Interaction of 6,6′′-bis(5,5,8,8-tetramethyl-5,6,7,8- tetrahydro-1,2,4-benzotriazin-3-yl)-2,2′:6′,2′′- terpyridine (CyMe4-BTTP) with some trivalent ions such as lanthanide(iii) ions and americium(iii)

Article abstract of DOI:10.1039/b924988e

The new ligand 6,6′′-bis(5,5,8,8-tetramethyl-5,6,7,8- tetrahydro-1,2,4-benzotriazin-3-yl)-2,2′:6′,2′′- terpyridine (CyMe₄-BTTP) has been synthesized in 4 steps from 2,2′:6′,2′′-terpyridine. Detailed NMR and mass spectrometry studies indicate th

Full text of DOI:10.1039/b924988e

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Interaction of 6,6¢¢-bis(5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-1,2,4-
benzotriazin-3-yl)-2,2¢:6¢,2¢¢-terpyridine (CyMe4-BTTP) with some trivalent
ions such as lanthanide(III) ions and americium(III)†
Frank W. Lewis,*a Laurence M. Harwood,*a Michael J. Hudson,a Michael G. B. Drew,a Giuseppe Modolo,b
Michal Sypula,b Jean F. Desreux,c Nouri Bouslimanic and Geoffrey Vidickc
Received 27th November 2009, Accepted 26th March 2010
First published as an Advance Article on the web 23rd April 2010
DOI: 10.1039/b924988e
The new ligand 6,6¢¢-bis(5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-1,2,4-benzotriazin-3-yl)-
2,2¢:6¢,2¢¢-terpyridine (CyMe4-BTTP) has been synthesized in 4 steps from 2,2¢:6¢,2¢¢-terpyridine.
Detailed NMR and mass spectrometry studies indicate that the ligand forms 1 : 2 complexes with
lanthanide(III) perchlorates where the aliphatic rings are conformationally constrained whereas 1 : 1
complexes are formed with lanthanide(III) nitrates where the rings are conformationally mobile. An
optimized structure of the 1 : 2 solution complex with Yb(III) was obtained from the relative magnitude
of the induced paramagnetic shifts. X-Ray crystallographic structures of the ligand and of its 1 : 1
2n+
complex with Y(III) were also obtained. The NMR and mass spectra of [Pd(CyMe4-BTTP)]n are
consistent with a dinuclear double helical structure (n = 2). In the absence of a phase-modifier,
CyMe4-BTTP in n-octanol showed a maximum distribution coefficient of Am(III) of 0.039 ( 20%) and a
maximum separation factor of Am(III) over Eu(III) of 12.0 from nitric acid. The metal(III) cations are
extracted as the 1 : 1 complex from nitric acid. The generally low distribution coefficients observed
compared with the BTBPs arise because the 1 : 1 complex of CyMe4-BTTP is considerably less
hydrophobic than the 1 : 2 complexes formed by the BTBPs. In M(BTTP)3+ complexes, there is a
competition between the nitrate ions and the ligand for the complexation of the metal.
The development of N-heterocyclic ligands which are capable of
separating actinides from lanthanides has thus been the subject of
Introduction
intensive research.4 Two classes of ligand have emerged that show
both high affinities and high selectivities towards the trivalent
actinides; the 2,6-bistriazinylpyridines (BTPs)5 1 and the 6,6¢-
bistriazinyl-2,2¢-bipyridines (BTBPs)6 2 (Fig. 1). One member
of the BTBP ligands (CyMe4-BTBP 3)7 has shown many of
the desirable qualities (stability towards hydrolysis and relative
stability towards radiolysis, reversible metal binding which allows
stripping, etc) for use in an industrial separation process. Its
suitability for a SANEX process has been demonstrated recently
Over the last three decades, research in Europe has resulted in
the development of the DIAMEX and SANEX processes for
the reprocessing of spent nuclear fuel produced in the PUREX
process. These are based on the co-separation of trivalent actinides
and lanthanides (DIAMEX process),1 followed by the subsequent
separation of actinide(III) from lanthanide(III) in the SANEX
process.2 If these trivalent actinides (particularly americium and
curium) are removed from the waste (partitioning) and converted
by neutron fission (transmutation) into shorter-lived or stable
elements, the remaining waste loses most of its long-term radiotox-
icity. Partitioning and transmutation are therefore considered
attractive options for reducing the burden on geological waste
disposal.3
aDepartment of Chemistry, The University of Reading, Whiteknights,
Reading, UK RG6 6AD. E-mail: f.lewis@reading.ac.uk, l.m.harwood@
reading.ac.uk; Fax: +44 (0) 118 378 6121; Tel: +44 (0) 118 378 7417
bForschungszentrum Ju¨lich GmbH, Sicherheitsforschung und Reaktortech-
nik, D-52425 Ju¨lich, Germany. E-mail: g.modolo@fz-juelich.de; Tel: + 492
46161 4896
cCoordination and Radiochemistry, University of Lie`ge, Sart Tilman B16,
B-4000 Lie`ge, Belgium. E-mail: jf.desreux@ulg.ac.be; Fax: + 324 366 4739;
Tel: + 324 366 3501
† Electronic supplementary information (ESI) available: Figures of NMR
spectra, mass spectra and graphs referred to in the text and X-ray
crystallographic data in CIF format. CCDC reference numbers 756181–
756182. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/b924988e
Fig. 1 The structures of the BTP, BTBP and BTTP molecules.
5172 | Dalton Trans., 2010, 39, 5172–5182
This journal is
The Royal Society of Chemistry 2010
©

View Online
in a ‘hot-partitioning demonstration test’ on genuine spent fuel
solution.8
The coordination chemistry and solvent extraction properties
of the related bistriazinyl-2,2¢:6¢,2¢¢-terpyridine (BTTP) ligands 4
with trivalent cations have not been previously studied. From the
point of view of ligand design, we wanted to know what effect the
introduction of an additional pyridine ring into the BTBP frame-
work would have on the solvent extraction capability of the ligand.
In addition, we anticipated that BTTPs could be capable of form-
ing helical metal complexes similar to those formed by the oligopy-
ridine ligands,9 which have found widespread use in coordination
and metallosupramolecular chemistry.10 Herein, we describe the
synthesis, selective extraction and coordination chemistry of the
novel ligand 6,6¢¢-bis(5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-1,2,4-
benzotriazin-3-yl)-2,2¢:6¢,2¢¢-terpyridine (CyMe4-BTTP) 5 which,
by virtue of its aliphatic tetramethylcyclohexenyl-rings, is designed
to be more soluble than many of the oligopyridine ligands in
organic solvents.
Scheme 1 Synthesis of CyMe4-BTTP 5
washed with distilled water (100 mL) and allowed to dry in the air
to afford the title compound 7 as a pale brown solid (1.13 g, 99%)
contaminated with minor amounts of compound 8 (see Fig. 2).
The full characterization of compound 7 was hampered by its
insolubility but 7 was identified by the successful preparation of
Experimental
Melting points were obtained on a Stuart SMP10 instrument.
5. Mp > 300 ◦C (from H2O/DMSO). vmax(Nujolꢀ)/cm 3438
R
-1
IR spectra were recorded as Nujolꢀ mulls on a Perkin Elmer
R
1
RX1 FT-IR instrument. 1H, 13C-{ H} and 13C NMR spectra were
(NH2), 3293 (NH2), 2916, 1650, 1567, 1454, 1376, 1265, 1169,
1072, 988, 927, 870, 795, 744 and 631. dH(249.8 MHz; DMSO-d6;
Me4Si) 5.49 (br s, NH2), 6.04 (br s, NH2), 7.50-7.54 (m), 7.93-
8.03 (m), 8.11-8.17 (m), 8.42-8.45 (m), 8.57-8.59 (m), 8.75-8.78
(m), 8.83-8.86 (m), 8.93-8.96 (m). m/z (CI) 348.1675 ([M + H]+);
C17H18N9 requires 348.1685.
recorded using either a Bruker AMX400, an Avance DFX400 or
an Avance DPX250 instrument. Chemical shifts are reported in
parts per million downfield from tetramethylsilane. Assignments
were verified with 1H-1H and 1H-13C COSY experiments as appro-
priate. Mass spectra were obtained under electrospray conditions
on a Thermo Scientific LTQ Orbitrap XL instrument. Elemental
microanalyses were carried out by Medac Ltd., Brunel Science
Centre, Surrey (UK). Solvent extraction studies were performed
at Forschungszentrum Ju¨lich (Germany). Solution phase NMR
studies were carried out at the University of Liege (Belgium). All
organic reagents were obtained from either Acros or Aldrich, while
inorganic reagents were obtained from either BDH or Aldrich and
used as received.
