Syntheses, Spectroscopy, and Redox Chemistry of Encapsulated Oxo-Mo(V) Centers: Implications for Pyranopterin-Containing Molybdoenzymes

Article abstract of DOI:10.1021/ic034821r

Coordination by at least four sulfur donors to an embedded molybdenum center has been found to be a common feature in the crystal structures of many mononuclear molybdenum enzymes. In an effort to model embedded molybdenum centers, we have synthesized dendritic thiolate ligands and their oxo-molybdenum complexes containing a [Mo^VOS₄] ^- core. These compounds have been isolated in pure form as blue solids or gummy materials, and the molecular nature of these compounds has been confirmed by electrospray ionization mass spectrometry and infrared, electron paramagnetic resonance, and UV-vis spectroscopies. The dendritic complexes exhibit little variation in their broad S → Mo charge transfer band (λ_max ～ 600 nm; ε ～ 6000 M^-1 cm ^-1), M=O vibration energy (941-943 cm^-1), and EPR g-values (g_∥ ～ 2.02; g^⊥ 1.98; g_av ～ 1.99). The spectroscopic data confirm the integrity of the square pyramidal [Mo^VOS₄]^- core with little geometric distortions, suggesting that the electronic structure at the metal center is not perturbed by the ligand architecture. The electronic structure of these complexes, calculated by the density functional theory, demonstrates a similar composition of the HOMO. In complexes 6 and 7a, the energy of the HOMO orbital might be modulated by the difference in the electronic structure of the ligands. The Mo(V/IV) reduction potentials vary as a function of the dielectric constant and the donor number of the solvent. The kinetics of the reduction is influenced by the reorganization of the geometry and the encapsulating effect. We suggest that protein structure imposed microenvironments may control the dielectric properties and hence the redox properties of the metal center in many metallobiomolecules.

Full text of DOI:10.1021/ic034821r

Inorg. Chem. 2003, 42, 7489−7501
Syntheses, Spectroscopy, and Redox Chemistry of Encapsulated
Oxo−Mo(V) Centers: Implications for Pyranopterin-Containing
Molybdoenzymes
Partha Basu,* Victor N. Nemykin, and Raghvendra S. Sengar
Department of Chemistry and Biochemistry, Duquesne UniVersity, Pittsburgh, PennsylVania 15282
Received July 14, 2003
Coordination by at least four sulfur donors to an embedded molybdenum center has been found to be a common
feature in the crystal structures of many mononuclear molybdenum enzymes. In an effort to model embedded
molybdenum centers, we have synthesized dendritic thiolate ligands and their oxo−molybdenum complexes containing
a [MoVOS ]- core. These compounds have been isolated in pure form as blue solids or gummy materials, and the
4
molecular nature of these compounds has been confirmed by electrospray ionization mass spectrometry and infrared,
electron paramagnetic resonance, and UV−vis spectroscopies. The dendritic complexes exhibit little variation in
their broad S f Mo charge transfer band (λmax ∼ 600 nm; ꢀ ∼ 6000 M-1 cm-1), ModO vibration energy (941−
943 cm-1), and EPR g-values (g ∼ 2.02; g ∼ 1.98; gav ∼ 1.99). The spectroscopic data confirm the integrity of
|
the square pyramidal [MoVOS ]- core with little geometric distortions, suggesting that the electronic structure at the
4
metal center is not perturbed by the ligand architecture. The electronic structure of these complexes, calculated by
the density functional theory, demonstrates a similar composition of the HOMO. In complexes 6 and 7a, the energy
of the HOMO orbital might be modulated by the difference in the electronic structure of the ligands. The Mo(V/IV)
reduction potentials vary as a function of the dielectric constant and the donor number of the solvent. The kinetics
of the reduction is influenced by the reorganization of the geometry and the encapsulating effect. We suggest that
protein structure imposed microenvironments may control the dielectric properties and hence the redox properties
of the metal center in many metallobiomolecules.
Introduction
Mononuclear molybdenum containing enzymes catalyze
a wide variety of reactions such as the reduction of nitrate
to nitrite, reduction of dimethyl sulfoxide to dimethyl sulfide,
reduction of arsenate to arsenite, or the oxidation of sulfite
to sulfate.1-3 These reactions play critical roles in the global
Figure 1. Structure of the cofactor and coodination environment of
dimethyl sulfoxide reductases (DMSOR), trimethylamine N-oxide reductases
cycling of elements such as nitrogen, sulfur, carbon, and
arsenic. The catalytic reactions take place at the molybdenum
center, which is coordinated by the ene-dithiolate moiety of
one or two pyranopterin cofactors (Figure 1). Hille3 classified
these enzymes into three distinct families: xanthine oxidase
family, sulfite oxidase family, and dimethyl sulfoxide re-
ductase (DMSOR) family; representative members of each
family have now been crystallographically characterized.
Using the inferred sequences and known biochemistry, the
(TMAOR), periplasmic nitrate reductases (NapA), and arsenite oxidase
(ArO).
nitrate reductases have been classified, and on the basis of
a detailed sequence analysis, the evolutionary lineages among
these enzymes have been proposed.4
The members of the DMSO reductase family are prokary-
otic enzymes in which the molybdenum centers are co-
ordinated by two pyranopterin cofactors as confirmed
crystallographically. For example, in the crystal structures
of dissimilatory nitrate reductase (i.e., NapA) from
* To whom correspondence should be addressed. E-mail: basu@duq.edu.
(1) Enemark, J. H.; Young, C. G. AdV. Inorg. Chem. 1993, 40, 1-88.
(2) Zumft, W. G. Microbiol. Mol. Biol. ReV. 1997, 61, 533-615.
(3) Hille, R. Chem. ReV. 1996, 96, 2757-2816.
(4) Stolz, J. F.; Basu, P. ChemBioChem 2002, 3, 198-206.
Inorganic Chemistry, Vol. 42, No. 23, 2003 7489
10.1021/ic034821r CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/24/2003

Basu et al.
DesulfoVibrio desulfuricans, the molybdenum center is
coordinated by two ene-dithiolates of two pyranopterin
cofactors and one cysteinato sulfur atom.5 The presence of
two pyranopterin cofactors per molybdenum atom has also
been confirmed in the crystal structures of DMSOR from
Rhodobacter sphaerodies6 and R. capsulatus,7,8 trimethy-
lamine oxide reductase (TMAOR) from Shewanella mas-
silia,9 arsenite oxidase from Alcaligenes faecalis,10 and
formate dehydrogenases (FDH-H and FDH-N) from Es-
cherichia coli.11 Additional coordination from the protein
side chain to the molybdenum atom is provided by a serinato
oxygen in DMSOR6-8 and TMAOR,9 and by selenocysteine
in FDH,11 while in arsenite oxidase no amino acid coordi-
nates the molybdenum center. Thus, a molybdenum center
coordinated by four sulfur donors emerges as the minimal
motif in members of the DMSOR family. On the basis of
the structural studies of enzymes such as NapA5 and DMSOR
from R. sphaerodies,6 it has been proposed that the desoxo-
molybdenum(IV) center is the catalytically competent state,
and in the fully oxidized state, the metal center is coordinated
by a single terminal oxo group. The high-resolution structure
of DMSOR exhibited a disordered (between penta- and
hexacoordinated forms) active site.6b In addition, structural
studies have demonstrated that the molybdenum centers are
buried within the protein matrix, and as such the coordinating
ligands are not exposed to the surface. They are, however,
associated with either a channel or a crevice presumably for
substrate entrance and product release.
to the metal center in the 5+ oxidation state, although a
recent pulsed EPR investigation17 suggests a Mo-OH unit
as the EPR active species. Taken together, it appears that
the minimal picture of molybdenum(V) centers that emerges
is of a [MoVOS4] core buried inside the protein. Interestingly,
when other biomimetic electroactive centers are encapsulated,
such as hemes and iron-sulfur clusters, they exhibit a
significant modulation in the reduction potentials.18-20 We
recently reported the synthesis of [MoVO(p-SC6H4-R)4]-
centers, where R represents a small to large organic
functionality.21 Herein we disclose the complete synthetic
and spectroscopic details, and solvent dependent electro-
chemistry of the oxo-molybdenum complexes coordinated
by dendritic thiolato ligands. Moreover, using density
functional theory, we discuss the electronic structure of
[MoOS4]- cores. We also have calculated the significance
of the steric encumbrance with respect to the protein
chemistry.
Experimental Section
The syntheses of all molybdenum complexes were carried out
in oxygen free dry argon atmosphere using dry distilled solvents
either following the Schlenk techniques or inside an inert (N2)
atmosphere drybox. Syntheses of dendron 1 and 2 were carried
out in air.
Materials and Supplies. Solvents were purchased either from
Aldrich Chemical Co. or Acros Chemical Company and were
purified before use. The solvents were purified as follows:
acetonitrile (MeCN) from CaH2 followed by Li2CO3-KMnO4 and
finally from P2O5; methylene chloride (CH2Cl2) from CaH2; toluene
and THF (THF) from Na-benzophenone; ethanol over sodium
ethoxide; and 2-methoxyethanol over K2CO3. 4-Mercaptobenzoic
acid, oxalyl chloride, dimethyl formamide (DMF), ferrocene,
tetrabutylammonium perchlorate (TBAP), tetraethylammonium
tetrafluoroborate, solium borohydride, and tetraphenyl phosphonium
chloride were purchased from Aldrich Chemical company and were
used without any purification. Silica gel (70-230 mesh) and Bio
Beads SX3 were purchased from Aldrich and Biorad, respectively.
Spectroscopy and Electrochemistry. Infrared spectra were
recorded on a Perkin-Elmer FT-IR 1760X spectrometer in KBr
pellets or NaCl plates. The electronic spectra were recorded on a
Cary 14 spectrophotometer with an OLIS 14 version 2.6.99
operating system or Cary 3 spectrophotometer connected with
constant temperature circulators. Electrochemical measurements
were carried out using a Bioanalytical Systems (BAS) model CV-
50W. Cyclic voltammograms were recorded with a standard three-
electrode system consisting of a Pt-disk working electrode, a Pt-
reference electrode, and a Pt-wire auxiliary electrode. TBAP was
used as the supporting electrolyte, and all voltammograms were
internally referenced to ferrocene. As such, the potentials are
reported with respect to the Fc+/Fc couple, without junction
correction. All cyclic voltammograms were simulated digitally to
During the course of catalysis, the molybdenum center
passes through a 5+ oxidation state, as determined by
electron paramagnetic resonance (EPR),12 magnetic circular
dichroism (MCD),13 and extended X-ray absorption fine
structure (EXAFS) spectroscopies.14-16 These studies also
indicate the presence of one terminal oxo group coordinated
(5) Dias, J. M.; Than, M. E.; Humm, A.; Huber, R.; Bourenkov, G. P.;
Bartunik, H. D.; Bursakov, S.; Calvete, J.; Caldeira, J.; Carneiro, C.;
Moura, J. J. G.; Moura, I.; Roma˜o, M. J. Structure 1999, 7, 65-79.
(6) (a) Schindelin, H.; Kiser, C.; Hilton, J.; Rajagopalan, K. V.; Rees, D.
C. Science 1996, 272, 1615-1621. (b) Li, H.-K.; Temple, C.;
Rajagopalan, K. V.; Schindelin, H. J. Am. Chem. Soc. 2000, 122,
7673-7680.
(7) McAlpine, A. S.; McEwan, A. G.; Bailey, S. J. Mol. Biol. 1998, 275,
613-623.
(8) Schneider, F.; Lo¨we, J.; Huber, R.; Schindelin, H.; Kiser, C.; Kna¨blein,
J. J. Mol. Biol. 1996, 263, 53-69.
(9) Czjzek, M.; Santos, J.-P. D.; Pommier, J.; Giordano, G.; Mejean, V.;
Haser, R. J. Mol. Biol. 1998, 284, 435-447.
(10) Ellis, P. J.; Conrads, T.; Hille, R.; Kuhn, P. Structure 2001, 9, 125-
132.
(11) (a) Boyington, J. C.; Gladyshev, V. N.; Khangulov, S. V.; Stadtman,
T. C.; Sun, P. D. Science 1997, 275, 1305-1308. (b) Jormakka, M.;
Tornroth, S.; Byrne, B.; Iwata, S. Science 2002, 295, 1862-1868.
(12) George, G. N.; Turner, N. A.; Bray, R. C.; Morpeth, F. F.; Boxer, D.
H.; Cramer, S. P. Biochem. J. 1989, 259, 693-700.
(13) (a) Helton, M. E.; Pacheco, A.; McMaster, J.; Enemark, J. H.; Kirk,
M. L. J. Inorg. Biochem. 2000, 80, 227-233. (b) Jones, R. M.; Inscore,
F. E.; Hille, R.; Kirk, M. L. Inorg. Chem. 1999, 38, 4963-4970. (c)
Benson, N.; Farrar, J. A.; McEwan, A. G.; Thomson, A. J. FEBS Lett.
1992, 307, 169-172. (d) Finnegan, M. G.; Hilton, J.; Rajagopalan,
K. V.; Johnson, M. K. Inorg. Chem. 1993, 32, 2616-2617.
(14) Baugh, P. E.; Garner, C. D.; Charnock, J. M.; Collison, D.; Davies,
E. S.; McAlpine, A. S.; Bailey, S.; Lane, I.; Hanson, G. R.; McEwan,
A. G. JBIC, J. Biol. Inorg. Chem. 1997, 2, 634-643.
(17) Raitsimring, A. M.; Astashkin, A. V.; Feng, C.; Enemark, J. H.; Nelson,
K. J.; Rajagopalan, K. V. JBIC, J. Biol. Inorg. Chem. 2003, 8, 95-
104.
(18) Cardona, C. M.; Mendoza, S.; Kaifer, A. E. Chem. Soc. ReV. 2000,
29, 37-42.
(19) Weyermann, P.; Gisselbrecht, J.-P.; Boudon, C.; Diederich, F.; Gross,
M. Angew. Chem., Int. Ed. 1999, 38, 3215-3219.
(20) Gorman, C. B.; Smith, J. C.; Hager, M. W.; Parkhurst, B. L.; Gracz,
H. S.; Haney, C. A. J. Am. Chem. Soc. 1999, 121, 9958-9966.
(21) Mondal, S.; Basu, P. Inorg. Chem. 2001, 40, 192-193.
(15) George, G. N.; Hilton, J.; Temple, C.; Prince, R. C.; Rajagopalan, K.
V. J. Am. Chem. Soc. 1999, 121, 1256-1266.
(16) George, G. N.; Hilton, J.; Rajagopalan, K. V. J. Am. Chem. Soc. 1996,
118, 1113-1117.
7490 Inorganic Chemistry, Vol. 42, No. 23, 2003

