Exploring electronic effects on the partitioning of actinides(III) from lanthanides(III) using functionalised bis-triazinyl phenanthroline ligands

Article abstract of DOI:10.1039/c6dt02474b

The first examples of 4,7-disubstituted 2,9-bis(5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-1,2,4-benzo-triazin-3-yl)-1,10-phenanthroline (CyMe₄-BTPhen) ligands are reported herein. Evaluating the kinetics, selectivity and stoichiometry of actinide(iii) and lanthanide(iii) radiotracer extractions has provided a mechanistic insight into the extraction process. For the first time, it has been demonstrated that metal ion extraction kinetics can be modulated by backbone functionalisation and a promising new CHON compliant candidate ligand with enhanced metal ion extraction kinetics has been identified. The effects of 4,7-functionalisation on the equilibrium metal ion distribution ratios are far more pronounced than those of 5,6-functionalisation. The complexation of Cm(iii) with two of the functionalised ligands was investigated by TRLFS and, at equilibrium, species of 1?:?2 [M?:?L] stoichiometry were observed exclusively. A direct correlation between the E_LUMO-E_HOMO energy gap and metal ion extraction potential is reported, with DFT studies reaffirming experimental findings.

Full text of DOI:10.1039/c6dt02474b

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: A. C. Edwards, C.
Wagner, A. Geist, N. A. Burton, C. A. Sharrad, R. W. Adams, R. Pritchard, P. J. Panak, R. Whitehead and L.
Harwood, Dalton Trans., 2016, DOI: 10.1039/C6DT02474B.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/dalton

Page 1 of 11
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02474B
Exploring Electronic Effects on the Partitioning of Actinides(III) from
Lanthanides(III) using Functionalised Bis-triazinyl Phenanthroline Lig-
ands
Alyn C. Edwards,† Christoph Wagner,‡,ℓ Andreas Geist,‡ Neil A. Burton,† Clint A. Sharrad,∆ Ralph W. Ad-
ams,† Robin G. Pritchard,† Petra J. Panak,‡,ℓ, Roger C. Whitehead†* and Laurence M. Harwood§*
† School of Chemistry, The University of Manchester, Oxford Road, Manchester, M13 9PL, U.K.
‡ Karlsruher Institut für Technologie (KIT-INE), Institut für Nukleare Entsorgung, Hermann-von-Helmholtz-Platz 1, D-76344
Eggenstein-Leopoldshafen, Germany.
ℓ University of Heidelberg, Department of Physical Chemistry, Im Neuenheimer Feld 253, 69120, Germany.
∆ School of Chemical Engineering, The University of Manchester, Oxford Road, Manchester, M13 9PL, U.K.
§ Department of Chemistry, The University of Reading, Reading, RG6 6AD, U.K.
ABSTRACT: The first examples of 4,7-disubstituted 2,9-bis(5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-1,2,4-benzo-triazin-3-yl)-1,10-
phenanthroline (CyMe4-BTPhen) ligands are reported herein.
Evaluating the kinetics, selectivity and stoichiometry of acti-
nide(III) and lanthanide(III) radiotracer extractions has provided a
mechanistic insight into the extraction process. For the first time,
it has been demonstrated that metal ion extraction kinetics can
be modulated by backbone functionalisation and a promising new
CHON compliant candidate ligand with enhanced metal ion ex-
traction kinetics has been identified. The effects of 4,7-
functionalisation on the equilibrium metal ion distribution ratios
are far more pronounced than those of 5,6-functionalisation. The
complexation of Cm(III) with two of the functionalised ligands was
investigated by TRLFS and, at equilibrium, species of 1:2 [M:L]
stoichiometry were observed exclusively. A direct correlation
between the ELUMO-EHOMO energy gap and metal ion extraction
potential is reported, with DFT studies reaffirming experimental findings.
geological repository.2,6,7 Consequently, substantial research
INTRODUCTION
has been devoted to developing the non-trivial separation of
Energy generation by nuclear fission is one of the likely key
the minor actinides, Am and Cm, from the chemically similar
components in an energy mix that will reduce greenhouse gas
lanthanides found as fission products in spent nuclear fuel by
(GHG) emissions while still providing sufficient energy for the
solvent extraction methods.8–10 Despite this extensive body of
world’s population at a reasonable cost.1–3 Next generation
work which has identified extractants that can partition minor
nuclear fission reactors are being developed with aims to be
actinides from the lanthanides with separation factors between
more energy intensive and accident tolerant while maximizing
the fuel resources available.4,5 Examples include the deploy-
20-2000, little is known with regards to the essential electronic
ment of fast breeder reactors.2,5 In order to support such activi-
and steric requirements of the molecular extractants that result
in this selectivity. 8–11
ty, full actinide recycle strategies are needed where the minor
It has been postulated that one of the major contributors to
this selectivity is the more radially extended nature of the 5f-
actinide orbitals versus the 4f-lanthanide orbitals.12,13 It is
believed that this subtle difference is exploited by the soft
donor ligands, leading to increased ligand-actinide bond cova-
lency and thus selectivity for the actinides. 15N NMR spectro-
scopic studies utilising a 15N labelled bis-triazinyl pyridine (BTP
1: Figure 1) analogue, a N-donor extractant that exhibits selec-
tivity for minor actinides over lanthanides, have provided some
actinides (MA: Np, Am and Cm) are partitioned from spent
nuclear fuel and made available for use as fuel in these ad-
vanced reactors.2,3 Despite only accounting for approximately 1
wt % of spent fuel, Pu and the minor actinides are the main
contributors to the long-term radiotoxicity and thermal power
in spent nuclear fuel from most current nuclear reactors. The
removal and burning of the minor actinides can significantly
reduce the volume and long-term thermal and radioactive load
of the highly active waste intended for disposal in a deep

