Synthesis and extraction studies with a rationally designed diamide ligand selectIVe to actinide(IV) pertinent to the plutonium uranium redox extraction process

Article abstract of DOI:10.1039/c6dt00748a

A new class of conformationally constrained oxa-bridged tricyclo-dicarboxamide (OTDA) ligand was rationally designed for the selective extraction of tetravalent actinides pertinent to the Plutonium Uranium Redox EXtraction (PUREX) process. Two of the designed diamide ligands were synthesized and extraction studies were performed for Pu(iv) from HNO₃ medium. The mechanism of extraction was investigated by studying various parameters such as feed HNO₃, NaNO₃ and OTDA concentrations. The nature of the extracted species was found to be [Pu(NO₃)₄(OTDA)]. One of the OTDA ligands was elaborately tested and showed the selective extraction of Pu(iv) and Np(iv) over other actinide species, viz., U(vi), Np(v), Am(iii), lanthanides and fission products contained in a nuclear waste from the PUREX process. DFT calculations predicted the charge density on each of the coordinating 'O' atoms of OTDA supporting its high Pu(iv) selectivity over other ions studied and also provided the energy optimized structure of OTDA and its Pu(iv) complex.

Full text of DOI:10.1039/c6dt00748a

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: S. Sharma, S.
Panja, A. Bhattacharyya, P. S. Dhami, P. Gandhi and S. K. Ghosh, Dalton Trans., 2016, DOI:
10.1039/C6DT00748A.
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 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
Synthesis and Extraction Studies with a Rationally Designed
Diamide Ligand Selective to Actinide(IV) pertinent to Plutonium
Uranium Redox Extraction Process
Shikha Sharma,a Surajit Panja,b Arunasis Bhattachariya,c Prem S. Dhami,b Preetam M. Gandhib
and Sunil K. Ghosh*a
a Bio-Organic Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400085, India
b Fuel Reprocessing Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400085,
India
c Radiochemistry Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400085, India
E-mail: ghsunil@barc.gov.in
Abstract
A new class of conformationally constrained oxa-bridged tricyclo-dicarboxamide (OTDA) was
rationally designed for selective extraction of tetravalent actinides pertinent to Plutonium
Uranium Redox EXtraction (PUREX) process. Two such designed diamide ligands were
synthesized and extraction studies were performed for Pu(IV) from HNO3 medium. The
mechanism of extraction was investigated by studying various parameters like feed HNO3,
NaNO3 and OTDA concentrations. The nature of extracted species was found to be
[Pu(NO3)4(OTDA)]. One of the OTDA ligand was elaborately tested which showed selective
extraction of Pu(IV) and Np(IV) over other actinide species viz. U(VI), Np(V), Am(III),
1

Dalton Transactions
Page 2 of 33
View Article Online
DOI: 10.1039/C6DT00748A
lanthanides and fission products contained in a nuclear waste from PUREX process. DFT
calculations predicted the charge density on each coordinating ‘O’ atoms of OTDA supporting
its high Pu(IV) selectivity over other ions studied, and also provided the energy optimized
structure of OTDA and its Pu(IV) complex.
Keywords: Ligand design, Conformationally Constrained Dicarboxamides, DFT calculation,
Solvent extraction, Plutonium selectivity, Nuclear waste.
Introduction
Current nuclear reactors use natural or enriched uranium where 235U is the primary fuel. After
power generation, the spent fuel contains mostly U isotopes (~95.5%), fission products (~3.5%)
and about 1% of transuranium elements mainly consist of Pu and minor actinides. Spent fuel
reprocessing technology in ‘Closed Fuel Cycle’ option1 separates major portion of U and Pu
from dissolver solution by PUREX process and generates raffinate solution known as high level
liquid waste (HLLW) comprised of fission products, minor actinides and lanthanides.2
Depending on the performance of the processes and extractants, there is always some loss of Pu
into various streams in nuclear reprocessing.3 239Pu is a strategic material and also known for its
chemical and radio-toxicity. Added to that, it has a long radioactive half life (2.41×104 years).4
Therefore, it is very much desirable to recover Pu from various lean sources drawn from PUREX
process for recycling as well as their safe disposal to the environment.
2

Page 3 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
Solvent extraction process is commonly used to recover important isotopes of elements
from the spent nuclear fuel. The extraction process is very much related to the chemistry of those
elements. A closer look to actinide chemistry reveals that oxidation state of a particular metal ion
has great influence on its extraction by an organic molecule (ligand) often called
solvent/extractant in nuclear extractions. The usual order of extractability of different actinides
+
by organic ligands is MO2 (aq) < M3+(aq) < MO22+(aq) < M4+(aq).5 This behavior is due to the
increase in complexation ability which also follows the same order.
A large group of ligands belonging to various classes have been introduced over the last
few decades for combined isolation of Pu, trivalent minor actinides and lanthanides from HLLW
followed by separation and isolation of Pu.6,7 Malonamide family of extractants were developed
for the extraction of actinides from PUREX raffinate (DIAMEX process) and N,N’-dimethyl-
N,N’-dioctylhexylethoxymalonamide (DMDOHEMA) was found to be the most promising
member of the family.8 Subsequently phosphorus based extractants9 such as phosphine oxides
(TRPO process),9a diisodecylphosphoric acid (DIDPA),9b carbamoyl organophoshorous
compounds (TRUEX process),9c-e etc have also been developed for the same purpose. These
phosphorous based molecules do not comply with the requirements of CHON principle of
extractant design. The multidentate diglycolamides (DGAs)10 such as N,N,N’,N’-
tetraoctyldiglycolamide
(TODGA),10c,d
and
N,N,N’,N’-tetra-2-ethylhexyldiglycolamide
(TEHDGA)10e,f have been found to be the most important extractants in this respect, although
they are known to extract Sr and Ru ions from the HLLW solution.10b,g-i,11 Ligands specific for
direct separation and isolation of Pu are rare. Some remarkably used ones are: (i)
organophosphorous based ligands, (ii) calix-arene based ligands and (iii) macrocyclic N-donor
ligands. Amongst them, tributyl phosphate (TBP) is the most popular and reliably used
3

