Selective uptake of thorium(IV) from nitric acid medium using two extraction chromatographic resins based on diglycolamide-calix[4]arenes: Application to thorium-uranyl separation

Journal of Chromatography A 1653 (2021) 462401

Contents lists available at ScienceDirect

Journal of Chromatography A
journal homepage: www.elsevier.com/locate/chroma

Selective uptake of thorium(IV) from nitric acid medium using two
extraction chromatographic resins based on
diglycolamide-calix[4]arenes: Application to thorium-uranyl separation
in an actual sample
Rajesh B. Gujar a, Prasanta K. Mohapatra a, Mudassir Iqbal b, Jurriaan Huskens b,
Willem Verboom b,∗
a
b

Radiochemistry Division, Bhabha Atomic Research Centre, Mumbai 400 085, India
Laboratory of Molecular Nanofabrication, MESA+ Institute for Nanotechnology, University of Twente, P. O. Box 217, 7500 AE Enschede, the Netherlands

a r t i c l e

i n f o

Article history:
Received 19 March 2021
Revised 1 July 2021
Accepted 3 July 2021
Available online 9 July 2021
Keywords:
Thorium(IV)
Extraction chromatography

Diglycolamide
Calix[4]arene

a b s t r a c t
Two novel extraction chromatography resins (ECRs) containing two diglycolamide (DGA) -functionalized
calix[4]arenes with n-propyl and isopentyl substituents at the amide nitrogen atom, termed as ECR-1
and ECR-2, respectively, were evaluated for the uptake of Th(IV) from nitric acid feed solutions. While
both the resins were having a quite high Th(IV) uptake ability (Kd >30 0 0 at 3 M HNO3 ), the uptake was
relatively lower with the resin containing the isopentyl DGA, which appeared magnified at lower nitric
acid concentrations. Kinetic modeling of the sorption data suggested fitting to the pseudo-second order
model pointing to a chemical reaction during the uptake of the metal ion. Sorption isotherm studies were
carried out showing a good fitting to the Langmuir and D-R isotherm models, suggesting the uptake conforming to monolayer sorption and a chemisorption model. Glass columns with a bed volume of ca. 2.5
mL containing ca. 0.5 g lots of the ECRs were used for studies to assess the possibility of actual applications of the ECRs. Breakthrough profiles obtained with feed containing 0.7 g/L Th(NO3 )3 solution resulted
in breakthrough volumes of 8 and 5 mL, respectively, for the ECR-1 and ECR-2 resins. Near quantitative
elution of the loaded metal ion was possible using a solution of oxalic acid and nitric acid. A method for
the separation of Th-234 from natural uranium was demonstrated for the possible application of ECR-1.
© 2021 The Author(s). Published by Elsevier B.V.
This is an open access article under the CC BY license ( />
1. Introduction
Solid phase extraction is considered as an alternative separation method to solvent extraction and can alleviate some of the
issues faced with the latter such as third phase formation, phase
entrainment, phase disengagement limitations, etc. Though solid
phase extraction (SPE) involves sorbents for neutral molecules or
ion exchange resins, there is a type of sorbents based on extraction
chromatography (EC) where, the organic extractant is impregnated
into an inert solid support material, is fast emerging as an efficient
SPE technique with highly promising results [1–4]. Furthermore,
the extractant inventory can be very low in case of EC and hence,
the cost of the separation method can be quite low. This suggests
that exotic extractants can be easily used in an EC method with-

∗

Corresponding author.
E-mail address: (W. Verboom).

out any significant cost consideration. A variety of solid support
materials have been used viz. XAD-4 [5], XAD-7 [5], voltalef [6],
Amberchrom – CG 71 [7], Chromosorb 102 [8], Chromosorb W [8],
etc. Out of these, Chromosorb W (dimethyl dichlorosilane treated
acid washed celite diatomaceous silica) is quite interesting since
the major constituent of this material is silica, which has good radiation stability and hence, can be easily used for radioactive feeds.
For the separation of actinide ions from acidic feeds, many
extractants have been used [9]. However, diglycolamide (DGA)
extractants such as TODGA (N,N,N’,N’-tetra-n-octyl diglycolamide)
are quite promising [10]. TODGA-based extraction chromatography resins (ECR) have been prepared by many researchers and
used for the separation of trivalent actinide ions from acidic feeds
[7,11,12]. It has been reported that multiple DGA extractants such
as the ones with a calix[4]arene scaffold display a much higher
separation efficiency than TODGA [13–15]. The DGA-functionalized
calix[4]arenes (termed as C4DGAs) with a branched alkyl chain offer a better selectivity, albeit a lower extraction than a linear alkyl

/>0021-9673/© 2021 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license ( />

R.B. Gujar, P.K. Mohapatra, M. Iqbal et al.

Journal of Chromatography A 1653 (2021) 462401

2.2. Radiotracers
233 U was used from the laboratory stock after carrying out

a fresh purification from its daughter products using a reported
method [18]. Carrier free 234 Th was obtained by freshly separating the radiotracer from an ‘aged’ solution of natural uranium in
6 M HNO3 by extractive separation of the latter into a solution
of Aliquat 336 in chloroform following a reported method [19].
The aqueous phase contained predominantly the carrier free 234 Th
tracer with <0.1% U. In all studies involving the 234 Th tracer, the
concentration of the metal ion was ca. 10−12 – 10−13 M, while the
concentration of U in the studies involving 233 U tracer was ca. 10−5
M.

2.3. Radiometric assay
Both 233 U and 234 Th were assayed radiometrically. While the
former was assayed by liquid scintillation counting using an instrument obtained from Hidex, Finland, 234 Th was assayed using a
well type NaI(Tl) scintillation counter (Para Electronics), which was
interphased with a multichannel analyzer (ECIL, India). The Ultima
Gold liquid scintillation cocktail was purchased from Perkin Elmer,
USA.