Dalton Transactions
Page 2 of 11
View Article Online
DOI: 10.1039/C6DT02474B
of the first direct evidence for this increased degree of covalen-
cy.14 Computational studies have identified possible discrepan-
cies with regards to the nature of covalency in actinide-ligand
bonding which shows the complexity of interpreting interac-
tions that involve the f-orbitals.15 Consequently, an improved
understanding on how systematic changes to an extractant
molecular framework impacts on minor actinide/lanthanide
separations performance is needed to understand further the
role of f-orbital covalency, if any, in such partitioning processes.
solvent extraction behavior of three, 4,7-disubstituted CyMe4-
BTPhen analogues and how electronic modulation of this
extractant template impacts on minor actinide/lanthanide
partitioning performance.
RESULTS
Synthesis of 4,7-disubstituted CyMe4-BTPhen analogues. To
introduce the desired functionality into the 1,10-
phenanthroline core, the versatile intermediate 4,7-dichloro-
2,9-dimethyl-1,10-phenanthroline 8 was prepared using the
synthetic protocol described by Ulven et al. (Scheme 1).22
Batches of intermediate 8 were functionalised with electron
donating methoxy-moieties via the in situ generation of sodium
methoxide (9). The introduction of a phenyl substituent was
achieved using a Pd-catalysed Suzuki-Miyaura cross coupling
with phenylboronic acid (10). The 4,7-diphenyl substituted
compound was designed to be an intermediate between the
stronger electronically modifying effects of the chloro- and
methoxy-substituents. It was reasoned that this modification
would also enhance solubility in hydrocarbon based solvents by
reducing molecular polarity and planarity.23
The set of extractants stemming from the initial bis-triazinyl
pyridine (BTP) molecules (1, Figure 1) reported by Kolarik et al.
showed considerable promise for minor actinide/lanthanides
separations.16,17 Subsequently, the tetradentate BTBP (2) and
BTPhen (3) ligand systems, bearing the radio- and hydrolytic
tolerant CyMe4 moiety have demonstrated very promising
partitioning capabilities, and have become benchmark extrac-
tion reagents for the European SANEX (Selective ActiNide
EXtraction) process.8–10,16–19 To explore the influence of elec-
tronic requirements on minor actinide/lanthanide separations
we have investigated alternatives to the cis-locked 1,10-
phenanthroline derived, CyMe4-BTPhen (3, Figure 1), which has
relatively fast complexation kinetics and profound selectivity
for the minor actinides (SFAm/Eu = 68-400).11,12,20
To date, all reported syntheses of CyMe4-BTPhen (3) and its
analogues (4-5) have involved the benzylic oxidation of com-
mercially available neocuproine using excess SeO2.20,21,24,25 In
the hands of Ulven and colleagues, application of this oxidation
procedure to substrate 8 provided an unsymmetrical oxy-chloro
compound.22 These findings combined with the highly toxic
nature of SeO2 and the reaction residues, rendered this particu-
lar route unsuitable for the synthesis of 4,7-Cl-CyMe4-BTPhen
(33) and a new synthetic protocol was developed (Scheme 2).
This alternative route was then applied to the parent CyMe4-
BTPhen (3), 4,7-MeO-CyMe4-BTPhen (34) and 4,7-phenyl-
CyMe4-BTPhen (35). This alternative approach to benzylic
functionalisation commenced with the per-chlorination of the
two methyl groups of substrates 8-11 using excess NCS, with
benzoyl peroxide (BPO) as initiator. Given that the oxidation
level of the symmetrical tetra-chlorinated intermediate is
equivalent to that of the bis-aldehyde, attempts were made to
stop the sequence at this stage to streamline the synthetic
route. Unfortunately, the symmetrical tetra-chlorinated inter-
mediated could not be isolated and numerous different chlo-
rinated species were obtained when the process was not al-
lowed to reach completion. The subsequent methanolysis of
intermediates 12-15 using concentrated H2SO4 and MeOH
afforded bis-esters 16-19. The synthesis of bis-amides 20-23
was carried out by stirring compounds 16-19 in concentrated
NH4OH at ambient temperature for 3 days.
Figure 1. Molecular Structures of CyMe4-BTP 1, CyMe4-BTBP 2 and
CyMe4-BTPhen analogues 3-5.
Screening of the 5 and 5,6-bromo-functionalised CyMe4-
BTPhen ligands 4 and 5 (Figure 1) revealed the substantial
impact that subtle electronic changes can have on the pre-
equilibrium partitioning behaviour, providing SFAm/Eu = 250‡ and
800‡ respectively.21 Based on these previous findings, it was
postulated that substitution at the 4,7-positions of the 1,10-
phenanthroline core would provide enhanced electronic modu-
lation due to the para-positioning of the substituents to the
phenanthroline N-atoms. Herein, we report the synthesis and

