Journal of Molecular Liquids 291 (2019) 111242

Contents lists available at ScienceDirect

Journal of Molecular Liquids

journal homepage: www.elsevier.com/locate/molliq

A rapid dispersive liquid-liquid microextraction based on hydrophobic

deep eutectic solvent for selective and sensitive preconcentration of
thorium in water and rock samples: A multivariate study
Susan Sadeghi ⁎, Abdolbaset Davami
Department of Chemistry, Faculty of Science, University of Birjand, P.O. Box. 97175/615, Birjand, Iran

a r t i c l e i n f o a b s t r a c t

Article history: Thorium [Th(IV)] and its compounds are highly toxic and cause severe damage to bone and kidney through long
Received 13 April 2019 term exposures. Conventional methods of Th(IV) detection are complicated and costly. In this study, we devel-
Received in revised form 9 June 2019 oped a new dispersive liquid-liquid microextraction (DLLME) based on hydrophobic deep eutectic solvent
Accepted 25 June 2019
(HDES) for preconcentration of Th(IV) in water and rock samples prior to UV–Vis. spectrophotometric detection.
Available online 11 July 2019
The microextraction procedure was carried out by applying thorin as a chelating agent, cethyltrimethyl ammo-
Keywords:
nium bromide (CTAB) as an ion pairing agent and a DES consisted of 1-hexyl-3-methylimidazolium and salicylic
Thorium acid components as an extraction solvent. Central composite design (CCD) was applied to obtain the best condi-
Dispersive liquid-liquid microextraction tion of the microextraction procedure. The developed method is a green, facile, fast, selective and sensitive that
Central composite design could detect Th(IV) in the concentration range of 10–600 ng mL−1 with a limit of detection (LOD) and
Hydrophobic deep eutectic solvent preconcentration factor of 2.1 ng mL−1 and 59, respectively. The relative standard deviation of five replicate mea-
Spectrophotometry surements of Th(IV) at 100 ng mL−1 was 1.7%. Finally, the proposed method was successfully applied to the anal-
ysis of Th(IV) in water and rock samples, and desirable results were obtained.
© 2019 Published by Elsevier B.V.

1. Introduction determination of Th(IV). However, it suffers from low sensitivity and

matrix interferences in determination of analytes in real samples. Due
Th(IV) as the radioactive element exists on monazite sand, soil, rock, to these drawbacks, separation and preconcentration are trivial steps
plant, atmosphere and water. It has extensive applications for various in trace analysis of metal ions. Several preconcentration approaches
industries such as metallurgy, optic, ceramic and especially in nuclear such as solid phase extraction (SPE) [9], liquid-liquid extraction (LLE)
energy sources [1,2]. People who are exposed to thorium through [10], cloud point extraction (CPE) [11] and flotation [12] have been
food, air and water are at risk of bone fractures and liver and kidney fail- established for the determination of Th(IV). But, these methods are te-
ure. According to the World Health Organization (WHO), concentration dious, provide insufficient preconcentration, consume a large amount
of Th(IV) in drinking water should be below 20 μg L−1 [3]. Hence, there of toxic organic solvents, time-consuming and produce secondary labs'
is a considerable attention to develop, simple, sensitive and selective waste that limits their applications. To overcome these drawbacks, min-
methods for determination of Th(IV) in environmental samples. Con- iaturized extraction techniques such as liquid phase microextraction
ventional techniques for detection of low concentration of Th(IV) are (LPME) [13] and solid phase microextraction (SPME) [14] have been de-
usually based on inductively coupled plasma-atomic emission spec- veloped. Until now, various miniaturized liquid phase extraction
trometry (ICP-AES) [4], inductively coupled plasma-mass spectrometry methods were developed such as emulsification microextraction
(ICP-MS) [5], capillary zone electrophoresis [6], ion chromatography [7] (EME), single-drop microextraction (SDME), hollow-fiber LPME (HF-
and X-ray fluorescence spectroscopy [8]. Although these techniques LPME), in situ solvent formation microextraction (ISFME) and disper-
provide reliable results, they require daily maintenance, well- sive liquid–liquid microextraction (DLLME) [15]. These methods pro-
controlled experimental conditions, costly instruments and in some vide important benefits to extraction of analytes such as simplicity,
cases time consuming. Because of simplicity, relatively low cost, and less consumption of toxic solvents, high preconcentration factor, being
availability in most laboratories, UV–Vis. spectrophotometry may be ef- an eco-friendly method and combine with different analytical measure-
fectively used as a reliable and an alternative technique to ment techniques. Among the LPME methods, DLLME is a well-
developed method to optimize microextration process. In this method,
⁎ Corresponding author. a mixture of a water-immiscible solvent and a small amount of a
E-mail address: [email protected] (S. Sadeghi). water miscible solvent as an extraction and disperser solvents,

