Synthesis, structure elucidation, and redox properties of 99Tc complexes of lacunary Wells-Dawson polyoxometalates: Insights into molecular 99Tc-metal oxide interactions

Article abstract of DOI:10.1021/ic102111t

The isotope ⁹⁹Tc (β_max, 293.7; half-life, 2.1 × 10⁵ years) is an abundant product of uranium-235 fission in nuclear reactors and is present throughout the radioactive waste stored in underground tanks at the Hanford and Savannah River sites. Understanding and controlling the extensive redox chemistry of ⁹⁹Tc is important in identifying tunable strategies to separate ⁹⁹Tc from spent fuel and from waste tanks and, once separated, to identify and develop an appropriately stable waste form for ⁹⁹Tc. Polyoxometalates (POMs), nanometer-sized models for metal oxide solid-state materials, are used in this study to provide a molecular level understanding of the speciation and redox chemistry of incorporated ⁹⁹Tc. In this study, ⁹⁹Tc complexes of the (α₂-P₂W₁₇O₆₁)^10- and (α₁-P₂W₁₇O₆₁) ^10- isomers were prepared. Ethylene glycol was used as a "transfer ligand" to minimize the formation of TcO₂? xH₂O. The solution structures, formulations, and purity of Tc ^VO(α₁/α₂-P₂W ₁₇O₆₁)^7- were determined by multinuclear NMR. X-ray absorption spectroscopy of the complexes is in agreement with the formulation and structures determined from ³¹P and ¹⁸³W NMR. Preliminary electrochemistry results are consistent with the EXAFS results, showing a facile reduction of the Tc^VO(α₁-P ₂W₁₇O₆₁)^7- species compared to the Tc^VO(α₂-P₂W₁₇O ₆₁)^7- analog. The α₁ defect is unique in that a basic oxygen atom is positioned toward the α₁ site, and the Tc^VO center appears to form a dative metal-metal bond with a framework W site. These attributes may lead to the assistance of protonation events that facilitate reduction. Electrochemistry comparison shows that the Re^V analogs are about 200 mV more difficult to reduce in accordance with periodic trends.

Full text of DOI:10.1021/ic102111t

1670 Inorg. Chem. 2011, 50, 1670–1681
DOI: 10.1021/ic102111t
Synthesis, Structure Elucidation, and Redox Properties of 99Tc Complexes of
Lacunary Wells-Dawson Polyoxometalates: Insights into Molecular 99Tc-Metal
Oxide Interactions
Donna McGregor,†,‡,# Benjamin P. Burton-Pye,†,# Robertha C. Howell,† Israel M. Mbomekalle,†,^
Wayne W. Lukens Jr.,§ Fang Bian,†,‡ Edward Mausolf,|| Frederic Poineau,|| Kenneth R. Czerwinski,|| and
Lynn C. Francesconi*,†,‡
†Hunter College of the City University of New York, 695 Park Avenue, New York, New York 10065,
United States, ‡Graduate Center of the City University of New York, New York, New York 10016,
United States, §Chemical Sciences Division, The Glenn T. Seaborg Center, E.O. Lawrence Berkeley National
Laboratory (LBNL), One Cyclotron Road, Berkeley, California 94720, United States, and ||Department of
Chemistry, University of Nevada Las Vegas, Las Vegas, Nevada 89154, United States . ^ Present address:
ꢀ
Institut Lavoisier, UMR8180, Universite de Versailles St. Quentin, 45 Avenue des Etats-Unis,
78035 Versailles Cedex, France. #These authors contributed equally to this work.
Received October 19, 2010
The isotope 99Tc (βmax, 293.7; half-life, 2.1 ꢀ 105 years) is an abundant product of uranium-235 fission in nuclear reactors
and is present throughout the radioactive waste stored in underground tanks at the Hanford and Savannah River sites.
Understanding and controlling the extensive redox chemistry of 99Tc is important in identifying tunable strategies to
separate 99Tc from spent fuel and from waste tanks and, once separated, to identify and develop an appropriately stable
waste form for 99Tc. Polyoxometalates (POMs), nanometer-sized models for metal oxide solid-state materials, are used in
this study to provide a molecular level understanding of the speciation and redox chemistry of incorporated 99Tc. In this
study, 99Tc complexes of the (R2-P2W17O61)10- and (R1-P2W17O61)10- isomers were prepared. Ethylene glycol was
used as a “transfer ligand” to minimize the formation of TcO2 xH2O. The solution structures, formulations, and purity of
3
TcVO(R1/R2-P2W17O61)7- were determined by multinuclear NMR. X-ray absorption spectroscopy of the complexes is in
agreement with the formulation and structures determined from 31P and 183W NMR. Preliminary electrochemistry results
are consistent with the EXAFS results, showing a facile reduction of the TcVO(R1-P2W17O61)7- species compared to the
TcVO(R2-P2W17O61)7- analog. The R1 defect is unique in that a basic oxygen atom is positioned toward the R1 site, and
the TcVO center appears to form a dative metal-metal bond with a framework W site. These attributes may lead to the
assistance of protonation events that facilitate reduction. Electrochemistry comparison shows that the ReV analogs are
about 200 mV more difficult to reduce in accordance with periodic trends.
Introduction
long half-life of 99Tc means that its waste form must be
resistant to degradation. Understanding and controlling the
extensive redox chemistry of Tc is important in identifying
strategies to separate it from the tank waste and spent nuclear
fuel and is important in the identification and development of
an appropriately stable waste form for 99Tc once separated
from spent nuclear fuel and from waste tanks.
The isotope 99Tc (βmax, 293.7 keV; half-life, 2.1 ꢀ 105 years)
is of interest and concern for two reasons: (1) 99Tc is a major
product of uranium-235 fission in nuclear reactors (∼6%
thermal neutron fission yield) and (2) large amounts of 99Tc,
formed during early plutonium production activities, are
present in the radioactive waste stored in underground tanks
at Hanford and Savannah River.1,2
The interplay between the specific oxidation state of 99Tc
and the “ligands” present in the tank waste or the coordina-
tion environment in a specific storage material determines the
speciation and stability of the Tc within those materials. One
crucial gap in the knowledge that needs to be filled to address
both the separation from tank waste and design of storage
matrices is a molecular level understanding of the factors that
control the oxidation state of 99Tc in metal-oxide matrices.
The physical properties of 99Tc and its complex redox
activity pose a problem for both closing the nuclear fuel cycle
and remediation of waste tanks. In a storage repository, the
*To whom correspondence should be addressed. E-mail: lfrances@
hunter.cuny.edu.
(1) Beals, D. M.; Hayes, D. W. Sci. Total Environ. 1995, 173, 101.
(2) Hunt, A. G.; Skinner, T. E. Hydrogeol. J. 2010, 18, 381.
r
pubs.acs.org/IC
Published on Web 01/26/2011
2011 American Chemical Society

