Oxo transfer reactions mediated by bis(dithiolene)tungsten analogues of the active sites of molybdoenzymes in the DMSO reductase family: Comparative reactivity of tungsten and molybdenum

Article abstract of DOI:10.1021/ja0036559

The discovery of tungsten enzymes and molybdenum/tungsten isoenzymes, in which the mononuclear catalytic sites contain a metal chelated by one or two pterin - dithiolene cofactor ligands, has lent new significance to tungsten - dithiolene chemistry. Reaction of [W(CO)₂(S₂C₂Me)₂′] with RO^- affords a series of square pyramidal desoxo complexes [W^IV(OR′)(S₂C₂Me₂) ₂]^1-, including R′ = Ph (1) and Prⁱ (3). Reaction of 1 and 3 with Me₃NO gives the cis-octahedral complexes [W^VIO(OR′)(S₂C₂Me₂) ₂]^1-, including R′ = Ph (6) and Prⁱ (8). These W(IV.VI) complexes are considered unconstrained versions of protein-bound sites of DMSOR and TMAOR (DMSOR = dimethylsulfoxide reductase, TMAOR = trimethylamine N-oxide reductase) members of the title enzyme family. The structure of 6 and the catalytic center of one DMSO reductase isoenzyme have similar overall stereochemistry and comparable bond lengths. The minimal oxo transfer reaction paradigm thought to apply to enzymes, W^IV + XO → W^VIO + X, has been investigated. Direct oxo transfer was demonstrated by isotope transfer from Ph₂Se¹⁸O. Complex 1 reacts cleanly and completely with various substrates XO to afford 6 and product X in second-order reactions with associative transition states. The substrate reactivity order with 1 is Me₃NO > Ph₃AsO > pyO (pyridine N-oxide) > R2SO ? Ph₃PO. For reaction of 3 with Me₃NO, k₂ = 0.93 M-¹ s-¹ and for 1 with Me₂SO, k₂ = 3.9 × 10^-5 M^-1 s^-1: other rate constants and activation parameters are reported. These results demonstrate that bis(dithiolene)W(IV) complexes are competent to reduce both N-oxides and S-oxides: DMSORs reduce both substrate types, but TMAORs are reported to reduce only N-oxides. Comparison of k_cat/K_M data for isoenzymes and k₂ values for isostructural analogue complexes reveals that catalytic and stoichiometric oxo transfer, respectively, from substrate to metal is faster with tungsten and from metal to substrate is faster with molybdenum. These results constitute a kinetic metal effect in direct oxo transfer reactions for analogue complexes and for isoenzymes provided the catalytic sites are isostructural. The nature of the transition state in oxo transfer reactions of analogues is tentatively considered. This research presents the first kinetics study of substrate reduction via oxo transfer mediated by bis(dithiolene)tungsten complexes.

Full text of DOI:10.1021/ja0036559

J. Am. Chem. Soc. 2001, 123, 1931-1943
1931
Oxo Transfer Reactions Mediated by Bis(dithiolene)tungsten
Analogues of the Active Sites of Molybdoenzymes in the DMSO
Reductase Family: Comparative Reactivity of Tungsten and
Molybdenum
Kie-Moon Sung and R. H. Holm*
Contribution from the Department of Chemistry and Chemical Biology, HarVard UniVersity,
Cambridge, Massachusetts 02138
ReceiVed October 12, 2000
Abstract: The discovery of tungsten enzymes and molybdenum/tungsten isoenzymes, in which the mononuclear
catalytic sites contain a metal chelated by one or two pterin-dithiolene cofactor ligands, has lent new significance
to tungsten-dithiolene chemistry. Reaction of [W(CO)2(S2C2Me2)2] with RO- affords a series of square
pyramidal desoxo complexes [WIV(OR′)(S2C2Me2)2]1-, including R′ ) Ph (1) and Pri (3). Reaction of 1 and
3 with Me3NO gives the cis-octahedral complexes [WVIO(OR′)(S2C2Me2)2]1-, including R′ ) Ph (6) and Pri
(8). These W(IV,VI) complexes are considered unconstrained versions of protein-bound sites of DMSOR and
TMAOR (DMSOR ) dimethylsulfoxide reductase, TMAOR ) trimethylamine N-oxide reductase) members
of the title enzyme family. The structure of 6 and the catalytic center of one DMSO reductase isoenzyme have
similar overall stereochemistry and comparable bond lengths. The minimal oxo transfer reaction paradigm
thought to apply to enzymes, WIV + XO f WVIO + X, has been investigated. Direct oxo transfer was
demonstrated by isotope transfer from Ph2Se18O. Complex 1 reacts cleanly and completely with various substrates
XO to afford 6 and product X in second-order reactions with associative transition states. The substrate reactivity
order with 1 is Me3NO > Ph3AsO > pyO (pyridine N-oxide) > R2SO . Ph3PO. For reaction of 3 with
Me3NO, k2 ) 0.93 M-1 s-1, and for 1 with Me2SO, k2 ) 3.9 × 10-5 M-1 s-1; other rate constants and
activation parameters are reported. These results demonstrate that bis(dithiolene)W(IV) complexes are competent
to reduce both N-oxides and S-oxides; DMSORs reduce both substrate types, but TMAORs are reported to
reduce only N-oxides. Comparison of kcat/KM data for isoenzymes and k2 values for isostructural analogue
complexes reveals that catalytic and stoichiometric oxo transfer, respectively, from substrate to metal is faster
with tungsten and from metal to substrate is faster with molybdenum. These results constitute a kinetic metal
effect in direct oxo transfer reactions for analogue complexes and for isoenzymes provided the catalytic sites
are isostructural. The nature of the transition state in oxo transfer reactions of analogues is tentatively considered.
This research presents the first kinetics study of substrate reduction via oxo transfer mediated by bis(dithiolene)-
tungsten complexes.
Introduction
include, among others, DMSO reductase (DMSOR) itself and
trimethylamine N-oxide reductase (TMAOR). A recent reex-
amination of the structure of oxidized Rhodobacter sphaeroides
(Rs) DMSOR at 1.3 Å resolution has demonstrated the presence
of two mutually disordered sites, square pyramidal [MoVIO2-
(O‚Ser)(S2pd)] and the distorted trigonal prismatic catalytic site
[MoVIO(O‚Ser)(S2pd)2], in an occupancy ratio of 0.6:0.4.7 This
work indicates that a previous structure determination of Rs
DMSOR was performed at a resolution (2.2 Å) insufficient to
resolve the two sites.8 While the matter is not settled, the same
behavior may apply to Rhodobacter capsulatus (Rc) DMSOR9,10
and Shewanella massilia TMAOR.11 EXAFS data for Rs
As observed recently,1 extensive advances in structure
elucidation by protein crystallography and X-ray absorption
spectroscopy provide a basis for the development of synthetic
analogues of the active sites of molybdenum2-4 and tungsten5,6
enzymes. Accurate analogues should be incisive in revealing
intrinsic structural and reactivity features. The enzyme sites
contain a single metal atom bound to one or two pterin-
dithiolene cofactor ligands (S2pd) shown in Figure 1, with
additional coordination provided by oxo, sulfido, hydroxo, aqua,
and protein-based ligands. Under the Hille classification,2
oxidized molybdoenzymes in the DMSO reductase family
(1) Lim, B. S.; Sung, K.-M.; Holm, R. H. J. Am. Chem. Soc. 2000, 122,
7410-7411.
(7) Li, H.-K.; Temple, C.; Rajagopalan, K. V.; Schindelin, H. J. Am.
Chem. Soc. 2000, 122, 7673-7680.
(2) Hille, R. Chem. ReV. 1996, 96, 2757-2816.
(3) Roma˜o, M. J.; Kna¨blein, J.; Huber, R.; Moura, J. J. G. Prog. Biophys.
Mol. Biol. 1997, 68, 121-144.
(8) Schindelin, H.; Kisker, C.; Hilton, J.; Rajagopalan, K. V.; Rees, D.
C. Science 1996, 272, 1615-1621.
(9) McAlpine, A. S.; McEwan, A. G.; Shaw, A. L.; Bailey, S. J. Biol.
Inorg. Chem. 1997, 2, 690-701.
(4) Kisker, C.; Schindelin, H.; Rees, D. C. Annu. ReV. Biochem. 1997,
66, 233-267.
(10) McAlpine, A. S.; McEwan, A. G.; Bailey, S. J. Mol. Biol. 1998,
275, 613-623.
(5) Johnson, M. K.; Rees, D. C.; Adams, M. W. W. Chem. ReV. 1996,
96, 2817-2839.
(11) Czjzek, M.; Dos Santos, J.-P.; Pommier, J.; Giordano, G.; Me´jean,
V.; Haser, R. J. Mol. Biol. 1998, 284, 435-447.
(6) Hagen, W. R.; Arendsen, A. F. Struct. Bonding 1998, 90, 161-192.
10.1021/ja0036559 CCC: $20.00 © 2001 American Chemical Society
Published on Web 02/10/2001