Encapsulated Oxo-Mo(V) Centers
obtain the half wave potentials. Mass spectra were collected on a
Micromas ZMD mass spectrometer using both negative and positive
ionization modes with the electrospray ionization (ESI) source.
Acetonitrile solutions of the samples were injected via a syringe
pump with a flow rate of 0.1 mL/min. X-band (∼9 GHz) CW-
EPR spectra were recorded on a Bruker 300 spectrometer equipped
with an Oxford ESR 910 cryostat for low temperature measurement.
Spectra were collected at 20 K with MeCN/toluene (50:50 v/v) as
and [MoO(p-SC6H4CONHCH3)4]- cores. A typical TDDFT cal-
culation on the [MoO(p-SC6H4CONHCH3)4]- core using 8 proces-
sors and 1.5 GB of RAM in HITAC Risc supercomputer took nearly
80 h. The solvation and reorganization effects were calculated by
using the isodensity polarized continuum model (IPCM)30 coupled
with self-consistent reaction field procedure implemented in Gauss-
ian 98 program.31 All calculations employed Becke’s three-
parameter hybrid exchange functional25 and the gradient corrections
of Perdew, along with his 1981 local correlation functional32
(B3P86). We have chosen this functional because our calcula-
tions33,34 as well as published data from the other laboratories35
suggest that the geometries and thermodynamic properties calculated
with the B3P86 functional are in good agreement. For all calcula-
tions, the 6-31G(d) basis set36 on all nonmetal atoms was used,
while the LANL2DZ basis set with LANL2 effective core
potential37 was used for molybdenum. For calculating the solvent
effects,geometryoptimizationswereperformedon[MoO(SCH3)4]1-/2-
using H3CS- to mimic the phenylthiolate ligands, and frequency
calculations were conducted to ensure a minimum for the optimized
geometry. The gas-phase optimized geometries were used for
electronic structure calculation, and solvation calculations were
conducted in respective solvents. The geometry reorganization
energy in the Mo(IV) state was estimated as a difference between
the energies for the molybdenum(IV) state as derived from the Mo-
(V) state and the optimized geometry. In the ligand calculations,
the geometries were optimized using the B3LYP exchange func-
tional with the 6-31G(d) basis set. The Mulliken38 charges of all
atoms of interest were calculated using standard procedures
implemented in the Gaussian family software package; details are
reported elsewhere.39 Last, the percent contribution of atomic
orbitals to their respective molecular orbitals was calculated by using
the VMOdes program.40
Syntheses. Tris[(cyanoethoxy)methyl]aminomethane, [G1-
CN] (1). Tris(hydroxymethyl) aminomethane (10 g, 82.5 mmol)
was added to a mixture of aqueous KOH (500 mg, 9 mmol in 2.5
mL of water) and dioxane (10 mL) at room temperature. Next,
acrylonitrile (17.6 mL, 268 mmol) was added to this mixture, and
the mixture was vigorously stirred for 24 h. Excesses of acrylonitrile
and dioxane were removed by evaporation to yield a yellow residue,
which was dissolved in CH2Cl2 (100 mL). The organic layer was
washed with water three times (total 300 mL) and dried over
anhydrous Na2SO4. A pale yellow liquid was recovered in 80%
yield (18.48 g) by evaporating the CH2Cl2 layer. 1H NMR
(CDCl3): δ ) 1.76 (br, 2H, NH2); 2.63 (t, 6H, J ) 6.0 Hz, CH2);
3.45 (s, 6H, CH2); 3.70 (t, 6H, J ) 5.5 Hz, CH2). 13C NMR
(CDCl3): δ ) 18.87, 65.82, 72.60 (CH2); 56.16 (C(CH2)3); 118.09
1
a glassy solvent. Room temperature H (at 300 MHz) and 13C (at
75.5 MHz) NMR spectra were recorded on a Bruker ACP-300
spectrometer with both CDCl3 and DMSO-d6.
Computational Details. The known crystal structure of (AsPh4)-
[MoO(SPh)4]22 was used for calculating the electronic structure of
[MoO(SPh)4]-. For the electronic structure calculation of [MoO-
(p-SC6H4CONHCH3)4]-, the geometry of the CONHCH3 group was
optimized by the PM3(tm) method,23 and the optimized structure
was added at the para position of the thiophenyl rings of (AsPh4)-
[MoO(SPh)4]. For calculations on [MoO(SPh)4]- and [MoO(p-
SC6H4CONHCH3)4]-cores, C4 symmetry was adopted. In another
set of calculations, the geometries of [MoO(SPh)4]- and [MoO(p-
SC6H4CONHCH3)4]-cores were optimized with the LANL2DZ
basis set at the same level of theory used for single-point
calculations (vide infra). All semiempirical calculations, using the
PM3(tm) method, were carried out using Hyperchem24 software
on a personal computer running on Windows operating system.
The density functional calculations were performed using Becke’s
three-parameter hybrid exchange functional25 and the Lee-Yang-
Parr nonlocal correlation functional26 (B3LYP) in the borders of
unrestricted or restricted-open Hartree-Fock formalism. For DFT
calculations, the molybdenum atom is described with a DGauss
full electron double-ú basis set with polarization having a (18s,-
12p,9d) f [6s,5p,3d] contraction scheme;27 for all other atoms the
standard 6-311G(d) basis set28 was used. All single point and
geometry optimization DFT calculations were performed using the
Gaussian 98 program family29 running on either the Windows or
the Linux operating system. Time-dependent DFT (TDDFT)
calculations were done using unrestricted-open shell calculations
using X-ray crystallography adapted geometries of [MoO(SPh)4]-
(22) Bradbury, J. R.; Mackay, M. F.; Wedd, A. G. Aust. J. Chem. 1978,
31, 2423-2430.
(23) Stewart, J. J. P. J. Comput.-Aided Mol. Des. 1990, 4, 1-105.
(24) HyperChem Pro. 6.03; HyperCube, Inc.: Gainesville, FL, 2001.
(25) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(26) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
(27) Basis sets were obtained from the Extensible Computational Chemistry
Environment Basis Set Database, Version 4/22/01, developed and
distributed by the Molecular Science Computing Facility, Environ-
mental and Molecular Sciences Laboratory which is part of the Pacific
Northwest Laboratory, P.O. Box 999, Richland, WA 99352, and
funded by the U.S. Department of Energy. The Pacific Northwest
Laboratory is a multiprogram laboratory operated by Battelle Memorial
Institute for the U.S. Department of Energy under Contract DE-AC06-
76RLO 1830. Contact David Feller or Karen Schuchardt for further
information.
(30) Foresman, J. B.; Keith, T. A.; Wiberg, K. B.; Snoonian J.; Frisch, M.
J. J. Phys. Chem. 1996, 100, 16098-16104.
(31) Wong, M. W.; Frisch, M. J.; Wiberg, K. B. J. Am. Chem. Soc. 1991,
113, 4776-4782. Wong, M. W.; Wiberg, K. B.; Frisch, M. J. J. Am.
Chem. Soc. 1992, 114, 523-529.
(28) McLean, A. D.; Chandler, G. S. J. Chem. Phys. 1980, 72, 5639-
5648. Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem.
Phys. 1980, 72, 650-654.
(32) Perdew, J. P. Phys. ReV. B 1986, 33, 8822-8824.
(33) Kail, B.; Nemykin, V. N.; Davie, S. R.; Carrano, C. J.; Hammes, B.;
Basu, P. Inorg. Chem. 2002, 41, 1481-1291.
(29) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone,
V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R.
L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.;
Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 98; Gaussian, Inc.: Pittsburgh, PA, 1998.
(34) Nemykin, V. N.; Basu, P. Inorg. Chem. 2003, 42, 4046-4056.
(35) Thoomson, L. M.; Hall, M. B. J. Am. Chem. Soc. 2001, 123, 3995-
4002.
(36) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54,
724-728.
(37) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270-283.
(38) Mulliken, R. S. J. Chem. Phys. 1955, 23, 1833-1840, 1841-1846,
2338-2342, 2343-2346.
(39) Sengar, R. S.; Nemykin, V. N.; Basu, P. New J. Chem. 2003, 27,
1115-1123.
(40) Nemykin, V. N.; Basu, P. VMOdes: Virtual Molecular Orbital
description program for Gaussian, GAMESS, and HyperChem, revision
B 6.2, 2001.
Inorganic Chemistry, Vol. 42, No. 23, 2003 7491