Article
Inorganic Chemistry, Vol. 50, No. 5, 2011 1671
Figure 1. Theplenary Wells-Dawson ion and lacunary Wells-Dawson isomers. Top: polyhedral representation. Bottom: ball and stick representation. (A)
6-
(R-P2W18O62
Wells-Dawson structure gives the “lacunary” (R1-P2W17O61
P2W17O61
10- isomer. The terminal oxygen atoms in the defects are shown in cyan. The defect structures possess distinct and different electronic and steric
features that impact chemical and redox speciation of 99Tc.
)
is the “plenary” or parent Wells-Dawson ion, ca. 1.2 nm in length and 0.6 nm in width. (B) Removal of a “belt” W from the plenary
10-
)
isomer. (C) Removal of a “cap” W from the plenary Wells-Dawson gives the (R2-
)
In nuclear waste and in aerobic environments, the most stable
Tc oxidation state is Tc(VII), whichis highly environmentally
mobile and generally not appropriate for disposal in a waste
repository. In most existing and proposed waste forms, the
desired oxidation state is Tc(IV) since Tc(IV) is not envi-
ronmentally mobile. Also, the ionic radius of Tc(IV) is
similar to that of Ti(IV) and Fe(III), which may allow Tc(IV)
to be incorporated into the lattices of titanium or iron
oxides.3-6
Polyoxometalates (POMs) are soluble nanometer-sized
models for metal oxide solid-state materials.7-9 We postulate
that POMs, polyanionic aggregates of early transition metals
(MoVI and WVI), may provide an understanding of the
coordination chemistry, oxidation state, and speciation of
99Tc incorporated into metal oxides. The MO6 octahedron is
the principal building block of POMs; the transition metals
are generally in their d0 oxidation states held together by
metal-oxygen bonds. The octahedra are condensed to form
aggregates primarily by sharing oxygen atoms in a μ-oxo
linkage at corners and in a di-μ-oxo linkage at edges of the
polyhedra. Oxygen atoms of the “plenary” structures of
POMs, such as the Wells-Dawson anion, (R-P2W18O62)6-
(Figure 1), are close packed, simulating solid oxide
structures.10 Polyoxometalates possess alternating bond or-
der differences, resulting in rings exhibiting alternating
short-long O-M-O bonds.10 Binding of counterions and
protonation dynamics of the POMs may be analogous to the
dynamics found in solid-state metal oxide materials.
In this study, the (R2-P2W17O61)10- and (R1-P2W17O61)10-
isomers that are derivatives of the “plenary” Wells-Dawson
(R-P2W18O62)6- POM, by removal of a WdO3þ unit from the
cap and belt regions, respectively, as shown in Figure 1, can be
considered “defect” structures. These “lacunary” polyoxome-
talates serve as models for defect structures on a solid surface.
The “mono-vacant” defect is comprised of four oxygen donor
atoms that are available for bonding to technetium, Figure 1,
and such oxide defects will bind “hard” Tc moieties.
The defects of POMs possess specific and distinct steric and
electronic features that can affect coordination chemistry and
impact the redox stability of technetium. Specifically, in this
study, monovacant defect structures of the Wells-Dawson
ion ((R2-P2W17O61)10- and (R1-P2W17O61)10- isomers) pos-
sess defects with clearly different steric and electronic proper-
ties, Figure 1. These R1 and R2 defects have different redox
properties and binding strengths to transition metals. Studies
on lanthanide and transition metal POM speciation support
the notion that the high basicity of the R1 vacancy requires
high charge/size cations for stabilization.11-22 The elevated
basicity of the R1 defect is ascribed by Contant and Ciabrini13
(11) Abbessi, M.; Contant, R.; Thouvenot, R.; Herve, G. Inorg. Chem.
1991, 30, 1695.
(12) Boglio, C.; Lenoble, G.; Duhayon, C.; Hasenknopf, B.; Thouvenot,
R.; Zhang, C.; Howell, R. C.; Burton-Pye, B. P.; Francesconi, L. C.; Lacote,
E.; Thorimbert, S.; Malacria, M.; Afonso, C.; Tabet, J.-C. Inorg. Chem.
2006, 45, 1389.
(13) Ciabrini, J.-P.; Contant, R. J. Chem. Research 1993, 391, 2720.
(14) Contant, R.; Abbessi, M.; Canny, J.; Belhouari, A.; Keita, B.; Nadjo,
L. Inorg. Chem. 1997, 36, 4961.
(15) Contant, R.; Ciabrini, J.-P. J. Chem. Res. 1982, 641.
(16) Contant, R.; Richet, M.; Lu, Y. W.; Keita, B.; Nadjo, L. Eur.
J. Inorg. Chem. 2002, 2587.
(17) Harmalker, S. P.; Leparulo, M. A.; Pope, M. T. J. Am. Chem. Soc.
1983, 105, 4286.
(3) Khalil, M. Y.; White, W. B. J. Am. Ceram. Soc. 1983, 66, C197.
(4) Langton, C. A. Challenging Applications for Hydrated and Chemically
Reacted Ceramics; Savannah River Lab: Aiken, SC, 1988.
(5) Langton, C. A. Mater. Res. Soc. Symp. Proc. 1988, 112, 61.
(6) Singh, D.; Mandalika, V. R.; Parulekar, S. J.; Wagh, A. S. J. Nucl.
Mater. 2006, 348, 272.
(7) Fang, X.; Hill, C. L. Angew. Chem., Int. Ed. 2007, 46, 3877.
(8) Day, V. W.; Klemperer, W. G.; Schwartz, C.; Wang, R.-C. In Surface
Organometallic Chemistry: Molecular Approaches to Surface Catalysis;
Kluwer Academic Publishers: New York, 1988; p 173.
(9) Long, D.-L.; Burkholder, E.; Cronin, L. Chem. Soc. Rev. 2007, 36,
105.
(10) Day, V. W.; Klemperer, W. G. Science 1985, 228, 533.
(18) Keita, B.; Girard, F.; Nadjo, L.; Contant, R.; Canny, J.; Richet, M.
J. Electroanal. Chem. 1999, 478, 76.
(19) Leparulo-Loftus, M. A.; Pope, M. T. Inorg. Chem. 1987, 26, 2112.