Dalton Transactions
Page 4 of 33
View Article Online
DOI: 10.1039/C6DT00748A
throughout the world.12 Although TBP is not so selective to Pu and has limitations,13-15 it is the
only ligand used in commercial scale for separation of Pu in the well known “PUREX” process12
for nearly six decades. Some other extractants explored for plutonium are alkylated monoamides,
catecholamides, pyrolidones, sulphoxide, terepthalimides tertiary and quaternary amines etc. but
none of them is specifically selective for plutonium over the other actinides, lanthanides and
fission products contained in a nuclear waste.16-20 It is therefore desirable to design and develop
Pu(IV) selective ligand21 for its extraction from different secondary waste streams pertinent to
PUREX process.
We have recently introduced a new class of conformationally constrained oxa-bridged
bicyclo-dicarboxamides (OBDA) A (Fig. 1) for extraction of actinides and lanthanides.22 The
special feature of these ligands was the conformational rigidity. The amide groups are placed at
β-position with respect to bridging oxygen in the same bicyclic system which ensured proper
orientation of the donor groups in space (two amides and ether oxygen) for complexation. These
special structural features resulted in facilitated complexation. The ligands also displayed good
selectivity for actinides and lanthanides over fission products. To suppress the complexation of
trivalent and hexavalent ions, the electron density on the coordinating oxygens of OBDA was
needed to be suppressed to some extent to increase its selectivity for Pu(IV). This could be
possible by bringing both the amidic ‘O’ atoms close to the bridging oxygen atom which
ultimately pushes back the electron density towards the ligand backbone. We felt that by
structural modifications of this scaffold, the orientation of the amide carbonyls can be modified
in a manner that would decrease the distance between three donor atoms for specific binding
maintaining the rigidity of the molecule. So choice of appropriate scaffold and stereochemistry
of the amide functionalities is desirable. We propose to fuse a five-membered carbocyclic ring
4

Page 5 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
into the OBDA structure and place the carboxamides on the fused ring at the γ position with
respect to bridged oxygen atom like in oxa-bridged tricyclo-dicarboxamide (OTDA) B (Fig. 1).
This proposed modification would make a tricyclic scaffold where the amidic oxygens would
orient closer to the bridging oxygen resulting in lowering in electron density and expected to
show selective extraction of Pu(IV) over other trivalent or low charged metal species. This paper
describes the synthesis of two such newly designed conformationally constrained OTDAs, DFT
calculation for estimating the electron density on the coordinating oxygens and their extraction
behavior for Pu(IV).
O
O
O
R2N
R2N
O
O
O
α
β
NR2
NR2
α
β
γ
α
β
α
γ
β
B
A
Fig. 1 Structures of conformationally constrained OBDA (A) and OTDA (B).
Results and discussion
Synthesis of OTDA ligands
Synthesis of OTDA B skeleton (Fig. 1) seems to be not very straight forward. We, therefore,
modified this designed structure of OTDA in such a way that the modified structure will retain
the required conformationally constrained oxa-bridged tricyclo-dicarboxamide functionalities as
core structure on a multicyclic scaffold that would be easily synthesized from cheap starting
materials. Based on this, a bridged tetracyclic OTDA 1 was designed and two derivatives of
5

Dalton Transactions
Page 6 of 33
View Article Online
DOI: 10.1039/C6DT00748A
OTDAs having difference in amine component of the cyclic imide part viz. 1a,b (Fig. 2) have
been synthesized from the commercially available starting materials as shown in Scheme 1.
O
O
R2N
O
O
R2N
R2N
α
O
O
β
O
R2N
α
γ
β
γ
H
O
α
γ
N
β
α
γ
R'
H
β
B
R = 2-Ethylhexyl, R' = n-Octyl; 1a
R = 2-Ethylhexyl, R' = i-Butyl; 1b
Fig. 2 Designed OTDA structures
The exo adduct 2 from the Diels-Alder (D-A) reaction of furan and maleic anhydride
acted as a dienophile in the D-A reaction with cyclopentadiene to give the endo adduct 3. This
anhydride 3 was condensed with one molecule of n-octylamine or isobutylamine in presence of
triethylamine to obtain the imides 4a and 4b, respectively, in very good yields. The double bond
in imides 4a,b was oxidatively cleaved under Sharpless conditions23 using RuO2.xH2O as
catalyst and NaIO4 as an oxidant to give diacids 5a,b which were further reacted with bis(2-
ethylhexyl)amine in the presence of diisopropylcarbodiimide (DIPC) and 4-
dimethylaminopyridine (DMAP) to obtain the final OTDA 1a,b (Scheme 1).
6

Page 7 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
O
O
RNH2, Et3N
O
O
Xylene, 120 o
C
H
H
O
O
CHCl3
40 oC, 41%
80-86%
H
H
O
O
3
2
O
HO2C
HO2C
O
NaIO4, RuO2
xH2O
.
O
O
H
N
CCl4/ H2O/
CH3CN, 28 o
H
N
R
H
C
R'
H
O
O
R' = n-Octyl; 4a
R' = i-Butyl; 4b
R' = n-Octyl; 5a
R' = i-Butyl; 5b
O
Bis(2-ethylhexyl)amine,
DIPC, DMAP, CH2Cl2
R2N
R2N
O
O
O
0 oC to 28 o
40-41%
C
H
O
N
R
H
R = 2-Ethylhexyl, R' = n-Octyl; 1a
R = 2-Ethylhexyl, R' = i-Butyl; 1b
Scheme 1 Synthesis of OTDAs.
Extraction studies with OTDAs
Optimization for time of contact
n-Dodecane is a preferred diluent in solvent extraction process in nuclear fuel cycle. The
solubility of OTDA 1a and 1b were not good in n-dodecane alone. For extraction studies with
such extractants, isodecyl alcohol (IDA) is commonly used to enhance the solubility as well as a
phase modifier to suppress the formation of third phase.24,25 Therefore, IDA was chosen as a co-
solvent to enhance the solubility and 30% IDA in n-dodecane was found to be suitable for
making a clear 0.1 M solution of the OTDA 1a and 1b. The time required to attain the
equilibrium for extraction of Pu(IV) was determined for 0.1 M OTDA 1a in 30% IDA/ n-
dodecane at 4 M feed HNO3 concentration. For this, equal volume of OTDA solution and Pu(IV)
solution (Pu concentration: ~5 mg/L) were shaken for different time intervals, centrifuged and
7

