Hexadentate terephthalamide(bis-hydroxypyridinone) ligands for uranyl chelation: Structural and thermodynamic consequences of ligand variation (1)

Article abstract of DOI:10.1021/ja201511u

Several linear, hexa- and tetradentate ligands incorporating a combination of 2,3-dihydroxy-terephthalamide (TAM) and hydroxypyridinone-amide (HOPO) moieties have been developed as uranyl chelating agents. Crystallographic analysis of several {UO₂[TAM(HOPO)₂]}^2- complexes revealed a variable and crowded coordination geometry about the uranyl center. The TAM moiety dominates the bonding in hexadenate complexes, with linker rigidity dictating the equality of equatorial U-O bonding. Hexadentate TAM(HOPO)₂ ligands demonstrated slow binding kinetics with uranyl affinities on average 6 orders of magnitude greater than those of similarly linked bis-HOPO ligands. Study of tetradentate TAM(HOPO) ligands revealed that the high uranyl affinity stems primarily from the presence of the TAM moiety and only marginally from increased ligand denticity. Uranyl affinities of TAM(HOPO)₂ ligands were within experimental error, with TAM(o-phen-1,2-HOPO)₂ exhibiting the most consistent uranyl affinity at variable pH.

Full text of DOI:10.1021/ja201511u

ARTICLE
pubs.acs.org/JACS
Hexadentate Terephthalamide(bis-hydroxypyridinone) Ligands for
Uranyl Chelation: Structural and Thermodynamic Consequences of
Ligand Variation1
Gꢀeza Szigethy and Kenneth N. Raymond*
Department of Chemistry, University of California at Berkeley, Berkeley, California 94720-1460, United States
Chemical Sciences Division, Glenn T. Seaborg Center, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States
S Supporting Information
b
ABSTRACT: Several linear, hexa- and tetradentate ligands incorporating a combination
of 2,3-dihydroxy-terephthalamide (TAM) and hydroxypyridinone-amide (HOPO) moieties
have been developed as uranyl chelating agents. Crystallographic analysis of several
{UO2[TAM(HOPO)2]}2ꢀ complexes revealed a variable and crowded coordination geometry
about the uranyl center. The TAM moiety dominates the bonding in hexadenate complexes,
with linker rigidity dictating the equality of equatorial UꢀO bonding. Hexadentate TAM-
(HOPO)2 ligands demonstrated slow binding kinetics with uranyl affinities on average 6 orders
of magnitude greater than those of similarly linked bis-HOPO ligands. Study of tetradentate
TAM(HOPO) ligands revealed that the high uranyl affinity stems primarily from the presence of the TAM moiety and only
marginally from increased ligand denticity. Uranyl affinities of TAM(HOPO)2 ligands were within experimental error, with TAM-
(o-phen-1,2-HOPO)2 exhibiting the most consistent uranyl affinity at variable pH.
’ INTRODUCTION
limitations make macrocycles better suited to the coordinative
inflexibility about actinyl cations. Macrocyclic actinyl complexes
are plentiful and coordinative saturation can be achieved with
conveniently placed O-, N-, and S-donating groups in both
flexible and rigid penta- and hexadenate macrocycles.8,26,27,30,31
The drawback to macrocyclic chelating agents, however, is that
their ligand conformation may be sensitive to steric agreement
between binding pocket and chelated ion, requiring a careful
approach to target ligand design.8
Our recent work on actinyl-specific sequestering agents has
addressed the incorporation of siderophore-inspired catechol-
amide (CAM), 2,3-dihyroxy-terephthalamide (TAM), and hy-
droxypyridinone-amide (HOPO) binding moieties into polybi-
dentate, equatorially coordinating ligands.3,32ꢀ34 These hard
Lewis basic chelating moieties are well-suited for binding the
Lewis acidic actinyls, and several studies exist on their crystal-
lographic, thermodynamic, and in vivo actinide decorporation
properties.3,33ꢀ36 These analyses indicated that bis-bidentate
ligands more effectively chelate the uranyl cation than mono-
bidentate ligands. However, recent studies suggested that bis-
bidentate siderophore-inspired ligands fail to saturate the uranyl
coordination plane, leaving a solvent-accessible area at the
uranium center whose size is inversely proportional to linker
length (Figure 1).32ꢀ34 While typically occupied by solvent
molecules, this open coordination site could allow undesired
interactions between the exposed uranyl center and the environ-
ment from which it is being extracted via coordination of
The uranyl cation (UO22þ) and the related dioxo cations
(actinyls, AnO2nþ, where An = U, Np, Pu, and n = 1, 2) pose an
unusual challenge for the rational design of selective sequestering
agents.2,3 These complications stem partly from the trans
orientation of their oxo atoms, which relegate ligand coordina-
tion to an equatorial plane perpendicular to the OdAndO
vector; significant deviations from actinyl linearity4ꢀ7 or equa-
torial coordinative planarity5,8 occur only in the presence of
unusually sterically demanding ligands. Additionally, the actinyl
oxo moieties generally exhibit poor Lewis basicity, and interac-
tions between oxo groups and Lewis acids are typically only
observed in the solid state,9ꢀ15 in carefully designed ligand
scaffolds,6,16ꢀ19 or with the lower-valent actinyls due to the
decreased Lewis acidity of the central actinide.20ꢀ22 As a result,
few actinyl chelating agents attempt targeted interaction with the
oxo moieties,18,19,23 with most instead focusing on optimizing
bonding interactions in the equatorial coordination plane.24 As a
natural example for the efficacy of equatorial chelation, oceanic
uranium exists as the [UO2(CO3)3]4ꢀ complex, with extremely
low concentrations of the uncoordinated uranyl cation (1.53 ꢁ
10ꢀ17 M).25
A special class of ligands investigated as high-denticity
f-element chelators are macrocycles, the design strategies for
which are discussed in several comprehensive reviews.26ꢀ29
Macrocycles targeted toward f-elements are by and large applied
toward lanthanide(III) chelation, with ring sizes of 18 atoms and
higher. However, they are often incapable of providing coordi-
native saturation to lanthanide(III) cations due to insufficient
coordinating atoms and limited flexibility. These structural
Received: February 17, 2011
Published: May 04, 2011
r
2011 American Chemical Society
7942
dx.doi.org/10.1021/ja201511u J. Am. Chem. Soc. 2011, 133, 7942–7956
|

Journal of the American Chemical Society
ARTICLE
Figure 1. (Left) Solvent accessible area in UO2(bis-Me-3,2-HOPO)
complexes (uranyl oxo atoms omitted for clarity). (Right) Tris-biden-
tate ligand design for coordinative saturation of actinyl cations.
Figure 2. Hexadentate TAM(HOPO)2 ligands.
mono- or bidentate ligands (e.g., hydroxide or acetate; bite angle
of 52°).37 The typical 66° bite angle of catecholates/hydroxypyr-
idinones at the uranyl center33,34,38,39 explains the lack of a
known uranyl complex saturated by three such moieties, but it
was surmised that a tris-bidentate ligand could achieve coordi-
native saturation by strategic employment of appropriately
shaped linker molecules.40 Such a ligand could capitalize upon
coordinative saturation observed in macrocyclic uranyl com-
plexes without the geometric constraints imparted by a rigidly
defined coordination pocket
The hexadentate ligand design employed in this study incor-
porates three bidentate, oxygen-donating TAM and HOPO
chelating moieties [TAM(HOPO)2 ligands, Figure 1]. This
ligand geometry is similar to the macrocyclic and linear ligands
of Nabeshima and co-workers,41ꢀ43 as well as the macrocycles
developed by MacLachlan,44 which have collectively been shown
to bind both transition and lanthanide metal cations. In contrast
to Schiff base macrocycles, the amide linkages employed in
Figure 1 allow only catecholate-type binding modes due to the
presence of amide protons which are known to stabilize the
negative charge of deprotonated o-phenolate oxygens via intra-
molecular hydrogen bonding.45 Macrocyclic tris-catecholamide
ligands have been synthesized and investigated by our group as
Fe(III) chelators,46 but the linear ligand design in Figure 1 is
intended to provide a greater level of coordinative flexibility than
that in macrocycles. This flexibility is ideal for the current study
because the steric consequences of uranyl chelation by synthetic,
hexadentate siderophore ligands are as yet unexplored. The work
discussed herein details the characterization of the solid-state and
solution-phase interactions of TAM(HOPO)2 ligands with the
uranyl cation.
at lower pH, and (2) they are monoprotic, resulting in dianionic
uranyl complexes as opposed to the tetraanionic species that
would result with a tris-TAM ligand.
Alkyl-linked TAM(HOPO)2 ligands have been previously
explored as high-affinity Fe(III) chelating agents and are capable
of adopting octahedral coordination geometries about Fe(III)
when the linkers contain three or more methylene units.49
However, the capacity for octahedral coordination is undesirable
in ligands designed for the purely equatorial coordination pre-
ferences of actinyl cations, suggesting the use of linkers with no
more than two carbon atoms between the amide nitrogens. The
shortest linkers observed in mononuclear bis-HOPO uranyl
complexes are 3,4-thiophene-, o-phenylene-, and ethylenedia-
mine (2Li; “Li” indicating a linear alkyl linker containing two
methylene units). These linkers also provided the largest solvent
accessible area about the uranyl cation in bis-Me-3,2-HOPO
structures, making them appropriate linkers for actinyl-specific
TAM(HOPO)2 ligands.33 On the basis of these observations,
ligands L1H4ꢀL5H4 were designed to incorporate o-phenylene-
and ethylenediamine linkers and both the 1,2- and Me-3,2-
HOPO moieties as terminal bidentate units (Figure 2). L1H4
through L5H4 are amorphous, beige solids that exhibit poor
solubility in water and most organic solvents with the exception
of DMF and DMSO. Deprotonation makes them significantly
more soluble in polar, organic solvents, although the o-pheny-
lene-linked ligands, L4H4 and L5H4, exhibit much lower solubi-
lity than the 2Li-linked ligands L1H4ꢀL3H4. Ligands containing
the 1,2-HOPO moiety exhibit slightly better solubility than those
containing the Me-3,2-HOPO moiety, independent of solvent
choice or degree of protonation.
