Gamma radiolysis of hydrophilic diglycolamide ligands in concentrated aqueous nitrate solution

Article abstract of DOI:10.1039/c9dt03918j

The radiation chemistry of a series of hydrophilic diglycolamides (DGAs: TEDGA, Me-TEDGA, Me₂-TEDGA, and TPDGA) has been investigated under neutral pH, concentrated aqueous nitrate solution conditions. A combination of steady-state gamma and time-resolved pulsed electron irradiation experiments, supported by advanced analytical techniques and multi-scale modeling calculations, have demonstrated that: (i) the investigated hydrophilic DGAs undergo first-order decay with an average dose constant of (-3.18 ± 0.23) × 10^-6 Gy^-1; (ii) their degradation product distributions are similar to those under pure water conditions, except for the appearance of NO_x adducts; and (iii) radiolysis is driven by hydroxyl and nitrate radical oxidation chemistry moderated by secondary degradation product scavenging reactions. Overall, the radiolysis of hydrophilic DGAs in concentrated, aqueous nitrate solutions is significantly slower and less structurally sensitive than under pure water conditions, similar to their lipophilic analogs. Acid hydrolysis, not radiolysis, is expected to limit their useful lifetime. These findings are promising for the deployment of hydrophilic DGAs as actinide aqueous phase stripping and hold-back agents, due to the presence of high concentrations of nitrate in envisioned large-scale process conditions.

Full text of DOI:10.1039/c9dt03918j

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Gamma radiolysis of hydrophilic diglycolamide
ligands in concentrated aqueous nitrate solution†
Cite this: Dalton Trans., 2019, 48,
17005
c
Gregory P. Horne, *a Andreas Wilden, *b Stephen P. Mezyk,
Liam Twight, c Michelle Hupert, d Andrea Stärk, d Willem Verboom,
e
Bruce J. Mincher a and Giuseppe Modolo
b
The radiation chemistry of
a series of hydrophilic diglycolamides (DGAs: TEDGA, Me-TEDGA,
Me2-TEDGA, and TPDGA) has been investigated under neutral pH, concentrated aqueous nitrate solution
conditions. A combination of steady-state gamma and time-resolved pulsed electron irradiation experi-
ments, supported by advanced analytical techniques and multi-scale modeling calculations, have demon-
strated that: (i) the investigated hydrophilic DGAs undergo first-order decay with an average dose constant
of (−3.18 0.23) × 10−6 Gy−1; (ii) their degradation product distributions are similar to those under pure
water conditions, except for the appearance of NOx adducts; and (iii) radiolysis is driven by hydroxyl and
nitrate radical oxidation chemistry moderated by secondary degradation product scavenging reactions.
Overall, the radiolysis of hydrophilic DGAs in concentrated, aqueous nitrate solutions is significantly
slower and less structurally sensitive than under pure water conditions, similar to their lipophilic analogs.
Acid hydrolysis, not radiolysis, is expected to limit their useful lifetime. These findings are promising for
the deployment of hydrophilic DGAs as actinide aqueous phase stripping and hold-back agents, due to
the presence of high concentrations of nitrate in envisioned large-scale process conditions.
Received 3rd October 2019,
Accepted 28th October 2019
DOI: 10.1039/c9dt03918j
rsc.li/dalton
as N,N,N′,N′-tetramethyl diglycolamide (TMDGA) and N,N,N′,
N′-tetraethyl diglycolamide (TEDGA) shown in Fig. 1.
Introduction
Lipophilic diglycolamides (DGAs) have been extensively
Many of these derivatives have been shown to be successful
studied as trivalent actinide (An(III)) and lanthanide (Ln(III)) aqueous phase stripping5–7 and hold-back agents.8–11
extractants for application in spent nuclear fuel reprocessing However, their potential for application as aqueous phase acti-
and recycling scenarios.1–3 These molecules are structurally nide complexants is hindered by a limited understanding of
flexible, exhibit relatively high affinities for An(III), and fulfil their radiolytic behaviour and the impact of their degradation
the CHON principle – being composed of only carbon, hydro- products on separation process performance. Actinide com-
gen, oxygen, and nitrogen elements – which is beneficial as it
enables incineration of waste without the generation of sec-
ondary waste streams.4 Their structural versatility has led to
the synthesis of a range of hydrophilic DGA derivatives, such
aIdaho National Laboratory, Center for Radiation Chemistry Research, Idaho Falls,
ID, P.O. Box 1625, 83415, USA. E-mail: gregory.horne@inl.gov
bForschungszentrum Jülich GmbH, Institut für Energie- und Klimaforschung –
Nukleare Entsorgung und Reaktorsicherheit (IEK-6), 52428 Jülich, Germany.
E-mail: a.wilden@fz-juelich.de
cCalifornia State University Long Beach, Department of Chemistry and Biochemistry,
Long Beach, CA 90804, USA
dForschungszentrum Jülich GmbH, Zentralinstitut für Engineering, Elektronik und
Analytik (ZEA-3), 52428 Jülich, Germany
eUniversity of Twente, Mesa+ Institute for Nanotechnology, Laboratory of Molecular
Nanofabrication, P.O. Box 217, 7500 AE Enschede, The Netherlands
†Electronic supplementary information (ESI) available: HPLC-ESI-MS/MS
solvent gradients, reaction kinetics figures, and degradation product tables. See
DOI: 10.1039/c9dt03918j
Fig. 1 Chemical structures of the hydrophilic DGAs: N,N,N’,N’-tetra-
methyl diglycolamide (TMDGA), N,N,N’,N’-tetrapropyl diglycolamide
(TPDGA), N,N,N’,N’-tetraethyl diglycolamide (TEDGA), 2-(2-(diethyl-
amino)-2-oxoethoxy)-N,N-diethyl propanamide (Me-TEDGA), and 2,2’-
oxybis(N,N-diethyl propanamide) (Me2-TEDGA).