Dalton Transactions
Page 8 of 33
View Article Online
DOI: 10.1039/C6DT00748A
the aqueous and organic layers were separated. The results are presented in Fig. 3 which also
shows that the equilibrium was attained within 5 min. A very similar equilibration time was
observed for 30% TBP in n-dodecane alone. The results indicate that Pu(IV) complexation by
OTDA 1a is thermodynamically favored because of the conformational rigidity of OTDA 1a
molecule that ensure the co-ordinating “O” atoms are appropriately oriented with fittingly
distanced as predicted by DFT calculations (vide infra). Therefore, in subsequent studies, 10 min
equilibration time was maintained.
18
15
12
D
9
6
3
0
4
8
12
16
Time (min)
Fig. 3 Kinetics of Pu(IV) extraction for OTDA 1a.
Choice of OTDA ligand
For comparison purpose, the extraction of Pu(IV) was studied with the OTDA 1a and 1b under
the same conditions using a freshly prepared 0.1 M 1a/b in 30% IDA/ n-dodecane at a 4 M feed
HNO3 concentration. The DPu for OTDA 1a and OTDA 1b were found to be very close, 15.5 and
17.5, respectively. The solution of OTDA 1a in 30% IDA/ n-dodecane remained homogeneous
8

Page 9 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
on standing for days. On the other hand, the solution of 0.1 M 1b in 30% IDA/ n-dodecane on
standing overnight showed gelatinous precipitate of the extractant but the chemical integrity of
OTDA 1b was intact. As OTDA 1a showed better solution properties in 30% IDA/n-dodecane
diluent system, further extraction studies were carried out with OTDA 1a only.26
Effect of nitric acid concentration
The efficiency of Pu(IV) extraction by OTDA 1a over a wide range of HNO3 concentration is
one of the key requirement for solvent extraction from various lean resources. The effect of feed
nitric acid concentration on the distribution ratio (DPu) of Pu(IV) for 0.1 M OTDA 1a in 30%
IDA/ n-dodecane was therefore investigated and is shown in Fig. 4. As can be seen, OTDA 1a
showed increasing trend in the DPu with increasing feed HNO3 concentration (0.5-6.0 M). A high
DPu value of ~65 was obtained with 6 M HNO3, while no third phase formation was observed
during the extraction process. The high extraction efficiency toward Pu(IV) indicates less
protonation at the ligand donation sites (vide infra). Based on these data and OTDA 1a being a
neutral extractant, the extraction mechanism can be explained via nitrate ion assisted complex
formation [eqn. (1)].
Pu4+ + x NO3 + y (OTDA 1a)org ꢀ [Pu(NO3)x(OTDA 1a)y]org (1)
−
The subscript ‘org’ refers to species in organic phase and without subscript refer to those
present in aqueous phase. So increasing nitric acid concentration leads to increased formation of
neutral metal-nitrate complex leading to higher DPu value.
9

Dalton Transactions
Page 10 of 33
View Article Online
DOI: 10.1039/C6DT00748A
60
40
20
0
D
0
1
2
3
4
5
6
HNO (M)
3
Fig. 4 DPu at varying feed HNO3 concentration.
Effect of nitrate ion concentration
In order to determine the role of nitrate ion in the formation of extractable metal species, the DPu
values were determined at different nitrate ion concentrations (1-6 M) at a fixed feed acidity (0.5
M HNO3) for 0.1 M 1a in 30% IDA/n-dodecane. The respective nitrate ion concentrations were
achieved by adding required amount of solid sodium nitrate into the solution. The DPu clearly
-
increases with increasing concentration of nitrate ion and the plot of log DPu vs log [NO3 ]
showed a straight line with a slope of 4.21 (Fig. 5) indicating four nitrate ions are involved in the
extraction of Pu(IV). These results indicate that nitrate ion played an important role in extraction
equilibrium by providing counter ions to balance the charge on Pu(IV).
10

Page 11 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
-
Fig. 5 Plot of log DPu vs log [NO3 ].
Effect of OTDA 1a concentration and stoichiometry of metal ligand complex
The extent of Pu(IV) extraction with different OTDA 1a concentration needs to be assessed to
find out the stoichiometry of metal ligand complex. For this, DPu was determined at varying
concentration of OTDA 1a in 30% IDA/n-dodecane at two fixed feed aqueous acidities (2 and 4
M HNO3) (Fig. 6). The DPu value increased with increasing ligand concentration in organic phase
and the plot of log DPu vs log [OTDA 1a]free gave a straight line with slope value ~1.08 indicating
that Pu(IV) forms a mono-ligated species with ligand OTDA 1a at both the acidities studied.
Although OTDA 1a is a neutral ligand, the amide carbonyls can coordinate to a proton from
nitric acid and may not be available for complexation to Pu(IV). So the actual free ligand
concentration will be slightly less than the molar concentration made just on the basis of solute
amount. The free concentration of ligand was obtained by NaOH titrations of the organic phases
comprised of pure diluents i.e. 30% IDA/n-dodecane and OTDA 1a solutions in 30% IDA/n-
11

Dalton Transactions
Page 12 of 33
View Article Online
DOI: 10.1039/C6DT00748A
dodecane after equilibration with 2 M and 4 M HNO3 solutions. Hence the equation for Pu(IV)
complexation with OTDA 1a can be written as:
−
Pu4+ + 4NO3 + (OTDA 1a)org ꢀ [Pu(NO3)4(OTDA 1a)]org (2)
As plutonium selective ligands are rare, a comparison of DPu has been made from the
literature data of two ‘O’-coordinting ligands viz. TODGA27 and Benzo-DODA.21 The ligands
were chosen on the basis of their use in solvent extraction process of HLLW originated from
PUREX process. The DPu values from 4 M HNO3 with 0.1 M ligand concentration in 30%
IDA/n-dodecane with OTDA 1a, TODGA and benzo-DODA were found to be closer to ~20,
~250 and ~7, respectively.
Slope- 1.07 0.044 (2M HNO )
3
Slope- 1.06 0.025 (4M HNO )
3
10
1
0.1
0.01
0.1
[OTDA 1a]
M
free,
Fig. 6 Plot of log DPu vs log [OTDA 1a].
12