1672 Inorganic Chemistry, Vol. 50, No. 5, 2011
McGregor et al.
to the orientation of the PO43- tetrahedron within the cavity
of the W-O framework, positioning a basic oxygen atom
near the R1 site. Transition metal cations substituted into the
R1 (belt) position are more readily reduced than when sub-
stituted in the R2 (cap) position.14,16,18,23,24
prepared as described in the literature.13,34 The preparation of
R1-[Fe(H2O)P2W17O61]7- (Fe-R1) and K7-nHn[ReVO(R2-P2W17-
O61)] (ReVO-R2) for electrochemistry was described prev-
iously.30,34 Analytical grade CH3COONa (Fisher) and glacial
CH3COOH (Acros) were used as received. Centrifugations were
performed with an International Equipment Co. Model CL
Clinical centrifuge. Pure water used throughout was obtained
using a Millipore Direct Q5 system (conductivity = 18 μΩ).
Infrared analyses were performed on a Perkin-Elmer 1625 FTIR
spectrometer. Negative-ion electrospray mass spectra were re-
corded on a VG Quattro at the University of Illinois School of
Chemical Sciences Mass Spectrometry Resource.
In this study, 99Tc complexes of the (R2-P2W17O61)10- and
(R1-P2W17O61)10- isomers are prepared and characterized by
multinuclear NMR and X-ray absorption spectroscopy, and
preliminary electrochemical studies are reported. The reduc-
tion and oxidation of TcV is compared in these isomeric
complexes along with the reduction and oxidation of the ReV
analogs. We attempt to identify the features of the polyoxo-
metalate that impact reduction potentials of technetium.
Key Prior Studies of Related Tc and Re POMs. Rhe-
nium, Re, the third row congener of Tc, has been used as a
nonradioactive surrogate for Tc. Re possesses higher re-
duction potentials, slower kinetics, and often an expanded
coordination sphere compared with Tc25-27 and should be
used with caution as a Tc substitute.
Collection of NMR Data. NMR data were collected on a JEOL
GX-400 spectrometer with 5 or 10 mm tubes fitted with a Teflon
insert that were purchased from Wilmad Glass. Resonance fre-
quencies are 161.8 MHz for 31P and 16.7 for 183W. Chemical shifts
are given with respect to external 85% H3PO4 for 31P and 2.0 M
Na2WO4 for 183W. Typical acquisition parameters for 31P spectra
includedthe following:spectralwidth, 10 000Hz;acquisitiontime,
0.8 s; pulse delay, 1 s; pulse width, 15 μs (50ꢀ tip angle). From 200
to 1000 scans were required. For 183W spectra, typical conditions
includedthe following:spectralwidth, 10 000Hz;acquisitiontime,
1.6 s; pulse delay, 0.5 s; pulse width, 50 μs (45ꢀ tip angle). From
1000 to 30 000 scans were acquired. For all spectra, the tempera-
ture was controlled to(0.2ꢀ. For the 31P and 183W chemical shifts,
the convention used is that the more negative chemical shifts
denote more upfield resonances.
The (ReVO)3þ and (TcVO)3þ units have been introduced
into the lacunary R-Keggin, R-XW11O39n- (X = P, n = 7;
X = Si, n = 8).28,29 The purity of these analogs has been
demonstrated by IR, electrochemistry, mass spectrometry,
and EPR (ReVI). We have isolated pure ReV,VI,VII (R2-
P2W17O61)10- complexes as verified by 31P and 183W NMR
spectroscopy along with the above-mentioned techniques.30
A slight modification of crystallization using acidic condi-
tions of the aqueous ReVO-R2-P2W17O61 complex resulted
in a 2:2 dimer.31
Extended X-Ray Absorption Fine Structure (EXAFS) Spec-
troscopy. Samples were dissolved in 1 mL of 18 MΩ water then
transferred to 2 mL screw-capped, polypropylene centrifuge
tubes, which where sealed inside two nested polyethylene bags.
Using these samples, data were acquired at room temperature in
transmission mode Ar-filled ion chambers at SSRL beamlines
11-2 and 4-1; data were collected using the locally written
program XAScollect. The harmonic content of the beam was
reduced by detuning the monochromator by 50%. The data were
processed using EXAFSPAK and Athena/ifeffit.35,36 Data were
fit using Artemis/ifeffit and theoretical phases and amplitudes
calculated using FEFF7.37 The initial model used in the FEFF
calculation was (NH4)6(P2W18O62) with one W atom replaced by
Tc.38 Additional scattering shellswereadded onlyif their inclusion
lowered the value of reduced χ-squared.
The F-test was used to analyze the significance of the fitting
parameters including the significance of adding a scattering
shell.39 The null hypothesis is that the additional shell of atoms
does not improve the fit. The result of the F-test is the probability,
p, that this hypothesis is correct. If p < 0.05, the null hypothesis is
rejected in favor of the alternative hypothesis that the additional
shell significantly improves the fit.40
Experimental Section
General. All materials were purchased as reagent grade and
used without further purification. Potassium hexachlororhenate
was purchased from Aldrich (99.99% purity). 99Tc is a weak
β-emitter with a half-life of 2 ꢀ 105 years. All syntheses and sample
preparations were performed in laboratories approved for low-
level use of radioactivity using appropriate radioactive material
handling procedures. 99TcO4- was purchased as the ammonium
salt from Oak Ridge National Laboratory and treated with H2O2
to oxidize any reduced Tc.32 (NBu4)TcOCl4 was prepared from
NH4TcO4 according to an established procedure.33 K9Li(R1-
P2W17O61) (R1) and K10(R2-P2W17O61) (R2) ligands were
(20) Sadakane, M.; Dickman, M. H.; Pope, M. T. Inorg. Chem. 2001, 40,
2715.
(21) Sadakane, M.; Ostuni, A.; Pope, M. T. J. Chem. Soc., Dalton Trans.
2002, 63.
(22) Zhang, C.; Howell, R. C.; Luo, Q.; Fieselmann, H. L.; Todaro, L.;
Francesconi, L. C. Inorg. Chem. 2005, 44, 3569.
(23) Contant, R.; Herve, G. Rev. Inorg. Chem. 2002, 22, 63.
(24) Contant, R.; Thouvenot, R. Inorg. Chim. Acta 1993, 212, 41.
(25) The Inorganic Chemistry of Technetium and Rhenium As Relevant to
Nuclear Medicine; Deutsch, E., Libson, K., Vanderheyden, J.-L., Eds.; Cortina
International: Verona, Italy, 1990; Vol. 3.
(26) Libson, K.; Woods, M.; Sullivan, J.; Watkins, J. W., II; Elder, R. C.;
Deutsch, E. Inorg. Chem. 1988, 27, 999.
Electrochemical Data Collection. Electrochemical data were
obtained using a BAS Voltammetric Analyzer System controlled
by the BAS CV-50W software (for PC). The cell used for cyclic
voltammetry (CV) contained a glassy-carbon working electrode
(BAS standard disk electrode, 3 mm OD), a Pt wire auxiliary
electrode (0.5 mm), and a BAS Ag/AgCl (3 M NaCl) reference
electrode.
The electrochemistry of the compounds was conducted at a
concentration of 0.2 mM in a pH 5 buffer (0.5 M NaSO4 in 0.01 M
NaOAc). Prior to obtaining electrochemical data, solutions were
(27) VanderHeyden, J.-L.; Heeg, M. J.; Deutsch, E. A. Inorg. Chem. 1985,
24, 1666.
(28) Abrams, M. J.; Costello, C. E.; Shaikh, S. N.; Zubieta, J. Inorg. Chim.
Acta 1991, 180, 9.
(29) Ortega, F.; Pope, M. T. Inorg. Chem. 1984, 23, 3292.
(30) Venturelli, A.; Nilges, M. J.; Smirnov, A.; Belford, R. L.; Francesconi,
L. C. J. Chem. Soc., Dalton Trans. 1999, 301.
(34) Contant, R. Inorg. Synth. 1990, 27, 71.
(35) Newville, M. J. Synchrotron Radiat. 2001, 8, 322.
(36) Ravel, B. Phys. Scr. 2005, T115, 1007.
(37) Rehr, J. J.; Albers, R. C.; Zabinsky, S. I. Phys. Rev. Lett. 1992, 69,
3397.
(31) Kato, C. N.; Hara, K.; Hatano, A.; Gogo, K.; Kuribayashi, T.;
Hayashi, K.; Shinohara, A.; Kataoka, Y.; Mori, W.; Nomiya, K. Eur.
J. Inorg. Chem. 2008, 2008, 3134.
(38) D’Amour, V. H. Acta Crystallogr. 1976, B32, 729.
(39) Bevington, P. R.; Robinson, K. D. Data Reduction and Error
Analysis for the Physical Sciences; McGraw-Hill: Boston, MA, 1992.
(40) Downward, L.; Booth, C. H.; Lukens, W. W.; Bridges, F. In AIP
Conference Proceedings 2007; American Institute of Physics: College Park, MD,
2007; Vol. 882, p 129.
(32) Boyd, G. E. J. Chem. Educ. 1959, 36, 3.
(33) Davison, A.; Trop, H.; DePamphilis, B.; Jones, A. Inorg. Synth.
1982, 21, 160.