1932 J. Am. Chem. Soc., Vol. 123, No. 9, 2001
Sung and Holm
isoenzymes, highly pertinent to this investigation, have been
described. These include TMAOR isoenzymes from E. coli;
W-TMAOR also reduces sulfoxides but Mo-TMAOR does not.26
The active site structures of these enzymes have not been
reported. Growth of a strain of Rc in the presence of Na2WO4
results in the production of a W-DMSOR isoenzyme, whose
crystal structure is very similar to that of Mo-DMSOR.27 Both
crystallography and tungsten EXAFS of the oxidized enzyme
are consistent with the six-coordinate site formulation [WVIO-
(O‚Ser)(S2pd)2], but the presence of another oxygen atom, as
described for the seven-coordinate oxidized Mo-DMSOR site,9,10
was not excluded. Molydenum and tungsten isoenzymes are
listed in Table 1 together with relative reactivities as ratios of
rate constants. These results are considered in relation to kinetics
data obtained in this work.
Given the structural results for Rs DMSOR,7 the operation
of a direct oxo transfer pathway by this enzyme,28,29 and the
existence of isoenzymes, we proceed on the basis of the minimal
reaction paradigm MIV + XO h MVIO + X (M ) Mo, W) in
the evolution of oxo transfer analogue reactions systems.1 Here
the stoichiometric reaction or catalytic cycle operates between
the desoxo M(IV) and monooxo M(VI) states as shown for
DMSOR isoenzymes in Figure 1. Our initial set of bis-
(dithiolene) active site analogues utilized benzene-1,2-dithiolate
(bdt) as a simulator of the pterin-dithiolene cofactor.20,21 These
and dithiolene complexes derived from the ligands in Figure
124,25 provide characteristic edge and EXAFS features for the
identification of desoxo, monooxo, and dioxo enzyme sites.23,30
To afford a more realistic electronic structure and, therewith, a
closer approach to the intrinsic reactivity imposed by the
cofactor ligandsitself essentially a dialkyldithioleneswe em-
ploy the ligands in Figure 1. In this investigation, we have
synthesized additional bis(dithiolene)W(IV,VI) complexes for
reactivity studies and investigated the kinetics and mechanism
of oxo transfer in analogue reaction systems. A parallel study
of oxo transfer mediated by molybdenum complexes is reported
elsewhere.31 Certain related reactivity results for both molyb-
denum and tungsten systems have been communicated recently.1
A more extensive comparison of molybdenum and tungsten oxo
transfer rates is presented here.
Figure 1. Proposed active site structures of reduced and oxidized Mo/
W-DMSOR isoenzymes with pterin-dithiolene cofactors based on
X-ray crystallography and EXAFS (top) and the oxo transfer reactions
mediated by bis(dithiolene)tungsten analogue system (bottom). The
structure of the pterin-dithiolene cofactor (R absent or a nucleotide)
is indicated.
DMSOR12and Rs biotin sulfoxide reductase13 are consistent with
the foregoing MoVIO site and the formulation [MoIV(O‚Ser)-
(OH2)(S2pd)2] for the dithionite-reduced enzymes.
Tungstoenzymes have been classified into two principal
families: AOR (aldehyde oxidoreductases) and F(M)DH (for-
mate dehydrogenases, FDH, and N-formylmethanofuran dehy-
drogenases, FMDH).5 Crystallographic information is currently
limited to the structures of aldehyde14 and formaldehyde15
ferredoxin oxidoreductases from the hyperthemophilic archaeon
Pyroccus furiosus. The active sites contain two chelating
cofactor ligands and one or two light atom, presumably
oxygenous, ligands. EXAFS results for P. furiosus AOR are
consistent with the X-ray structure.5 The extent to which various
tungstoenzymes bind two cofactor ligands is not known.
Because of the existence of molybdenum and tungsten
isoenzymes, originally detected in the F(M)DH family,16-18 and
a more general interest in the comparative properties of these
elements, we havesinsofar as possiblesdeveloped the chemistry
of bis(dithiolene)molybdenum and tungsten complexes in
parallel.1,19-25,30 Since these studies were initiated, additional
Experimental Section
Preparation of Compounds. All operations were carried out under
a pure dinitrogen atmosphere with an inert atmosphere box or standard
Schlenk techniques. Solvents were distilled, dried, and degassed prior
to use; ether and THF were distilled from Na/benzophenone ketyl,
acetonitrile, and dichloromethane from CaH2, and methanol from
magnesium. Acetonitrile-d3, dimethyl sulfoxide-d6, and CD2Cl2 were
stored over 4-Å molecular sieves for at least 1 d. Lithium 2-adaman-
(12) George, G. N.; Hilton, J.; Temple, C.; Prince, R. C.; Rajagopalan,
K. V. J. Am. Chem. Soc. 1999, 121, 1256-1266.
(13) Temple, C. A.; George, G. N.; Hilton, J. C.; George, M. J.; Prince,
R. C.; Barber, M. J.; Rajagopalan, K. V. Biochemistry 2000, 39, 4046-
4052.
(14) Chan, M. K.; Mukund, S.; Kletzin, A.; Adams, M. W. W.; Rees,
D. C. Science 1995, 267, 1463-1469.
(15) Hu, Y.; Faham, S.; Roy, R.; Adams, M. W. W.; Rees, D. C. J.
Mol. Biol. 1999, 286, 899-914.
(21) Donahue, J. P.; Goldsmith, C. R.; Nadiminti, U.; Holm, R. H. J.
Am. Chem. Soc. 1998, 120, 12869-12881.
(22) Goddard, C. A.; Holm, R. H. Inorg. Chem. 1999, 38, 5389-5398.
(23) Musgrave, K. B.; Donahue, J. P.; Lorber, C.; Holm, R. H.; Hedman,
B.; Hodgson, K. O. J. Am. Chem. Soc. 1999, 121, 10297-10307.
(24) Lim, B. S.; Donahue, J. P.; Holm, R. H. Inorg. Chem. 2000, 39,
263-273.
(25) Sung, K.-M.; Holm, R. H. Inorg. Chem. 2000, 39, 1275-1281.
(26) Buc, J.; Santini, C.-L.; Giordani, R.; Czjzek, M.; Wu, L.-F.;
Giordano, G. Mol. Microbiol. 1999, 32, 159-168.
(27) Stewart, L. J.; Bailey, S.; Bennett, B.; Charnock, J. M.; Garner, C.
D.; McAlpine, A. S. J. Mol. Biol. 2000, 299, 593-600.
(28) Schultz, B. E.; Holm, R. H.; Hille, R. J. Am. Chem. Soc. 1995,
117, 827-828.
(29) Garton, S. D.; Hilton, J.; Oku, H.; Crouse, B. R.; Rajagopalan, K.
V.; Johnson, M. K. J. Am. Chem. Soc. 1997, 119, 12906-12916.
(30) Musgrave, K. B.; Lim, B. S.; Sung, K.-M.; Holm, R. H.; Hedman,
B.; Hodgson, K. O. Inorg. Chem. 2000, 39, 5238-5247.
(31) Lim, B. S.; Holm, R. H. J. Am. Chem. Soc. 2001, 123, 1920-1930.
(16) (a) Schmitz, R. A.; Richter, M.; Linder, D.; Thauer, R. K. Eur. J.
Biochem. 1992, 207, 559-565. (b) Bertram, P. A.; Schmitz, R. A.; Linder,
D.; Thauer, R. K. Arch. Microbiol. 1994, 161, 220-228. (c) Bertram, P.
A.; Karrasch, M.; Schmitz, R. A.; Bo¨cher, R.; Albracht, S. P. J.; Thauer,
R. K. Eur. J. Biochem. 1994, 220, 477-484.
(17) Schmitz, R. A.; Albracht, S. P. J.; Thauer, R. K. Eur. J. Biochem.
1992, 209, 1013-1018.
(18) Hochheimer, A.; Schmitz, R. A.; Thauer, R. K. Eur. J. Biochem.
1995, 234, 910-920.
(19) Tucci, G. C.; Donahue, J. P.; Holm, R. H. Inorg. Chem. 1998, 37,
1602-1608.
(20) Lorber, C.; Donahue, J. P.; Goddard, C. A.; Nordlander, E.; Holm,
R. H. J. Am. Chem. Soc. 1998, 120, 8102-8112.