Synthesis and Structure of Uranyl Complexes. Mononuc-
lear {UO2[TAM(HOPO)2]}2ꢀ complexes were synthesized in
either DMF or methanol using a combination of KOH and
NMe4OH as bases. The use of hydroxide bases ensured complete
ligand deprotonation, since alkyl-substituted TAM moieties
exhibit second pKa values of 10.3ꢀ11.0.47 Unlike the orange or
red color of uranyl complexes with tetradentate bis-Me-3,2-
HOPO ligands,32ꢀ35 {UO2[TAM(HOPO)2]}2ꢀ complexes
are dark brown in solution and in the solid state, independent
of HOPO moiety and linker geometry. Because they are dianio-
nic and thus highly soluble, the uranyl complexes are difficult to
separate from the mixture of potassium, tetramethylammonium,
’ RESULTS AND DISCUSSION
Hexadentate Ligand Design. Two properties of the TAM
moiety make it ideal for inclusion in the ligand topology in
Figure 1: (1) the presence of two amide groups allows it to act as
a functional ligand extension (whereas the HOPO moiety can
only act as a terminal group), and (2) TAM is known to form
strong complexes with hard Lewis acidic ions such as Fe(III)47
and Th(IV).48 The terminal binding moieties in Figure 1 were
chosen to be HOPO moieties for two reasons: (1) they exhibit
lower acidity than the TAM moiety,35,47 helping the ligands bind
7943
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
Figure 3. Top and side views of X-ray diffraction structures of {UO2[TAM(HOPO)2]}2ꢀ complexes: (a) [UO2(L2)]2ꢀ; (bꢀe) [UO2(L4)]2ꢀ #1ꢀ#4
respectively; (f) [UO2(L5)]2ꢀ. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms, counterions, and solvent molecules are
omitted for clarity. Oxygen atoms are red, carbons gray, nitrogens blue, and uranium is silver.
and nitrate salts that are byproducts of their complexation. Clean
isolation of {UO2[TAM(HOPO)2]}2ꢀ complexes was afforded
in most cases by solvent layering or vapor diffusion techniques
that selectively deposited the uranyl complexes, often in crystal-
line form. Isolation of crystalline, X-ray-quality samples of
[UO2(L2)]2ꢀ, [UO2(L4)]2ꢀ, and [UO2(L5)]2ꢀ was facilitated
by the poor solubility imparted by the Me-3,2-HOPO moieties
and the o-phenylene linkers, whereas crystals of [UO2(L1)]2ꢀ
and [UO2(L3)]2ꢀ could not be cleanly isolated. The six crystal
structures collected from these samples (one each of [UO2-
(L2)]2ꢀ and [UO2(L5)]2ꢀ, and four of [UO2(L4)]2ꢀ) are shown
in Figure 3, with crystallographic parameters listed in Table 1.
The TAM(HOPO)2 ligands bind the uranyl cation in a
hexadentate fashion, with all coordinating oxygens occupying
the equatorial coordination plane. This behavior is independent
of HOPO moiety or linker rigidity, although there exists a
variable amount of ligand ruffling out of the coordination plane
in the Supporting Information].34 Typical of UO2(bis-Me-3,2-
HOPO) complexes, (phen12HP)2ꢀ binds uranyl with four
oxygen atoms in the equatorial coordination plane, with coordi-
native saturation provided by a molecule of DMSO.33,34
The equatorial UꢀO bond lengths in the {UO2[TAM-
(HOPO)2]}2ꢀ and UO2(bis-HOPO) complexes are labeled
schematically in Figure 5 and are compared in Table 2. Although
the bond lengths vary significantly between the investigated
structures, the UꢀOTAM distances in the hexadentate structures
are very similar, with an average of about 2.40(3) Å. This bond
distance is consistent with the 2.39ꢀ2.49 Å MꢀO bond lengths
in [ML4]4ꢀ complexes, where M is U/Th(IV) and L is a
bidentate, untethered TAM or catechol.48,51 The invariance in
UꢀOTAM bond lengths suggests that the TAM moiety dom-
inates the chelation behavior in {UO2[TAM(HOPO)2]}2ꢀ
complexes.
In tetradentate bis-HOPO uranyl complexes the UꢀOphenolate
bond lengths are typically about 0.1 Å shorter than the UꢀOamide
bond lengths, as observed in UO2(2Li32HP) and UO2-
(phen32HP)(DMSO). This is consistent with a stronger Uꢀ
O bond to the more negatively charged phenolate oxygen than to
the neutral amide oxygen.33,34 However, the UꢀOHOPO dis-
tances in the hexadentate [UO2(L2)]2ꢀ complex are nearly
equal, and both are longer than the UꢀOamide bonds in UO2-
(2Li32HP) by 0.1 Å. This suggests that the geometric con-
straints of hexadentate uranyl coordination require a sacrifice in
bond strength to the HOPO oxygens even when utilizing the
flexible 2Li linker. The constraints of uranyl chelation also
generate a mild helical twist in [UO2(L2)]2ꢀ, indicating that fully
planar coordination is not quite achievable given the bite angles and
geometric constraints of the TAM(HOPO)2 ligand design.
In contrast to the mild ligand distortion in [UO2(L2)]2ꢀ, the
four [UO2(L4)]2ꢀ structures exhibit a wide variety of ligand
which differs even between the four structure of [UO2(L4)]2ꢀ
.
We suggest that the inclusion of the third bidentate moiety is
responsible for the observed ligand distortions, since tetradentate
bis-HOPO ligands utilizing similar linkers adopt very planar
coordination modes about the uranyl cation.33
To better understand the bonding in {UO2[TAM(HO-
PO)2]}2ꢀ complexes, comparison against uranyl complexes with
analogous tetradentate bis-HOPO ligands was required. Appro-
priate tetradentate analogues for L2H4, L4H4, and L5H4 are 2Li-
Me-3,2-HOPO [(2Li32HP)H2],34 o-phenylene-1,2-HOPO
[(phen12HP)H2],50 and o-phenylene-Me-3,2-HOPO [(phen-
32HP)H2],33 respectively (Figure 4). The UO2(2Li32HP) and
UO2(phen32HP)(DMSO) complexes have been reported,33 and
the UO2(phen12HP)(DMSO) complex was synthesized following
standard synthetic procedures [the UO2(phen12HP)(DMSO)
crystal structure and crystallographic parameters are provided
7944
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
7945
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
Figure 4. Tetradentate ligands for structural comparison of uranyl complexes.33,34,50
Figure 5. Schematic of equatorial UꢀO bond lengths, angles, and binding pocket layout in uranyl complexes with TAM(HOPO)2 and bis-Me-3,2-
HOPO ligands.
Table 2. Equatorial UꢀO Bond Lengths in {UO2[TAM(HOPO)2]}2ꢀ and UO2(bis-HOPO) Complexes Labeled According to
Figure 5c
a The two halves of the molecule are crystallographically identical. b Two unique uranyl complexes exist in the asymmetric unit. c Tetradentate complexes
are listed in shaded cells.
orientations about the uranyl cation. In [UO2(L4)]2ꢀ structures
#1 and #4 (Figure 3b and e), helical twists similar to that in
[UO2(L2)]2ꢀ are observed, while in [UO2(L4)]2ꢀ structures #2
and #3 (Figure 3c and d), the TAM moiety bends completely to
one side of the uranyl coordination plane. A range of UꢀOHOPO
bond distances accompany these ligand distortions (inter-
molecular Δdmax = 0.2 Å). However, all of the UꢀOHOPO bonds
in [UO2(L4)]2ꢀ are 0.1ꢀ0.2 Å longer than in the tetradentate
UO2(phen12HP)(DMSO) complex, in which the UꢀOamide
and UꢀON-oxide bonds are nearly identical. The UꢀOamide and
UꢀON-oxide bond equality in UO2(phen12HP)(DMSO) is in
fact typical of coordination complexes of 1,2-HOPO with cations
such as Fe(III), Co(III), UO22þ and Th(IV), in which MꢀON-
oxide and MꢀOamide bonds are the same to within 0.03 Å.38,52,53
This is due to a resonance form of the deprotonated 1,2-HOPO
moiety that populates the formally neutral amide oxygen with
negative charge and establishes aromaticity. Such a resonance
form is insignificant in the Me-3,2-HOPO moiety, given the
propensity for MꢀOphenolate bonds to be shorter than MꢀOamide
bonds in complexes with the Me-3,2-HOPO ligand.33,34,54ꢀ58
Surprisingly, the hexadentate [UO2(L4)]2ꢀ structures #1ꢀ#3
display UꢀON-oxide bond lengths that are actually longer than
their UꢀOamide bonds. In addition, the UꢀON-oxide bonds vary
up to 0.18 Å within each complex. As in [UO2(L2)]2ꢀ, these
bond distances are considered to be the result of geometric
constraints imposed by the TAM-dominated chelation with the
additional constraints of the o-phenylene linker rigidity.
The [UO2(L5)]2ꢀ structure displays a severe helical ligand
twist similar to that in [UO2(L2)]2ꢀ and [UO2(L4)]2ꢀ struc-
tures #1 and #4. The UꢀOphenolate bond length is comparable to
7946
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
Figure 6. Uranyl binding pockets in {UO2[TAM(HOPO)2]}2ꢀ complexes with ligands: (a) (L2)4ꢀ; (b)ꢀ(e) (L4)4ꢀ, #1ꢀ#4 respectively; (f) (L5)4ꢀ
.
Structures are viewed down the uranyl OdUdO vector, with uranyl oxo atoms omitted for clarity. Oxygen atoms are oriented as illustrated in Figure 5.
Thermal ellipsoids are drawn at the 50% probability level. Oxygen atoms are red and uraniums are silver.