Page 13 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
Selectivity studies
To assess our design hypothesis that the designed OTDA ligands could be potentially useful for
selective extraction of tetravalent actinides in general and Pu(IV) in particular, we have carried
out few experiments under various conditions. Therefore, the extraction of Pu(IV) by OTDA 1a
was studied from a plutonium bearing solution such as HLLW containing other metal ions of the
nuclear waste pertaining to PUREX process. OTDA 1a showed selective extraction of Pu over
other actinides and lanthanides viz Am, Eu, Ce, Np(V), U and fission products viz Zr, Cs, Sr, Ru
+
(Table 1). Neptunium, which exists in variable valency state with NpO2 being the most stable
one under normal acidic conditions, negligible extraction of the element was observed. But,
when the valency of Np was adjusted to Np(IV), OTDA 1a showed similar extraction behavior
with that of Pu(IV) and gave very similar distribution ratios. The extractant showed small D
value for U (for D values at other acidities, see the Supporting Information) which can be
removed from the loaded organic phase by scrubbing with suitable scrubbing acid. The
separation factors for Pu(IV) with respect to different elements studied at different acidities are
presented in Table 2. Very high values of separation factors for Pu(IV) with respect to all the
other elements [except Np(IV)] present in the nuclear waste at a wide range of acidities clearly
showed the selective nature of OTDA 1a for Pu(IV). The above results also proved in principle
that the design concept we adopted is correct.
13

Dalton Transactions
Page 14 of 33
View Article Online
DOI: 10.1039/C6DT00748A
Table 1 Selectivity of 0.1 M OTDA 1a in 30% IDA/ n-dodecane for Pu(IV) in presence of other
metal ions present in HLLW at different acidities.
Distribution ratio (DM)
+
[HNO3],M
Pu(IV) U(VI) Am(III) Eu(III) NpO2
Np(IV) Ce(III) Ru
Cs(I)
Sr(II)
Zr(IV)
2
4
6
0.95 0.25 <0.01 <0.01 <0.01 0.71
<0.01 <0.01 <0.01 <0.01 <0.1
15.39 0.42 <0.01 <0.01 <0.01 13.32 <0.01 <0.01 <0.01 <0.01 <0.1
64.89 0.62 <0.01 <0.01 <0.01 53.22 <0.01 <0.01 <0.01 <0.01 <0.1
Table 2 Separation factors for different elements with respect to Pu(IV) for 0.1M OTDA 1a in
30% IDA/n-dodecane at different acidities
Separation factor (DPu/DM)
+
U(VI)
3.8
Am(III)
>95
Eu(III) NpO2
Np(IV)
1.34
Ce(III) Ru
Cs(I)
Sr(II)
>95
Zr(IV)
>95
[HNO3],M
2
4
6
>95
>95
>95
>95
>95
36.64
104.66
>103
>104
>103 >103
>104 >104
1.16
>103 >103 >103
>104 >104 >104
>103
>104
>103
>104
1.22
Stripping studies
Low concentration of HNO3 or its combination with oxidation state adjustment by a redox active
reagent is widely used as stripping agents for Pu from the extracted organic phase. Therefore,
14

Page 15 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
studies were also carried out to understand the stripping behavior of Pu(IV) loaded into 0.1 M
OTDA 1a in 30% IDA/n-dodecane with two different strippants. A reductive strippant mixture
of 0.3 M hydroxyl ammonium nitrate (HAN) and 0.2 M HNO3 was found to be very effective
(Table 3). Three successive contacts of HAN mixture resulted in quantitative stripping of Pu
from OTDA 1a where as only ~86 % of the loaded Pu(IV) could be stripped by four successive
contacts using 0.5 M HNO3 alone. The better stripping ability of HAN-HNO3 combination is due
to conversion of Pu(IV) to Pu(III) by the strippant and very poor complexation of the latter by
OTDA 1a.
Table 3 Percentage stripping of Pu from 0.1 M OTDA 1a in 30% IDA in successive contacts
% Cumulative stripping
No of contacts
0.3 M HAN + 0.3 M HNO3
0.5 M HNO3
40.2
1
2
3
4
80.2
95.9
99.9
-
68.9
81.1
85.6
Stability of the ligand and functionalities involved in extraction of Pu
To ascertain the stability under acidic conditions, 0.1 M OTDA 1a in 30% IDA/ n-dodecane was
stirred with equal volume of 4 M HNO3 over a week prior to Pu(IV) extraction. Almost similar
DPu value was found (14.9 vs 15.5) with the pretreated extractant indicating the good stability of
OTDA 1a under the extraction conditions. The chemical stability of OTDA 1a was further
15

Dalton Transactions
Page 16 of 33
View Article Online
DOI: 10.1039/C6DT00748A
monitored by thin layer chromatography and no degradation was observed even after 15 days of
contact with 4 M HNO3.
To ascertain if any role played by the cyclic imide group of OTDA 1a for the extraction
of Pu(IV), the imides 6a-c devoid of additional amide groups were synthesized as presented in
Scheme 2. For this, the known anhydride 722 was reacted with different primary amines to give
the imides 6a-c.28 When the extraction experiments were performed with 0.1 M solution of these
imides in 30% IDA/n-dodecane, no extraction of Pu(IV) and other metal ions present in nuclear
waste was observed from a feed acidity of 1-4 M HNO3. Therefore, it can be inferred that the
cyclic imide group in OTDA 1a also does not participate in the extraction of metal ions from
nuclear waste solution and the selectivity in extraction of Pu(IV) is solely due to oxa-diamide
functionalities present in the ligand.
O
O
O
O
RNH2,Et3N,
Xylene
O
O
H2, Pd/C
EtOAc
H
H
H
28 o
C
O
O
120 o
82-90%
C
N
H
R
H
H
98%
O
O
O
2
7
R = i-Butyl; 6a
R = 2-Ethylhexyl; 6b
R = n-Dodecyl; 6c
Scheme 2 Synthesis of imides 6a-c.
Computational Study
Natural charges on six different types of oxadiamides OTDA I-VI were calculated using natural
population analysis (Fig. 7). The result indicated a decrease in electron density on the oxygen
16