ComparatiVe ReactiVity of Tungsten and Molybdenum
J. Am. Chem. Soc., Vol. 123, No. 9, 2001 1933
Table 1. Relative Reactivities of Molybdenum and Tungsten Isoenzymes
tyloxide was prepared by the lithiation of 2-adamantanol at -78 °C
with BunLi in THF. The compounds NaOC6H4-p-X (X ) CN, Br, Me,
OMe, NH2) were prepared from the reaction of NaOMe with p-XC6H4-
OH (purified by sublimation) in methanol. Trimethylamine N-oxide,
pyridine N-oxide, and 4-methylmorpholine N-oxide were sublimed
under reduced pressure. The compound Ph2Se18O (62% 18O-enriched)
was synthesized by a reported method.20 Dimethyl sulfoxide (Aldrich,
99%) was dried over 4-Å molecular sieves and tetramethylene sulfoxide
(Aldrich, 96%) was dried over CaH2 and distilled under 0.3 Torr at
100 °C. Other compounds were of commercial origin and used as
received. Solvent removal and drying steps were performed in vacuo.
Filtrations were through Celite; compounds were isolated by filtration
unless otherwise stated.
(acetonitrile): λmax (ꢀM) 284 (10 000), 309 (8500), 340 (sh, 3900), 380
(sh, 1400), 423 (850), 545 (270), 670 (120) nm. Anal. Calcd for C19H39-
NOS4W: C, 37.43; H, 6.45; N, 2.30; S, 21.04. Found: C, 37.54; H,
6.38; N, 2.30; S, 21.12.
(Et4N)[W(2-AdO)(S2C2Me2)2]. To a suspension of 21.3 mg (0.13
mmol) of 2-AdOLi (Ad ) adamantyl) in 0.5 mL of acetonitrile was
added a violet solution of 64.5 mg (0.14 mmol) of [W(CO)2(S2C2Me2)2]
in 2 mL of THF. A red-brown color developed immediately; the reaction
mixture was stirred for 30 min. After the treatment with 22.2 mg (0.13
mmol) of Et4NCl, the solvents were removed. The resulting brown solid
was redissolved in 1.5 mL of acetonitrile and the solution was filtered.
Onto the brown filtrate was layered 10 mL of ether. Over 6 h, needlelike
crystals deposited. These were isolated, washed with ether (3 × 1 mL),
1
(Et4N)[W(OPh)(S2C2Ph2)2]. To a dark purple solution of 107 mg
(0.15 mmol) of [W(CO)2(S2C2Ph2)2]22 in 1.5 mL of THF was added
18.1 mg (0.16 mmol) of solid NaOPh. A green-brown solution
developed within 5 min and was stirred for 1 h. This solution was
treated with 26.9 mg (0.16 mmol) of Et4NCl in 1 mL of acetonitrile
and the solvents were removed. The resulting brown solid was
redissolved in 2 mL of acetonitrile and the solution was filtered. Ether
(60 mL) was layered on the filtrate, producing 45.5 mg (34%) of green-
brown crystalline solid over 1 d, which was isolated, washed with ether
and dried to afford 50.5 mg (53%) of dark brown crystals. H NMR
(CD3CN, anion): δ 1.41-1.66 (m, 14), 2.50 (s, 12), 3.49 (s, 1).
Absorption spectrum (acetonitrile): λmax (ꢀM) 283 (11 000), 307 (9100),
340 (sh, 4500), 380 (sh, 1700), 423 (920), 544 (300), 670 (110) nm.
Anal. Calcd for C26H47NOS4W: C, 44.50; H, 6.75; N, 2.00; S, 18.28.
Found: C, 44.68; H, 6.71; N, 1.98; S, 18.37.
(Et4N)[W(O-p-C6H4X)(S2C2Me2)2]. These compounds were pre-
pared on a 0.10-0.22 mmol scale by a procedure analogous to that for
(Et4N)[W(OPh)(S2C2Me2)2]25 with NaOC6H4-p-X instead of NaOPh.
X ) CN: Dark brown block-shaped crystals. Yield: 35%. IR
1
(3 × 1 mL), and dried. H NMR (CD3CN, anion): δ 6.6-7.4 (m).
1
Absorption spectrum (acetonitrile): λmax (ꢀM) 310 (26 000), 380 (sh,
6700), 500 (sh, 1800), 680 (980) nm. Anal. Calcd for C42H45NOS4W:
C, 56.56; H, 5.09; N, 1.57; S, 14.38. Found: C, 56.42; H, 5.15; N,
1.47; S, 14.30.
(KBr): ν(CN) 2221(m) cm-1. H NMR (CD3CN, anion): δ 2.62 (s,
12), 6.56 (d, 2, J ) 8.8 Hz), 7.38 (d, 2, J ) 8.8 Hz). Absorption
spectrum (acetonitrile): λmax (ꢀM) 282 (21 000), 330 (sh, 11 000), 360
(sh, 3900), 408 (3000), 490 (820) 660 (320) nm. Anal. Calcd for
C23H36N2OS4W: C, 41.31; H, 5.43; N, 4.19; S, 19.18. Found: C, 41.39;
H, 5.40; N, 4.08; S, 19.29.
(Et4N)[W(OPri)(S2C2Me2)2]. To a solution of 205 mg (0.43 mmol)
of [W(CO)2(S2C2Me2)2]22 in 2 mL of THF was added 37.1 mg (0.45
mmol) of solid NaOPri. The solution turned from dark violet to brown
immediately. The reaction mixture was stirred for 10 min and treated
with 74.7 mg (0.45 mmol) of Et4NCl in 1 mL of acetonitrile, and the
solvents were removed. The resulting brown solid was redissolved in
1.5 mL of acetonitrile and the solution was filtered. Ether (50 mL)
was layered on the brown filtrate affording 132 mg (50%) of brown
microcrystalline solid over 1 d, which was isolated, washed with ether
X ) Br: Dark brown block-shaped crystals. Yield: 46%. 1H NMR
(CD3CN, anion): δ 2.62 (s, 12), 6.33 (d, 2, J ) 8.8 Hz), 7.21 (d, 2, J
) 8.8 Hz). Absorption spectrum (acetonitrile): λmax (ꢀM) 291 (15 000),
319 (sh, 12 000), 347 (sh, 4800), 398 (2700), 478 (760) 658 (380) nm.
Anal. Calcd for C22H36NOBrS4W: C, 36.57; H, 5.02; N, 1.94; Br, 11.06;
S, 17.75. Found: C, 36.47; H, 5.10; N, 1.86; Br, 10.98; S, 17.87.
1
X ) Me: Brown needle-shaped crystals. Yield: 60%. H NMR
1
(3 × 1 mL), and dried. H NMR (CD3CN, anion): δ 0.84 (d, 6, J )
(CD3CN, anion): δ 2.12 (s, 3), 2.60 (s, 12), 6.30 (d, 2, J ) 8.4 Hz),
6.80 (d, 2, J ) 8.4 Hz). Absorption spectrum (acetonitrile): λmax (ꢀM)
6.0 Hz), 2.52 (s, 12), 3.75 (heptet,1, J ) 6.4 Hz). Absorption spectrum

1934 J. Am. Chem. Soc., Vol. 123, No. 9, 2001
Sung and Holm
1
IR (KBr): ν(WO) 895 cm-1. H NMR (CD3CN, anion): δ 2.19 (s,
12), 6.79 (t, 1, J ) 7.2 Hz), 6.94 (d, 2, J ) 7.0 Hz), 7.18 (t, 2, J ) 8.8
Hz). Absorption spectrum (acetonitrile): λmax (ꢀM) 299 (sh, 11 000),
337 (8000), 408 (sh, 5000), 514 (5000), 637 (2500) nm. Anal. Calcd
for C22H37NO2S4W: C, 40.06; H, 5.65; N, 2.12; S, 19.44. Found: C,
40.14; H, 5.61; N, 2.06; S, 19.51.
Chart 1. Abbreviations and Designation of Tungsten
Complexes
(Et4N)[WO(OPri)(S2C2Me2)2]. To a brown solution of 37.2 mg
(0.061 mmol) of (Et4N)[W(OPri)(S2C2Me2)2] in 2 mL of acetonitrile
was added 6.87 mg (0.091 mmol) of Me3NO in 0.2 mL of acetonitile
at room temperature. The solution color changed to green-brown over
1 min; the solution was stirred for 10 min and 30 mL of ether was
added. A black crystalline solid deposited over 2 d, which was isolated
by decantation of the pale green-brown supernatant, washed with ether
(3 × 1 mL), and dried to give 11.3 mg (30%) of brown-black crystalline
solid. IR (KBr): ν(WO) 885 cm-1. 1H NMR spectrum (CD3CN, anion):
δ 1.06 (d, 6, J ) 5.6 Hz), 2.13 (s, 12), 4.74 (heptet, 1, J ) 5.6 Hz).
Absorption spectrum (acetonitrile): λmax (ꢀM) 331 (4400), 410 (2200),
487 (3000), 632 (1800) nm. Anal. Calcd for C19H39NO2S4W: C, 36.48;
H, 6.28; N, 2.24; S, 20.50. Found: C, 36.40; H, 6.35; N, 2.40; S, 20.56.
18O Transfer Reactions. To a solution of (Et4N)[W(OPh)(S2C2-
Me2)2] (6.3 mg, 9.8 µmol) in 0.5 mL of acetonitrile was added a solution
of Ph2Se18O (3.2 mg, 13 µmol) in 0.5 mL of acetonitrile. This mixture
was shaken for 3 min. The solvent was removed and the residue was
dried in vacuo. IR (KBr): 895 (W16O), 848 (W18O) cm-1
.
In the sections that follow, tungsten complexes are referred to by
the numerical designations in Chart 1, which also includes abbreviations
used.
X-ray Structure Determinations. The structures of the five
compounds in Table 2 were determined. Single crystals were obtained
by layering ether onto acetonitrile solutions at ambient temperature
((Et4N)[2,3,4,5e]) or at -15 °C ((Et4N)[6]). Selected crystals were
coated in silicone grease and mounted on a Siemens (Bruker) SMART
CCD diffractometer equipped with an LT-2 low-temperature apparatus
operating at 213 K. Mo KR radiation (λ ) 0.71073 A) was used for
data collection. Data were collected with ω scans of 0.3°/frame for 30
s, 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 significant decay was detected for any com-
pounds. Data only out to 2θ of 50° for (Et4N)[2] were used because of
the low-quality high-angle data while data out to 2θ of 56.4-56.6°
were employed for the other compounds. Cell parameters were retrieved
by using SMART software and refined by using SAINT software on
all reflections. Data reduction was performed with the SAINT software,
which corrects for Lorentz polarization and decay. Absorption correc-
tions were applied with SADABS. The space groups for all compounds
were assigned unambiguously by analysis of symmetry and systematic
absences determined by the program XPREP. Missing symmetry was
checked by the program PLATON; no significant missing symmetry
was observed. Crystal parameters are given in Table 2.
295 (14 000), 319 (sh, 10 000), 347 (sh, 4000), 393 (2000), 478 (520),
580 (200) nm. Anal. Calcd for C23H39NOS4W: C, 42.00; H, 5.98; N,
2.13; S, 19.50. Found: C, 42.68; H, 6.10; N, 1.95; S, 19.41.
1
X ) OMe: Brown needle-shaped crystals. Yield: 53%. H NMR
(CD3CN, anion): δ 2.60 (s, 12), 3.62 (s, 3), 6.34 (d, 2, J ) 8.8 Hz),
6.55 (d, 2, J ) 8.8 Hz). Absorption spectrum (acetonitrile): λmax (ꢀM)
295 (16 000), 319 (sh, 11 000), 347 (sh, 4100), 390 (2000), 478 (550),
569 (250). Anal. Calcd for C23H39NO2S4W: C, 41.01; H, 5.84; N, 2.08;
S, 19.04. Found: C, 41.05; H, 5.92; N, 2.05; S, 19.13.
Structures were solved by direct methods with SHELXS-97 and
subsequently refined against all data by full-matrix least squares on F2
with SHELXL-97. Asymmetric units contain one-half ((Et4N)[4]), one
((Et4N)[5e,6]), and two ((Et4N)[2,3]) formula weights; equivalent atoms
were generated by symmetry transformations for (Et4N)[4]. In (Et4N)-
[4], atoms C5, C7, and C11 of the adamantyl group are severely
disordered over two positions and were refined with a site occupancy
factor of 0.5; adamantyl carbon atoms were refined isotropically. All
methylene groups of the cations in (Et4N)[4,5e,6] are highly disordered
over two positions and were refined with a site occupancy factor of
0.5. All non-hydrogen atoms including those of the disordered cations
were refined anisotropically. In the final refinement, hydrogen atoms
were attached at idealized positions on carbon atoms. Selected bond
distances and angles are listed in Table 3. (See the paragraph at the
end of the article for Supporting Information available.)
X ) NH2: Green-brown block-shaped crystals. Yield: 49%. IR
1
(KBr): ν(NH2) 3327 (m), 3397 (m) cm-1. H NMR (CD3CN, anion):
δ 2.09 (s, 2), 2.59 (s, 12), 6.18 (d, 2, J ) 8.8 Hz), 6.27 (d, 2, J ) 8.8
Hz). Absorption spectrum (acetonitrile): λmax (ꢀM) 297 (18 000), 316
(sh, 16 000), 394 (sh, 2500), 478 (sh, 720), 572 (310) nm. Anal. Calcd
for C22H38N2OS4W: C, 40.12; H, 5.82; N, 4.25; S, 19.47. Found: C,
40.16; H, 6.02; N, 4.33; S, 19.55.
(Et4N)[WO(OPh)(S2C2Me2)2]. To a frozen solution of 83.4 mg (0.13
mmol) of (Et4N)[W(OPh)(S2C2Me2)2]25 in ca. 4 mL of acetonitrile in a
dry ice-acetone bath was injected a solution of 15.0 mg (0.20 mmol)
of Me3NO in 0.15 mL of MeCN. The frozen mixture was gradually
warmed to -15 °C over 30 min, resulting in a color change to dark
violet from green-brown. This solution was treated with 50 mL of cold
ether (-15 °C) and maintained at -20 °C over 1 d during which black
needlelike crystals separated. The solid was isolated by decantation of
the pale brown supernatant, washed with 10 mL of cold ether (-15
°C), and dried to give the product as 65.1 mg (76%) of black crystals.
The structure of (Et4N)[8] was also investigated. This compound
crystallizes in monoclinic space group P21/n with a ) 9.176(1) Å, b
) 17.641(1) Å, c ) 16.139(1) Å, â ) 98.126(1)°, V ) 2586.1(4) Å3,
and Z ) 4. Because of disorder, an accurate structure was not obtained.
However, the overall structure of the anion is the same as that of 6.