Table 3. Equatorial OꢀUꢀO Bite Angles in {UO2[TAM(HOPO)2]}2ꢀ Complexes; Angle Labels Refer to Those in Figure 5
complex
angle 1, [deg]
angle 2, [deg]
angle 3, [deg]
angle 4, [deg]
angle sum, [deg]
[UO2L2]2ꢀ
57.9(1)a
59.0(3)
60.5(1)
60.0(2)
64.3(2)
59.58(9)a
60.9(1)a
59.6(1)a
63.3(1)a
65.0(2)
64.4(1)
64.3(1)
65.2(2)
65.40(9)a
362.2(2)a
364.7(5)
362.2(2)
369.0(3)
383.7(5)
378.3(2)a
[UO2L4]2ꢀ, #1
[UO2L4]2ꢀ, #2
[UO2L4]2ꢀ, #3
[UO2L4]2ꢀ, #4
[UO2L5]2ꢀ
59.8(2), 60.7(2)
59.3(1), 59.9(1)
60.5(1), 61.1(2)
59.4(2), 60.7(2)
60.71(6)a
59.4(2), 60.8(2)
58.87(9), 59.25(9)
60.2(1), 62.9(1)
66.6(2), 67.5(2)
65.93(7)a
a The two halves of the molecule are crystallographically identical, giving rise to single values for angles 2 and 3.
those in the other {UO2[TAM(HOPO)2]}2- structures, but the
UꢀOamide distance of 2.7 Å is longer even than those of uranyl-
coordinated DMSO or DMF molecules (typically about 2.4 Å),
suggesting a very weak interaction with the HOPO amide oxygen
atoms.33,34 This disparity in UꢀOHOPO bond lengths is again
evidence that the TAM moiety dominates the binding interaction
between the uranyl cation and TAM(HOPO)2 ligands.
complexes with mono- and bis-bidentate 1,2- and Me-3,2-HOPO
ligands.33,34,38 The decrease in bite angle is caused by the
relatively long equatorial UꢀOHOPO bond distances discussed
above. In contrast, the TAM bite angles range between 63° and
65°, which are closer to the 65° TAM bite angle in the
unconstrained [Th(ETAM)4]4ꢀ,consistent with the above con-
clusion that the TAM moiety dominates the binding event in
{UO2[TAM(HOPO)2]}2ꢀ complexes.48
The total equatorial angle sum in the uranyl complexes is a
metric by which coordinative crowding can be assessed. This sum
is ideally 360° for completely planar coordination and is expected
to increase upon coordinative crowding, which would in turn be
suggestive of poor geometric agreement between the uranyl
cation and coordinated ligands. Equatorial angle sums in uranyl
complexes deviate little from 360° when using mono- and
bidentate ligands, but may do so much more in the presence of
macrocyclic or coordinatively saturating ligands. The
[UO2(L2)]2ꢀ structure exhibits the smallest angle sum of
362°, but upon exchanging the flexible 2Li linker for the rigid
o-phenylene in (L4)4ꢀ and (L5)4ꢀ, the angle sums generally
increase, with a maximum of 384° observed in [UO2(L4)]2ꢀ
structure #4. These deviations illustrate the significant geometric
strain the o-phenylene linkers introduce to the {UO2[TAM-
(HOPO)2]}2ꢀ complexes which the 2Li linker in (L2)4ꢀ more
effectively mitigate.
The tendency for uranyl complexes with ligands (L4)4ꢀ and
(L5)4ꢀ to exhibit more pronounced ligand distortions than those
with (L2)4ꢀ suggests that the o-phenylene linker imposes too
much steric constraint for unhindered uranyl chelation. The
gauche conformation of the linear 2Li linker may explain the
relatively small deviation from planarity in the [UO2(L2)]2ꢀ
structure; this natural alkyl conformation allows a closer ap-
proach of the HOPO moieties to the uranyl center than does the
rigid o-phenylene linker. Figure 6 illustrates the effect ligand
rigidity has on uranyl position in the TAM(HOPO)2 binding
pocket: the ethylene-linked (L2)4ꢀbinds the uranium most
equally, while the o-phenylene linked (L4)4ꢀ and (L5)4ꢀ bind
the uranium atom farther back toward the TAM moiety, with
UꢀOTAM and UꢀOHOPO bond lengths becoming more unequal
with increasing helical ligand twist (Figure 6e and f).
2þ
Additional information on the interaction between UO2
and TAM(HOPO)2 ligands can be gleaned from the equatorial
OꢀUꢀO angles listed in Table 3. The HOPO bite angle (angle
2) is constant at about 60° in all {UO2[TAM(HOPO)2]}2ꢀ
structures, which is approximately 6° smaller than in uranyl
The interplanar TAM/HOPO/uranyl angles listed in Table 4
yield additional information about the details of uranyl chelation
7947
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
Table 4. Interplanar Angles in {UO2[TAM(HOPO)2]}2ꢀ Complexes
complex
HOPOꢀHOPO [deg]
HOPOꢀTAM [deg]
HOPOꢀuranyl [deg]
TAMꢀuranyl [deg]
[UO2L2]2ꢀ
2.5(3) a
30.8(5)
14.9(2)
45.3(2)
81.8(2)
71.49(8) a
12.3(2) a
3.8(2) a
8.7(2) a
16.9(4)
24.54(8)
29.0(2)
11.4(3)
13.1(2) a
[UO2L4]2ꢀ, #1
[UO2L4]2ꢀ, #2
[UO2L4]2ꢀ, #3
[UO2L4]2ꢀ, #4
[UO2L5]2ꢀ
16.4(5), 25.9(5)
23.2(1), 37.5(1)
11.6(3), 56.6(2)
46.0(3), 49.0(2)
48.3(1) a
13.9(5), 17.0(4)
5.1(1), 13.1(2)
17.8(2), 27.6(2)
41.9(2), 44.5(2)
41.1(1) a
a The two halves of the molecule are crystallographically identical, giving rise to only one HOPOꢀTAM and HOPOꢀuranyl angle.
Figure 7. Water-soluble PEG-TAM(HOPO)2 ligands.
by hexadentate TAM(HOPO)2 ligands. Interplanar angles are
calculated using the mean squared planes defined by the six ring
atoms in the TAM and HOPO moieties and the six coordinating
oxygen atoms from the TAM(HOPO)2 ligands, respectively. Assum-
ing that the unconstrained coordination geometry of a HOPO or
TAM moiety is nearly coplanar with the uranyl coordination plane,33
the ideal values for all the angles in Table 4 are 0°. The differences in
interplanar angles necessarily result from a combination of twists
through and bends out of the uranyl coordination plane by the
chelating moieties, and thus vary significantly between {UO2[TAM-
(HOPO)2]}2ꢀ structures. One notable result is that (L2)4ꢀ imposes
the most planar orientation, with a maximum interplanar angle of
12.3° between the HOPO and TAM moieties; this value is only
slightly larger than interplanar angles observed in tetradentate UO2-
(bis-Me-3,2-HOPO) complexes.33 Ligands (L4)4ꢀ and (L5)4ꢀ
generate interplanar angles between 5.1° and 81.8°, illustrating the
steric consequences of the o-phenylene linker. Significantly, the range
of interplanar angles observed in the o-phenylene-linked ligands
supports our belief that the crystal structures in Figure 3 do not
necessarily represent lowest energy conformations and that uranyl
complexes with (L4)4ꢀ and (L5)4ꢀ in fact experience considerable
flexibility in solution despite their rigid linkers. Such flexibility must
therefore also be assumed in complexes with (L2)4ꢀ due to the
flexibility of the ethylene linker.
Soluble Ligand Design. Aqueous thermodynamic measure-
ments were performed to determine the uranyl affinity of
TAM(HOPO)2 ligands. However, more soluble TAM(HOPO)2
ligands were needed for such measurements because while L1H4,
L2H4, and L3H4 were soluble enough for protonation constant
measurements at 50 μM concentrations, L4H4 and L5H4 were
too insoluble due to the o-phenylene linkers. Water solubility was
introduced to TAM(HOPO)2 ligands using the methyl-pro-
tected triethyleneglycol moiety (referred to hereon as PEG) on
the linker moieties.
PEG group location in TAM(HOPO)2 ligands was important
because minimal electronic or steric influence was desirable in
water-soluble ligands for consistency with their unsubstituted
analogues. Substitution on the linker moieties was thus the only
plausible location for the PEG moieties because they would be
removed from the ligand coordination pocket and have the least
possible affect on the electronic properties of the TAM and
HOPO aromatic moieties. Some degree of inductive influence
upon PEG substitution was unavoidable,35 but structural and
electronic influences are known to be mild in similar bis-Me-3,2-
HOPO ligands. Four PEG-containing TAM(HOPO)2 ligands
were synthesized (L6H4 through L9H4, Figure 7), and all exhibit
better aqueous solubility than L1H4 through L5H4 in both their
neutral and deprotonated forms.
The variety in ligand conformation and bond distances
observed in the {UO2[TAM(HOPO)2]}2ꢀ structures naturally
introduces the question of which molecule most strongly binds
the uranyl cation: The relatively long UꢀOHOPO bonds suggest
that individual chelate strength may suffer in TAM(HOPO)2
topologies compared to tetra- or bidentate ligands, but the
increased chelate effect should yield high uranyl affinities.
The uranyl affinities of TAM(HOPO)2 ligands were expected
to be significantly greater than those of bis-Me-3,2-HOPO
moieties previously reported,35 but to determine if denticity or
TAM moiety inclusion is responsible for this effect, two tetra-
dentate TAM-2Li-Me-3,2-HOPO ligands were designed, one
with and one without a PEG moiety (L10H3 and L11H3,
Figure 8). Ligands L10H3 and L11H3 present a tetradentate
7948
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
Table 5. pKa Values for Tetra- and Hexadentate Ligands
ligand
pKa1
pKa2
pKa3
pKa4
ΣpKa
L1H4
L2H4
L3H4
L6H4
L7H4
L8H4
L9H4
4.91(3)
5.62(8)
4.94(1)
3.90(5)
4.5(1)
6.56(8)
6.65(8)
6.48(1)
5.53(3)
6.35(2)
5.10(7)
6.7(1)
8.7(1)
8.12(6)
8.44(3)
7.78(1)
9.22(2)
7.1(1)
10.7(2)
11.16(9)
11.12(4)
9.95(1)
10.89(4)
9.8(2)
10.07(5)
ꢀ
30.9(2)
31.6(2)
30.98(5)
27.16(6)
30.96(5)
25.2(2)
28.6(2)
23.3(2)
20.9(1)
3.20(8)
4.7(1)
7.13(9)
10.6(1)
10.46(6)
L
L
10H3
11H3
5.89(5)
4.3(1)
6.8(1)
6.15(5)
ꢀ
Figure 8. Tetradentate TAM-2Li-Me-3,2-HOPO ligands.
neutral aqueous solution, so spectrophotometric titrations with
these ligands were performed at 2 μM. Ligand titrations were
typically performed between pH 3ꢀ11. Ligands containing the
more acidic 1,2-HOPO moiety required measuring as low as pH
2.5, and some experiments needed to be brought as high as pH
11.5 because of the basic TAM phenols. Titrations were per-
formed from acidic to basic pH, but subsequent return to acidic
pH revealed poor reversibility which only worsened with in-
creased equilibration time (see Supporting Information for
reversibility plots). Irreversibility was presumably caused by the
oxidation of the TAM moiety at very basic pH by trace oxygen in
the titration cell; as a result, no reverse titrations (basic to acidic pH)
were used in determining ligand protonation constants. The pKa
values determined by spectrophotometric titrations are listed in
Table 5 along with the pKa sums (ΣpKa), which are an overall
measure of ligand acidity; lower values indicate decreased proton
affinity.
