10
Å. At this long distance, the Mo-N waves are, however, not
dominant contributors to the total EXAFS signal and thus are
not easily determined. For complexes 8 and 9, the same
coordination spheres at distances essentially identical to those
seen in 6 and 7 were established. Individual EXAFS contribu-
tions and the total EXAFS signals and fits for 7 and 9 are shown
in Figure 10. The excellent fits to the data evident here apply
(38) Lapasset, J.; Chezeau, N.; Belougne, P. Acta Crystallogr. 1976, B32,
3087.
(39) To test the effect of ring formation on Mo-N distances, the sulfido
derivative of 5 was sought. The reaction product proved to be
somewhat unstable, and suitable crystals were not obtained.
(40) Eagle, A. A.; Harben, S. M.; Tiekink, E. R. T.; Young, C. G. J. Am.
Chem. Soc. 1994, 116, 9749.
(41) Potvin, C.; Manoli, J. M.; Marzak, S.; Secheresse, F. Acta Crystallogr.
1988, C44, 369.
(42) Coucouvanis, D.; Al-Ahmad, S.; Kim, C. G.; Mosier, P. E.; Kampf,
J. W. Inorg. Chem. 1993, 32, 1533.
(43) Kawaguchi, H.; Yamada, K.; Lang, J.-P.; Tatsumi, K. J. Am. Chem.
Soc. 1997, 119, 10346.
(44) Cotton, F. A.; Schmid, G. Inorg. Chem. 1997, 36, 2267.
(45) Xin, X.; Morris, N. L.; Jameson, G. B.; Pope, M. T. Inorg. Chem.
1985, 24, 3482.
(46) Coucouvanis, D.; Koo, S.-M. Inorg. Chem. 1989, 28, 2.
(47) Do, Y.; Simhon, E. D.; Holm, R. H. Inorg. Chem. 1985, 24, 2827.
(48) Serezhkin, V. N. Russ. J. Inorg. Chem. (Engl. Transl.) 1977, 22, 845.
(49) Gonschorek, W.; Hahn, T.; Mu¨ller, A. Z. Kristallogr. 1973, 138, 380.
(50) Shulman, R. G.; Eisenberger, P.; Blumberg, W. E.; Stombaugh, N.
A. Proc. Natl. Acad. Sci. U.S.A. 1975, 72, 4003.

The Oxo-Sulfido Functional Group cis-MoVIOS
Table 3. GNXAS Fit Results of Mo K-Edge EXAFSa
Inorganic Chemistry, Vol. 38, No. 18, 1999 4111
[MoVIO2-
[MoVIOS-
[MoVIOS(OSiPh3)2-
[MoVIOS(OSiPh3)2-
[MoVIOS(OSiPh3)2-
(OSiPh3)2(phen)], 1
(OSiPh3)2(phen)], 6
(Me4phen)], 7
(4,4′-Me2bipy)], 8
(5,5′-Me2bipy)], 9
E0
S0
20 018.4
0.953
1.72
2
0.001
1.94
2
20 017.7
0.954
1.72
1
0.001
1.95
2
0.005
2.19
1
0.002
2.37
2
20 017.9
0.953
1.71
1
0.001
1.95
2
0.004
2.19
1
0.002
2.37
2
20 017.6
0.951
1.71
20 017.0
0.950
1.71
2
R(ModO), Å
Na
1
1
σ2, Å2
0.002
1.94
2
0.002
2.19
1
0.002
2.37
2
0.005
1.65
2
0.001
1.94
2
0.003
2.18
1
0.001
2.40
2
0.003
1.68
2
R(Mo-O), Å
Na
σ2, Å2
0.003
R(ModS), Å
Na
σ2, Å2
R(Mo-N), Å
2.38
2
0.003
1.62
2
0.001
3.52
161 ( 2
1.33
Na
σ2, Å2
0.004
0.004
R(O-Si), Å
Na
σ2, Å2
0.001
3.53
159 ( 2
1.15
0.001
3.54
156 ( 3
1.12
R(Mo-OSi), Å
MoOSi, deg
R(N-C), Å
Na
2
2
2
σ2, Å2
0.002
3.23
0.001
3.28
0.001
3.28
R(Mo-NC), Å
MoNC, deg
R (fit)
120 ( 2
134 ( 4
134 ( 2
0.687 × 10-7
0.273 × 10-7
0.305 × 10-6
0.315 × 10-7
0.106 × 10-6
2
a In the fits, the following parameters were refined: E0, E0 , R(ModO, Mo-O, ModS, Mo-N, O-Si, N-C), and the corresponding σ2 values
as well as the six covariance matrix elements. The coordination numbers (N) were systematically varied throughout the analysis but were not
allowed to vary in a given fit. The Γ0 and experimental resolution parameters were fixed to physically reasonable values throughout the analysis.