ComparatiVe ReactiVity of Tungsten and Molybdenum
Table 2. Crystallographic Dataa
J. Am. Chem. Soc., Vol. 123, No. 9, 2001 1935
(Et4N)[2]
(Et4N)[3]
(Et4N)[4]
(Et4N)[5e]
(Et4N)[6]
formula
fw
crystal system
space group
Z
C42H45NOS4W
891.88
monoclinic
P21/c
C19H39NOS4W
609.60
monoclinic
P21/n
C26H47NOS4W
701.74
orthorhombic
Pnma
C22H38N2OS4W
658.63
monoclinic
P21/c
C22H37NO2S4W
659.62
monoclinic
P21/c
8
8
4
4
4
a, Å
b, Å
c, Å
15.109(3)
19.512(7)
28.167(6)
102.537(18)
8105(4)
9.1267(5)
39.135(2)
14.2182(7)
95.647(1)
5053.7(5)
1.602
23.918(1)
11.4484(5)
11.1165(5)
18.149(2)
8.8873(12)
18.151(2)
114.667(7)
2660.6(5)
1.664
10.0883(4)
15.8029(6)
16.8607(6)
98.602(1)
2657.77(17)
1.648
â, deg
V, Å3
3044.0(2)
1.531
d
calc, g/cm3
1.462
2θ range, deg
2.6-50
2.1-56.6
3.4-56.4
2.5-56.6
3.6-56.5
GOF (F2)
1.082
1.007
1.038
1.018
1.059
c
R1b (wR2 )
0.0614 (0.1129)
0.0417 (0.0796)
0.0366 (0.0922)
0.0277 (0.0626)
0.0361 (0.0661)
a Obtained with graphite-monochromatized Mo KR (λ ) 0.71073 Å) radiation at 213 K. b R1 ) ∑||Fo| - |Fc||/∑|Fo|. c wR2 ) {∑[w(Fo
-
2
Fc2)2]/∑[w(Fo )2]}1/2
.
2
Table 3. Selected Bond Distances (Å) and Angles (deg)
2a 3a 4b
5e
were linear over at least 3 half-lives and the initial rate constants kobs
were obtained from the plots by using data for the first 30 min. Plots
of kobs vs [XO]0 were linear and yielded overall second-order rate
constants k2 at each temperature. In the case of mixed-second-order
conditions,32 the second-order rate constants were obtained from the
plots of {ln([XO]t/[WIV]t) - ln([XO]0/[WIV]0)}/([XO]0 - [WIV]0) vs
time, where the values of [W]t and [XO]t were calculated from the
equations of [WIV]t ) [WIV]0 - (At - A0)/(A∞ - A0) × [WIV]0 and
[XO]t ) [XO]0 - (At - A0)/(A∞ - A0) × [WIV]0, respectively. In mixed-
second-order kinetics, k2 was averaged from the values that were
obtained from multiple independent runs to minimize experimental
errors. For the hydrolysis of 1 and 3, Michaelis-Menten kinetics were
applied and plots of 1/kobs vs 1/[H2O or D2O] were linear to yield k2′
and KM. Activation parameters were determined from plots of the Eyring
equation k ) (kBT/h)[exp(∆Sq/R - ∆Hq/RT)]. Standard deviations for
rate constants were estimated by using linear least-squares error analysis,
with uniform weighting of the data points.33
6
WdOc
1.728(3)
W-Od
1.827(8) 1.840(4) 1.826(4)
2.309(3) 2.322(2) 2.327(1)
2.318(3) 2.319(2) 2.336(1)
2.320(3) 2.338(2)
2.317(3) 2.343(2)
1.79(1) 1.776(5) 1.771(3)
1.868(2) 1.994(4)
2.332(1) 2.492(1)e
2.326(1) 2.432(1)
2.307(1) 2.402(1)
2.320(1) 2.413(1)
1.781(9) 1.74(2)
W-S1
W-S2
W-S3
W-S4
S-Cf
C-Cf,g
1.338
1.330
1.333
1.311
1.336
93.3(2)
146.0(1)
O-W-O
O-W-Sh
W-O-Cd
145.0(8) 141.9(4) 119.3(8)i 141.3(2) 141.0(3)
S1-W-S2j 82.7(1) 82.66(6) 82.79(6)k 82.04(4) 77.26(5)
S3-W-S4j 82.7(1) 82.56(6) 81.86(6)l 82.38(3) 79.13(5)
S1-W-S4m 142.6(1) 144.59(6) 142.78(4)n 144.64(4) 96.00(5)
S2-W-S3m 140.8(1) 139.24(6)
136.47(4) 153.49(5)
0.782
δo
θd
0.756
128.2
0.760
128.4
0.744
129.8
Other Physical Measurements. All measurements were performed
under anaerobic conditions. Absorption spectra were recorded with a
p
126.5
99.8
1
a One of two anions in the asymmetric unit; both have very similar
Varian Cary 50 Bio or a Cary 3 spectrophotometer. H NMR spectra
metric parameters. b Symmetry transformations used to generate equiva-
lent atoms: x, -y + 1/2, z. c Terminal oxo ligand. d Unidentate ligand.
e Trans to oxo. f Mean value. g Chelate ring. h O1-W-S1. i One of the
disordered carbon atoms considered. j Chelate ring bite angle. k S1-
W-S1A. l S2-W-S2A. m Transoid angle. n S1-W-S2A and S1A-
W-S2 are identical. o Perpendicular displacement of W atom from S4
least-squares plane. p Dihedral angle between WS2 planes.
were obtained with Varian Mercury 400/500 spectrometers. IR spectra
were measured with KBr pellets with a Nicolet Nexus 470 FT-IR
instrument.
Results and Discussion
In this investigation, we seek W(IV,VI) complexes that serve
as analogues of the sites of tungsten isoenzymes of the DMSOR
family of molybdoenzymes. Consequently, we require syntheses
of desoxo and monooxo complexes [WIV(OR′)(S2C2R2)2]1- and
[WVIO(OR′)(S2C2R2)2]1-, respectively, of which only two
examples (125 and 61) have been reported earlier. Synthetic
methods are outlined in Figure 2. All complexes were isolated
as crystalline Et4N+ salts.
Synthesis and Structures of W(IV) Complexes. These
complexes are readily prepared by carbonyl displacement
reaction 1 (R ) Me, Ph), which affords products with aryloxide
(1, 2, 5) and alkoxide (3, 4) terminal ligands in yields of ca.
Kinetics Measurements. Reactions were performed under anaerobic
conditions in acetonitrile solutions, and were monitored with a Varian
Cary 3 spectrophotometer equipped with a cell compartment thermo-
stated to (0.5 °C and a multicell changer. Thermal equilibrium was
reached by placing a quartz cell (1 mm path length) containing 0.3
mL of a W(IV) compound solution into the cell compartment at least
5 min prior to reaction initiation. Reactions were initiated by the
injection of substrate (or its solution) through the rubber septum cap
of the quartz cell using a gastight syringe (50 µL) followed by rapid
shaking of the reaction mixture.
Reductions of N-oxide and S-oxide substrates by W(IV) complexes
1-3 and 5a-e and hydrolysis of 1 and 3 were monitored spectropho-
tometrically in the region 280-700 nm. Sharp isosbestic points were
observed for all reaction systems. Reactions were run under pseudo-
first-order conditions for the reduction of relatively weak oxo donors
such as pyO, S-oxides, and Ph3AsO, and for hydrolysis. Initial W(IV)
complex concentrations [WIV]0 ) 0.45-1.2 mM; oxo donor substrates
were present at [XO]0 ) 150-4000 equiv (DMSO, TMSO), 7-40 equiv
(pyO, Ph3AsO), ca. 7500 equiv (neat TMSO), and 1000-20000 equiv
(H2O, D2O). For faster reactions, mixed-second-order conditions were
applied with [WIV]0 ) 0.97-1.4 mM and 2-3 equiv of N-oxide.
Reactions were conducted at five or six temperatures in the range of
290-333 K. At each temperature, at least five independent runs with
different initial concentrations [XO]0 were performed in the case of
pseudo-first-order conditions. Plots of ln[(At - A∞)/(A0 - A∞)] vs time
[W(CO)2(S2C2R2)2] + R′O- f
[WIV(OR′)(S2C2R2)2]1- + 2CO (1)
35-70%. The set 5a-e was prepared to ascertain the effect of
para-substituent X on substrate reduction rates. In solution, the
W(IV) complexes are highly sensitive to trace amounts of water
(32) Espenson, J. H. Chemical Kinetics and Reaction Mechanisms;
McGraw-Hill: New York, 1995.
(33) Clifford, A. A. MultiVariate Error Analysis; Halsted Press: New
York, 1973.