The first two protonation constants for TAM(HOPO)2
ligands (pKa1 and pKa2) and the first for L10H3 and L11H3
(pKa1) correspond to deprotonation of the HOPO moieties,35,36
while the last two protonation constants for both ligand types are
those of the TAM moiety.47 The 1,2-HOPO moieties in L1H4,
L6H4, and L8H4 are responsible for the depressed pKa1 and pKa2
values compared to their Me-3,2-HOPO-containing analogues
L2H4, L7H4, and L9H4, consistent with prior results.36 Addition-
ally, the presence of the PEG-amide group in L6H4, L7H4, and
L11H3 lowers each pKa value compared against those of their
unsubstituted analogues (L1H4, L2H4, and L10H3). This effect
has been observed in PEG-substituted bis-Me-3,2-HOPO
ligands and is most likely caused by a hydrogen-bonding
stabilization of the deprotonated phenol or N-hydroxide oxygens
of the TAM and/or HOPO moieties.35 This interaction, in turn,
is enabled by the close proximity of the PEG-amide functionality
to the HOPO and TAM moieties.
ligand geometry toward the uranyl cation similarly to bis-Me-3,2-
HOPO ligands,33ꢀ35 but the TAM moiety makes their electro-
static properties more similar to TAM(HOPO)2 ligands. By
comparing the uranyl affinities of L10H3 and L11H3 against those
of tetradentate PEG-2Li-Me-3,2-HOPO35 and hexadentate
PEG-TAM(HOPO)2 ligands, it can be determined whether
increased TAM(HOPO)2 uranyl affinity is caused by high
denticity and chelate effect or solely by the inclusion of the
TAM moiety.
Solution Thermodynamics. Because TAM and HOPO
moieties require deprotonation for efficient metal chelation,
their metal affinity is necessarily pH dependent. The most
common pH in biological applications is that of blood serum
(pH 7.4), but it is also desirable to investigate the uranyl affinity
of polybidentate ligands in acidic and caustic media, as such condi-
tions are encountered in industrial and waste remediation appli-
cations.59,60 In the presence of dissolved metal ion (Maþ) and
protonated ligand (LHb, where L is ligand with b removable
protons), the pH-dependent metalꢀligand complex MmLlHh
forms according to the equilibrium shown in Eq. 1. The relative
amount of each species in solution is determined by Eq. 2, the
rearrangement of which provides the standard formation con-
stant notation of log βmlh (Eq. 3). The log βmlh value describes a
cumulative formation constant, and a stepwise formation con-
stant, log K, can be calculated from log βmlh values using Eq. 4.
When addressing protonation constants, the stepwise formation
constants are commonly reported as dissociation constants
(ꢀlog K, or pKa), and are done so in this study.
βmlh
þ
ꢀ
mMa þ lLb þ hHþ
s
M L H
m l
ðEq. 1Þ
ðEq. 2Þ
ðma ꢀ lb þ hÞþ
F
s
h
R
m
l
h
½MmLlHhꢂ ¼ βmlh½Mꢂ ½Lꢂ ½Hꢂ
Another observable trend is that the o-phenylene linkers make
L8H4 and L9H4 more acidic compared to their ethylenediamine-
linked analogues L1H4, L2H4, L6H4, and L7H4. This can be
attributed to the extended conjugation introduced by the o-
phenylene linker, which may better stabilize negative charge
buildup upon deprotonation. Because poor solubility prevents
thermodynamic evaluation of L4H4 and L5H4, it is impossible to
discern whether increased acidity is due to the extended ligand
conjugation or alkoxy substitution.
The TAM deprotonation constants (pKa3 and pKa4) in TAM-
(HOPO)2 ligands display an interesting trend when compared to
the published monobidentate TAM pKa values of 6.0ꢀ6.5 and
10.3ꢀ11.0.47 Namely, while the second TAM proton in TAM-
(HOPO)2 ligands exhibits the expected high pKa value of ∼11,
!
½MmLlHhꢂ
h
log βmlh ¼ log
ðEq. 3Þ
m
l
½Mꢂ ½Lꢂ ½Hꢂ
!
ꢀ
ꢁ
½LHnꢂ
β01n
log K01n ¼ log
¼ log
½LHn ꢀ 1ꢂ½Hꢂ
β01ðn ꢀ 1Þ
¼ log β01n ꢀ log β01ðn ꢀ 1Þ
ðEq. 4Þ
Protonation constants were determined using spectrophoto-
metric techniques at ligand concentrations of 4ꢀ6 μM with a
starting DMSO concentration of ∼5%. Despite their PEG
substituents, L8H4 and L9H4 were only sparingly soluble in
7949
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
Figure 9. Spectrophotometric reversibility plots of uranyl titrations with L1H4: (a) 10-min equilibration time illustrating lack of ligand/complex
decomposition; (b) overlay of two separate, identical titrations illustrating hysteresis between 10- and 120-min equilibration times.
the first TAM proton has a pKa between 1 and 3 orders of
magnitude higher than expected. In contrast, both TAM pKa
values in tetradentate L10H3 and L11H3 are very close to those of
monomeric TAM moieties. We hypothesize this discrepancy
stems from the different position of the TAM moiety in the two
ligand types: the terminal amide group in L10H3 and L11H3 can
provide dedicated hydrogen-bond stabilization to only one
phenolate oxygen, whereas the linking TAM amides in TAM-
(HOPO)2 ligands can provide stabilization to both a phenolate
TAM oxygen as well as a N-hydroxide/phenolate oxygen on the
HOPO moiety. Thus the TAM phenols in L10H3 and L11H3 are
on average better stabilized upon deprotonation than those in
TAM(HOPO)2 ligands, lowering their pKa2 values compared to
pKa3 in TAM(HOPO)2 ligands.
Uranyl titrations were monitored spectrophotometrically
using a 1:1 metal-to-ligand ratio to avoid decomposition of free
ligand at high pH. It was assumed that maintaining this ratio
would also ensure mononuclear complexes of the sort observed
in crystallographic analysis. Ligand concentrations during these
measurements ranged between 2ꢀ6 μM with a starting DMSO
concentration of ∼5% to facilitate solvation of less soluble,
neutral uranyl species during the course of the titration. Equili-
bration time between data points was found to be of great
importance: uranyl titrations with ligands L1H4, L2H4, L3H4,
L6H4, and L7H4 using a 10-min equilibration time between data
points displayed terrible reversibility which was found to be a
kinetic effect and not caused by ligand or complex decomposition
(Figure 9a). Extending equilibration time to 2 h revealed a kinetic
hysteresis between pH 2.5 and 5.5 (Figure 9b). It was estimated
that the equilibration time needed between pH 2.5 and 5.5 was at
least 12ꢀ18 h, so batch titrations were performed with all
TAM(HOPO)2 ligands using 48ꢀ72-h equilibration times. In
batch titrations [UO22þ] = [L] = 1.3ꢀ2.0 μM to accommodate
for the 10-cm path length quartz-window UV/vis cell used. The
extended equilibration times resulted in precipitation of either
protonated ligand or neutral uranyl complexes at low pH when
using L2H4 and L3H4 so that uranyl affinities could not be
determined for these ligands. In contrast to the slow kinetics with
TAM(HOPO)2 ligands, uranyl titrations with tetradentate
ligands L10H3 and L11H3 exhibited excellent reversibility with
equilibration times of 10ꢀ20 min, and thus, titrations with these
ligands were performed using incremental titrant addition meth-
ods with [UO22þ] = [L] = 6 μM and a 6.6 cm path length cell.
The uranyl titration spectra also reveal a significant difference
between hexadentate and tetradentate complexes at high pH.
Little change in the UVꢀvisible spectra was observed above pH
8ꢀ9 with TAM(HOPO)2 ligands regardless of equilibration
time, suggesting that fully deprotonated, coordinatively saturated
{UO2[TAM(HOPO)2]}2ꢀ complexes can form at relatively low
pH and that no partial hydrolysis of the uranyl cation occurs
upon increasing hydroxide concentration. In contrast, uranyl
titrations with L10H3 and L11H3 exhibited dynamic spectro-
scopic behavior above pH 9, requiring titrations to be run as high
as pH 11.4. Because the coordination modes of L10H3 and L11H3
are similar to that of bis-Me-3,2-HOPO ligands and do not
saturate the uranyl coordination plane, it was expected that
partial hydrolysis of the uranyl cation occurs at elevated pH, as
it does with bis-Me-3,2-HOPO ligands.33ꢀ35 Excellent reversi-
bility at high pH indicated that the spectrum change was not due
to ligand decomposition. Low pH titrations between pH 3.0 and
1.6 were also performed with L10H3 and L11H3 to help more fully
characterize the complex formation.
One notable result was that L9H4, like the other TAM-
(HOPO)2 ligands, inhibited hydrolysis of the uranyl cation at
elevated pH. This was surprising because the structurally analo-
gous [UO2(L5)]2ꢀ exhibited two very long UꢀOHOPO bonds in
Figure 3f, presumably due to ligand constraints. These weak
UꢀO bonds were expected to yield to hydroxide coordination at
high pH, but it seems that even these bonds successfully prevent
uranyl hydrolysis at high pH. It is also possible that these bonds
do not yield because the strained geometry in Figure 3f is only a
solid-state effect and that in solution the average bond strength of
the amide oxygens to the uranyl cation is greater than the crystal
structure would suggest.
