Actinide redox-active ligand complexes: Reversible intramolecular electron-transfer in U(dpp-BIAN)2/U(dpp-BIAN)2(THF)

Article abstract of DOI:10.1021/ic901636f

Actinide complexes of the redox-active ligand (dpp-BIAN)^2- (dpp-BIAN = 1, 2-bis(2, 6-diisopropylphenylimino)acenaphthylene), U(dpp-BIAN)₂ (1), U(dpp-BIAN)₂(THF) (1-THF), and Th(dpp-BIAN)₂(THF) (2-THF), have been prepared. Solid-state magnetic and single-crystal X-ray data for complex 1 indicate a ground-state U ^IV-π*³ configuration, whereas a (dpp-BIAN) ^2-? to-uranium electron transfer occurs for 1-THF, resulting in a U^III-π*³ ground configuration. The solidstate magnetic data also indicate that interconversion between the two forms of the complex is possible, limited only by the ability of tetrahydrofuran (THF) vapor to penetrate the solid upon cooling of the sample. In contrast to those in the solid state, spectroscopic data acquired in THF indicate only the presence of the U^IV -π*⁴ form for 1-THF in solution, evidenced by electronic absorption spectra and by measurement of the solution magnetic moment in THF-O8 using the Evans method. Also reported is the electrochemistry of the complexes collected in CH₂CI₂, CF₃C₆H₅, and THF. As expected from the solution spectroscopic data, only small differences are observed in half-wave potentials of ligandbased processes in the presence of THF, consistent with the solution U^IV-π*⁴ configuration of the complexes in all cases. Density functional theory calculations were undertaken for complexes 1 and 1-THF to determine if intrinsic energetic or structural factors underlie the observed charge-transfer process. While the calculated optimized geometries agree well with experimental results, it was not possible to arrive at a convergent solution for 1 -THF in the U^III-*³ configuration. However, perturbations in the orbital energies in 1 versus 1-THF for the U^IV-π*⁴ configuration do point to a diminished highest occupied molecular orbital-lowest unoccupied molecular orbital energy gap in 1-THF, consistent with the solidstate magnetic data. These results represent the first example of a stable and well-defined, reversible intramolecular electron transfer in an actinide complex with redox-active ligands.

Full text of DOI:10.1021/ic901636f

924 Inorg. Chem. 2010, 49, 924–933
DOI: 10.1021/ic901636f
Actinide Redox-Active Ligand Complexes: Reversible Intramolecular
Electron-Transfer in U(dpp-BIAN)2/U(dpp-BIAN)2(THF)
Eric J. Schelter, Ruilian Wu, Brian L. Scott, Joe D. Thompson, Thibault Cantat, Kevin D. John, Enrique R. Batista,
David E. Morris, and Jaqueline L. Kiplinger*
Los Alamos National Laboratory, Los Alamos, New Mexico 87545
Received August 14, 2009
Actinide complexes of the redox-active ligand (dpp-BIAN)2- (dpp-BIAN = 1,2-bis(2,6-diisopropylphenylimino)acena-
phthylene), U(dpp-BIAN)2 (1), U(dpp-BIAN)2(THF) (1-THF), and Th(dpp-BIAN)2(THF) (2-THF), have been prepared.
Solid-state magnetic and single-crystal X-ray data for complex 1 indicate a ground-state UIV-π*4 configuration, whereas
a (dpp-BIAN)2--to-uranium electron transfer occurs for 1-THF, resulting in a UIII-π*3 ground configuration. The solid-
state magnetic data also indicate that interconversion between the two forms of the complex is possible, limited only by
the ability of tetrahydrofuran (THF) vapor to penetrate the solid upon cooling of the sample. In contrast to those
in the solid state, spectroscopic data acquired in THF indicate only the presence of the UIV-π*4 form for 1-THF in
solution, evidenced by electronic absorption spectra and by measurement of the solution magnetic moment in THF-d8
using the Evans method. Also reported is the electrochemistry of the complexes collected in CH2Cl2, CF3C6H5, and THF.
As expected from the solution spectroscopic data, only small differences are observed in half-wave potentials of ligand-
basedprocesses in the presence of THF, consistentwiththesolutionUIV-π*4 configurationofthecomplexesinallcases.
Density functional theory calculations were undertaken for complexes 1 and 1-THF to determine if intrinsic energetic
or structural factors underlie the observed charge-transfer process. While the calculated optimized geometries agree
well withexperimentalresults, it was not possibletoarriveata convergent solutionfor1-THFinthe UIII-π*3 configuration.
However, perturbations in the orbital energies in 1 versus 1-THF for the UIV-π*4 configuration do point to a diminished
highest occupied molecular orbital-lowest unoccupied molecular orbital energy gap in 1-THF, consistent with the solid-
state magnetic data. These results represent the first example of a stable and well-defined, reversible intramolecular
electron transfer in an actinide complex with redox-active ligands.
Introduction
spaced electronic states.6,7 In the context of their electronic
properties, we have recently become interested in 5f-element
complexes of redox-active ligands and how ligand “noninno-
cence” might manifest in these complexes in which the metal-
orbital participation in bonding is often considerably smaller
than that observed in transition metal complexes.
Reports of f-element complexes supported by classical
redox-active ligands such as catecholates have appeared;
however, the focus of these studies has generally been on
their structural features rather than their physicochemical
properties.8-12 Considerably more effort has been devoted
Metal complexes possessing redox-active ligands constitute
a rapidly growing area of interest in inorganic and organo-
metallic chemistry.1-4 Redox-active ligands display reversible
redox changes between two or more stable oxidation states
that introduce the possibility for electron-transfer processes
between the metal ion and ligand.5 The electronic structures
of transition metal complexes exhibiting intramolecular elec-
tron transfer often display interesting phenomena including
thermal, pressure, or photoinduced valence changes (i.e.,
valence tautomerism) as a consequence of their closely
(7) Sato, O.; Cui, A.; Matsuda, R.; Tao, J.; Hayami, S. Acc. Chem. Res.
2007, 40, 361–369.
*To whom correspondence should be addressed. E-mail: kiplinger@
lanl.gov.
(8) Xu, J.; Gorden, A. E. V.; Raymond, K. N. Eur. J. Org. Chem. 2004,
3244–3253.
(9) Belkhiri, L.; Arliguie, T.; Thuery, P.; Fourmigue, M.; Boucekkine, A.;
Ephritikhine, M. Organometallics 2006, 25, 2782–2795.
(10) Hitchcock, P. B.; Huang, Q.; Lappert, M. F.; Wei, X.-H.; Zhou, M.
Dalton Trans. 2006, 2991–2997.
(11) Weis, E. M.; Barnes, C. L.; Duval, P. B. Inorg. Chem. 2006, 45,
10126–10130.
(12) Zhu, D.; Kappel, M. J.; Raymond, K. N. Inorg. Chim. Acta 1988,
147, 115–121.
(1) Vasudevan, K.; Cowley, A. H. Chem. Commun. 2007, 3464–3466.
(2) Blackmore, K. J.; Lal, N.; Ziller, J. W.; Heyduk, A. F. J. Am. Chem.
Soc. 2008, 130, 2728–2729.
(3) Trovitch, R. J.; Lobkovsky, E.; Bill, E.; Chirik, P. J. Organometallics
2008, 27, 1470–1478.
(4) Lu, C. C.; Bill, E.; Weyhermueller, T.; Bothe, E.; Wieghardt, K. J. Am.
Chem. Soc. 2008, 130, 3181–3197.
(5) Ray, K.; Petrenko, T.; Wieghardt, K.; Neese, F. Dalton Trans. 2007,
1552–1566.