Fig. 1. Structure of the C4DGA ligands LI and LII .

chain substituted ligand [15]. The uptake of trivalent actinide / lanthanide ions by a C4DGA-based ECR is much superior to that of
the corresponding TODGA-based resin [16]. Recently, we reported
the extraction behavior of tetravalent actinide ions such as Np(IV)
and Pu(IV) with linear and branched chain alkyl group containing
C4DGA ligands [15]. However, the extraction behavior of Th(IV) is
not studied so far. It is of interest from the point of view of U(VI)
/ Th(IV) separation, which is relevant for AHWR (advanced heavy
water reactor) applications [17].
Therefore, we embarked on a study to the uptake behavior of
Th(IV) and U(VI) with two novel extraction chromatography resin

(ECR) materials termed as ECR-1 and ECR-2 containing the C4DGA
ligands LI and LII , respectively (Fig. 1) impregnated into the pores
of the solid support material Chromosorb W. The separation behavior was studied by batch studies and subsequently, separation
was demonstrated using a mixture of U and Th. Kinetic modeling and sorption isotherm studies were also carried out. To our
mind, this is the first ever report on the separation of Th(IV) from
U(VI) using an ECR, impregnated with calix[4]arene-based DGA
ligands.

2.4. Preparation of the EC resins
The EC resins were prepared by a previously reported method
[20] which required vortex mixing of known quantities of the ligands LI and LII and Chromosorb W taken in a stoppered 100 mL
Erlenmeyer containing 20 mL acetone, for 24 h. Afterwards, the
slurry was settled, and the acetone was removed by careful flushing of N2 gas through a jet. After the evaporation of >99% of
the solvent, the ECRs were kept in a vacuum desiccator for drying overnight to yield free flowing materials for subsequent uptake
studies. The weight difference between the ECRs and the support
material (Chromosorb W) suggested about 8% (w/w) ligand loading
for both the resins.
2.5. Characterization of the SPE resins
The ECRs were characterized by thermogravimetry (TGA) for
ligand loading and by SEM for the surface morphology; the results are given in a previous paper [21]. The presence of the extractants inside the resin pores was characterized by FTIR recorded
on a Bruker Alpha II ATR-FTIR spectrometer.

2. Experimental
2.6. Batch uptake studies
2.1. Chemicals
The batch uptake studies involved acidic feeds containing Th(IV)
and UO2 2+ ions. For the batch studies, ca. 10 mg of the ECR was
taken in a stoppered Pyrex glass tube (10 mL capacity) containing
the radiotracer spiked dilute nitric acid solution. The tubes were
agitated in a thermostated water bath at 25 ± 0.1 °C for about 1

hour. The time taken to attain equilibrium metal ion uptake values
was much lower than the employed time which was done to ensure attainment of equilibrium. The tubes were then rested for 5
minutes and centrifuged at 30 0 0 rpm before removal of the samples for radiometric assay. Usually, 100 μL aliquots were removed
for assaying purpose using Eppendorf fixed volume micropipettes.
The weight distribution coefficient (Kd ) values of the metal ions
were calculated using the following equation:

The DGA-functionalized calix[4]arene (C4DGA) ligands LI and LII
(Fig. 1) have been described before [13]. The same batch was used,
again characterized by comparison of both the 1 H NMR spectra
(Fig. S1) with the previous data and of the distribution coefficients
for the extraction of Am(III) and Eu(III) using 1 mM ligand solutions of 5% isodecanol and 95% n-dodecane at 3 M HNO3 . For ligand LI DAm and DEu values were obtained of 78±3.1 and 242±12
versus 79.5 and 225, respectively, in ref. [13]. For ligand LII the DAm
and DEu values were 314±10 and 357±14 versus 325 and 370, respectively, in ref. [13]. Chromosorb-W (mesh size 60-80) was purchased from John Manville, USA and was cleaned and dried prior
to use by rinsing with methanol and air drying at 70 °C. Suprapur nitric acid (Merck) was used for the preparation of dilute nitric acid solutions for the uptake studies. Thorium nitrate was purchased from Indian Rare Earths limited, while uranyl nitrate hexahydrate was obtained from Uranium Extraction Division, BARC. All
other chemicals were of AR grade.

Kd =

(Co − C )
C

·

V
, mL/g
W

(1)

where Co and C are the initial and equilibrium concentrations of
the metal ions, respectively, V is the aqueous phase volume (in
2


R.B. Gujar, P.K. Mohapatra, M. Iqbal et al.

Journal of Chromatography A 1653 (2021) 462401

Table 1
Various parameters for the column studies using the ECR-1 and ECR-2 resins.
Parameter
Resin weight

ECR-1 (n-propyl)
0.503 g

ECR-2 (isopentyl)
0.501 g

Bed height
Bed volume
Bed density
Flow rate

20.5 cm
2.57 mL
0.022 g/cm3
0.05 mL per minute

20 cm
2.51 mL
0.022 g/cm3
0.05 mL per minute

Fig. 3. Th(IV) uptake from 3 M HNO3 as a function of equilibration time using the
two C4DGA containing resins ECR-1 and ECR-2.

the other hand, the C4DGA loaded resins show the >C=O bands
at 1650 cm−1 indicating the presence of the carbonyl groups of
the C4DGA ligands present in the pores of the ECRs. For comparison purpose, the two ECRs were contacted with Th(IV) solution
and the FTIR spectra of the Th(IV) loaded resins are also presented
in Fig. 2. A clear shift in the >C=O stretching frequency to lower
values (to ca. 1610 cm−1 ) suggested binding of Th(IV) ions to the
carbonyl groups present in the resins.