1936 J. Am. Chem. Soc., Vol. 123, No. 9, 2001
Sung and Holm
Figure 2. Synthesis of desoxo bis(dithiolene)W(IV) (1-5) and monooxo W(VI) (6-10) complexes. Complexes 7, 9, and 10 were generated in
solution but not isolated.
147°, including 1.25 The square pyramidal stereochemistry
observed here is now established as a general structural motif
for molybdenum and tungsten complexes of the type [MIV(QR)-
(S2C2Me2)2]1- (Q ) O, S, Se),1,24,25,31 In all such complexes,
the chelate ring C-C and C-S bond lengths are consistent with
an ene-dithiolate ligand electron distribution.
Synthesis and Structures of W(VI) Complexes. The
diamagnetic monooxo W(VI) complexes 6-10 have been
prepared by oxo transfer reaction 2. The green-brown or brown
acetonitrile solution of a W(IV) complex undergoes an immedi-
ate color change to red-violet (6, 7, 10) or green-brown (8, 9)
[WIV(OR′)(S2C2R2)2]1- + Me3NO f
[WVIO(OR′)(S2C2R2)2]1- + Me3N (2)
upon the addition of 1.0-1.5 equiv of the N-oxide. Complexes
6 and 8 were isolated in substance; the remainder were generated
in solution by oxo transfer reactions and identified by their
absorption spectra. These complexes are also extremely dioxy-
gen and water sensitive and must be handled under rigorously
dry, anaerobic conditions. Complexes 6 and 8 in acetonitrile at
Figure 3. Structures of desoxo complexes 2-4 and 5e showing 50%
probability ellipsoids and partial atom labeling schemes. The disordered
carbon atoms of the adamantyl group in 4 were refined isotropically.
ambient temperature slowly autoreduce over 1-2 days to
[WVO(S2C2Me2)2]1-. (The corresponding molybdenum com-
plexes are too unstable to isolate.31) Addition of 0.5 equiv of
the N-oxide to a solution of 1 (δ 2.61) generates methyl NMR
signals corresponding to a 50% conversion to 6 (δ 2.19). No
other signals are observed, indicating that the conceivable
reaction WIV + WVIO f WV-O-WV is not favored. Related
MoV-O-MoV complexes are readily formed, particularly from
and to dioxygen, which causes rapid oxidation to blue
[WVO(S2C2R2)2]1- 22
.
The structures of four desoxo W(IV) complexes are set out
in Figure 3; metric parameters are summarized in Table 3. These
complexes adopt a square pyramidal structure in which the
tungsten atoms are 0.74-0.78 Å above the S4 mean plane
formed by planar dithiolene ligands disposed at dihedral angles
of 126-130°. The bond distances W-OR′ are little affected
by R′ and fall into the range 1.826(4)-1.868(2) Å. The bond
angles W-O-C are somewhat variable because of packing
interactions, but for the most part occur in the interval 142-
(34) Craig, J. A.; Harlan, E. W.; Snyder, B. S.; Whitener, M. A.; Holm,
R. H. Inorg. Chem. 1989, 28, 2082-2091.
(35) Doonan, C. J.; Slizys, D. A.; Young, C. G. J. Am. Chem. Soc. 1999,
121, 6430-6436.

ComparatiVe ReactiVity of Tungsten and Molybdenum
J. Am. Chem. Soc., Vol. 123, No. 9, 2001 1937
tungsten complexes of the type [MVIQ(OR′)(S2C2R2)2]1- (Q )
O, S). This structure is chiral with four inequivalent methyl
groups. In dichloromethane at room temperature, 6 exhibits one
sharp methyl signal. The fluxional nature of this molecule is
shown by the broadening of this signal as the temperature is
decreased, until at 203 to 183 K two equally intense methyl
signals appear at δ 1.9-2.4 (not shown). This observation is
consistent with a species of mirror symmetry, such as a trigonal
prismatic intermediate in the inversion of the complex by a
trigonal twist mechanism that would equilibrate the four methyl
groups.
The structure of 6 is compared in Figure 4b with that of Rc
W-DMSOR determined with data up to 2.0 Å resolution,27 from
which it is seen that the enzyme site in the six-coordinate model
has the common feature of two cis oxygen atoms. The bond
distances W-O ) 1.76 and 1.89 Å and W-S ) 2.44 Å from
tungsten EXAFS correspond rather closely to those of 6 (Table
3). Analysis of positional deviations of the WS4O2 coordination
spheres of both structures affords a weighted root-mean-square
deviation of 0.39 Å at this stage of resolution of the protein
structure. While we cannot be precise about the structural
differences, we can assert that 6 and the protein site have a
similar overall stereochemistry with comparable bond lengths.
Kinetics of Oxo Transfer with S-Oxide and N-Oxide
Substrates. With desoxo W(IV) and monooxo W(VI) com-
plexes synthesized and structurally characterized, a kinetics
study of reaction 3 with biological substrates XO ) N-oxide
and S-oxide becomes feasible. Our first-generation Mo/W
analogues [MIV(OSiR3)(bdt)2]1- are not suitable in this regard
because of the slow to nil reaction rates with S-oxides at ambient
temperature.20,21
Figure 4. (a) Structure of monooxo complex 6 showing 50%
probability ellipsoids and a partial atom labeling scheme. (b) Best-fit
superposition of the cores of 6 and the active site of oxidized Rc
W-DMSOR.27 X-ray structural data for the enzyme in PDB format were
obtained from the Brookhaven Protein Data Bank. The WS4O2 portions
of 6 and the enzyme were fitted by using the subroutine OFIT of the
SHELXS package.
[WIV(OR′)(S2C2R2)2]1- + XO f
[WVIO(OR′)(S2C2R2)2]1- + X (3)
-d[WIV]/dt ) k2[WIV][XO]
(4)
1
Reactivity with S-oxides was first examined by H NMR.
Addition of equimolar DMSO to a solution of complex 1 in
acetonitrile resulted in no appreciable conversion to 6 over ∼10
h. However, complete conversion to 6 in neat DMSO occurred
within 4 h accompanied by development of the red-violet color
and the upfield-shifted methyl signal (δ 2.19) characteristic of
this complex. Thereafter, reaction 3 with XO ) S-oxide was
examined under pseudo-first-order conditions and monitored by
UV-visible absorbance changes. The time course of the system
1/DMSO is depicted in Figure 5. The strong absorptions of 1
at 299 and 319 nm decrease in intensity with time and the
features at 408, 514, and 637 nm increase. A tight isosbestic
point is observed at 337 nm. The final spectrum is identical
with that of 6 measured separately. Because of the relatively
slow reaction rate of this system (k2 ) 3.9 × 10-5 M-1 s-1 at
298 K), the only other S-oxide examined was TMSO, which
under the same conditions reacts 23 times faster. In this and all
other reaction systems with S-oxide substrates, kobs is linearly
dependent on [S-oxide]0 even when the substrate is present in
large excess. Saturation effects are not observed; Michaelis-
Menten kinetics, as found in certain molybdenum-based oxo
transfer systems, do not apply.36,38 All reactions adhere to the
second-order rate law 4. Rate constants and activation param-
eters are collected in Table 4. The latter were evaluated from
the Eyring equation; representative linear plots of ln(k/T) vs
1/T for three reactions are set out in Figure 6.
neutral MoIVO and MoVIO2 complexes,34,35 but thus far not from
anionic bis(dithiolene) complexes with these oxometal groups.36,37
Using the strong oxo donor Ph2Se18O, we have established direct
oxygen atom transfer to complex 1 with formation of
[W18O(OPh)(S2C2Me2)2]1-, demonstrated by IR spectroscopy.
On this basis, we assume direct transfer from N-oxide and
S-oxide substrates to this and other W(IV) complexes in the
reactions that follow.
The structure of complex 6, described briefly earlier,1 is
shown in Figure 4a; bond angles and distances are available in
Table 3. The molecule exhibits an irregularly distorted octa-
hedral stereochemistry with cis oxo (O1) and phenolate (O2)
ligands. Atoms S(2,3) are mutually trans and cis to O(1,2).
Distortions from an octahedral arrangement include a transoid
S2-W-S3 angle of 153.49(5)° and a dihedral angle between
dithiolene rings of 99.8°. Values for octahedral/trigonal prismatic
stereochemistry are 180°/136° and 90°/120°. The trans influence
of the oxo ligand is evident from the W-S1 bond distance of
2.492(1) Å, 0.07 Å longer than the mean of the three cis W-S
bonds. Complex 6 is isostructural with [WO(OSiButR2)(bdt)2]1-
(R ) Me, Ph),20 [WVIS(OSiButPh2)(bdt)2]1- 20
(OSiButPh2)(bdt)2]1- 21
a relationship that establishes the dis-
,
and [MoO-
,
torted cis-octahedral structural motif for molybdenum and
(36) Das, S. K.; Chaudhury, P. K.; Biswas, D.; Sarkar, S. J. Am. Chem.
Soc. 1994, 116, 9061-9070.
(37) Oku, H.; Ueyama, N.; Kondo, M.; Nakamura, A. Inorg. Chem. 1994,
33, 209-216.
Several variables of the W(IV) reactant were examined with
TMSO as the substrate. Replacement of the chelate ring methyl