Article
Inorganic Chemistry, Vol. 50, No. 5, 2011 1673
183W NMR, δ (ppm)
Table 1. Multinuclear NMR Data for TcV-R1/R2-P2W17O61 Complexesa
compound
solvent
D2O
31P NMR, δ (ppm)
TcVO-R2
-11.60 (1), -13.16 (1)
-131.06 (2), -161.51 (1), -179.91 (2),
-181.72 (2), -192.19 (2), -195.23 (2),
-199.69 (2), -214.82 (2), -228.78 (2)
37.21 (1), -117.85 (1), -120.78 (1), -132.83 (1),
-147.71 (1), -172.40 (1), -173.08 (1), -174.01 (1),
-185.67 (1), -188.84 (1), -189.09 (1), -190.01 (2),
-203.29 (1), -204.70 (2), -229.39 (1)
35.17(1), -108.55(1), -111.43(1),
TcVO-R1
ReVO-R1
D2O
D2O
-11.60 (1), -12.72 (1)
-12.02 (1), -12.33 (1)
-157.35(1), -167.90(1), -180.02(2),
-184.51(1), -197.41(1), -208.1 (2), -228.77(1),
-237.95(1), -242.69(1), -258.18(1), -308.79(1),
-398.52(1).
a See text for conditions. All spectra were taken at room temperature, 25 ꢀC.
deaerated for at least 30 min with high purity Ar. A positive
pressure of this gas was maintained during subsequent work.
Preparation, including fine polishing of the glassy-carbon working
electrode, was adapted from the procedure of Keita and co-
by filtration and recrystallized twice from 50 mL of hot water. The
product was 95% pure as judged by 31P NMR. Yield: 1.08 g, 0.22
mmol. In order to remove impurities arising from the ReVO-R2
congener, 0.4 g of recrystallized product was dissolved in 5 mL
of distilled water and loaded onto a silica column (5.5 cm
(diameter) ꢀ 60 cm filled to a height of 40 cm with silica) primed
with methanol and eluted with methanol. The dark blue fraction
containing ReVO-R1 was reduced in volume by rotary evapora-
tion and filtered to remove any silica. The solvent was then
removed by rotary evaporation to yield pure ReVO-R1, free from
ReVO-R2.
workers.41 Unless indicated otherwise, scan rates were 10 mV s-1
,
and all experiments were carried out at ambient temperature
under an atmosphere of Ar.
Synthesis of Compounds.
K
7-nHn[TcVO(r1-P2W17O61)]
(TcVO-r1). To a 20 mL scintillation vial containing a yellow/
green solution of (NBu4) TcOCl4 (100 mg, 0.2 mmol) dissolved in
1 mL of MeOH was added 60 μL (0.98 mmol) of ethylene glycol
(CH2OH)2 (eg) to produce a blue/green TcO(eg)2- complex. This
solution was quickly added to a clear, colorless solution of R1
(447 mg, 0.099 mmol) dissolved in 5 mL of H2O containing LiCl
(10 mg, 0.2 mmol). The resulting dark red solution and brown
suspension was stirred for 3 min. The reaction mixture was
centrifuged (15 min, 70 000 rpm) and the red supernatant de-
canted. KCl (0.16 g, 2 mmol) was added to the supernatant and
was stored at room temperature for up to 48 h to allow any
unreacted R1 (as monitored by 31P NMR) to precipitate. The
mixture exhibiting a clean, two peak 31P NMR was filtered and
20 mL of ethanol added to the filtrate to precipitate a dark red
solid, which was collected by vacuum filtration.
The 31P NMR and 183W NMR data for the complexes are
given in Table 1.
Results and Discussion
Synthesis of Complexes. The synthesis of pure transi-
tion metal polyoxometalates is often complicated by the
difficulty in separating unreacted POM ligands from the
metal POM complex. Usually, separations are achieved
by fractional crystallizations and are complicated by the
similar solubilities of the metal complexes of POMs and
the POM ligands. Separations are rendered even more
difficult due to the small volumes manipulated in frac-
tional crystallizations when working with milligram
quantities of 99Tc. The small volumes, multiple oxidation
Yield: 40-60% based on (NBu4)TcOCl4. IR (KBr, cm-1):
790 cm-1 (strong, broad), 953 cm-1 (weak), 1086 cm-1 (strong).
K7-nHn[TcVO(r2-P2W17O61)] (TcVO-r2). To a 20 mL scin-
tillation vial containing a yellow/green solution of (NBu4)TcOCl4
(100 mg, 0.2 mmol) dissolved in 1 mL of MeOH was added 60 μL
-
states accessible to 99Tc, and hydrolysis of the 99TcOCl4
-
starting reagent generally compromise the separation of
the reactant from the products, reproducibility, and yield
of the reactions
(0.98 mmol) of ethylene glycol to produce a blue/green TcO(eg)2
complex. This solution was added quickly to a clear, colorless
solution of R2 (447 mg, 0.099 mmol) dissolved in 5 mL of H2O at
80 ꢀC. The resulting dark red solution and brown suspension was
stirred for 3 min. While still warm, the reaction mixture was
centrifuged (15 min, 70 000 rpm), the dark red supernatant
decanted, and KCl (0.16 g, 2 mmol) added. The red solution
was allowed to sit at room temperature for up to 72 h to allow any
unreacted R2 (as monitored by 31P NMR) to precipitate. The
mixture exhibiting a clean, two peak 31P NMR spectrum of the
product was filtered, and 10 mL of ethanol was added to the
filtrate to precipitate a dark red solid, which was collected by
vacuum filtration.
The (NBu4)TcOCl4 starting reagent readily hydrolyzes
in water and in methanol; in order to stabilize this source
of TcVO, (NBu4)TcOCl4 was treated with ethylene glycol,
which acts as a “transfer ligand”42 to inhibit the hydro-
lysis of TcOCl4-, stabilize the TcVO core, and facilitate
transfer of the TcVO moiety for isolation of TcVO-R1 and
TcVO-R2. The TcO(eg)2 complex43 was formed in situ,
-
and the ethylene glycol ligands were replaced by the
tetradentate R1/R2 ligands to form the TcV complexes.
The transfer ligand protocol used here is similar to those
used in radiopharmaceutical kits (that employ the clinical
isotope 99mTc on the tracer level) where ligands such
as glucoheptonate form weak complexes to stabilize
the 99mTcVO core for eventual substitution into the
desired 99mTcVO radiopharmaceutical.
Yield: 40-60% based on (NBu4)TcOCl4. IR (KBr, cm-1):
790 cm-1 (strong, broad), 953 cm-1 (weak), 1086 cm-1 (strong).
K7-nHn[ReVO(r1-P2W17O61)] (ReVO-r1). A 200 mL round-
bottom flask was charged with K2ReCl6 (0.285 g, 0.6 mmol) and
R1 (2.39 g, 0.48 mmol), and deoxygenated water was added (30
mL) to form a slurry. The slurry was heated at 40 ꢀC for 30 min,
producing a navy blue solution. The solution was cooled to room
temperature, 125 mL of ethanol added, and the sample placed in
the freezer for 10 min. The dark microcrystalline solid was isolated
(42) DePamphilis, B. V.; Jones, A. G.; Davison, A. Inorg. Chem. 1983, 22,
2292.
(43) Davison, A.; DePamphilis, B. V.; Jones, A. G.; Franklin, K. L.;
Lock, C. J. L. Inorg. Chim. Acta 1987, 128, 161.
(41) Keita, B.; Girard, F.; Nadjo, L.; Contant, R.; Belghiche, R.; Abessi,
M. J. Electroanal. Chem. 2001, 508, 70.