Fig. 2. FTIR spectra of the ECRs. Top: Pristine Chromosorb W (green); Freshly made
ECR-1 (black); Freshly made ECR-2 (red). Bottom: Th(IV) loaded ECR-1 (black);
Th(IV) loaded ECR-2 (red). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).

3.2. Batch uptake studies
mL), and W the weight of the resin (in g). As radioactive solutions were used in the present study, the concentrations are conveniently expressed in terms of counts per unit time per unit volume. All batch studies were carried out in duplicate and the mass
balance was found to be within 5%.

The uptake of the metal ions was negligible (Kd values being
2 ± 0.2 and 5 ± 0.4, respectively, for U(VI) and Th(IV) at 3 M
HNO3 ) when pristine Chromosorb W (without any ligand loaded
onto it) was used as the sorbent. On the other hand, a very large
increase in the Kd values was observed, especially for Th(IV), when
the two ECRs were used at 3 M HNO3 (vide infra). The Kd values

of U(VI) were extremely low and were very close to the value obtained with bare Chromosorb W as the sorbent (vide infra), suggesting weak binding of the metal ion with the C4DGA ligands.
This is in line with the results obtained from the solvent extraction studies reported before [14].

2.7. Column studies
The column studies were carried out using small glass columns
of the dimension: 30 cm (length) × 0.4 cm (dia); bed volume ca.
2.5 mL. The columns were filled with a slurry containing the resins
in distilled water and subsequently allowed to settle to get resin
beds without any air bubble. The top of the columns was fitted
with a glass cup (10 mL), while the bottom part was fitted with
flexible rubber tubing and pinch cocks for maintaining a constant
flow rate. The detailed column parameters are given in Table 1.
The columns were conditioned prior to the loading of the metal
ions by passing 10 mL of HNO3 . For the breakthrough studies, the
column was loaded with a feed containing 0.7 g/L Th solution in
3 M HNO3 . The feed was spiked with 234 Th tracer for easy column data analysis. After loading of the Th solution, the column
was washed with 3 M HNO3 prior to the elution of the loaded Th
using a mixture of 0.5 M oxalic acid and 0.5 M HNO3 . The loading
of the uranium solution was performed in an identical manner, but
as mentioned below the effect of the loading was quite insignificant.

3.2.1. Time of equilibration
For the column operations discussed below the time taken to
attain equilibrium was determined. Fig. 3 gives the plots of Kd vs
time for both the resins for Th(IV), showing an equilibrium time
of about 30 minutes. U(IV) data are of less significance as the Kd
values obtained were very low (in the range of 5 to 30, which are
about three orders of magnitude lower than the Kd values obtained
with Th(IV)). The equilibrium Kd values for U(IV) were attained in
ca. 20 minutes but are not included in the plot.

3. Results and discussion

3.2.2. Effect of feed acid concentration
It is well known from the solvent extraction data reported by
Zhu et al. [22] that the extraction of Th(IV) is highly favourable
with DGA-based extractants such as TODGA. The extraction of the
metal ion increases sharply with the feed acid concentration as per
the following equation:

3.1. Characterization of the extraction chromatography resins

Th4+ aq + n TODGAorg + 4 NO3 − aq = Th(NO3 )4 •nTODGAorg

The FTIR spectra of the resins were recorded (Fig. S2). However, for the sake of checking C4DGA loading in the resins, the FTIR
bands in the range of 1700 – 1550 cm−1 are magnified and given
in Fig. 2. Pristine Chromosorn W has no bands in this region. On

where the subscripts ‘aq’ and ‘org’ refer to the species present in
the aqueous and the organic phases, respectively, and n is 3.39
for TODGA. On the other hand, the extractant dependence for the
C4DGA ligands LI and LII is not known. Since there are four DGA
3

(2)


R.B. Gujar, P.K. Mohapatra, M. Iqbal et al.

Journal of Chromatography A 1653 (2021) 462401

Fig. 4. Th(IV) and U(VI) uptake profiles as a function of aqueous feed nitric acid concentration using the C4DGA-containing resins ECR-1 and ECR-2. The separation factor
(SF) data are also presented.

Table 2
Bach extraction of Th(IV) using various stripping solutions from ECR-1 and ECR-2
resins.

moieties present in the C4DGA ligands, the value of n could be 1.
Consequently, Eq. (1) can be extended to the following equation
when the ligand is impregnated in the ECR.

Th4+ aq + C4DGAR + 4 NO3 − aq = Th(NO3 )4 .C4DGAR

(3)

where the subscript ‘R’ represents species in the resin phase. It is
clear from the above equations that the extraction of Th(IV) ion
has a positive dependence with the acid concentration of the feed
and hence, should increase with the feed acid concentration. The
Th(IV) uptake data (quantified in terms of Kd ) as a function of the
feed HNO3 concentration (in the range of 0.5 M – 6 M) are presented in Fig. 4. The Kd values for U(VI), obtained under identical
experimental conditions, are also plotted in the same figure. Uptake data were not obtained at nitric acid concentrations < 0.5 M
due to the possibility of hydrolysis of the tetravalent metal ion at
these conditions.
As shown in Fig. 4, the Kd values for Th(IV) increase steeply at
lower acid concentrations and reach a plateau at nitric acid concentrations > 3 M. On the other hand, U(VI) uptake shows a steady
increase in the entire range of nitric acid concentrations. The separation factor (SF = Kd,Th /Kd,U ) values were calculated and plotted in
the same figure for a better impression of the results. The SF values exhibit a rapid rise at lower acid concentrations and a sharp
decline at higher acidities. However, at 3 M HNO3 , which is the

acid concentration of interest in most radioactive wastes, SF values
of >100 suggest that the resins can be employed for the separation
of Th(IV) from U(VI), which has relevance in THOREX type feeds
[23] for application in AHWR. Finally, comparing the efficiency of
the two resins, ECR-1 is superior to ECR-2, the reason being a less
efficient extraction of the metal ions with C4DGA ligand LII containing branched alkyl chains, as discussed before [15].