1938 J. Am. Chem. Soc., Vol. 123, No. 9, 2001
Sung and Holm
Figure 6. Eyring plots affording activation parameters for the reaction
[WIV(O-p-C6H4X)(S2C2Me2)2]1- + TMSO f [WVIO(O-p-C6H4X)(S2C2-
Me2)2]1- + TMS (X ) CN, H, NH2) in acetonitrile solutions.
Figure 5. Absorption-spectral changes in the pseudo-first-order reaction
1 + Me2SO f 6 + Me2S in acetonitrile at 25 °C with [1]0 ) 1.2 mM
and [Me2SO]0 ) 2.3 M; spectra were recorded every 30 min. The final
spectrum is identical with that of isolated complex 6.
Table 4. Kinetics Data for Substrate Reduction by Oxo Transfer
Figure 7. The time course of the reaction system with [3]0 ) 1.1 mM
and [NMMO]0 ) 2.8 mM in acetonitrile at 298 K. The reaction was
monitored at 487 nm. The value of (At - A0)/(A∞ - A0) is equal to the
mole fraction of product 8; the second-order rate constant k is directly
obtained from the linear plot (inset).
enhancement. Replacement of phenolate with isopropoxide or
2-adamantyloxide, alkoxide ligands with a closer electronic
resemblance to serinate, results in markedly diminshed reactiv-
ity. Complexes 3 and 4 do not react with either DMSO or TMSO
in ca. 1000-fold excess in acetonitrile at room temperature. At
373 K, these complexes are partially oxidized to 8 and 9 in
very slow reactions; rate constants were not determined.
Complex 3 is cleanly oxidized in neat TMSO (11.1 M) at 298
K. The time course of the reaction over 24 h is very similar to
that in Figure 5 but is much slower; a tight isosbestic point
occurs at 334 nm. The second-order rate constant k2 ≈ 4 ×
10-6 M-1 s-1 is obtained. Although the comparison is inexact
because of difference in solvent, the ratio k2(1)/k2(3) ≈ 220
emphasizes the appreciable reduction in rate upon replacement
of phenolate with the more electron-rich isopropoxide ligand.
The relatively slow reactivity of complex 3 allowed kinetics
measurements with Me3NO and NMMO under second-order
conditions. Spectrophotometric monitoring of the reaction
system 3/NMMO is shown in Figure 7. Adherence to second-
order kinetics is apparent; the rate constant is obtained directly
from the linear plot. Rate constants are ca. 103-105 larger than
for the reactions of 1 and 3 with S-oxides. In addition to reaction
a Acetonitrile solution, 298 K. b For the reaction of 2 with (CH2)4SO,
k2 ) 3.0(1) × 10-3 M-1 s-1 at 298 K. c Measured in neat substrate at
298 K.
substituents (σI ) -0.05) with the less electron-releasing phenyl
groups (σI ) 0.10) in complex 2 results k2(2)/k2(1) ) 3.3.
Substitution of the terminal phenolate ligand in 6 with a range
of para-substituted phenolate ligands in 5a-e affords a ca. 90-
fold range of rates, with the most electron-withdrawing sub-
stituent (X ) CN, σp ) 0.69) producing the largest rate

ComparatiVe ReactiVity of Tungsten and Molybdenum
J. Am. Chem. Soc., Vol. 123, No. 9, 2001 1939
3, Me3NO has been employed previously in the oxidations
[WIVO(bdt)2]2- f [WVIO2(bdt)2]2- 20,39 and [WIV(OSiR3)(bdt)2]1-
f [WVIO(OSiR3)(bdt)2]1- 20
.
The reactions reported here contribute to the relatively small
body of well-defined direct oxo transfer reactions mediated by
tungsten.19,20,39-42 A majority of these reactions have been
tabulated.19,40 Other than our initial report,1 kinetics data for
tungsten-mediated reactions have been confined to the systems
[WO(bdt)2]2-/Me3NO, for which only kobs at room temperature
was reported,39 and [WO2(mnt)2]2-/(MeO)2PhP,(MeO)3P, for
which rate constants and activation parameters were deter-
mined.19 Our thermodynamic oxo transfer reaction scale is based
on ∆G or ∆H values for the reaction couples X(g) + 1/2O2(g) f
XO(g). For the couples involving Me3NO, Me2SO, and Ph3AsO,
∆H ) -2, -27, and -43 kcal/mol, respectively.43 The
spontaneous occurrence of reaction 5 demonstrates that complex
1 is a stronger oxo acceptor than Ph3As, and that for the couple
Figure 8. Hammett plot of log(kX/kH) vs σp for the reaction series 1
and 5a-e + TMSO f 6 and 10a-e + TMS in acetonitrile at 298 K.
1
1 + /2O2(g) f 6, ∆H < -43 kcal/mol. This marks 1 as a
relatively strong oxo acceptor, comparable to, e.g., MoCl4 (∆H
Table 5. Properties of Substrates XO (the quantities GB and PA
refer to the reaction M(g) + H+ ) MH+(g) at temperature T)
(g)
[W(OPh)(S2C2Me2)2]1- + Ph3AsO f
[W(OPh)(S2C2Me2)2]1- + Ph3As (5)
) -41 kcal/mol). In an attempt to set a lower limit of 1 on the
oxo transfer scale, further reactions were investigated. No
reaction was observed between 1 and 100 equiv of Ph3PO (∆H
) -74 kcal/mol) at 50 °C for 1 day. Under similar conditions
with Ph3P and 6, the reverse reaction of substrate oxidation did
not occur over a 10 h period. With 200-700 equiv of the more
basic Et3P, a slow reaction at room temperature proceeded to
form a yellow solution, the absorption spectrum of which
suggests that the product is the bis(phosphine) complex [W(PEt3)2-
(S2C2Me2)2], analogous to [W(P(OEt)3)2(bdt)2] and [Mo-
(PMePh2)2(bdt)2], which have been isolated.20,21 While the lack
of a reaction cannot establish a lower limit, the implication is
that complex 1 is a less strong oxo acceptor than Ph3P or WCl4
(∆H ) -67 kcal/mol)
Mechanistic Considerations. The most thorough experi-
mental analysis of the mechanism of a molybdenum- or
tungsten-mediated oxo transfer is that of [MoIVO(ButL-NS)2]
with N-oxides, S-oxides, and Ph3AsO.44 Although the minimal
reaction in question, MoIVO + XO f MoVIO2 + X, is not that
examined here (nor a known enzymatic process), certain
considerations are pertinent as in the corresponding bis-
(dithiolene)molybdenum systems.31 Reference is made to the
rate constants and activation parameters in Table 4. All reactions
are second order with |∆H| > |T∆S| and large negative entropies
of activation for N-oxides (-19 to -27 eu), S-oxides (-24 to
-33 eu), and Ph3AsO (-24 eu), which indicate associative
transition states. For the most part T∆Sq is approximately
constant, such that ∆Hq is 55-68% of ∆Gq. We look into the
nature of the transition state by examining the effects of ligand
and substrate variation on reaction rates which are expected to
influence primarily activation enthalpies. Ligand effects are
considered first.
a D ) -∆HX/XO + ∆fH°m(O(g)); ref 43. b Gas basicity, GB(M) )
-∆G°; ref 46. c Proton affinity, PA(M) ) -∆H° ) ∆fH°(M) +
∆fH°(H+) - ∆fH°(MH+); ref 46. d Reference 47. e Reference 48.
f Reference 49. g Reference 50. h Reference 51. i Reference 52. j Ref-
erence 53. k Reference 54. l Reference 55. m Reference 56.
making, is significant in the transition state. The first ratio
suggests that reducing the basicity of the dithiolene by substitut-
ing methyl with phenyl leads to a small rate enhancement. The
second ratio indicates that substitution of a more basic oxygen
ligand, PriO- for PhO-, without an increase in steric hindrance
at the metal site significantly, retards the reaction rate. There-
after, in reactions with TMSO the electronic effect of the axial
ligand was examined by using the set of complexes 1 and 5a-
e, in which the para-substituent X of the axial phenolate ligands
p-XC6H4O- was varied without introducing significant steric
differences. The reaction products were shown by spectral
comparison with authentic samples to be 6 and 10a-e (Figure
2). As shown in Figure 8, the rate constants fit the Hammett
equation log(kX/kH) ) Fσp. The positive slope F ) 2.1 shows
that electron-withdrawing groups accelerate the rate, with
The relative rates k2(2)/k2(1) ) 3.3 and k2(3)/k2(1) ≈ 0.0045
with TMSO imply that substrate binding, i.e., W‚‚‚OX bond
limiting ratios being kCN/kH ) 39 and kH/kNH ) 2.4. The effects
2
(38) Berg, J. M.; Holm, R. H. J. Am. Chem. Soc. 1985, 107, 925-932.
(39) Ueyama, N.; Oku, H.; Nakamura, A. J. Am. Chem. Soc. 1992, 114,
7310-7311.
are not large but the trend is clear from these and the preceding
rate constant ratios: rates increase in a manner qualitatively
consistent with diminution of electron density at the tungsten
center.
(40) Holm, R. H. Chem. ReV. 1987, 87, 1401-1449.
(41) Yu, S.-B.; Holm, R. H. Inorg. Chem. 1989, 28, 4385-4391.
(42) Lee, S.; Staley, D. L.; Rheingold, A. L.; Cooper, N. J. Inorg. Chem.
1990, 29, 4391-4396.
Assembled in Table 5 are certain thermodynamic properties
of substrates XO, including ∆HX/XO in the thermodynamic
reactivity scale, bond dissociation energies D, gas basicities GB,
(43) Holm, R. H.; Donahue, J. P. Polyhedron 1993, 12, 571-589.
(44) Schultz, B. E.; Holm, R. H. Inorg. Chem. 1993, 32, 4244-4248.