Basu et al.
(CN). ESI+-MS (CH3CN): m/z 281.1 [M + H]+ (M ) C13H20N4O3,
280.2); 303.1 [M + Na]+. IR (NaCl): 3371 cm-1, 3350 cm-1 (νNH);
2916 cm-1, 2877 cm-1 (νCH); 2251 cm-1 (νC≡N).
(NaCl): 3416 cm-1 (νNH); 2916 cm-1, 2887 cm-1 (νNH); 2252 cm-1
(νC≡N), 1733 cm-1 and 1661 cm-1 (νCONH).
Disulfide of [G1-Ester]-amide (4c). This compound was syn-
thesized following the procedure outlined for 4b using 3 (1.0 gm,
3.27 mmol), 2 (3.44 g, 8.17 mmol), and triethylamine (1.7 mL 12.3
mmol) in CH2Cl2 (25 mL). The crude product was purified by
adsorption chromatography on silica gel using a 1:2 mixture of
MeCN and toluene to yield 89% (3.24 g).1H NMR (CDCl3): δ )
1.22 (t, 18H, J ) 7.0 Hz, CH3), 2.53 (t, 12H, J ) 6.3 Hz, CH2),
3.71 (t, 12H, J ) 6.2 Hz, CH2), 3.82 (s, 12H, CH2), 4.10 (qt, 12H,
J ) 7.0 Hz, CH2), 6.55 (s, 2H, NH), 7.50 (d, 4H, J ) 8.5 Hz, arom
CH), 7.73 (d, 4H, J ) 8.5 Hz, arom CH). 13C NMR (CDCl3): δ )
13.47 (CH3); 34.23, 59.55, 66.10, 68.44 (CH2); 50.69 (C(CH2)3);
125.63, 127.29, 133.44, 139.22 (arom C); 165.59 (CONH), 170.64
(CO2Et). FAB+-MS: m/z: 1113.0 [M + H]+ (M ) C52H76N2O20S2,
1112). 1023.0 [M - 2OEt]+ (M ) C52H76 N2O20S2, 1112). ESI+-
MS (CH3CN): m/z: 1113.3 [M + H]+; 1135.2 [M + Na]+. IR
(NaCl): 3418 cm-1 (νNH); 2982 cm-1, 2877 cm-1 (νCH); 1734 cm-1
(νCO); 1666 cm-1 (νCONH).
Thiol of [G1-CN]-Amide (5b). Compound 4b (1.65 g, 1.99
mmol) was dissolved in a mixture of dry EtOH/THF (15 mL/15
mL), and NaBH4 (0.3 g, 7.96 mmol) was added in small portions
at 0 °C under Ar atmosphere. After the complete addition of the
NaBH4, the temperature was raised to room temperature, and the
reaction mixture was allowed to stir for 4 h. After removing the
solvents under reduced pressure, the resulting yellow solid was
dissolved in distilled water (10 mL), and 0.5 N HCl was added to
acidify the solution to pH ∼ 4. A yellowish oily material separated
out from the solution, which was extracted with EtOAc (60 mL)
and dried over anhydrous MgSO4. Evaporation of the EtOAc
solution yielded a yellow liquid, which was purified by adsorption
chromatography on silica gel using a 1:3 mixture of MeCN and
toluene to yield 90% (1.49 g) of final product. 1H NMR (CDCl3):
δ ) 2.62 (t, 6H, J ) 6.1 Hz, CH2); 3.72 (t, 6H, J ) 5.7 Hz, CH2);
3.96 (s, 6H, CH2); 6.46 (s, 1H, NH); 7.23 (d, 2H, J ) 7.9 Hz,
arom CH); 7.65 (d, 2H, J ) 8.2 Hz, arom CH). 13C NMR
(CDCl3): δ ) 18.32, 65.32, 68.66 (CH2); 59.52 (C(CH2)3); 117.72
(CN); 127.33, 127.93, 131.16, 136.03 (arom C); 166.51 (CONH).
ESI+-MS (CH3CN): m/z: 417.2 [M + H]+ (M ) C20H24N4O4S,
416.2); 439.2 [M + Na]+. ESI--MS (CH3CN): m/z: 415.2 [M -
H]-. IR (NaCl): 3417 cm-1 (νNH); 2937 cm-1, 2883 cm-1 (m, νCH);
2559 cm-1 (νSH); 2252 cm-1 (νC≡N); 1734 cm-1 (νCdO); 1659 cm-1
(νCONH).
Tris[((ethoxycarbonyl)ethoxy)methyl]aminomethane, [G1-
Ester] (2). p-Toluenesulfonic acid monohydrate (21.96 g, 0.12 mol)
and absolute ethanol (6.77 mL, 0.12 mol) were added to 1 (10.79
gm, 0.04 mol), and refluxed with stirring for 6 h. During the
reaction, the ammonium salt of p-toluenesulfonic acid formed.
Toluene (10 mL) was added into the reaction for smooth stirring,
and the precipitate was filtered and washed with toluene. After it
was washed, the filtrate was dissolved in CH2Cl2 and washed with
saturated NaHCO3 solution several times. The organic layer was
removed and dried over anhydrous MgSO4 and evaporated to give
a yellow liquid at a 50% (8.42 gm) yield. 1H NMR (CDCl3): δ )
1.27 (t, 9H, J ) 6.6 Hz, CH3); 1.74 (br, 2H, NH2); 2.55 (t, 6H, J
) 6.4 Hz, CH2); 3.32 (s, 6H, CH2); 3.69 (t, 6H, J ) 6.1 Hz, CH2);
4.15 (qt, 6H, J ) 6.8 Hz, CH2). 13C NMR (CDCl3): δ ) 14.06
(CH3); 34.97, 60.22, 66.74, 72.66 (CH2); 55.86 (C(CH2)3); 171.36
(CO2Et). ESI+-MS (CH3CN): m/z 422.2 [M + H]+ (M ) C19H35-
NO9, 421). IR (NaCl): 3379 cm-1(νNH);, 2982 cm-1, 2873 cm-1
(νCH); 1736 cm-1 (νCdO).
4,4′-Dithiobenzoic acid (3). A saturated solution of I2 in 95%
ethanol was slowly added to an ethanol solution (100 mL) of
4-mercaptobenzoic acid (1.5 g, 9.74 mmol). As the reaction
progressed, the brown color of iodine disappeared with a precipita-
tion. The addition of iodine solution was continued until a light
yellow color of the reaction mixture persisted. The off-white
precipitate was collected and washed with ethanol, followed by
drying under reduced pressure at 50 °C to remove any adsorbed
iodine to obtain the final product at a 98% yield (1.47 g). 1H NMR
(DMSO-d6): δ ) 7.63 (d, 4H, J ) 8.2 Hz, arom CH); 7.93 (d, 4H,
J ) 8.5 Hz, arom CH). 13C NMR (DMSO-d6): δ ) 126.12, 129.70,
130.23, 140.74 (arom C); 166.58 (CO2H). ESI--MS (CH3OH): m/z
305.0 [M - H]- (M ) C14H10O4S2, 306). IR (KBr): 3444 cm-1
(νOH); 2981 cm-1 (νCH); 1686 cm-1 (νCO); 759 cm-1 (νCH bend).
Disulfide of G0-Amide (4a). This compound has been synthe-
sized according to the literature procedure.39 Yield 65%. 1H NMR
(DMSO-d6): δ ) 2.74 (d, 6H, J ) 4.5 Hz, CH3); 7.60 (d, 4H, J )
8.5 Hz, arom CH); 7.82 (d, 4H, J ) 8.5 Hz, arom CH); 8.46 (m,
2H, NH). 13C NMR (DMSO-d6): δ ) 24.08, 26.15 (CH3); 126.28,
128.14, 133.54, 138.58 (arom C); 165.72 (CONH). ESI+-MS (CH3-
OH): m/z: 333.0 [M + H]+ (M ) C16H16 N2O2S2, 332). ESI--MS
(CH3OH): m/z 330.9 [M - H]- (M ) C16H16 N2O2S2, 332). IR
(KBr): 3321.5 cm-1 (νNH); 1637 cm-1 (νCO); 1593 cm-1 (νCONH).
Disulfide of [G1-CN]-Amide (4b). The acid chloride of 339 (1.5
g, 4.9 mmol) was reacted with the dendron, 1 (3.5 g, 12.5 mmol),
in CH2Cl2 (20 mL) in the presence of triethylamine (2.62 mL, 18.75
mmol) in CH2Cl2 (15 mL) as described for 4a. After 24 h of stirring,
the reaction was quenched by adding 10 mL of water, and the
organic layer was withdrawn and washed with 15 mL of (1 N)
HCl solution, followed by 15 mL of 10% NaHCO3 solution, and
drying over anhydrous Na2SO4. The organic layer was collected
by filtration and was evaporated to dryness to yield a light yellow
oily compound. Next, the oil was purified by adsorption chroma-
tography on silica gel with a 1:3 mixture of MeCN and toluene.
Yield is 85% (3.45 g).1H NMR (CDCl3): δ ) 2.61 (t, 12H, J )
5.9 Hz, CH2); 3.72 (t, 12H, J ) 6.0 Hz, CH2); 3.95 (s, 12H, CH2);
6.43 (s, 2H, NH); 7.52 (d, 4H, J ) 8.3 Hz, arom CH); 7.71 (d, 4H,
J ) 8.6 Hz, arom CH). 13C NMR (CDCl3): δ ) 18.81, 65.79,
69.15 (CH2); 60.01 (C(CH2)3); 117.99 (CN); 126.67, 127.82, 133.51,
140.50 (arom C); 166.76 (CONH). ESI+-MS (CH3CN): m/z: 831.2
[M + H]+ (M ) C40H46N8O8S2, 830.3); 853.2 [M + Na]+. IR
Thiol of [G1-Ester]-amide (5c). This compound was synthesized
as 5b using 4c (1.0 g, 0.9 mmol) and NaBH4 (0.14 g, 3.6 mmol) in
a mixture of dry EtOH/THF (10 mL/10 mL). The purified
compound was isolated in 96% yield (0.95 mg) as a yellow liquid.
1H NMR (CDCl3): δ ) 1.24 (t, 9H, J ) 7.0 Hz, CH3), 2.55 (t, 6H,
J ) 6.4 Hz, CH2), 3.73 (t, 6H, J ) 6.4 Hz, CH2), 3.82 (s, 6H,
CH2), 4.12 (qt, 6H, J ) 6.8 Hz, CH2), 6.53 (s, 1H, NH), 7.26 (d,
2H, J ) 5.7 Hz, arom CH), 7.64 (d, 2H, J ) 7.8 Hz, arom CH).
13C NMR (CDCl3): δ ) 13.74 (CH3); 34.54, 59.87, 66.34, 68.76
(CH2); 59.61 (C(CH2)3); 127.39, 127.80, 131.70, 135.49 (arom C);
166.19 (CONH); 170.99 (CO2Et). ESI+-MS (CH3CN): m/z: 580.2
[M + Na]+ (M ) C26H39NO10S, 557.2). ESI--MS (CH3CN): m/z:
556.2 [M - H]-. IR (NaCl): 3421 cm-1 (νNH); 2981 cm-1, 2877
cm-1 (m, νCH), 2559 cm-1 (νSH), 1735 cm-1 (νCdO), 1664 cm-1
(νCONH).
(PPh4)[MoO(SPh)4] (6). This compound was synthesized ac-
cording to the published procedure41 by reacting MoOCl3(THF)2
(41) Boyd, I. W.; Dance, I. G.; Murray, K. S.; Wedd, A. G. Aust. J. Chem.
1978, 31, 279-284.
7492 Inorganic Chemistry, Vol. 42, No. 23, 2003