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17005–17013 | 17005

View Article Online
Paper
Dalton Transactions
plexants are exposed to a multi-component radiation field that which is more than an order of magnitude lower than that
induces their radiolysis and degradation product formation. with the •OH.16,19,23,24
While the radiation chemistry of the lipophilic DGAs has been
ꢁ
Haqþ þ eaq ! H• k ¼ 2:3 ꢂ 1010 Mꢁ1 sꢁ1
ð2Þ
ð3Þ
ð4Þ
ð5Þ
well characterized,12–15 to date only a single experimental
investigation has explored the radiation chemistry of the
hydrophilic DGAs.16 Koubský and Lustinec studied hydrophilic
DGAs using computational methods and confirmed the experi-
mentally observed trend in radiolytic stability with increasing
molecular weight.17 Wilden et al. reported on the radiolytic
and hydrolytic degradation of a series of hydrophilic DGAs
under pure water conditions.16 Despite the formation of similar
degradation product distributions, the hydrophilic DGAs were
much more susceptible to radiolysis compared to their lipo-
philic analogues. For example, the dose constant18 (d) for
NO3ꢁ þ eaq ! NO3•2ꢁk ¼ 9:7 ꢂ 109 Mꢁ1 sꢁ1
ꢁ
NO3ꢁ þ H• ! HNO3•ꢁk ¼ 5:6 ꢂ 106 Mꢁ1 sꢁ1
HNO3 þ OH ! NO3 þ H2O k ¼ 1:9 ꢂ 107 Mꢁ1 sꢁ1
•
•
:
Further, ionizing radiation energy is partitioned amongst
solution components (solutes and solvent molecules) pro-
portional to their electron fraction of the total medium’s elec-
tron density.25 In pure water and dilute aqueous solutions, the
majority of radiation energy is deposited into the electrons of
the water molecules. However, for concentrated HNO3 solu-
tions (≥1.0 M), an appreciable fraction of radiation energy is
partitioned into the electrons of the NO3−/HNO3 molecules,
promoting direct generation of additional radiolysis
products:26
TMDGA (d = (−14.9 1.2) × 10−6 Gy−1 16
)
is considerably larger
0.3) × 10−6
than the equivalent for TODGA (d = (−4.1
Gy−1).12,14 These observations were rationalized by reaction
kinetics measurements highlighting the affinity of the hydro-
philic DGAs for the highly oxidizing hydroxyl radical (•OH, E° =
2.7 V) from water radiolysis:19
NO3ꢁ=HNO3 ⇝ eꢁ; •NO3; NO2ꢁ=HNO2; O; Haq
:
ð6Þ
þ
H2O ⇝ ½0:28ꢀ OH þ ½0:06ꢀH• þ ½0:27ꢀeaqꢁ þ ½0:05ꢀH2
•
These changes in aqueous phase radiation chemistry may
reduce the extent of hydrophilic DGA degradation, leaving acid
hydrolysis as their useful lifetime limiting factor.
Consequently, the nitrate radiation chemistry of the hydro-
philic DGAs needs to be quantitatively established to deter-
mine their radiolytic lifetime and viability in the absence of
acid hydrolysis.
To this end, we present a systematic investigation into the
fundamental radiation chemistry of a series of hydrophilic
DGAs in neutral pH, concentrated aqueous nitrate solutions.
An extensive suite of steady-state gamma and time-resolved
pulsed electron irradiations, complemented by advanced
analytical techniques and predictive multi-scale modelling cal-
culations to provide mechanistic discrimination between
radiolytic and hydrolytic effects under envisioned process con-
ditions, is presented.
þ
þ ½0:07ꢀH2O2 þ ½0:27ꢀHaq
;
ð1Þ
where the numbers in brackets are the radiolytic yields of de-
posited energy (G-values in µmol J−1) for species production.
For example, the rate coefficient (k) for the reaction of TEDGA
•
•
with the OH is very fast, k(TEDGA + OH) = (2.91 0.10) × 109
M−1 s−1 16
The lipophilic DGAs have been shown to react with
.
•
the organic phase equivalent of the OH, the organic radical
cation (R•+), at similar diffusion limited rates, k(TODGA + R•+)
= (9.72 0.10) × 109 M−1 s−1 12
However, lifetimes of these
.
organic transients are typically far shorter than that of the
•OH, making them less available to induce DGA radiolysis. For
example, the n-dodecane radical cation undergoes rapid ion-
•
recombination within ∼2 ns,20 while the OH is typically still
available for reaction at microsecond timescales in pure water.
Envisioned large-scale process conditions employ nitric
acid (HNO3) solutions as the aqueous phase, the chemistry of
which is drastically different from that of pure water.
