Journal of Alloys and Compounds 436 (2007) 187194

Extraction mechanism of Gd3+ by BTMPPA

Md. Fazlul Bari a, , Bahruddin Saad b , Ismail Abdur Rahman b , Muhammad Idiris Saleh b
a

School of Materials Engineering, Northern Malaysia University College of Engineering, KUKUM, Taman Muhibbah 02600, Arau, Perlis, Malaysia
b School of Chemical Sciences, Universiti Sains Malaysia, 11800 Minden, Penang, Malaysia
Received 16 March 2006; received in revised form 29 June 2006; accepted 1 July 2006
Available online 4 August 2006

Abstract
The extraction behavior of Gd(III) from acidic nitrate-acetato medium by bis(2,4,4-trimethylpentyl)phosphinic acid (BTMPPA, H2 A2 ) in toluene
either alone or in combination with trioctylphospine oxide (TOPO, B) has been investigated as a function of contact time, concentrations of Gd3+ ,
H+ , NO3 and Ac (acetate) in the aqueous phase, the concentration of H2 A2 and TOPO in the organic phase and temperature. Dependencies of H+ ,
BTMPPA, TOPO and Ac concentrations on extraction ratio suggest that the extracted species are [Gd(Ac)(HA2 )2 H2 A2 ] and [Gd(Ac)A2 HA2B]
for without and with TOPO, respectively at lower loading, while GdA3 is extracted at higher loading in both cases. The synergistic effect of TOPO
is observed only at higher TOPO concentrations region. The equilibrium constants are evaluated to be 100.63 and 102.79 for the system without and
with TOPO, respectively. The La(III)/Gd(III) separation factors are found to be 348 and 393 for the system without and with TOPO, respectively.
The temperature dependence of the extraction data from acidic nitrate-acetato gives the enthalpy of extraction, 56 kJ/mol. The loading capacity for
100 g BTMPPA is found to be 17.89 g Gd(III).
2006 Elsevier B.V. All rights reserved.
Keywords: Extraction mechanism; Gadolinium; Cyanex 272

1. Introduction
The wide applications of gadolinium have increased the
search for suitable process to recover high purity, and large
amount of the metal. Gadolinium yttrium garnets are used in
microwave applications and gadolinium compounds are used in
making phosphors for color TV tubes. About 1% of gadolinium
in iron, chromium, and related alloys improves the workability
and resistance to high temperatures and oxidation. The extreme
low noise characteristics of gadolinium ethyl sulfate makes it use
in duplicating the performance of amplifiers, such as the maser.
Since gadolinium is ferromagnetic, it can be used as a magnetic
component that senses hot and cold. Numerous gadolinium(III)
complexes are used as active constituents of pharmaceuticals
employed as contrast agents in clinical MRI diagnostics. The
revealing of extraction mechanism and separation possibility
of gadolinium will contribute to the various advance materials
manufacturing.

Corresponding author. Tel.: +604 9798608; fax: +604 9798178.

E-mail address: fazlul@kukum.edu.my (Md.F. Bari).

0925-8388/$ see front matter 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2006.07.004

Using the method of slope analysis for the extraction of Gd(III) from aqueous nitrate medium by di-2ethylhexylphosphoric acid (HA), the extracted species was
found to be [GdA3 (HA)] with the extraction equilibrium
constant (Kex ) = (1.48 0.042) 1012 mol/dm3 [22]. It was
found for the extraction and separation of La(III) and Gd(III)
from aqueous solutions by hydroxyquinoline and Kelex 100
using cloud point extraction method that extraction efficiency and selectivity depend on pH of the solution and
the structure of the chelating agent [23]. The relationship
between the complex structure and extractabilities of the ligand was discussed in the extraction of Gd(III) with 1,1 -(3,6,9trioxaundecanedionyl)diphenothiazine in a waternitrobenzene
[24].
Equilibrium and kinetic studies of lanthanum extraction have
been studied earlier [1,2]. Di-(2,4,4-trimethylypentyl) phosphinic acid is known to extract several metal ions from different
acid solutions [38]. It has been observed that the extraction
behavior varies with concentrations of metal ions and hydrogen
ion, types of anion, counter ion, extractant and diluents. Consequently, selectivity can also be changed by varying the extraction
conditions. Previous studies showed that different species can