See the text for definitions of the above-mentioned parameters. The estimated standard deviations in bond distances are on the order of (0.01-0.02
Å.
Figure 9. Comparison of the non-phase-shift-corrected Fourier trans-
forms of [MoO2(OSiPh)2(phen)] (1, s), [MoOS(OSiPh3)2(Me4phen)]
(7, ‚‚‚), and [MoOS(OSiPh3)2(5,5′-Me2bipy)] (9, - - -).
Figure 8. Comparison of the experimental EXAFS data (s) and fits
beat pattern remained in the EXAFS fit residual for all three
to the data (- - -) for [MoO2(OSiPh3)2(phen)] (1, top), [MoOS(OSiPh3)2-
complexes that was assumed to originate from Mo-N-C
(Me4phen)] (7, middle), and [MoOS(OSiPh3)2(5,5′-Me2bpy)] (9, bot-
scattering from either the phenanthroline (1) or the bipyridyl
tom). Significant differences in the EXAFS beat pattern around k ) 7
and k ) 13 Å-1 for 1 vs 7/9 are seen. (The ordinate scale is 10 units
(8, 9) ring structure. Upon inclusion of this additional Mo-
N-C multiple scattering pathway, a less significant decrease
in the R value resulted. As an angle deviates further from
linearity, the multiple-scattering signal contributes less to the
total EXAFS wave and limits the possibility of detection. In all
cases, the angles determined were low (120 and 134° for 1 and
8/9, respectively) as was the multiplicity (2). All distances and
their corresponding bond variances, as well as the nonstructural
parameters that were allowed to float during the fitting, remained
essentially constant in both the single- and multiple-scattering
fits, implying that in either case the parameters were well fit.
It is interesting that an Mo-O-Si multiple-scattering interac-
tion is not detected for complexes 6 and 7, as may be deduced
from the Fourier transform of 7 (Figure 9). In order that multiple
between each tick mark.)
also to 1, 6, and 8. The estimated standard deviations in the
bond distances in Table 3 are ((0.01-0.02) Å.
The Fourier transforms of the data for 1 and 8/9 (Figure 9)
showed a distinctive peak at R ∼ 3.2 Å. It is readily shown
that, by inclusion of an Mo-O-Si multiple-scattering pathway
modeled from the crystal structure (for 1) and a similar
arrangement (for 8/9), this feature is well fit, with accompanying
significant decreases in the fit function R from 0.112 × 10-6
to 0.790 × 10-7, 0.940 × 10-7 to 0.467 × 10-7, and 0.274 ×
10-6 to 0.167 × 10-6 for 1, 8, and 9, respectively, with resulting
angles of ∼160°. While a strong improvement was seen, a small

4112 Inorganic Chemistry, Vol. 38, No. 18, 1999
Thapper et al.
Figure 10. Individual EXAFS contributions and the total EXAFS
signal (s) compared with the experimental EXAFS data (- - -) for
[MoOS(OSiPh3)2(Me4phen)] (7, top) and [MoOS(OSiPh3)2(5,5′-
Me2bipy)] (9, bottom). In both cases, the low residual is indicative of
an excellent overall fit. (The ordinate scale is 15 units between each
tick mark.)
Figure 11. Normalized Mo K-edge spectra of [MoO2(OSiPh3)2(phen)]
(1, s), [MoOS(OSiPh3)2(Me4phen)] (7, - - -), and [MoOS(OSiPh3)2-
(5,5′-Me2bipy)] (9, ‚‚‚) with the corresponding second-derivative spectra
shown below. See the text for a detailed description of the edge
structure.
scattering contribute strongly to the EXAFS wave, it must
involve rigidly fixed atoms in an almost linear arrangement.
The bond variance parameters for the Mo-O distance in 6 and
7 are relatively high (0.004-0.005 Å2) and are larger than those
for the other three complexes (0.002-0.003 Å2) in which the
Mo-O-Si interaction was found to be significant. This may
indicate that there is more static disorder present which decreases
the multiple-scattering contribution. It is also possible that the
reduced intensity signal is destructively interfering with the Mo-
N-C multiple-scattering pathways, as opposed to the more
constructive interference patterns seen in complexes 1, 8, and
9. This effect produces cancellation of the contribution and
prevents its detection. In any case, entirely satisfactory fits were
obtained for both 7 and 9 (Figures 8 and 10).