(6) Hendrickson, D. N.; Pierpont, C. G. Top. Curr. Chem. 2004, 234, 63–95.
r
pubs.acs.org/IC
Published on Web 12/29/2009
2009 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 925
to ytterbium diimine complexes, which present redox
activity derived from the readily accessible YbII/YbIII redox
couple; YbIII reduction in aqueous solution versus SHE
is -1.15 V.13-15 Recent reports have also included the
observation of temperature-dependent redox isomerism in
solution for a ytterbium dpp-BIAN complex.16 In all cases,
these electronic structure effects are ascribed to steric pres-
sure in the ytterbium coordination sphere,15 a clear manifes-
tation of the sterically induced reduction concept developed
by Evans.17 Our work has also shown that energetically
proximate charge-transfer electronic excited states, derived
from intramolecular electron-transfer processes, mix with
lower-lying f-orbital-based ligand-field states in uranium
complexes, resulting in significant intensity enhancement of
the f-f transitions.18-24 In light of these observations, and of
our interest in probing the effects of ligand redox activities
when coordinated to f-element metal ions, we initiated an
investigation of actinide complexes of the redox-active li-
gand, dpp-BIAN = 1,2-bis(2,6-diisopropylphenylimino)-
acenaphthylene.
The redox activity of dpp-BIAN is provided primarily
through its R-diimine backbone and secondarily by its
acenaphthylene framework; it can accept up to four electrons
and be conveniently isolated as the alkali metal salts,
[Na]1-4[dpp-BIAN].25 Various dpp-BIAN salts and metal
complexes have been reported.26,27 A report of lanthanide-
BIAN complexes by Cowley highlights the ability of this
ligand to form neutral, mono-, and dianionic complexes by
the formation of (C5Me5)xLn(BIAN)(THF)y complexes of
Ln=EuII, x=2, y=0; YbIII, x=2, y=0; and SmIII, x=1, y =1
(Scheme 1).1
Scheme 1. Lanthanide-BIAN Complexes Reported by Cowley1
Showing the Formation of Complexes with the BIAN Ligand in the
Neutral, Monoanionic, and Dianionic Configurations
Results and Discussion
Structural Chemistry. The divalent salt of the dpp-
BIAN ligand, Na2dpp-BIAN, can be conveniently
prepared in situ from the comproportionation reac-
tion of Na4[dpp-BIAN] (prepared from a heterogeneous
reaction of dpp-BIAN with excess sodium metal) with
1 equiv of neutral dpp-BIAN.25 Subsequent reaction
with UCl4, workup, and crystallization from hexanes/
THF produced the intense purple U(dpp-BIAN)2(THF)
(1-THF) in 59% isolated yield (Scheme 2). A similar
reaction with ThBr4(THF)4 and workup afforded the
related purple thorium complex Th(dpp-BIAN)2(THF)
(2-THF) in 49% isolated yield. Heating crystalline
solid 1-THF to 125 °C under an inert atmosphere
followed by successive evacuation and backfilling of
the flask with argon gas produced the THF-free form
of the complex, U(dpp-BIAN)2 (1), in quantitative
yield. X-ray-quality single crystals of 1 were similarly
grown from slow evaporation of a concentrated hexane
solution.
The solid-state molecular structures of 1 and 1-THF are
shown in Figure 1 and reveal that the complexes contain
tetra- and pentacoordinate uranium ions, respectively,
coordinated by two bidentate dpp-BIAN ligands. The
molecular structure of complex 2-THF is presented in
Figure 2 and is isostructural with that of 1-THF. The
metal ion in 1 adopts a distorted tetrahedral geometry,
whereas 1-THF and 2-THF adopt distorted square pyr-
amidal geometries, as observed in the N-U-N angles
at the uranium ions: 1, N(1)-U(1)-N(3) = 122.8(3)°,
N(2)-U(1)-N(4)= 120.8(3)°; 1-THF, N(1)-U(1)-
N(4) = 127.27(9)°, N(2)-U(1)-N(3) =150.18(9)°. Both
complexes 1 and 2-THF crystallize with two independent
molecules in their asymmetric units; the U-N bonds
We have found that dpp-BIAN also exhibits extensive
redox activity in its complexes with the 5f elements. Herein
we describe our investigations and report the synthesis and
characterization of the first examples of dpp-BIAN uranium
and thorium complexes, U(dpp-BIAN)2 (1), U(dpp-BIAN)2-
(THF) (1-THF), and Th(dpp-BIAN)2(THF) (2-THF), and
show that the uranium system undergoes a facile and rever-
sible metal ion redox change in the solid state.
(13) Berg, D. J.; Boncella, J. M.; Andersen, R. A. Organometallics 2002,
21, 4622–4631.
(14) Kuehl, C. J.; Da Re, R. E.; Scott, B. L.; Morris, D. E.; John, K. D.
Chem. Commun. 2003, 2336–2337.
(15) Trifonov, A. A. Eur. J. Inorg. Chem. 2007, 3151–3167.
(16) Fedushkin, I. L.; Maslova, O. V.; Baranov, E. V.; Shavyrin, A. S.
Inorg. Chem. 2009, 48, 2355–2357.
˚
in 1 range from 2.244(7) to 2.283(7) A, and the U-N
˚
bonds in 1-THF range from 2.273(3) to 2.304(2) A. Little
(17) Evans, W. J.; Davis, B. L. Chem. Rev. 2002, 102, 2119–2136.
(18) Da Re, R. E.; Jantunen, K. C.; Golden, J. T.; Kiplinger, J. L.; Morris,
D. E. J. Am. Chem. Soc. 2005, 127, 682–689.
(19) Morris, D. E.; Da Re, R. E.; Jantunen, K. C.; Castro-Rodriguez, I.;
Kiplinger, J. L. Organometallics 2004, 23, 5142–5153.
(20) Schelter, E. J.; Yang, P.; Scott, B. L.; Thompson, J. D.; Martin, R. L.;
Hay, P. J.; Morris, D. E.; Kiplinger, J. L. Inorg. Chem. 2007, 46, 7477–7488.
(21) Graves, C. R.; Scott, B. L.; Morris, D. E.; Kiplinger, J. L. J. Am.
Chem. Soc. 2007, 129, 11914–11915.
(22) Graves, C. R.; Yang, P.; Kozimor, S. A.; Vaughn, A. E.; Clark, D. L.;
Conradson, S. D.; Schelter, E. J.; Scott, B. L.; Thompson, J. D.; Hay, P. J.;
Morris, D. E.; Kiplinger, J. L. J. Am. Chem. Soc. 2008, 130, 5272–5285.
(23) Graves, C. R.; Scott, B. L.; Morris, D. E.; Kiplinger, J. L. Organo-
metallics 2008, 27, 3335–3337.
(24) Graves, C. R.; Vaughn, A. E.; Schelter, E. J.; Scott, B. L.; Thompson,
J. D.; Morris, D. E.; Kiplinger, J. L. Inorg. Chem. 2008, 47, 11879–11891.
(25) Fedushkin, I. L.; Skatova, A. A.; Chudakova, V. A.; Fukin, G. K.
Angew. Chem., Int. Ed. 2003, 42, 3294–3298.
(26) Schumann, H.; Hummert, M.; Lukoyanov, A. N.; Chudakova,
V. A.; Fedushkin, I. L. Z. Naturforsch., B: Chem. Sci. 2007, 62, 1107–1111.
(27) Schumann, H.; Hummert, M.; Lukoyanov, A. N.; Fedushkin, I. L.
Chem.;Eur. J. 2007, 13, 4216–4222.
structural variation wasfound between the R-diimine back-
bonesof the dpp-BIAN ligands among all of the complexes.
˚
The An-OTHF distance is 2.541(2) A in 1-THF and
˚
2.556(5) and 2.574(5) A in 2-THF. The Th-N distances
in 2-THF are consistent with expectations defined by the
U-N distances in 1. These Th-N distances range from
2.284(6) to 2.370(6) A; the lengthening of these distances is
˚
IV
˚
consistent with the ∼0.05 A size difference between Th
and UIV ions.28
Description of the Solid-State Electronic Structures of 1,
1-THF, and 2-THF. The electronic structures of the
crystalline forms of 1, 1-THF, and 2-THF were evaluated
with magnetic susceptibility studies. These results, as well
as those obtained in solution, are described below and
summarized in Scheme 3.