Previously, TEDGA was found to be susceptible to acid hydro-
lysis, and other short side-chain DGAs are believed to behave
Experimental
Materials
similarly, further limiting their useful lifetime.16 However, the N,N,N′,N′-tetraethyl diglycolamide (TEDGA, >99%), 2-(2-(di-
presence of high solute concentrations of protons (Haq+) and ethylamino)-2-oxoethoxy)-N,N-diethyl
propanamide (Me-
nitrate anions (NO3−) alters the suite of water radiolysis pro- TEDGA, >99%), and 2,2′-oxybis(N,N-diethyl propanamide (Me2-
ducts available for reaction with the hydrophilic DGAs. Both TEDGA, >99%) were supplied by Technocomm Ltd (Wellbrae,
Haq+ and NO3− are efficient scavengers of the hydrated electron Falkland, Scotland, UK). N,N,N′,N′-tetrapropyl diglycolamide
(eaq−, reactions (2) and (3), respectively), and thus capable of (TPDGA, >99%) was prepared by Twente University (Enschede,
completely inhibiting eaq reactions with the DGAs.19 Further, The Netherlands), the synthetic method for which has been
−
at high acidities there is an appreciable concentration of previously described.27 Sodium nitrate (NaNO3, ≥99.995%
−
undissociated HNO3 (HNO3 ⇌ NO3 + Haq+, pKa ≈ 1.37),21 trace metals basis), potassium thiocyanate (KSCN, ≥99.0% ACS
•
which can scavenge OH, converting it into the relatively less Reagent Grade), para-chlorobenzoic acid (pCBA, 99%), per-
reactive nitrate radical (•NO3, E° = 2.3–2.6 V),22 reaction (5). chloric acid (HClO4, ≥99.999% trace metals basis) potassium
This is consistent with the reported rate of reaction for TEDGA phosphate monobasic (99.99% trace metals basis), tertiary-
•
•
with NO3, k(TEDGA + NO3) = (6.66 0.63) × 107 M−1 s−1
,
butanol (tBuOH, ≥99.5%), n-dodecane (99%), and dichloro-
17006 | Dalton Trans., 2019, 48, 17005–17013
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
methane (DCM, ≥99.8%) were sourced from MilliporeSigma. TEDGA, and TPDGA irradiations were performed in static
All chemicals were used without further purification. Ultra- cuvettes at ambient room temperature. Kinetic traces were gen-
pure water (18.2 MΩ cm) was used to prepare all aqueous solu- erated through averaging 6–20 individual measurements.
tions. Unless otherwise specified, all solvents for analyses were Quoted errors for the reaction rate coefficients are a combi-
Merck (Darmstadt, Germany) LC/MS grade or VWR nation of measurement precision and sample concentration
(Darmstadt, Germany) LC/MS grade.
errors.
Irradiations
Quantification
Steady-state. A Shepherd 109-68 cobalt-60 irradiator unit at Irradiated samples were analyzed at Forschungszentrum Jülich
the Notre Dame Radiation Laboratory (NDRL) was used to (FZJ) by high performance liquid chromatography – electro-
perform all steady-state gamma irradiations. Samples con- spray ionization mass spectrometry/mass spectrometry
sisted of the aforementioned hydrophilic DGAs dissolved in (HPLC-ESI-MS/MS). The chromatographic separations varied
aqueous 5.0 M NaNO3 solutions, irradiated in triplicate as slightly with the DGA analyte. All samples were diluted
single phases in sealed screw-cap 0.5 dram glass vials to doses 1 : 500 000 and measured in triplicate.
up to 200 kGy. TPDGA was not fully soluble in 5.0 M NaNO3 to
For TEDGA, TPDGA and Me-TEDGA, a Thermo Phenyl-X
the desired concentration. Therefore, the stock solution was (100 × 4.6 mm; 2.6 µm particle size) column was used with a
diluted with water until a clear solution was achieved, with gradient of acetonitrile + 0.1% formic acid (A) in 0.1% formic
final concentrations of 0.034 M TPDGA and 3.39 M NaNO3. A acid in H2O (B) at 35 °C and a flow-rate of 800 µL min−1. The
single set of pure water and n-dodecane samples of TPDGA gradient is described in ESI Tables S1 and S2.† Calibration was
were irradiated under the same conditions for comparison. performed by dilution of unirradiated TEDGA, TPDGA, and
Dosimetry was determined using Fricke solution,28,29 sub- Me-TEDGA samples. The linearity was found to be good in the
sequently corrected for 60Co decay (τ1/2 = 5.27 years) and solu- region of 10 to 500 nM with R2 = 0.9955, 0.9989, and 0.9991,
tion density, affording an average dose rate of 7.13 kGy h−1
.
respectively. The variation coefficients of a 100 nM standard
Time-resolved. Hydrophilic DGA reaction kinetics were were 7.4, 6.0, and 6.2%, respectively. It was found that TPDGA
determined using the NDRL nanosecond pulsed electron sample dilutions in acetonitrile were unstable, but stable in
linear accelerator facility and the Brookhaven National methanol.
Laboratory (BNL) Laser Electron Accelerator Facility (LEAF). A
For Me2-TEDGA, a Phenomenex Luna Omega C18 Polar
detailed description of these systems has been given (100 × 4.6 mm; 2.6 µm particle size) column was used with a
previously.30–32 Dosimetry was determined using N2O satu- gradient of acetonitrile + 0.1% formic acid (A) in 0.1% formic
rated solutions of 10 mM KSCN at λmax = 470–475 nm (Gε = 5.2 acid in H2O (B) at 35 °C and a flow-rate of 800 µL min−1. The
× 10−4 m2 J−1).33 Isolation and reaction kinetics of specific rad- gradient is described in ESI Table S3.† Calibration was per-
icals was achieved using the following conditions:
formed by dilution of unirradiated Me2-TEDGA samples. The
linearity was found to be good in the region of 1 to 500 nM
−
• Hydrated electron (eaq−). Direct decay kinetics of the eaq
were observed at 720 nm, using N2-saturated aqueous solu- with R2 = 0.9997. The variation coefficient of a 100 nM stan-
tions of 0.5 M tBuOH and 10 mM phosphate buffer.
dard of Me2-TEDGA was 8.2%.
• Hydrogen atom (H•). pCBA competition kinetics using the
HPLC-ESI-MS/MS for quantification was performed with a
direct growth kinetics of the H-pCBA adduct transient absor- Qtrap6500 instrument (ABSciex, Darmstadt) coupled with an
bance at 370 nm, using N2-saturated solutions containing Agilent 1260 HPLC system consisted of a binary pump system,
100 μM pCBA, 0.01 M HClO4, and 20 mM tBuOH.
an autosampler and a thermostated column compartment.