Fig. 2 Minor by-products obtained in the synthesis of compounds 7
and 5.
3,3,6,6-Tetramethylcyclohexane-1,2-dione11 was obtained by
the acyloin reaction of diethyl 2,2,5,5-tetramethylhexanedioate
with sodium/trimethylchlorosilane,12 followed by oxidation with
bromine.13 Diethyl 2,2,5,5-tetramethylhexanedioate11 was ob-
tained by the dimerization of pivalic acid using Fenton’s reagent,14
followed by esterification of 2,2,5,5-tetramethylhexanedioic acid
with conc. sulfuric acid in ethanol.11 2,2¢:6¢,2¢¢-Terpyridine-6,6¢¢-
dicarbonitrile15 6 was obtained by the oxidation of 2,2¢:6¢,2¢¢-
terpyridine with m-CPBA,16 followed by modified Reissert-Henze
reaction with N,N-dimethylcarbamyl chloride and trimethylsilyl
cyanide in dichloromethane.17 Caution! Trimethylsilyl cyanide is
a volatile hydrogen cyanide equivalent. The synthesis of CyMe4-
BTTP 5 is shown in Scheme 1.
6-Cyano-6¢¢-(5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-1,2,4-
benzotriazin-3-yl)-2,2¢:6¢,2¢¢-terpyridine 9 and 6,6¢¢-bis(5,5,8,8-
tetramethyl-5,6,7,8-tetrahydro-1,2,4-benzotriazin-3-yl)-2,2¢:6¢,2¢¢-
terpyridine 5
2,2¢:6¢,2¢¢-Terpyridine-6,6¢¢-dicarbohydrazonamide
7 (1.10 g,
3.17 mmol) was suspended in THF (70 mL) and 3,3,6,6-
tetramethylcyclohexane-1,2-dione (1.06 g, 6.34 mmol) was added.
Triethylamine (6 mL) was added and the suspension was heated
under reflux for 24 h. The resulting solution was allowed to cool
to room temperature and stirring was continued for a further 2 d.
The solvent was removed in vacuo and the residue was purified by
chromatography, eluting first with DCM, then with 2.5% MeOH
in DCM to afford two products. The first product to elute was
compound 9 as a yellow solid (0.11 g, 7.8%). Mp 231–234 ◦C (from
2,2¢:6¢,2¢¢-Terpyridine-6,6¢¢-dicarbohydrazonamide 7
To a suspension of 2,2¢:6¢,2¢¢-terpyridine-6,6¢¢-dicarbonitrile 615
(0.93 g, 3.28 mmol) in DMSO (25 mL) was added hydrazine
hydrate (64% aq. soln, 12 mL). The suspension was stirred at
room temperature for 7 d, then warmed to 50 ◦C for a further 24 h
and was then allowed to cool to room temperature. Distilled water
(500 mL) was added and the precipitated solid was filtered and
MeOH–DCM). Found: C 72.21, H 5.89, N 21.82%; C27H25N7
-1
R
requires C 72.46, H 5.63, N 21.90%. vmax(Nujolꢀ)/cm 2925,
2235 (CN), 1577, 1562, 1459, 1376, 1269, 1146, 1099, 989, 799,
739, 719. dH(400.1 MHz; CDCl3; Me4Si) 1.48 (6 H, s, 2 ¥ Me),
1.53 (6 H, s, 2 ¥ Me), 1.90 (4 H, s, 2 ¥ CH2), 7.73 (1 H, dd, J = 7.8
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5172–5182 | 5173
©

View Online
and 0.6 Hz, 5-H), 8.01 (1 H, t, J = 7.8 Hz, 4-H), 8.03 (1 H, t,
J = 7.8 Hz, 4¢-H), 8.06 (1 H, t, J = 7.8 Hz, 4¢¢-H), 8.53 (1 H, dd,
J = 7.8 and 0.6 Hz, 3¢-H), 8.56 (1 H, dd, J = 7.8 and 0.6 Hz,
3¢¢-H), 8.75 (1 H, dd, J = 7.8 and 0.6 Hz, 5¢¢-H), 8.87 (1 H, dd,
J = 7.8 and 0.6 Hz, 5¢-H), 8.90 (1 H, dd, J = 7.8 and 0.6 Hz,
3-H). dC(100.6 MHz; CDCl3; Me4Si) 29.3 (2 ¥ Me), 29.8 (2 ¥ Me),
33.3 (CH2), 33.7 (CH2), 36.5 (quat), 37.3 (quat), 117.4 (CN), 121.7
(C-3¢), 122.1 (C-5¢¢), 122.7 (C-5¢), 123.8 (C-3¢¢), 124.3 (C-3), 128.1
(C-5), 133.2 (C-6), 137.8 (C-4¢), 137.8 (C-4), 138.3 (C-4¢¢), 153.1
(quat), 153.1 (quat), 155.3 (quat), 156.0 (quat), 157.8 (quat), 160.8
(quat), 163.1 (quat), 164.4 (quat). m/z (CI) 448.2240 ([M + H]+);
C27H26N7 requires 448.2250.
7.8 Hz, 4¢-H), 8.33 (1 H, t, J = 7.8 Hz, 4-H), 8.52 (1 H, dd, J =
7.8 and 0.8 Hz, 3¢ or 5¢-H), 8.70 (1 H, dd, J = 7.8 and 1.2 Hz, 3 or
5-H), 8.75 (1 H, dd, J = 7.8 and 1.2 Hz, 3¢¢ or 5¢¢-H), 8.86 (1 H, dd,
J = 7.8 and 1.2 Hz, 3¢¢ or 5¢¢-H), 8.91 (1 H, t, J = 7.8 Hz, 4¢¢-H).
dC(100.6 MHz; acetone-d6; Me4Si) 28.6 (2 ¥ Me), 29.2 (Me), 29.4
(Me), 30.1 (Me), 30.2 (Me), 30.5 (Me), 30.6 (Me), 32.3 (CH2), 32.6
(2 ¥ CH2), 33.3 (CH2), 38.1 (quat), 38.3 (quat), 39.5 (quat), 39.6
(quat), 126.4 (C-3 or C-5), 126.8 (C-3¢ or C-5¢), 128.6 (C-3¢¢ or C-
5¢¢), 129.3 (C-3¢ or C-5¢), 129.3 (C-3¢¢ or C-5¢¢), 130.2 (C-3 or C-5),
141.2 (C-4¢), 144.4 (C-4), 146.4 (C-4¢¢), 147.5 (quat), 151.1 (quat),
155.3 (quat), 156.7 (quat), 156.8 (quat), 160.1 (quat), 162.2 (quat),
166.3 (quat), 167.4 (quat), 171.2 (quat), 175.0 (quat), 176.6 (quat).
m/z (CI) 358.6264 ([5 + Pd]2+); (C37H41N9Pd)2+ requires 358.6259.