188

Md.F. Bari et al. / Journal of Alloys and Compounds 436 (2007) 187194

be formed due to the presence of different anion in the aqueous phase. Detailed studies on the extraction of lanthanum(III)
in our laboratory revealed several new information. Since lanthanides have similar chemical properties, individual lanthanide
is a problem employing solvent extraction technique, separation
of these metals in a pure form was achieved. This paper discusses
the extraction behavior of gadolinium(III) from acidic nitrateacetato medium by BTMPPA in detail and suggest the possibility
to separate gadolinium from lanthanum. Although in nature
many lanthanides are coexist instead of only two (lanthanum and
gadolinium), separation of gadolinium and lanthanum will give
the idea for the others. In our previous studies, the extraction of
lanthanum by BTMPPA was conducted [1]. The present studies
is an extension of that work where the extraction of gadolinium
was investigated.
2. Experimental
Gadolinium oxide (99.9% purity) obtained from Sigma was used
to prepare the stock solutions (10 g/L). Commercial grade bis-(2,4,4trimethylpentyl)phosphinic acid (kindly supplied by Cytec Canada Inc.) which
purity was determined by the potentiometric titration method to be 92% and it
was purified to about 99.5% by the middle phase micro-emulsion method [9].
The stock solution of 1.0 M dimeric BTMPPA and 0.50 M TOPO (Merck, 99%)
were prepared in toluene.
A double-beam spectrophotometer (Model, U2000, Hitachi) with 1.0 cm
corex cuvette was used for colorimetric estimation of Gd(III). A pH meter (Orion,
Model EA940) equipped with a combination glass electrode, was used for pH
measurements and potentiometric titrations. A wrist-action flask shaker (Model
SFI, Stuart Scientific) was used for phase agitation in extraction. A Perkin-Elmer
(Model, 200-FT-IR, Perkin-Elmer) spectrophotometer was used for IR spectra
of the isolated solid spectra in KBr.
The extraction was performed at ambient temperature (29 2 C) as
described earlier work [1]. The extraction was carried out by agitating 10 mL of
the aqueous phase containing definite amounts of Gd3+ , NO3 and Ac at a particular pH and an equal aliquot of the organic phase containing definite amounts
of BTMPPA alone or in mixture with TOPO taken in a 100 mL stoppered conical
flask. For adjusting the concentrations of Ac and NO3 and the pH of the aqueous phase, acetic acid, nitric acid and 1M NaOH, respectively, were used. The
mixing was done for 15 min and the mixed mass was allowed to settle for 30 min
and then disengagged. The aqueous phase was subjected to pH determination as
well as the determination of Gd(III)-content using the Arsenazo III method [10].
The gadolinium content of the organic phase was estimated by difference. Occasionally, gadolinium concentration in the organic phase was checked by stripping
with 2 M HNO3 . Especially when gadolinium concentration in the organic phase
is <30 ppm. The distribution ratio, D = [Gd3+ ](o) /[Gd3+ ](a) , was determined. For
the temperature variation, extraction was conducted in the thermostated water
bath.
Loading of 0.05 M H2 A2 (25 mL) was done by repeated contacts with fresh
25 mL of the Gd3+ Solution. After loading, the organic phase was separated
and centrifuged, and evaporated to remove toluene under vacuum. A white GdBTMPP complex obtained was kept in oven at 70 C for 4 days, and stored in a
dessicator. KBr pellets were prepared for IR spectrum.