Encapsulated Oxo-Mo(V) Centers
Scheme 1
with thiophenol in a Et3N/MeCN solution. UV-vis spectra [λmax
,
Results
1. Synthesis of Ligands and Metal Complexes. The
nm (ꢀ, M-1 cm-1), MeCN]: 602 (6960). EPR g parameter: (g|,
2.017); (g , 1.979); (gav, 1.992). IR (KBr): 943 cm-1 (νModO). ESI--
MS (CH3CN): m/z 550.7 [M - PPh4]- (M ) C24H20OS4Mo, 550).
(PPh4)[MoO(p-SC6H4CONHCH3)] (7a). A mixture of 6 (40
mg, 0.45 mmol) and 4a (120 mg, 0.36 mmol) was stirred in THF
(12 mL) at room temperature for 48 h, resulting in a dark precipitate.
The precipitate was filtered from the reaction mixture and washed
with 30 mL of 1,2-dimethoxyethane (DME), and cold THF (30
mL). Finally, the residue was recrystallized from a MeCN/DME
mixture. Yield 40% (20 mg). UV-vis [λmax, nm (ꢀ, M-1 cm-1),
MeCN]: 599 (4800), 330 (20906), 265 (75262), 230 (96168). EPR
g parameter: (g|, 2.018); (g , 1.988); (gav, 1.998). IR (KBr): 3342
cm-1 (νNH), 3048 cm-1, 2941 cm-1 (νCH), 1659 cm-1 (νCdO), 1588
cm-1 (νCONH), 943 cm-1 (νModO). ESI--MS (CH3CN): m/z 778 [M
- PPh4]- (M ) C32H32N4O5S4Mo, 778).
synthetic methodology of dendritic units (dendrons) closely
followed those reported in the literature.42 The synthetic
approach is shown in Scheme 1. The polyether dendritic
nitrile (1) was synthesized by reacting tris(hydroxymethyl)
aminomethane (tris) and acrylonitrile in dioxane, which was
then purified by a series of solvent extractions using CH2Cl2
and water. The ester-terminated dendron, 2, was synthesized
by hydrolyzing the nitrile group with p-toluenesulfonic acid
in ethanol and toluene. Although Newkome and co-workers
originally used hydrochloric acid to hydrolyze the nitrile
groups,43 in our hands p-toluenesulfonic acid afforded cleaner
reactions. The reaction was probed by IR spectroscopy
monitoring the disappearance of the nitrile stretching. Both
dendrons were synthesized in spectroscopically pure forms
in very good yields. To synthesize the dendritic ligands, the
amine terminals were linked with the carboxyl groups of
derivatized benzene thiols. The thiol group was first protected
through oxidation to the corresponding disulfide using iodine
as an oxidant. The disulfide (3) precipitated out of the
solution and was collected by filtration and dried under
reduced pressure. In some cases, an excess of adsorbed iodine
complicates subsequent reactions, and required removal by
sublimation.
The carboxyl groups of 3 were reacted with oxalyl chloride
in dry THF and a trace amount of DMF; the solvent was
evaporated to dryness affording the corresponding acid
chloride that was reacted in CH2Cl2 directly with the
dendrons in the presence of triethylamine. The coupling
reactions were conducted overnight and quenched with water.
The amide compounds were extracted with CH2Cl2 and
chromatographed on silica gel. With the exception of 4a,
the target disulfide amide compounds were eluted as yellow
bands using a 1:3 mixture of MeCN-toluene as eluent. All
disulfides were synthesized in very good to excellent yields
(82-98%). The G0-disulfide, (p-SC6H4CONHCH3)2 (4a),
was isolated as a white solid, whereas first generation nitrile
(PPh4)[MoO(p-SC6H4-G1CN)4] (7b). Cold dry degassed THF
(10 mL) was added to a precooled Schlenk flask containing MoCl5
(55 mg, 0.2 mmol) at -78 °C. After the complete addition, the
reaction mixture was slowly warmed to room temperature, and
during this period, the initial brown-red color of the solution
changed to green. After 10 min, a solution containing 5a (500 mg,
1.2 mmol) and triethylamine (0.16 mL, 1.2 mmol) in dry degassed
THF (10 mL) was added over a period of 5 min. Upon addition,
the color of the reaction mixture turned blue and was stirred for an
additional 10 min and filtered. To the filtrate a solution of PPh4Cl
(110 mg, 0.3 mmol) in 10 mL of methanol was added, followed
by evaporation under reduced pressure. The resultant residue was
redissolved in THF and purified by size exclusion chromatography
using BioBeads SX-3 using THF as an eluent. Yield 28% (120
mg). UV-vis [λmax, nm (ꢀ, M-1 cm-1), MeCN]: 601 (5390), 412
(13766), 335 (45894), 267 (133918), 225 (152485). EPR g
parameter: (g|, 2.022); (g , 1.984); (gav, 1.997). IR (NaCl): 3426
cm-1 (m, νNH), 3066 cm-1, 2955 cm-1 (νCH), 2249 cm-1 (νC≡N),
1665 cm-1 (νCdO), 1588 cm-1 (νCONH), 948 cm-1 (νModO). ESI--
MS (CH3CN): m/z: 1774 [M - PPh4]- (M ) C80H92N16O17S4-
Mo, 1774).
(PPh4)[MoO(p-SC6H4-G1Ester)4] (7c). This compound was
synthesized following the procedure described for 7b. Yield 31%.
UV-vis [λmax, nm (ꢀ, M-1 cm-1), MeCN]: 602 (5840), 410 (9259),
336 (33777), 267 (74800), 225 (118518). EPR g parameter: (g|,
2.020); (g , 1.983); (gav, 1.995). IR (NaCl): 3426 cm-1 (m, νNH),
2981 cm-1, 2960 cm-1 (νCH), 1660 cm-1, 1731 cm-1 (νCdO), 1588
cm-1 (νCONH), 944 cm-1 (νModO). ESI--MS (CH3CN): m/z 2339
[M - PPh4]- (M ) C104H152N4O41S4Mo, 2339).
(42) Newkome, G. R.; Lin, X. Macromolecules 1991, 24, 1443-1444.
(43) Newkome, G. R.; Lin, X.; Young, J. K. Synth. Lett. 1992, 53-54.
Inorganic Chemistry, Vol. 42, No. 23, 2003 7493