The second product to elute was compound 5 as a yellow solid
(1.38 g, 71%). Mp 257–262 ◦C (from MeOH–DCM). Found: C
72.41, H 6.56, N 20.44%; C37H41N9 requires C 72.64, H 6.75, N
Solvent extraction studies
-1
R
20.60%. vmax(Nujolꢀ)/cm 2921, 1565, 1508, 1458, 1376, 1268,
1244, 1143, 1079, 1049, 1018, 989, 853, 802, 749, 682 and 631.
dH(400.1 MHz; CDCl3; Me4Si) 1.48 (12 H, s, 4 ¥ Me), 1.53 (12
H, s, 4 ¥ Me), 1.89 (8 H, s, 4 ¥ CH2), 8.03 (1 H, t, J = 7.8 Hz,
4¢-H), 8.07 (2 H, t, J = 7.8, 4-H and 4¢¢-H), 8.55 (2 H, dd, J =
7.8 and 1.0 Hz, 3-H and 3¢¢-H), 8.83 (2 H, d, J = 7.8 Hz, 3¢-H
and 5¢-H), 8.84 (2 H, dd, J = 7.8 and 1.0 Hz, 5-H and 5¢¢-H).
dC(100.6 MHz; CDCl3; Me4Si) 29.3 (4 ¥ Me), 29.8 (4 ¥ Me),
33.3 (2 ¥ CH2), 33.8 (2 ¥ CH2), 36.5 (2 ¥ quat), 37.3 (2 ¥ quat),
121.9 (C-3¢ and C-5¢), 122.3 (C-5 and C-5¢¢), 123.7 (C-2 and C-
2¢¢), 137.8 (C-4 and C-4¢¢), 138.0 (C-4¢), 152.9 (2 ¥ quat), 154.9
(2 ¥ quat), 156.5 (2 ¥ quat), 161.0 (2 ¥ quat), 163.0 (2 ¥ quat),
164.5 (2 ¥ quat). m/z (CI) 612.3574 ([M + H]+); C37H42N9 requires
612.3563.
The aqueous solutions were prepared by spiking nitric acid
solutions (0.001–4 mol dm-3) with stock solutions of 241Am and
152Eu tracers (10 mL) in nitric acid. The radiotracers 241Am and
152Eu were supplied by Isotopendienst M. Blaseg GmbH, Wald-
burg (Germany). Solutions of CyMe4-BTTP 5 (0.01 mol dm-3)
were prepared by dissolving 5 in n-octanol, with or without an
additional phase modifier. Each organic phase (500 mL) was
shaken separately with each of the aqueous phases (500 mL) for
one hour at 22 ◦C using an IKA Vibrax Orbital Shaker Model
VXR (2,200 rpm). The contact time of one hour was sufficient
to attain the distribution equilibrium. After phase separation by
centrifugation, 200 mL aliquots of each phase were withdrawn for
radio analysis. Activity measurements of the g-ray emitters 241Am
and 152Eu were performed with a HPGe g-ray spectrometer, EG–
G Ortec. The g-lines at 59.5 keV, and 121.8 keV were examined
for 241Am, and 152Eu, respectively. The acidities of the initial
and final aqueous solutions were determined by potentiometric
titration against sodium hydroxide solution (0.1 mol dm-3) using
a Metrohm 751 GPD Titrino device. The distribution ratio D was
measured as the ratio between the radioactivity in the organic and
the aqueous phase. Distribition ratios between 0.1 and 100 exhibit
a maximum error of 5%. The error may be up to 20% for
smaller and larger values.
Yttrium(III) complex of CyMe4-BTTP 5
CyMe4-BTTP 5 (0.0142 g, 0.0232 mmol) was dissolved in
dichloromethane (5 mL). To this solution was added a solution
of Y(ClO4)3 (0.0112 g, 40 wt% solution in H2O, 0.5 equiv.) in
acetonitrile (2 mL). The two solutions were mixed and allowed to
evaporate to dryness over several days to afford crystals suitable
for X-ray analysis.
Palladium(II) complex of CyMe4-BTTP 5
Nuclear magnetic resonance and electrospray mass studies of
lanthanide complexes
CyMe4-BTTP 5 (0.1 g, 0.1636 mmol) was suspended in MeOH
(20 mL) and Pd(OAc)2 (0.036 g, 1 equiv.) was added. The flask
was heated under reflux until a clear solution was obtained. After
20 min of refluxing, a saturated solution of NH4PF6 in MeOH
(10 mL) was added and the solution was evaporated to a volume
of approx. 15 mL. The flask was cooled to 0 ◦C and the precipitate
was filtered and washed with MeOH (10 mL) and allowed to dry
in air to afford the complex as a pale green solid (0.12 g, 75%).
Mp 298–300 ◦C (decomposed). Found: C 43.83, H 3.87, N 12.58,
P 5.92, Pd 10.37%; C74H82N18F24P4Pd2 requires C 44.08, H 4.10,
Anhydrous perchlorate and nitrate lanthanide salts were prepared
by metathesis in dried CH3CN between the corresponding anhy-
drous silver salt (Aldrich) and an anhydrous lanthanide trichloride
(Aldrich). CH3CN was replaced by dried CD3CN (Aldrich)
by evaporation of the solutions to dryness and dissolution in
the deuterated solvent. If needed, the procedure was repeated
several times to ensure that all the non-deuterated solvent was
eliminated. All manipulations were performed in an inert atmo-
sphere glove box (less than 0.5 ppm of water) and the solutions
were prepared by weighing with a XP56 Mettler Toledo XP56
balance (0.002 mg readability). All spectra were recorded on a
250 MHz and/or a 400 MHz Avance Bruker spectrometer at 25 ◦C.
Electrospray mass spectra (ES-MS) of the lanthanide complexes
were recorded with a MicrOTOF mass spectrometer. The absence
of traces of silver complexes after each metathesis procedure was
verified with the same instrument.
N 12.50, P 6.14, Pd 10.56%. vmax(Nujolꢀ)/cm-1 3086, 2925, 1610,
R
1532, 1460, 1377, 1251, 1150, 1111, 1065, 1000, 938, 853 (PF6), 833
(PF6), 742, 556. dH(400.1 MHz; acetone-d6; Me4Si) 0.85 (3 H, s,
Me), 0.89 (3 H, s, Me), 1.29 (3 H, s, Me), 1.34 (3 H, s, Me), 1.35
(3 H, s, Me), 1.66 (3 H, s, Me), 1.67–1.72 (2 H, m, CH2), 1.74 (3
H, s, Me), 1.84 (3 H, s, Me), 1.93–1.99 (2 H, m, CH2), 2.09–2.21
(4 H, m, 2 ¥ CH2), 7.87 (1 H, dd, J = 7.8 and 1.2 Hz, 3 or 5-H),
8.15 (1 H, dd, J = 7.8 and 0.9 Hz, 3¢ or 5¢-H), 8.24 (1 H, t, J =
5174 | Dalton Trans., 2010, 39, 5172–5182
This journal is
The Royal Society of Chemistry 2010
©

View Online
Crystallography
obtained following the addition of a saturated methanolic solution
of ammonium hexafluorophosphate.
X-Ray diffraction data were collected with MoKa radiation at
150 K using the Oxford Diffraction X-Calibur CCD System. The
crystals were positioned at 50 mm from the CCD. 321 frames were
measured with a counting time of 10 s. Data analyses were carried
out with the CrysAlis program.18 The structures were solved
using direct methods with the SHELXS-97 program.19 The non-
hydrogen atoms were refined with anisotropic thermal parameters.