3. Results and discussions

Variation of distribution ratio with contact time showed that
distribution ratio increases time up to 2 min and then remains
almost constant. However, in subsequent experiments, contact
time of 15 min was used to ensure equilibration at all parameters
used.
Distribution ratio was found to be decreased with increasing
Gd(III) concentration up to 3.82 mM of initial Gd(III) concen-

Fig. 1. Dependence of distribution ratio on Gd(III) ion concentration in aqueous

phase and plot of log[Gd(III)]](o) vs. log[Gd(III)](eq) : [H2 A2 ] = 0.05 M, pH 3.00,
[Ac ] = 0.25 M, [NO3 ] = 0.10 M (); log[Gd(III)](o) vs. log[Gd(III)](eq) plot:
pH 3.0, [H2 A2 ] = 0.05 M, [Ac ] = 0.25 M, [NO3 ] = 0.10 M, s = 0.42 ().

tration and after that it increases a small extent with increasing

Gd(III) up to a maximum of 5.09 mM, and followed by decrease
thereafter (Fig. 1). Decrease of D with increasing of Gd(III) concentration is due to in the case of decreasing free extractant with
gadolinium loading or non extractable species is formed and
amount increases with increasing the concentration of Gd(III)
in the aqueous phase, or polymerization is occurred in the aqueous phase. An emulsion-like third phase was observed in the
organic phase to be formed when the initial gadolinium concentration in the aqueous phase >3.82 M. It was reported in the
earlier that distribution ratio decreases with metal ion concentration in the case of increasing of polymerization in the aqueous
phase [15]. Due to the formation of emulsion like third phase,
there is a decreasing of metal ions in the organic phase, consequently more metal ions transfer from the aqueous phase to
the organic phase. Thus distribution values increase since D values were calculated by estimating the Gd(III) experimentally in
the aqueous phase. However, the D values again decrease above
5.09 mM of Gd(III) concentration and this may due to the fact
of saturation of third phase is occurred. Unlike the extraction of
lanthanum(III), the increases of distribution ratio after 3.82 mM
of initial Gd(III) concentration is not as high as the system of
lanthanum extraction [1]. These results conclude that the amount
of formation of third phase in organic phase is not too high for
the extraction of gadolinium at concentration >3.82 mM similar
to that of lanthanum. The log[Gd3+ ](o) versus log[Gd3+ ](eq) was
plotted in Fig. 1 and a slope of 0.46 was obtained, also indicating
that mononuclear species is not extracted into the organic phase
since due to the formation of mononuclear species one slope
is obtained [11]. Polymerization observed for the extraction of
yttrium and lanthanides was reported [1,1214].
With increasing nitrate ion in the aqueous phase D values
slightly decrease suggesting that nitrate ion does not participate
in the extraction.

Md.F. Bari et al. / Journal of Alloys and Compounds 436 (2007) 187194

Fig. 2. Dependence of distribution ratio on aqueous phase hydrogen ion concentration for the extraction of Gd(III): [Gd(III)] = 0.64 mM, [NO3 ] = 0.1 M,
[Ac ] = 0.25 M. Without TOPO: () [H2 A2 ] = 0.025 M, s = 1.97, I = 5.50; ()
[H2 A2 ] = 0.05 M, s = 1.96, I = 4.55; () [H2 A2 ] = 0.10 M, s = 1.99, I = 3.75.
With TOPO: () [H2 A2 ] = 0.05 M, [TOPO] = 0.1 M, s = 1.90, I = 4.63.

For the extraction of Gd(III) by BTMPPA alone, the effect

of hydrogen ion concentration on D values (Fig. 2) showed the
D increases with increasing pH up to 4.0 and then decreases.
The slope of the straight line 2 (pH 1.04.0) is due to the
release of two H+ on complex formation. At higher pH (>4.0)
gadolinium ion gets hydrolyzed [1,25]. With the mixture of
BTMPPA and TOPO, similar plots were obtained suggesting
that below pH 4.0, two H+ ions are liberated developing the
complex, as was reported earlier for the La(III)BTMPPA and
La(III)BTMPPATOPO systems [1].
The variations of D values with extractant concentration were
given at three different pH both without and with TOPO (Fig. 3).
Experimental distribution ratio was corrected at expected pH
using pH dependence results. The straight line plots with slopes
of about 3 and 1.5 were obtained without and with TOPO, respectively. This apparently indicates that the extraction dependences
are 3 and 1.5 in the system without and with TOPO, respectively.
In the earlier work, extractant dependence of 3 was reported for
the extraction of Ln3+ from chloride medium by BTMPPA in
systems without TOPO and that of 2 in the system with TOPO
[3,6,16]. These results are probably due to differences in the
experimental conditions used in the respective studies, especially with regards to the metal ion concentration, anions. In our
previous work on lanthanum extraction, dependences of 2 and 1
were found in the system without and with TOPO, respectively.
These may arise due to the differences of electronic configuration and atomic sizes.
TOPO (B) was used as synergistic agent while exhibiting
chelating agent properties [3,16,1]. Variation of distribution ratio
with TOPO was investigated and given in Fig. 4. All points did