1674 Inorganic Chemistry, Vol. 50, No. 5, 2011
McGregor et al.
We found that a 1:1 Tc/R1 and R2 stoichiometry and a
Tc/eg ratio of between 1:3.5 and 1:6 were required to obtain
an optimal yield and purity. A ratio of less than 1:3.5
resulted in the formation of an increased amount of both
Table 2. Mass Spectrometry of the TcV-R1/R2-P2W17O61 Complexes
compound
ion
m/z
2-
TcVO-R2
K4HTcO[P2W17O61
K4TcO[P2W17O61]
K3HTcO[P2W17O61
]
2217.9
1478.5
1465
1453
1440
1427
1089.6
1080.2
1070.7
2217.8
1477.6
1465
3-
a dark solid, presumed to be mostly TcO2 H2O, and freeR1
3-
3
]
K2H2TcO[P2W17O61
3-
or R2 ligand. A ratio of more than 1:6 resulted in a low yield
and increased difficulty when isolating a pure product as a
solid. The synthetic conditions balance the hydrolysis of the
]
3-
KH3TcO[P2W17O61
]
3-
H4TcO[P2W17O61]
K2HTcO[P2W17O61
KH2TcO[P2W17O61
4-
TcOCl4- startingmaterialtoTcO2 H2O and incorporation
]
]
4-
3
of Tc(V)O into the R1 and R2 frameworks. The develop-
4-
H3TcO[P2W17O61]
K4HTcO[P2W17O61
2-
TcVO-R1
]
ment of the synthesis, inhibition of TcO2 H2O formation,
3
and formation of a TcIV-(μ-O)2-TcIV R1/R2 dimer, or a
TcIV-(μ-O)2-TcIV ethyleneglycol complex, as found from
EXAFS and XANES, will be discussed in a separate
publication.
3-
K4TcO[P2W17O61]
K3HTcO[P2W17O61
3-
]
4-
K3TcO[P2W17O61
]
1099
4-
K2HTcO[P2W17O61
KH2TcO[P2W17O61
]
]
1089.8
1079.1
1070.3
4-
ReVO-R1 was initially prepared in the same fashion as
reported for ReVO-R2;however, thepreparation resulted in
the formation of the parent Wells-Dawson ion, (R-
P2W18O62)6-, and ReVO-R2 impurities. Elevated tempera-
tures increased the amount of P2W18O626- and ReVO-R2.
Running the reaction at lower temperatures reduced the
amount of P2W18 and ReVO-R2 but did not eliminate the
formation of these impurities entirely. Multiple recrystalli-
zations were attempted to isolate pure ReVO-R1, but trace
amounts of the ReVO-R2 isomer were visible by 31P NMR.
Column chromatography utilizing silica gel as the station-
ary phase and employing methanol as an eluent separated
the ReVO-R2 from ReVO-R1. Figure S1 (Supporting In-
formation) illustrates the purification process as monitored
by 31P NMR and cyclic voltammetry.
H3TcO[P2W17O61
]
4-
spectral data confirm the formulation of the technetium
complexes.
Multinuclear NMR Spectroscopy. Multinuclear NMR
spectroscopy, given in Table 1, is used to establish the purity
and structure of TcVO-R1, TcVO-R2, and ReVO-R1. 31P
NMR is a well-established technique to assess impurities
of polyoxometalates at the 2% level46-48 and has been used
to identify pure Wells-Dawson complexes of transition
metals46-50 and lanthanides.12,20-22,49,51-56 (We also turn
to electrochemistry to clearly observe signature patterns
that identify the R1 and R2 isomer ligands18,57 and to
corroborate TcV and ReV substitution into the R1 and R2
frameworks, vide infra.) The 31P NMR spectra along with
the 183W spectra are presented in Figure 2 and Figure S4
(Supporting Information) for TcVO-R1 and TcVO-R2. The
31P NMR spectra establish that the purities of the TcV
complexes of R1 and R2 are greater than 98%. Two
resonances are observed for the TcVO-R1 and TcVO-R2
species. The upfield resonance is assigned to P1, the phos-
phorus atom close to the TcV center, and the remote
phosphorus, P2, is assigned to the downfield resonance.
183W NMR is used for structural identification of the TcV
complexes in solution. Each magnetically inequivalent
tungsten atom presents one resonance, with P coupling
and W-O-W coupling if the concentration, and thus
intensity and resolution, is high enough. The TcVO-R2
Characterization of Complexes. Infrared Spectroscopy.
The infrared spectra of TcVO-R1 and TcVO-R2 show char-
acteristic stretches at 790 cm-1 (strong, broad), 907 cm -1
,
953 cm-1 (weak), and 1086 cm-1(strong). Representative
IR spectra (KBr pellet) are shown in the Supporting
Information, Figure S2. The IR spectra are, not surpris-
ingly, similar to the parent (P2W18O62)6-. In a study by
Ortega and Pope, the IR spectra for ReV, ReVI, and ReVII
n-
substituted into the XW11O39 framework are similar to
the parent XW12O40 .
(nþ2)- 29 We observed also that the IR
spectra for ReV, ReVI, and ReVII substituted into the (R2-
P2W17O61)10- framework are similar to that of the parent
(P2W18O62)6- 30
. The TcdO stretch is expected to occur at
900-960 cm-1, and that would overlap with the bands
observed in the W framework.
(46) Finke, R. G.; Droege, M. W.; Domaille, P. J. Inorg. Chem. 1987, 26,
3886.
(47) Finke, R. G.; Lyon, D. K.; Nomiya, K.; Sur, S.; Mizuno, N. Inorg.
Chem. 1990, 29, 1784.
(48) Finke, R. G.; Rapko, B.; Saxton, R. J.; Domaille, P. J. J. Am. Chem.
Soc. 1986, 108, 2947.
(49) Bartis, J.; Kunina, Y.; Blumenstein, M.; Francesconi, L. C. Inorg.
Chem. 1996, 35, 1497.
(50) Jorris, T. L.; Kozik, M.; Casan-Pastor, N.; Domaille, P. J.; Finke,
R. G.; Miller, W. K.; Baker, L. C. W. J. Am. Chem. Soc. 1987, 109, 7402.
(51) Bartis, J.; Dankova, M.; Lessmann, J. J.; Luo, Q.-H.; Horrocks,
W. D., Jr.; Francesconi, L. C. Inorg. Chem. 1999, 38, 1042.
(52) Bartis, J.; Sukal, s.; Dankova, M.; Kraft, E.; Kronzon, R.; Blumenstein,
M.; Francesconi, L. C. J. Chem. Soc., Dalton Trans. 1997, 1937.
(53) Howell, R. C.; Perez, F. G.; Jain, S.; Horrocks, W. D., Jr.;
Rheingold, A. L.; Francesconi, L. C. Angew. Chem., Int. Ed. 2001, 40, 4301.
(54) Salmonte, J. L.; Pope, M. T. Can. J. Chem. 2001, 79, 802.
(55) Zhang, C.; Bensaid, L.; McGregor, D.; Fang, X.; Howell, R. C.;
Burton-Pye, B.; Q., L.; Todaro, L.; Francesconi, L. C. J. Cluster Sci. 2006,
17, 389.
Mass Spectrometry. Negative-ion electrospray mass
spectrometry provides data with minimal fragmentation
and has been found to be a useful tool for the analysis of
polyoxometalates.12,30,44,45 The data for the mass spectro-
metry analysis of TcVO-R1 and TcVO-R2 are presented in
Table 2. The highly negatively charged molecules form
adductswith cations, andthusdifferentchargestatescan be
formed. In aqueous solution, the -4, -3, and -2 charge
state clusters are observed. Table 2 and Figure S3
(Supporting Information) show the corresponding m/z
values for the various charge states of the molecules with
variable amounts of potassium and proton adducts. Taken
with the multinuclear NMR data, vide infra, these mass
(44) Bonchio, M.; Bortolini, O.; Conte, V.; Sartorel, A. Eur. Jour. Inorg.
Chem. 2003, 4, 699.
(45) Deery, M. J.; Howarth, O. W.; Jennings, K. R. J. Chem. Soc., Dalton
Trans. 1997, 4783.
(56) Zhang, C.; Howell, R. C.; Scotland, K. B.; Perez, F. G.; Todaro, L.;
Francesconi, L. C. Inorg. Chem. 2004, 43, 7691.
(57) Keita, B.; Mbomekalle, I.-M.; Nadjo, L.; Contant, R. Eur. J. Inorg.
Chem. 2002, 473.

Article
Inorganic Chemistry, Vol. 50, No. 5, 2011 1675
Figure 2. Multinuclear NMR Data for TcVO-R1 and TcVO-R2 taken at 25 ꢀC in D2O. Top: 31P NMR data; bottom: 183W NMR data.
clearly shows nine well-resolved resonances with appropri-
ate integrations for incorporation of the TcVO center
into the cap region of the R2 lacunary polyoxometalate
(Figure 2). This spectrum is different from free R2.49,58
Incorporation of the TcVO center into R1 is clearly identi-
fied by the 183W NMR spectrum (Figure 2) as well. In this
case, R1 possesses C1 symmetry, and 17 resonances are
observed in the 183W NMR spectrum of the TcV complex.
The pattern observed in Figure 2 is clearly different from
the TcVO-R2 species and from R1 and R2.49,50 Although the
resonances have not been assigned to specific tungsten
atoms, the downfield resonance is likely attributed to a W
atom close to the vacancy, vide infra. The cause of the
downfield shift is discussed in detail in the EXAFS section
of this manuscript. In summary, the mass spectroscopy
confirms the expected formulations, and the multinuclear
(31P and 183W) NMR spectroscopy is consistent with the
TcVO centers incorporating into the cap and belt regions of
R2 and R1 ligands, respectively. The isolated, pure species
were carried forward for comparison in electrochemical
and X-ray absorption spectroscopy experiments, vide infra.
X-Ray Absorption Fine Structure. The X-ray absorption
,
-
near edge structure (XANES) spectra of TcO4
TcO2 2H2O, TcVO-R1, and TcVO-R2 are shown in Figure 3.
3
In this case, the energy of the absorption edge is a reflection
of the amount of screening experienced by the 1s electrons.
For this reason, the absorption edges of complexes with
higher oxidation states typically occur at higher energies
than those of complexes with lower oxidation states. This
trend can, however, be complicated by the effects of
covalency, which can shift the energy of the absorption
edge from what would be expected on the basis of formal
oxidation state alone. In the case of the complexes shown in
Figure 3, the effects of covalency are anticipated to be
minor compared with the effects of oxidation state. The
half-height of the edges of TcO2 2H2O, TcVO-R1, and
3
TcVO-R2 with respect to that of TcO4- are -6.7 eV, -3.2 eV,
and -3.7 eV, respectively. The relative position of the Tc
K-edge absorption edges of TcVO-R1 and TcVO-R2 with
respect to TcO2 2H2O are consistent with the presence of
3
Tc(V). In addition, the Tc center in TcVO-R2 is slightly
more electron rich than that of TcVO-R1.
(58) Acerete, R.; Hammer, C. F.; Baker, L. C. W. J. Am. Chem. Soc. 1982,
104, 5384.