The hydrogen atoms bonded to carbon were included in geometric
positions and given thermal parameters equivalent to 1.2 times
those of the atom to which they were attached. The hydrogen
atoms bonded to oxygen in water molecules in the yttrium complex
could not be located. One perchlorate in the yttrium complex
was disordered over 3 different sites and refined with occupancy
factors of 0.4, 0.4 and 0.2 with distance constraints. An absorption
correction was applied using the ABSPACK program.20 The
structures were refined on F2 using SHELXL-97.19
Solvent extraction studies
The distribution coefficients and the separation factors for CyMe4-
BTTP 5 in n-octanol as a function of nitric acid concentration
are shown in Fig. 3. In the absence of a phase-modifier, the
distribution coefficients for Am(III) are greater than those for
Eu(III) at nitric acid concentrations of 1.0–4.0 mol dm-3 and
reach a maximum value of 0.039 at 2.0 mol dm-3. In comparison,
CyMe4-BTBP 3 gives a distribution coefficient of approximately
4.5 at a nitric acid concentration of 0.5 mol dm-3.7 The resulting
separation factors were uniform over this range of nitric acid
concentration increasing to a maximum of 12.0 before decreas-
ing to 6.6 in 4.0 molar nitric acid. However, some emulsion
formation was observed at all nitric acid concentrations except
4.0 molar.
Results and discussion
Our aim was to synthesize CyMe4-BTTP 5, which is a symmetrical
molecule similar to CyMe4-BTBP 36 (Fig. 1). The molecule 5 has
two outer 1,2,4-triazinyl moieties bound to a central terpyridine
core meaning that there is one additional pyridine ring compared
with 3. It was expected that the comparison between 5 and
3 would prove useful in the future design of selective actinide
extractants. Moreover, the coordination chemistry of 5 has not
been previously reported. Accordingly, 2,2¢:6¢,2¢¢-terpyridine was
converted in two steps to the dinitrile 6 in 17% overall yield
by formation of the bis-N-oxide using m-CPBA, followed by a
modified Reissert-Henze reaction (Scheme 1). The reaction of
6 with hydrazine hydrate21 afforded the novel dicarbohydrazon-
amide 7, which on treatment with 3,3,6,6-tetramethylcyclohexane-
1,2-dione in refluxing THF in the presence of triethylamine gave
the novel bistriazinylterpyridine 5 as a yellow solid. To our
knowledge, this is the first reported example of this class of
ligand.
Fig. 3 Extraction of Am(III) and Eu(III) by 5 in n-octanol as a function
of nitric acid concentration.
In addition to 5, a minor amount (ca. 10% by 1H NMR analysis
of the crude product) of nitrile 9 was also obtained which pre-
sumably arose from contamination of the dicarbohydrazonamide
7 with the monocarbohydrazonamide 8 (Fig. 2). Fortunately, 9
could be separated from 5 by careful column chromatography.
The formation of 9 indicates that the reaction of dinitrile 6 with
hydrazine does not reach completion. Despite employing long
reaction times, a large excess of hydrazine and heating, minor
amounts of 9 were always obtained in the subsequent condensation
reaction. Ligand 5 showed low solubility in dodecane but showed
good solubility in n-octanol (up to 0.01 mol L-1). The solvent
extraction studies reported herein were subsequently carried out
in n-octanol.
Admixture of a dichloromethane solution of 5 (2 equiv.) with an
aqueous solution of yttrium(III) perchlorate (1 equiv.) in CH3CN,
followed by slow evaporation over several days afforded crystals
of the complex suitable for X-ray analysis. In order to compare
the coordination chemistry of 5 with that of the quinquepyridine
ligands, we synthesized a palladium(II) complex. The reaction
of 5 with Pd(OAc)2 (1 equiv.) in refluxing methanol afforded
a pale green solution from which a pale green precipitate was
We then carried out the extractions in the presence of
certain additives that confer a greater hydrophobicity on the
extracted species. In the presence of 2-bromohexanoic acid,22
higher distribution coefficients were obtained for both Am(III)
and Eu(III) at low nitric acid concentrations but these values
decreased rapidly at higher acidities. The distribution coeffi-
cients for Am(III) were greater than those for Eu(III), while
the separation factors increased to a maximum value of 20.5
before decreasing sharply at higher nitric acid concentrations
(Fig. 4). This decrease at higher nitric acid concentrations is
due to the inability of 2-bromohexanoic acid to dissociate and
form a more lipophilic complex with the Am(CyMe4-BTTP)3+
cation. Employing N,N¢-dimethyl-N,N¢-dioctyl-2-(2-hexoxyethyl)
malondiamide (DMDOHEMA)7-8,23 as a phase-modifier did not
lead to any significant improvement. (Fig. 5). The decrease in the
separation factor at higher nitric acid concentrations is presumably
due to the non-selective extraction of both Am(III) and Eu(III) by
DMDOHEMA itself, rather thanby 5. This reduction in selectivity
in the presence of DMDOHEMA is also observed with the BTBP
ligands.6
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5172–5182 | 5175
©

View Online
planar with successive torsion angles (from left to right in
Fig. 6) between the aromatic rings of 19.6(1), -9.1(1), 1.5(1) and
-24.7(1)◦. As expected the conformation of the central pyridine
ring can be characterised as trans, trans from the relative positions
of the pyridine nitrogen atoms. When bound to a metal, terpyridine
will have the cis, cis conformation. The trans, trans conformation
however is usually found when terpyridine is not chelating as the
ortho hydrogens in adjacent rings do not clash.
Fig. 4 Extraction of Am(III) and Eu(III) by 5 + 2-bromohexanoic acid
(1 mol L-1) in n-octanol as a function of nitric acid concentration.
Fig. 6 The structure of 5 with ellipsoids at 50% probability.
The conformations of the outer triazine rings show that N(11)
is trans to N(21) and N(51) is trans to N(41). This is somewhat
unexpected as in previous work on the BTPs 15 and BTBPs 26
the relative positions of adjacent triazine and pyridine rings were
cis. However, this cis conformation is found when chelating to a
metal, as it is N(15) and N(55) that binds rather than N(11) and
N(51). The crystal packing of 5 is shown in Fig. 7. As is apparent,
there is p–p stacking between the central aromatic rings with the
˚
closest distances ranging upwards from 3.42 A while the aliphatic
tetramethylcyclohexenyl rings form vertical columns.
Fig. 5 Extraction of Am(III) and Eu(III) by 5 + DMDOHEMA
(0.25 mol L-1) in n-octanol as a function of nitric acid concentration.
Whilst CyMe4-BTTP 5 clearly shows some selectivity for
Am(III) over Eu(III), the distribution coefficients for Am(III) are
significantly lower than those for the BTPs and the BTBPs. It is
likely that the lower extraction efficiency exhibited by 5 is due to
its inability to completely enclose the coordination sphere of the
metal, leaving vacant coordination sites to which other ligands
(e.g.: nitrate, water) can bind.
Crystallography
Fig. 7 The packing of CyMe4-BTTP 5 in the ‘a’ projection.
The X-ray crystal structure of CyMe4-BTTP 5 was determined and
is shown in Fig. 6 together with the atomic numbering scheme.‡
The five aromatic rings in CyMe4-BTTP 5 are approximately
We next attempted to prepare complexes of 5 with various
lanthanides by ad-mixture of DCM solutions of 5 with solutions
of lanthanide(III) nitrates in acetonitrile in both 1 : 1 and 2 : 1 ratios
of ligand : metal, followed by slow evaporation over several days.
However, in the case of europium, cerium and lanthanum(III)
nitrates, suitable crystals were not obtained despite repeated
attempts and any crystals that were obtained turned out to be of
‡ Crystal Data for 5: C37H41N9, Mr = 611.79, monoclinic, space◦group
˚
P21/c, a = 9.760(2), b = 10.731(2), c = 31.542(7) A, b = 95.13(2) , U =
3
-3
˚
3290.3(12) A , Dc = 1.235 g cm , 9533 independent reflections, 4741 data
(I > 2s(I)), Rint = 0.0578, R1 = 0.1556, wR2 (all data) = 0.1880. CCDC
756181. Crystal Data for [Y(5)(H2O)4](ClO4)3·2H2O: C37H53Cl3N9O18Y,
Mr = 1107.14, triclinic, Z = 2, a = 13.0611(5), b = 13.7135(◦7), c =
˚
15.7947(6) A, a = 108.832(4), b = 105.958(3), g = 98.035(4) , U =
data (I > 2s(I)), Rint = 0.0344, R1 = 0.2522, wR2 (all data) = 0.2828.