189

Fig. 3. Dependence of distribution ratio on extractant concentration in

the organic phase for the extraction of Gd(III): [Gd(III)] = 0.64 mM,
[NO3 ] = 0.10 M, [Ac ] = 0.25 M. Without TOPO: () pH 2.06, s = 3.04,
I = 3.38; () pH 2.91, s = 3.04, I = 4.87; () pH 4.03, s = 3.03, I = 6.79. With
TOPO: () pH 2.50, s = 1.47, I = 1.87; () pH 3.14, s = 1.52, I = 3.13; () pH
3.80, s = 1.51, I = 4.37.

Fig. 4. Dependence of distribution ratio on TOPO in the organic phase

for the extraction of Gd(III): [Gd(III)] = 0.64 mM, pH(eq) 3.14, [H2 A2 ] =
0.05 M, [Ac ] = 0.25 M, [NO3 ] = 0.1 M. The solid line is log DGd = log Kex
where
[H2 A2 ]1.5 [H+ ]2 (1 + KAcTOPO [Ac ]2 )1 + log(1 + KTOPO [TOPO]2 ),
KTOPO is a proportionality constant. The dotted and dashed lines are asymptotes
at higher and lower concentration region of TOPO.

190

Md.F. Bari et al. / Journal of Alloys and Compounds 436 (2007) 187194

variation of Ac concentration in the aqueous phase can be

expressed mathematically by the following empirical relationship:
log DGd = log Kex [H2 A2 ]3 [H+ ]2 log(1 + KAc [Ac ]2 )
(without

TOPO)

(2)

The equation has two asymptotes (Fig. 5); the point of intersection of these two asymptotes gives the value of KAc as 3.5. So,
under the experimental conditions, the equation takes the form:
log DGd = 0.8 log(1 + 3.5[Ac ]2 )

(3)

(4)

log DGd = 1.55 log(1 + 3.5[Ac ] )

and
log DGd = 2.28 log(1 + 3.5[Ac ]2 )

Fig. 5. Dependence of distribution ratio on acetate ion concentration in the aqueous phase for the extraction of Gd(III): [Gd(III)] =
0.64 mM, [NO3 ] = 0.1 M. Without TOPO: pH 3.00, [H2 A2 ] = 0.025 M
(); [H2 A2 ] = 0.05 M (), [H2 A2 ] = 0.10 M (). The solid line is
log DGd = log Kex [H2 A2 ]3 [H+ ]2 log(1 + KAc [Ac ]2 ),
where
KAc
is proportionality constant. With TOPO: [TOPO] = 0.1 M, [H2 A2 ] =
0.05 M, pH 3.14 (O). The solid line is log DGd = log Kex
[H2 A2 ]1.5 [H+ ]2 (1 + KTOPO [TOPO]2 ) log(1 + KAcTOPO [Ac ]2 ),
where
KAcTOPO is a proportionality constant. The dotted and dashed lines are
asymptotes at higher and lower concentration region of Ac .

not fall on straight line, a curve was obtained. Distribution ratio

increases with increasing TOPO concentration at higher concentration region and at lower concentration region, distribution
ratio was independent. The solid line was obtained from the
equation:
log DGd = log Kex [H2 A2 ]1.5 [H+ ]2 (1 + KAcTOPO [Ac ]2 )1
+ log(1 + KTOPO [TOPO]2 )