• Hydroxyl radical (•OH). KSCN competition kinetics using The MS parameters used for all methods were optimized by
the direct growth kinetics of the (SCN)2•− transient observed at performing a Flow Injection Analysis (FIA) with standards and
470–475 nm, using N2O saturated solutions containing 100 μM led to the following settings for all analysis: curtain gas (N2) 40
KSCN and 10 mM phosphate buffer.
arbitrary units (a.u.), temperature of the source 350 °C, nebuli-
•
• Nitrate radical (•NO3). Direct decay kinetics of the NO3 zer gas (N2) 40 a.u. and heater gas (N2) 80 a.u. Quantification
observed at 630–640 nm using aerated or N2O-saturated after HPLC was performed using ESI-MS/MS detection in the
aqueous solutions of 5.0 M NaNO3.
multiple reaction-monitoring (MRM) mode in positive ioniza-
• Dodecane radical cation (R•+). Direct decay kinetics of R•+ tion mode. MRM transitions involving precursor ions (M + H)+
observed at 800 nm, using aerated solutions of 0.5 M DCM in and the two most abundant product ions were used for quanti-
n-dodecane.
fication of all analytes as shown in ESI Table S4.† All LC-MS/
All transient absorption measurements were made using MS data acquisition and processing was carried out using the
1.0 cm optical pathlength quartz cells. TEDGA stock solutions Software Analyst 1.6.1 (AB Sciex, Darmstadt). Quantification
used a flow cell, with flow rate regulated to ensure that each was performed with the Software Multiquant (AB Sciex,
electron pulse irradiated a fresh sample. Solution temperatures Darmstadt).
were directly measured by an in-flow thermocouple placed
Product analysis was performed at FZJ using a hybrid linear
immediately above the irradiation cell. The temperature stabi- ion trap Fourier-Transform Ion Cyclotron Resonance (FTICR)
lity of the system was better than 0.3 °C. Me-TEDGA, Me2- mass spectrometer Linear Tandem Quadrupole Fourier
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17005–17013 | 17007

View Article Online
Paper
Dalton Transactions
Transform (LTQFT) UltraTM (Thermo Fisher Scientific,
Bremen) coupled with an Agilent 1200 HPLC system consisted
of a binary pump system, an autosampler, a thermostated
column compartment, and a Diode-Array detector (Agilent,
Waldbronn). The mass spectrometer was first tuned and cali-
brated in the positive mode following the standard optimiz-
ation procedure for all voltages and settings: Source Type: ESI,
Ion Spray Voltage: 3.8 kV, Capillary Voltage: 37.00 V, Tube
Lens; 130.00 V, Capillary Temp: 275.00 °C, Sheath Gas Flow:
60.00. Mass spectra were recorded in full scan from 100 to
1000 Da with a resolution of 100 000 at m/z 400. All data were
processed using the Xcalibur software version 2.0.
Multi-scale modelling
Kinetic calculations were performed using a multi-scale mod-
elling methodology.34 A combination of stochastic35–37 models
were used to calculate representative gamma radiation track
yields for aerated ([O2] = 2.5 × 10−4 M) aqueous 3.39 and 5.0 M
NaNO3 solutions. Radiation track yields from fits to stochastic
data (1.0, 4.0, and 6.0 M NaNO3(aq)), in conjunction with
nitrate/nitric acid direct effect yields,38 were used as initial
parameters for a deterministic model and solved using the
FACSIMILE numerical algorithm (MCPA software). For a com-
plete description of this modelling methodology see ref. 34.
Employed kinetic reaction sets are from water radiolysis com-
pilations by Buxton et al.19 and Elliot and Bartels,39 a similar
nitrate radiolysis reaction scheme as presented in ref. 34, and
experimental reaction kinetics determined here.
Fig. 2 Determination of the •OH rate coefficient with Me-TEDGA by
KSCN competition kinetics. (A) Transient growth kinetics are for 98.5 μM
KSCN dissolved in N2O-saturated 10 mM phosphate buffered solution at
pH 6.98 and 24.1 °C in the presence of zero (black), 63.4 (red), 128
(blue), 196 (green) and 322 (magenta) μM added Me-TEDGA. (B)
Transformed kinetic plot using the peak absorbances of transient kine-
tics obtained for different Me-TEDGA concentrations. Solid line is a
weighted linear fit to transformed data, with a slope of 0.141 0.002,
Results and discussion
Reaction kinetics
The kinetic data measured in this study are summarized in
Table 1, and corresponding kinetic plots given in ESI.† For
comparison purposes, equivalent values for TMDGA are also
intercept of 0.997
corresponds to the second-order reaction rate coefficient for •OH and
Me-TEDGA of (1.71 0.02) × 109 M−1 s−1
0.005, and R2 = 0.997, where the slope value
•
given.16 Typical kinetic data is shown in Fig. 2 for the OH
.
reaction with Me-TEDGA using KSCN competition kinetics.19
The highest reactivity for these hydrophilic DGAs is seen to
be with the •OH, of the order of 109 M−1 s−1. Consistency
between the •OH rate coefficients for TEDGA and TMDGA,
implies that little •OH reactivity occurs at the C–N alkyl chains;
rather, this radical is predominantly reacting with the bridging
methylene (–CH2–) groups. Based upon analogous reactions in
the literature,19 we presume that this reaction occurs by hydro-
gen atom abstraction from one of these methylene moieties:
•
•OH þ TEDGA ! H2O þ ½TEDGAꢀ :
ð7Þ
Preference for reaction at these methylene centers can be
explained in terms of the carbon centered radical product
([TEDGA]•), which is stabilized, relative to radical formation on
the C–N alkyl chains, by a combination of electron withdrawal
by the ether linkage and delocalization into the carbonyl
group. Substitution of these methylene group hydrogen atoms
with one (Me-TEDGA) or two (Me2-TEDGA) methyl groups
Table 1 Summary of measured rate coefficients for reaction of the
hydrophilic DGAs (TMDGA, TEDGA, Me-TEDGA, Me2-TEDGA, and
−
TPDGA) with •OH, •NO3, H•, and eaq
Rate coefficient (k, M−1 s−1
)
•
decreases the OH rate coefficient. This decrease is attributed
Ligand
•OH (109)
•NO3 (108)
H• (108)
eaq− (108)
to greater steric hindrance imposed by the methyl substituents
on this bridging group, impeding the abstraction reaction.