1676 Inorganic Chemistry, Vol. 50, No. 5, 2011
McGregor et al.
The XANES spectra also contain information about the
symmetry of the coordination environment in TcVO-R1
and TcVO-R2. The small, well-defined peaks below
21050 eV are pre-edge features due to the 1s-4d transition,
which is symmetry forbidden in centrosymmetric com-
plexes. The presence of this feature in the spectra of
TcVO-R1 and TcVO-R2 is consistent with the postulated,
noncentrosymmetric structure.
Extended X-ray absorption fine structure (EXAFS)
spectra can be used to determine the local structure of a
particular chemical element. The Tc K-edge solution EX-
AFS of TcVO-R1 and TcVO-R2 are shown Figure 4 along
with the EXAFS spectra of model complexes derived by
fitting the data. To determine the local structure of the Tc
sites, model complexes were created by replacing a tungsten
atom in the cap (TcVO-R2) or the belt (TcVO-R1) of the
Wells-Dawson plenary ion by a Tc center. The data were
fit using shells of scattering atoms at increasing distances
from the Tc center. If the value of reduced χ-squared
decreased, the scattering shells were retained. The numbers
of heavy atom (W, P) neighbors and nearest O neighbors
were determined from the model, but the number of more
distant oxygen neighbors were determined by varying this
number to obtain the best fit.
TcVO-r2. The local structure obtained by fitting the
EXAFS spectrum of TcVO-R2 complexes is consistent with
the structure of the cap W site of the Wells-Dawson ion.
The distances from the Tc atom to the neighboring atoms
are given in Table 3.59 In this case, p(F) values for all
scattering atoms are less than 0.05, so all atoms can be
considered to be observed in the EXAFS experiment. The
structure is consistent with a square pyramidal Tc(V). The
oxygen coordination environment contains a very short
˚
Tc-O bond, 1.64 A, which is consistent with the terminal
oxo ligand of a Tc(V) ion, and the best fit is obtained with
˚
four oxygen ligands at 2.00 A and a more distant oxygen
˚
neighbor at 2.53 A. Scattering from more distant atoms,
particularly P and W, is consistent with coordination of the
Tc(V) ion by (R2-P2W17O61)10-. The best fit is obtained
with two neighboring W atoms at 3.43 A, one neighboring
˚
˚
P atom at 3.54 A, and two more neighboring W atoms at
˚
3.57 A, in good agreement with the distances predicted if a
W atom in the cap of the Wells-Dawson ion is replaced by
a Tc atom. The major difference is that the terminal oxo of
Tc(V) is much shorter than the terminal oxo of W(VI).
EXAFS data collected at the Advanced Photon Source,
Argonne National Laboratory (Figure S5, Table S1, Sup-
porting Information), on the TcVO-R2 are in excellent
Figure 3. Tc K-edge XANES spectra of, from lowest energy to highest
energy: TcO2 2H2O (black), TcVO-R2 (green), TcVO-R1 (orange), and
3
TcO4- (blue). Data obtained using SSRL beamline 4-1.
agreement with the above results. The Tc atom is 6-co-
V
ordinate with oxygen atoms at 1.63(1) A (Tc O), 1.93(2) A,
˚
˚
˚
and 2.36(2) A. The results from the APS further show W
˚
atoms at 3.43(3) A and 3.57 (3) A, a P atom at 3.48(3) A,
˚
˚
˚
and an O atom at 3.98(4) A, which is consistent with Tc(V)
coordinated to (R2-P2W17O61)10-
.
TcVO-r1. Surprisingly, the local structure obtained by
fitting the EXAFS spectrum of TcVO-R1 is not consistent
with simply replacing a W atom in the belt of the Well-
s-Dawson ion by a Tc atom, as shown by the comparison
of the bond distances derived by fitting the EXAFS spec-
trum with the local structure of the belt W atoms of the
Wells-Dawson ion given in Table 4. Initial refinements
using this model suggested that Tc possessed a neighboring
Figure 4. Tc K-edge EXAFS spectra (left) and Fourier transforms
(right) for TcVO-R1 (upper) TcVO-R2 (lower). Data are shown in color,
and fits are given in black. Data obtained using SSRL beamline 4-1.
˚
atom at ∼3 A. Furthermore, fitting of the nearest oxygen
Table 3. Best Fit Parameters for the EXAFS Spectrum for Kþ Salt of TcVO-R2
a
2
2
c
# of neighborsb
σ (A )
p(F)b
˚
distance (A)
˚
˚
local structure of the “cap” tungsten site (A)
neighbor
O
O
O
W
P
W
1
4
1
2
1
2
6
2
6
1.638(4)
1.996(4)
2.53(2)
3.43(2)
3.54(5)
3.57(2)
3.97(2)
4.28(7)
4.47(1)
0.0014(5)
0.0025(2)
0.0025(2)d
0.003(1)
0.003(1)d
0.004(2)
0.002(1)
0.002(1)d
0.002(1)d
<0.001
<0.001
0.016
0.011
0.018
<0.001
0.002
0.034
1.72
1.93
2.33
3.38
3.52
3.62
O
6 from 3.7 to 4.1
4.05
8 from 4.3 to 4.6 A
O (MS)e
O
˚
<0.001
a Fit range: 2 < k < 15, 1.1 < R < 3.9, 30 independent data. Fifteen parameters, S02 = 0.9, ΔE0 = 71 eV, χ2 = 25, χν2 = 2, R = 0.09. b F-test
determines the significance of the improvement to the fit created by adding an additional set of atoms. If the p value is less than 0.05, the additional atoms
have significantly improved the fit and can be considered “observed” in the EXAFS experiment. c Dawson, B. Acta Crystallogr. 1953, 6, 113-126.
d Debye-Waller of this shell constrained to be equal to that of the preceding shell. e Multiple scattering from the mutually trans oxygen ligands

Article
Inorganic Chemistry, Vol. 50, No. 5, 2011 1677
Table 4. Best Fit Parameters for the EXAFS Spectrum for TcVO-R1a
2
2
# of neighborsb
σ (A )
p(F)b
local structure of the “belt” tungsten sitec
˚
distance (A)
˚
neighbor
O
O
W
P
O
W
1
5
1
1
5
3
6
4
1.640(3)
1.974(2)
3.087(4)
3.394(5)
3.31(4)
3.456(7)
4.08(2)
4.44(2)
0.0013(4)
0.0033(2)
0.0022(5)
0.0022(15)d
0.02(1)
<0.001
<0.001
0.014
0.042
0.054
<0.001
0.001
0.032
1.72
2.01
3.37
3.51
8 from 3.5 to 4.1
3.72
4.05
0.0068(7)
0.004(3)
O (multiple scatter)e
O
0.005(2)2
5 from 4.4 to 4.5 A
˚
a Fit range: 2 < k < 14, 1 < R < 5; 33 independent data. Sixteen parameters, S02 = 0.9, ΔE0 = 4.2(9) eV, χ2 = 428, χν2 = 26, R = 0.009. b F-test
determines thesignificance of the improvementtothe fit createdby addingan additional set of atoms. If thep value is less than 0.05, the additional atomshave
significantly improved the fit and can be considered “observed”in the EXAFSexperiment.c Dawson, B. Acta Crystallogr. 1953, 6, 113-126. d Debye-Waller
of this shell constrained to be equal to that of the preceding single-scattering shell. e Multiple scattering due to mutually trans oxygen ligands.
shells with a square pyramidal model did not fit the data
accurately; an acceptable fit could only be obtained with the
typical short Tc(V) bond plus five next-nearest oxygen
neighbors. The oxygen neighbors suggested that the Tc
was in a pseudo-octahedral rather than square pyramidal
geometry. Therefore, the initial Tc model was replaced by
rich than the Tc in TcVO-R2, as suggested by the XANES
spectrum. The presence of a similar peak in the 183W NMR
spectrum of ReVO-R1 suggests that this complex possesses
a similar metal-metal interaction.
Electrochemistry. As reported by Nadjo and co-
workers,18 cyclic voltametry (CV) is a sensitive technique
used to identify the R1 and R2 isomers and their complexes
with transition metals and to assess their purity. In this
study, CV has been conducted in aqueous 0.1 M Na-
COOCH3 buffer solutions containing 0.5 M NaSO4 at
pH 5. The stability of both transition metal-substituted
R1/R2 complexes and R1/R2 ligands with this combination
of supporting electrolyte and pH has been previously
demonstrated.16,18,58
˚
one with the TcO unit translated by 0.3 A toward the
phosphate oxygen in the center of the Wells-Dawson ion.
The revised model is in good agreement with the EXAFS
V
˚
data. Tc O-R1 still possesses the short Tc-O bond, 1.64 A,
typical of Tc(V) but has five next-nearest oxygen neighbors
˚
at 1.97 A, rather than the four next-nearest oxygen neigh-
bors found in TcVO-R2. The distance to the nearest W atom
is only 3.09 A, which is much shorter than the nearest
˚
distance between neighboring W atoms in the belt of the
˚
Figures 5 and 6 compare the CVs of R2 and R1 lacunary
POMs (solid lines) and their respective TcVO complexes
(dotted lines). In each case, the scans are restricted to a
region where no derivitization of the electrode surface is
observed; proceeding to more negative potentials can lead
to the adsorption of reduced species to the electrode
surface. The CVs of TcVO-R2 and TcVO-R1 complexes
(Figures 5 and 6, respectively) show new waves that are not
observed in the lacunary parents. These waves can be
attributed to Tc3þ/Tc4þ, Tc4þ/Tc5þ, and Tc5þ/Tc6þ redox
couples within the substituted complex. This is analogous
to complexes of the R2 and R1 isomer containing Cu3þ or
Fe3þ, where the reduction of the copper or iron occurs
before the reduction of the tungsten centers.57,65 The
observed rest potentials for the samples, and the observa-
tion of sharply resolved signals in the 31P NMR (suggestive
of diamagnetic species) allow us to assign the waves ca. -33
and -175 mV (vs Ag/AgCl) as Tc4þ/Tc5þ redox couples
within the R1 and R2 complexes, respectively. Comparison
with the Fe2þ/Fe3þ wave in the compound Fe-R1 reveals
that this first step consumes one electron per molecule (see
Figure S6, Supporting Information). Waves occurring at
more negative potentials are attributed to the reduction of
the tungsten framework.
Wells-Dawson ion, 3.37 A. Likewise, the distances to the
phosphorus at the next nearest W atoms are shorter than
expected. Overall, the data suggest that the Tc(V) center in
TcVO-R1 has been “pulled into” the Wells-Dawson ion
relative to the position of the belt W atom.
A potential explanation for this unexpected structure is
suggested by the peak at 37.21 ppm in the 183W NMR
spectrum of TcVO-R1. A strong positive shift in the 183W
NMR resonances is associated with two scenarios: two-
electron reduced diamagnetic polyoxometalates where the
electrons are “delocalized” on the NMR or EPR time scales
have been reported for the 2e-reduced R-[P2W18O62]6-
,
[W10O32]6-, and [γ-SiW12O40]6- species.60-63 Strong posi-
tive shifts are also characteristic of the presence of a
metal-metal bonding association. The latter was found
for the case of [γ-SiW12S2O38]6- wherethedeshieldedpeaks
at þ1041 ppm correspond to WV atoms of the {W2O2S2} frag-
ment.64 The distances between these WV atoms are quite short
˚
(2.815 (1) A), characteristic of a metal-metal bond involving
both the WV centers.
In the TcVO-R1 case, a similar metal-metal bonding
may be occurring: the bonding association would involve a
dative bond from the occupied Tc dxy orbital into the
˚
unoccupied dxy orbital of the W atom at 3.09 A with which
the Tc shares an edge. The presence of the Tc-W interac-
The redox waves occurring at more positive potential
are assigned as Tc5þ/Tc6þ redox couples within the sub-
stituted complexes (þ990 mV for TcVO-R1 and þ813 mV
for TcVO-R2). Extending the electrochemical window to
more positive potentials allows us to observe a quasi-
reversible wave in both compounds. This is assigned to
the Tc6þ/Tc7þ redox process; however, the Tc(VII) spe-
cies is unstable. Prolonged exposure to positive potentials
tion also explains why the Tc in TcVO-R1 is less electron
(59) Dawson, B. Acta Crystallogr. 1953, 6.
(60) Kazansky, L. P.; Yamase, T. J. Phys. Chem. A 2004, 180, 6437.
ꢀ ꢀ
ꢀ
(61) Teze, A.; Canny, J.; Gurban, L.; Thouvenot, R.; Herve, G. Inorg.
Chem. 1996, 35, 1001.
(62) Duncan, D. C.; Hill, C. L. Inorg. Chem. 1996, 35, 5828.
(63) Kozik, M.; Hammer, C. F.; Baker, L. C. W. J. Am. Chem. Soc. 1986,
108, 2748.
ꢀ ꢀ
ꢀ
ꢀ
(64) Teze, A.; Cadot, E.; Bereau, V.; Herve, G. Inorg. Chem. 2001, 40,
(65) Keita, B.; Belhouari, A.; Nadjo, L.; Contant, R. J. Electroanal.
Chem. 1998, 442, 49.
2000.