(1)

where KTOPO is a proportionality constant. All experimental

points were found to be closed to the theoretical solid line. Therefore it can be concluded that the dependency of TOPO is 0 and
2 at lower and higher concentration region, respectively. Synergistic property of TOPO was found only at higher concentration
region. Dependency of TOPO was found to be 0 and 1 at lower
and higher concentration region, respectively for the extraction
of lanthanum(III) [1].
Dependence of acetate ion concentration on D values was
given in Fig. 5 both without and with TOPO. It was found
that all points did not fall on the straight line and a curve was
obtained instead. In both cases, the distribution ratio decreases
with increasing acetate ion concentration in the aqueous phase at
higher acetate ion concentration region (0.251.0 M) while the
distribution ratio was independent with the acetate ion concentration at lower acetate ion concentration region (0.010.25 M)
when the initial Gd3+ , BTMPPA and H+ concentrations were
kept constant. The variation of distribution ratio data with the

(5)

for 0.025, 0.05 and 0.10 M BTMPPA, respectively. Using Eqs.

(3)(5), the values of log DGd can be calculated for any Ac
concentration chosen in the aqueous phase. Using these values, the solid lines in Fig. 5 were obtained. It was found that
the experimental points fallen very close to these curves. The
decrease of the distribution ratio of Gd(III) with an increase in
the acetate ion concentration suggests that at higher Ac concentration, Gd(III) exists as [Gd(Ac)3 ] species and this species
forms extractable [Gd(Ac)(HA2 )2 2HA] species with liberation
of two Ac . As the concentration of Ac is increased in the
aqueous phase, the non-extractable [Gd(Ac)3 ] species is gradually formed. The formation of the [Gd(Ac)3 ] species in the
aqueous phase reduces the concentration of Gd(III), resulting in
the reduction of the D-value. At low Ac concentration, Gd(III)
may exists as [Gd(Ac)2 ]+ , [Gd(Ac)]2+ or Gd3+ species and these
species form extractable complex without liberation of Ac . It
was shown that at higher acetate concentrations, it is possible to
form the [Gd(Ac)2 ]+ , [Gd(Ac)]2+ or [Gd(Ac)3 ] species [26].
The variation of distribution ratio data with the variation of
Ac concentration with TOPO in the aqueous phase (Fig. 5) can
also be expressed mathematically by the following empirical
relationship:
log DGd = log Kex [H2 A2 ]1.5 [H+ ]2 (1 + KTOPO [TOPO]2 )
log(1 + KAcTOPO [Ac ]2 )
(with

TOPO)

(6)

where KAcTOPO is proportionality constant. Therefore it can be

concluded that two acetate ions are liberated only at higher
acetate ion concentration and no acetate ion is liberated at lower
acetate ion concentration in both systems.
These results suggest that Gd(III) is extracted by BTMPPA
via following reaction equations:
At higher acetate ion concentration region:
[Gd(Ac)3 ] + 3H2 A2
= [Gd(Ac)(HA2 )2 H2 A2 ] + 2H+ + 2Ac
(without

TOPO)

(7)

Md.F. Bari et al. / Journal of Alloys and Compounds 436 (2007) 187194

191

[Gd(Ac)3 ] + 1.5 H2 A2 + 2B
= [Gd(Ac)A2 HA 2B] + 2H+ + 2Ac
(with

TOPO)

(8)

At lower acetate ion concentration region:

[Gd(Ac)]2+ + 3H2 A2 = [Gd(Ac)(HA2 )2 H2 A2 ] + 2H+
(without

TOPO)

(9)

[Gd(Ac)]2+ + 1.5 H2 A2 + 2B
= [Gd(Ac)A2 HA 2B] + 2H+

(with

TOPO)
(10)