The difference in equilibration times, coupled with the depro-
tonation behavior of the tetra- and hexadentate uranyl complexes
in this study suggest yet another important difference in uranyl
complex formation stemming from ligand geometry: We
7950
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
Scheme 1. Hypothesized Complexation Behavior of TAM(HOPO)2 Ligands (L1H4 shown as example)a
a Uranyl oxo atoms are omitted for clarity and hydrogen-bonding interactions are only indicated for the TAM moiety.
Scheme 2. Proposed Speciation in Uranyl Titrations with
TAM-2Li-Me-3,2-HOPO Ligands (L10H3 shown as example)a
hypothesize that the kinetic barrier observed in uranyl titrations
with TAM(HOPO)2 ligands is associated with TAM moiety
rotation upon complete deprotonation (Scheme 1). It is assumed
that both HOPO moieties simultaneously deprotonate at low
pH35 to form a UO2LH2 complex in which only HOPO moieties
chelate the uranium atom, with the uranyl coordination plane
possibly completed by a coordinated solvent molecule. The
protonated TAM moiety would most likely be oriented away
from the uranyl cation and hydrogen bonded to the amide
oxygen atoms as observed in protonated TAM species.45,47,61
Upon increasing pH the TAM phenols sequentially deprotonate,
forming first a [UO2LH]ꢀ complex in which the hydrogen-
bonding network is disrupted, and allowing slow rotation and
displacement of coordinated solvent at the uranyl center by the
TAM moiety. Upon full deprotonation and uranyl coordination
to form a [UO2L]2ꢀ complex, the TAM phenolates regain a
favorable intramolecular hydrogen-bonding interaction with the
ortho amide protons, explaining the kinetic barrier observed upon
lowering pH and reversing the process. Following this reasoning,
uranyl titration data with TAM(HOPO)2 ligands were refined using
UO2LH2, [UO2LH]ꢀ, and [UO2L]2ꢀ species, assuming that the
HOPO protons are of such similar pKa values that they simulta-
neously deprotonate upon initial uranyl chelation.
In contrast to the slow kinetics observed in TAM(HOPO)2
metalation titrations, the kinetic time scales observed with L10H3
and L11H3 were comparable to those with bis-Me-3,2-HOPO
ligands,35 suggesting that there is little to no kinetic barrier for
tetradentate uranyl chelation. The relatively low value of one
TAM pKa in both of these ligands is comparable to the pKa of Pr-
Me-3,2-HOPO (pKa = 6.12),62 and because of its similarity to
bis-Me-3,2-HOPO ligands, it is reasonable to suppose that initial
uranyl chelation occurs via simultaneous and complete deproto-
nation of the HOPO and TAM moieties to generate an anionic
[UO2L(solv.)]ꢀ complex. Data refinement using more stepwise
deprotonation was unstable and not supported by factor analysis.
A partial hydrolysis of the uranyl center occurs upon increasing
pH, resulting in a [UO2L(OH)]2ꢀ species, consistent with other
tetradentate ligands that inadequately saturate the uranyl co-
ordination plane (Scheme 2).35,63
a Uranyl oxo atoms are omitted for clarity.
Table 6. Log βmlh and pUO2 Values for TAM-Containing
Ligands
a
pUO2
ligand log β11ꢀ1 log β110 log β111 log β112 pH 3.0 pH 7.4 pH 9.0
L1H4
L6H4
L7H4
L8H4
L9H4
—
—
—
—
—
21.95(4) 26.86(8) 30.79(2) 6.9(3) 18.2(3) 21.0(3)
17.9(3) 24.8(2) 29.4(2) 9.2(3) 15.9(3) 17.8(3)
21.5(5) 28.7(4) 32.0(4) 8.1(5) 17.3(5) 20.1(6)
19.1(6) 25.0(4) 30.0(3) 11.5(4) 17.5(8) 19.2(8)
19.0(5) 25.7(2) 29.5(4) 7.9(5) 17.1(5) 18.8(6)
L
10H3 11.92(6) 19.75(1) —
—
6.6(1) 17.5(1) 20.3(2)
L
11H3 10.31(9) 17.9(1)
—
—
7.0(2) 16.0(2) 18.8(1)
a pUO2 = ꢀlog[UO2 free]; [UO22þ] = 10ꢀ6 M and [L] = 10ꢀ5 M
2þ
2þ
for that purpose is pUO2, where pUO2 = ꢀlog[UO2 free].
2þ
“UO2 free” refers to solvated uranyl ion free of complexation by
ligand or hydroxide, with a higher pUO2 corresponding to a
lower concentration of uncomplexed uranyl in solution. As a
pertinent benchmark, pUO2 in oceanic conditions is 16.8 due to
the high affinity of carbonate for the uranyl cation.25,64 pUO2
values in this study are calculated using standard conditions of
[UO22þ] = 10ꢀ6 M and [L] = 10ꢀ5 M (L:M = 10), and thus the
minimum pUO2 value of 6.0 corresponds to completely uncom-
plexed uranyl cation. While typically reported at physiological
pH, pUO2 can be calculated at any pH upon determination of
pKa and log βmlh values; pUO2 values at pH 3.0, 7.4, and 9.0 are
listed for each ligand in Table 6. The relatively large errors in
The uranyl formation constants (log βmlh) for the TAM-
(HOPO)2 and TAM-2Li-Me-3,2-HOPO ligands are reported in
Table 6. Because log βmlh values are species dependent, a species-
independent metric is needed to compare uranyl affinities of the
bis- and tris-bidentate ligands studied here. The metric employed
7951
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
Figure 10. Representative speciation diagrams for uranyl complexes with TAM(HOPO)2 (L1H4, left) and TAM-2Li-Me-3,2-HOPO (L10H3, right)
ligands at standard pUO2 conditions ([UO22þ] = 10ꢀ6 M, [L] = 10ꢀ5 M).
pUO2 values result from a combination of the ∼0.1ꢀ0.2 error in
ligand pKa values and the ∼0.1ꢀ0.6 error in log βmlh values,
which are themselves a consequence of the low concentrations
and large volumes required in the uranyl titrations to achieve
aqueous solubility and allow UVꢀvisible measurements in long
path length cells, respectively.
in pUO2 of 2 to 3 log units between pH 7.4 and pH 9.0 is
primarily caused by hydrolysis of free uranyl ion. In contrast, the
rise in pUO2 (typically 6 to 10 log units) between pH 3.0 and pH
7.4 is primarily caused by increased uranyl affinity due to TAM
deprotonation which increases chelate strength. Increased pUO2
of TAM-2Li-Me-3,2-HOPO ligands upon increasing pH can be
explained in a similar manner. In particular, it should be noted
that although hydrolysis of the uranyl cation occurs at elevated
pH with L10H3 and L11H3, this does not significantly influence
ligand affinity, and the increase in pUO2 between pH 7.4 and pH
9.0 is again primarily caused by an increase in hydrolysis of free
uranyl cation.
The pUO2 values of most TAM(HOPO)2 ligands at pH 7.4
and 9.0 are within experimental error of each other, so the effect
of subtle changes in ligand geometry are unfortunately indis-
cernible from this study. It was expected that the o-phenylene
backbone in L8H4 and L9H4 would at the very least cause a
measurable difference in pUO2 compared to L6H4 and L7H4
because of the significant geometric variations observed in the
structures of their respective uranyl complexes. However, the
differences in pUO2 values are below the error associated with
the batch titration methods. In contrast, pUO2 values at pH 3.0
do show significant variation between TAM(HOPO)2 ligands.
As mentioned above, the TAM moiety does not coordinate the
uranyl cation at low pH, leaving only the HOPO moieties
available to bind the uranyl cation. In support of this theory,
higher pUO2 values at pH 3.0 are observed in ligands incorpor-
ating the more acidic 1,2-HOPO moiety (L6H4 versus L7H4;
L8H4 versus L9H4). Ligand L8H4 demonstrated the highest
pUO2 at pH 3 due to a combination of the intrinsically low
pKa of 1,2-HOPO moieties coupled with the tendency for the
PEG-substituted o-phenylene linkers to lower the pKa of their
associated ligand moieties.35
Representative speciation diagrams of the uranyl complexes
with TAM(HOPO)2 and TAM-2Li-Me-3,2-HOPO ligands are
illustrated in Figure 10; comprehensive speciation diagrams and
UVꢀvisible titration spectra of all ligands are provided in the
Supporting Information. Notably, initial uranyl chelation by
TAM(HOPO)2 ligands L1H4 through L9H4 is complete below
pH 2.5, while for TAM-2Li-Me-3,2-HOPO ligands L10H3 and
L11H3 this is still incomplete at pH 3. We hypothesize this
difference exists because initial complex formation with TAM-
(HOPO)2 ligands requires deprotonation only of the acidic
HOPO moieties, while complexation by TAM-2Li-Me-3,2-
HOPO ligands also requires deprotonation of the more basic
TAM moiety. In fact, the complete deprotonation of the TAM
moiety in L10H3 and L11H3 represents an effective maximum
pKa shift of ∼8 log units in the presence of the uranyl cation as
compared to the free ligand. This illustrates the strong inductive
effect of the uranyl cation that is known to cause effective pKa
shifts as high as 13 orders of magnitude in chelating ligands.65
Such an extreme inductive effect on TAM moiety pKa is not
observed with TAM(HOPO)2 ligands because of the geometric
consequences discussed above.