https://doi.org/10.1016/j.molliq.2019.111242
0167-7322/© 2019 Published by Elsevier B.V.
2 S. Sadeghi, A. Davami / Journal of Molecular Liquids 291 (2019) 111242

respectively, is injected rapidly into the sample tube to produce fine interactions between factors. To overcome these drawbacks, response
droplets of the extractant which are dispersed entirely in the aqueous surface methodology (RSM) has been applied. The main benefits of
phase and increases mass transfer of the analyte into the extraction sol- RSM include determination of significant factors, interactions among
vent. Finally, the extraction phase can be settled in the bottom of the factors and reducing the number of experiments [28].
tube which may be removed by a syringe [16]. Great efforts also were In the present study, we persuaded to evaluate a new hydrophobic
made in DLLME to use less amount of organic extraction solvent or elim- DES as a green solvent in a simple, selective and sensitive DLLME ap-
ination of disperser solvent to simple phase separation by air-assisted proach for the microextraction of Th(IV) in aqueous solution for the
DLLME or ultrasound assisted DLLME. Also, to accelerate elmulsification first time. RSM based central composite design (CCD) was employed
procedure, vortex assisted DLLME was introduced whereby the extrac- for optimization of the most significant variables in the DLLME proce-
tion solvents are dispersed into aqueous samples by vortex mixing [15]. dure. The validity of the developed method was checked by determin-
Thus, these methods simplify the operation and speed up the extraction ing Th(IV) in water and rock samples.
procedure. Instead of organic solvents, different types of extraction sol-
vents such as supramolecular solvents, switchable solvents, ionic liq- 2. Experimental
uids, and deep eutectic solvents have been used in DLLME for
extraction and preconcentration of heavy metal ions [17]. Utilization 2.1. Chemical and reagents
of new DLLME based on surfactants as surfactant-assisted DLLME (SA-
DLLME), ion pair-based surfactant assisted microextraction (IP-SA- All of the reagents used in the present work were of analytical grade
ME), and surfactant-enhanced emulsification microextraction (SE-ME) and supplied by Merck (Darmstadt, Germany). 1-methylimidazole, 1-
combined with ultrasonic radiation, vortex agitation, and solidification bromobutane, 1-bromohexane, salicylic acid (SA) and benzoic acid (BA)
improvements have been reported [15,17]. Few methods are reported were used to synthesize DESs. Th(NO3)4.5H2O, cetyltrimethylammonium
in the literature for separation and preconcentration of Th(IV) using bromide (CTAB) and sodium salt of 2-(2-hydroxy-3, 6-disulfo-1-
various DLLME methods. Elmasry et al. used supramolecular solvent- naphthylazo)-benzenearsonic acid (thorin) were applied as the analyte,
based liquid-liquid microextraction method for preconcentration of ion pairing and chelating agents, respectively. The stock standard solu-
thorium(IV) in various environmental samples [13]. Veyseh and Niazi tion of Th(IV) (1000 mg L−1) was prepared by dissolving a suitable
applied flotation-assisted homogeneous LLME for determination of tho- amount of thorium salt in deionized water. The working standard solu-
rium in water samples [5]. Hydari and Ramazani reported a new stirring tions were freshly prepared by diluting the stock solution with deionized
assisted LLME for the analysis of trace amount of thorium ions [18]. water. The pH of the solution was adjusted with 0.1 M HCl or NaOH.
Room temperature ionic liquids (RTILs) are a novel class of green
solvents consisted of an imidazolium or pyridinium cation and organic
2.2. Instrumentation and software
or inorganic anion which have been considered as an alternative for
the organic solvents [19]. Most of the hydrophobic RTILs are contained
Absorption measurements were performed with a UV–Vis. spectro-
[PF6]− as an hydrophobic anion but it produces toxic HF and POF3 com-
photometer 2510-PC (Shimadzu, Japan) equipped with a 1 cm quartz
pounds by hydrolysis. Besides, these solvents are expensive and non-