Similar behaviour is observed for H• reactivity with these
hydrophilic DGA’s, albeit that these reactions occur about an
order of magnitude slower than for the •OH reactions. The
TMDGA16
TEDGA
Me-TEDGA
Me2-TEDGA
TPDGA
3.06 0.09
2.91 0.10
1.71 0.02
1.37 0.02
3.81 0.01
3.05 0.12
11.19 0.46
7.79 0.38
3.29 0.20
4.83 0.36
1.22 0.03
1.60 0.09
0.46 0.06
0.36 0.07
0.41 0.02
n.d.
2.24 0.10
2.71 0.15
0.99 0.08
0.80 0.01
17008 | Dalton Trans., 2019, 48, 17005–17013
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
four extra terminal methylene groups in TEDGA, as compared dation product was monitored. Although this method is not pre-
to TMDGA, facilitate faster hydrogen atom abstraction at these cisely quantitative (semi-quantitative), as ionization potentials can
sites:
be different for different fragments and depend on the total ion
count, an estimate of the change in abundance of a product in
the samples can be given. The m/z ratios of the protonated degra-
dation product ions were used, although usually sodiated product
ions, adducts with NH4 ions as well as higher complexes (e.g.,
1 : 2 Na : ligand complexes) were also observed.
•
H• þ TEDGA ! H2 þ ½TEDGAꢀ :
ð8Þ
However, the substantial decrease in H• rate coefficient for
Me-TEDGA and Me2-TEDGA again demonstrates that most of
the reactivity is at the bridging methylene groups, which have
greater steric hindrance and less reactive sites upon methyl-
group substitution. Lastly, it would be expected that the
TPDGA reaction with this radical would be at least as fast as
TEDGA. Unfortunately, this latter measured value is anoma-
lously slower, which we cannot explain at this time.
+
The degradation products observed in this study are given
in ESI Tables S6–S9,† and are generally identical to those
observed in previous DGA radiolysis studies.12–14,16 A sche-
matic representation of possible degradation reactions and the
corresponding degradation products is shown in Fig. 3.
Comparable to the radiolysis study of hydrophilic DGAs in
water,16 products of single and double de-alkylation reactions,
de-amination reactions (with subsequent oxidation to form the
carboxylic acid), and Oether–Cether bond breaks were observed.
Again, no products of a Camide–Cether bond cleavage could be
identified, although such products were observed for lipophilic
DGAs with low abundance.12–14
The alcohols formed through Oether–Cether bond cleavage
were identified in all irradiated samples. For Me-TEDGA two
different products can be formed, depending on which of the
two different Oether–Cether bonds are broken. In this study, we
observed that the degradation product of the unsubstituted
−
Based on analogous reactions in the literature,19 the eaq
reaction with these DGAs is believed to occur with the carbonyl
groups in these molecules:
•ꢁ
eaqꢁ þ TEDGA ! ½TEDGAꢀ
:
ð9Þ
These reactions are slow, but slightly faster than the
measured H• reactions. For this radical, the presence of a
single methyl group does not appear to significantly impact
reactivity. However, having two methyl groups present, as in
Me2-TEDGA, again provides greater steric hindrance to this
reductive process.
•
The new NO3 reaction rate coefficients for the methylated
O
ether–Cether bond break is present in much higher intensity
TEDGA compounds were considerably faster than TEDGA
itself.16 A re-measurement of this rate coefficient using a new
batch of TEDGA gave a much higher rate constant, (1.12
0.05) × 109 M−1 s−1, see Table 1 and ESI.† This new value is far
more consistent with additional reactivity at the four extra
methylene groups in this molecule as compared to TMDGA,
with decreasing reactivity shown for the degree of methylation
(ca. 20-times higher). Additionally, the reaction products of
the single de-alkylation and the alcohol radical formed by
O
ether–Cether bond cleavage (see Fig. 4) were detected with
the product deriving from the unsubstituted Oether–Cether
bond break again being found in higher intensity. Therefore,
•
of this molecule, as expected. The reactions between the NO3
and these hydrophilic DGAs also suggests that the •NO3 is
reacting by a combination of both electron and hydrogen atom
abstraction:40
•þ
•NO3 þ TEDGA ! NO3ꢁ þ ½TEDGAꢀ
ð10Þ
and
•
•NO3 þ TEDGA ! HNO3 þ ½TEDGAꢀ :
ð11Þ
The dependence of all these radicals’ reaction rate coeffi-
cients on the availability of the methylene centers is observed,
which presents an opportunity to incorporate radiation resis-
tance/robustness into the design methodology for ligands
intended for radiation environments.
Fig. 3 Schematic representation of possible degradation reactions and
corresponding degradation products.