The pUO2 values of both tetra- and hexadentate ligands are
significantly higher than those of bis-Me-3,2-HOPO ligands at all
pH values.35 This was expected for the TAM(HOPO)2 ligands
due to their higher denticity. However, the pUO2 values of the
tetradentate TAM-2Li-Me-3,2-HOPO ligands reveal the TAM
moiety to have a very strong effect on pUO2, which is necessarily
a significant influence in the hexadentate ligands as well. This
general result is consistent with the known strong affinity of
TAM moiety toward hard Lewis acids such as Fe(III).47 It is
The pUO2 of ligands L10H3 and L11H3 at pH 3.0 are generally
lower than those of TAM(HOPO)2 ligands and similar to those
of tetradentate bis-Me-3,2-HOPO ligands.35 These relatively low
pUO2 values result from the need of L10H4 and L11H3 to lose
three protons to achieve initial uranyl chelation, while TAM-
(HOPO)2 ligands need only lose their two most acidic HOPO
protons. In contrast, the pUO2 values of L10H3 and L11H3 are
surprisingly similar at physiological and basic pH to those of
important to note that any increase in pUO2 upon increasing pH
2þ
is partially due to ligand-independent reduction of [UO2
]
free
caused by increased uranyl hydrolysis. Because speciation dia-
grams of uranyl with TAM(HOPO)2 ligands indicate formation
of a [UO2(L)]2ꢀ complex by pH 8, we hypothesize the increase
7952
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
TAM(HOPO)2 ligands, suggesting that the most influential
chelating moiety in these ligands is the TAM moiety. Namely,
bis- and tris-bidentate ligands are not expected to exhibit
similar metal affinities assuming that all possible coordinating
atoms are capable of simultaneous metal coordination. How-
ever, the high pUO2 values of L10H3 and L11H3 reveal that
the change in geometry from tetradentate to hexadentate is
not as significant as the inclusion of the more basic and
strongly binding TAM moiety into the ligand scaffold. This
conclusion is supported by the crystallographic studies of
{UO2[TAM(HOPO)2]}2ꢀ complexes, which revealed an
invariance in the UꢀOTAM distance despite linker and HOPO
moiety variation.
’ EXPERIMENTAL SECTION
General. All reactions brought to reflux were done so with an
efficient condenser attached to the reaction flask. NMR spectra
were collected using Bruker AMX-400 and AM-400 spectrometers
(1H 400 MHz, 13C 100 MHz) in DMSO-d6. Mass spectrometry and
elemental analyses were performed at the Microanalytical Facility,
College of Chemistry, University of California, Berkeley. Yields indicate
the amount of isolated compound, and reactions are unoptimized.
UO2(o-phen-1,2-HOPO), UO2(phen12HP). A solution of
o-phen-1,2-HOPO50 (31 mg, 0.081 mmol) and 3 drops of pyridine were
dissolved in 15 mL of MeOH, and a solution of UO2(NO3)2 6H2O
3
(37 mg, 0.074 mmol) in 2 mL of MeOH was added. The mixture was stirred
at reflux overnight. After cooling to room temperature, the precipitate
was filtered, washed with MeOH, and dried under vacuum to yield 40 mg
of an orange solid, 83%. Anal. Calcd (Found) for C18H12N4O8U: C,
33.24 (33.19); H, 1.86 (1.90); N, 8.62 (8.47). 1H NMR: δ 7.33
’ CONCLUSIONS
3
3
(dd, J(H,H) = 6.4, 3.6 Hz, 2H; HOPO H), 7.38 (dd, J(H,H) = 8.4,
1.6 Hz, 2H; arom. H), 7.73, (dd, 3J(H,H) = 7.6, 1.6 Hz, 2H; arom. H),
7.96 (t, 3J(H,H) = 8.0 Hz, 2H; HOPO H), 8.48 (dd, 3J(H,H) = 6.4, 3.6
Hz, 2H; HOPO H), 12.51 (s, 2H; NH). 13C NMR: δ 115.46, 118.74,
123.48, 125.15, 128.45, 136.75, 139.13, 157.66, 163.63. MS (FABþ):
m/z 651 (MHþ).
Coordinative saturation of the uranyl cation has been achieved
by the development of hexadentate TAM(HOPO)2 ligands
utilizing short alkyl and aromatic linkers. Hexadentate chelation
is accompanied by a variety of molecular distortions, throughout
which the TAM moiety maintains the shortest and most con-
sistent equatorial UꢀO bond distances. The dominance of the
TAM moiety in uranyl binding is supported by solution phase
thermodynamic measurements with TAM-2Li-Me-3,2-HOPO
ligands, which have remarkably high pUO2 values despite their
lower denticity. Unfortunately, uranyl affinities of the investi-
gated ligands were within experimental error at physiological and
basic pH, obscuring the effect of minor geometric differences
between ligands.
Future ligand design strategies for selective actinyl complexa-
tion are informed by several of the results reported here. First, the
speciation of UO2[TAM-2Li-Me-3,2-HOPO] complexes sup-
ports the known phenomenon that basic chelating moieties are
capable of undergoing complete deprotonation at low pH if
coupled to an acidic chelating moiety that simultaneously binds
the metal cation.65 Second, the spatial arrangement of binding
moieties of differing pKa can result in undesirably slow binding
kinetics, especially in cations with well-defined coordination
modes. A method to avoid the slow binding kinetics of TAM-
(HOPO)2 ligands would be to put the most basic moiety at the
terminal position of a linear, polybidentate ligand. This arrange-
ment may utilize the best traits of both the TAM(HOPO)2 and
TAM-2Li-Me-3,2-HOPO ligands without their respective draw-
backs. Namely, the increased denticity and TAM inclusion
would enhance pUO2, but would also impart flexibility to the
TAM moiety such that it could bind in concert with the lower-
acidity chelating moieties. This geometric property should in
turn depress the effective pKa of the TAM moiety and allow
high efficiency chelation at low pH.
UO2[TAM(2Li-Me-3,2-HOPO)2] 2(NMe4), UO2(L2) 2(NMe4).
3
3
A solution of L2H4 MeOH (101 mg, 0.164 mmol) and NMe4OH 5H2O
3
3
(120 mg, 0.662 mmol) in 5 mL of MeOH was added to a stirred solution of
82.4 mg (0.164 mmol) of UO2(NO3)2 6H2O in 2 mL MeOH. The dark-
3
red solution was heated at reflux overnight, and the volume was reduced to
3 mL. Insoluble material was removed by filtering through glass wool, and
the solution was layered with acetone to yield 106 mg of dark-red crystals
which were filtered and dried by aspiration. These crystals were used for
NMR and X-ray crystallographic analysis and were shown to be UO2-
(L2) 2NMe4 Me2CO, 58%. Anal. Calcd (Found) for C26H24N6O12U
3
3
3
2N(CH3)4 2C3H6O (%): C, 43.09 (42.87); H, 5.42 (5.33); N, 10.05
3
(10.00). 1H NMR: δ 2.08 (s, 6H; CH3), 3.05 (s, 24H; CH3), 3.36 (s, 8H;
OH2), 3.59ꢀ3.63 (m, 8H; CH2), 3.73 (s, 6H; CH3), 6.81 (d, 3J(H,H) =
7.2 Hz, 2H; HOPO H), 6.90 (d, 3J(H,H) = 7.2 Hz, 2H; HOPO H), 6.92
(s, 2H; TAMH), 11.55 (t, 3J(H,H) = 5.2 Hz, 2H; NH), 11.75 (t, 3J(H,H) =
5.2 Hz, 2H; NH). 13C NMR: δ 30.74, 36.46, 38.79, 54.30, 54.34, 54.38,
107.11, 113.34, 116.01, 117.30, 120.84, 161.80, 165.51, 166.90, 167.04,
167.49.
UO2[TAM(o-phen-1,2-HOPO)2] 2NMe4, UO2(L4) 2NMe4.
3
3
A solution of L4H4 1/2H2O 1/2CH3OH (50 mg, 0.074 mmol) and
3
3
NMe4(OH) 5H2O (54 mg, 0.298 mmol) in 4 mL of MeOH was added
to a stirred solution of UO2(NO3)2 6H2O (37 mg, 0.074 mmol) in
3
3
1 mL of MeOH. The deep-red solution was heated at reflux overnight
and cooled to room temperature, and the solvent was removed under
vacuum. The residue was dissolved in 0.5 mL of DMSO, and THF was
diffused into the solution at room temperature. Initial precipitates were
generally colorless, so the crystallization solution was filtered every week
to remove what was suspected to be NMe4NO3. Once dark material
began to precipitate, the solution was filtered one last time and allowed
to continue to diffuse at 4 °C. The crop of dark crystals and amorphous
material was filtered, washed with THF, and allowed to dry by
aspiration for 2 days, yielding 37 mg of crystalline and amorphous,
dark solid that elemental and NMR analyses indicated was UO2-
(L4) 2NMe4 DMSO H2O 1/5THF 1/3NMe4NO3. Anal. Calcd
The substitution strategies used to solubilize the ligands in this
study could similarly be employed as tethers to a variety of larger
molecules or solid supports for various extraction or sensing
applications. Although substitution on the linker moieties does
affect ligand pKa, the influence on uranyl affinity is small, leaving
the efficacy of these ligand platforms largely preserved. Finally, a
conclusion taken from the asymmetric location of the uranyl
cation within the TAM(HOPO)2 binding pocket is that this
ligand system may be more applicable toward actinyl cations of
larger radii, such as UO2þ and NpO2þ, in which ligand distor-
tions may be lessened and M-O bond equality could be better
realized.
3
3
3
3
3
(Found) for C32H20N4O12 2N(CH3)4 C2H6OS H2O 1/5C4H8O
3
3
3
3
3
1/3N(CH3)4NO3 (%): C, 43.35 (43.49); H, 4.75 (4.45); N, 9.93
(9.59); S, 2.62 (2.67). 1H NMR: δ 1.74ꢀ1.77 (m, 0.9H; THF H), 2.54
(s, 6H; DMSO CH3), 3.04 (s, 29H; CH3), 3.58ꢀ3.61 (m, 0.9H; THF
H), 7.11ꢀ7.22 (m, 8H; TAM H þ HOPO H þ arom. H), 7.55 (d, 3J(H,
H) = 6.8 Hz, 2H; HOPO H), 7.72 (t, 3J(H,H) = 8.0 Hz, 2H; HOPO H),
8.43 (d, 3J(H,H) = 8.4 Hz, 2H; arom. H), 8.67 (d, 3J(H,H) = 8.4 Hz, 2H;
7953
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
arom. H), 13.05 (s, 2H; NH), 14.16 (s, 2H; NH). 13C NMR: δ 25.13,
40.42, 54.30, 54.34, 54.38, 67.03, 111.60, 114.33, 117.95, 122.01, 122.36,
122.51, 124.46, 126.85, 130.69, 133.45, 138.10, 158.18, 162.78, 166.49,
166.57. (MS (ESIꢀ): 459.1 (M2ꢀ). X-ray-quality crystals could also be
grown by layering a similarly prepared crude DMSO solution of the
complex and accompanying salts with dioxane. After diffusion at room
temperature, these aliquots yielded three X-ray-quality crystals reported
above. These other crystals were grown from crude materials, and the
mixtures of precipitates from which they were isolated were not suitable
for NMR or elemental analysis measurements.
under Ar. Carbonate-free 0.1 M KOH was prepared from Baker Dilut-It
concentrate and was standardized by titrating against potassium hydro-
gen phthalate using phenolphthalein as an indicator. Solutions of 0.1 M
HCl were similarly prepared and were standardized by titrating against
sodium tetraborate decahydrate to Methyl Red end point.