Page 3 of 11
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02474B
Scheme 1: Synthesis of 4,7-disubstituted CyMe4-BTPhen
analogues.
h; (ii) MeOH, 65 °C, 2 h, 74 %; (h) NH4OH (29 %), NH4Cl, 25 °C,
72 h, 81 %; (i) (i) (COCl)2, DMF, 0 °C, 6 h; (ii) pyridine, 25 °C, 1 h,
50 %; (j) N2H4.H2O, EtOH, 25 °C, 72 h, 54 %; (k) 31 (2.2 eqvs),
EtOH, 78 °C, 3 h, 34 %. Synthesis of 4,7-phenyl-CyMe4-BTPhen
35. Reagents and conditions: (f) (PhCOO)2, NCS, CHCl3, 62 °C, 24
h, 80 %; (g) (i) H2SO4, 95 °C, 2 h; (ii) MeOH, 65 °C, 1 h, 76 %; (h)
NH4OH (29 %), NH4Cl, 25 °C, 72 h, 67 %; (i) (i) (COCl)2, DMF, 0 °C,
6 h; (ii) pyridine, 25 °C, 1 h, 69 %; (j) N2H4.H2O, EtOH, 25 °C, 72
h, 77 %; (k) 31 (2.2 eqvs), EtOH, 78 °C, 3 h, 38 %. Synthesis of
CyMe4-BTPhen (3). Reagents and conditions: (f) Neocuproine,
(PhCOO)2, NCS, CHCl3, 62 °C, 72 h, 99 %; (g) (i) H2SO4, 95 °C, 4 h;
(ii) MeOH, 65 °C, 1.5 h, 98 %; (h) NH4OH (29 %), NH4Cl, 25 °C, 72
h, 67 %; (i) (i) (COCl)2, DMF, 0 °C, 6 h; (ii) pyridine, 25 °C, 0.5 h,
54 %; (j) N2H4.H2O, EtOH, 25 °C, 72 h, 78 %; (k) 32 (2.2 eqvs),
EtOH, 78 °C, 72 h, 63 %.
The extended reaction periods of these transformations is
attributed to the insoluble nature of the substrates in concen-
trated aqueous NH4OH. Use of elevated temperatures in sealed
tubes, and increased loadings of NH4Cl, failed to enhance the
reaction rate. Dehydration of compounds 20-23 using the
Vilsmeier-Haack reagent afforded bis-nitrile compounds 24-27
in respectable yields (>50 %). The penultimate step was carried
out in accordance with the procedure reported by Case et
al.18,26,27 This procedure involved stirring compounds 24-27 in
N2H4.H2O and EtOH for 3 days at ambient temperature, provid-
ing bis-aminohydrazide compounds 28-31. The final step re-
quired the coupling of 28-31 with di-ketone 32 (purchased from
Technocomm Ltd).28 This step proved particularly problematic
with compound 28, presumably due to the electron-
withdrawing nature of the chloro-substituents. When the
literature procedure, involving refluxing in THF and Et3N, was
applied to substrate 28, it provided <1 % of target 33.18,20,27
Extensive optimization efforts found that heating to reflux in
EtOH for 3 hours provided a slightly improved conversion to 33
(11 %).
X-Ray Crystallography. Crystals suitable for the crystallo-
graphic analysis of ligand 33 were obtained by slow evapora-
tion of a dichloromethane solution. As can be seen from the
resulting crystal structure (Figure 2), a water molecule is pre-
sent in the cavity of the ligand. This observation is consistent
with the first reported X-ray crystal structure of a CyMe4-
BTPhen ligand system.20 One major difference between the two
structures is that ligand 33 was found to be in the c-c confor-
mation whereas 3 was observed in the t-t conformation (with
regards to the two triazine rings).29 A second major difference
between the structures is that the cavity of ligand 33 also
contains a dichloromethane molecule, which was found to form
non-conventional hydrogen bonds to nitrogen atoms (19) and
(33). Analysis of the structure, also highlighted the non-planar
geometry of ligand 32, with the N(1)-C-C-(N12) and N(10)-C-C-
N(26) torsion angles being measured as -154.7(6)° and -
165.5(6)° respectively, facilitating the hydrogen bonding.
*Compound numbers follow relative trend shown for 33-
3.aReagents and conditions: (a) (i) 109 °C, 15 mins; (ii) o-
phenylenediamine, 109 °C, 2 h; (iii) 25 °C, 16h, 65 %; (b) Ph2O,
260 °C, 30 mins, 95 %; (c) POCl3, 90 °C, 3.5 h, 98 %; (d) Na
metal, MeOH, 65 °C, 72 h, 96 %; (e) K2CO3, Pd(PPh3)4, Ph-
B(OH)2, 66 °C, 24 h, 98 %. Synthesis of 4,7-Cl-CyMe4-BTPhen
(33). Reagents and conditions: (f) (PhCOO)2, NCS, CHCl3, 62 °C,
16 h, 93 %; (g) (i) H2SO4, 95 °C, 1 h; (ii) MeOH, 65 °C, 2 h, 66 %;
(h) NH4OH (29 %), NH4Cl, 25 °C, 72 h, 85 %; (i) (i) (COCl)2, DMF,
0 °C, 6 h; (ii) pyridine, 25 °C, 1 h, 72 %; (j) N2H4.H2O, EtOH, 25 °C,
72 h, 84 %; (k) 32 (2.2 eqvs), EtOH, 78 °C, 3 h, 11 %. Synthesis
of 4,7-MeO-CyMe4-BTPhen 34. Reagents and conditions: (f)
(PhCOO)2, NCS, CHCl3, 62 °C, 48 h, 95 %; (g) (i) H2SO4, 110 °C, 48

Dalton Transactions
Page 4 of 11
View Article Online
DOI: 10.1039/C6DT02474B
Figure 2. X-ray crystallographic structure of 4,7-Cl-CyMe4-BTPhen
(33) (c-c conformer). Thermal ellipsoids are shown at 30 % proba-
bility. Additional solvent molecules are omitted for clarity. Pro-
posed hydrogen bonds are represented by dotted lines and bond
distances are quoted in Å.
Figure 3. Application of equation 1 to the recorded chemical shifts
of the 5,6 and 3,8 protons of 34 vs. pD. r2 = 0.9934.
linear curve fitting. Determining the inflection point of the
resulting sigmoidal curves provided the methanolic pKa values
of the ligands.30
pKa Measurements. The pKa values of ligands 3, 33 and 34
were determined by exploring the acid dependence of the H
1
The second approach involved logarithmic analysis using the
Henderson-Hasselbalch approximation. The relationship be-
NMR spectra in methanol. All three compounds exhibited very
limited solubilities in a range of water containing binary and
ternary solvent mixtures, e.g. D2O: MeOD (v:v) therefore,
definitive pKa values could not be determined. The compounds
were however, sufficiently soluble in methanol to allow “meth-
anolic” pKa values to be determined using a modified proce-
dure to that reported by Harwood et al. for parent 3.20 Unfor-
tunately, the solubility of ligand 35 in methanol was too low to
yield the necessary spectroscopic data. The chemical shift
dependence of the aromatic reporter protons in compounds 3,
33 and 34 (0.2 mM) on pD were recorded following potentiom-
etric titration (full experimental details can be found in the
Supporting Information). The resulting data sets were analysed
using two different approaches, the results of which are sum-
marised in Table 1.
tween the pH and chemical shift (δ) expressed by this approxi-
31–33
mation is shown in equation 1.
The x-axis intercept of the
linear, double logarithmic plots was recorded as the first pKa
values of the ligands. The resulting plot for compound 34 is
shown in Figure 3, with the x-axis intercept providing a pKa
value of ≈ 3.9. The plots for ligands 3 and 33 are provided in the
Supporting Information.34
ꢂδꢃmaxꢄ ꢅ δꢃobsꢄꢆ
pH ꢀ pKa ꢁ log
ꢃꢇꢄ
ꢂδꢃobsꢄ ꢅ δꢃminꢄꢆ
As anticipated during the design of these electronically modi-
fied systems, the 4,7-methoxy-substituents were found to
increase basicity by ≈ 1.4 orders of magnitude versus 3. Con-
versely, the 4,7-chloro-substituents reduced basicity by ≈ 0.8
orders of magnitude, showing that overall basicity within the
series, increases in the order 33<3<<34. This trend is in good
agreement with that reported for the 4-functionalised BTP
series in ethanol by Trumm et al.35
Table 1. pKa values of compounds 3 and 33-34 in d4-methanol
at 298 K.
Ligand
Reporter
Protonsa
Inflection
Point
H-Hb
Mean
Both analytical methodologies provided very similar data sets
with differences of ≤0.1 pKa units observed. Furthermore,
reasonable agreement with the literature value of parent
compound 3 (2.5 vs 3.1) was achieved, despite the literature
methodology not accounting for the likely change in solution
pD upon dissolution of the ligand.20
33
3
5,6 and 3,8
1.7
1.7
2.5
3.9
1.7
2.5
3.9
5,6
2.4
3.9
34
5,6 and 3,8
aNumbering scheme in Figure 1. bDetermined by Henderson-
Hasselbalch analysis
Solvent Extraction Studies. Extraction capabilities were as-
sessed by measuring the extraction kinetics and HNO3 depend-
ence of ligands 33-35 at equilibrium. To allow for direct com-
parison with existing literature, parent ligand 3 was also evalu-
ated. All of the two phase extraction experiments were imple-
mented using solutions of ligands 3 and 33-35 in 1-octanol
(0.0045 - 0.01 M) and tracer quantities of 241Am and 152Eu, and
each 5mg/L of all the lanthanides (La(III)-Lu(III) excluding
The first of these approaches involved plotting the observed
chemical shifts (δ) against the measured pD, yielding a sigmoid-
al plot that was subsequently treated to non-