3
-3
˚
2491.91(19) A , Dc = 1.476 g cm , 13 491 independent reflections, 7005
CCDC 756182.
5176 | Dalton Trans., 2010, 39, 5172–5182
This journal is
The Royal Society of Chemistry 2010
©

View Online
the free ligand 5 rather than the desired complexes. We then used
the corresponding perchlorate salts but again suitable crystals of
the complexes could not be obtained with europium, lanthanum
or ytterbium(III). One notable exception however was yttrium(III)
perchlorate. Its complex with 5 is shown in Fig. 8. A 1 : 1 complex
was obtained despite two equivalents of ligand 5 being used during
the crystal growing. As shown the complex is a mononuclear
helicate with all five ligand coordination sites bound to the metal in
addition to four aqua ligands. The five donor nitrogen atoms form
a distorted equatorial plane with deviations of 0.498(3), -0.480(4),
-0.009(4), 0.503(4) and -0.512(3), respectively. The metal is --
˚
0.026(3) A from the plane with two water molecules above and
two below this plane. The helicity of the complex is indicated by
sucessive N–C–C–N torsion angles (from left to right in Fig. 8)
of 1(9), -15.6(9), -10.5 (9) and 1.7(1)◦. The bond lengths from
˚
the metal are Y–O 2.331(5), 2.356(5), 2.378(5), 2.381(5) A and
Y–N(11) 2.614(6), Y–N(21) 2.518(6), Y–N(31) 2.504(6), Y–N(41)
Fig. 9 Aromatic region of the 1H NMR spectrum of [Pd(5)(PF6)2]2 in
acetone-d6. Assignments were verified by H–H and C–H COSY.
˚
2.535(6), Y–N(51) 2.584(6) A.
in the aliphatic region while in the aromatic region, the methine
carbons appeared as 9 singlets.
These data are clearly consistent with an asymmetric environ-
ment for the ligand 5 in the complex. Two structures are then
possible. One is a dinuclear double-helicate (2 : 2 complex) similar
to the Pd(quinquepyridine)2+ complex published by Constable
et al.24 where each metal centre is in a square pyrimidal coordi-
nation geometry and each ligand 5 is divided into two binding
domains; a bidentate triazinylpyridyl domain and a tridentate
triazinylbipyridyl domain. This can be achieved by twisting about
the interannular axis between C-2 and C-2¢ of 5. The other
possible structure is a mononuclear 1 : 1 complex where the metal
is in a square planar coordination geometry and one terminal
triazene ring is non-coordinated. The electrospray ionisation mass
spectrum showed a mass peak at m/z = 358.6264 which is
consistent with the formula [Pd(5)]n2n+. The isotope distribution
pattern of this peak (see ESI) †is consistent with that calculated
for a 1 : 1 complex (n = 1) but additional mass peaks were also
observed which corresponded to those calculated for the 2 : 2
complex (n = 2). It thus appears that a mixture of 1 : 1 and 2 : 2
complexes is present under the electrospray ionisation conditions
used. Although definitive evidence in the form of a solid-state
structure could not be obtained, it appears that the complex is
a dinuclear double-helicate of 2 : 2 stoichiometry similar to the
quinquepyridine complex published by Constable et al.24
Fig. 8 Y3+ complex of 5 with ellipsoids at 50% probability. Counterions
have been omitted for clarity. Hydrogen atoms on water molecules were
not located and are not included.
We then attempted to obtain an X-ray structure of the palladium
complex of 5. However, despite repeated attempts at crystal
growing (slow evaporation from acetone solutions, diffusion of
diethyl ether into acetonitrile solutions), crystals suitable for X-
ray analysis could not be obtained. Subsequently, we recorded the
NMR and mass spectra of the complex. The aromatic region of the
1H NMR spectrum of the complex in deuterated acetone is shown
in Fig. 9. A total of 9 well-resolved resonances were observed
(three triplets and six double doublets). As in the parent ligand 5,
the three triplets were assigned to the protons at the 4-positions of
each ring while the remaining signals were assigned to the protons
at the 3- and 5-positions of each ring. The tentative assignments
were made as shown in Fig. 9 based on a comparison with the 1H
NMR spectrum of the corresponding quinquepyridine complex.24
In the aliphatic region, 8 singlets were observed corresponding to
each of the methyl groups while the methylene protons appeared
as poorly-resolved multiplets. In addition, the 13C NMR spectrum
displayed 7 methyl environments and 3 methylene environments
Nuclear magnetic resonance and theoretical studies
Solution-phase NMRstudieswere undertaken in orderto establish
the nature of the complexed species in solution. The stoichiometry
of the lanthanide complexes of 5 was deduced from nuclear
magnetic relaxation dispersion titrations as shown in Fig. 10.25,26
The decrease in relaxation rate (or relaxivity) upon formation of
a paramagnetic complex was used to establish its stoichiometry.25
As solvent molecules are removed from the paramagnetic centres,
their protons relax more slowly in the bulk of the solution.
Relaxivity is thus decreasing and a plateau is finally reached once
a complex is fully formed. This procedure has been successfully
used for the BTPs 125 but in the present case, the perchlorate
complex precipitated once a 2 : 1 5/Gd(III) ratio was reached. The
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5172–5182 | 5177
©

View Online
Fig. 10 NMR titration of Gd3+ by 5 (anhydrous CD3CN, 25 ◦C); ꢀ:
perchlorate salt, ᭢: nitrate salts. Solubility limits are indicated by vertical
dotted lines and relaxivity plateaux are tentatively presented as horizontal
dash lines.
Fig. 11 1H NMR spectrum of a 2 : 1 5:Eu(ClO4)3 mixture in anhydrous
CD3CN at 25 ◦C. Peak assignments: meta and para protons in the central
pyridine ring (᭹): -2.45 and -0.25 ppm; meta and para protons in the
outer pyridine rings (᭹): 14.85, 11.90 and 4.0 ppm; methyl protons (+):
2.90, 1.53, -0.29 and -4.83 ppm; methylene protons (¥): 1.53, 0.62, 0.45
and -0.50 ppm.
exact value of the relaxation plateau is thus uncertain but it seems
clear that a 2 : 1 ligand : metal complex is essentially formed as
we obtained a linear relaxivity decrease until this concentration
ratio was reached. Similarly, a precipitate was observed once the
ligand : metal ratio was between 0.8 and 1.5 in the presence of
nitrate ions. This could mean that a 1 : 1 complex is the major
species in anhydrous acetonitrile.
In view of the uncertainty of the stoichiometries of the
complexes, we initiated an electrospray mass spectrometry study
of the perchlorate solutions. The ES-MS spectrum of a solution of
Gd(ClO4)3 with 2.5 equivalents of 5 is shown in the ESI. †Intense
mass peaks were observed for the protonated free ligand and for its
sodium complex together with a peak assigned to the bis-complex
at m/z = 460.19. In contrast, the electrospray mass spectrum of a
1 : 1 mixture of 5 and Yb(NO3)3 displayed a peak at m/z = 909.27
corresponding to the 1 : 1 complex [Yb(5)(NO3)2]+ in addition to
mass peaks due to the protonated free ligand and its sodium
complex. No peak corresponding to the bis-complex was observed
even for 1 : 3 mixtures of Yb(NO3)3 and 5. Expanded views of the
experimental and computed mass patterns of the metal complex
peaks are shown in the ESI.†
We next recorded the NMR spectrum of the Eu(III) bis-complex
in perchlorate medium. The Eu(III) ion is not suitable for a
conformational analysis because the dipolar contribution to the
paramagnetic shifts is comparable to the contact contribution, but
this ion causes minimal broadening of the NMR peaks and well-
resolved spectra are usually obtained.26,27 A qualitative analysis of
the NMR spectra is thus greatly facilitated. As shown in Fig. 11,
this was indeed the case for the bis-complex of 5. As expected,
the spectrum of the Eu(III) complex exhibited five resonances due
to the terpyridine moiety. Surprisingly, four resonances due to the
methyl groups and four resonances due to four different types of
methylene protons were also observed. In contrast, the free ligand
5 and its Lu(III) complex exhibited a pair of singlets for the methyl
groups and an 8-proton multiplet for the methylene protons as
expected for rapidly inverting cyclohexenyl groups.
used in the present work. Rapid inversion of the cyclohexenyl
groups appears to be prevented by steric hindrance, a phenomenon
that is readily observed for the Eu(III) complex because of the large
induced paramagnetic shifts. An expanded view of the methylene
resonances that clearly shows the methylene coupling patterns is
shown in the ESI. †Two resonances only would be expected if the
tetramethylcyclohexenyl groups were conformationally mobile.