Stripping solution

% Stripping
(ECR-1)

% Stripping
(ECR-2)

0.5 M oxalic acid
0.5 M oxalic acid + 0.5 M nitric acid
0.05 M HEDTA + 0.05 M nitric acid

84.7
86.5
87.1

84.4
86.1
86.3

Fig. 5. Reusability of the ECRs based on the Th(IV) uptake data after regeneration
using stripping conditions of 0.5 M HNO3 and 0.5 M oxalic acid.

Though HEDTA was marginally better as a stripping agent as
compared to the mixture of oxalic acid and nitric acid, we have
used the latter in view of some previous reports where the stripping mixture has been successfully used. In our previous studies involving tetravalent ions such as Np(IV) and Th(IV), stripping was achieved using oxalic acid or a mixture of it with nitric acid [24,25]. This is due to the rather strong complexation of
these tetravalent ions with oxalate anion [26]. Subsequently, the
reusability of the resin was studied by carrying out repeated uptake and stripping for five cycles (Fig. 5). ECR-1 resin showed almost no change in the Kd values even after about four cycles,
though a clear deterioration was seen after the fourth cycle. Even
the Kd value obtained after four cycles of uptake and stripping was

3.2.3. Back extraction and reusability
The back extraction (stripping) of the loaded Th(IV) was attempted by three stripping solutions, viz. a) 0.5 M oxalic acid; b)
0.5 M oxalic acid + 0.5 M nitric acid; c) 0.05 M HEDTA (N-(2hydroxyethyl)ethylenediamine triacetic acid) + 0.05 M nitric acid.
The back extraction of the loaded metal ion was done by first loading the resin (ca. 20 mg) with 234 Th radiotracer from 3 M HNO3
followed by careful removal of the aqueous phase completely from
the tube. After this, the stripping solution was added (0.5 mL) to
the tube and aqueous phase samples were removed after 1 hour
of equilibration. The results of the back extraction are given in
Table 2.
4


R.B. Gujar, P.K. Mohapatra, M. Iqbal et al.

Journal of Chromatography A 1653 (2021) 462401

Fig. 6. Kinetic modeling of the Th(IV) uptake data as fitted to (a) pseudo-first and (b) pseudo-second order models.

Table 3
Parameters obtained from the straight-line plots obtained by data fitting to Eqs. (4)
and (5).

ca. 20 0 0, which is quite high for metal ion uptake. On the other
hand, the ECR-2 resin appeared stable (based on the Kd values)
only up to two cycles of uptake and stripping runs. The Kd values with ECR-2 fall sharply from ca. 2400 in the second cycle to
ca. 750 in the fifth cycle, suggesting clear degradation which is attributed to the leaching of the extractant from the resin pores.

Pseudo-first order kinetic model
Resin
k1 (min−1 )
ECR-1
0.018 ± 0.005
ECR-2
0.014 ± 0.005
Pseudo-second order kinetic model
Resin
k2 (g/cpm.min)
ECR-1
(2.67 ± 0.01) × 10−6
ECR-2
(3.48 ± 0.01) × 10−6

3.3. Kinetic modeling of the uptake data
To understand the mechanism of sorption, kinetic modeling of
the Th4+ ion uptake data was done. Uptake studies were carried
out at different time intervals (for 3-4 hours) using the two extraction chromatographic resins, where the feed contained 700 mg/L
Th in 3 M HNO3 and 234 Th radiotracer for easy assaying by radiometry (vide supra). The uptake data were fitted using the following
equations:
Pseudo first-order kinetic model:

ln (qe − qt ) = ln qe − k1 · t

R2
0.999
0.999

Table 4
Linearized form of different sorption isotherm models.a

(4)

Pseudo second-order kinetic model:

t/ = 1/
t
qt
k2 qe 2 + /qe

R2
0.7228
0.6967

Model

Linearized form model equation

Langmuir
Freundlich
D-R
Temkin

Ce

qe

[1]
b·qmax

Ce
qmax

+
=
log qe = log K f + 1n log Ce
ln qe = ln qmax − βε 2
qe = BT ln AT + BT ln Ce

Plot

[Ref]

Ce
qe

[31]
[32]
[33,34]
[35]

vs · Ce
log qe vs · log Ce
ln qe vs · ε 2
qe vs · ln Ce

a
qe is the concentration of metal ion sorbed per gram of the solid at equilibrium;
Ce is the equilibrium concentration of the metal ion in the aqueous phase; qmax
and Kf are the maximum sorbed mass of Th(IV) at saturation and the Freundlich
constant, respectively; β and ε represent the D-R constant and Polanyi potential,
respectively; BT and AT are the Temkin constants.

(5)

where qe and qt are the mass of the metal ion retained per unit
mass of the resin at equilibrium and at time ‘t’, respectively. The
pseudo-first order and pseudo-second order rate constants are denoted as k1 and k2 , respectively.
The pseudo first-order kinetic model or the Lagergren model
[27] is operating if the Ln (qe -qt ) vs t plot should conform to
straight-line fits with negative slopes to give the pseudo-first order rate constant (Fig. 6a). On the other hand, when t/qt vs t plots
conform to straight-line fits, it is presumed that the data obey the
Ho’s pseudo second-order model [28]. From Fig. 6 it is clear that
the fit to the Lagergren model is very poor (R2 = 0.72 and 0.70,
for ECR-1 and ECR-2, respectively), while near perfect straight-line
fits are obtained for the pseudo second-order kinetic model, suggesting chemisorption as the rate-limiting step [29] i.e., a chemical
reaction controls the uptake mechanism. This is only logical as the
uptake reaction mainly involves the extraction of the metal ion as
per Eq. (2). The rate constants obtained from the kinetic modeling along with the fitting parameters are listed in Table 3. As indicated from the square of the correlation coefficient (R2 ) values, the
pseudo second-order kinetics has a better fitting and hence, should
be valid for the present case.

the different isotherm models used for data analysis are given in
Table 4. The Th(IV) uptake studies were carried out using a 234 Th
tracer spiked carrier solution containing 1.2 g/L Th in 3 M HNO3 .