Page 17 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
atoms when both the amidic ‘O’ atoms are forcefully fixed close to the bridging oxygen atom by
fixing its stereochemistry. This is probably due to the presence of all the three electron rich 'O'
atoms in close proximity which ultimately pushes back the electron density towards the ligand
backbone. This reduction in electronic charge density on the donor ‘O’ atoms will certainly
reduce the extraction capability in case of the oxadiamide ligands having the entire three donor
‘O’ atoms in close proximity (OTDA III and OTDA VI) as compared to the ligand (OBDA I)
reported in our previous work.22 However, there is a chance of enhancement in the selectivity for
the tetravalent actinides [viz. Pu(IV) over the trivalent one viz. Am(III)]. The interatomic
distances between the donor ‘O’ atoms of the OTDA VI is shown in Fig. 8. For comparison
purpose, the interatomic distances between the donor ‘O’ atoms of OBDA I, OBDA II and
OBDA V are also shown in Fig. 9 which show that OTDA VI is having close proximity of
coordinating ‘O’ atoms.
-0.625
-0.586
-0.593
-0.513
-0.552
-0.620
-0.580
-0.635
-0.635
OBDA I
OBDA II
OTDA III
-0.560
-0.539
-0.588
-0.514
-0.561
-0.570
-0.549
-0.608
-0.592
-0.580
-0.548
-0.557
-0.557
OBDA IV
OBDA V
OTDA VI
Fig. 7 Structures and natural charges on the donor ‘O’ atoms for four different OBDAs (I, II, IV
and V) and two different OTDAs (III and VI) calculated using natural population analysis.
17

Dalton Transactions
Page 18 of 33
View Article Online
DOI: 10.1039/C6DT00748A
OTDA VI
Fig. 8 Structure of OTDA VI showing the interatomic distances between the three donor ‘O’
atoms.
OBDA II
OBDA I
OBDA V
Fig. 9 Structures of OBDAs (I, II and V) showing the interatomic distances between the three
donor ‘O’ atoms.
18

Page 19 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
The geometry of the Pu(IV) complex was optimized considering the metal ligand
stoichiometry of 1:1 as observed from the solvent extraction technique. The optimized
geometries of the Pu(IV) complexes with two oxa-diamides OBDA V (complex ‘a’) and OTDA
VI (complex ‘b’) are shown in Fig. 10. In complex “a”, the OBDA V possesses flexible
stereochemistry of the donor ‘O’ atoms while complex ‘b’ comprised of the donor oxygen atoms
stereochemically fixed in one direction from OTDA VI. It was observed that the ligand could
approach closer to the Pu(IV) ion in case of complex ‘b’ as compared to that in case of complex
‘a’. In order to further understand the metal-ligand bonding the natural charges on the Pu(IV) ion
and coordinating sites of the ligand molecules were calculated and the results are shown in Table
4. In case of complex ‘a’, the electron density on the amidic and ethereal oxygen atom decreased
from -0.60 e to -0.583 e and -0.570 e to -0.538 e after complexation. On the other hand, these
electron densities were found to be increased upon complexation in case of complex ‘b’. As the
charge densities on ‘O’ atoms in OTDA III and VI are almost same, this can also be concluded
from the theoretical study that the scaffold imide functionality does not affect the charge density
on coordinating ‘O’ atoms of the OTDA VI ligand thus also does not change the extraction
properties of the ligand.
19

Dalton Transactions
Page 20 of 33
View Article Online
DOI: 10.1039/C6DT00748A
Fig. 10 Optimized geometries of Pu(IV) complexes of OBDA V (a) and OTDA VI (b).
Table 4 Natural charges on the carbonyl and ethereal oxygen and carbonyl carbon atoms before
and after complexation and on Pu(IV) ion in the complexes ‘a’ and ‘b’.
Q(Pu)
Q(O)amide
Q(C)amide
Q(O)ether
Complex
In complex In free ligand
In complex
-0.583 0.007
-0.611 0.003
In free ligand
In complex
0.72 0.01
In free ligand
In complex
-0.538
1.099
1.129
-0.60 0.01
-0.55 0.05
0.69 0.01
-0.570
-0.549
a
0.659 0.006
0.736 0.005
-0.565
b
Conclusions
A new class of conformationally constrained OTDAs were rationally designed and synthesized.
The molecules showed good selectivity for Pu(IV) over other metal ions contained in a nuclear
waste solution. DFT calculations supported the lowering of electron density on coordinating ‘O’
atoms of OTDAs resulting in their selectivity for Pu(IV) over other actinides, lanthanides and
20

Page 21 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
fission products. Stochiometry of extracted complex was found to be 1:1 with respect to
plutonium and ligand. The new extractant showed excellent extraction behavior for Pu with rapid
extraction, high selectivity, easy stripping of the extracted Pu and good chemical stability under
extraction conditions. This is a remarkable example of designing of C,H,O,N based Pu selective
ligand which is rare in literature. The work provided the basic data for the feasibility of this
designed novel family of molecules for selective separation of plutonium from HLLW as well as
various Pu bearing solutions generated during nuclear fuel reprocessing.
Experimental
Preparation of OTDA ligands
General details
All reactions were performed in oven-dried (120 °C) or flame-dried glass apparatus under dry N2
or argon atmosphere. Solvent removal was performed with a rotary evaporator that was
connected to a dry ice condenser. Triethylamine, octylamine, isobutylamine and bis(2-
ethylhexyl)amine were obtained from Aldrich and distilled over CaH2 prior to use. Chloroform
was dried over P2O5. Diisopropylcarbodiimide was freshly distilled before use. Other chemicals
were used as obtained from Aldrich. Anhydride 2 was prepared following the procedure
described previously.22 TLC was carried out using Merck silica gel 60 F254 plates. Column
chromatography was performed on silica gel (230-400 mesh). 1H NMR spectra were recorded on
13
200, 300, 500 MHz and C NMR spectra were recorded with 50, 75, 125 MHz spectrometers
using CDCl3 as the solvent. The spectra were referenced to residual chloroform (δ 7.25 ppm, 1H;
21