(28) Shannon, R. D. Acta Crystallogr. 1976, A32, 751–767.

926 Inorganic Chemistry, Vol. 49, No. 3, 2010
Schelter et al.
Figure 1. Molecular structures of 1 (left) and 1-THF (right) with ellipsoids projected at the 50% probability level. Hydrogen atoms are omitted and
˚
diisopropyl-aryl group carbon atoms are depicted as spheres of arbitrary radius for clarity. Selected bond distances (A) for 1: U(1)-N(1), 2.252(7);
U(1)-N(2), 2.260(7); U(1)-N(3), 2.283(7); U(1)-N(4), 2.247(7); N(1)-C(1), 1.415(11); N(2)-C(2), 1.395(10); N(3)-C(37), 1.378(11); N(4)-C(41),
1.407(10); C(1)-C(2), 1.410(12); C(37)-C(41), 1.412(12). For 1-THF: U(1)-N(1), 2.273(3); U(1)-N(2), 2.291(2); U(1)-N(3), 2.280(2); U(1)-N(4),
2.304(2); N(1)-C(13), 1.405(4); N(2)-C(14), 1.396(4); N(3)-C(49), 1.393(4); N(4)-C(50), 1.404(4); C(13)-C(14), 1.407(4); C(49)-C(50), 1.400(4).
Scheme 2. Synthesis of the dpp-BIAN Uranium and Thorium Complexes 1, 1-THF, and 2-THF
The χT product of 1 exhibits typical magnetic behavior
for the UIV ion (Figure 3).20 The χT product decreases
gradually from 300 to 100 K, beginning at 0.96 emu K
mol-1 (2.77 μB), then descends more quickly beginning at
100 K to attain a χT product of near zero at 2 K. This
temperature-dependent behavior is a reflection of the
Boltzmann thermal population of the nine crystal field
(CF) states for the nominal 3H4 manifold in the UIV ion,
through reduction of the χT product by quenching of the
orbital angular momentum of the J=4 ground term.29
The vanishing χT product at 2 K is consistent with a
singlet, nonmagnetic ground state for this ion in its low
symmetry crystal field. The temperature-dependent mag-
netic data for 1-THF were found to be surprisingly diffe-
rent from those of 1, as shown in Figure 3. The χT product
for 1-THF at 300 K exhibits a value of 1.71 emu K mol-1
.
Figure 2. Molecular structure of 2-THF with ellipsoids projected at the
50% probability level. Hydrogen atoms omitted and diisopropyl-aryl
group carbon atoms depicted as spheres of arbitrary radius for clarity.
Selected bond distances for 2-THF: Th(1)-N(1), 2.348(6); Th(1)-N(2),
2.338(6); Th(1)-N(3), 2.370(6); Th(1)-N(4), 2.309(6); N(1)-C(13),
1.428(9); N(2)-C(53), 1.394(8); N(3)-C(49), 1.396(9); N(4)-C(59),
1.446(9); C(13)-C(53), 1.412(10); C(49)-C(59), 1.400(10); Th(1)-O(1),
2.556(5).
Given the value of 0.96 emu K mol-1 for 1, the identical
CF symmetry between 1 and 1-THF, and the predicted
maximum value of 1.6 emu K mol-1 for the UIV ion, it is
not possible that the χT product for 1-THF is due to a
single UIV ion. The temperature-dependent character of
χT versus T for 1-THF (vide infra) is also dissimilar to
that for 1; the χT product of 1-THF remains fairly
constant up to ∼50 K, where it decreases precipitously,
reaching a small finite value of 0.21 emu K mol-1 at 2 K.
(29) Boudreaux, E. A.; Mulay, L. N. Theory and Applications of Molec-
ular Paramagnetism; John Wiley & Sons: New York, 1976.

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 927
Figure 3. χT versus T data for 1 (blue trace) and 1-THF (red trace) from 2 to 300 K (left) and field-dependent magnetization data (right) for 1 and 1-THF.
Scheme 3. Summary of the Electronic Configurations Determined for the 1/1-THF System under Varying Experimental Conditions
These data indicate that an electron-transfer process has
occurred with the presence of THF in the uranium coordi-
nation sphere such that the ground state for 1-THF is best
represented as UIII-π*3 (f3 plus dpp-BIAN- ligand
radical), while complex 1 exhibits a UIV-π*4 ground
state.30 It is commonly held that the use of magnetic data
is an imprecise tool for the differentiation of UIII and
UIV complexes, due to the similar predicted χT values
unequivocally preclude the assignment of both complexes
as UIV.
Evidence for a change in metal oxidation states between
complexes 1 and 1-THF is also found in their solid-state
molecular structures. As noted above, the U-N bonds
˚
are longer in 1-THF than in 1, consistent with the ∼0.14 A
larger ionic radius of the UIII ion in 1-THF.28 The
structural data also suggest that the position of the ligand
radical hole present in 1-THF is disordered over the two
ligand sites. The U-OTHF distances are usually uninfor-
for the free ions using L-S coupling (UIV, 1.60; UIII
,
1.64 emu K mol-1). Empirically, low-symmetry crystal
fields imposed by the ligands, state mixing, and partial
failure of the L-S coupling scheme always reduce the
value of the χT products from those predicted for these
ions.31 With similar CF strengths and symmetries, phe-
nomenological oxidation state assignment is possible.32,33
The difference in the χT products for 1 and 1-THF at
300 K is 0.75 emu K mol-1. This value is consistent
with an additional two unpaired electrons per complex in
1-THF compared with 1. Coupled with the essentially
identical CF effects between 1 and 1-THF, these data
mative with respect to the uranium oxidation state; values
IV
˚
˚
range from 2.42 to 2.65 A for U and 2.46 to 2.74 A for
U
III, as determined by a Cambridge Structural Database
search.34
The temperature-dependent character of the 1-THF
susceptibility data is also consistent with assignment of
UIII in this complex. Specifically, the five Kramer’s
doublet crystal field states that arise from the nominal
4I9/2 (5f3) ground term of UIII are significantly fewer than
the nine, nondegenerate crystal field levels of UIV. As
such, UIII complexes show a sharper decrease (due to five
doublets versus nine crystal field states in the absence of
magnetic field) through quenching of the orbital angular
(30) 5f3-π*3 and 5f2-π*4 correspond to UIII(dpp-BIAN•)(dpp-BIAN)
and UIV(dpp-BIAN)2 configurations, respectfully.
(31) Stewart, J. L.; Andersen, R. A. New J. Chem. 1995, 19, 587–595.
(32) Castro-Rodriguez, I.; Meyer, K. Chem. Commun. 2006, 1353–1368.
(33) Schelter, E. J.; Morris, D. E.; Scott, B. L.; Thompson, J. D.;
Kiplinger, J. L. Inorg. Chem. 2007, 46, 5528–5536.
(34) Cambridge Structural Database; Cambridge University: Cambridge,
England (accessed August 2009).

928 Inorganic Chemistry, Vol. 49, No. 3, 2010
Schelter et al.
us to examine other signatures for electronic differences
between the complexes.
Description of the Electronic Structures of 1 and 1-THF
in Solution. The low symmetry of the complexes in the
solid state creates the expectation of a localized electron
hole on one of the two dpp-BIANn- ligands in 1-THF. A
common molecular orbital between the π* NCCN back-
bones is not expected given the orthogonality of the
N(1)-U(1)-N(2) plane to the N(3)-U(1)-N(4) plane
in complex 1-THF as is observed in the coordination
angles at U(1) that describe its pseudotetrahedral coordi-
nation: N(1)-U(1)-N(4) = 127.27(9)° and N(2)-U(1)-
N(3)=150.18(9)°. The complexes were investigated in this
context using electrochemistry in an effort to detect
nominal ligand- and metal-based redox processes. Cyclic
and square-wave voltammetry experiments were per-
formed on complexes 1 and 1-THF, as well as the baseline
complex 2-THF and neutral dpp-BIAN, in order to
elucidate the origins of the redox waves. Measurements
were performed on ∼1-2 mM solutions of complexes in
THF, dichloromethane, or R,R,R-trifluorotoluene (TFT)
using a 0.1 M [Bu4N][B(3,5-(CF3)2-C6H3)4] or [Bu4N]-
[B(C6F5)4] supporting electrolyte.20,35,36 The numerical
results of these studies, reported versus an internal ferro-
cenium/ferrocene standard, are provided as a table in the
Supporting Information.