Basu et al.
Scheme 2
contrast, the thiophenolato complexes of the o-substituted
ligands afford pure materials with only a slight excess of
the ligand.44 One possible reason for such a difference is
that the conformation of the o-substituted ligand geometry
allows for hydrogen bond formation between the NH proton
and the thiolato sulfur, which significantly alters the redox
potential of the ligands. Fortunately, complex 7a can be
separated from the reaction mixture by recrystallization from
MeCN/DME solutions. When the same methodology was
applied for the preparation of 7b and 7c, an incomplete
exchange of the thiophenolato group was observed, presum-
ably due to the steric constraint. Instead of using the
disulfides (4b and 4c), complexes 7b and 7c were synthesized
by using excess (∼6 times) of the thiols (5b and 5c), followed
by repeated cycles of size exclusion chromatography from
the evaporated residues of the reaction mixture. In each case,
the target compound could be eluted with THF as the second
dark blue band, and the evaporation of the organic solvent
resulted in a pure complex. Because this procedure requires
substantial time to purify the target compounds, we exploited
the alternative halogen exchange reaction from MoOCl3-
(THF)2 with the dendritic thiols. Using this strategy, we have
synthezised [MoVO(SAr)4]- cores without the complication
of incomplete thiol substitution reaction. The dendritic
molecules (7b and 7c) were finally purified by size exclusion
chromatography. Although the yields of the complexes were
lower as compared to those of the thiol exchange reactions,
the purity was higher, and the purification procedure was
more convenient. All of the oxo-molybdenum(V) complexes
were synthesized as tetraphenyl phosphonium salts. Com-
plexes 6 and 7a were isolated in solid form whereas
compounds 7b and 7c were isolated as gummy materials.
All molybdenum compounds reported here are soluble in
common organic solvents such as MeCN, DMF, CH2Cl2, and
CH3OH, although the molybdenum complexes of dendritic
ligands are unstable in chlorinated solvents or protic solvents
and are extremely sensitive to air.
2. Spectral Characterization of Molecules. All dendrons
and ligands were characterized by ESI-MS, 1H and 13C NMR,
and IR spectroscopies. For disulfides, the chemical shifts of
the aromatic protons were slightly different from those of
the thiols, serving as good indicators of the purity of the
materials. The 13C NMR chemical shift is known to correlate
with the Hammett constants, and we have used the 13C NMR
chemical shifts to parametrize the Hammett constants for
the dendritic molecules.39 In the positive mode ESI-MS, often
the ions were found to be associated with a sodium ion,
which is a common phenomenon when bulky or biological
molecules are investigated. The IR spectra of the nitrile-
terminated dendrons and ligands exhibit a strong absorption
∼2250 cm-1 that served as a diagnostic feature for the
presence of a -CN group, which was used to probe the
progress of reactions involving -CN group.
terminated [p-SC6H4CONHC(CH2O(CH2)2CN)3]2 (4b) and
ester terminated [p-SC6H4CONHC(CH2O(CH2)2CO2Et)3]2
(4c) disulfides were isolated as yellow liquids. The disulfides
(4b and 4c) were reduced to dendritic thiols [p-HSC6H4-
CONHC(CH2O(CH2)2CN)3] (5b) and [p-HSC6H4CONHC-
(CH2O(CH2)2CO2Et)3] (5c) with sodium borohydride in a
1:1 mixture of ethanol and THF. The corresponding thiols
were isolated as light yellow oils in excellent yields (90-
95%). The thiols were found to be moderately air sensitive,
especially in solution.
In general, three distinct routes have been described for
synthesizing the oxo-molybdenum(V) tetrathiolato com-
plexes. The first method involves a halogen substitution
reaction from either [MoVOCl5]2- or MoOCl3(THF)2 with
thiolato ligands in the presence of a base, usually triethy-
lamine. The second method also involves exchange of
thiophenolato ligands with substituted aryl thiols. This
method has been utilized to synthesize a wide variety of
complexes including alkanethiols. Finally, the tetrathiolate
oxo-molybdenum(V) complexes have been synthesized by
exchanging the thiophenolato groups in [MoVO(SPh)4]- with
aryl disulfides via redox-coupled ligand exchange reaction.44
Such redox-coupled ligand exchange reactions have been
described for synthesizing other metal-sulfur cores such as
iron-sulfur clusters.45 The redox coupled ligand exchange
reactions have been attempted for the synthesis of 7a, 7b,
and 7c. However, only complex 7a was successfully
synthesized by exchanging the thiophenolato groups of [PPh4]-
[MoVO(SPh)4] (6) using disulfide 4a (Scheme 2). The ESI
MS of the crude product suggests a mixture of the general
formula [MoV(SPh)n(p-SC6H4CONHCH3)4-n
-
]
even in the
presence of a large excess (7 times) of the disulfide, 4a. In
All oxo-molybdenum(V) compounds (6, 7a-c) have been
characterized by mass spectrometry, IR spectroscopy, EPR
and electronic spectroscopy. The molecular mass for com-
pounds 6 and 7a-c has been determined by negative ion
ESI-MS from their MeCN solutions (Figure 2). The mass-
(44) Ueyama, N.; Okamura, T.; Nakamura, A. J. Am. Chem. Soc. 1992,
114, 8129-8137.
(45) Que, L. J.; Bobrik, M. A.; Ibers, J. A.; Holm, R. H. J. Am. Chem.
Soc. 1974, 96, 4168-4178.
7494 Inorganic Chemistry, Vol. 42, No. 23, 2003

Encapsulated Oxo-Mo(V) Centers
Figure 2. ESI-MS of 7c and the theoretical distribution of the anion.
to-charge ratio and the isotope distribution pattern agreed
well with the theoretical value for tetrasubstituted dendritic
complexes. Furthermore, mass spectrometry confirmed that
there is no branching defect in the ligand architecture.
Interestingly, both the fast atom bombardment mass spec-
trometry (FABMS) and the matrix assisted laser desorption
ionization (MALDI) mass spectrometry were unsuccessful
in providing molecular ion peaks. Also at higher tempera-
tures, ESI-MS exhibited spectra with polymeric patterns,
which are currently under investigation. In addition to Ct
N, CdO, and NsH vibrations, the ModOt vibration has also
been probed by infrared spectroscopy. Infrared spectra of
gummy samples were measured on a salt plate, while those
for the solid samples were recorded in KBr pellets. The Mod
Ot vibration has been observed at the same frequency as those
of other tetrathiophenolato complexes.46 Practically no varia-
tion was observed in the ModOt stretching frequency (943-
948 cm-1) within the complexes (6, 7a-c) reported here,
suggesting that the ModOt bond strength in these molecules
is very similar. The oxo-molybdenum(V) complexes 6 and
7a-c exhibit an intense low energy absorption band at ∼600
nm (ꢀ ∼ 6000 M-1 cm-1) (Figure 3) with the largest variation
in the band maxima being less than 5 nm. This is a very
intriguing result, which is in contrast to the peripherally
substituted thiophenolato or structurally rigid hydrotris-
(pyrazolyl)borate complexes that show a linear dependence
of the low energy charge transfer transition as a function of
substituents.46,47 In our molecules, the maximum of the broad
charge transfer band, however, is invariant with the Hammett
constant of the ligand. In order to understand the solvochro-
mic nature of this low energy band, we have recorded optical
spectra of [MoVO(SPh)4]- in CH2Cl2, MeCN, propylene
carbonate (PC), DMSO, and DMF. In all cases, the band
maxima appeared at the same position, and the peak width
also remained similar indicating no appreciable solvochromic
effect (Figure S1). It is instructive to point out that this broad
band is a composite of several overlapping transitions
involving the low lying sulfur based orbitals and the
d-manifolds. We have investigated the nature of this transi-
Figure 3. Room temperature electronic spectra of Mo complexes 6 and
7a-c in MeCN solutions.
tion using magnetic circular dichroism (MCD) spectros-
copy;48 general features are in agreement with those reported
for [MoVO(SPh)4]-.49
This low energy transition is a sensitive reporter of
distortions at the metal center. For example, in [MoO(S2-
bph)2] (where S2bph ) bis-thiophenol, a rigid ligand), this
transition shifted by 30 nm.50 It is important to note that the
position of the low-energy transition is not affected by the
architecture of ligands, suggesting the orientation of the S π
orbitals with respect to the Mo dxy orbital is similar. The
orientation not only modulates the energy of the transition
but also the oscillator strengths. A similar orbital control is
now established for other metal centers such as in the Cu-
cysteine in blue copper centers51 and in heme.52 The near
invariant band position as well as similar oscillator strength
strongly argues that the addition of bulkier dendritic ligands
does not alter the orientation of the π orbitals with respect
to the metal dxy orbital. Thus, relative orientation and position
of the sulfurs are almost similar to each other.
The oxo-molybdenum(V) complexes (7a-c) are paramag-
1
netic with an S ) /2 (d1) ground state and are amenable to
EPR spectroscopy (Figure 4). The molybdenum complexes
exhibit axial EPR spectra similar to those observed by
Hanson et al. for [MoVO(SPh)4]-.53 In addition, the charac-
teristic molybdenum hyperfine structures, due to the presence
5
of naturally abundant I )
/ isotopes (25% of total
2
contribution), have also been observed. No significant
variation in the g-values has been observed in these
complexes, indicating that the integrity of the square
(48) McNaughton, R. L.; Nemykin, V. N.; Mondal, S.; Basu, P.; Kirk, M.
L. Manuscript in preparation.
(49) (a) McMaster, J.; Carducci, M. D.; Yang, Y. S.; Solomon, E. I.;
Enemark, J. H. Inorg. Chem. 2001, 40, 687-702. (b) McNaughton,
R. L.; Tipton, A. A.; Rubie, N. D.; Conry, R. R.; Kirk, M. L. Inorg.
Chem. 2000, 39, 5697-5706.
(50) Conry, R. R.; Tipton, A. A. JBIC, J. Biol. Inorg. Chem. 2001, 6, 359-
366.
(51) Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. ReV. 1996, 96,
(46) Ellis, S. R.; Collision, D.; Garner, C. D. J. Chem. Soc., Dalton Trans.
1989, 413-417.
(47) Chang, C. S. J.; Collison, D.; Mabbs, F. E.; Enemark, J. H. Inorg.
Chem. 1990, 29, 2261-2267.
2239-2314.
(52) Walker, F. A. Coord. Chem. ReV. 1999, 185-186, 471-534.
(53) Hanson, G. R.; Wilson, G. L.; Bailey, T. D.; Pilbrow, J. R.; Wedd, A.
G. J. Am. Chem. Soc. 1987, 109, 2609-2616.
Inorganic Chemistry, Vol. 42, No. 23, 2003 7495