Dalton Transactions
Page 22 of 33
View Article Online
DOI: 10.1039/C6DT00748A
13
77.00 ppm, C). IR spectra were carried out on the FT-IR-spectrometer using KBr pellets, and
wave numbers were given in cm-1. Melting points (mp) were measured in a Büchi B-540
apparatus. Elemental analyses (C, H, N) were carried out by Elementar, varioMICRO CHNS
instrument.
Synthesis of Diels-Alder adduct 3
Freshly distilled cyclopentadiene (37.8 mL, 450 mmol) was added to a suspension of exo-3,6-
epoxy-1,2,3,6-tetrahydrophthalic anhydride 2 (4.98 g, 30 mmol) in dry CHCl3 and the mixture
was heated at 40 °C for 4 d. The resulting reaction mixture was allowed to stand overnight. The
white crystals of the product was filtered and washed with 2% ethyl acetate-petroleum ether
mixture to give 3 (2.85 g, 41%).
M.P. 183 °C. 1H NMR (300 MHz, CDCl3): δ 1.20 (1 H, d, J = 8.1 Hz, CHAHB), 1.40 (1 H, d, J =
8.1 Hz, CHAHB), 2.40 (2 H, s, 2 × CH), 2.97 (2 H, s, 2 × CH), 3.08 (2 H, s, 2 × CHCO), 4.64 (2
13
H, s, CHOCH), 5.97 (2 H, s, CH=CH); C NMR (50 MHz, CDCl3): δ 44.9 (2 C), 48.3 (2 C),
51.3, 51.9 (2 C), 80.7 (2 C), 133.4 (2 C), 171.0 (2 × CO); IR (KBr): 3074, 2978, 2871, 1859,
1835 (C=O), 1770 (C=O), 1252, 1234, cm-1; Anal. Calcd. for C13H12O4; C, 67.23; H, 5.21;
Found C, 67.17; H, 5.123.
Synthesis of imide 4a
Triethylamine (1.24 mL, 8.9 mmol) and octylamine (1.24 mL, 7.5 mmol) was added to a stirred
suspension of anhydride 3 (1.72 g, 7.4 mmol) in xylene (7 mL). The reaction mixture was stirred
for 0.5 h at room temperature, diluted with xylene (25 mL) and heated at 120 °C for 4 h. The
solvent and excess amine was evaporated under reduced pressure and the residue was purified by
22

Page 23 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
column chromatography using 15% ethyl acetate-petroleum ether mixture to obtain product 4a
(2.03 g, 80%).
M.P. 70 °C. 1H NMR (300 MHz, CDCl3): δ 0.85 (3 H, t, J = 7.2 Hz, CH3), 1.22 (11 H, bs, 5 ×
CH2, CHAHB), 1.36 (1 H, d, J = 8.4 Hz, CHAHB), 1.40-1.50 (2 H, m, 2 × CH), 2.37 (2 H, s, 2 ×
CH), 2.75 (2 H, s, 2 × CH), 2.91 (2 H, s, 2 × CHCO), 3.35 (2 H, t, J = 7.5 Hz, NCH2), 4.46 (2 H,
s, CHOCH), 5.95 (2 H, s, CH=CH); 13C NMR (75 MHz, CDCl3): δ 14.0, 22.6, 26.7, 27.6, 29.0 (2
C), 31.7, 39.0, 44.9 (2 C), 48.7 (2 C), 51.2 (2 C), 51.5, 79.5 (2 C), 133.4 (2 C), 176.9 (2 × CO);
IR (KBr): 2962, 2930, 2857, 1690 (C=O), 1407 cm-1; Anal. Calcd. for C21H29NO3; C, 73.44; H,
8.51; N, 4.08%; Found C, 73.58; H, 8.236; N, 4.31%.
Synthesis of imide 4b
Following the procedure described for the synthesis of 4a, triethylamine (1.3 mL, 9.3 mmol),
isobutylamine (0.93 mL, 9.3 mmol) and anhydride 3 (1.78 g, 7.7 mmol) gave imide 4b (1.9 g,
86%).
M.P. 140 °C. 1H NMR (300 MHz, CDCl3): δ 0.82 (6 H, d, J = 6.6 Hz, 2 × CH3), 1.2 (1 H, d, J =
7.8 Hz, CHAHB), 1.37 (1 H, d, J = 7.8 Hz, CHAHB), 1.85-2.03 (1 H, m, CH), 2.38 (2 H, s, 2 ×
CH), 2.77 (2 H, s, 2 × CH), 2.92 (2 H, s, 2 × CHCO), 3.22 (2 H, d, J = 7.5 Hz, NCH2), 4.47 (2 H,
s, CHOCH), 5.96 (2 H, s, CH=CH); 13C NMR (75 MHz, CDCl3): δ 20.0 (2 C), 26.9, 44.9 (2 C),
46.2, 48.8 (2 C), 51.2 (2 C), 51.6, 79.5 (2 C), 133.4 (2 C), 177.2 (2 × CO); IR (KBr): 2966, 2930,
2864, 1696 (C=O), 1405 cm-1; Anal. Calcd. for C17H21NO3; C, 71.06; H, 7.37; N, 4.87%; Found
C, 71.11; H, 7.047; N, 4.67%.
23