Page 5 of 11
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02474B
Pm(III)) and Y(III) in HNO3. Following contact on an orbital
Acid Dependence Studies. In contrast to both the BTP (1)
and BTBP ligand series (2), the extraction of Am(III) and Eu(III)
by ligand 3 has been found to exhibit little dependence on
HNO3 concentration.8,20 To determine whether this counter-
intuitive behavior was true for all CyMe4-BTPhen analogues,
several acid dependency experiments were implemented using
ligands 33-35. Once again, ligand 3 was studied as a ‘standard’.
The distribution ratios from the γ-spectroscopy and ICP-MS
measurements are shown in Figures 5a-d.
shaker (2500 min-1, T=20°C) and phase separation, the distribu-
tion ratios (DM(III) = [M(III)]org / [M(III)]aq) and separation factors
(SFAm/Eu = DAm(III)/DEu(III)) were calculated for each instance. Full
experimental details are provided in the Supporting Infor-
mation.
Kinetic Studies. The metal ion extraction kinetics of ligands 3
and 33-35 were investigated through a series of experiments
using 0.5 mL of each phase with contact times of between 0.5 -
16 hours. Previous studies have reported ligand 3 to display
enhanced metal ion extraction kinetics at increased HNO3
concentrations.20 In order to explore the impact of the various
ligand substituents on extraction kinetics, an aqueous HNO3
phase concentration of 0.1 M was selected. The Dc values
calculated from γ-spectroscopy measurements are shown as a
function of time in Figure 4. Each experiment was carried out in
duplicate and the relevant ICP-MS data for all of the Ln(III)
species can be found in the Supporting Information.
The results of the HNO3 dependence study using parent lig-
and 3 (Figure 5a), are in good agreement with previous litera-
ture as no dependence of DAm and DEu on acid concentration
was observed. The measured SFAm/Eu ≈ 120 is also in good
agreement with that of the literature.20 Analysis of the Ln(III)
distribution ratios show that the late-mid to early-heavy
(Tb(III)-Ho(III)) species are preferentially extracted, with lower
distribution ratios observed for the light and late-heavy species.
Very good agreement between the γ-spectroscopy and ICP-MS
distribution ratios for Eu(III) was also achieved. Conversely, for
ligand 34 it can be seen that the presence of the electron
donating methoxy-substituents at the 4,7-positions appears to
induce an acid dependence (Figure 5b). In this instance, almost
an order of magnitude decrease in the distribution ratios was
observed for all metallic species as the HNO3 concentration
increased from 0.1 to 3.0 M. This observation may be attribut-
ed to the increased basicity of 34, resulting in increased ligand
protonation and decreased free ligand concentration. Similarly
to ligand 3, separation factors (SFAm/Eu) of ≈ 120 were recorded.
Comparable trends in the Ln(III) distributions to those of 3 were
also observed.
Interestingly, it can be seen from Figure 5c that ligand 33 also
has a dependence on HNO3 concentration. In this instance, an
order of magnitude increase in the distribution ratios was
witnessed as the aqueous phase HNO3 concentration was
increased to 3 M. It is postulated that this may be a result of
the 4,7-chloro substituents decreasing the basicity of the 1,10-
phenanthroline core N-atoms, thus decreasing the extent of the
competing protonation. Comparison of Figures 5a, 5b and 5c
highlights the substantial impact the 4,7-substituents have on
the magnitude of the distribution ratios, with 33 having signifi-
cantly reduced affinity for all of the trivalent species. Very little
change in the selectivity of 33 versus 3 was recorded with the
Am(III)/Eu(III) separation for 33 ≈ 110. The Ln(III) distribution
ratios shown in Figure 5d vary slightly to those observed for
ligands 3 and 33 as 35 appears to have increased relative affini-
ty for the lighter Sm(III) and Eu(III) species.
Figure 4. Distribution ratios from the extraction of Am(III) and
Eu(III) from 0.1 M HNO3 by ligands 3, 33, 34 (0.01 M) and 35
(0.0045 M) in 1-octanol as a function of contact time. Each ex-
periment was shaken at 2500 min-1 at 20°C.
Analysis of the obtained distribution ratios (Figure 4) showed
that diphenyl system (35) had attained equilibrium within 1
hour, whilst both the parent and 4,7-dimethoxy compounds (3
and 34 respectively) took twice as long, requiring approximate-
ly 2.5 hours of contact time; whereas even after 16 hours, it
was apparent that the 4,7-dichloro compound (33) had still not
attained equilibrium. These results provide the first evidence
that metal ion extraction kinetics of ligand 3 can be modulated
by functionalisation of the 1,10-phenanthroline core, paving
the way for future rationale design. Figure 4 also highlights the
significant impact 4,7-functionalisation has on the equilibrium
distribution ratios, with 3 orders of magnitude difference
observed between 33 (DAm ≈ 1.9) and 34 (DAm ≈ 1800), with 33
showing very poor extraction properties and kinetics.
The solubility of 35 in 1-octanol was found to be limited and
so a 4.5 mM stock solution was prepared (versus 10 mM for 3,
33 and 34). Moreover, the addition of small quantities of kero-
sene (≤20 %) was found to increase the solubility of 35 substan-
tially. The extraction behavior of ligand 35 was largely found to
mirror that of parent compound 3, exhibiting no overall HNO3
dependence (Figure 5d).