Basu et al.
Figure 6. Calculated molecular orbital diagrams for the optimized and
the X-ray derived structure for [MoO(SPh)4]-.
Table 1. Mo(V/IV) Reduction Potentials E1/2, mV (∆Ep, mV), at 25 Ca
solvent
6
7a
7b
7c
MeCN
DMF
PC
-1321 (122)
-1383 (89)
-1324 (114)
-1171 (70)
-1213 (68)
-1084 (69)
-1097 (122)
-1251 (167)
-1052 (138)
-1117 (137)
-1247 (335)
-1099 (269)
Figure 4. X-band CW-EPR spectra of frozen solutions of 7b (thick line
labeled as G1CN) and 7c (thin line labeled as G1-ester) at 20K.
a Potentials vs Fc+/Fc.
The anion displays a well-defined couple due to the reduction
of Mo(V) to Mo(IV), similar to those observed by Bradbury
et al.,54 but no oxidative response could be observed within
the solvent window. In all compounds (6, 7a-c), the
reversibility of this reduction was supported by different scan
rate experiments, which show a linear relationship between
square root of the scan rate and current observed (Figure
S2). In a separate experiment, the influence of the electrolyte
concentration on the reduction potentials of complex 6 was
investigated. It has been found that, in the 0.1-0.4 M
electrolyte concentration range, the reduction potential is
practically constant (the maximum deviation found to be 16
mV). With increasing steric bulk at the ligand, the irrevers-
ibility also increased as demonstrated by the peak-to-peak
separation. For all compounds, the peak-peak separation was
larger in propylene carbonate and in DMF as compared to
acetonitrile.
Figure 5. Cyclic voltammograms of 6 (A), 7a (B), 7b (C), and 7c (D) in
4. Electronic Structure Calculations. The electronic
structures of [MoO(SPh)4]- and [MoO(p-SC6H4CONHCH3)4]-
complexes were calculated both in the restricted and
unrestricted formalism using the DFT method with the
B3LYP EC functional. Calculations were conducted for both
X-ray idealized and optimized geometries in order to
understand the effect of the geometry optimization on the
electronic structure. The results of these calculations are
presented in Figure 6 and in Tables 2, 3, and S1. The average
bond distances and angles of the optimized geometry deviate
from the X-ray crystallographically determined geometry by
0.07 Å and 0.93-1.22°, for the most important distances
and angles, respectively (Table 2). In the optimized struc-
tures, the terminal oxo bonds are found to be longer than
those in the X-ray structures. Such elongation stabilizes the
orbitals aligned along the ModO vector; thus, the energy
MeCN solution (scan rate 100 mV/s).
pyramidal [MoVOS4]- core (C4V local symmetry) is retained.
Thus, again, the EPR data in agreement with the UV-vis
spectra of complexes 6 and 7a-c suggest that the addition
of bulky and symmetric dendritic architecture does not
impose any appreciable distortion at the metal center in the
frozen solution.
3. Electrochemical Behavior of the Molybdenum Com-
plexes. The redox chemistry of the molybdenum complexes
was investigated using cyclic voltammetry in MeCN, DMF,
and propylene carbonate. Figure 5 shows cyclic voltammo-
grams recorded in MeCN, and all potentials are listed in
Table 1. The solvents were chosen to span a wide range of
dielectric constant and stability of the molybdenum com-
plexes. In each case, the half wave potential was obtained
by digitally simulating cyclic voltammograms. An EC
mechanism (electron transfer followed by a chemical trans-
formation) was used to simulate the cyclic voltammograms
(a detailed study of digital simulation will be communicated
elsewhere). The potentials are internally calibrated and
expressed with respect to the ferrocenium/ferrocene couple.
2
2
2
of difference between the Mo dz and Mo dx -y orbitals is
reduced. Calculations based on the X-ray geometry show
2
2
2
that the difference in energy between the dz and the dx -y
(54) Bradbury, J. R.; Masters, A. F.; McDonell, A. C.; Brunette, A. A.;
Bond, A. M.; Wedd, A. G. J. Am. Chem. Soc. 1981, 103, 1959-
1964.
7496 Inorganic Chemistry, Vol. 42, No. 23, 2003

Encapsulated Oxo-Mo(V) Centers
Table 2. Selected Geometric Parameters [MoO(SAr)4]-
optimized geometrya
exptl geometry22
[MoO(SPh)4]-
[MoO(p-SC6H4-
[MoO(SPh)4]-
CONHMe)4]-
ModO
MosS
1.67
2.40
1.72
2.49
1.72
2.49
SsC
1.76
1.84
1.84
OsMosS
MosSsC
OsMosSsC
av deviationb
109.86
109.82
58.81
109.46
109.33
56.04
109.74
108.97
56.99
0.07/1.22
0.07/0.93
Figure 7. Pictorial representation of the HOMO for 6 and 7a.
a Optimized using LANL2DZ basis set for all the atoms. b rmsd for
distances/angles compared to the crystallographic geometry. In the case of
[MoO(p-SC6H4CONHMe)4]-, the crystal structure of [MoO(SPh)4]- was
used.
making it higher in energy than other d orbitals. Strong σ
interactions between the equatorial sulfur and molybdenum
2
2
dx -y orbital lead to a displacement of the molybdenum atom
orbitals is about 0.15 eV for [MoO(SPh)4]-, while it is only
0.03 eV for the geometry-optimized structure. Because the
2
2
from the S4 plane. The dx -y orbital moves away from the
σ
2
2
Mo-S vectors decreasing the S -Mo(dx -y ) interaction,
2
2
2
dz orbital is mixed with several ligand based orbitals in
which results in stabilization of the dx -y orbital. The terminal
oxo group contributes to the degenerate dxz,yz orbitals, i.e.,
the LUMOs for both [MoO(SPh)4]- and [MoO(p-SC6H4-
CONHMe)4]-. The contribution of the Mo dxy to the HOMO
orbital is similar in [MoO(SMe)4]-, [MoO(SPh)4]-, and
[MoO(p-SC6H4CONHMe)4]-; however, the presence of
aromatic rings in the later complexes decreases the contribu-
tion from the sulfur atoms to the HOMO.
[MoO(p-SC6H4CONHMe)4]-, the energy difference between
2
2
2
the dz and the dx -y orbitals could not be quantified. In all
cases, the composition of the HOMO (primarily Mo dxy) is
very similar suggesting that the geometry optimization
primarily affects the virtual orbitals in this system. Thus,
the ground state calculations on X-ray derived geometry
provide a reasonable approach for understanding the proper-
ties of these orbitals.
In recent years, time-dependent density functional theory
(TDDFT) has become a popular tool for calculating the
vertical excitation energies in inorganic and organometallic
complexes. A few examples of calculations on closed-shell
molybdenum-containing systems have been reported in the
literature.55 Recently, a detailed investigation of the influence
of the basis set, exchange-correlation functional, and starting
geometries on the vertical excitation energies in an open-
shell [MoOCl4]- complex has been conducted.34 In the
present case, TDDFT calculations on 6 and 7a have been
carried out. The results are presented in Table 4. In each
case, four ligand-to-metal charge transfer (two of them are
degenerate) and two predominantly d-d transitions (one of
The highest occupied molecular orbitals of [MoO(SPh)4]-
and [MoO(p-SC6H4CONHMe)4]- are presented in Figure 7
while their compositions are listed in Tables 3 and S1. In
both cases, the HOMO orbitals are primarily dxy in character
with contributions from sulfur pseudo-σ orbitals. Four sulfur
π orbitals with e, a, and b symmetries appeared below the
HOMO. The sulfur pseudo-σ orbitals have lower energy as
compared to the sulfur π orbitals. This orbital description is
analoguous to those describing the ground state DFT
calculations on a truncated model, [MoO(SMe)4]-.49 Like
other square-pyramidal oxo-molybdenum complexes, the
2
terminal oxo-group strongly destabilizes the dz orbital
Table 3. Calculated Orbital Composition for 6 and 7a Using Experimental Geometry
ROB3LYPa
UB3LYP
[MoO(SPh)4]-, [MoO(p-SC6H4CONHMe)4]-
[MoO(SPh)4]- b
orbital composition
[MoO(p-SC6H4CONHMe)4]- b
E, eV
Mo
Mo(d)
O
S
E, eV
-3.664; 14.1;
-3.635 13.1
-3.797; 66.9;
-3.002 27.3
-3.367; 8.9;
Mo
Mo(d)
O
S
E, eV
59.5; -4.225; 13.9;
61.0 -4.196 13.0
11.5; -4.372; 66.3;
Mo
Mo(d)
O
S
7e
-3.642;
-4.203
-3.146;
-3.722
-3.277;
-3.844
-2.96;
-3.534
-0.881;
-1.272
1.300;
13.6;
13.5
27.0;
25.9
9.1;
8.7
4.2;
3.8
74.1;
73.6
56.9;
51.8
32.0;
21.1
39.9;
10.9
4.8;
4.7
27.0;
25.9
0.3;
0.3
0.6;
0.6
0.0;
0.0
1.5;
1.5
5.1;
5.0
0.0;
0.0
10.8;
10.0
0.0;
0.0
60.3;
59.2
40.4;
39.3
64.1;
61.9
61.6;
59.8
18.0;
17.2
11.3;
9.3
5.0;
4.6
66.9;
27.3
0.2;
0.3
3.5;
4.2
35.5;
66.2
57.0;
55.8
33.5;
29.0
18.2;
11.3
0.7;
0.5
0.0;
0.0
1.4;
1.5
5.3;
5.0
0.0;
0.0
11.3; 11.7;
10.1
0.0;
0.0
4.3;
2.6
4.8;
4.5
66.3;
26.2
0.2;
0.3
0.7; 58.3;
0.5 59.9
0.0; 10.7;
0.0 39.1
1.4; 60.3;
1.5 62.5
5.1; 59.2;
4.8 60.0
0.0; 43.6;
0.0 18.4
10.7; 9.9;
9.1 8.3
0.0; 13.2;
0.0 10.3
0.9; 3.1;
0.6 2.9
7b
9a
40.2
62.8; -3.93;
64.6 -3.798
-3.582
26.2
8.5;
8.9
-3.228
-3.054;
-2.911
9.3
3.7
4.5
8e
3.9;
3.6
61.0; -3.624; 3.5;
61.7 -3.487 4.2
45.3; -2.741; 34.7;
3.2;
3.9
8bc
9e
74.1;
73.6
56.8;
51.7
32.0;
21.1
15.7;
3.3
-2.156; 35.5;
34.7;
64.5
52.7;
48.9
22.5;
19.3
3.6;
2.8
-0.091
1.225;
1.432
2.048;
2.143
66.2
57.1;
55.9
33.5;
29.0
43.5;
33.4
19.8
-0.674
0.576;
0.75
1.225;
1.297
1.239;
1.252
64.5
52.7;
49.0
22.5;
19.3
11.5;
10.1
0.640
2.081;
1.247
2.227;
10.7
66.5;
17.4
12.2;
9.1
9b
10a
20.8;
11.7
11.1;
3.0
3.6;
0.8
2.199;
2.269
1.243
a Top line for [MoO(SPh)4]- and the bottom line for [MoO(p-SC6H4CONHMe)4]-. b The R and the â orbitals are presented sequentially. c SOMO or R
HOMO orbital.
Inorganic Chemistry, Vol. 42, No. 23, 2003 7497