The feed acid concentration was chosen since most of the radioactive waste feeds contain ca. 3-4 M HNO3 [30].
The sorption isotherm data when fitted to the Langmuir adsorption isotherm equation (Table 4) by plotting Ce /qe vs Ce yielded
straight lines (Fig. 7a) with very good correlation coefficient values of >0.99 (Table 5). This indicates that the uptake of Th(IV) by
both the ECRs conform to the monolayer model. The qmax values
for ECR-1 and ECR-2 were found to be 12.4 ± 1.3 and 5.1 ± 1.1 mg,
respectively, as against the experimentally determined saturation
Th(IV) uptake values of 13.2 ± 1.4 and 5.8 ± 1.2 mg, showing a
reasonably good agreement. The intercept of the straight-line plots
yield ‘b’, which is related to a dimensionless equilibrium constant
RL (also known as the separation factor [36]) as given by:

RL = 1/(1 + bC )
o

(6)

where a value of RL > 1 suggests an unfavourable sorption. On the
other hand, a favourable sorption process is considered where RL
falls in between 0 and 1, while RL = 0 points to an irreversible uptake process. The experimentally obtained b values were found to
be 0.21 and 0.95, respectively, for ECR-1 and ECR-2, which when

3.4. Sorption isotherms
To understand the nature of the interaction of the metal ion
(Th(IV)) with the ligands present in the resin pores, it was imperative to carry out sorption isotherm studies. The linear forms of
5


R.B. Gujar, P.K. Mohapatra, M. Iqbal et al.

Journal of Chromatography A 1653 (2021) 462401

Fig. 7. Th(IV) uptake data fitted to the linearized forms of (a) Langmuir, (b) Freundlich, (c) D-R, and (d) Temkin sorption isotherms.

Table 5
Parameters calculated from Langmuir, Freundlich, D-R, and Temkin isotherm models
for the sorption of Th(IV) onto the extraction chromatographic resins ECR-1 and
ECR-2; Aqueous phase: 25 mg/L to 300 mg/L Th(IV) solution in 3 M HNO3 .
Isotherms

Parameters

Values at 25 °C
ECR-1

ECR-2

Langmuir

b (mL/mg)
qmax (mg/g)
R2
Kf (mg/g)
n
R2
Xm
(mmol/g)
E (kJ/mole)
RT2
A
BT

R2

0.21 ± 0.01
12.4 ± 1.3 (13.2 ± 1.4)a
0.999
3.18 ± 0.05
2.95 ± 0.06
0.829
0.089 ± 0.010
17.9 ± 1.7
0.996
6.05 ± 0.31
1.85 ± 0.10
0.980

0.95 ± 0.02
5.1 ± 1.1 (5.8 ± 1.2)a
0.996
2.49 ± 0.04
4.82 ± 0.03
0.839
0.024 ± 0.018
54.2 ± 1.8
0.981
3.94 ± 0.4
0.66 ± 0.13
0.816

Freundlich

D-R

Temkin

a

Values in parentheses refer to the experimental values.
Fig. 8. Breakthrough profiles for the Th(IV) uptake columns containing ECR-1 and
ECR-2 resins. Feed contained a 234 Th tracer spiked solution of 3 M HNO3 containing
0.7 g/L Th.

used in Eq. (6) resulted in RL values of 0.799 and 0.467, respectively, and suggest a favourable uptake of the metal ion into the
resins.
When the uptake data were fitted to the Freundlich isotherm
model [32] by plotting log qe vs log Ce , the scatter (Fig. 7b) could
not be fitted well to a linear regression line, the correlation coefficient (R2 ) values being very poor (Table 5). This suggests the
absence of the multilayer sorption phenomenon for the Th(IV) uptake in case of both resins. On the other hand, the linear fitting
of the sorption data to the D-R isotherm model [34] by plotting
ln qt vs ε 2 gave reasonably good correlation coefficients (0.996 for
ECR-1 and 0.981 for ECR-2; Table 5; Fig. 7c). Other parameters of

the fitting of the sorption data to the D-R model are also listed in
Table 5. The slope gives the quantity β which can be correlated
to E (mean sorption energy), which is defined as the free energy
needed to transfer one mole of the Th(IV) ions from infinity to the
surface of the ECRs [35] and is given as:

E = 1/ −2β

(7)

Depending on the value of E one can get a rough idea about
the mechanism of sorption, such as chemisorption (E >8 kJ/mol) or
physisorption (E <8 kJ/mol) [37]. The values of E for Th(IV) uptake
6


R.B. Gujar, P.K. Mohapatra, M. Iqbal et al.