Dalton Transactions
Page 6 of 11
View Article Online
DOI: 10.1039/C6DT02474B
a)
b)
c)
d)
Figure 5. Distribution ratios from the extraction of Am(III), Eu(III), Y(III) and all of the lanthanides (La(III)-Lu(III)) excluding Pm(III) by 3 (0.01
M) (a), 34 (0.01 M) (b), 33 (0.01 M) (c) and 35 (0.0045 M) (d) in 1-octanol as a function of HNO3 concentration. Shaken at 2500 min-1 for 20
hours at 20°C and duplicated.
The magnitudes of the distribution ratios were also compa-
rable to 3. Given its CHON compliant composition and faster
extraction kinetics (Figure 4), 35 is a promising new candidate
for an Am(III)/Eu(III) partitioning process.
studied using TRLFS. The fluorescence spectra presented in
Figure 6 are a result of the 6D’7/2 → 8S’7/2 transition over time.
For comparative purposes, all spectra are normalised to the
same peak area. Full experimental TRLFS details can be found in
the Supporting Information. Figure 6a shows the normalised
Cm(III) emission spectra from Cm(ClO4)3 (1.0 x 10-7 M) after the
addition of 3.3 equivalents of 33 as a function of time. Initially
three emission bands are observed.
Time Resolved Laser-Induced Fluorescence Spectroscopy
(TRLFS) Studies. The first X-ray crystal structure of a CyMe4-
BTPhen complex was reported to be of [Eu(3)2NO3]2+ stoichi-
ometry, with the presence of a nitrate ion in the inner coordi-
nation sphere of the metal centre.20 This finding was later
reaffirmed in 2013 by Bremer et al. using Time Resolved Laser
Fluorescence Spectroscopy (TRLFS).36 To probe the complexa-
tion kinetics and solution phase stoichiometry of the di-
substituted CyMe4-BTPhen analogues, a series of mono-phasic
experiments using Cm(III) and ligands 33 and 34 in EtOH was

Page 7 of 11
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02474B
a)
b)
Figure 6: Evolution of the normalised fluorescence spectra of the Cm(III)-33 (a) and Cm(III)-34 (b) complexes over time. Spectra were
obtained in EtOH containing 2 mol% ultrapure H2O (Millipore, Billerica, USA; 18.2 MΩcm), (c(Cm(III)) = 1.0 x 10-7 M) and (c(33) = 3.3 x 10-7
M) or (c(34) = 2.0 x 10-7 M).
The emission band with a maximum at 599.0 nm can be at-
tributed to the solvated Cm(III) species. The red shift of the
emission bands to 609.9 nm and 619.2 nm result from a large
ligand field splitting due to the complexation of Cm(III) with 33
and are characteristic for 1:1 and 1:2 [M:L] complex species of
Cm(III) with BTPhen type ligands, respectively.36–38 Only 1 min
after addition of 33 the emission bands of the solvent species
and the 1:1 [M:L] complex show similar intensities with a less
prominent emission band corresponding to the 1:2 [M:L] com-
plex. For the first 3 hours, the relative intensities of both the
[Cm(solv.)] and [Cm(33)2] emissions decreased whilst the
[Cm(33)] peak increased. After this period, a substantial in-
crease in the intensity of the [Cm(33)2] emission was observed
over a period of 11 days. Comparison of the spectra after 11
and 14 days suggested that the system had finally reached
equilibrium. This 11 day equilibration period is extremely slow,
with 3 reported to have reached equilibrium after approximate-
ly 24 hours.36
validated that complexes of 1:2 [M:L] stoichiometry are formed
preferentially (see Supporting Information). Both of these data
sets are in agreement with the findings of Bremer et al. for
parent compound 3.36
Computational studies of free ligands. The conformational
preferences of parent 3 have been intensively studied at the
DFT level of theory, with the c-c conformer being reported as
the most favorable (by ≈4.1 kcal mol-1).39 This conformational
preference is consistent with that observed in the crystallo-
graphic structure of 33 (Figure 2). In order to explore the theo-
retical effects of 4,7-functionalisation on the electronic struc-
ture and potential for metal binding, the lowest energy, c-c
conformers of ligands 3 and 32-35 were optimised using DFT
using the B3LYP exchange-correlation functional and the 6-
31G(d) basis set. Proton affinities (PA), natural atomic charges
and orbital energies (EHOMO and ELUMO) were computed for these
structures at the higher B3LYP/6-311G(2d,p) level.
Proton Affinities. Proton affinities (EProt) in the gas phase are
commonly used to quantify ligand basicity in preference to pKa
values where solvation effects are influential.40–43 In this in-
stance, the EProt values of 3 and 33-35 were calculated from the
relative differences in energies of the protonated and un-
protonated species in the c-c conformations. In accord with the
obtained experimental data (see pKa measurements), the EProt
of compound 34 was found to be the most favorable of the
series (-276.1 kcal mol-1), followed by 35 and parent 3 (-273.2
and -267.9 kcal mol-1) respectively. The 4,7-chloro compound
33 was found to be the least favorable with EProt = -263.3 kcal
mol-1. These values are also in good agreement with those
reported by Benay et al. for CyMe4-BTPhen (3) and 4-t-Bu-
CyMe4-BTPhen (-271.3 and -275.4 kcal mol-1 respectively).41
Similar observations to those obtained with 34 were noted
for 33, with initially high concentrations of [Cm(34)] (606.8 nm)
decreasing to form [Cm(34)2] (618.2 nm) complex Figure 6b. In
this instance, the system took approximately 3 days to reach
equilibrium; which is significantly faster than analogue 33 but
slower than 3.
The kinetic experiments were performed at very low ligand
concentrations (≈ 10-7 M) in order to observe changes in the
species distribution. Analogous TRLFS experiments using ligand
concentrations likely to be used in extraction processes (≈ 10-2
M) resulted in the 1:2 [M:L] complex being observed exclusive-
ly. Therefore, mono-phasic TRLFS experiments highlight that
both ligands 33 and 34 preferentially form complexes of 1:2
[M:L] stoichiometry with Cm(III). To verify this observed stoi-
chiometric preference, a series of extraction experiments
utilising varying concentrations of 3 and 33-35 was implement-
ed. Determining the slopes of resulting double logarithmic plots