Figure 4. Elevated temperature scans of χT versus T for 1-THF showing
conversion of the complex to 1 with the associated change in oxidation
state of the metal ion, and its hysteretic, partial recovery. The red trace
shows the first temperature scan, while the blue trace shows the second
complete scan after the complex recovers ∼17% of its liberated THF.
Sample aging, with concomitant loss of THF, accounts for a small
reduction of the χT signal prior to onset of wholesale desolvation
compared to Figure 3. Incomplete recovery is attributed to low partial
pressure of THF in the sealed sample container.
All of the complexes, as well as the free ligand, exhibit
varying degrees of decomposition in the electrolyte/sol-
vent mixtures on the time scale of the voltammetric
studies, as evidenced by the appearance of new waves
having lower currents than found for the starting ana-
lytes. Despite the instability, we were able to generalize
the appearance of reversible or quasi-reversible redox
processes for the complexes. For the neutral dpp-BIAN
ligand, in either THF or TFT, a reversible one-electron
reduction wave is observed at -2.12 V, which is assigned
to the reduction of the R-diimine backbone of the dpp-
BIAN ligand to generate dpp-BIAN- (π*1). A second,
quasi-reversible reduction is also observed for the ligand
at -2.56 V (π*2). On complexation to either uranium or
thorium ions, two reversible or quasi-reversible waves are
evident for the complexed dpp-BIAN2- ligands. Assign-
ment of these waves to ligand-based processes in all cases
can be made from comparison of the data for the d0f0
thorium complex 2-THF, which is not expected to exhibit
metal-based redox processes. The first primary redox
process of the complexes, a ligand-based oxidation, oc-
curs between -0.37 and -0.48 V, depending on solvent
and electrolyte identity. The reversibility of this ligand
oxidation is also dependent on experimental conditions.
The second primary redox process occurs within a wider
window between complexes, showing a range of -2.31
to -2.55 V and a greater sensitivity to experimental
conditions. The presence of THF, either in the starting
complex or as the solvent, has little effect on the observed
half-wave potentials.
momentum with decreasing temperature. This difference
is present between 1 and 1-THF; χT versus T for 1 begins
to decrease gradually at ∼150 K, while χT versus T for 1-
THF decreases precipitously at ∼50 K. Field-dependent
magnetization data collected for the complexes at 2 K are
also presented in Figure 3. While neither complex achieves
saturation behavior to the experimental limit of 6.5 T, the
1-THF complex clearly shows a larger response compared
to 1 over the field range probed, also supportive of a
UIII-π*3 configuration.32 The thorium congener 2-THF
was found to be diamagnetic over the temperature range
of these experiments, as expected for the ostensible f0-π*4
(ThIV-π*4) electronic configuration.
To further probe the origin and nature of the charge-
transfer ground state in 1-THF, the magnetic response of
this complex was evaluated in an expanded temperature
range. At T ∼ 360 K (Figure 3), the χT product of 1-THF
began to decrease with the onset of THF loss from the
complex (consistent with the temperature employed to
synthesize and isolate 1 from 1-THF, as noted above).
That is, coordinated THF was thermally liberated in this
temperature regime to effect the back-electron transfer
to generate complex 1. The drop in χT with its midpoint
of ∼380 K clearly illustrates this effect. As the sample was
self-contained in a sealed tube (see the Experimental
Section), we were also able to observe slight hysteresis
with a second temperature scan as the volatilized THF
was reincorporated into the structure to regenerate the
charge-transfer state of 1-THF (Figure 4), presumably
with initiation of a solid-vapor equilibrium. The partial
reformation of 1-THF was estimated at ∼17%, as judged
by the recovery of the χT product. The assignment of the
UIV-π*4 and UIII-π*3 ground configurations in the
crystalline forms of 1 and 1-THF, respectively, prompted
The lack of readily identifiable metal-based redox pro-
cesses for the uranium complexes 1 and 1-THF under
any experimental conditions means that the voltam-
metric data provide no additional insight into the metal
(35) Barriere, F.; Geiger, W. E. J. Am. Chem. Soc. 2006, 128, 3980–3989.
(36) Ohrenberg, C.; Geiger, W. E. Inorg. Chem. 2000, 39, 2948–2950.

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 929
Figure 5. UV-visible, near-IR electronic absorption data for 1 (blue trace), 1-THF (red trace), and 2-THF (purple trace) in the π-π* (left) and f-f (right)
transition energy regions.
oxidation state or the possibility for an oxidation-state
transition upon solvent coordination, as seen for the solid-
state complexes. This apparent lack of metal-based redox
processes in 1 and 1-THF is noteworthy but not without
precedent. In our previous studies on multimetallic
complexes derived from the uranium(IV) ketimide com-
plex (C5Me5)2U[-NdC(Bn)(tpy)]2 (where Bn = CH2-
C6H5, tpy = terpyridyl), UIV-based redox chemistry is
seen for the monometallic precursor, but when additional
(C5Me4Et)2UIII moieties are ligated by the two tpy func-
tional groups, evidence for redox chemistry associated
with the bridging UIV ion is no longer observed.37,38
Since the electrochemistry of the complexes revealed no
significant differences between 1 and 1-THF, in the presence
or absence of THF, further investigations of the behavior of
the complexes in solution were pursued. Solution 1H NMR
solution. The identity of 1-THF as a uranium(IV) complex in
solution is consistent with the results of the electrochemical
experiments for the ligand-based processes, which showed
little difference between measurements collected across the
various solvents and electrolyte combinations. These results
support an expected subtle difference between redox states
for 1-THF, which exists in a UIII-π*3 configuration in the
solid state but a UIV-π*4 configuration in solution
(Scheme 3).
Solution UV-vis-NIR measurements for complexes 1,
1-THF, and 2-THF collected in toluene and THF are
presented in Figure 5. The spectroscopic features for these
species are differentiated into high- and low-energy regimes
between 12 000 and 25000 and 4500-12 000 cm-1, respec-
tively. Above 12 000 cm-1, all of the complexes exhibit a
broad, intense feature. Complex 2-THF exhibits this transi-
tion centered at 17 800 cm-1. The absence of valence d or f
electrons in the thorium complex 2-THF, as well as its
broad, intense character, affords assignment of this transi-
tion as ligand-based π-π* (or π*-π*), an assignment
that can be extended to the analogous transitions in 1 and
1-THF. Complex 1 shows this transition at 18 900 cm-1
with a higher-energy shoulder at ∼20 400 cm-1, while
this primary transition in complex 1-THF is shifted to
18 300 cm-1, and the shoulder is no longer evident. A clean
isosbestic point appears at 17 100 cm-1 on titration of the
solutions of 1 with THF, showing the conversion of 1 to
1-THF (see the Supporting Information), although non-
linearity in the concentration dependence of the data pre-
cludes the derivation of an equilibrium constant for the
formation of the solvent-coordinated complex. The rela-
tively small shift of ∼550 cm-1 of the primaryband between
complexes 1 and 1-THF is consistent with a reorganiza-
tion of the coordination sphere geometry upon the coordi-
nation of THF40-44 and agrees with the orbital energy
1
provided several important insights. The H NMR reso-
nances for 1 and 1-THF, collected in benzene-d6 and tetra-
hydrofuran-d8, respectively, display the typical paramagneti-
cally broadened and shifted signals observed for trivalent and
tetravalent uranium coordination complexes. The range of
signals for the two complexes is remarkably similar: 1displays
resonances from δ þ40 to -45 ppm, while the resonances for
1-THF range from δ þ37 to -43 ppm. Both complexes
possess C2 symmetry in solution. The solution magnetic
moment of 1-THF was determined in tetrahydrofuran-d8
using the Evans method to have a χT product of 0.97 emu
K mol-1 (2.79 μB).39 (The solution magnetic moment of 1
could not be measured reliably by this method because of its
low solubility in both benzene-d6 and toluene-d8.) Compar-
ison with the temperature-dependent magnetic data for
crystalline 1-THF (1.71 emu K mol-1) and 1 (0.96 emu
K mol-1) at ambient temperature reveals that 1-THF clearly
exhibits a UIV-π*4 ground-state electronic configuration in
(37) Schelter, E. J.; Veauthier, J. M.; Thompson, J. D.; Scott, B. L.; John,
K. D.; Morris, D. E.; Kiplinger, J. L. J. Am. Chem. Soc. 2006, 128, 2198–
2199.