microcell (0.35 mL). The pH values were determined with BP3001
biodegradable [20]. In recent years, deep eutectic solvents (DESs) have
benchtop (Trans instruments, Singapore) pH meter with a combined
been used as green solvents for extraction of different compounds.
glass electrode. For the phase separation a centrifuge model EBA20
DESs are consisted of two or more components including a hydrogen
(Hettich, Germany) was used. A Magnetic stirrer (Heidolph, Germany)
bond donor (HBD) and a hydrogen bond acceptor (HBA), which can
and a R2062 rotary evaporator (Heidolph, Germany) were employed
be associated with each other at moderate temperature via hydrogen
in the synthesis of DESs. Deionized water was obtained from Aqua
bonding interaction to form an eutectic mixture with lower melting
max water purification system (Young–Lin, South Korea). The central
point than each component [21]. DESs offer solvents with excellent
composite design was accomplished with the Design-Expert®Software
physiochemical properties such as non-flammable, negligible vapor
Version 10-Stat-Ease (Minneapolis, USA).
pressure, extensive liquid range and thermal stability similar to RTILs.
However, facile syntheses, cost effective, not require to furthering puri-
fication, biodegradability and biocompatibility make them potential al- 2.3. Synthesis of DES
ternatives to replace conventional organic solvents as well as traditional
ILs [22–24]. Most of the DESs are hydrophilic that limits their applica- In the first step, the 1-hexyl-3-methyl imidazolium bromide
tions as extractants to recover of analytes from aqueous solutions. Re- ([HMIM]Br) and 1-butyl-3-methyl imidazolium bromide ([BMIM]Br)
cently, some literatures have been reported on production of were synthesized according to the previous report [29]. Briefly, an ap-
hydrophobic DESs for extraction of inorganic or organic targets from propriate amount of 1-methylimidazole and an excess amount of 1-
water samples [21]. Habibollahi and coworker applied hydrophobic bromohexane or 1-bromobutane were poured into a round bottom
DESs as the extraction solvent in DLLME based on the solidified deep eu- flask and the mixture refluxed at 80 °C for 48 h until a viscous liquid
tectic solvent for extraction of heavy metal ions in soil and vegetables ir- was produced. Finally, the products were washed three times with
rigated with treated municipal wastewater [25]. Akramipour et al. ethyl acetate and dried under vacuum at 80 °C. In the second step, to
extended the 1-octyl-3-methylimidazolium chloride and 1-dodecanol synthesize a DES, a HBA and a HBD in certain molar ratio were mixed
as hydrophobic extraction solvent of vortex assisted DLLME based on in a screw-capped bottle and stirred at room temperature until a homo-
the freezing of deep eutectic solvent, and applied it for speciation of or- geneous liquid was formed. In this study, seven DESs (DES-1 to DES-7)
ganic/inorganic mercury and total mercury in blood samples [26]. Zhu were prepared by employing two HBAs, namely [HMIM]Br and [BMIM]
et al. demonstrated that hydrophobic DESs could be used as an extrac- Br which could be combined with salicylic acid (SA) or benzoic acid
tion solvent in LLME to analysis of synthetic pigments in beverages [27]. (BA) as HBDs with different molar ratios (Table S1).
Investigation of main factors that control a process via minimal ex-
periments associated with confident results is desirable in each study. 2.4. Preparation of real samples
In the most conventional multifactor experiments, optimization process
is usually performed by changing a single factor in each step while all Two types of water samples were used including river water (Band-
other factors are set at constant levels (one factor at a time). Besides, a e Dareh, Birjand, Iran) and seawater (Oman Sea, Chabahar, Iran). These
large number of experiments should be performed that consuming water samples were collected, filtered through a 0.45 μm membrane
high amount of materials and times without considering the and finally stored at 4 °C before use.
 S. Sadeghi, A. Davami / Journal of Molecular Liquids 291 (2019) 111242 3