1678 Inorganic Chemistry, Vol. 50, No. 5, 2011
McGregor et al.
Figure 5. Cyclic voltammograms of R2 (solid line) and TcVO-R2 (dashed line) in 0.1 M CH3COONa/CH3COOH containing 0.5 M NaSO4, pH = 5.00.
Working electrode, glassy carbon; auxiliary electrode, platinum wire; reference electrode, Ag/AgCl. Scan rate = 10 mV s-1
.
Figure 6. Cyclic voltammograms of R1 (solid line) and TcVO-R1 (dashed line) in 0.1 M CH3COONa/CH3COOH containing 0.5 M NaSO4, pH = 5.00.
Working electrode, glassy carbon; auxiliary electrode, platinum wire; reference electrode, Ag/AgCl. Scan rate = 10 mV s-1
.
Table 5. Halfwave Potentials for the Electroactive TcVO and ReVO Substituted
a more negative potential (-386 mV), which we are
attributing to a Tc3þ/Tc4þ redox couple.
within the R1 and R2 Frameworks at pH 5, Scan Rate 10 mV s-1
A comparison of the Tc4þ/Tc5þ and Tc5þ/Tc6þ redox
couples in the TcVO-R1 and TcVO-R2 complexes is pre-
sented in Figure 7 and summarized in Table 6. The Tc
within TcVO-R1 is reduced at more positive potentials
than the Tc within TcVO-R2. This is consistent with the
trend observed for complexes incorporating first row
transition metals.14,16,18,23 The Tc center located in the
belt region of the POM (R1) undergoes more facile
reduction than when located in the cap region (R2). This
is observed as a difference of 177 mV for the Tc5þ/Tc6þ
redox process and 142 mV for the Tc4þ/Tc5þ redox
process. The more facile reduction of TcVO-R1 relative
compound
M6þ/7þ (mV)
M5þ/6þ (mV)
M4þ/5þ (mV)
TcVO-R1
TcVO-R2
ReVO-R1
ReVO-R2
n/a
n/a
811.5
630
990
813
477.5
254
-33
-175
-295
-284
causes dissociation of the complex into a combination of
plenary (P2W18O62)6- and lacunary (P2W17O61)10- Well-
s-Dawson anions and TcO4- (as monitored by 31P and
99Tc NMR). A summary of the observed half-wave
potentials is presented in Table 5. In comparison with
TcVO-R1, the TcVO-R2 complex exhibits an extra wave at

Article
Inorganic Chemistry, Vol. 50, No. 5, 2011 1679
Figure 7. Cyclic voltammograms of TcVO-R1 (solid line) and TcVO-R2 (dashed line) in 0.1 M CH3COONa/CH3COOH, containing 0.5 M NaSO4, pH =
5.00. Working electrode, glassy carbon; auxiliary electrode, platinum wire; reference electrode, Ag/AgCl. Scan rate = 10 mV s-1
.
Table 6. Summary of the Differences in the Half-Wave Potentials between TcVO-
(P2W17O61)10- Wells-Dawson anions and ReO4 (as
judged by 31P NMR and electrochemistry). A comparison
of the various waves is presented in Table 6. The redox
processes in the belt region consistently occur at potentials
about 200 mV more positive than those of the cap region.
This is parallelto whatis observed in the Tc examples, albeit
the reduction/oxidation of the Re is decidedly more im-
pacted by the substitution, i.e, the average E1/2(ReVO-
-
R1/TcVO-R2 and ReVO-R1/ReVO-R2 Complexes and the Differences between the
Tc and Re Complexes
compound
M6þ/7þ (mV) M5þ/6þ (mV) M4þ/5þ (mV)
TcVO-R1-TcVO-R2
ReVO-R1-ReVO-R2
TcVO-R1-ReVO-R1
TcVO-R2-ReVO-R2
n/a
177
142
-11
262
109
181.5
n/a
n/a
223.5
512.5
559
R1) - E1/2(ReVO-R2) = 202 mV and the average E1/2
-
to TcVO-R2 may be due the to the Tc-W interaction in
TcVO-R1, which should stabilize the lower Tc oxidation
states. This highlights the inequivalency of the two sites,
and their impact on the chemical properties of transition
metal cations substituted in these positions; in this case
specifically, the electron transfer properties are unmis-
takably different.
(TcVO-R1) - E1/2 (TcVO-R2)=160 mV.
Upon comparison of the redox couples for ReVO-R1 and
ReVO-R2 (Table 6), a difference of 223.5 and 181.5 mV is
observed for the Re5þ/Re6þ and the Re6þ/Re7þ waves,
respectively. As with the Tc complexes, this trend is con-
sistent with stabilization of the lower oxidation states in
ReVO-R1 by formation of a Re-W interaction. This trend
is not apparent for what are assigned as Re3þ/Re4þ and
Re4þ/Re5þ couples. The difference in the position of the
redox waves differs by ca. -11 mV (vs Ag/AgCl), which
indicates that either (i) these waves are actually due to redox
Figure 8 presents a comparison of the CV traces of the
Re-substituted R1 and R2 complexes, and a summary of the
half wave potentials is presented in Table 5. A direct
assessment of the Re analogues under the same conditions
as those of the Tc analyses allows us to correlate the
relationships between the substitution position and redox
behavior of the metal centers. As with the Tc analogues,
new waves at more positive potentials than those of the
redox waves of the tungsten framework of the lacunary R1
and R2 are observed. On the basis of the observed rest
potentials and the observation of sharp, resolute lines in
the 31P NMR spectra (indicative of diamagnetic species),
we can assign the waves centered at -445 and -443 mV as
Re3þ/Re4þ redox couples, those at -295 and -284 mV as
Re4þ/Re5þ redox couples, the waves at þ477 and þ254 mV
as Re5þ/Re6þ redox processes, and the waves at þ822 and
þ627 mV as Re6þ/Re7þ redox couples (for ReVO-R1 and
ReVO-R2, respectively). The assignment of the quasi-rever-
sible waves at þ822 and þ627 mV as Re6þ/Re7þ is analo-
gous to those in the Tc complexes.
processes involving the tungsten framework and the Re3þ
Re4þ and Re4þ/Re5þ couples are eclipsed or (ii) the Re3þ
/
/
Re4þ and Re4þ/Re5þ couplesareunaffected by substitution
position. This is currently undergoing investigation using in
situ spectroelectrochemistry.
Figure 9 presents a comparison of the CVs of the family
of R2 and R1 complexes. Figure 9A shows the voltammo-
grams of TcVO-R2 (dotted line) and ReVO-R2 (solid line);
Figure 9B shows the voltammograms of TcVO-R1 (dotted
line) and ReVO-R1 (solid line). It can be seen that the
reduction of Tc(VI) to Tc(V) is more facile than that of
Re(VI) to Re(V) by 559 mV (Table 6) within the cap-
substituted R2 complexes. This trend is paralleled in the R1
complexes, but the difference between the M(VI) to M(V)
reduction is 512 mV. When comparing the differences in
potential for the reduction of M(V) to M(IV), the more
favorable process again lies with the Tc complexes, but the
ease with which this occurs compared to the Re congener is
Prolonged exposure to oxidative potentials (greater that
þ870 mV) causes the complexes to dissociate irreversibly
into a combination of plenary (P2W18O62)6- and lacunary