(38) Schelter, E. J.; Wu, R.; Scott, B. L.; Thompson, J. D.; Morris, D. E.;
Kiplinger, J. L. Angew. Chem., Int. Ed. 2008, 47, 2993–2996.
(39) Naklicki, M. L.; White, C. A.; Plante, L. L.; Evans, C. E. B.;
Crutchley, R. J. Inorg. Chem. 1998, 37, 1880–1885.
(40) Malek, C. K.; Krupa, J. C.; Genet, M. Spectrochim. Acta, Part A
1986, 42A, 907–912.
(41) Van Der Sluys, W. G.; Berg, J. M.; Barnhardt, D.; Sauer, N. N. Inorg.
Chim. Acta 1993, 204, 251–256.
(42) Fielden, J.; Long, D.-L.; Cronin, L. Chem. Commun. 2004, 2156–
2157.

930 Inorganic Chemistry, Vol. 49, No. 3, 2010
Schelter et al.
predictions from geometry-optimized DFT calculations
(vide infra).
for 1-THF, as shown in Scheme 3. Recent results have
also demonstrated the ability of a Yb-(dpp-BIAN)
complex to show temperature-dependent valence changes
in solution.16 In addition, a self-contained two-electron
oxidation state change has been reported for a GeII
complex of the redox-active tetraphenylporphyrin2-
(TPP) ligand, which converts to the GeIV upon the
coordination of pyridine.50 The oxidation state change
in this case results from the conversion of the aromatic
GeII(TPP) complex to the antiaromatic GeIV(TPP)(py)2
on the coordination of pyridine. The accessibility of the
two-electron charge-transfer state in GeIV(TPP)(py)2
clearly regulates this conversion. Together with the 1/1-
THF system, this body of work indicates significant
electronic flexibility of the f-element redox couples in
complexes supported by redox-active ligand sets. More
specifically, the redox change at the uranium metal center
upon conversion of 1-THF to 1 with release of the THF
ligand is likely due to stabilization and relaxation of the
coordination sphere, specifically the redox-active π* or-
bitals of the R-diimine backbone of the dpp-BIANn-
ligands.
Compared to the higher-energy bands for 1 and 1-
THF, the transitions appearing between 4500 and 12 000
cm-1 are significantly less intense and narrower, consis-
tent with their assignment as intraconfigurational f-f
bands. Although there is some variation between the
spectra of 1 and 1-THF in this region, specifically the
appearance of new bands at 9330, 6270, and 5270 cm-1
,
the energy range and intensity of the f-f bands between
these complexes also indicate minimal alteration, incon-
sistent with expectations for a change in metal oxidation
state with THF coordination. Lower-energy ligand-field
transitions are not observed for 2-THF, consistent with
the d0f0 configuration of the ThIV ion. The absence of an
interconfigurational f3 f f2d1 transition in the spectrum
of 1-THF also provides supporting evidence for its assign-
ment as a UIV complex in solution. Overall, changes in the
absorption spectrum of 1 with the addition of THF are
also inconsistent with a change in the oxidation state of
the complex in solution; rather, the changes in the pre-
sence of THF are due to a relatively minor modification
of the local geometry of the metal ion by coordination of
the solvent.
Summary of the Electronic Structures of 1 and 1-THF in
Solution and the Solid State. A representation of the
electronic configurations for the 1/1-THF system under
the experimental conditions is presented in Scheme 3.
Complex 1-THF exists as a UIII-π*3 complex in the solid
state and can be converted to the UIV-π*4 complex 1
reversibly with thermal cycling. However, in both the
presence and absence of THF, the electronic configura-
tions of the complexes in solution are consistent with a
UIV-π*4 assignment. Recent reports have shown that
“charge separation” complexes result from one-electron
reduction chemistry starting from a UIII complex, which
becomes oxidized with the coordination of small mole-
cules such as CO2,45 diazomethane,46 benzophenone,47
or pyrazine.48 Of these, the bimetallic pyrazine reduc-
tion is the most relevant given its reported reversi-
bility; however, the uniqueness of the 1/1-THF system is
that the donor molecule is not involved in the electron
transfer.
DFT and CASSCF Calculations of 1 and 1-THF. The
electronic structures of 1 and 1-THF were examined
using density functional theory calculations to gain
further information about the ground states of these
complexes. No simplified model compounds were used
in the calculations. Overall, the optimized geometries
were in good agreement with the X-ray diffraction
results, with the experimentally observed lengthening
of the average U-N bond distances for 1-THF com-
pared to 1 being reproduced in the calculated structures
˚
within 0.01 A. The distinction between the ground states
for 1 and 1-THF was expected to be subtle, stemming
from different degrees of electron localization between
the metal center and one or both dpp-BIAN ligands.
Using hybrid density functional theory (B3LYP), which
corrects the overdelocalization introduced by the LDA
and GGA functionals, the calculated gas-phase electro-
nic structures of 1 and 1-THF both yield a UIV ground
state with a net spin of 2.2, which is contrary to the
experimental magnetic data for 1-THF in the solid state.
This is not surprising given the difficulties with perform-
ing electronic structure calculations on electronic bro-
ken-symmetry systems. The binding energy of the THF
ligand in complex 1 was calculated to be -2.3 kcal/mol,
indicating that it is weakly bound to the uranium metal
center and can undergo rapid exchange at room tem-
perature. The presence of free THF is evident experi-
mentally in the 1H NMR spectrum of 1-THF collected in
benzene-d6 at 297 K.
A striking aspect of the 1/1-THF system is the solution
configuration of 1-THF as UIV-π*4 (Scheme 3). A rever-
sible, THF-mediated redox reaction has been reported
for (C5Me5)Eu(DADC6F5) (DADC6F5=C6F5NC(Me)-
C(Me)NC6F5).49 The solid-state electronic configuration
of this complex is EuIII-π*1, whereas dissolution results
in the generation of EuII with a concomitant loss of the
neutral DADC6F5 ligand, similar to the results observed
Significant differences between 1 and 1-THF do arise in
the electronic structure calculations, which may support
the observed reversible charge-transfer state. An energy
level diagram with the metal orbital components is shown
in Figure 6. The unpaired spin density is located in the fσ
and fφ uranium orbitals, which are the two orbitals that
minimize the effect of the ligand field. HOMO and
HOMO-1 are orbitals based on the dpp-BIAN ligand
with more than 50% of the charge density in the R-diimine
(43) Livoreil, A.; Sauvage, J.-P.; Armaroli, N.; Balzani, V.; Flamigni, L.;
Ventura, B. J. Am. Chem. Soc. 1997, 119, 12114–12124.
(44) Sato, Y.; Ohkoshi, S.; Arai, K.; Tozawa, M.; Hashimoto, K. J. Am.
Chem. Soc. 2003, 125, 14590–14595.
(45) Castro-Rodriguez, I.; Nakai, H.; Zakharov, L. N.; Rheingold, A. L.;
Meyer, K. Science 2004, 305, 1757–1760.