The rock sample was collected from Shah-Kooh area (Nehbandan, thorin has the negative charge and can not extract into the hydrophobic
South Khorasan, Iran). The digestion of rock sample was performed as extraction solvent. Therefore, the presence of an ion pairing agent with
the following: 5 g of rock sample were weighted and transferred into the positive charge is necessary [30]. For this purpose, CTAB was used as
a Teflon reactor containing 50 mL concentrate HF. The reactor was ion pairing agent that could facilitate interaction of Th(IV) with thorin
heated at 100 °C to digest the sample completely. Then, the sample and resulted in an association of Th(IV)-Thorin-CTAB. This association
was heated slowly to evaporate the solvent and thereafter, cooled was very insoluble in the aqueous phase but could easily extract into hy-
down to room temperature. In order to dissolution the residue, a mix- drophobic extraction solvents. The UV–Vis. absorption spectra of
ture containing 5 mL HNO3 (65%, w/v) and 5 mL HCl (37%, w/v) was Thorin, Th(IV)-Thorin, Thorin-CTAB and Th(IV)-Thorin-CTAB complex
added. Finally, the mixture was diluted to 50 mL with deionized water after DLLME are presented in Fig. 1. Thorin exhibited an absorption
and was applied to the DLLME process. spectrum with a maximum wavelength of 473 nm. It is obvious that
Th(IV)-Thorin complex or Th(IV)-CTAB extracted poorly in the hydro-
2.5. DLLME procedure phobic solvent, while in the presence of the CTAB, an ion-pair complex
of Th(IV)-thorin-CTAB was produced which extracted into the DES; pro-
An aliquot of 10 mL sample or standard solution containing viding a significant increase in the absorption with a red shift in the
250 ng mL−1 of Th(IV) (pH 3.2), thorin (60 μM) as chelating agent, maximum wavelength to 515 nm. Thus, the hydrophobic DES could ex-
CTAB (50 μM) as ion pairing agent and NaNO3 (0.5%, w/v) to adjust tract the ternary complex efficiently.
ionic strength, were introduced to a conical bottom centrifuge tube.
Then, 170 μL of [HMIM]-SA (DES-4) as hydrophobic extraction solvent 3.1. Selection of DES
was injected into the solution and immediately was shaken manually
for 1 min. During shaking, fine droplets of the DES were produced so The choice of DES as the extraction solvent is a crucial step in design-
that the red color ternary complex Th(IV)-thorin-CTAB could simplify ing the DLLME method. The proper DES should have low solubility in
transfer from aqueous solution to the DES phase. To separate two aqueous solution, high affinity to extract the target analyte and easily
phases, the cloudy solution was centrifuged at 3500 rpm for 5 min. disperse into aqueous solution. In this work, four types of DES in similar
Then, the aqueous phase was removed by a syringe, and the settled molar ratio (1:1) including [HMIM]-BA (DES-1), [BMIM]-BA (DES-2),
phase was diluted to 500 μL with acidic ethanol and transferred to the [BMIM]-SA (DES-3) and [HMIM]-SA (DES-4) were employed. The pre-
cuvette. The absorbance was recorded in the wavelength range from pared hydrophobic DESs were evaluated for the extraction of Th(IV)
450 to 650 nm against a blank solution which was prepared in the from aqueous solutions. The results (Fig. 2) demonstrated that
same way without the analyte for control. The calibration curve was [HMIM]-SA provided the highest extraction of Th(IV) among the used
constructed based on the measurement of absorbance against the dif- DESs. Therefore, [HMIM]-SA was selected for further experiments.
ferent extracted concentration of Th(IV). After selection of an appropriate DES as extraction solvent, four DESs
of [HMIM]-SA with different molar ratios of [HMIM] to SA (1:1, 2:1,
2.6. Experimental design methodology 1.5:1, 1:1.5; DES-4 to DES-7) were prepared and examined for the ex-
traction of Th(IV). As seen in Fig. 3, changing the molar ratio of
In the present study, a design of experiments based on CCD was [HMIM]Br to SA from 2:1 to 1:1 led to an increase in extraction effi-
employed. Based on the preliminary experiments, four independent ciency and absorbance of the extracted and remained nearly constant
variables were found that affected performance of the microextraction at molar ratio of 1:1.5. Therefore, the [HMIM]-SA with 1:1 molar ratio
process, i.e. sample pH, the concentration of thorin, the concentration (DES-4) was chosen as the best composition of extraction solvent in fur-
of CTAB and the amount of DES. In this design, the absorbance of the ex- ther experiments.
tracted complex of Th(IV)-thorin-CTAB was considered as the analytical
response. The effects of independent variables and their interactions on 3.2. Statistical evaluation of the proposed procedure
the response were studied in five levels and three blocks. Other vari-
ables such as the concentration of NaNO3 (0.5%, w/v), extraction time The effectiveness of the developed regression model based on inde-
(1 min), centrifuge time (5 min) and centrifuge speed (3500 rpm) pendent variables, and the interactions between the variables was eval-
were set at fixed values. The high and low levels of each variable were uated by the analysis of variance (ANOVA). The high F-value (111.42)
selected according to the preliminary experiments (Table 1). The and low P-value (b0.0001) of the model in Table S3 suggesting the
coded values of the variables in the CCD matrix and their corresponding model was significant. The lack of fit (LOF) P-value of 0.1542 was
responses are presented in Table S2. All experiments were performed in more than the critical value (0.05), meaning that the LOF was not
a random manner using 10 mL sample solution containing 250 ng mL−1
Th(IV). 0.6