1680 Inorganic Chemistry, Vol. 50, No. 5, 2011
McGregor et al.
Figure 8. Cyclic voltammograms of ReVO-R1 (solid line) and ReVO-R2 (dotted line) in 0.1 M CH3COONa/CH3COOH containing 0.5 M NaSO4, pH =
5.00. Working electrode, glassy carbon; auxiliary electrode, platinum wire; reference electrode, Ag/AgCl. Scan rate = 10 mV s-1
.
Figure 9. Cyclic votammograms of Tc and Re complexes. (A) ReVO-R2 (solid line) and TcVO-R2 (dotted line). (B) ReVO-R1 (solid line) and TcVO-R1
(dotted line) in 0.1 M CH3COONa/CH3COOH containing 0.5 M NaSO4, pH = 5.00. Working electrode, glassy carbon; auxiliary electrode, platinum wire;
reference electrode, Ag/AgCl. Scan rate = 10 mV s-1
.
different between the R2 (109 mV) and the R1 (262 mV)
complexes.
problem from a molecular-level vantage point by employing
polyoxometalates to address the coordination chemistry and
redox stability of Tc within metal oxide matrices.
All of the electrochemical studies, described above, along
with multinuclear NMR and mass spectrometry clearly
confirm that the technetium center present within the
structure of the compounds TcVO-R1 and TcVO-R2 is in
the þV oxidation state. This Tc(V) center can be reduced to
Tc(IV) or oxidized to Tc(VI). Other oxidation states (þIII
and þVII) have been identified in other electrolytes, and
these studies will be published elsewhere.
In this study, we report the synthesis and physical chemical
properties of TcVO incorporated into the lacunary R1 and R2
polyoxometalates. These syntheses rely on the use of a
“transfer ligand”, ethylene glycol, which facilitates the stabi-
lization of the TcVO core as the complexes are formed. The
complexes were characterized by multinuclear NMR spec-
troscopy, X-ray spectroscopy, and electrochemistry. To-
gether, these techniques provide compelling evidence that
the Tc(V) is incorporated in the R1 and R2 frameworks.
Transition metal cations substituted into the R1 and R2
positions clearly show differences in electrochemistry; speci-
fically, transition metal cations substituted into the R1 posi-
tion are more readily reduced than when substituted into the
R2 position.14,16,18,23 These differences may be explained by
the differences in the local structures of Tc in these complexes
Conclusion
Understanding and controlling the extensive redox chem-
istry of 99Tc is one of the obstacles in identifying separation
strategies, and the identification and development of appro-
priate waste forms for containing 99Tc once separated from
nuclear fuel and waste tanks. We are approaching this

Article
Inorganic Chemistry, Vol. 50, No. 5, 2011 1681
Figure 10. Representations of the defect sites of R2 and R1. The tungsten atoms that create the defect are shown as balls. The tungsten atoms not closely
associated with the defect site are represented as octahedra. The oxygen atoms bound to the W atoms of the defect are shown in green. All other oxygen
atoms bound to W atoms are shown in red. The P atoms are yellow. The oxygen atom bound to the phosphorus of the defect is shown in off-white. In the R2
site, this oxygen atom is bound to two W atoms and the P atom; in the R1 site, this oxygen atom is bound to one W atom and the P atom.
as determined by EXAFS. The ease of reduction for metal
ions coordinated in the R1 position is consistent with the
formation of a dative metal-metal bond between Tc(V) and
the nearest tungsten atom in the lacunary Wells-Dawson
ion. The defects of R1 and R2 are shown in Figure 10. The
basic oxygen of the R1 defect (gray in Figure 10) is bound to
the phosphorus atom and one tungsten atom; in contrast, the
corresponding oxygen of the R2 defect is bound to the one
phosphorus atom and two tungsten atoms. The oxygen of the
R1 site would thus favor stronger interaction of the transition
metal with the W atoms in thebelt. Theoreticaltreatments are
consistent with these findings; the LUMO in the frontier
orbitals of the parent (R-P2W18O61)6- (a001) consists of 96%
R1 character.66
This reduction behavior is illustrated in this study wherein
TcVO is substituted into the R1 and R2 defects. The reduction
of Tc(V) to Tc(IV) indeed is facilitated in the TcVO-R1 isomer
compared to the TcVO-R2 isomer (þ157 mV). To evaluate
and quantify the influence due to the specific isomers, we
must operate under conditions where the influence of pro-
tonation is negligible, i.e., at pH equal to or greater than 7. A
full electrochemical study as a function of pH and compar-
ison with DFT calculations to compare the energies of TcVO-
R1 and TcVO-R2 will be forthcoming.
Portions of this research were carried out at the Stanford
Synchrotron Radiation Laboratory, a national user facil-
ity operated by Stanford University on behalf of the U.S.
DOE, Office of Basic Energy Sciences. Use of the Ad-
vanced Photon Source at Argonne was supported by the
U.S. Department of Energy, Office of Science, Office of
Basic Energy Sciences, under Contract No. DE-AC02-
06CH11357. We are grateful to the NSF (Grant Nos.
CHE 0414218 and CHE 0750118) for the research per-
formed at Hunter College. We are also grateful to DE-
FG02-09ER16097 (Heavy Element Chemistry, Office of
Science, Department of Energy) for support of this work.
The research Infrastructure at Hunter College is partially
supported by NIH-Research Centers in Minority Institu-
tions Grant RR03037-08.
Supporting Information Available: Figures showing the col-
umn purification of ReVO-R1 (Figure S1); infrared spectra of
TcVO-R1 and TcVO-R2 (Figure S2); mass spectra for TcVO-R1
and TcVO-R2 (Figure S3); 183W NMR for an independent synthe-
sis of TcVO-R1 (Figure S4); EXAFS data for TcVO-R2 taken at
the Advanced Photon Source, Argonne National Laboratory
(Figure S5). Also included are the CVs of TcVO-R1 and Fe-R1
(Figure S6). Table S1 shows the best fit parameters for TcVO-R2
taken at the APS at Argonne National Laboratory. This material
is available free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. Part of this work was performed at
Lawrence Berkeley National Laboratory and was
supported by the Director, Office of Science, Office of
Basic Energy Sciences of the U.S. Department of Energy
(DOE) under Contract No. DE-AC02-05CH11231.
Note Added after ASAP Publication. This paper was
published on the Web on January 26, 2011, with the
incorrect number of half-life years in the first line of
the Abstract. The corrected version was reposted on
February 2, 2011.
(66) Lopez, X.; Bo, C.; Poblet, J. M. J. Am. Chem. Soc. 2002, 124, 12574.