(46) Lam, O. P.; Feng, P. L.; Heinemann, F. W.; O’Connor, J. M.; Meyer,
K. J. Am. Chem. Soc. 2008, 130, 2806–2816.
(47) Lam, O. P.; Anthon, C.; Heinemann, F. W.; O’Connor, J. M.; Meyer,
K. J. Am. Chem. Soc. 2008, 130, 6567–6576.
(48) Mehdoui, T.; Berthet, J.-C.; Thuery, P.; Ephritikhine, M. Eur. J.
Inorg. Chem. 2004, 1996–2000.
(49) Moore, J. A.; Cowley, A. H.; Gordon, J. C. Organometallics 2006, 25,
5207–5209.
(50) Cissell, J. A.; Vaid, T. P.; Yap, G. P. A. J. Am. Chem. Soc. 2007, 129,
7841–7847.

Article
Inorganic Chemistry, Vol. 49, No. 3, 2010 931
complex, we performed complete active space self-
consistent field (CASSCF) calculations on the structure
optimized by DFT. The active space included six elec-
trons in the top four occupied and three lowest virtual
orbitals. The isopropyl groups were removed, and corre-
lation corrections were not included, as the calculations
were performed at the limit of our computational capa-
city. Overall, the ground state obtained from CASSCF
for 1-THF was identical to the one obtained from DFT-
B3LYP. A full output of the CASSCF calculations is
included as Supporting Information. The two lowest
excited states predicted by CASSCF were found to be
2.84 and 3.74 eV higher in energy than the ground state,
respectively; however, these energies are likely overesti-
mated by the missing correlation energy. The electronic
state of the first excited state was found to have significant
contributions from several determinants; therefore, it is
difficult to give a single particle picture. The most promi-
nent determinants involve charge-transfer excitations
from the diimine ligand into the uranium 5f orbitals and
into the acenaphthylene moiety with the spin state of the
f electrons corresponding to a combination of quartet
multiplicity (determinant 106) and doublet states (deter-
minant 15). As with the DFT-B3LYP, the excited state
structure points to a complex charge-transfer picture for
1-THF, lying energetically proximate to, but unmixed
with, the 5f2 configuration.
Figure 6. Orbital level diagram obtained from hybrid DFT for com-
plexes 1 and 1-THF. The left column presentsa picture ofthe orbital metal
component, and the orbital character is shown on the right column.51
Upon coordination of a THF solvent molecule to complex 1, the
R-diimine-based orbital is destabilized while the unoccupied metal 5f
orbitalsare stabilized, facilitatingthe observedcharge-transferfor1-THF.
For clarity, only the R atoms are shown, although no simplified model
complexes were used in the electronic structure calculations.
Conclusions
We have presented a uranium complex of the redox-active
ligand dpp-BIAN that undergoes a facile and reversible
redox change upon desolvation. The complexes 1-THF and
2-THF can be easily prepared and isolated in good yields;
heating/evacuation of samples of 1-THF produced the un-
solvated complex 1 in quantitative yields. Magnetic and
structural data reveal that complex 1 exhibits a UIV-π*4
ground-state electronic configuration, while 1-THF is
UIII-π*3. Temperature cycling of 1-THF affords the libera-
tion and partial recoordination of THF to effect reversible
redox changes in the complex in the solid state. This repre-
sents a rare example of a well-defined reversible intramole-
cular electron transfer for an f-element complex and further
exemplifies the unique electronic structure aspects of ura-
nium complexes.
We postulate that the presence of THF in the coordination
sphere of 1-THF causes steric pressure on the R-diimine
backbone of the dpp-BIAN ligands, resulting in a destabili-
zation that finds an energetic minimum on electron transfer
and concomitant elongation of the UIII-N bonds in 1-THF.
This is supported by the DFT calculations and spectroscopic
data, which suggest that a charge transfer from the ligand to
the metal is more feasible for 1-THF compared to complex 1.
Although preliminary reactivity studies of 1 and 1-THF with
small molecules have shown complex decomposition in all
cases, the physicochemical studies presented here represent
an interesting look at the properties of 5f-element complexes
of redox-active ligands.
fragments and 10-20% mixed uranium 5f character.51 The
lowest unoccupied orbitals consist of the five uranium 5f
orbitals plus two π* orbitals from the dpp-BIAN ligands
(LUMOþ1 and LUMOþ2 in complex 1). Upon the co-
ordination of THF, the metal orbitals in 1-THF are stabi-
lized relative to what they are in complex 1. In particular, the
LUMOþ3 orbital, which is mostly a metal 5fδ orbital in
complex 1, is stabilized and mixes with the ligand orbital
LUMOþ1 to yield the 1-THF orbitals LUMO and
LUMOþ3, leading to a lowering in energy of up to
0.3 eV. All of the other metal 5f orbitals also are stabilized
upon the coordination of THF, but to a smaller extent.
Simultaneously, upon the coordination of THF, the
HOMO-1 orbital in 1-THF is raised in energy relative to
what it is in 1. Linear response theory was used to
calculate the lowest triplet-triplet excitations to compare
the energy of the charge-separated states in the two
molecules. For complex 1, the lowest charge-transfer
state is a UIII-π*3 configuration due to a HOMO f
LUMO charge transfer. In complex 1-THF, three charge-
transfer states were found to be lower in energy than that
in complex 1, by 0.12, 0.11, and 0.02 eV.51 Although
convergence to a ground-state UIII-π*3 configuration
was not attained for 1-THF, the calculations suggest that
a charge transfer from the ligand to the metal is more
feasible for 1-THF compared to complex 1.
To further scrutinize the results of the DFT calcula-
tions for 1-THF and to evaluate for significant admixture
of multiple configurations in the ground state of the
Experimental Section
General Synthetic Considerations. Unless otherwise noted,
reactions and manipulations were performed at 25 °C in a
recirculating Vacuum Atmospheres NEXUS inert atmosphere
(51) See Tables S4-S7 in the Supporting Information for a detailed
output of the exact orbital populations for complexes 1 and THF-1.

932 Inorganic Chemistry, Vol. 49, No. 3, 2010
Schelter et al.
(N2) drybox equipped with a 40CFM Dual Purifier NI-Train, or
using standard Schlenk line techniques. Glassware was dried
overnight at 150 °C before use. All NMR spectra were obtained
in either tetrahydrofuran-d8 or benzene-d6 using a Bruker
packed coarse porosity frit. The combined dark purple solutions
were chilled overnight at -30 °C to deposit crystalline 1-THF.
The supernatant was decanted off and the crystals washed
with bis(trimethylsilyl)ether (4 ꢀ10 mL) and dried under reduced
pressure for 10 min. Further chilling the mother liquor at -30 °C
produced two additional batches of purple crystals (combined
1
Avance 300 MHz spectrometer. Chemical shifts for H NMR
spectra were referenced to solvent impurities. Elemental ana-
lyses were performed at the University of California, Berkeley
Microanalytical Facility, on a Perkin-Elmer Series II 2400
CHNS analyzer.
Unless otherwise noted, reagents were purchased from com-
mercial suppliers and used without further purification. Celite
˚
(Aldrich), alumina (Brockman I, Aldrich), and 4 A molecular
1
yield: 0.48 g, 0.36 mmol, 59%). H NMR (tetrahydrofuran-d8,
298 K): δ 36.56 (br s, 2H, -CH(CH3)2), 26.32 (br s, 2H, -CH-
(CH3)2), 23.92 (d, J = 7.8 Hz, 2H, Ar-H), 21.28 (s, 6H, -CH-
(CH3)2), 19.07 (d, J=7.5Hz, 2H, Ar-H), 18.17 (d, J=9.0 Hz, 2H,
Ar-H), 15.64 (t, J = 7.5 Hz, 2H, Ar-H), 13.87 (s, 6H, -CH-
(CH3)2), 7.34 (s, 6H, -CH(CH3)2), 7.22 (s, 2H, Ar-H), 6.72
(d, J = 6.9 Hz, 2H, Ar-H), -0.58 (s, 6H, -CH(CH3)2), -0.63
(s, 6H, -CH(CH3)2), -1.48 (d, J = 7.8 Hz, 2H, Ar-H), -1.87
(t, J = 7.2 Hz, 2H, Ar-H), -2.50 (d, J = 6.9 Hz, 2H, Ar-H),
-14.45 (s, 6H, -CH(CH3)2), -18.36 (br s, 2H, -CH(CH3)2),
-20.01 (d, J = 3.0 Hz, 2H, Ar-H), -42.38 (s, 6H, -CH(CH3)2),
-42.48 (s, 6H, -CH(CH3)2); additional resonances for 1-THF
were not observed. 1H NMR of 1-THF (297 K) recorded in
benzene-d6 showed the presence of free THF at δ 3.90 and 1.88.