Thorin
3. Results and discussion 0.5
Th(IV) - Thorin - CTAB
Thorin - CTAB
In this study, thorin was used as a selective chelating agent for Th 0.4
Absorbance

Th(IV) - Thorin
(IV) microextraction. The hydrated free thorium has positive charge,
and the charge of free thorin is negative due to the loss of the sulfonic 0.3
acid hydrogens in acidic media at pH b 6. Thus, the complex of Th(IV)-
0.2

Table 1 0.1
Variables and levels used in the central composite design.

Variable Coded variable levels 0

450 475 500 525 550 575 600 625 650
Symbol -α −1 0 +1 +α
Wavelength (nm)
pH A 1 2 3 4 5
Concentration of thorin (μM) B 5 25 45 65 85
Fig. 1. Absorption spectra of Thorin, Th(IV)-Thorin-CTAB, Thorin-CTAB, and Th(IV)-Thorin
Concentration of CTAB (μM) C 5 30 55 80 105
after microextraction: Conditions: pH 3.2; Sample volume, 10 mL; Th(IV), 250 ng mL−1;
Amount of DES (μL) D 50 100 150 200 250
Thorin, 60 μM; CTAB, 50 μM; DES, 170 μL, [HMIM]- SA DES with molar ratio of 1:1.
4 S. Sadeghi, A. Davami / Journal of Molecular Liquids 291 (2019) 111242

0.6

0.5
0.60
0.4
Absorbance

0.3
A
0.50

0.2
0.40

Predicted
0.1
0.30
0
[HMIM]-SA [HMIM]-BA [BMIM]-SA [BMIM]-BA
0.20
Extraction solvent

Fig. 2. Effect of type of DES on the microextraction of Th(IV) ions. Conditions: pH 3.2; 0.10
Sample volume, 10 mL; Th(IV), 250 ng mL−1; Thorin, 60 μM; CTAB, 50 μM; DES, 170 μL,
DES with molar ratio of 1:1. 0.10 0.20 0.30 0.40 0.50 0.60