1940 J. Am. Chem. Soc., Vol. 123, No. 9, 2001
Sung and Holm
proton affinities PA, and solution basicity constants.43,46-56 We
take the faster reaction of TMSO than DMSO with complex 1,
expressed by k2TMSO/k2DMSO ) 23, to reflect a small steric effect
owing to the ring structure of TMSO (and possibly a slightly
higher basicity), which should facilitate binding to the tungsten
center. The D values of the two substrates are essentially the
same and aqueous basicities are quite similar. Properties are
conveniently considered in terms of the series 6-8. The bond
energy order P-O > As-O > S-O > N-O in series 6 is
clear-cut and is based mainly on thermochemical data. In the
proton affinity series 7, the two N-oxides are the most basic.
The order is unclear for Ph3PO and Ph3AsO whose PA values
are quoted from “bracketing” experiments intended to define a
limited range of values for a series of bases.57 Aqueous
protonation constants for the reaction XO + H+ a XOH+ were
not all measured by the same technique. Examination of all
available data indicates that the order of series 8 is appropriate.
These series may be compared to the rate constant orders 9 and
10. In series 9, Me3NO is placed at the fast end because its
reaction with 1 under the same conditions for Ph3AsO is too
rapid to measure by the spectrophotometric method used. Ph3-
PO is placed at the slow end because of the lack of reaction
between between 1 and this compound in excess. The order in
series 10 is experimental.
Figure 9. Mechanistic pathways for oxygen atom transfer from
substrate XO to [WIV(OR′)(S2C2R2)2]1-. If ∆Hq1 > ∆Hq , the transition
2
state is early and is dominated by W‚‚‚OX bond-making with little
X-O bond-weakening. The opposite situation leads to a late transition
state with substantial X-O bond weaking and concomitant atom transfer
and oxidation of initial W(IV).
another ligand. The indicated intermediate and by the Hammond
postulate the two transition states are expected to have structures
approaching that of product complex 6.58 The observation that
series 9 parallels series 8 suggests an early transition state
involving considerable WIV‚‚‚OX bond making.
Hydrolysis Reactions. As noted earlier, the W(IV,VI)
complexes isolated in this work are highly sensitive to water.
When the green-brown and brown acetonitrile solutions of 1
and 3, respectively, are treated with 1000-20000 equiv of water,
they slowly turn to the characteristic orange-brown color of 11.
The rates of the hydrolysis of complexes 1 and 3 have been
determined at room temperature. Spectrophotometric monitoring
of the reactions reveals well-defined isosbestic points at 632 (1
f 11) and 624 nm (3 f 11). As shown in Figure 10, pseudo-
first-order plots of kobs vs [H2O] are nonlinear, showing
saturation kinetics at high substrate concentrations. The data
are well-fitted to Michaelis-Menten kinetics reaction 11; double
reciprocal plots generate straight lines which afford first-order
rate constant k2′ and KM. Data are contained in Table 6. The
reaction involves a preequilibrium between 1 or 3 and an aqua-
bound species followed by release of PhOH or PriOH in the
rate-determining step. The rate constant k2′ ) 7.8 × 10-4 s-1
for the hydrolysis of 1 indicates a rather slow reaction. The
value k2′/KM ) 1.7 × 10-4 M-1 s-1 is comparable to rate
constants for S-oxides (Table 4), and emphasizes the need for
D: Ph3PO > Ph3AsO > R2SO > Me3NO, pyO
PA: Me3NO > pyO J Ph3PO ≈ Ph3AsO > R2SO
pKa: Me3NO > Ph3AsO J pyO > R2SO J Ph3PO
k2(1): Me3NO > Ph3AsO > pyO > R2SO . Ph3PO
k2(3): R3NO > R2SO
(6)
(7)
(8)
(9)
(10)
The inverse of series 6 would be expected to apply to k2
values if substrate X-O bond weakening dominated the
transition state. However, Ph3AsO has a much higher reactivity
than would be expected on this basis. Series 7 or 8 is pertinent
if proton affinities in the gas or aqueous phase reflect the
substrate binding affinities toward W(IV) centers. Possible
mechanistic pathways are summarized in Figure 9. If ∆Hq
>
1
∆Hq , the higher transition state is early and is dominated by
2
W‚‚‚OX bond making with little X-O bond weakening. If ∆Hq
2
> ∆Hq , the higher transition state is late and involves
1
considerable X-O bond weakening or dissociation and incipient
and concomitant oxygen atom transfer and oxidation of the
initial W(IV) center. Either situation involves an extent of
structural rearrangement as the five-coordinate complex acquires
[W(OR′)(S2C2Me2)2]1- + H2O yk1′z
k-1
′
[W(OR′)(S2C2Me2)2‚(H2O)]1- 9k2′8
[WO(S2C2Me2)2]2- + R′OH + H+ (11)
(45) Laughlin, L. J.; Young, C. G. Inorg. Chem. 1996, 35, 1050-1058.
(46) Hunter, E. P.; Lias, S. G. J. Phys. Chem. Ref. Data 1998, 27, 413-
656 and references therein.
(47) Acree, W. E., Jr.; Tucker, S. A.; Ribeiro da Silva, M. D. M. C.;
Matos, A. R.; Gonc¸alves, J. M.; Ribeiro da Silva, M. A, V. J. Chem.
Thermodyn. 1995, 27, 391-398.
(48) Nyle´n, P. Z. Anorg. Allg. Chem. 1941, 246, 227-330.
(49) Haaland, A.; Thomassen, H. J. Mol. Struct. 1991, 263, 299-310.
(50) Chmurzyn´ski, L.; Pawlak, Z. J. Chem. Thermodyn. 1998, 30, 27-
35.
k2′[H2O]
k-1′ + k2′
k1′
kobs
)
, KM )
KM + [H2O]
dry conditions when examining the reduction of these substrates.
The rate constant for 3 is nearly the same as that for 1. The
substantial primary isotope effects k2′H/k2′D ) 3.1 and 5.6 reflect
(51) Shaofeng, L.; Pilcher, G. J. Chem. Thermodyn. 1988, 20, 463-
465.
(52) (a) Jaffe´, H.; Doak, G. O. J. Am. Chem. Soc. 1955, 77, 4441-
4444. (b) Kubota, T. J. Am. Chem. Soc. 1965, 87, 458-468.
(53) Landini, D.; Modena, G.; Scorrano, G.; Taddei, F. J. Am. Chem.
Soc. 1969, 91, 6703-6707.
(56) Klofutar, C.; Krasˇovec, F.; Kusˇar, M. Croat. Chim. Acta 1968, 40,
23-28 and references therein.
(57) Tran, V. T.; Munson, B. Org. Mass Spectrom. 1986, 21, 41-46.
(58) A somewhat related discussion has been presented for the reaction
of oxo donors with carbonyl ligands: Kelly, A. M.; Rosini, G. P.; Goldman,
A. S. J. Am. Chem. Soc. 1997, 119, 6115-6125.
(54) Jenks, W. S.; Matsunaga, N.; Gordon, M. J. Org. Chem. 1996, 61,
1275-1283.
(55) Curci, R.; Furia, F.; Levi, A.; Lucchini, V.; Scorrano, G. J. Chem.
Soc. Perkin Trans. 2 1975, 341-344.
(59) Lorber, C.; Plutino, M. R.; Elding, L. I.; Nordlander, E. J. Chem.
Soc., Dalton Trans. 1997, 3997-4003.