Anal. Calcd for C76H88N4OU (1311.57 g/mol): C, 69.60; H, 6.76;
N, 4.27. Found: C, 69.49; H, 6.79; N, 4.45.
sieves (Aldrich) were dried under a dynamic vacuum at 250 °C
for 48 h prior to use. Hexane (anhydrous, Aldrich), tetrahydro-
furan (anhydrous, Aldrich), toluene (anhydrous, Aldrich), and
bis(trimethylsilyl)ether (Aldrich) were dried over KH for 24 h,
passed through a column of activated alumina under nitrogen,
˚
and stored over activated 4 A molecular sieves prior to use. R,R,
R-Trifluorotoluene (anhydrous, Aldrich) was passed through a
column of activated alumina under nitrogen and stored over
˚
activated 4 A molecular sieves prior to use. Benzene-d6 (Aldrich)
and tetrahydrofuran-d8 (Cambridge Isotope Laboratories) were
˚
purified by storage over activated 4 A molecular sieves under N2
prior to use. UCl4,52 ThBr4(THF)4,53 [Bu4N][B(C6F5)4],54-57
[Bu4N][B(3,5-(CF3)2-C6H3)4],35 and 1,2-bis(2,6-diisopropylphe-
nylimino)acenaphthylene (dpp-BIAN)58 were prepared accord-
ing to literature procedures.
Synthesis of U(dpp-BIAN)2 (1). A 50 mL thick-walled Schlenk
tube equipped with a Teflon valve was charged with 1-THF
(0.10 g, 0.078 mmol). The vessel was heated to 125 °C under Ar
for 1 h, then evacuated, and backfilled, resulting in quantita-
tive conversion of 1-THF to 1. X-ray-quality single crystals of 1
were grown from slow evaporation of a concentrated hexane
solution. 1H NMR (benzene-d6, 296 K): δ 40.28 (m, 2H, -CH-
(CH3)2), 27.49 (m, 2H, -CH(CH3)2), 16.08 (d, J = 7.5 Hz, 2H,
Ar-H), 12.22 (d, J = 5.7 Hz, 2H, Ar-H), 12.09 (d, J = 4.5 Hz,
6H, -CH(CH3)2), 9.66 (d, J =7.2 Hz, 2H, Ar-H), 9.14 (two
overlapping doublets, 12H, -CH(CH3)2), 8.85 (overlapping
multiplets, 6H, -CH(CH3)2 and Ar-H), 7.30 (d, J=8.1 Hz,
2H, Ar-H), 7.04 (t, J = 7.5 Hz, 2H, Ar-H), 5.29 (t, J = 7.5 Hz,
2H, Ar-H), 5.19 (d, J = 6.6 Hz, 2H, Ar-H), 4.94 (d, J = 6.9
Hz, 2H, Ar-H), 4.40 (t, J = 7.2 Hz, 2H, Ar-H), 3.10 (d, J = 7.2
Hz, 2H, Ar-H), 1.09 (d, J=6.9 Hz, 2H, Ar-H), 0.61 (d, J=5.1
Hz, 6H, -CH(CH3)2), -0.68 (s, 6H, -CH(CH3)2), -8.07 (s, 6H,
-CH(CH3)2), -22.85 (s, 6H, -CH(CH3)2), -32.38 (s, 6H,
-CH(CH3)2), -45.09 (br s, 2H, -CH(CH3)2). Anal. Calcd for
C72H80N4U (1239.46 g/mol): C, 69.77; H, 6.51; N, 4.52. Found:
C, 69.47; H, 6.74; N, 4.85.
Caution! Depleted uranium (primarily isotope 238U) and nat-
ural thorium (232Th) are weak R-emitters with half-lives of 4.47 ꢀ
109 years and 1.41 ꢀ 1010 years, respectively; manipulations and
reactions should be carried out in monitored fume hoods or in an
inert atmosphere drybox in a radiation laboratory equipped with
R- and β-counting equipment.
Note: Complexes 1-THF, 1, and 2-THF were observed to be
unstable, even on storage in the solid state at -30 °C, noted
by the appearance of two sets of H resonances (doublets) at
1
δ ∼1.19 and 1.09 ppm (depending on the solvent). These peaks
are attributed to a form of uncomplexed dpp-BIAN. All char-
acterization measurements were performed on fresh samples
whose purity was checked by 1H NMR spectroscopy.
Synthesis of U(dpp-BIAN)2(THF) (1-THF). A 20 mL scintil-
lation vial was charged with dpp-BIAN (0.31 g, 0.62 mmol),
sodium metal (0.34 g, 14.57 mmol, 23.5 equiv), and tetrahydro-
furan (10 mL). The resulting clear, orange solution was stirred
for 16 h to produce a dark brown suspension. The brown
suspension was decanted from the remaining sodium metal
and added to a clear, orange tetrahydrofuran solution of dpp-
BIAN (0.31 g, 0.62 mmol, 1.0 equiv) in a 20 mL scintillation vial,
which initiated an immediate change to a dark orange/green
dichroic solution. This mixture was stirred for 30 min, and solid
UCl4 (0.23 g, 0.62 mmol, 0.5 equiv) was added, resulting in an
immediate color change to dark purple. The dark purple solu-
tion was stirred for 2 h and filtered through a Celite-packed
coarse porosity frit and the volatiles removed from the filtrate
under reduced pressure. The dark purple, wet residue was extr-
acted with hexanes (2 ꢀ 20 mL) and filtered through a Celite-
Synthesis of Th(dpp-BIAN)2(THF) (2-THF). The synthesis
of 2-THF from ThBr4(THF)4 (0.43 g, 0.51 mmol), dpp-BIAN
(0.51 g, 1.02 mmol, 2 equiv), and sodium metal (0.32 g, 13.70
mmol, 27 equiv) is identical to 1-THF. Yield of purple crystalline
2-THF: 0.32 g, 0.25 mmol, 49%. 1H NMR (tetrahydrofuran-d8,
298 K): δ 7.24 (m, 4H, Ar-H), 7.04 (m, 4H, Ar-H), 6.88 (t, J =
7.8 Hz, 2H, Ar-H), 6.81 (t, J=7.8 Hz, 2H, Ar-H), 6.40 (d, J =
6.9 Hz, 2H, Ar-H), 6.30 (d, J = 6.6 Hz, 2H, Ar-H), 3.63 (s, 8H,
β-H-THF), 3.48 (m, 6H, -CH(CH3)2), 3.23 (m, 2H, -CH-
(CH3)2), 1.29 (s, 8H, R-H-THF), (br d, J = 7.2 Hz, 12H, -CH-
(CH3)2), 1.13 (d, J = 6.6 Hz, 6H, -CH(CH3)2), 0.96 (two over-
lapping doublets, 12H, -CH(CH3)2), (br, d, J = 6.3 Hz, 12H,
-CH(CH3)2), 0.19 (d, J=6.0 Hz, 6H, -CH(CH3)2); additional
aryl resonances overlap with the protio impurity solvent peak.
Anal. Calcd for C76H88N4OTh (1305.58 g/mol): C, 69.92; H,
6.79; N, 4.29. Found: C, 69.69; H, 7.08; N, 4.28.