Separation factor can be defined between two metals as the

ratio of D values of one metal to that of another. Higher separation factor is desirable to enhance the separation and to reduce
the number of stages of extraction. In this work, separation factor was calculated from the pH and extractant dependencies
(Figs. 2 and 3) and from the extraction of La(III) in earlier investigation [1]. The calculated separation factor was plotted against
pH and extractant concentration in both system with and without TOPO and given in Fig. 6. From the plot, it was seen that
separation factor increases with increasing the value of pH and
extractant concentration and maximum separation factor was
found at 0.158 M BTMPPA, pH 3.74 and 0.1 M TOPO. TOPO
was found to show the synergistic effect and also enhance the
separation.
From the intercepts of the lines in Figs. 25, the values of
equilibrium constant for the extraction system of gadolinium
by BTMPPA at lower loading were evaluated and shown in
Tables 1 and 2 in the system without and with TOPO, respectively. The average values of log Kex are 0.63 with a standard
deviation of 0.14 and 2.79 with a standard deviation of 0.09
in the system without and with TOPO, respectively. The values of log Kex were 2.16 with a standard deviation of 0.10
and 2.58 with a standard deviation of 0.21 without and with
TOPO, respectively for the extraction of lanthanum by BTMPPA
reported earlier [1]. It should be pointed out that these equilibrium constants were obtained under the investigated conditions.
Effect of temperature on D values was given at three different BTMPPA concentrations (Fig. 7). The straight line plots

Fig. 6. Effect of extractant concentration and pH on separation factor: [La(III)]/[Gd(III)] = 100 ppm, [Ac ] = 0.25 M, [NO3 ] = 0.10 M. Without
TOPO: () pH 2.5; () [H2 A2 ] = 0.158 M. With TOPO: () pH 2.5; ()
[H2 A2 ] = 0.158 M, [TOPO] = 0.10 M.

with slopes of the lines were 2.93 103 , 2.92 103

and 2.88 103 giving H values of 55.56, 55.91 and
55.67 kJ/mol for 0.025, 0.05 and 0.10 M H2 A2 concentrations,
respectively. The positive H value indicated that the gadolinium extraction process under investigation is endothermic in
nature. The value of H was reported to be 60.12 kJ/mol for the
extraction of lanthanum [1].
The loading capacity, defined as the amount of metal ion in
grams extracted per 100 g of pure extractant, is a very important
factor for its commercial applicability. Higher loading capacity
is desirable for any extractant-metal complex system. More-

Table 1
Evaluation of equilibrium constant, Kex for the extraction of Gd(III) by BTMPPA in toluene without TOPO
Parameter
H2 A2

pH

Ac
a

[H2 A2 ] (M)

pH

[Ac ] (M)

Intercepts

log Kex

Average log Kex

S.D.

0.025
0.05
0.10
0.025
0.05
0.10

2.06
2.91
4.03

3.00
3.00
3.00

0.25
0.25
0.25
0.25
0.25
0.25

3.38
4.49
6.79
5.50
4.55
3.75
0.80
1.55
2.28

0.66
0.87
1.19a
0.61
0.56
0.66
0.39
0.55
0.72

0.63

0.14

Value is not included to calculate average due to different extracted species being formed at that pH.

192

Md.F. Bari et al. / Journal of Alloys and Compounds 436 (2007) 187194

Table 2
Evaluation of equilibrium constant, Kex for the extraction of Gd(III) by BTMPPA in toluene with TOPO
Parameter
H2 A2
pH
Ac
TOPO

[H2 A2 ] (M)

pH

[Ac ] (M)

[TOPO] (M)

Intercepts

log Kex

Average log Kex

S.D.