We also recorded a H–H COSY spectrum of the bis-complex
with Eu(III) in order to show that there are indeed four types
of methylene protons that are all coupled together (see ESI).
†Crosspeaks are found for the four types of methylene protons
as well as for the two groups of terpyridine protons.
Fig. 12 presents the 1H NMR spectrum of the Yb(III) perchlorate
bis-complex of 5. There are again resonances for four different
methylene protons and four different methyl groups. Coupling
patterns are observed for the four methylene peaks (see ESI), †an
unusual feature because the NMR resonances of Yb(III) complexes
are usually so broad that J couplings cannot be observed. The
COSY spectrum clearly shows all the expected cross peaks and
a
13C-1H HSQC correlation confirms the peak assignments and
clearly shows which of the methylene peaks are due to two
protons on the same carbon atom (see ESI). †The latter spectrum
displays two 13C peaks each connected with two methylene protons
as expected for rigid cyclohexenyl units. The methylene proton
coupling patterns (see ESI) †are also in keeping with a rigid
structure. No spectral changes were observed in the limited
temperature range available with acetonitrile and we were thus
unable to estimate the energy barrier for conformation inversion.
Yb(III) is the ideal ion for a conformational analysis as it
induces essentially pure dipolar shifts.26,28 The solution structures
of its complexes can be inferred from these shifts provided the
complexes under study display some form of symmetry and are
conformationally rigid.27 We thus decided to deduce the solution
structure of the bis-complex of 5 from the relative magnitude of
the paramagnetic shifts induced by Yb(III). A molecular model of
the bis-complex was obtained starting with structures built with
It thus seems that the aliphatic tetramethylcyclohexenyl groups
of the metal complexes are constrained on the NMR time scale
5178 | Dalton Trans., 2010, 39, 5172–5182
This journal is
The Royal Society of Chemistry 2010
©

View Online
3+
Fig. 14 Space-filling model of the Yb(5)2 complex. The arrow shows
one of the interatomic distances that is smaller than the sum of the van
˚
der Waals radii (2.75 instead of 3.05 A).
Fig. 12 1H NMR spectrum of a 2 : 1 5:Yb(ClO4)3 mixture in anhydrous
CD3CN at 25 ◦C. Peak assignments: meta and para protons in the central
pyridine ring (᭹): -14.38 and -14.69 ppm; meta and para protons in the
outer pyridine rings (᭹): 40.69, 25.51 and 8.52 ppm; methyl protons (+):
4.84, 1.83, -5.12 and -16.38 ppm; methylene protons (¥): -0.32, -2.40,
-3.22 and -6.04 ppm.
ring is located above the center of a square face formed by the
nitrogen atoms of the two adjacent pyridine rings and by the
nitrogen atoms of two triazine units belonging to the other ligand
in the bis-complex. A bicapped square-antiprismatic arrangement
is the most favored geometry when one sphere is surrounded
by 10 spheres31 and the optimized geometry of Yb(5)2 is in
3+
the Chem3D program (CambridgeSoft, MA, USA) with either
two planar ligands perpendicular to each other or at an angle or
with the aromatic rings in a twisted conformation so as to obtain
keeping with the stereochemical arrangement predicted by theory.
It should also be noted that some of the aromatic rings in
the ligands are slightly bowed, a feature that is often found in
complexes with high coordination numbers even in phenanthroline
rings.32 A bowed arrangement of 5 is also observed in the crystal
structure of Y(5)(H2O)4](ClO4)3 (Fig. 8). A ligand deformation is
of course detrimental to the thermodynamic stability of complexes
but is probably imposed by the necessity to rearrange one or
more ligands so as to completely encapsulate the metal ion.
The optimised geometry of the [Yb(5)2]3+ complex is similar to
the coil structure found for a Eu(III) homoleptic bis-complex
with the terpytz ligand (Fig. 15) that was published by Giraud
et al. (EFAXUF in the Cambridge Structural Database).33 In this
crystallographic structure, the metal ion occupies the center of an
irregular bicapped trigonal prismatic assembly of nitrogen atoms
that is close to a square antiprismatic arrangement.
˚
metal-nitrogen distances close to 3 A.
All structures were optimized by a force field approach using
parameters published by Cundari et al. for Gd(III) complexes.29
Finally, full optimizations were performed using the SPARKLE
Yb(III) parameter set that was proposed by Simas et al.30 and that
is included in MOPAC 2009. The minimum energy geometry with
˚
expected Yb(III)–N distances (2.41–2.43 A) was obtained starting
from the twisted arrangement.
Different views of the optimized structure are presented below
in Fig. 13 and 14. The structure is a mononuclear double-
helicate with the metal ion at the center of a bicapped square
antiprismatic arrangement of donor atoms. Each central pyridine
Fig. 15 Structure of the terpytz ligand.
The induced paramagnetic shifts di were computed from the
3+
3+
spectra of Yb(5)2 and of diamagnetic Lu(5)2 that was used
as a reference. The geometric factors in eqn (1) that relate these
shifts with magnetic susceptibility terms were deduced from the
structure shown in Fig. 13.
⎡
⎤
⎥
⎛
⎞
⎟
⎠
1
1
2
xx + cyy 3cos2 q −1 + cxx − cyy sin2 qi cos2yi
⎟
⎜
Fig. 13 Optimized structure of the Yb(5)23+ bis-complex. The protons on
one of the ligands have been removed for clarity. The two square faces of
the square antiprism are shown.
⎢
di =
c
−
c
⎟
(
)
(
)
⎜
zz
i
12pr3
⎟
⎢⎜
⎥
⎦
⎝
⎣
(1)
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5172–5182 | 5179
©

View Online
where cxx cyy and czz are magnetic susceptibility terms character-
istic of the investigated complex and thus identical for all nuclei.
The factors ri, qi and ji are the polar coordinates of proton i
giving rise to chemical shift di in the set of axes of the magnetic
susceptibility tensor with the metal ion at the centre. The full
the structure of the lanthanide nitrate complex of 5. The NMR
titration data (see Fig. 10) suggest that a 1 : 1 complex is formed
in the presence of nitrate ions and the crystallographic structure
of the 1 : 1 complex with Y(ClO4)3 has been obtained (Fig. 8). One
may then wonder whether this structure is maintained in solution.
The spectrum of a 1 : 1 mixture of 5:Yb(NO3)3 in acetonitrile
is presented in Fig. 17. This spectrum displays resonances for the
Yb(III) complex in addition to peaks due to the free ligand. In our
experimental conditions, there is a competition between the nitrate
ions and the ligand 5 for the complexation of the metal ion. Adding
an excess of Yb(III) displaces the equilibrium towards the 1 : 1
complex and the free ligand resonances completely disappear. The
Yb(5)3+ peaks remain broad in all conditions and their assignment
is difficult except from the relative areas. Moreover, none of the
2D NMR techniques applied to the perchlorate complex could be
used (COSY, HSQC) because of the broadness of the peaks. Only
the protons due to the methyl groups and the para proton in the
central pyridine group can be assigned with certainty. Moreover,
the number of resonances clearly indicates that the tetramethyl-
cyclohexenyl groups are inverting rapidly on the NMR scale we
used in contrast with what we observed for the Yb(III) perchlorate
complex. As expected for a mono-complex, the structure is thus
much less crowded and the tetramethylcyclohexenyl rings are free
to invert rapidly (it should however be noted that the shift range
is 25 ppm instead of 60 ppm for the perchlorate complex. Peak
coalescence is thus more easily reached).