A series of fits were performed for complexes 6-9 in which
either the ModO or ModS interaction was not included, as
well as fits in which the ModO, Mo-O, and ModS coordina-
tion numbers were varied. When the coordination number of
either the ModO or the ModS interaction was increased above
1 or that of the Mo-O interaction was increased above 2, a
large increase (to 0.008-0.011 Å2) was seen in the bond
variance parameter, essentially eliminating the coordination
number increase. In fits for 7, there was a decrease of
approximately an order of magnitude in the R value, from a
high of 0.301 × 10-5, when two ModO interactions were
included, to a low of 0.376 × 10-6, for the best fit, in which
oxo/sulfido ligation was included. In all cases, visually unac-
ceptable fits to the data resulted without inclusion of both the
ModO and ModS interactions. The conclusion is inescapable
that both ModO and ModS interactions are absolutely neces-
sary to achieVe acceptable EXAFS fits for complexes 6-9.
(b) Molybdenum K Edges. Edge data for 1, 7, and 9 are
shown in Figure 11 together with their second derivatives. The
edges of complexes 6 and 7 are almost entirely superimposable
as are those of 8 and 9. These two sets of complexes are also
strongly similar in the shapes and intensities of the edge
transitions. The rising edges of 7 and 9 are shifted to lower
energy by ∼2 eV relative to that of 1. Such a shift is consistent
with the higher electronegativity of oxygen, which results in
differentially larger removal of electron density from the Mo-
(VI) center in 1. The edge structure of this complex consists of
at least four clearly displayed transitions (as confirmed by the
second derivative) which are superimposed on the rising edge
at ∼20 008, 20 017, 20 020, and 20 036 eV. As expected, the
intensities and shapes of these edge transitions are very similar
for 7 and 9. The first three transitions of complex 1 are shifted
to higher energy by ∼2 eV but have intensities relatively similar
to those of complexes 7 and 9 with the exception of the
transition at ∼20 010 eV, for which a substantial shift to higher
energy as well a large increase in the intensity is seen.

The Oxo-Sulfido Functional Group cis-MoVIOS
Inorganic Chemistry, Vol. 38, No. 18, 1999 4113
Molybdenum complexes with ModO ligation show a char-
acteristic edge feature around 20 008 eV. This arises from a
formally dipole-forbidden 1s f 4d bound state transition to
antibonding orbitals directed along the ModO bond(s) and gains
intensity in noncentrosymmetric geometries through p mix-
ing.17,51 The ModS unit also contributes to this edge feature,
but to a lesser extent with broader and less resolvable transi-
tions.17 The intensity and position of this transition can be
correlated to the presence of one or two ModO groups.52,53
The present set of complexes exhibit this “oxo” transition, which
is most easily seen in the second-derivative spectra (Figure 11).
The differences described above are thus in agreement with 1
containing two ModO bonds and 7 and 9 containing one Mod
O and one ModS bond, consistent with the EXAFS results.
Taken together, the EXAFS and edge analyses as well as the
Fourier transforms are consistent with the existence of the oxo-
sulfido functional group cis-MoVIOS, which is found in the
xanthine oxidase family of molybdoenzymes. The ModO and
ModS bond lengths determined in the X-ray refinement of
disordered structures differ by 0.11 and 0.07 Å for 6 and by
0.06 and 0.01 Å for 7 from the respective EXAFS distances.
While the X-ray distances cannot be said to be implausible,
particularly those of 7, they do not exhibit the conformity
expected for such closely related molecules. Note that, for the
set 6-9, the individual ModO (1.71-1.72 Å) and the ModS
values (2.18-2.19 Å) from EXAFS are not distinguishable.