Journal of Chromatography A 1653 (2021) 462401

Fig. 9. Th(IV) breakthrough data fitted to the Thomas kinetic model as given by Eq. (8).

onto ECR-1 and ECR-2 were calculated to be 17.9 ± 1.7 kJ/mol and
54.2 ± 1.8 kJ/mol, respectively (Table 5), indicating that the metal
ion uptake conforms to a chemisorption process via some type of
chemical interactions.
Finally, an attempt was made to fit the sorption data to the
Temkin isotherm, which considers the effects of indirect adsorbateadsorbent interactions on the Th(IV) ion uptake onto the two
resins (Table 5). The main assumption of the Temkin isotherm
model is a multi-layer sorption phenomenon which takes care of
the indirect adsorbate–adsorbent interactions and where the heat
of sorption would more often decrease linearly with coverage. The
Temkin sorption fit lines are presented in Fig. 7d and the fitted parameters are listed in Table 5.
3.5. Column studies
3.5.1. Breakthrough profile of Th(IV)
For obtaining the breakthrough profiles, the Th(IV) feed solution
was loaded onto the column at 1 mL at a time and a total of 14 mL
each of the feed solutions were passed through the two columns.
The breakthrough profiles are plotted in Fig. 8 which show that

ECR-1 is a better sorbent than ECR-2, 8 vs 5 mL, from the Th(IV)
loading point of view. The breakthrough volumes were found to
be 4 and 8 bed volumes for the ECR-1 and ECR-2 based columns,
respectively.
To determine the characteristic parameter of the columns used
in this study, the data obtained from the breakthrough loading
studies were subjected to fitting with the Thomas kinetic model
[38,39] given by the following equation:

Ct
=
C0

1
1 + exp

KT h qads m
Q

− KT hC0 t

Fig. 10. Elution profiles of Th(IV) from columns containing ECR-1 and ECR-2 resins.
Feed: 0.7 g/L Th in 3 M HNO3 and spiked with 234 Th tracer. Eluent: mixture of 0.5
M HNO3 and 0.5 M oxalic acid.

containing ECR-2. In view of this, it is recommended that the ECR1-based column can be used for actual applications to feeds containing Th(IV) in nitric acid.
3.5.2. Separation of 234 Th(IV) from natural U
The batch uptake studies show that U(VI) ion uptake is negligible in view of the very low Kd values at 3 M HNO3 , while the
uptake of Th(IV) is rather high (Table 6). This was used for the separation of 234 Th from natural uranium. A feed containing 1 g/mL of
natural U was first used for the 234 Th as well as 235 U contents as

identified by their gamma ray peaks (234 Th: 63 keV and 93 keV;
235 U: 186 keV) as indicated in Fig. 11a. The loaded U was not held
in the column. The 234 Th tracer held in the column was eluted as
mentioned above using a mixture of oxalic acid and nitric acid. The
separation of Th from U was quite efficient as indicated from the
gamma ray spectra of the eluted fractions from ECR-1 (Fig. 11b)
and ECR-2 (Fig. 11c) which suggested the U content in the Th fractions being below the detection limit of U (0.3 counts per minute).
The decontamination factor (DF) values (defined as the ratio of the
Th / U ratio in the product to that in the feed) were calculated
to be 178 and 165, respectively, for the resins ECR-1 and ECR-2.
Table 6 gives a comparison of the results reported in the literature.
We have previously evaluated an Aliquat 336-based resin material
for the same purpose [25]. However, the present sets of resins are
found to be superior to that in the previous report.

(8)

where Ct and C0 are the effluent and influent concentration, respectively, at time ‘t’ (minutes), KTh is the Thomas rate constant
(mL/mg.min), qads is the sorption capacity of the column (mg/g),
‘m’ is the amount of adsorbent (g) in the column, and Q (flow rate)
is 0.05 mL/minute. The plots of Ct /C0 against time for a given flow
rate were done and the data fitted using Eq. (8) and presented in
Fig. 9. The value of C0 was 0.7 mg/mL and that of m was 0.5 g.
Values for the Thomas constant KTh were calculated to be 0.302 ±
0.005 and 0.030 ± 0.002, respectively for ECR-1 and ECR-2.
Subsequently, the loaded Th was eluted using a mixture of 0.5
M HNO3 and 0.5 M oxalic acid; the elution curves for both resins
are presented in Fig. 10. It is clearly seen from the elution profiles that there is near quantitative elution of the loaded metal
ions in about 7 mL of the eluent. Furthermore, the elution profile
is sharper for the column containing ECR-1 as compared to that

7


R.B. Gujar, P.K. Mohapatra, M. Iqbal et al.

Journal of Chromatography A 1653 (2021) 462401

Table 6
Comparative presentation of Th/U separation from literature reports.
Aqueous medium

Method

Comments

Ref.

Nitric acid

Combination of leaching, solvent extraction,
precipitation and chromatography
UTEVA® resin

99.5% pure Th-230 was obtained. But the method is very
complicated. F− is used which is corrosive.
70% Th recovery is reported. But HCl use is the major drawback
as it is corrosive in nature.
Kd value of Th(IV) >105 ; Kd value of U(VI) ~ 102 . Use of 90%
glacial acetic acid is needed for loading
Apparently simple method. DF ~ 500.

DF values of 178 and 165 are obtained

[40]

Nitric acid / HCl
Nitric acid / acetic
acid
Nitric acid
3 M HNO3

Dowex 1 × 8 (nitrate form)
Aliquat 336-based ECR
ECR-1 and ECR-2

[41]
[42]
[25]
This work

was attempted and DF (Th/U) values of 178 and 165 were obtained
with the ECR-1 and ECR-2 resins, respectively.
Declaration of Competing Interest
The authors have no conflict of interest to declare.
CRediT authorship contribution statement
Rajesh B. Gujar: Formal analysis, Investigation. Prasanta K.
Mohapatra: Conceptualization, Methodology, Writing – original
draft. Mudassir Iqbal: Investigation. Jurriaan Huskens: Methodology, Supervision. Willem Verboom: Methodology.
Acknowledgement
The authors (RBG and PKM) thank Dr. P.K. Pujari, Head, Radiochemistry Division, Bhabha Atomic Research Centre for his constant encouragement.
Supplementary materials

Fig. 11. Separation of U(VI) from Th(IV) as measured from their gamma ray spectra.
(a) Natural uranium solution in the feed; (b) Eluted fraction from the ECR-1 column;
(c) Eluted fraction from the ECR-2 column.

Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.chroma.2021.462401.
References

4. Conclusions

[1] O.B. Mokhodoeva, G.V. Myasoedova, E.A. Zakharchenko, Solid-phase extractants
for radionuclide pre-concentration and separation. New possibilities, Radiochemistry 53 (2011) 35–43.
[2] I. Akaza, Extraction Chromatography, Elsevier, NY, 1975, p. 17.
[3] S. Risticevic, V.H. Niri, D. Vuckovic, J. Pawliszyn, Recent developments in
solid-phase micro-extraction, Anal. Bioanal. Chem. 393 (2001) 781–795.
[4] S.A. Ansari, P.K. Mohapatra, A review on solid phase extraction of actinides and
lanthanides with amide based extractants, J. Chromatogr. A 1499 (2017) 1–20.
[5] J.A. Drader, L. Zhu, P. Smith, K. McCann, S. Boyes, J.C. Braley, Assessment of
monoamide extractants and solid supports as new extraction chromatographic
materials, Sep. Purif. Technol. 163 (2016) 352–356.
[6] J.M. Flanagan, R.K. Fujimura, K.B. Jacobson, Nucleic acid chromatography: substitute solid support material for RPC-5 columns, Anal. Biochem. 153 (1986)
299–304.
[7] E.P. Horwitz, D.R. McAlister, A.H. Bond, R.E. Barrans, Novel extraction of chromatographic resins based on tetraalkyldiglycolamides: Characterization and
potential applications, Solvent Extr. Ion Exch. 23 (2005) 319–344.
[8] P.K. Mohapatra, S. Sriram, V.K. Manchanda, L.P. Badheka, Uptake of
metal ions by extraction chromatography using dimethyl dibutyl tetradecyl-1,3-malonamide (DMDBTDMA) as the stationary phase, Sep. Sci. Technol.
35 (20 0 0) 39–55.
[9] A. Leoncini, J. Huskens, W. Verboom, Ligands for f-element extraction used in
the nuclear fuel cycle, Chem. Soc. Rev. 46 (2017) 7229–7273.
[10] S.A. Ansari, P.N. Pathak, P.K. Mohapatra, V.K. Manchanda, Chemistry of diglycolamides: promising extractants for actinide partitioning, Chem. Rev. 112 (2012)

1751–1772.
[11] K. Van Hecke, G. Modolo, Separation of actinides from low level liquid wastes
(LLLW) by extraction chromatography using novel DMDOHEMA and TODGA
impregnated resins, J. Radioanal. Nucl. Chem. 261 (2004) 269–275.
[12] S.A. Ansari, P.N. Pathak, M. Hosain, A.K. Prasad, V.S. Parmar, V.K. Manchanda,
Extraction chromatographic studies of metal ions using N,N,N’,N’-tetraoctyl
diglycolamide as the stationary phase, Talanta 68 (2006) 1273–1280.
[13] M. Iqbal, P.K. Mohapatra, S.A. Ansari, J. Huskens, W. Verboom, Preorganization
of diglycolamides on the calix[4]arene platform and its effect on the extraction
of Am(III)/Eu(III), Tetrahedron 68 (2012) 7840–7847.

Uptake of Th(IV) from nitric acid was studied by two novel extraction chromatography resins containing two DGA-functionalized
calix[4]arene ligands (ECR-1 and ECR-2) using 234 Th radiotracer for
easy radiometric assay. The Kd values for Th(IV) were found to
be significantly higher than those obtained for U(VI) in the entire
range of acid concentrations studied. In view of the very low ligand loading into the ECRs (ca. 8%), the resins are quite efficient for
Th(IV) ion uptake. These resins are far superior to the known ECRs
reported in the literature. A previously reported Aliquat 336-based
resin displayed much lower (about 10 to 20 times) Kd values at
3 M HNO3 demonstrating that the present set of resins are much
more superior. On the other hand, the uptake of U(VI) was negligible suggesting that the resins can be used for the separation of
234 Th tracer from natural U for possible applications as a radiotracer.
Kinetic modeling of the Th(IV) uptake data pointed to pseudosecond order kinetics, which is the possible mode for chemical interactions. The batch uptake data were fitted to various isotherms
and conformed both to the Langmuir isotherm, suggesting a monolayer adsoption mechanism, and to the D-R isotherm, indicating
chemisorption. The column studies were carried out by loading
Th(IV) in a feed containing 3 M HNO3 and the breakthrough profiles showed much better loading with ECR-1 than with ECR-2.
Sharp elution peaks were obtained while using an eluent mixture
of 0.5 M oxalic acid and 0.5 M nitric acid. The separation of 234 Th
8

R.B. Gujar, P.K. Mohapatra, M. Iqbal et al.

Journal of Chromatography A 1653 (2021) 462401

[14] A. Sengupta, P.K. Mohapatra, M. Iqbal, J. Huskens, W. Verboom, A highly efficient solvent system containing functionalized diglycolamides and an ionic
liquid for americium recovery from radioactive wastes, Dalton Trans. 41 (2012)
6970–6979.
[15] R.B. Gujar, P.K. Verma, P.K. Mohapatra, M. Iqbal, J. Huskens, W. Verboom, Extraction of tetra- and hexavalent actinide ions from nitric acid solutions using
some diglycolamide functionalized calix[4]arenes, Radiochim. Acta 109 (2021)
167–176.
[16] S.A. Ansari, P.K. Mohapatra, M. Iqbal, J. Huskens, W. Verboom, Two novel extraction chromatography resins containing multiple diglycolamide-functionalized ligands: preparation, characterization and actinide uptake properties, J.
Chromatogr. A 1334 (2014) 79–86.
[17] R.K. Sinha, A. Kakodkar, Design and development of the AHWR—the Indian thorium fuelled innovative nuclear reactor, Nucl. Eng. Des. 236 (2006)
683–700.
[18] S.S. Rattan, A.V.R. Reddy, V.S. Mallapurkar, R.J. Singh, S.A. Prakash, A method
for the simultaneous estimation of 228 Th and 229 Th, J. Radioanal. Chem. 67
(1981) 95–99.
[19] P.
Thakur,
R.
Veeraraghavan,
P.K.
Mohapatra,
V.K.
Manchanda,
K.C. Dash, Extraction of ternary complexes of thorium(IV) with
3-phenyl-4-benzoyl-5-isoxazolone and neutral donors from nitric acid
medium, Talanta 43 (1996) 1305–1312.
[20] A. Bhattacharyya, P.K. Mohapatra, V.K. Manchanda, Solvent extraction and extraction chromatographic separation of Am3+ and Eu3+ from nitrate medium