Degradation product distribution
The degradation products from irradiated hydrophilic DGAs
were identified using HPLC-MS with a high-resolution mass
spectrometer, allowing identification of these products by
their m/z ratios and the corresponding chemical formulae. The
intensity of the identified degradation products were followed
by measuring the area of the corresponding peaks in the chro-
Fig. 4 Reaction scheme of the combination reaction of the single de-
alkylation product radical with the alcohol radical formed by Oether
–
matograms when only the m/z ratio of the respective degra- Cether bond cleavage.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17005–17013 | 17009

View Article Online
Paper
Dalton Transactions
we assume that the methyl-substituted Oether–Cether bond nificant impact upon separation process performance than
reacts slower with the primary radicals from solvent radiolysis those degradation products formed under pure water con-
(although the overall degradation rate of the molecule may be ditions alone. These NOx adducts need to be isolated and their
less affected). These observations are consistent with our rate respective chemistries elucidated.
coefficient trends, and have been previously postulated for the
lipophilic analogue of Me-TEDGA.14
Ligand degradation rates
Addition products of the hydrophilic DGAs with radiolytic The measured and calculated rates of gamma-induced radio-
NOx species (e.g., NO, •NO3, and •NO2) were also observed. lytic degradation of TEDGA, Me-TEDGA, Me2-TEDGA, and
Interestingly, for the unsubstituted TEDGA and TPDGA mole- TPDGA in concentrated, aqueous nitrate solution are shown in
cules mainly NO and NO2 adducts were detected, while NO3 Fig. 5. All four molecules are susceptible to gamma radiolysis
adducts were seen only with little intensity. All three NOx and exhibit first order decays with increasing absorbed dose,
addition products were observed for Me-TEDGA and Me2- the dose constants for which are given in Table 2.18 As with
TEDGA. Similar reaction products of radiolytic NOx species their lipophilic analogues, the hydrophilic DGA dose
with the single de-alkylation products were also measured.
constants in concentrated, aqueous nitrate solution are rela-
Overall, similar product distributions were observed for the tively insensitive to alkyl substitution, affording an average
radiation-induced degradation of the hydrophilic DGAs in con- value of (−3.18 0.23) × 10–6 Gy−1. The average hydrophilic
centrated NaNO3(aq) as in pure water conditions, indicating DGA dose constant is much lower than the complimentary
similar radiolytic pathways. However, in concentrated dose constants reported for pure water.16 This is indicative of
NaNO3(aq) solutions, reaction products with radiolytic NOx significant differences in radiation chemistry between the two
species are observed. These addition products may have metal systems. As previously discussed, the presence of concentrated
complexing ability of their own, and thus may have more sig- nitrate significantly alters the available radiolytic species for
Fig. 5 Gamma radiolysis of hydrophilic DGAs in concentrated, aqueous nitrate solution as a function of absorbed dose: (A) TEDGA; (B) Me-TEDGA;
(C) Me2-TEDGA; and (D) TPDGA. Curves are from multi-scale calculations predicting DGA decay without secondary degradation product reactions
•
(dashed black) and with secondary degradation product reactions (solid black) and corresponding accumulated reaction kinetics for OH (red) and
•NO3 (blue).
17010 | Dalton Trans., 2019, 48, 17005–17013
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Table 2 Pseudo-first-order dose constants for the gamma irradiation
of TEDGA, Me-TEDGA, Me2-TEDGA, and TPDGA under pure water and
concentrated, aqueous nitrate conditions
Dose constant (×10−6 Gy−1
)
Hydrophilic DGA
Pure water
Concentrated aqueous nitrate
TEDGA
−10.3 1.1316
−7.8 0.716
−7.3 2.116
−8.39 0.29
−2.36 0.25
−2.93 0.17
−3.90 0.21
−3.54 0.29
Me-TEDGA
Me2-TEDGA
TPDGA
reaction with the hydrophilic DGAs. Multi-scale calculations pre-
sented in Fig. 5 show that radiolytic degradation is driven by reac-
•
•
tion with OH and, to a lesser extent, NO3, which is consistent
with their kinetics in Table 1. Reaction of the hydrophilic DGAs
Fig. 6 Hydrophilic and lipophilic DGA gamma radiolysis dose constants
as a function of molecular weight and solvent formulation: pure water
•
■
with the OH was invoked to predominantly explain their rate of
(
, this work and ref. 16), 5.0 and 3.39 M NaNO3(aq) ( , this work),
degradation under pure water conditions.16
n-dodecane ( , this work and ref. 12, 13, 14, and 16), and n-dodecane/
HNO3(org) , ref. 12, 13, and 14).
(
However, the differences between dose constants for pure
water and concentrated, aqueous nitrate must be due to:
1. Chemistries associated with the reducing products of
water radiolysis (i.e., the eaq and the H•). In concentrated,
−
Overall, the presence of concentrated, aqueous nitrate sig-
nificantly reduces the rate of hydrophilic DGA degradation,
relative to pure water. When hydrophilic and lipophilic DGA
dose constants are compared in Fig. 6, we can see that the
radiation robustness of the hydrophilic DGAs in concentrated,
aqueous nitrate solution is similar to that of their lipophilic
analogues. This is most clearly demonstrated by TPDGA
(301.5 g mol−1), which is soluble in both aqueous and organic
media, and found to react with R•+ with a rate coefficient of
aqueous nitrate solutions, there is negligible contribution from
the eaq− and H•, as both are rapidly scavenged by NO3− within the
lifetime of the radiation chemical track; reactions (3) and (4).
•
•
2. Reduced reactivity of the NO3 relative to the OH, which
is clear from the measured rate coefficients given in Table 1. In
addition to reactivity differences, the •NO3 rate coefficients exhibit
less selectivity for the methylene sites, which may provide some
explanation as to why the hydrophilic DGA dose constants are
less sensitive to alkyl substitution under concentrated, aqueous
nitrate conditions compared with pure water conditions.