Basu et al.
Scheme 3
Table 4. Vertical Transition Energies for [MoO(SPh)4]- and
[MoO(p-SC6H4CONHCH3)4]- Calculated by TDDFT Approach
band
transition description
[MoO(SPh)4]-
polarization E, cm-1
f
1
2
3
4
5
6
7
137(b)â f 140(b)â (100%)
138(e)â f 140(b)â (92%)
139(e)â f 140(b)â (92%)
140(b)R f 142(e)R (76%)
140(b)R f 141(e)R (76%)
136(a)â f 140(b)â (96%)
134(e)â f 140(b)â (17%);
135(e)â f 140(b)â (81%)
134(e)â f 140(b)â (81%);
135(e)â f 140(b)â (17%)
140(b)R f 143(b)R (44%);
138(e)â f 141(e)â (10%);
139(e)â f 142(e)â (10%)
z
x,y
x,y
x,y
x,y
forbidden
x,y
14948
15198
15198
16667
16667
17825
19685
0.0000
0.0370
0.0370
0.0288
0.0288
0.0000
0.0021
Figure 8. Molecular orbital diagrams for 6 (left) and 7a (right).
8
9
x,y
19685
23809
0.0021
0.0002
with a typical deviation ∼50-100 cm-1. Thus, the stabiliza-
tion of the occupied orbitals in 7a does not change the
interorbital energy difference that results in very similar
vertical excitation energies. Indeed, the calculated relative
orbital energies for complexes 6 and 7a at unrestricted open
shell formalism are very close to each other (Figure 8).
5. The Solvent Effect. In order to understand the nature
of the solvent effect observed in the reduction potentials of
complexes 6 and 7a-c, we have calculated the energy of
the molecule under a variety of conditions (Scheme 3). The
calculations for Scheme 3 were performed on reduced-size
models, [MoO(SMe)4]n- (n ) 1, 2). Each reduction processes
of Scheme 3 in specific solvent can be presented as a sum
of two individual steps. The first step is the addition of one
electron to the molybdenum(V) center keeping the same
geometry. In the second step, the molybdenum(IV) complex
is allowed to settle into a lower energy conformation. The
calculations show that the primary difference in the two Mo-
(IV) geometries is in the O-Mo-S-C torsion angle (∼56°
for MoV and ∼87° for MoIV). Scheme 3 fits well with the
results obtained from the digital simulations of cyclic
voltammograms using an EC mechanism, where the first step
is the Mo(V) to Mo(IV) electron transfer step and the second
chemical transformation step can be referred to the reorga-
nization step in the Mo(IV) state. The energy of the molecule
associated with both the steps was calculated in acetonitrile
as well as propylene carbonate. The geometric rearrangement
is under investigation and will be discussed elsewhere.57 The
solvent effect for the individual steps was calculated as the
difference between the energy of the same compound in
different solvents, while the solvent effect for the complete
reduction process was taken as the addition of the individual
components and then taking a difference. The calculated
solvent effect ∼ -66 mV for the model system is comparable
z
[MoO(p-SC6H4CONHCH3)4]-
197(b)â f 200(b)â (100%)
1
2
3
4
5
6
7
z
x,y
x,y
x,y
x,y
forbidden
x,y
15049
15191
15191
16605
16605
17790
19608
0.0000
0.0455
0.0455
0.0404
0.0404
0.0000
0.0032
198(e)â f 200(b)â (91%)
199(e)â f 200(b)â (92%)
200(b)R f 201(e)R (81%)
200(b)R f 202(e)R (81%)
196(a)â f 200(b)â (96%)
194(e)â f 200(b)â (11%);
195(e)â f 200(b)â (86%)
194(e)â f 200(b)â (86%);
195(e)â f 200(b)â (11%)
200(b)R f 203(b)R (48%);
198(e)â f 201(e)â (9%);
199(e)â f 202(e)â (9%)
8
9
x,y
19608
23402
0.0032
0.0005
z
which is degenerate) have been calculated in the low energy
region. All calculated band positions are in the good
agreement with the experimental assignments given for
[MoO(SPh)4]- and related complexes on the basis of UV-
vis and MCD spectroscopies and ground state DFT calcula-
tions.49,50,48 The calculated oscillator strengths are also in
good agreement with the experimental data except the first
formal d-d transition. Here, because of the spin-polarization
for the R orbital set,56 the calculated intensity is probably
overestimated. More importantly, despite the difference
between electronic structure of the para-substituents in these
two compounds (-H versus -CONHCH3), the calculated
vertical excitation energies are very similar to each other
(55) Broclawik, E.; Borowski, T. Chem. Phys. Lett. 2001, 339, 433-437.
Adamo, C.; Barone, V. Theor. Chem. Acc. 2000, 105, 169-172.
Wakamatsu, K.; Nishimoto, K.; Shibahara, T. Inorg. Chem. Commun.
2000, 3, 677-679. Rosa, A.; Baerends, E. J.; van Gisbergen, S. J. A.;
van Lenthe, E.; Groeneveld, J. A.; Snijders, J. G. J. Am. Chem. Soc.
1999, 121, 10356-10365.
(56) Li, J.; Noodleman, L.; Case, D. A. In Inorganic Electronic Structure
and Spectroscopy; Solomon, E. I., Lever, A. B. P., Eds.; Wiley: New
York, NY, 1999; pp 661-724.
(57) Sengar, R. S.; Basu, P. Unpublished results.
7498 Inorganic Chemistry, Vol. 42, No. 23, 2003

Encapsulated Oxo-Mo(V) Centers
to the experimentally observed values for complexes 6 and
7a-c (from -18 to -87 mV). The calculated energy for
the reduction step has been found to the dominant factor in
Scheme 3 (∼76 kcal/mol), with a significant contribution
from the reorganization energy (∼23 kcal/mol). Interestingly,
models for molybdenum(V) and molybdenum(IV) states are
more stable in propylene carbonate (dielectric constant 64.4),
with the stability of the Mo(IV) state is three times higher.
molecules. Interestingly, the solid-state structures of [MoO-
(SAr)4]2- complexes differ from the [MoO(SAr)4]- structures
with opposite phenyl rings, changing conformation away
from the propeller type arrangements, thereby lowering gross
symmetry of the molecule.22
The electronic contribution was further evaluated by
electronic structure calculations at the DFT level. While the
electronic structures of 6 and 7a can be directly compared,
it is inferred for 7b and 7c from the calculations on respective
ligands. The calculated energy difference between the SOMO
(in the case of restricted open formalism) or â-LUMO (in
the case of unrestricted formalism) orbitals of complexes 6
and 7a is at least 0.391 eV, indicating that the electron-
withdrawing -CONHCH3 group stabilizes the occupied
orbitals in 7a compared to 6 (Table 3). The DFT calculations
on dendritic molecules are computationally expensive and
have not been attempted, the electronic term could not be
directly compared for complexes 7b and 7c as in the case
of 6 and 7a. However, the following two observations can
provide information about the energy of this orbital. First,
the computed Mulliken charges on sulfur for the dendritic
ligands are similar (-0.269, -0.295, and -0.290 au for
p-HSC6H4CONHMe, 5b, and 5c, respectively) and differ
significantly from that calculated for HSPh (-0.18 au).39 The
Mulliken charge on sulfur atoms has been shown to vary
linearly with the reduction potentials in oxo-molybdenum-
(V) ene-dithiolate complexes.60 However, in the present case
the charges on the sulfur atoms do not correlate as well with
the experimental reduction potentials of 6 and 7a-c, and if
the relation derived for the ene-dithiolate is used, a ∼4 mV
difference in the reduction potential between 7a and 7b is
predicted. Furthermore, the MoV/MoIV reduction potentials
of substituted thiophenolato complexes of general formula
[MoO(SAr)4]- exhibit a modest correlation with the Hammett
Discussion
1. Thermodynamic Consideration. A fundamental goal
of the present investigation is to delineate the factors that
contribute to the reduction potential of oxo-molybdenum
centers. Although this is a complex issue, understanding the
factors that influence the redox potentials can provide insight
into enzymatic function. It has been reported that the
reduction potential (E0) of an active site in a metalloenzyme
can be described with the help of eq 1, where 4.5 corrects
the potential to that of NHE, IE is the ionization energy,
and U is a solvation term.51
E0 ) IE + U - 4.5
(1)
The ionization energy term is dependent on the electronic
structure, i.e., the energy of the redox orbital. Similarly, the
energy of the redox orbital is dependent on the effective
nuclear charge, and the geometry of the active site. The
solvation term is composed of factors such as specific
interaction with charged residues, the H-bonding that can
preferentially stabilize either the reduced or oxidized state
of the metal, and the general control of the dielectric relative
to the aqueous solution.58
One possible way to change the energy of the redox orbital
is by imposing distortion at the metal center. It has been
demonstrated that a sterically demanding ligand, e.g., bis-
(thiophenol) imposes a distortion at the oxomolybdenum-
(V) center resulting in the change of the energy of the redox
orbital and hence the reduction potential.50 Even for relatively
rigid iron-porphyrinate molecules, 2-Me-imidazoles as axial
ligands rather than the N-Me-imidazole change the ground
state electronic structure. This change is manifested in their
EPR spectra.59 In the present system, if the ligand bulk can
effectively reduce the local symmetry from C4, the reduced
symmetry should be manifested in their EPR spectra. The
axial EPR spectra, with similar g-values, clearly demonstrate
that the local symmetry is unperturbed even with bulkier
ligands. As mentioned, optical spectroscopy is a sensitive
probe for evaluating subtle distortion at the metal center such
as those due to a difference in the torsion angle. In the present
case, the low energy band offers discerning features of
distortion that can be evaluated in MCD spectra. A detailed
magnetic circular dichroism investigation, reported else-
where48 that shows little change in the band positions, also
supports this conclusion; i.e., in the +5 oxidation state the
geometry of the metal center is very similar in these
+
(σp ) constants,46 and inclusion of the dendrimer data does
not improve the correlation significantly. Taken together, it
is apparent that while there is a significant change in the
electronic structure (imposed by the ligand electronic envi-
ronment) from 6 to 7a, such a change cannot account
completely for the variation in the redox potential from 7a
to 7c.
All compounds are easier to reduce in propylene carbonate,
whose dielectric constant is considerably larger (64.4) than
that of MeCN (37). We believe that this stabilization is purely
electrostatic in origin, with the larger dielectric constant in
propylene carbonate preferentially stabilizing the Mo(IV)
state. Interestingly, complex 7a is easier to reduce than
complex 7c. This behavior suggests a destabilization due to
the dendritic architecture in 7c. Such a trend was also
observed in other redox-active dendritic systems. In these
molecules, the ModO unit provides a direction of the dipole,
and organization of the solvent molecules with respect to
this dipole can influence the reduction potential. The IPCM
calculations on a simplified model suggest that the energy
for the reduced form is ∼76 kcal/mol more stable in MeCN
than the corresponding Mo(V) complex. This stabilization
(58) Kassner, R. J.; Yang, W. J. Am. Chem. Soc. 1977, 99, 4351-4355.
(59) Basu, P.; Raitsimring, A. M.; Enemark, J. H.; Walker, F. A. Inorg.
Chem. 1997, 36, 1088-1094.
(60) Helton, M. E.; Gruhn, N. E.; McNaughton, R. L.; Kirk, M. L. Inorg.
Chem. 2000, 39, 2273-2278.
Inorganic Chemistry, Vol. 42, No. 23, 2003 7499