using Cyanex 301, Solvent Extr. Ion Exch. 25 (2007) 27–39.
[21] R.B. Gujar, P.K. Mohapatra, M. Iqbal, J. Huskens, W. Verboom, Highly efficient
uptake of neptunium(IV) from acidic feeds using two solid phase extraction
resins containing diglycolamide-functionalized calix[4]arene ligands, J. Chromatogr. A 1642 (2021) 462037.
[22] Z. Zhu, Y. Sasaki, H. Suzuki, S. Suzuki, T. Kimura, Cumulative study on solvent
extraction of elements by N,N,N’,N’-tetraoctyl-3-oxapentanediamide (TODGA)
from nitric acid into n-dodecane, Anal. Chim. Acta 527 (2004) 163–168.
[23] P.V. Achuthan, A. Ramanujam, Aqueous reprocessing by THOREX process, in:
D. Das, S.R. Bhardwaj (Eds.), Thoria Based Nuclear Fuels: Thermophysical and
Thermodynamic Properties, Fabrication, Reprocessing, and Waste Management,
Springer – Verlag, London, 2013, pp. 279–334.
[24] B.N. Mahanty, A. Bhattacharyya, P.K. Mohapatra, Separation of neptunium (IV)
from actinides by solid phase extraction using a resin containing Aliquat 336,
J. Chromatogr. A 1564 (2018) 94–101.
[25] B.N. Mahanty, P.K. Mohapatra, Highly efficient separation of thorium from uranium in nitric acid feeds by solid phase extraction using Aliquat 336, Sep. Purif. Technol. 237 (2020) 116318.
[26] Z. Yoshida, S.G. Johnson, T. Kimura, J.R. Krsul, Neptunium, in: L.R. Morss,
N.M. Edelstein, J. Fuger, J.J. Katz (Eds.), The Chemistry of the Actinide and
Transactinide Elements, 2, third ed., Springer, Dordrecht, 2006, pp. 771–782.

[27] S. Lagergren, Zur Theorie der sogenannten Adsorption gelöster Stoffe: Kungliga
Svenska Vetenskapsakademiens, Handlingar 24 (1898) 1–39.
[28] Y.S. Ho, Second-order kinetic model for the sorption of cadmium onto tree
fern: a comparison of linear and non-linear methods, Water Res. 40 (2006)
119–125.
[29] C.-V. Gherasim, G. Bourceanu, Removal of chromium(VI) from aqueous solutions using a polyvinyl-chloride inclusion membrane: experimental study and
modeling, Chem. Eng. J. 220 (2013) 24–34.
[30] G.R. Choppin, Technology for nuclear reprocessing: Present and future directions, Sep. Sci. Technol. 41 (2006) 1955–1963.
[31] I. Langmuir, The constitution and fundamental properties of solids and liquids,
J. Am. Chem. Soc. 38 (1916) 2221–2295.
[32] H.M.F. Freundlich, Über die Adsorption in Lösungen, Z. Phys. Chem. 57A (1906)

385–470.
[33] A.H. Gemeay, A.S. El-Sherbiny, A.B. Zaki, Adsorption and kinetic studies of the
intercalation of some organic compounds onto Na+ -montmorillonite, J. Colloid
Interface Sci. 245 (2002) 116–125.
[34] M.M. Dubinin, The potential theory of adsorption of gases and vapors for
adsorbents with energetically non-uniform surfaces, Chem. Rev. 60 (1960)
235–241.
[35] A. El Nemr, Potential of pomegranate husk carbon for Cr(VI) removal from
wastewater: kinetic and isotherm studies, J. Hazard. Mater. 161 (2009)
132–141.
[36] N. Ayawei, A.N. Ebelegi, D. Wankasi, Modelling and interpretation of adsorption isotherms, J. Chem. (2017) 3039817 Article ID11 pages, doi:10.1155/2017/
3039817.
[37] S. Kundu, A.K. Gupta, Investigation on the adsorption efficiency of iron oxide coated cement (IOCC) towards As(V)—kinetics, equilibrium and thermodynamic studies, Colloids Surf. A 273 (2006) 121–128.
[38] H.C. Thomas, Heterogeneous ion exchange in a flowing system, J. Am. Chem.
Soc. 66 (1944) 1664–1666.
[39] R. Han, Y. Wang, X. Zhao, Y. Wang, F. Xie, J. Cheng, M. Tang, Adsorption
of methylene blue by phoenix tree leaf powder in a fixed-bed column: experiments and prediction of breakthrough curves, Desalination 245 (2009)
284–297.
[40] K.N. Kmak, D.A. Shaughnessy, J. Vujic, Separation of thorium from uranium ore,
J. Radioanal. Nucl. Chem. 323 (2020) 931–945.
[41] M. Mola, A. Nieto, A. Peñalver, F. Borrull, C. Aguilar, Uranium and thorium sequential separation from norm samples by using a SIA system, J. Environ. Radioact. 127 (2014) 82–87.
[42] J. Korkisch, G. Arrhenius, Separation of uranium, thorium, and earth elements
by anion exchange, Anal. Chem. 36 (1964) 850–854.

9