(6.18 0.39) × 1010 M−1 s−1
:
•þ
R
•þ þ TPDGA ! R þ ½TPDGAꢀ
:
ð13Þ
Agreement between experiment and calculation was
achieved by the incorporation of additional •OH and •NO3 reac-
tions with degradation products from hydrophilic DGA radioly-
sis. The importance of these secondary scavenging processes
has been recently highlighted by Horne et al. for similarly
aqueous miscible sulfonated bis-triazinyl-bipyridine ligands,41
and is clearly demonstrated by comparison of calculations
with (solid black) and without (dashed black) secondary degra-
dation product scavenging reactions in Fig. 5. In their absence,
multi-scale calculations predict a faster rate of DGA degra-
dation than is observed. This demonstrates that certain degra-
dation products (e.g., the alcohol species shown in Fig. 3)
accumulate and compete for the oxidizing radical species, in
turn reducing the extent to which they react with the parent
DGA compounds. In the absence of a comprehensive study of
the radiolytic behaviour of all identified degradation products,
presented multi-scale calculations employ a simple degra-
dation product (DP) scavenging reaction with an optimized
rate coefficient to achieve agreement with experiment:
The similarity between concentrated, aqueous nitrate and
organic media DGA dose constants is most promising for
large-scale separation process applications and indicates
that in the presence of concentrated nitrate, acid hydrolysis,
n̲o̲t̲ r̲a̲d̲i̲o̲l̲y̲s̲i̲s̲, will limit their useful lifetime. Complementary
nitric acid radiolysis studies are underway to further evaluate
the potential of these hydrophilic DGAs.
Conclusions
Hydrophilic DGAs are promising actinide and lanthanide strip-
ping and hold-back agents, although their useful lifetime is
constrained by water radiolysis a̲n̲d̲ nitric acid hydrolysis –
both are significant factors under expected large-scale separ-
ation process conditions. However, the speciation of proble-
−
matic primary water radiolysis products (i.e., eaq
,
•H, and
•OH) is significantly different under concentrated nitrate/nitric
acid conditions.
•
DGA DP þ OH=•NO3 ! Other products:
ð12Þ
To discriminate between radiolytic and hydrolytic effects,
the irradiation of a series of hydrophilic DGAs (TEDGA, Me-
However, this does not preclude these steady-state radioly- TEDGA, Me2-TEDGA, and TPDGA) in concentrated, aqueous
sis products (e.g., hydrogen peroxide and the nitrite anion) nitrate solution, has shown that the rate of radiolytic degra-
from being involved in radical capping processes.
dation is significantly reduced, relative to pure water con-
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17005–17013 | 17011

View Article Online
Paper
Dalton Transactions
ditions. Further, their radiation-induced decay affords an
average dose constant ((−3.18 0.23) × 10−6 Gy−1) similar to
that of their lipophilic analogues, many of which have been
successfully employed in scaled-up separation processes.
Multi-scale calculations, supported by experimental reac-
tion kinetics and degradation product distribution measure-
ments, have demonstrated that hydrophilic DGA degradation
in the presence of concentrated, aqueous nitrate is driven by
3 S. A. Ansari, P. Pathak, P. K. Mohapatra and
V. K. Manchanda, Chem. Rev., 2012, 112, 1751.
4 C. Musikas and H. Hubert, Solvent Extr. Ion Exch., 1987, 5,
877.
5 S. Chapron, C. Marie, G. Arrachart, M. Miguirditchian and
S. Pellet-Rostaing, Solvent Extr. Ion Exch., 2015, 33, 236.
6 S. Chapron, C. Marie, V. Pacary, M. T. Duchesne,
G. Arrachart, S. Pellet-Rostaing and M. Miguirditchian,
Procedia Chem., 2016, 21, 133.
7 L. Klaß, A. Wilden, F. Kreft, C. Wagner, A. Geist,
P. J. Panak, I. Herdzik-Koniecko, J. Narbutt and G. Modolo,
Solvent Extr. Ion Exch., 2019, 37, 297.
8 Y. Sasaki, Y. Sugo, Y. Kitatsuji, A. Kirishims, T. Kimura and
G. R. Choppin, Anal. Sci., 2007, 23, 727.
9 G. J. Lumetta, A. V. Gelis, J. C. Carter, C. M. Niver and
M. R. Smoot, Solvent Extr. Ion Exch., 2014, 32, 333.
10 C. Rostaing, C. Poinssot, D. Warin, P. Baron and B. Lorrain,
Procedia Chem., 2012, 7, 367.
•
•OH and, to a lesser extent, NO3 oxidation, as the reducing
−
products of water radiolysis are inhibited by NO3 scavenging.
Further, net DGA decomposition is progressively inhibited by
•
reaction of their own degradation products with the OH and
•
the NO3. Consequently, we can expect for acid hydrolysis, n
̲
̲
̲
r̲a̲d̲i̲o̲l̲y̲s̲i̲s̲, to limit their useful lifetime, especially in concen-
trated HNO3 solutions, where the •NO3 is the predominant
oxidant due to •OH scavenging by HNO3.
11 S. Lange, A. Wilden, G. Modolo, F. Sadowski, M. Gerdes
and D. Bosbach, Solvent Extr. Ion Exch., 2017, 35, 161.
12 C. A. Zarzana, G. S. Groenewold, B. J. Mincher, S. P. Mezyk,
A. Wilden, H. Schmidt, G. Modolo, J. F. Wishart and
A. R. Cook, Solvent Extr. Ion Exch., 2015, 33, 431.
13 K. M. Roscioli-Johnson, C. A. Zarzana, G. S. Groenewold,
B. J. Mincher, A. Wilden, H. Schmidt, G. Modolo and
B. Santiago-Schubel, Solvent Extr. Ion Exch., 2016, 34, 439.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
This research has been funded by the US Department of
Energy (US-DOE) Assistant Secretary for Nuclear Energy, under 14 H. Galan, C. A. Zarzana, A. Wilden, A. Nunez, H. Schmidt,
the Fuel Cycle Research & Development Radiation Chemistry
program; DOE-Idaho Operations Office Contract DE-AC07-
05ID14517 and Nuclear Energy Universities Program (NEUP)
DE-NE0008406 grant.