ComparatiVe ReactiVity of Tungsten and Molybdenum
J. Am. Chem. Soc., Vol. 123, No. 9, 2001 1941
Figure 10. Dependence of the observed rate constants on substrate concentration for hydrolysis reaction 11 involving (a) 1 and (b) 3.
Table 6. Kinetics Data of the Hydrolysis of Complexes 1 and 3 at 298 K
complex
substrate
k2′ (s-1
)
KM (M)
k2′/KM (M-1 s-1
)
k2′H/k2′D
[W(OPh)(S2C2Me2)2]1- (1)
H2O
D2O
H2O
D2O
7.8(2) × 10-4
2.5(1) × 10-4
5.2(1) × 10-4
9.3(1) × 10-5
4.5
8.4
7.1
7.0
1.7 × 10-4
3.0 × 10-5
7.3 × 10-5
1.3 × 10-5
3.1
[W(OPri)(S2C2Me2)2]1- (3)
5.6
proton transfer in the transition state. In the proposed reaction
12, water may first associate with the complex by hydrogen
bonding to the axial ligand, followed by formation of a concerted
transition state with W‚‚‚OH2 and H‚‚‚OR′ bond making and
W-OR′ bond weakening. Thereafter, proton transfer is com-
pleted, R′OH dissociates, and the WIVdO bond is developed.
The larger isotope effect is associated with the more basic axial
ligand. The rate comparison k2Mo/(k2′/KM)W ≈ 70, involving 1
other classes of compounds, isoelectronic molybdenum and
tungsten molecules are isostructural and essentially isometric.
We have already noted that the active site formulation
[WVIO(O‚Ser)(S2pd)2] is consistent with crystallographic and
EXAFS results27 and the structure is related to 6 (Figure 4).
DMSORs reduce both DMSO and Me3NO,62,63 whereas
TMAORs thus far reduce only N-oxides at a measurable rate.26
Although the small database renders any inquiry preliminary,
we have become interested in whether relative rates of isoen-
zymes are intrinsic to the metal. To proceed, we examine Table
7, which contains kinetic data for oxo transfer by pairs of
isoelectronic molybdenum and tungsten complexes. Upon
comparison of data for isoenzymes and synthetic complexes, a
parallel behavior emerges: oxo transfer from substrate to metal
is faster with tungsten,64 and oxo transfer to substrate from metal
is faster with molybdenum. Oxo transfer from and to substrate
and [Mo(OPh)(S2C2Me2)2]1- 31,60
stronger W-OR′ bond.
,
is consistent with the expected
(60) Unlike 1, [Mo(OPh)(S2C2Me2)2]1- does not exhibit saturation
behavior over the water concentration range that afforded measurable rates
by the spectrophotometric method used. Consequently, second-order rate
constants are compared.
Comparative Molybdenum and Tungsten Reactivities.
Summarized in Table 1 are relative reactivities of three types
of molybdenum/tungsten isoenzymes. Structural information is
available only for Rc Mo- and W-DMSOR. A seemingly
reasonable but unproven assumption is that catalytic sites at
the same oxidation level in these isoenzymes have the same
structure. As demonstrated with bis(dithiolenes)1,19-25,37,61 and
(61) Das, S. K.; Biswas, D.; Maiti, R.; Sarkar, S. J. Am. Chem. Soc.
1996, 118, 1387-1397.
(62) Adams, B.; Smith, A. T.; Bailey, S.; McEwan, A. G.; Bray, R. C.
Biochemistry 1999, 38, 8501-8511.
(63) Simala-Grant, J. L.; Weiner, J. H. Eur. J. Biochem. 1998, 251, 510-
515 and references therein.

1942 J. Am. Chem. Soc., Vol. 123, No. 9, 2001
Sung and Holm
Table 7. Comparative Kinetics Data of M ) Mo and W Oxo Transfer at 298 K
a k2′/KM from Michaelis-Menten kinetics performed in MeCN/H2O ) 1/1 (v/v) at 20 °C. b This work.
leads to metal oxidation and reduction, respectively. Elsewhere
we have enumerated properties that may contribute to dif-
ferential reaction rates of molybdenum/tungsten pairs,19 among
which are redox potentials. It is always true that EMo > EW, a
behavior illustrated by EMo - EW ) 0.29, 0.20 V for the
toward the axial ligand and dithiolene ligands disposed at
dihedral angles of 126-130°. The complex [WVIO(OPh)-
(S2C2Me2)2]1- adopts a distorted cis-octahedral structure and
is fluxional in solution. These structures are taken as those of
oxidized and reduced protein-bound sites when unconstrained
by the protein. Comparison with the structure of Rc W-DMSOR
reveals similarities in overall stereochemistry and W-O and
W-S bond lengths.
reversible redox couples [MO(S2C2Me2)2]2-/1-;1-/0 22,24
While
.
it is conceivable that parallel reactivity ratios, in the sense of
kW/kMo >1 or <1, in the two tables is coincidental, the reactivity
ratios could also reflect the inherent oxidizability or reducibility
of the metal in a consistent manner. Given that there is a “metal
effect” on rates, the preceding concept of the transition state
requires alteration. Following intermediate formation, the system
may pass through a highly concerted transition state with X-O
bond-weakening and W-O bond-making with incipient atom
transfer and metal oxidation. Further consideration of the nature
of the transition state may be assisted by theoretical studies of
the reaction coordinate, as has been done for the reactions
MoVIO2 + R3P f MoIVO + R3PO (R ) H, Me).65
Summary. This investigation presents the synthesis and
structures of potential or real analogues of the active sites of
tungsten-containing DMSOR, TMAOR, and possibly FMDH
(iso)enzymes, and the first extensive kinetics study of substrate
reduction via oxo transfer mediated by bis(dithiolene)tungsten
complexes. The following are the principal results and conclu-
sions of this research.
(3) The complexes [WIV(OR′)(S2C2Me2)2]1- (R′ ) Ph,
p-XC6H4, Pri) engage in direct oxo transfer reactions with N-,
As-, S-, and Se-oxides to afford [WVIO(OR′)(S2C2Me2)2]1- in
clean transformations. Direct atom transfer was demonstrated
with Ph2Se18O.
(4) The reactions in (3) are second order and proceed through
associative transition states (∆Sq ) -19 to -33 eu). For
[WIV(OPh)(S2C2Me2)2]1-, rates or rate constants k2 follow the
substrate order Me3NO > Ph3AsO > pyO > R2SO . Ph3PO.
The W(IV) complexes are intrinsically capable of reducing both
N-oxides and S-oxides, although at very different rates. This
behavior contrasts with the apparent lack of reduction of
S-oxides by TMAORs, and suggests possible enzymic reduction
of S-oxides but at very slow rates relative to N-oxide reduction.
(5) Comparison of relative reactivities of Mo/W isoenzymes
and pairs of isostructural Mo/W complexes at parity of ligation
reveals that oxo transfer from substrate to metal is faster with
tungsten and to substrate from metal is faster with molybdenum.
These results demonstrate a kinetic metal effect in direct oxo
transfer reactions for analogue complexes and for isoenzymes
provided the catalytic sites are isostructural.
(6) The results in (4) and (5) and the influence on rates by
variation of axial oxygen ligands leads to the tentative concep-
tion of a highly concerted transition state involving primarily
W-OX bond-making and incipient atom transfer and metal
oxidation.
(7) The W(IV,VI) complexes reported here are highly water
sensitive. The kinetics of hydrolysis of [WIV(OR′)(S2C2Me2)2]1-
(R′ ) Ph, Pri) exhibit saturation kinetics with k2′ ∼ 10-3 s-1
and primary isotope effects indicative of proton transfer in the
transition state.
(1) The desoxo complexes [WIV(OR′)(S2C2R2)2]1- (R ) Me,
Ph) are readily prepared by carbonyl substitution reactions of
[W(CO)2(S2C2R2)2], further emphasizing the utility of the
dicarbonyls as precursors to variously substituted bis(di-
thiolene)W(IV) species. These complexes readily react with Me3-
NO to afford the monooxo complexes [WVIO(OR′)(S2C2R2)2]1-
,
two of which ([WVIO(OR′)(S2C2Me2)2]1-, R ) Ph, Pri) have
been isolated.
(2) The structures of the desoxo W(IV) complexes are square
pyramidal with the metal atom 0.74-0.78 Å above the S4 plane
(64) We have observed one possible deviation from this behavior. In
the reduction of TMSO by 3 (Table 4) and [Mo(OPri)(S2C2Me2)2]1-31 in
Mo
neat TMSO, k2W/k2 ∼ 0.8. Given the errors in the rate constants of these
extremely slow reactions, we conclude that the rates are essentially the same.
(65) (a) Pietsch, M. A.; Hall, M. B. Inorg. Chem. 1996, 35, 1273-1278.
(b) Thomson, L. M.; Hall, M. B. J. Am. Chem. Soc. Submitted for
publication.
Last, we reemphasize that the databases of comparative Mo/W
kinetics used in (5) are not extensive, the similarities or

ComparatiVe ReactiVity of Tungsten and Molybdenum
J. Am. Chem. Soc., Vol. 123, No. 9, 2001 1943
differences in enzyme active site structures are largely unknown,
rate-limiting features of catalytic cycles have not been defined,
and reactivity differences are relatively small for substrate
reduction reactions. It will be most interesting to learn if further
data sustain a parallel reactivity between isoenzymes and
analogue complexes. Current research is directed at clarifying
(6) and investigating other reactions of tungstoenzymes.
and Professors H. Schindelin and M. B. Hall for access to refs
7 and 65b, respectively, prior to publication.
Supporting Information Available: X-ray crystallographic
data for structure determination of the five compounds listed
in Table 2 (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
Acknowledgment. This research was supported by NSF
Grant CHE-98-76457. We thank B. S. Lim for useful discussions
JA0036559