0.05
0.05
0.05

2.50
3.14
3.80

3.14
3.14

0.25
0.25
0.25
0.25

0.25

0.10
0.10
0.10
0.10
0.10

1.87
3.13
4.37
4.63
1.28
1.00

2.84
2.79
2.80
2.62
2.90
2.80

2.79

0.09

over, the species extracted at higher loading may be easily

converted to pure (usually solid) complexes for its structure to
be determined by instrumental methods and also for its elemental composition. Aliquots of 25 mL of 0.05 M BTMPPA
were repeatedly extracted for 20 min using equal volume of
aqueous solution at pH 4.5 containing 3.18 mM Gd3+ . After
equilibration, the aqueous phase was analyzed for Gd3+ content colorimetrically. The amount of Gd3+ transferred into the
organic phase for each stage of extraction was then calculated by
difference and cumulative concentration of Gd3+ in the organic
phase (cumulative [Gd(III)](o) , M) after each contact was determined. The cumulative [Gd(III)](o) versus contact number plot
(Fig. 8) suggests 25 mL of 0.05 M H2 A2 solution can extract
as much as 0.1297 g gadolinium, which amounts to a loading
capacity 17.89 g Gd3+ /100 g BTMPPA. The values of loading capacity of D2EHPA for titanium and iron, and BTMPPA
for lanthanum were 7.31 g Ti4+ /100 g D2EHPA [17], 8.389 g

Fig. 7. Dependence of distribution ratio on temperature. Plot of log DGd

against the inverse of absolute temperature. [Gd(III)] = 0.64 mM, pH 3.00,
[Ac ] = 0.25 M, [NO3 ] = 0.1 M; [H2 A2 ] = 0.10 M, s = 2.88 103 ();
[H2 A2 ] = 0.05 M, s = 2.92 103 (); [H2 A2 ] = 0.025 M, s = 2.93 103
().

Fe3+ /100 g D2EHPA [18] and 15.97 g La3+ /100 g BTMPPA [1],
respectively.
The Gd3+ /BTMMPA ratio in the extracted species was calculated to be 1:1.5. The ratio suggests that the reaction at higher
loading proceeds according to the equation:
[Gd(Ac)2 ]+ + 1.5H2 A2 = GdA3 + 2Ac + 3H+

(11)

The ratio of phosphorus to gadolinium of the solid complex

obtained was 3. This ratio indicates that the composition of
the complex is GdA3 . The IR spectra of BTMPPA and the
gadolinium solid complexes were shown in Fig. 9. The presence of P(OH) and P O(OH) stretching in the BTMPPA at
27502550 cm1 and the absence of P(OH) and P O(OH)
stretching in the complex at 27502550 and 27002550 cm1
suggests a strong chelation of Gd3+ by monomeric BTMPPA
[1921]. Similar observation has been reported by Biswas and
Begum for the extraction of tetravalent titanium by D2EHPA
[17].

Fig. 8. Loading of the organic phase by Gd3+ : [Gd(III)] = 3.18 mM,

[H2 A2 ] = 0.05 M, pH 4.50, [NO3 ] = 0.10 M, [Ac ] = 0.25 M.

Md.F. Bari et al. / Journal of Alloys and Compounds 436 (2007) 187194

193

Fig. 9. IR spectra for BTMPPA and Gd-BTMPPA complex.

4. Conclusion
BTMPPA can extract Gd3+ from acetato-nitrate solution.
The equilibrium time is 2 min and distribution ratio slightly
decreases with increasing nitrate ion concentration. Distribution ratio decreases with increasing Gd3+ concentration in the
aqueous phase. However, the distribution ratio slightly increases
between the concentration of 3.82 and 5.09 mM. These results
suggest the polymerization in the aqueous phase. The loglog
plot of [Gd3+ ](o) versus [Gd3+ ](eq) indicates that mononuclear
species is not extracted, but different types of species is extracted
instead. The pH and log extractant dependences suggest that
[Gd(Ac)(HA2 )2 H2 A2 ] and [Gd(Ac)A2 HA2B] are extracted in
the system without and with TOPO, respectively. Distribution
ratio is proportional to the square of acetate ion concentration
at higher concentration region and almost independent at lower
concentration region in the system both without and with TOPO.
Distribution ratio is proportional to square of TOPO concentration at higher TOPO concentration region and independent at

lower TOPO concentration region. The equilibrium constant,

Kex is 100.63 with a standard deviation of log Kex of 0.14 in
the system without TOPO and 102.79 with a standard deviation
of log Kex of 0.09 for the extraction system with TOPO. The
maximum separation factor between Gd and La is calculated to
be 393 at 0.158 M BTMPPA, pH 3.74 and 0.1 M TOPO. The
temperature dependence data give the enthalpy of extraction,
56 kJ/mol. The loading capacity is 17.89 g Gd(III)/100 g HA. It
can be concluded from the loading tests that the phosphorus to
gadolinium ratio in the solid complex as obtained is 3, indicating
formation of GdA3 at higher loading. Evidence from IR spectra
also suggests the strong chelation of Gd3+ by A .