The time-resolved pulsed electron irradiations reported
herein were performed at the Notre Dame Radiation
R. J. M. Egberink, A. Leoncini, J. Cobos, W. Verboom,
G. Modolo, G. S. Groenewold and B. J. Mincher, Dalton
Trans., 2015, 44, 18049.
15 Y. Sugo, Y. Izumi, Y. Yoshida, S. Nishijima, Y. Sasaki,
T. Kimura, T. Sekine and H. Kudo, Radiat. Phys. Chem.,
2007, 76, 794.
Laboratory, supported by the Division of Chemical Sciences, 16 A. Wilden, B. J. Mincher, S. P. Mezyk, L. Twight,
Geosciences and Biosciences, Basic Energy Sciences, Office of
Science, US-DOE through Award No. DE-FC02-04ER15533, and
the Brookhaven National Laboratory (BNL) Laser Electron
K. M. Rosciolo-Johnson, C. A. Zarzana, M. E. Case,
M. Hupert, A. Stärk and G. Modolo, Solvent Extr. Ion Exch.,
2018, 36, 347.
Accelerator Facility of the BNL Accelerator Center for Energy 17 T. Koubský and J. Lustinec, J. Radioanal. Nucl. Chem., 2018,
Research, supported by the US-DOE Office of Basic Energy 318, 2407.
Sciences, Division of Chemical Sciences, Geosciences, and 18 B. J. Mincher and R. D. Curry, Appl. Radiat. Isot., 2000, 52,
Biosciences under contract DE-SC0012704. 189.
All gamma irradiation analyses were performed at 19 G. V. Buxton, C. L. Greenstock, W. P. Helman and
Forschungszentrum Jülich, with financial support from the A. B. Ross, J. Phys. Chem. Ref. Data, 1988, 17, 513.
European Commission through the Safety of Actinide 20 Y. Yoshida, T. Ueda, T. Kobayashi, H. Shibata and
Separation Processes (SACSESS) project, under contract No. S. Tagawa, Nucl. Instrum. Methods Phys. Res., 1993, 327, 41.
FP7-Fission-2012-323-282, and the European GEN IV 21 W. Davis and H. J. De Bruin, J. Inorg. Nucl. Chem., 1964, 26,
Integrated Oxide Fuels Recycling Strategies (GENIORS) project,
grant agreement No. 730227.
1069.
22 P. Y. Jiang, Y. Katsumura, K. Ishigure and Y. Yoshida, Inorg.
Chem., 1992, 31, 5135.
23 S. P. Mezyk and D. M. Bartels, J. Phys. Chem. A, 1997, 101,
6233.
24 T. Logager and K. Sehested, J. Phys. Chem., 1993, 97, 6664.
Notes and references
1 R. Taylor, Reprocessing and Recycling of Spent Nuclear Fuel, 25 A. J. Swallow and M. Inokuti, Radiat. Phys. Chem., 1988, 32,
Woodhead Publishing, Oxford, 1st edn, 2015. 185.
2 D. Whittaker, A. Geist, G. Modolo, R. Taylor, M. Sarsfield 26 Y. Katsumura, The Chemistry of Free Radicals: N-Centered
and A. Wilden, Solvent Extr. Ion Exch., 2018, 36, 223.
Radicals, John Wiley & Sons, Chichester, 1998.
17012 | Dalton Trans., 2019, 48, 17005–17013
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
27 A. Leoncini, J. Huskens and W. Verboom, Synlett, 2016, 27, 35 S. M. Pimblott, J. A. LaVerne and A. Mozumder, J. Phys.
2463.
Chem., 1996, 100, 8595.
28 H. Fricke and E. J. Hart, J. Chem. Phys., 1935, 3, 60.
29 N. F. Barr and R. H. Schuler, J. Phys. Chem., 1959, 63, 808.
36 S. M. Pimblott and J. A. LaVerne, J. Phys. Chem. A, 2002,
106, 9420.
30 K. Whitman, S. Lyons, R. Miller, D. Nett, P. Treas, A. Zante, 37 P. Clifford, N. J. B. Green, M. J. Oldfield, M. J. Pilling and
R. W. Fessenden, M. D. Thomas and Y. Wang, Linear S. M. Pimblott, J. Chem. Soc., Faraday Trans., 1986, 82, 2673.
Accelerator for Radiation Chemistry Research at Notre 38 P. Y. Jiang, R. Nagaishi, T. Yotsuyanagi, Y. Katsumura and
Dame, Proceedings of the ‘95 Particle Accelerator Conference K. Ishigure, J. Chem. Soc., Faraday Trans., 1994, 90, 93.
and International Conference on High Energy Accelerators, 39 A. J. Elliot and D. M. Bartels, The Reaction Set, Rate
Texas, USA, 1996.
Constants and G-Values for the Simulation of the
Radiolysis of Light Water Over the Range 20° to 350 °C
Based on Information Available in 2008, AECL Nuclear
Platform Research and Development − Report 153-127160-
450-001, 2009.
31 G. L. Hug, Y. Wang, C. Schöneich, P. Y. Jiang and
R. W. Fessenden, Radiat. Phys. Chem., 1999, 54, 559.
32 J. F. Wishart, A. R. Cook and J. R. Miller, Rev. Sci. Instrum.,
2004, 75, 4359.
33 G. V. Buxton and C. R. Stuart, J. Chem. Soc., Faraday Trans., 40 S. P. Mezyk, T. D. Cullen, K. A. Rickman and B. J. Mincher,
1995, 92, 279. Int. J. Chem. Kinet., 2017, 49, 635.
34 G. P. Horne, T. A. Donoclift, H. E. Sims, R. M. Orr and 41 G. P. Horne, S. P. Mezyk, N. Moulton, J. R. Peller and
S. M. Pimblott, J. Phys. Chem. B, 2016, 120, 11781.
A. Geist, Dalton Trans., 2019, 48, 4547.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 17005–17013 | 17013