Spectrophotometric Titration Methods. Ligand stock solu-
tions were made by dissolving a weighed amount of ligand accurate to
0.01 mg in DMSO in a volumetric flask. These stock solutions were frozen when
not in use to prevent ligand decomposition. A stock uranyl solution in 1.2 wt %
nitric acid was purchased from Aldrich (4.22 mM) and used as received.
All titrations were performed with a ∼5% starting concentration of DMSO
added to the KCl solution to promote the solvation of protonated ligands
and their neutral uranyl complexes. Spectrophotometric titrations were
carried out in the presence of 10ꢀ20 equiv (as compared to ligand
concentration) of NH4Cl, MES, and HEPES buffers in order to dampen
the pH change between incremental additions of titrant. Each addition of
acid or base was followed by an equilibration period before pH and
absorbance data were collected: 600 s for free ligand and 600ꢀ1200 s for
titrations in the presence of UO22þ. Low-concentration batch titrations
were allowed to equilibrate for three days with constant agitation before the
pH was measured on a freshly calibrated electrode and the spectra were
recorded. Spectra were recorded between 250ꢀ550 nm; The UVꢀvisible
silent region above ∼420 nm was monitored for baseline drift as an
indication of precipitated material.
Ligand concentrations for spectrophotometric titrations with a
6.6 cm path length cell and incremental addition of titrant were
approximately 2ꢀ6 μM. Ligand concentrations for batch titrations using
a 10 cm path length cell were 1.3ꢀ2 μM. All uranyl titrations were
conducted with a 1:1 ligand:metal ratio to avoid decomposition of metal-
free TAM ligands at high pH. Metal-to-ligand ratios were controlled by
careful addition of a ligand solution of known concentration in DMSO
and a standardized uranyl solution in 1.2 wt % nitric acid to the titration
apparatus. All titrations (including long-equilibration time batch
titrations) were repeated a minimum of three times. Each titration
involving incremental addition of titrant was run forward and backward
(from acid to base and reverse) when the titrations were deemed
reversible. Titrations with TAM(HOPO)2 ligands were only performed
down to pH 2.4, while those with tetradentate TAM-2Li-Me-3,2-HOPO
ligands were performed down to pH 1.6 by performing two strong acid
titrations between pH 3.0 and 1.6. Data from these titrations were
combined with those from higher pH titrations to yield the reported
values.
Titration Data Treatment. Spectrophotometric titration data
were analyzed using the program pHab,73 utilizing nonlinear least-
squares regression to determine formation constants. Values for the
hydrolysis product of the uranyl cation were taken from a recent
literature publication.74 Wavelengths between 250ꢀ400 nm were
typically used for data refinement, although batch titration data was
often truncated to ∼270ꢀ400 nm due to large errors in the data at the
lower wavelengths that typically had much stronger absorbance that the
rest of the spectrum. The number of absorbing species to be refined
upon was determined by factor analysis within the pHab program
suite.73 Reversibility of the titrations was determined by comparison
of the species- and concentration-independent value A*v (absor-
bance*volume) at selected wavelengths for the forward and reverse
titrations. Speciation diagrams were generated using HYSS75,76 titration
simulation software and the protonation and metal complex formation
constants determined by potentiometric and spectrophotometric titra-
tion experiments.
UO2[TAM(o-phen-Me-3,2-HOPO)2] 2NMe4, UO2(L5) 2NMe4.
3
3
A solution of L5H4 1/2H2O 1/2HCl (100 mg, 0.141 mmol) and NMe4-
3
3
(OH) 5H2O (106 mg, 0.585 mmol) in 10 mL of MeOH was added to a
stirred solution of UO2(NO3)2 6H2O (73.6 mg, 0.147 mmol) in 5 mL of
3
3
MeOH. The resultant red suspension was stirred at reflux overnight, then
cooled to room temperature and filtered. The solid was dried under vacuum,
yielding 132 mg of brown powder isolated as the methanolic hydrate, 78%.
Anal. Calcd (Found) for C34H24N6O12U 2[N(CH3)4] CH3OH H2O
3
3
3
(%): C, 44.41 (44.40); H, 4.85 (4.56); N, 9.64 (9.44). Crystals of this
complex were formed by diffusion of MeOH into a DMSO complex solution.
X-ray crystallography revealed the crystals to be of the composition UO2-
[TAM(3ꢀ5)2] 2NMe4 2MeOH. NMR analysis was performed on these
3
3
crystals. 1H NMR: δ 3.02 (s, 24H; CH3), 3.18 (d, 3J(H,H) = 4.8 Hz, 6H;
CH3OH), 3.77 (s, 6H; CH3), 4.10(q,3J(H,H) = 4.8 Hz, 2H; CH3OH), 6.97
(q, 3J(H,H) = 7.2 Hz, 4H; arom. H), 7.07ꢀ7.18 (m, 6H; arom. H þ TAM
H), 12.89 (s, 2H; NH), 13.24 (s, 2H; NH). 13C NMR: δ 36.65, 48.61, 54.28,
54.32, 54.36, 106.94, 114.13, 115.98, 117.43, 121.54, 121.72, 122.04, 122.94,
123.43, 127.84, 130.52, 161.29, 164.31, 166.38, 166.59. MS (ESIꢀ): m/z
473.1 (M2ꢀ).
X-ray Diffraction Data Collection. Uranyl complex crystals were
mounted on captan loops with oil and cooled under a controlled
temperature stream of liquid nitrogen boil-off during data collection.
X-ray diffraction data were collected using either Bruker SMART 1000
or APEX I detectors with Mo KR radiation at the UC Berkeley X-ray
crystallographic facility. All data were integrated by the program
SAINT.66,67 The data were corrected for Lorentz and polarization
effects. Data were analyzed for agreement and possible absorption using
XPREP, and an empirical absorption correction was applied in
SADABS.68,69 Equivalent reflections were merged without an applied
decay correction. All structures were solved using direct methods and
were expanded with Fourier techniques using the SHELXL package.70
Least squares refinement of F2 against all reflections was carried out to
convergence with R[I > 2σ(I)]. Least squares planes and angles between
them were calculated using the SHELXL package.70 Further details on
the crystallographic refinement of the crystal structures are provided in
the Supporting Information.
Titration Solutions and Equipment. Corning high perfor-
mance combination glass electrodes (response to [Hþ] was calibrated
before each titration71) were used together with either an Accumet pH
meter or a Metrohm Titrino to measure the pH of the experimental
solutions. Metrohm autoburets (Dosimat or Titrino) were used for
incremental additions of acid or base standard solution to the titration
cell. The titration instruments were fully automated and controlled using
LabVIEW software.72 Titrations were performed in 0.1 M KCl support-
ing electrolyte under positive Ar gas pressure. The temperature of the
experimental solution was maintained at 25 °C by an externally
circulating water bath. UVꢀvisible spectra for incremental titrations
were recorded on a Hewlett-Packard 8452a spectrophotometer (diode
array), while those for batch titrations were recorded on a Cary 300 Scan
UVꢀvis spectrophotometer using customized quartz-windowed cells.
Solid reagents were weighed on a Metrohm analytical balance accurate
to 0.01 mg. All titration solutions were prepared using distilled water that
was purified by passing through a Millipore Milli-Q reverse osmosis
cartridge system and degassed by boiling for 1 h while being purged
’ ASSOCIATED CONTENT
S
Supporting Information. Ligand design and synthesis de-
b
tails, crystallographic figures and refinement details, crystallographic
7954
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
(25) Schwochau, K. Top. Curr. Chem. 1984, 124, 91–133.
information files (CIF), titration data, and figures. This material
(26) Alexander, V. Chem. Rev. 1995, 95, 273–342.
(27) Fenton, D. E.; Vigato, P. A. Chem. Soc. Rev. 1988, 17, 69–90.
(28) Dam, H. H.; Reinhoudt, D. N.; Verboom, W. Chem. Soc. Rev.
2007, 36, 367–377.
is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
(29) Sessler, J. L.; Melfi, P. J.; Pantos, G. D. Coord. Chem. Rev. 2006,
250, 816–843.
(30) Casellato, U.; Sitran, S.; Tamburini, S.; Vigato, P. A.; Graziani,
R. Inorg. Chim. Acta 1986, 114, 111–117.
Corresponding Author
raymond@socrates.berkeley.edu
’ ACKNOWLEDGMENT
(31) Sessler, J. L.; Vivian, A. E.; Seidel, D.; Burrell, A. K.; Hoehner,
M.; Mody, T. D.; Gebauer, A.; Weghorn, S. J.; Lynch, V. Coord. Chem.
Rev. 2001, 216ꢀ217, 411–434.
We thank Drs. Rebecca Abergel and Trisha Hoette for
assistance with the titration measurements and data treatment.
This research is supported by the Director, Office of Science,
Office of Basic Energy Sciences, and the Division of Chemical
Sciences, Geosciences, and Biosciences of the U.S. Department
of Energy at LBNL under Contract No. DE-AC02-05CH11231.
(32) Szigethy, G.; Raymond, K. N. Inorg. Chem. 2009, 48,
11489–11491.
(33) Szigethy, G.; Raymond, K. N. Inorg. Chem. 2010, 49, 6755–6765.
(34) Xu, J.; Raymond, K. N. Inorg. Chem. 1999, 38, 308–315.
(35) Szigethy, G.; Raymond, K. N. Chem.—Eur. J. 2011,
17, 1818–1827.
(36) Durbin, P. W.; Kullgren, B.; Ebbe, S. N.; Xu, J.; Raymond, K. N.
Health Phys. 2000, 78, 511–521.
(37) Templeton, D. H.; Zalkin, A.; Ruben, H.; Templeton, L. K. Acta
Crystallogr., Sect. C 1985, C41, 1439–1441.