3+
dipolar equation has to be used because Yb(5)2 is not an axially
symmetric complex.26,27 However, the orientation of the magnetic
susceptibility axes is well-defined as one of them must coincide
with the C2 axis joining the two central pyridine nitrogen atoms
and the metal ion. The two other magnetic susceptibility axes must
be located in the plane perpendicular to the C2 axis and must be
oriented so as to obtain identical geometric factors for protons
giving rise to the same NMR resonance in the two ligands. The x
(or y) axis thus bisects one of the square faces of the antiprism.
The magnetic susceptibility terms in eqn (1) were computed
by solving a set of 13 linear equations established for the 13
resonances displayed by [Yb(5)2]3+ for which the geometric factors
were deduced from the geometric model shown in Fig. 13. An
excellent correlation was obtained between the calculated and the
experimental shifts as shown in Fig. 16 with
1
(2)
czz
−
c
xx + cyy = (−2883 187) A3
(
)
2
cxx - cyy = (2490 301) A3
(3)
Fig. 16 Correlation between the calculated and experimental paramag-
netic shifts in the 1H NMR spectrum of [Yb(5)2](ClO4)3. 99% intervals are
indicated by dotted lines. (slope: 1.00, intercept: 2.13).
Fig. 17 1H NMR spectrum of a 1 : 1 mixture of 5:Yb(NO3)3 in anhydrous
CD3CN at 25 ◦C. Tentative peak assignments: meta and para protons of the
pyridine rings (᭹): 16.18, 15.96, 9.96, 0.91 and -2.08 ppm; methyl protons
(+): 7.20 and -0.79 ppm; methylene protons (¥); 4.06 and 3.11 ppm. The
narrow resonances between 8 and 9 ppm and between 1 and 2 ppm are due
to the free ligand, acetonitrile and a trace of water.
The NMR analysis fully supports the structure obtained by
the force field calculations and the geometry of the bis-complex
3+
Yb(5)2 is thus very close to or identical to the structure shown
in Fig. 13 and 14 above. The ring-inversion barrier of cyclohexene
is rather low (23 kJ mol-1) and the aliphatic substituents of the
triazine moieties of 5 were not expected to be rigidified. Several
inter-ligand distances involving the cyclohexenyl units are smaller
than the sum of the van der Waals distances as shown in Fig. 14.
Although it is partially stabilized by p–p interactions between
aromatic rings, the Yb(III) bis-complex is sterically crowded and
is unable to withstand the competition with other coordinating
ligands as shown below in the case of the nitrate ion.
Despite these uncertainties, we deduced the dipolar geometric
factors in eqn (1) above from the crystallographic structure of
the Y(5)3+ mono-complex (Fig. 8) and we made a fit between
these factors and the induced paramagnetic shifts of the Yb(III)
nitrate complex. A reasonably good agreement was obtained (see
ESI) †and the solution and solid state structures are therefore
probably similar. The magnetic susceptibility axes were oriented
as follows: the z axis joins the metal ion with the nitrogen atom
and the para carbon atom of the central pyridine unit, the x
(or y) axis is perpendicular to the plane formed by the central
pyridine ring and the metal ion. This analysis may explain the
The partitioning of actinides from lanthanides is performed
in nitric acid solutions and information is thus also needed on
5180 | Dalton Trans., 2010, 39, 5172–5182
This journal is
The Royal Society of Chemistry 2010
©

View Online
relatively low distribution coefficients observed in nitrate media
(Fig. 3) as the 1 : 1 complex of 5 with Am(III) would be expected
to be considerably less hydrophobic than the corresponding 1 : 2
complex.
Better agreement between the experimental and the calculated
paramagnetic shifts could probably be obtained by orienting
differently the magnetic susceptibility axes. However, the uncer-
tainties in the assignment of the NMR resonances and on the
exact orientation of the magnetic susceptibility tensor in a low
symmetry structure would remain and efforts in that direction
would not be very fruitful. At this stage, one can only say that the
solution and solid state structures of Yb(5)3+ are probably close.
2 C. Madic, M. J. Hudson, J.-O. Liljenzin, J.-P. Glatz, R. Nannicini, A.
Facchini, Z. Kolarik and R. Odoj, Prog. Nucl. Energy, 2002, 40, 523–
526; C. Hill, D. Guillaneux, L. Berthon and C. Madic, J. Nucl. Sci.
Technol., 2002, (Sup. 3), 309–312.
3 J. Magill, V. Berthou, D. Haas, J. Galy, R. Schenkel, H.-W. Wiese, G.
Heusener, J. Tommasi and G. Youinou, Nuclear Energy, 2003, 42, 263–
277; International Atomic Energy Agency, Technical Reports Series,
ISSN 0074-1914 no. 435, Vienna, Austria, 2004.
4 C. Madic, B. Boullis, P. Baron, F. Testard, M. J. Hudson, J-O. Liljenzin,
B. Christiansen, M. Ferrando, A. Facchini, A. Geist, G. Modolo, A. G.
Espartero and J. De Mendoza, J. Alloys Compd., 2007, 444–445, 23–27;
C. Ekberg, A. Fermvik, T. Retegan, G. Skarnemark, M. R. S. Foreman,
M. J. Hudson, S. Englund and M. Nilsson, Radiochim. Acta, 2008, 96,
225–233; Z. Kolarik, Chem. Rev., 2008, 108, 4208–4252.
5 C. Boucher, M. G. B. Drew, P. Giddings, L. M. Harwood, M. J. Hudson,
P. B. Iveson and C. Madic, Inorg. Chem. Commun., 2002, 5, 596–599;
M. G. B. Drew, C. Hill, M. J. Hudson, P. B. Iveson, C. Madic and
T. G. A. Youngs, Dalton Trans., 2004, 244–251.
6 M. G. B. Drew, M. R. S. J. Foreman, C. Hill, M. J. Hudson and
C. Madic, Inorg. Chem. Commun., 2005, 8, 239–241; M. Nilsson, C.
Ekberg, M. Foreman, M. Hudson, J-O. Liljenzin, G. Modolo and G.
Skarnemark, Solvent Extr. Ion Exch., 2006, 24, 823–843; M. R. S.
Foreman, M. J. Hudson, M. G. B. Drew, C. Hill and C. Madic, Dalton
Trans., 2006, 1645–1653.
7 A. Geist, C. Hill, G. Modolo, M. R. St, J. Foreman, M. Weigl, K.
Gompper and M. J. Hudson, Solvent Extr. Ion Exch., 2006, 24, 463–
483.
Conclusions
In conclusion, we have synthesized the first example of a (1,2,4-
triazin-3-yl)-2,2¢:6¢,2¢¢-terpyridine ligand and its solvent extraction
chemistry has been examined. Low binding affinities for Am(III)
but good selectivities for Am(III) over Eu(III) are observed in
n-octanol in the absence of a phase-modifier. The addition of
2-bromohexanoic acid increases the distribution coefficient at
low acidities. The stoichiometries and solution structures of the
complexes with Gd(III), Eu(III) and Yb(III) have been determined
by nuclear magnetic relaxation dispersion titrations, 1D and 2D
NMR techniques and mass spectrometry. In the presence of
perchlorate ions the ligand forms highly crowded 1 : 2 complexes
where the aliphatic substituents are conformationally immobile
on the NMR timescale (250–400 MHz). The solution structure
of the 1 : 2 complex of 5 with Yb(ClO4)3 has been deduced from
the relative magnitude of the induced paramagnetic shifts. The
complex is a mononuclear double helicate where the metal adopts
a bicapped square antiprism coordination geometry. In contrast,
in the presence of nitrate ions the ligand forms less-crowded 1 : 1
complexes where the aliphatic substituents are conformationally
mobile on the NMR time scale. The nitrate ions compete
effectively with the ligand for coordination sites on the metal
leading to poorly-extracted complexes of low hydrophobicity.