Dalton Transactions
Page 8 of 11
View Article Online
DOI: 10.1039/C6DT02474B
Figure 7: Walsh diagram for ligands 3 and 33-35 in the c-c conformations, showing the frontier orbitals and their relative energies.
It is well established that ionisation equilibrium is largely de-
pendent on the polarity and ion-solvating ability of the sur-
rounding solvent.41,44,45 To assess the influence (if any) of such
effects on the proton affinities of 3 and 33-35, the DFT calcula-
tions were re-executed using the polarisable continuum solva-
tion model (PCM) to include the effects of the aqueous envi-
ronment (EProt-Water). It was rationalized that any effects would
be most significant in highly polar media such as water and less
so for moderately-polar media such as 1-octanol; thus calcula-
tions in water were selected to provide the full extent of such
effects. In the case of ligands 3 and 33-35, only changes in the
absolute EProt-Water values c.f. EProt were observed, with the
relative trend in the EProt-Water values remaining unchanged:
33<3<35<34.
These energy gaps can be compared with the metal ion dis-
tribution ratio (Dc) values at low acid concentration, in order to
correlate the hardness with the metal binding potential of the
ligand. We used the ln(Dc) values, recognizing the exponential
relationship between equilibrium and energy difference, at the
lowest experimental HNO3 concentration of 0.1 mol L-1, thus
minimising the dependence due to the ligand-acid equilibrium
(Figure 8).
Frontier Orbital Analysis. The frontier orbital energies of lig-
ands 3 and 33-35 in the c-c conformation were calculated.
Analysis of the resulting orbital depictions (Figure 7) revealed
that 4,7-functionalisation has only a very subtle effect on the
occupation of the HOMO and LUMO orbitals. In each case, the
HOMO is largely occupied by the lone pair electrons of the
triazine N-atoms, with a small contribution from the lone pair
of the phenanthroline N-atoms. Conversely, the LUMO in each
was found to be evenly contributed to, by both the 1,10-
phenanthroline core and adjacent triazine rings. These observa-
tions are in excellent agreement with those reported by Lewis
et al. and reinforce that the triazine N-atoms play a dominant
role in the donation of electrons to M3+ ions.39
Figure 8: Plot of the averaged Dc values of 241Am, 152Eu and all of
the lanthanides (La(III)-Lu(III) excluding Pm(III)) and Y(III) vs. fron-
tier orbital energy gap for ligands 3 and 33-35. r2 = 0.9713. (dark
green circle = 3, orange triangle = 34, green square = 35 and red
diamond = 33).
Evaluating the HOMO-LUMO energy gaps, ∆E = ELUMO – EHOMO
,
of ligands 3 and 33-35 (Figure 7), directly reveals the impact
4,7-functionalisation has on the chemical hardness of the
ligand.46–49 Despite the calculated changes between parent 3
and 34-35 being relatively modest (≈ 0.9 kcal mol-1), considera-
ble differences were observed between the softest ligand 33
and the hardest ligand 34 (5 kcal mol-1). The magnitude of this
difference is analogous to that reported for the Me2(t-bipy) and
Me2(t-phen) compounds by Mohapatra et al.42
Natural Atomic Charge Analysis. To examine the electronic
impact of 4,7-functionalisation on the BTPhen framework
further, natural bond orbital (NBO) analysis was performed.
The natural atomic charges (Table 2) resulting from this analysis
confirmed that functionalisation of the 1,10-phenanthroline