Dalton Transactions
Page 24 of 33
View Article Online
DOI: 10.1039/C6DT00748A
Synthesis of OTDA 1a from imide 4a
Sodium periodate (335 mg, 1.57 mmol) was added to a stirred solution of imide 4a (120 mg,
0.35 mmol) in a 2:2:3 mixture of CCl4, CH3CN and H2O (5.25 mL). After 5 min of stirring,
RuO4.xH2O (1.8 mg, 1.5 wt%) was added to the reaction mixture and stirred vigorously for 3 h.
The reaction mixture was concentrated, triturated with ethyl acetate and filtered through celite.
The filtrate was evaporated under reduced pressure to give the acid 5a which was dissolved in
dry CH2Cl2 (2 mL). Bis(2-ethylhexyl)amine (210 µL, 0.7 mmol) and DMAP (17 mg, 0.14 mmol)
was added to the solution of the acid 5a at room temperature. The reaction mixture was cooled
on an ice-water bath and diisopropylcarbodiimide (110 ꢀL, 0.7 mmol) was added drop-wise to
the stirred reaction mixture. The reaction mixture was allowed to attain room temperature and
stirred overnight. The reaction mixture was diluted with 1/1 ethyl acetate-petroleum ether
mixture and washed with 5% citric acid solution, with water and concentrated under reduced
pressure. The residue was purified with column chromatography to give the diamide 1a (123 mg,
41%).
1
M.P. 183-196 °C (waxy solid). H NMR (300 MHz, CDCl3): δ 0.80-1.05 (27 H, m, 9 × CH3);
1.05-1.51 (44 H, m, 22 × CH2), 1.52-1.68 (4 H, m), 1.68-1.88 (2 H, m) 2.49 (2 H, bs, 2 × CH),
2.73 (2 H, s, 2 × CHCO), 2.80-3.40 (10 H, m, 5 × NCH2), 3.55-3.68 (1 H, m, CHCO), 3.69-3.82
(1 H, m, CHCO), 4.82 (2 H, s, CHOCH); IR (KBr): 2960, 2930, 2874, 2860, 1704 (C=O), 1630
(C=O), 1461, 1377 cm-1; Anal. Calcd. for C53H95N3O5; C, 74.51; H, 11.21; N, 4.92%; Found C,
74.46; H, 11.02; N, 4.87%.
24

Page 25 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
Synthesis of OTDA 1b from imide 4b
Following the procedure described for the synthesis of 1a, sodium periodate (335 mg, 1.57
mmol), imide 4b (101 mg, 0.35 mmol), RuO4.xH2O (1.5 mg, 1.5 wt%) gave acid 5a which on
further reaction with bis(2-ethylhexyl)amine (210 µL, 0.7 mmol), DMAP (17 mg, 0.14 mmol)
and diisopropylcarbodiimide (110 ꢀL, 0.7 mmol) gave diamide 1b (112 mg, 40%).
M.P. 184-197°C (waxy solid). 1H NMR (300 MHz, CDCl3): δ 0.82 (6 H, d, J = 6.6 Hz, 2 × CH3),
0.83-1.05 (24 H, m, 8 × CH3), 1.05-1.51 (34 H, m, 17 × CH2), 1.52-1.85 (4 H, m, 4 × CH), 1.85-
2.0 (1 H, m, CH), 2.49 (2 H, s, 2 × CHCO), 2.73 (2 H, s, 2 × CHCO), 2.80-3.30 (10 H, m, 5 ×
NCH2), 3.51-3.65 (1 H, m, CHCO), 3.65-3.8 (1 H, m, CHCO), 4.84 (2 H, s, CHOCH); IR (KBr):
2960, 2929, 2874, 2859, 1705 (C=O), 1630 (C=O), 1451 cm-1; Anal. Calcd. for C49H87N3O5; C,
73.73; H, 10.99; N, 5.26%; Found: C, 73.49; H, 10.74; N, 5.13%.
Synthesis of imides 6a-c
Imides 6a and 6b/6c were synthesized from the known anhydride 718 following the procedure
described for the synthesis of 4b and 4a, respectively.
Imide 6a
1
Yield: 1.56 g (90%). H NMR (200 MHz, CDCl3): δ 0.83 (6 H, d, J = 6.8 Hz, 2 × CH3), 1.51-
1.63 (2 H, m, 2 × CHAHB), 1.75-1.88 (2 H, m, 2 × CHAHB), 1.88-2.10 (1 H, m, CH), 2.84 (2 H, s,
2 × CHCO), 3.25 (2 H, d, J = 7.4 Hz, CH2N), 4.82-4.85 (2 H, m, CHOCH); 13C NMR (50 MHz,
CDCl3): δ 19.8 (2 C), 26.8, 28.5 (2 C), 46.1, 49.7 (2 C), 79.0 (2 C), 177.4 (2 × CO); Anal. Calcd.
for C12H17NO3; C, 64.55; H, 7.67; N, 6.27%; Found C, 64.61; H, 7.8; N, 6.20%.
25

Dalton Transactions
Page 26 of 33
View Article Online
DOI: 10.1039/C6DT00748A
Imide 6b
Yield: 1.7 g (82%). 1H NMR (200 MHz, CDCl3): δ 0.77-0.95 (6 H, m, 2 × CH3), 1.10-1.35 (8 H,
m, 4 × CH2), 1.53-1.93 (5 H, m), 2.85 (2 H, s, 2 × CHCO), 3.35 (2 H, d, J = 7.4 Hz, CH2N),
4.84-4.87 (2 H, m, CHOCH); 13C NMR (50 MHz, CDCl3): δ 10.2, 13.9, 22.8, 23.4, 28.1, 28.5 (2
C), 30.0, 37.0, 42.7, 49.7 (2 C), 79.0 (2 C), 177.4 (2 × CO); Anal. Calcd. for C16H25NO3; C,
68.79; H, 9.02; N, 5.01%; Found C, 68.74; H, 9.45; N, 5.3%.
Imide 6c
Yield: 2.16 g (87%). 1H NMR (500 MHz, CDCl3): δ 0.86 (3 H, t, J = 7 Hz, CH3), 1.2-1.3 (18 H,
s, 9 × CH2), 1.47-1.53 (2 H, m, 2 × CHAHB), 1.53-1.6 (2 H, m, 2 × CHAHB), 1.80-1.87 (2 H, m,
CH2), 2.83 (2 H, s, 2 × CHCO), 3.43 (2 H, t, J = 7 Hz, CH2N), 4.85 (2 H, s, CHOCH); 13C NMR
(50 MHz, CDCl3): δ 14.0, 22.6, 26.6, 27.5, 28.5 (2 C), 29.0, 29.2, 29.3, 29.4, 29.5 (2 C), 31.8,
39.0, 49.8 (2 C), 78.9 (2 C), 177.1 (2 × CO); Anal. Calcd. for C20H33NO3; C, 71.6; H, 9.91; N,
4.18; Found C, 71.5; H, 10.09; N, 4.18%.
Solvent extraction experiments
Extraction studies
All aqueous solutions were prepared using Milli-Q deionized water. Anhydrous n-dodecane
(99+%) and IDA (99+%) were used for the preparation of the organic phases. The nitric acid
concentration was determined by titration with standard NaOH.
Solutions of desired concentration of OTDA 1a,b were prepared in 30% IDA/n-
dodecane. Solvent extraction studies were carried out by equilibrating equal volume (2 mL each)
26