A mononuclear helical 1 : 1 complex is formed between 5 and
Y(ClO4)3 and its X-ray structure was determined. The Pd(II)
complex of 5 has been synthesized and its NMR and mass spectra
are consistent with a dinuclear double-helical structure.
8 D. Magnusson, B. Christiansen, M. R. S. Foreman, A. Geist, J.-P.
Glatz, R. Malmbeck, G. Modolo, D. Serrano-Purroy and C. Sorel,
Solvent Extr. Ion Exch., 2009, 27, 97–106.
9 E. C. Constable, Adv. Inorg. Chem., 1986, 30, 69–121; E. C. Constable,
Tetrahedron, 1992, 48, 10013–10059; E. C. Constable, in Comprehensive
Supramolecular Chemistry, ed. J. P. Sauvage, and M. W. Hosseini,
Elsevier, Oxford, 1996, vol. 9, pp. 213-252; C. Piguet, G. Bernardinelli
and G. Hopfgartner, Chem. Rev., 1997, 97, 2005–2062; M. Albrecht,
Chem. Rev., 2001, 101, 3457–3498; R. P. Thummel, in Comprehensive
Coordination Chemistry II, ed. J. A. McCleverty and T. J. Meyer,
Elsevier, Oxford, 2004, vol. 1, pp. 41-53; R.-A. Fallahpour, Curr. Org.
Synth., 2006, 3, 19–39.
10 E. C. Constable, Prog. Inorg. Chem., 1994, 42, 67–138; E. C. Constable,
Pure Appl. Chem., 1996, 68, 253–260; R. Ziessel, Synthesis, 1999, 1839–
1865; E. C. Constable and C. E. Housecroft, Chimia, 1999, 53, 187–191;
R.-A. Fallahpour, Chimia, 1999, 53, 195; R. Ziessel, Coord. Chem. Rev.,
2001, 216–217, 195–223; E. C. Constable, V. Chaurin, C. E. Housecroft
and A. Wirth, Chimia, 2005, 59, 832–835.
11 P. Jones, G. B. Villeneuve, C. Fei, J. DeMarte, A. J. Haggarty, K. T. Nwe,
D. A. Martin, A-M. Lebuis, J. M. Finkelstein, B. J. Gour-Salin, T. H.
Chan and B. R. Leyland-Jones, J. Med. Chem., 1998, 41, 3062–3077;
K. Kikuchi, S. Hibi, H. Yoshimura, N. Tokuhara, K. Tai, T. Hida, T.
Yamauchi and M. Nagai, J. Med. Chem., 2000, 43, 409–419.
12 K. Ru¨hlmann, Synthesis, 1971, 236–250.
13 J. Strating, S. Reiffers and H. Wynberg, Synthesis, 1971, 209–211; J.
Strating, S. Reiffers and H. Wynberg, Synthesis, 1971, 211–212.
14 D. D. Coffman, E. L. Jenner and R. D. Lipscomb, J. Am. Chem. Soc.,
1958, 80, 2864–2874; D. H. Bremner, S. R. Mitchell and H. Staines,
Ultrason. Sonochem., 1996, 3, 47–52; D. H. Bremner, A. E. Burgess and
F-B. Li, Green Chem., 2001, 3, 126–130.
Acknowledgements
15 V-M. Mukkala, C. Sund, M. Kwiatkowski, P. Pasanen, M. Ho¨gberg,
J. Kankare and H. Takalo, Helv. Chim. Acta, 1992, 75, 1621–1632; C.
Galaup, J-M. Couchet, S. Bedel, P. Tisne`s and C. Picard, J. Org. Chem.,
2005, 70, 2274–2284; J. M. Veauthier, C. N. Carlson, G. E. Collis, J. L.
Kiplinger and K. D. John, Synthesis, 2005, 2683–2686.
16 R. P. Thummel and Y. Jahng, J. Org. Chem., 1985, 50, 3635–
3636.
17 W. K. Fife, J. Org. Chem., 1983, 48, 1375–1377.
We thank the Nuclear Fission Safety Program of the European
Union for support under the ACSEPT (FP7-CP-2007-211267)
contract. We also thank the EPSRC and the University of Reading
for funds for the X-Calibur system. The Belgian team also
acknowledge the financial support of the Fonds de la Recherche
Scientifique (FNRS) of Belgium.
18 CrysAlis, (2006) Oxford Diffraction Ltd., Abingdon, UK.
19 G. M. Sheldrick, Acta. Crystallogr., 2008, A64, 112–122.
20 ABSPACK program (2007), Oxford Diffraction Ltd., Abingdon, UK.
21 F. H. Case, J. Org. Chem., 1965, 30, 931–933.
22 Z. Kolarik, U. Mullich and F. Gassner, Solvent Extr. Ion Exch., 1999,
17, 23–32.
23 D. Serrano-Purroy, P. Baron, B. Christiansen, R. Malmbeck, C. Sorel
and J.-P. Glatz, Radiochim. Acta, 2005, 93, 351–355.
24 E. C. Constable, S. M. Elder, J. Healy, M. D. Ward and D. A. Tocher,
J. Am. Chem. Soc., 1990, 112, 4590–4592.
References
1 A. Facchini, L. Amato, G. Modolo, R. Nannicini, C. Madic and P.
Baron, Sep. Sci. Technol., 2000, 35, 1055–1068; D. Serrano-Purroy, B.
Christiansen, J.-P. Glatz, R. Malmbeck and G. Modolo, Radiochim.
Acta, 2005, 93, 357–361; G. Modolo, H. Vijgen, D. Serrano-Purroy, B.
Christiansen, R. Malmbeck, C. Sorel and P. Baron, Sep. Sci. Technol.,
2007, 42, 439–452.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5172–5182 | 5181
©

View Online
25 M. J. Hudson, C. E. Boucher, D. Braekers, J. F. Desreux, M. G. B. Drew,
M. R. St, J. Foreman, L. M. Harwood, C. Hill, C. Madic, F. Marken
and T. G. A. Youngs, New J. Chem., 2006, 30, 1171–1183.
26 I. Bertini, C. Luchinat, G. Parigi, Solution NMR of Paramagnetic
Molecules, Elsevier, Amsterdam, 2001.
27 J. F. Desreux and C. N. Reilley, J. Am. Chem. Soc., 1976, 98, 2105–2109;
R. S. Ranganathan, N. Raju, H. Fan, X. Zhang, M. F. Tweedle, J. F.
Desreux and V. Jacques, Inorg. Chem., 2002, 41, 6856–6866.
28 J. A. Peters, J. Huskens and D. J. Raber, Prog. Nucl. Magn. Reson.
Spectrosc., 1996, 28, 283–350.
29 T. R. Cundari, E. W. Moody and S. O. Sommerer, Inorg. Chem., 1995,
34, 5989–5999.
30 G. B. Rocha, R. O. Freire, N. B. da Costa Jr., G. F. de Sa´ and A. M.
Simas, Inorg. Chem., 2004, 43, 2346–2354.
31 D. L. Kepert, Inorganic Stereochemistry, Springer-Verlag, Berlin,
1982.
32 N. E. Dean, R. D. Hancock, C. L. Cahill and M. Frisch, Inorg. Chem.,
2008, 47, 2000–2010.
33 M. Giraud, E. S. Andreiadis, A. S. Fisyuk, R. Demadrille, J. Pecaut, D.
Imbert and M. Mazzanti, Inorg. Chem., 2008, 47, 3952–3954.
5182 | Dalton Trans., 2010, 39, 5172–5182
This journal is
The Royal Society of Chemistry 2010
©