Page 9 of 11
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02474B
core largely impacts the electron density of the metal-
= 2.5). This behavior is also true for compound 35, whose
predicted pKa is expected to vary only slightly from that of 3.
For ligand 34, the observed decrease in distribution ratios with
increasing HNO3 concentration suggests that the effects of the
increased basicity of the ligand (pKa = 3.9) supersedes the
coordinating phenanthroline N-atoms, with only small changes
in the triazine rings recorded. The results also correlate with
the experimentally measured Dc values, with ligand 34 having
the largest total electron density (-2.664) and Dc values (DAm
≈
-
700) whilst 33 was found to have the lowest theoretical elec-
effects of the increased NO3 concentration, leading to an
tron density (-2.602) and DC values (DAm ≈ 10).
inverse dependence. Conversely, for ligand 33 the observed
increase in distribution ratios with increased HNO3 concentra-
tion is most likely due to the decreased ligand basicity (pKa =
-
1.7), thus decreased protonation combined with increased NO3
Table 2. Natural atomic charges of the N-atoms in ligands 3
concentration. This behavior supports the supposition that it is
the un-protonated species that ultimately undergoes metal
complexation, and is therefore the extractant form that under-
takes the transfer of the metal ions into the organic phase.
and 33-25.
Donor
Atomsa
N 1, N 10
H(3)
Cl(33)
MeO(34)
Ph(35)
CONCLUSIONS
-0.397
-0.488
-0.212
-0.211
-2.616
-0.395
-0.489
-0.207
-0.210
-2.602
-0.418
-0.492
-0.210
-0.212
-2.664
-0.398
-0.487
-0.214
-0.212
-2.622
In summary, we have developed a modified synthetic route
for the large scale preparation of BTPhen based ligand systems
which avoids the use of hazardous SeO2. Significantly, we have
also demonstrated that CyMe4-BTPhen (3) and three novel
analogues, exhibit predictable extraction behavior in equili-
brated systems. This predictability substantially enhances their
suitability for future industrial scale partitioning processes such
as the SANEX. Order of magnitude differences in the distribu-
tion ratios of the four ligands were observed, yet all were found
N 12, N 26
N 20, N 34
N 19, N 33
∑
aNumbering scheme in Figure 2.
The NBO data also highlights the fundamental role the 1,2,4-
triazine moieties play in creating the soft and thus highly selec-
tive molecular cavity of the ligands.19 The presence of neighbor-
ing (α) triazine N-atoms can be seen to reduce the point charge
of both binding N-atoms significantly (-0.212 vs -0.488 for 3).
This observation is reaffirmed by Rao et al. who reported the
1,3-diazine containing analogue of BTPhen (BQPhen) to have
significantly diminished selectivity for the trivalent actinides
(SFAm/Eu = 10 vs. 110 for 3).43
to have comparable selectivity for Am(III) over Eu(III) (SFAm/Eu
110) at equilibrium. This behavior is in contrast to that of a
series of substituted BTP’s (1) for which both distribution
ratios and selectivity for Am(III) over Eu(III) increased as
≈
a
function of ligand basicity.34 Furthermore, a less pro-
nounced influence of substitution on extraction properties was
found for BTPhens substituted at the 5- or 5,6-positions.20 For
the first time, we have shown that complexation kinetics are
substantially affected by backbone functionalisation. This paves
the way for the design of future ligand systems that provide
enhanced extraction kinetics. Evaluating the basicity, preferred
stoichiometry and HNO3 dependence behavior of the ligands
has provided a preliminary insight into the potential extraction
mechanisms taking place, allowing the previously unexplained
HNO3 independence of parent ligand 3 to be qualitatively
rationalised. 4,7-phenyl-CyMe4-BTPhen (35) is a CHON compli-
ant ligand which demonstrates extremely promising properties,
including enhanced metal ion extraction kinetics, and provides
a promising alternative to lead compound 3.
DISCUSSION
It is now well established that CyMe4-BTPhen (3) and its ana-
logues, extract trivalent metal ions from nitrate solutions
according to equation 2:
-
M3+(aq) + 3NO3 (aq) + 2L(org) ⇋ [ML2(NO3)](NO3)2(org)
(2)
However, if this were the only equilibrium process taking
place, all systems including ligand 3 would exhibit distribution
-
ratios strongly increasing with the NO3 and thus HNO3 concen-
tration. Clearly this is not the case as analysis of Figures 5a-5c
implies that a second, competing process must also be taking
place. It was also clear that this second equilibrium is substan-
tially affected by variation of the 4,7-substituents, suggesting
that the competing process is most likely a result of the varying
pKa values of 3 and 33-35 and thus their relative protonation
states (equation 3).35
ASSOCIATED CONTENT
Supporting Information available: Full experimental details, key
NMR spectra, solvent extraction procedures, TRLFS setup and
procedures and X-ray crystallographic data are provided. This
material is available free of charge via the Internet at
http://pubs.acs.org.”
L + H+ ⇋ LH+
(3)
Analysis of the extraction data presented in Figures 5a-5c,
along with both the pKa measurements and stoichiometric
preferences of ligands 3 and 33-35 (equations 2 and 3) allows
the extraction behavior to be qualitatively rationalised. During
AUTHOR INFORMATION
Corresponding Author
*Email: l.m.harwood@reading.ac.uk
*Email: roger.whitehead@manchester.ac.uk
-
extractions utilising parent 3, the effects of increased NO3
concentration are predominantly cancelled out by the decrease
in free ligand concentration due to increased protonation (pKa
Notes