(38) Casellato, U.; Vigato, P. A.; Tamburini, S.; Graziani, R.; Vidali,
M. Inorg. Chim. Acta 1983, 72, 141–147.
(39) Casellato, U.; Guerriero, P.; Tamburini, S.; Vigato, P. A.;
Graziani, R. J. Chem. Soc., Dalton Trans. 1990, 1533–1541.
(40) Andersen, O. Chem. Rev. 1999, 99, 2683–2710.
(41) Akine, S.; Taniguchi, T.; Nabeshima, T. Tetrahedron Lett. 2001,
42, 8861–8864.
’ REFERENCES
(1) This paper is #64 in the series “Specific Sequestering Agents for
the Actinides.” For the previous paper, see Ni, Chengbao; Shuh, D. K.;
Raymond, K. N. Chem. Commun. DOI: 10.1039/c1cc11329a.
ꢀ
(2) Ansoborlo, E.; Amekraz, B.; Moulin, C.; Moulin, V.; Taran, F.;
Bailly, T.; Burgada, R.; Hengꢀe-Napoli, M.-H.; Jeanson, A.; Den Auwer,
C.; Bonin, L.; Moisy, P. C. R. Chim. 2007, 10, 1010–1019.
(3) Gorden, A. E. V.; Xu, J.; Raymond, K. N. Chem. Rev. 2003,
103, 4207–4282.
(4) Maynadiꢀe, J.; Berthet, J.-C.; Thuery, P.; Ephritikhine, M. Chem.
Commun. 2007, 486–488.
(5) Berthet, J.-C.; Thuery, P.; Dognon, J.-P.; Guillaneux, D.;
Ephritikhine, M. Inorg. Chem. 2008, 47, 6850–6862.
(6) Sarsfield, M. J.; Helliwell, M. J. Am. Chem. Soc. 2004,
126, 1036–1037.
(42) Akine, S.; Taniguchi, T.; Nabeshima, T. J. Am. Chem. Soc. 2006,
128, 15765–15774.
(43) Akine, S.; Sunaga, S.; Taniguchi, T.; Miyazaki, H.; Nabeshima,
T. Inorg. Chem. 2007, 46, 2959–2961.
(44) Gallant, A. J.; Hui, J. K.-H.; Zahariev, F. E.; Wang, Y. A.;
MacLachlan, M. J. J. Org. Chem. 2005, 70, 7936–7946.
(45) Garrett, T. M.; Cass, M. E.; Raymond, K. N. J. Coord. Chem.
1992, 25, 241–253.
(46) Rodgers, S. J.; Ng, C. Y.; Raymond, K. N. J. Am. Chem. Soc.
1985, 107, 4094–4095.
(47) Jurchen, K. M. C.; Raymond, K. N. Inorg. Chem. 2006,
45, 1078–1090.
(48) Gramer, C. J.; Raymond, K. N. Inorg. Chem. 2004, 43,
(7) Wilkerson, M. P.; Burns, C. J.; Morris, D. E.; Paine, R. T.; Scott,
B. L. Inorg. Chem. 2002, 41, 3110–3120.
(8) Sessler, J. L.; Seidel, D.; Vivian, A. E.; Lynch, V.; Scott, B. L.;
Keogh, D. W. Angew. Chem., Int. Ed. 2001, 40, 591–594.
(9) Thuꢀery, P.; Masci, B.; Takimoto, M.; Yamato, T. Inorg. Chem.
Commun. 2007, 10, 795–799.
(10) Zhang, W.; Zhao, J. Inorg. Chem. Commun. 2006, 9, 397–399.
(11) Masci, B.; Thuꢀery, P. CrystEngComm. 2007, 9, 582–590.
(12) Sarsfield, M. J.; Helliwell, M.; Raftery, J. Inorg. Chem. 2004,
43, 3170–3179.
(13) Burns, C. J.; Clark, D. L.; Donohoe, R. J.; Duval, P. B.; Scott,
B. L.; Tait, C. D. Inorg. Chem. 2000, 39, 5464–5468.
(14) Danis, J. A.; Lin, M. R.; Scott, B. L.; Eichhorn, B. W.; Runde,
W. H. Inorg. Chem. 2001, 40, 3389–3394.
(15) Thuꢀery, P.; Masci, B. J. Chem. Soc., Dalton Trans.
2003, 2411–2417.
6397–6402.
(49) Jurchen, K. M. C.; Raymond, K. N. J. Coord. Chem. 2005,
58, 55–80.
(50) D’Aleo, A.; Xu, J.; Moore, E. G.; Jocher, C. J.; Raymond, K. N.
Inorg. Chem. 2008, 47, 6109–6111.
(51) Sofen, S. R.; Abu-Dari, K.; Freyberg, D. P.; Raymond, K. N.
J. Am. Chem. Soc. 1978, 100, 7882–7887.
(52) Scarrow, R. C.; Riley, P. E.; Abu-Dari, K.; White, D. L.;
Raymond, K. N. Inorg. Chem. 1985, 24, 954–967.
(53) Riley, P. E.; Abu-Dari, K.; Raymond, K. N. Inorg. Chem. 1983,
22, 3940–3944.
(16) Arnold, P. L.; Blake, A. J.; Wilson, C.; Love, J. B. Inorg. Chem.
2004, 43, 8206–8208.
(17) Arnold, P. L.; Patel, D.; Blake, A. J.; Wilson, C.; Love, J. B. J. Am.
Chem. Soc. 2006, 128, 9610–9611.
(54) Xu, J.; Weisenhunt, D. W., Jr.; Veeck, A. C.; Uhlir, L. C.;
Raymond, K. N. Inorg. Chem. 2003, 42, 2665–2674.
(55) Xu, J.; Franklin, S. J.; Whisenhunt, D. W. J.; Raymond, K. N.
J. Am. Chem. Soc. 1995, 117, 7245–7246.
(56) Xu, J.; O’Sullivan, B.; Raymond, K. N. Inorg. Chem. 2002,
41, 6731–6742.
(18) Franczyk, T. S.; Czerwinski, K. R.; Raymond, K. N. J. Am. Chem.
Soc. 1992, 114, 8138–8146.
(19) Sather, A. C.; Berryman, O. B.; Rebek, J., Jr. J. Am. Chem. Soc.
2010, 132, 13572–13674.
(20) Copping, R.; Gaunt, A. J.; May, I.; Sharrad, C. A.; Collison, D.;
Helliwell, M.; Fox, O. D.; Jones, C. J. Chem. Commun. 2006,
36, 3788–3790.
(21) Nocton, G.; Horeglad, P.; Pꢀecaut, J.; Mazzanati, M. J. Am. Chem.
Soc. 2008, 130, 16633–16645.
(22) Madic, C.; Begun, G. M.; Hobart, D. E.; Hahn, L. Radiochim.
Acta 1983, 34, 195–202.
(23) Walton, P. H.; Raymond, K. N. Inorg. Chim. Acta 1995,
240, 593–601.
(24) Beer, S.; Berryman, O. B.; Ajami, D.; Rebek, J., Jr. Chem. Sci.
2010, 1, 43–47.
(57) Xu, J.; Radkov, E.; Ziegler, M.; Raymond, K. N. Inorg. Chem.
2000, 39, 4156–4164.
(58) Gorden, A. E. V.; Shuh, D. K.; Tiedemann, B. E. F.; Wilson,
R. E.; Xu, J.; Raymond, K. N. Chem.—Eur. J. 2005, 11, 2842–2848.
(59) Committee on Cesium Processing Alternatives for High-Level
Waste at the Savannah River Site; Board on Radioactive Waste Manage-
ment; Board on Chemical Sciences and Technology; National Research
Council Alternatives for High-Level Waste Salt Processing at the Savannah
River Site; National Academies Press: Washington, DC, 2000.
7955
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956

Journal of the American Chemical Society
ARTICLE
(60) Committee on Disposition of High-Level Radioactive Waste
Through Geological Isolation; Board on Radioactive Waste Manage-
ment; National Research Council Disposition of High-Level Waste and
Spent Nuclear Fuel; The Continuing Societal and Technical Challenges;
National Academies Press: Washington, DC, 2001.
(61) Karpishin, T. B.; Stack, T. D. P.; Raymond, K. N. J. Am. Chem.
Soc. 1993, 115, 182–192.
(62) Xu, J.; Durbin, P. W.; Kullgren, B.; Raymond, K. N. J. Med.
Chem. 1995, 38, 2606–2614.
(63) Mullen, L.; Gong, C.; Czerwinski, K. R. J. Radioanal. Nucl.
Chem. 2007, 273, 683–688.
(64) Martell, A. E.; Smith, R. M. Critical Stability Constants; Plenum:
New York, 1977; Vol. 5.
(65) Szabꢀo, Z.; Grenthe, I. Inorg. Chem. 2000, 39, 5036–5043.
(66) SAINT: SAX Area-Detector Integration Program, V.6.40; Bruker
Analytical X-ray Systems, Inc.: Madison, WI, 2003.
(67) SAINT: SAX Area-Detector Integration Program, V.4.024;
Siemens Industrial Automation, Inc.: Madison, WI, 1995.
(68) XPREP, Part of SHELXTL Crystal Structure Determination
Package, V.6.12; Bruker Analytical X-ray Systems, Inc.: Madison, WI,
2001.
(69) SADABS: Bruker Nonius Area Detector Scaling and Absorption,
V.2.05; Bruker Analytical X-ray Systems, Inc.: Madison, WI, 2003.
(70) SHELXTL, SHELXTL Crystal Structure Determination Package,
V.5.10; Bruker Analytical X-ray Systems, Inc.: Madison, WI, 1997.
(71) Gans, P.; O’Sullivan, B. Talanta 2000, 51, 33–37.
(72) National Instruments Corp.; Bishop, R. H. LABVIEW, version
8.0; Pearson Prentice Hall: Upper Saddle River, NJ, 2007.
(73) Gans, P.; Sabatini, A.; Vacca, A. Ann. Chim. 1999, 89, 45–49.
(74) Brown, P. L. Radiochim. Acta 2002, 90, 589–593.
(75) Alderighi, L.; Gans, P.; Ienco, A.; Peters, D.; Sabatini, A.; Vacca,
A. Coord. Chem. Rev. 1999, 184, 311–318.
(76) Gans, P. HySS2; University of Leeds: Leeds, U.K., 2000.
7956
dx.doi.org/10.1021/ja201511u |J. Am. Chem. Soc. 2011, 133, 7942–7956