(52) Kiplinger, J. L.; Morris, D. E.; Scott, B. L.; Burns, C. J. Organome-
tallics 2002, 21, 5978–5982.
(53) Clark, D. L.; Frankcom, T. M.; Miller, M. M.; Watkin, J. G. Inorg.
Chem. 1992, 31, 1628–1633.
(54) Barriere, F.; LeSuer, R. J.; Geiger, W. E. Trends Mol. Electrochem.
2004, 413–444.
(55) LeSuer, R. J.; Buttolph, C.; Geiger, W. E. Anal. Chem. 2004, 76,
6395–6401.
(56) LeSuer, R. J.; Geiger, W. E. Angew. Chem., Int. Ed. 2000, 39, 248–
250.
(57) Nafady, A.; Costa, P. J.; Calhorda, M. J.; Geiger, W. E. J. Am. Chem.
Soc. 2006, 128, 16587–16599.
(58) Paulovicova, A.; El-Ayaan, U.; Shibayama, K.; Morita, T.; Fukuda,
Y. Eur. J. Inorg. Chem. 2001, 2641–2646.
Electronic Absorption Spectroscopy. Electronic absorption
spectral data were obtained for THF and toluene solutions of
complexes 1, 1-THF, and 2-THF over the wavelength range
280-2500 nm on a Perkin-Elmer Model Lambda 950 UV-
visible-NIR spectrophotometer. Alldata were collectedin 1 mm
path length cuvettes loaded in the Vacuum Atmospheres drybox
system described above. Samples were run at multiple concen-
trations to optimize absorbance in the UV-visible and near-
infrared, respectively. Spectral resolution was typically 2 nm in
the visible region and 4-6 nm in the near-infrared.

Article
Computational Methodology. The B3LYP hybrid density
Inorganic Chemistry, Vol. 49, No. 3, 2010 933
from quartz wool and the NMR tube were measured indepen-
dently and subtracted from the total measured signal. Diamag-
netic corrections were made with the use of Pascal’s constants.
Measurements on samples of fresh batches of the compound
were duplicated to ensure veracity of the results.
functional59 was employed to study the electronic structure of
the model complexes. Both complexes, 1 and 1-THF, were
optimized with no symmetry constraint using an unrestricted
open shell formalism. The Stuttgart RSC 1997 ECP60-62 was
employed for the uranium center, which incorporates scalar
relativistic effects and replaces 60 core electrons. The valence
electrons are represented as [8s/7p/6d/4f]; 6-31G* basis sets were
used for carbon, hydrogen, and nitrogen. For the CASSCF
calculations, the converged molecular orbitals from B3LYP
hybrid density functional theory were used when defining the
active space, and the geometry of the molecule was kept fixed at
the DFT optimized one. The same basis set and effective core
potentials were used as described above. All calculations were
carried out using the Gaussian 03 suite of codes.63
Crystallography. Crystals were mounted in a nylon cryoloop
using Paratone-N oil under an argon gas flow. The data were
collected on a Bruker D8 APEX II charge-coupled-device dif-
fractometer, with a KRYO-FLEX liquid nitrogen vapor cooling
device. The instrument was equipped with graphite monochro-
˚
matized Mo KR X-ray source (λ = 0.71073 A), with MonoCap
X-ray source optics. Hemispheres of data were collected using ω
scans. Data collection and initial indexing and cell refinement
were handled using APEX II software.64 Frame integration,
including Lorentz-polarization corrections, and final cell para-
meter calculations were carried out using SAINTþ software.65
The data were corrected for absorption using the SADABS
program.66 Decay of the reflection intensity was monitored by
analysis of the redundant frames. The structure was solved using
direct methods and difference Fourier techniques. Unless other-
wise noted, non-hydrogen atoms were refined anisotropically
and hydrogen atoms were treated as idealized contributions.
Residual electron density originating from solvent contributions
was removed using SQUEEZE/PLATON.67 Structure solution,
refinement, graphics, and creation of the publication materials
were performed using SHELXTL.68
Electrochemistry. Cyclic wave voltammetric data were ob-
tained in the Vacuum Atmospheres drybox system described
above. All data were collected using a Perkin-Elmer Princeton
Applied Research Corporation (PARC) Model 263 potentiostat
under computer control with PARC Model 270 software. All
sample solutions were∼2-3 mM in complex with a 0.1 M [Bu4N]-
[B(C6F5)4] or [Bu4N][B(3,5-(CF3)2-C6H3)4] supporting electrolyte
in THF or R,R,R-trifluorotoluene (TFT). All data were collected
with the positive-feedback IR compensation feature of the soft-
ware/potentiostat activated to ensure minimal contribution to
the voltammetric waves from uncompensated solution resistance
(typically ∼1 kΩ under the conditions employed). Solutions
were contained in PARC Model K0264 microcells consisting of
a ∼3-mm-diameter Pt disk working electrode, a Pt wire counter-
electrode, and a Agwire quasi-reference electrode. Scan rates from
20 to 5000 mV/s were employed in the cyclic voltammetry scans to
assess the chemical and electrochemical reversibility of the ob-
served redox transformations. Half-wave potentials were deter-
mined from the peak values in the square-wave voltammograms
or from the average of the cathodic and anodic peak potentials in
the reversible cyclic voltammograms. Potential calibrations were
performed at the end of each data collection cycle using the
ferrocenium/ferrocene couple as an internal standard. Electronic
absorption and cyclic voltammetric data were analyzed using
Wavemetrics IGOR Pro (version 6.04) software on a Macintosh
platform.
Acknowledgment. For financial support of this work, we
acknowledge the Division of Chemical Sciences, Office of Basic
Energy Sciences, Heavy Element Chemistry program, LANL
(Director’s and Frederick Reines PD Fellowships), the LANL
G. T. Seaborg Institute for Transactinium Science, and the
LANL Laboratory Directed Research & Development pro-
gram. Finally, we thank Drs. Stosh A. Kozimor and Rebecca
M. Chamberlin (LANL) for helpful discussions.
Supporting Information Available: Crystallographic informa-
tion in CIF format; electrochemical data for complexes 1, 1-
THF, and 2-THF; variable THF concentration UV-vis absorp-
tion spectra for isosbestic point determination for 1; tables
presenting the full orbital populations for complexes 1 and 1-
THF and a comparison of the U-N bond distances obtained for
experimental and calculated structures of complexes 1 and 1-
THF; and full output for the ground state and first excited state
CASSCF calculations of 1-THF. This material is available free
of charge via the Internet at http://pubs.acs.org.
Magnetic Susceptibility. Magnetic susceptibility data were
collected using a Quantum Design Superconducting Quantum
Interference Device (SQUID) magnetometer at 6.5 T from 2 to
400 K. The samples were sealed in a 5 mm Wilmad 505-PS NMR
tube along with a small amount of quartz wool, which held the
sample near the tube center. Contributions to the magnetization
(64) APEX II; Bruker Analytical X-Ray Systems, Inc.: Madison, WI, 1997.
(65) SAINT 7.03; Bruker Analytical X-Ray Systems, Inc.: Madison, WI,
2003.
(59) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
€
(60) Kuchle, W.; Dolg, M.; Stoll, H.; Preuss, H. J. Chem. Phys. 1994, 100,
€
€
(66) Sheldrick, G. M. SADABS 2.03; University of Gottingen: Gottingen,
7535–7542.
(61) Fuentealba, P.; Preuss, H.; Stoll, H.; Szentpaly, L. v. Chem. Phys.
Lett. 1982, 89, 418–422.
Germany, 2001.
(67) Spek, A. L. PLATON; Utrecht University: Utrecht, The Netherlands,
€
(62) Kuchle, W.; Dolg, M.; Stoll, H.; Preuss, H. Mol. Phys. 1991, 74,
2005.
(68) Sheldrick, G. M. SHELXTL 5.10; Bruker Analytical X-Ray Systems,
1245–1263.
(63) Frisch, M. J. Gaussian03; Gaussian, Inc.: Wallingford, CT, 2004.
Inc.: Madison, WI, 1997.