Actual

significant relatives to the pure error. The suggested second-order poly-

nomial equation that approximated the relationship between the inde-
4.00
pendent variables and the response after backward mode selection in 3.77363

terms of coded variables could be expressed as the following equation

Externally Studentized Residuals
(Eq. (1)):
2.00
B
Absorbance ¼ þ0:520 þ 0:019A þ 0:034B þ 0:012C þ 0:022D
þ 0:019AB–0:021 BC–0:013BD–0:015CD–0:074A2 ð1Þ
–0:025B2 –0:025C2 –0:020D2 0.00
0

where linear coefficients (A, B, C and D), quadratic coefficients (A2, B2,
C2 and D2) and the interaction coefficients (AB, BC, BD and CD) were sig-
-2.00
nificant in the microextraction process of Th(IV). The value of the coef-
ficient of determination suggests that the proposed polynomial model is
suitable to analyze the importance of individual variables, and interac- -3.77363
-4.00
tion effects of the selected variables in the response. According to
ANOVA results, the predicted and adjusted R2 values (0.9495 and
0.10 0.20 0.30 0.40 0.50 0.60
0.9797) were in reasonable agreement which implied an excellent rela-
tionship between the response and the fitted model. The value of ade-
Predicted
quate precision, defined as a signal-to-noise ratio, for the model was
significantly N4 (41.8), indicating to suitability of the model
(Table S3).The predicted values versus actual data are also shown in
Fig. 4A. As it can be observed, the most actual values were lying on a
straight line, which represented an excellent relationship between the
fitted model and actual data. Also, the random pattern of residual values
99
C
versus the predicted values indicated the high efficiency of the model 95
Normal % Probability

90
80
0.6
70

0.5 50

30
0.4
Absorbance

20
10
0.3
5

0.2 1

0.1

0 -3.00 -2.00 -1.00 0.00 1.00 2.00 3.00

2:1 1.5:1 1:1 1:1.5
Molar ratio of HBD/HBA Externally Studentized Residuals

Fig. 3. Effect of composition of [HMIM]-SA on the microextraction of Th(IV) ions. Fig. 4. (A) A plot of predicted response against observed response. (B) A plot of the
Conditions: pH 3.2; Sample volume, 10 mL; Th(IV), 250 ng mL−1; Thorin, 60 μM; CTAB, internally studentized residuals vs. the predicted response. (C) Normal probability plot
50 μM; DES, 170 μL. of residuals for the absorbance.
 S. Sadeghi, A. Davami / Journal of Molecular Liquids 291 (2019) 111242 5

for good presentation of actual data (Fig. 4B). In addition, Fig. 4C illus- 2–4 on the formation the Th(IV)-thorin-CTAB complex was examined.
trated a normal distribution of residuals in which they were along a As it can be seen from Fig. 5A, the absorbance was increased with the el-
straight line approximately. evation of pH from 2.0 to 3.2 and then decreased from 3.2 to 4.0. This
could be justified by the fact that at higher pH values than 3.2, the com-
3.3. Interaction effects through response surface plots plex was probably incompletely formed or hydrolyzed. Thus, pH of 3.2
was chosen for further experiments. Also, the effect of thorin concentra-
Response surface plot shows relationship between independent and tion as chelating agent on the absorbance of the complex was explored.
dependent (response) variables and their effects on the response graph- According to Fig. 5A, the absorbance was raised up with increasing of
ically in two (2D) or three (3D) dimensions and finding the optimum thorin concentration from 25 to 60 μM in the pH range of 2.0–3.2 and
amounts of the variables that maximize the response. The response sur- thereafter, the absorbance was nearly constant probably due to com-
face plots were depicted the influence of two significant factors on the plete formation of the Th(IV)-thorin-CTAB complex in this pH range.
response in the experimental range, while other factors were remained Consequently, the concentration of 60 μM thorin was selected as the op-
fixed at their central levels (Fig. 5). The observed curvatures of the 3D timum amount. CTAB as cationic surfactant enhanced the extraction of
response surface showed that there was a possible interaction between the complex to the DES extractor through electrostatic interaction
the individual factors. Fig. 5A illustrates the effect of interaction of thorin with Th(IV)-thorin complex. Thus, in the concentration range of thorin
concentration and pH on the response. from 25 to 60 μM, with increasing the CTAB concentration from 30 to 50
The sample pH played a vital role in formation of the metal-chelate- μM, the strength of non-covalent interactions amplified, resulted in high
ion pair complex which in turn affects the extraction efficiency. In the extraction to the DES. The excess concentration of CTAB N50 μM, dimin-
acidic pH, the complex of thorin-Th(IV) is stable, so the pH range of ished stability of the ternary complex and the absorbance was