Page 27 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
of the organic and aqueous phases by agitation in a rotary thermostated water bath for 10 min at
o
25.0 0.1 C followed by centrifugation for phase separation. The phases were separated and
dispensed into glass vials from which 100 to 500 µL of each phase was taken for radiometric
assay. Each experiment was done in duplicate and results agree within 5%.
A high purity germanium detector coupled to a 4096 channel analyzer was used to assay
the activity of the radio nuclides during HLW experiment. Energies used for the various radio
nuclides were as follows: 106Ru – 621 keV, 137Cs – 662 keV, 144Ce – 133 keV, 241Am – 59 keV,
152Eu – 121.8 keV and 85+89Sr – 514 keV. Np analysis was performed by alpha spectrometry. The
results were reported as distribution ratio (DM) and are calculated as radioactivity of the
corresponding radionuclide in the organic phase divided by that in the aqueous phase. The
separation factor is defined as the ratio of the distribution ratios of the corresponding
radionuclides.
The valency of Pu during extraction was maintained as Pu(IV). For preparation of Pu(IV)
stock, NaNO2 was used as oxidant in 1 M HNO3 solution. The oxidized Pu(IV) solution was
subsequently extracted by 0.5 M 2-thenoyltrifluoroacetone (TTA) in xylene where Pu(IV) is
quantitatively extracted. The loaded Pu was stripped by 7 M HNO3 and equilibrated 3 times with
xylene to remove dissolved TTA from the aqueous phase. The stripped Pu solution was used as
stock (concentration of ~ 500 mg/L which was diluted during the experiments) and stabilized as
Pu(IV) using NaNO2. Np(V) aqueous stock was prepared from solid NpO2 using standard
methods, which have been described previously.29 Typically, known quantities of the solid NpO2
were dissolved by refluxing in concentrated HCl in a radiochemical glove box followed by
evaporation and re-dissolution in ∼1 M HNO3. The pink solution of Np(VI) was evaporated to
27

Dalton Transactions
Page 28 of 33
View Article Online
DOI: 10.1039/C6DT00748A
dryness and redissolved to give a greenish Np(V) solution in nitric acid medium. By this way a
working stocks of Np(V) (2 × 10−4 M) was prepared. This would be sampled for acidity and Np
analyses and checked spectrophotometrically for its valence purity. A stock solution of natural
uranium was prepared by dissolving UO2 in nitric acid (1.2 M U(VI) at 4 M HNO3) and used
after appropriate dilutions. Solutions of pure Np(IV) were obtained by reducing aliquots of the
Np stock solution with large excess FeCl2, diluted with HNO3 such that the overall acidity was
usually between 2-3 M HNO3. The resultant Np(IV) solution was extracted by 30% TBP in
kerosene (3 times) and backwashing into 0.01 M HNO3.30 The extraction experiments were
carried out immediately after backwashing to prevent oxidation to Np(V).
For extraction studies of various fission products, HLLW solution generated from
PUREX process was diluted to the desired nitric acid concentration (2 or 4 or 6 M HNO3) and
was used as feed. HLLW originated from research reactor fuel reprocessing plant from a
particular batch having the composition of uranium (predominantly 238U)- 6.34 g/L, plutonium
(predominantly 239Pu)- 4.22 mg/L, 137Cs- 7.98 Ci/L, 106Ru- 7.82 Ci/L,144Ce- 26.51 Ci/L, 90Sr-
3.87 Ci/L, 125Sb- 0.2 Ci/L, Nb- 14.32 mCi/L was used. In view of high activity of HLLW, it
95
was diluted by a factor of 100 to bring down the activity to a measurable level. Consequently, the
activity of the radionuclide 152+154Eu became too low to determine and hence, had to be spiked.
Owing to pure β activity of 90Sr present in the diluted solution, 85+89Sr was spiked where a
gamma emitter, namely, 85Sr was used as a tracer. All distribution studies were carried out in
duplicate and the data were reproducible within an error limit of 5%.
28

Page 29 of 33
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT00748A
Computational methodology
Gas phase geometries of the free oxy diamide ligands and Pu(IV) complexes of OBDA-V and
OTDA-VI were optimized at the GGA level of density functional theory (DFT) by using Becke’s
exchange functional31 in conjunction with Perdew’s correlation functional32 (BP86) with
generalized gradient approximation (GGA) where 60 electron core pseudopotentials (ECPs)
along with the corresponding def-SV(P) basis set were selected for the Pu(IV) ion. All other
lighter atoms were treated at the all electron (AE) level. All the calculations were performed
using the TURBOMOLE program package (TURBOMOLE is program package developed by
the Quantum Chemistry Group at the University of Karlsruhe, Germany, 1988).33 Natural
charges were calculated using natural population analysis as implemented in TURBOMOLE
program using the same basis function and DFT functional as used for the geometry
optimization.
References
1 B. A. Semenov and N. Oi, Nuclear Fuel Cycles- Adjusting to New Realities, IAEA
Bulletin, Vienna, Austria, 1993.
2 J. L. Swanson, PUREX Process Flowsheets, in Science and Technology of Tributyl
Phosphate, ed. W. W. Schulz, L. L. Burger, J. D. Navratil and K. P. Bender, CRC
Press Inc., Boca Raton, 1984, p. 55.
29