Basu et al.
is more in propylene carbonate that has a higher dielectric
constant. Along with the stabilization of the Mo(IV) state,
the energy associated with the relaxation in the O-Mo-
S-C torsion angle is also substantial, although less than the
reduction step. This geometric rearrangement has an impor-
tant kinetic consideration that is discussed later.
In DMF, the situation is quite different. Here, the dendritic
effect upon the reduction potential of complexes 7a-c is
small (Table 1). This is surprising, because it is expected
that an increase in the dendrimer size should more effectively
encapsulate the redox-active molybdenum atom from the
solvent molecules. The observed behavior can be explained
by taking into consideration the possibility of the hydrogen
bond formation between amide proton of the 7a-c com-
plexes and the DMF oxygen atom.61 Indeed, among the
solvents tested, DMF is more effective in forming hydrogen
bonds and thus providing the solvent specific interaction.
This interaction leads to a smaller separation in the reduction
potentials for compounds 7a-c and contributes to the shift
of the reduction potentials of 7a-c by ∼100 mV than in
MeCN. Interestingly, the reduction potential of 6 is shifted
only 60 mV in DMF. The larger shift in complexes 7a-c is
clearly indicative of solvent specific interactions. Hydrogen
bonding would increase the partial negative charge on the
nitrogen atom, effectively enhancing the electron donating
properties and making the reduction more difficult. Then,
why do we observe a 50 mV shift in reduction potential in
DMF (relative to MeCN) for compound 6, when both DMF
and MeCN have similar dielectric constants? We believe that
the explanation lies in the donor numbers for these solvents.
Among the three solvents, DMF has the highest donor
number, which should make ionic complexes less ion paired.
The oxo-molybdenum complexes, isolated as salts of PPh4,
are more “ion paired” in MeCN and propylene carbonate,
while they are more ionized in DMF. Thus, the shift in the
reduction potential for 6 from MeCN to DMF may reflect
the difficulty in reducing a monoanionic species to a
dianionic species.
2. Kinetic Consideration. Molecules 7a-c display an
increasingly larger peak-to-peak separation potential differ-
ence between the current maxima of the reduction and return
oxidation waves (∆E), indicative of increasing kinetic
difficulty of reduction and oxidation processes. The elec-
trochemical data suggest that increasing the steric bulk of
the ligand far from the metal center induces sluggish redox
chemistry, a behavior exhibited by other encapsulated redox-
active metal centers.18 Certainly, such an insulating effect
can be operative in this system. In addition, it is possible
that the reorganization energy, detailed in Marcus theory,
can also contribute to the observed peak-to-peak separation.
This would mean that significant differences exist within
these molecules in the Mo(IV) state. We have already
mentioned that upon reduction the reduced species relaxes
its geometry by increasing the O-Mo-S-C torsion angle.
It has been demonstrated that the Mo(IV) complexes of
o-substituted thiolphenolato ligands are in an equilibrium
Figure 9. Reduction potential of selected molybdenum enzymes at pH ∼
7: circles (b) represent the Mo(VI/V) couple, and triangles (2) represent
Mo(V/IV) couple.
between conformers of different symmetry.44 At lower
temperatures, all four amide groups are placed syn to the
ModO vector, while at higher temperatures two of the amide
groups point away from the ModO group lowering the
symmetry. Thus, at high temperature, a low symmetry
conformer is preferred while at lower temperatures, a high
symmetry conformation is favored. Such a stereochemical
rearrangement could also be operating in the dendritic
system. Because stereochemical rearrangements in dendritic
ligands would require more energy, it can attenuate the rate
of electron transfer.
3. Relation to Enzymes. The reduction potentials are
exquisitely sensitive to applied conditions such as temper-
ature, pH, and source. Figure 9 shows the reduction potentials
of Mo centers of mononuclear molybdoenzymes at pH 7.
Note that the Mo(VI/V) and Mo(V/IV) potentials of E. coli
NarG are 220 and 180 mV, respectively, but for DMSOR,
isolated from the same organism, they are -75 and -90 mV.
These two enzymes have the same Mo centers, yet the redox
potentials are ∼300 mV apart. Another interesting feature
is that, with the exception of chicken liver sulfite oxidase,
the Mo(VI/V) and Mo(V/IV) couples are separated by only
15-20 mV.62 In contrast, the redox potentials of discrete
inorganic complexes are often ∼1500 mV apart.63
In all of crystal structures, the Mo centers have been found
to be encapsulated inside the protein matrix (typically ∼10-
15 Å from the surface) and are not exposed to the solvent
(Figure 10). In general, there is a channel associated with
the Mo center for substrate entrance and product release.
This feature is in contrast to other electron transfer proteins
such as cytochrome b,64 plastocyanin,65 or ferredoxin66 where
either the metal ion or a part of the coordinating ligand frame
is exposed to the exterior surface. Furthermore, the crystal
structure of the substrate bound form of DMSOR demon-
strates that the molybdenum is not exposed to the catalytic
pocket. This is interesting because for other protein-
encapsulated metal centers, such as the Zn in carbonic
anhydrase, the metal is exposed to the catalytic pocket. We
hypothesize that the microenvironment created by the protein
backbone can modulate the reduction potential.
(62) Spence, J. T.; Kipke, C. A.; Enemark, J. H.; Sunde, R. A. Inorg. Chem.
1991, 30, 3011-3015.
(63) Graff, J. N.; McElhaney, A. E.; Basu, P.; Gruhn, N. E.; Chang, C.-S.
J.; Enemark, J. H. Inorg. Chem. 2002, 41, 2642-2647.
(64) Rodriguez-Maranon, M. J.; Qiu, F.; Stark, R. E.; White, S. P.; Zhang,
X.; Foundling, S. I.; Rodriguez, V.; Schilling, C. L.; Bunce, R. A.;
Rivera, M. Biochemistry 1996, 35, 16378-16390.
(65) Yafeng, X.; Mats, O.; Orjan, H.; Simon, Y. Protein Sci. 1998, 7, 2099-
2105.
(66) Tsukihara, T.; Fukuyama, K.; Mizushima, M.; Harioka, T.; Kusunoki,
M.; Katsube, Y.; Hase, T.; Matsubara, H. J. Mol. Biol. 1990, 216,
399-410.
(61) Stone, D. L.; Smith, D. K.; McGrail, P. T. J. Am. Chem. Soc. 2002,
124, 856-864.
7500 Inorganic Chemistry, Vol. 42, No. 23, 2003

Encapsulated Oxo-Mo(V) Centers
growth (i.e., the degree of encapsulation).68 These studies
clearly indicate that encapsulation has a significant effect
on the redox behavior of iron porphyrins and 4Fe4S clusters.
Similar conclusions were also drawn from studies of
encapsulated abiological cores such as ferrocene.18 Thus, it
is provocative to suggest that protein structure imposed
microenvironments at the metal center may be a general
feature in many metalloenzymes, and dielectric control of
such center may be a general mode of operation in nature.
Concluding Remarks
In conclusion, we have discussed the synthetic route to a
novel architecture for investigating the encapsulated oxo-
molybdenum(V) centers. The effect of the encapsulation on
the redox properties of the [MoVOS4]- core without imposing
significant distortion is demonstrated. The electronic structure
of oxo-Mo(V) centers, calculated by a DFT method, shows
a very similar composition of the redox orbital. Furthermore,
the redox chemistry of the molybdenum complexes was
investigated in three different solvents with varying dielectric
constants and donor numbers. In all solvents, the dendritic
Mo complexes show sluggish electrochemistry. The NH
hydrogen appears to support hydrogen bond formation in
DMF, while the reduced state is better stabilized in propylene
carbonate. Thus, the influence of solvent on redox potential
without significantly altering the geometric contribution to
electronic structure is demonstrated.
Figure 10. The encapsulated Mo-center derived from the crystal structure
of DMSO reductases (reference 7).
In this regard, dendritic architecture has been thought to
generate microenvironments. It has been argued that branched
dendritic shells can generate microenvironments that can
modify the properties of the functional cores. While such
microenvironments can be used for generating new functional
materials with modified cores, this emulates the microenvi-
ronments created by protein superstructures. Although only
the first generation dendritic ligands were used here, the
mononuclear metal core has four branching points originating
from four thiophenolato ligands, which when compared to
the bulk, creates a different environment at the metal center.
As mentioned, in all mononuclear molybdoenzymes, the
molybdenum center is deeply buried, and intense hydrogen
bonding networks create microenvironments differing from
that of the bulk. The present molecules provide the first
attempt at topological modeling of pterin-containing mono-
nuclear molybdoenzymatic centers.
Acknowledgment. We are grateful to Professor Michael
Hendrich for recording the EPR spectra, and greatly ap-
preciate Dr. R. Belosludov’s assistance in TDDFT computa-
tion. We also thank Professors F.A. Schultz, M. Rivera, J.
Madura, and M. L. Kirk for numerous helpful discussions,
and Dr. Sujit Mondal for preliminary experiments. Financial
support from Research Corporation and the National Insti-
tutes of Health (GM 6155501) is gratefully acknowledged.
Supporting Information Available: UV-vis spectra of 6 in
different solvents (Figures S1) and the linear relations between the
square root of scan rate and peak current (Figure S2). The orbital
composition of 6 and 7a calculated from optimized geometries are
listed in Table S1. Optimized coordinates, energies, and frequencies
for 6, 7a, and [MoO(SMe)4]n- (n ) 1,2). This material is available
free of charge via the Internet at http://pubs.acs.org.
The dielectric control of the redox potential in heme
proteins and ferredoxins has been discussed using synthetic
molecules. As a model for globular heme proteins, the
Diederich group has reported that the redox potential of a
second generation dendritic iron porphyrinate molecule is
420 mV more positive than the corresponding first generation
complex in CH2Cl2, whereas in water the shift is only 12
mV.67 This study suggests that the solvent dielectric can play
an important role in modulating the redox potential through
its effects on the ligand backbone. Gorman et al. have
observed that the reduction of the encapsulated 4Fe4S
clusters become progressively more difficult, both thermo-
dynamically and kinetically, as a function of dendrimer
IC034821R
(67) Weyermann, P.; Gisselbrecht, J.-P.; Boudon, C.; Diederich, F.; Gross,
M. Angew. Chem., Int. Ed. 1999, 38, 3215-3219. Dandliker, P. J.;
Diederich, F.; Gisselbrecht, J.-P.; Louati, A.; Gross, M. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 2725-2728. Dandliker, P. J.; Diederich, F.;
Zingg, A.; Gisselbrecht, J.-P.; Gross, M.; Louati, A.; Sanford, E. HelV.
Chim. Acta 1997, 80, 1773-1801.
(68) Gorman, C. B.; Parkhurst, B. L.; Su, W. Y.; Chen, K.-Y. J. Am. Chem.
Soc. 1997, 119, 1141-1142.
Inorganic Chemistry, Vol. 42, No. 23, 2003 7501