Acknowledgement
A postdoctoral fellowship to one of us (M.F. Bari) from
the Ministry of Science and Environment, Malaysia is kindly
acknowledged.

194

Md.F. Bari et al. / Journal of Alloys and Compounds 436 (2007) 187194

References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]

M.I. Saleh, M.F. Bari, B. Saad, Hydrometallurgy 63 (1) (2002) 7584.

M.I. Saleh, M.F. Bari, B. Saad, Hydrometallurgy 67 (13) (2002) 4552.
K. Li, H. Freiser, Solvent Extr. Ion Exch. 4 (4) (1986) 739755.
X. Fu, Z. Hu, Y. Liu, J.A. Golding, Solvent Extr. Ion Exch. 8 (45) (1990)
573595.
K. Akiba, H. Yamada, S. Nakamura, J. Chem. Soc. Jpn. 5 (1993) 555560
(in Japanese).
X. Zhang, D. Li, Acta Metall. Sin. 10 (4) (1993) 7274.
J. Sun, D. Li, Chin. J. Appl. Chem. 11 (3) (1994) 4953.
S. Paatero, T. Lantto, P. Ernola, Solvent Extr. Ion Exch. 8 (3) (1990)
371388.
Z. Hu, Y. Pan, W. Ma, X. Fu, Solvent Extr. Ion Exch. 13 (5) (1995) 965976.
S.B. Savvin, Talanta 8 (1961) 673685.
T. Francis, R.T. Prasada, M.L.P. Reddy, Hydrometallurgy 57 (3) (2000)
263268.
T. Harada, M. Smutz, J. Inorg. Nucl. Chem. 32 (1970) 649662.
T. Harada, M. Smutz, R.G. Bautista, J. Chem. Eng. Data 17 (2) (1972)
203204.

[14] E. Antico, A. Masana, M. Hidalgo, V. Salvado, M. Iglesias, M. Valiente,

Anal. Chim. Acta 327 (1996) 267276.
[15] Z. Kolarik, R. Grimm, J. Inorg. Nucl. Chem. 38 (1976) 17211727.
[16] Y. Komatsu, H. Freiser, Anal. Chim. Acta 227 (2) (1989) 397404.
[17] R.K. Biswas, D.A. Begum, Hydrometallurgy 49 (3) (1998) 263274.
[18] R.K. Biswas, D.A. Begum, Hydrometallurgy 50 (2) (1998) 153168.
[19] D.F. Peppard, J.R. Ferraro, J. Inorg. Nucl. Chem. 10 (1959) 275288.
[20] L.J. Bellamy, The Infra-red Spectra of Complex Molecules,
Wiley/Chapman and Hall, New York, 1975, pp. 381392.
[21] L.W. Daasch, D.C. Smith, Anal. Chem. 23 (1951) 853868.
[22] N.E. El-Hefny, S.I. El-Dessouky, J. Chem. Technol. Biotechnol. 81 (3)
(2006) 394400.
[23] N. De Jong, M. Draye, A. Favre-Reguillon, G. LeBuzit, G. Cote, J. Foos,
J. Colloid Interf. Sci. 291 (1) (2005) 303306.
[24] X. Li, Y. Wen, W. Liu, S. Liu, M. Tan, K. Yu, J. Coord. Chem. 56 (18)
(2003) 15371547.
[25] S. Deberdt, S. Castet, J.L. Dandurand, J.L. Harrichoury, I. Louiset, Chem.
Geol. 151 (1998) 349372.
[26] S. Deberdt, S. Castet, J.L. Dandurand, J.L. Harrichoury, Chem. Geol. 167
(2000) 7588.