X-ray structures for 8 and 9 are unavailable. We conclude that
the more reliable ModO and ModS bond lengths are those
from the EXAFS analysis. X-ray diffraction defines the overall
stereochemistry and other distance and angular features of the
molecules. Bond lengths from both X-ray and EXAFS analyses
are presented here as an illustration of a problem associated
with O/S disorder and an experimental recourse necessary to
obtain accurate values. The preferred distances compare favor-
ably with those for the EXAFS consensus structure of xanthine
oxidase (ModO 1.68 Å, ModS 2.15 Å)1,4-6 and with those
calculated from density functional theory for the five-coordinate
unit [MoVIOS(S2C2H2)(OH2)] (ModO 1.73 Å, ModS 2.18 Å).54
Oxo Abstraction of the MoVIOS Group. This group presents
two atoms that are susceptible to nucleophilic attack with
attendant reduction to Mo(IV). To ascertain which atom is
removed by a nucleophile that is both oxophilic and thiophilic,
equimolar amounts of 7 and Ph3P were allowed to react in CD2-
Cl2 solution. After ca. 1 h, the 31P NMR spectrum showed a
strong signal for Ph3PS (δ 46.7) and no signals for Ph3PO (δ
35.3) or unreacted phosphine (δ -2.0).55 When the experiment
was conducted on a preparative scale, 37 µmol of 7 and 1.0
equiv of Ph3P were allowed to react in 2 mL of dichloromethane.
Within minutes, the orange solution became red; addition of
ether resulted in precipitation of the reaction product, which
after drying was obtained as a red solid. The product was
identified as the Mo(V) complex 10 from the structure deter-
mination shown in Figure 12. As indicated in Figure 2, this
complex is presumably formed by atom transfer to produce
Figure 12. Structure of [MoOCl(OSiPh3)2(Me4phen)] (10) showing
50% probability ellipsoids and the atom-labeling scheme.
[MoIVO(OSiPh3)2(Me4phen)], which is then trapped by a
chlorine atom transfer reaction with solvent to afford 10 in 63%
yield.
Complex 10 has a distorted octahedral configuration in which
the two silyloxy groups are cis to each other and cis to the oxo
ligand, which itself is cis to chlorine. Consequently, the molecule
does not have pseudo-2-fold symmetry and is not disordered.
The ModO (1.703(4) Å) and Mo-Cl (2.434(2) Å) bond lengths
and the Cl-Mo-O bond angle (95.4(1)°) are normal, as are
the remaining metric parameters listed in Table 2. A small trans
influence of the oxo ligand (0.08 Å) is evident in the Mo-N1
bond. In dichloromethane solution at 4.2 K, 10 exhibits a
rhombically broadened EPR signal with g ≈ 1.95, 1.93, 1.89.
The oxo abstraction reaction of 10 is controlled by the relative
energies of the MoVIdO and MoVIdS bonds rather than the
difference in P-element bond dissociation energies. The BDE
of gaseous Ph3PO is 133 kcal/mol56 and that of Ph3PS in
benzene/toluene solution is 88 kcal/mol,57 a difference of 45
kcal/mol. Recently, the BDE’s of the four-coordinate Mo(V)
complexes [MoQ(N[R]Ar)3] were calorimetrically determined
in solution.58 The values are 156 kcal/mol for Q ) O and 104
kcal/mol for Q ) S. Other ModO BDE’s have been determined
or theoretically estimated to fall in the 126-158 kcal/mol
range.59-63 Density functional calculations afford BDE’s of 126
and 79 kcal/mol for [MoOCl4] and [MoSCl4], respectively.63
Although the data are limited and do not apply to six-coordinate
Mo(VI), a reasonable conclusion at this stage is that ModO/
ModS bond energy differences favor sulfur atom abstraction
in the general case. Investigation of the reaction of 6 with Ph3P
and cyanide, a common deactivator of xanthine oxidase, in the
absence of a trapping reagent is currently in progress.
Summary. The following are the principal results and
conclusions of this investigation.
1. The complexes [MoOS(OSiPh3)2(L-L)] with L-L a phenan-
throline (6, 7) or bipyridyl (8, 9) ligand can be prepared by
silylation of [MoO3S]2- in the presence of L-L and isolated as
orange solids in good yield.
(57) Capps, K. B.; Wixmerten, B.; Bauer, A.; Hoff, C. D. Inorg. Chem.
1998, 37, 2861.
(51) Kutzler, F. W.; Natoli, C. R.; Misemer, D. K.; Doniach, S.; Hodgson,
K. O. J. Chem. Phys. 1980, 73, 3274.
(52) George, G. N.; Colangelo, C. M.; Dong, J.; Scott, R. A.; Khangulov,
S. V.; Gladyshev, V. N.; Stadtman, T. C. J. Am. Chem Soc. 1998,
120, 1267.