Dalton Transactions
Page 10 of 11
View Article Online
DOI: 10.1039/C6DT02474B
J. Bezençon, M. B. Wittwer, B. Cutting, M. Smieško, B. Wagner,
‡ Pre-equilibrium values.
30
M. Kansy and B. Ernst, J. Pharm. Biomed. Anal., 2014, 93, 147–
155.
*The authors declare no competing financial interest.
31
E. Breitmaier and K. H. Spohn, Tetrahedron, 1973, 29, 1145–
1152.
ACKNOWLEDGMENT
We thank the EPSRC for funding a Nuclear Fission Research, Sci-
ence and Technology DTC (Nuclear FiRST) studentship
EP/G037140/1 (A.C.E) and core capability grant EP/K039547/1. This
work was also supported by the SACESS project funded by the
European Atomic Energy Community Seventh Framework under
grant agreement No. 323282.
32
33
34
J. Henderson, Am. J. Physiol., 1907, 21, 250–255.
K. A. Hasselbalch, Biochem. Z., 1917, 78, 112.
L. Grycová, R. Dommisse, L. Pieters and R. Marek, Magn. Reson.
Chem., 2009, 47, 977–981.
S. Trumm, G. Wipff, A. Geist, P. J. Panak and T. Fanghänel,
Radiochim. Acta, 2011, 99, 13–16.
A. Bremer, D. M. Whittaker, C. Sharrad, A. Geist and P. J. Panak,
Dalton Trans., 2013, 43, 2684–2694.
S. Trumm, G. Lieser, M. R. Foreman, P. J. Panak, A. Geist and T.
Fanghänel, Dalton Trans., 2010, 39, 923–929.
C. Wagner, U. Müllich, A. Geist and P. J. Panak, Solvent Extr. Ion
Exch., 2015, 6299, 1–11.
F. W. Lewis, L. M. Harwood, M. J. Hudson, M. G. B. Drew, V.
Hubscher-Bruder, V. Videva, F. Arnaud-Neu, K. Stamberg and S.
Vyas, Inorg. Chem., 2013, 52, 4993–5005.
T. Wróblewski, L. Ziemczonek, A. M. Alhasan and G. P. Karwasz,
Eur. Phys. J. Spec. Top., 2007, 144, 191–195.
G. Benay, R. Schurhammer and G. Wipff, Phys. Chem. Chem.
Phys., 2011, 13, 2922–34.
A. Bhattacharyya, T. Gadly, P. K. Mohapatra, S. K. Ghosh, D.
Manna, T. K. Ghanty and V. K. Manchanda, RSC Adv., 2012, 2,
7066.
35
36
37
38
39
REFERENCES
1
2
3
4
K. L. Nash and M. Nilsson, Reprocessing and Recycling of Spent
Nuclear Fuel, Elsevier, 2015.
C. Poinssot, B. Boullis and S. Bourg, Reprocessing and Recycling
of Spent Nuclear Fuel, Elsevier, 2015.
The Royal Society Science Policy Centre Report, ISBN
9780854038916, 2011.
S. J. Zinkle, K. A. Terrani, J. C. Gehin, L. J. Ott and L. L. Snead, J.
Nucl. Mater., 2014, 448, 374–379.
40
41
42
5
6
S. Monti, in International Atomic Energy Agency, 2015.
K. L. Nash and G. R. Choppin, Sep. Sci. Technol., 1997, 32, 255–
274.
7
D. D. Sood and S. K. Patil, J. Radioanal. Nucl. Chem. Artic., 1996,
203, 547–573.
43
44
Y. Yang, J. Liu, L. Yang, K. Li, H. Zhang, S. Luo and L. Rao, Dalt.
Trans., 2015, 8959–8970.
8
9
P. J. Panak and A. Geist, Chem. Rev., 2013, 113, 1199–236.
C. A. Sharrad and D. M. Whittaker, Reprocessing and Recycling
of Spent Nuclear Fuel, Elsevier, 2015.
G. Modolo, A. Geist and M. Miguirditchian, in Reprocessing and
Recycling of Spent Nuclear Fuel, Elsevier, 2015, pp. 245–287.
B. A. Moyer, G. J. Lumetta and B. J. Mincher, Reprocessing and
Recycling of Spent Nuclear Fuel, Elsevier, 2015.
N. Kaltsoyannis, P. J. Hay, J. B. Li and B. E. J P Bursten, The
Chemistry of the Actinide and Transactinide Elements, Springer
Netherlands, Dordrecht, 2006.
William M. Nelson, Green Solvents for Chemistry: Perspectives
and Practice, Oxford University Press, USA, 2003.
G. Benay and G. Wipff, J. Phys. Chem. B, 2013, 117, 1110–22.
R. G. Pearson, Proc. Natl. Acad. Sci. U. S. A., 1986, 83, 8440–
8441.
W. Yang and R. G. Parr, Proc. Natl. Acad. Sci. U. S. A., 1985, 82,
6723–6726.
P. W. Ayers and R. G. Parr, J. Am. Chem. Soc., 2000, 122, 2010–
2018.
10
11
12
45
46
47
48
49
13
14
G. R. Choppin, J. Alloys Compd., 1995, 223, 174–179.
C. Adam, P. Kaden, B. B. Beele, U. Müllich, S. Trumm, A. Geist,
P. J. Panak and M. A. Denecke, Dalton Trans., 2013, 42, 14068–
74.
Y. Li and J. N. S. Evans, J. Am. Chem. Soc., 1995, 117, 7756–
7759.
15
16
N. Kaltsoyannis, Inorg.Chem., 2012, 52, 3407.
Z. Kolarik, U. Müllich and F. Gassner, Solvent Extr. Ion Exch.,
1999, 17, 23–32.
17
18
Z. Kolarik, U. Müllich and F. Gassner, Solvent Extr. Ion Exch.,
1999, 17, 1155–1170.
M. J. Hudson, C. E. Boucher, D. Braekers, J. F. Desreux, M. G. B.
Drew, M. R. Foreman, L. M. Harwood, C. Hill, C. Madic, F.
Marken and T. G. A. Youngs, New J. Chem., 2006, 30, 1171.
M. J. Hudson, L. M. Harwood, D. M. Laventine and F. W. Lewis,
Inorg. Chem., 2013, 52, 3414–28.
F. W. Lewis, L. M. Harwood, M. J. Hudson, M. G. B. Drew, J. F.
Desreux, N. Bouslimani, G. Modolo, A. Wilden, M. Sypula, T. Vu
and J. Simonin, J. Am. Chem. Soc., 2011, 133, 13093–13102.
A. Afsar, D. M. Laventine, L. M. Harwood, M. J. Hudson and A.
Geist, Chem. Commun., 2013, 49, 8534–6.
19
20
21
22
23
A. F. Larsen and T. Ulven, Org. Lett., 2011, 13, 21–23.
L. M. Harwood, F. W. Lewis, M. J. Hudson, J. John and P. Distler,
Solvent Extr. Ion Exch., 2011, 29, 551–576.
24
25
A. Afsar, L. M. Harwood, M. J. Hudson, J. Westwood and A.
Geist, Chem. Commun., 2015, 51, 5860–5863.
N. T. Coogan, M. A. Chimes, J. Raftery, P. Mocilac and M. A.
Denecke, J. Org. Chem., 2015, 80, 8684–8693.
F. H. Case, J. Org. Chem., 1965, 30, 931–3.
A. Geist, C. Hill, G. Modolo, M. R. Foreman, M. Weigl, K.
Gompper and M. J. Hudson, Solvent Extr. Ion Exch., 2006, 24,
463–483.
26
27
28
29
http://www.technocomm.co.uk.
D. M. Whittaker, T. L. Griffiths, M. Helliwell, A. N. Swinburne, L.
S. Natrajan, F. W. Lewis, L. M. Harwood, S. A. Parry and C. A.
Sharrad, Inorg. Chem., 2013, 52, 3429–44.

Page 11 of 11
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02474B
11