(58) Johnson, A. R.; Davis, W. M.; Cummins, C. C.; Serron, S.; Nolan, S.
P.; Musaev, D. G.; Morokuma, K. J. Am. Chem. Soc. 1998, 120, 2071.
N[R]Ar ) -NBut(3,5-C6H3Me2).
(59) Glidewell, C. Inorg. Chim. Acta 1977, 24, 149.
(60) Pedley, J. B.; Marshall, E. M. J. Phys. Chem. Ref. Data 1983, 12,
967.
(53) Musgrave, K. B.; Donahue, J. P.; Lorber, C.; Holm, R. H.; Hedman,
B.; Hodgson, K. O. J. Am. Chem. Soc., submitted for publication.
(54) Voityuk, A. A.; Albert, K.; Ko¨stlmeier, S.; Nasluzov, V. A.; Neyman,
K. M.; Hof, P.; Huber, R.; Roma˜o, M. J.; Ro¨sch, N. J. Am. Chem.
Soc. 1997, 119, 3159.
(61) Pershina, V.; Fricke, B. J. Phys. Chem. 1995, 99, 144; 1996, 100,
8748.
(62) The only exception is [MoOCl4]; D(ModO) ) 101 kcal/mol is
calculated from thermodynamic data.57
(55) 31P chemical shifts are referenced to 85% D3PO4 in D2O.
(56) Holm, R. H.; Donahue, J. P. Polyhedron 1993, 12, 571.
(63) Gonza´lez-Blanco, O.; Branchadell, V.; Monteyne, K.; Ziegler, T. Inorg.
Chem. 1998, 37, 1744.

4114 Inorganic Chemistry, Vol. 38, No. 18, 1999
Thapper et al.
2. The complexes in point 1 have been shown to contain the
cis-MoVIOS functional group from the collective evidence of
elemental analyses, 1H NMR spectra, mass spectrometry, X-ray
structure determinations (6, 7), Mo K-edge spectra, and EXAFS
analyses. The orange color derives from a LMCT band at 424-
434 nm.
6. The results presented here demonstrate access to the
unperturbed cis-MoVIOS functional group obligatory for the
activity of the xanthine oxidase family of enzymes. Complexes
6-9 or related species should allow further exploration of the
structural and reactivity properties of this group. The method
of synthesis may allow introduction of other, more biologically
relevant ligands.
3. Dioxo complexes [MoO2(OSiPh3)2(L-L)] approach or
possess C2 symmetry, with trans silyloxy groups cis to mutually
cis oxo atoms. Oxo-sulfido complexes [MoOS(OSiPh3)2(L-
L)] have the same overall stereochemistry but exhibit O/S
disorder in the crystalline state. X-ray structure refinements with
O/S half-occupancies for complexes 6 and 7 afford ModO and
ModS distances which, while not unreasonable, lack the internal
consistency expected for such closely related molecules.
4. The distances in point 3 obtained for 6-9 by EXAFS
analysis using the GNXAS protocol (ModO 1.71-1.72 Å,
ModS 2.18-2.19 Å) are considered to be more reliable and
are in good agreement with EXAFS values for xanthine oxidase.
Both ModO and ModS interactions are absolutely necessary
to obtain acceptable fits of the EXAFS data. Differences in
Mo-N distances indicated by X-ray structures are not resolvable
by EXAFS.
Acknowledgment. This research was supported by Grants
NSF CHE 94-23181 and NIH RR-01209 (to K.O.H.) and NSF
CHE 94-23830 (to R.H.H.) and by grants from the Swedish
Natural Science Research Council and the Swedish Department
of Education (to E.N.). The Stanford Synchrotron Radiation
Laboratory is supported by the Department of Energy, Office
of Basic Energy Sciences. The Biotechnology Program is
supported by the National Institutes of Health, National Center
for Research Resources, Biomedical Technology Program, and
by the DOE Office of Biological and Environmental Research.
Supporting Information Available: Listings of crystallographic
data for the compounds in Tables 1 and 2, including intensity collection
details, positional and thermal parameters, interatomic distances and
angles, and calculated hydrogen atom positions. This material is
available free of charge via the Internet at http://pubs.acs.org.
5. Reaction of [MoOS(OSiPh3)2(Me4phen)] with Ph3P in
dichloromethane affords [MoVOCl(OSiPh3)2(Me4phen)]. Sulfur
vs oxygen abstraction is favored by relative ModO/ModS bond
energies and is likely to be a general result.
IC990440V