(B)
(A) 0.550
0.550

0.488
0.488
Absorbance

Absorbance

0.425 0.425

0.362 0.362

0.300 0.300
65.0 4.0 80.0 65.0
55.0 3.5 67.5 55.0
45.0 3.0 55.0 45.0
35.0 2.5
B:Thorin Conc. (µM) A: pH C:CTAB Conc. (µM)42.5 35.0
B:Thorin Conc. (µM)
25.0 2.0 30.0 25.0

(C) (D)
0.550 0.550

0.488 0.488
Absorbance

Absorbance

0.425 0.425

0.362 0.362

0.300 0.300
200.0 80.0 200.0 65.0
175.0 67.5 175.0 55.0
150.0 55.0 150.0 45.0

D: DES volume (µL)125.0 42.5

C:CTAB Conc. (µM) D: DES volume (µL)125.0 35.0
B:Thorin Conc. (µM)
100.0 30.0 100.0 25.0

Fig. 5. The 3D central composite design curves for investigation of the effects of variables on analytical response: (A) Effect of the concentration of thorin and pH on the absorbance.
(B) Effect of the concentration of CTAB and concentration of thorin on the absorbance. (C) Effect of the DES volume and concentration of CTAB on the absorbance. (D) Effect of the DES
volume and concentration of thorin on the absorbance.
6 S. Sadeghi, A. Davami / Journal of Molecular Liquids 291 (2019) 111242

Table 2
Effect of interfering ions on the absorbance 250 ng mL−1 Th(IV) from the aqueous solution.

Ion Added as Tolerance limita Ion Added as Tolerance limita

Na+ NaNO3 1000 Cd2+ Cd(NO3)2.6H2O 800

K+ KNO3 1000 Co2+ Co(NO3)2.6H2O 800
Ca2+ Ca(NO3)2.4H2O 1000 SO2−
4 Na2SO4 700
Mg2+ Mg(NO3)2.6H2O 1000 c 4+
Zr ZrCl4 500
Ba2+ Ba(NO3)2 1000 MoO2− 4 Na2MoO4.2H2O 500
Al3+ Al(NO3)3.2H2O 1000 WO2− 4 Na2WO4.2H20 500
−
Mn2+ Mn(NO3)2 1000 VO3 NaVO3.H2O 400
Ni2+ Ni(NO3)2.6H2O 1000 d
Cu2+ Cu(NO3)2.3H2O 300
Pb2+ Pb(NO3)2 1000 Ta5+ TaCl5 300
Cr3+ Cr(NO3)3.9H2O 1000 Gd3+ Gd2O3 250
Fe2+ Fe(NO3)2.9H2O 1000 Er3+ Er2O3 250
Zn2+ Zn(NO3)2.6H2O 1000 Ho3+ Ho(NO3)3.6H2O 150
b
Fe3+ Fe(NO3)3.9H2O 1000 Ce3+ Ce(NO3)3.6H2O 120
NO− 3 NaNO3 1000 La3+ La(NO3)3.6H2O 120
Cl− NaCl 1000 UO22+ UO2(NO3)2.6H2O 70
Sr2+ Sr(NO3)2 800
a
Amount of foreign ion/amount of Th(IV).
b
Masked with ascorbic acid 0.005 M.
c
Masked with tartaric acid 0.002 M.
d
Masked with thiourea 0.002 M.

decreased (Fig. 5B). As aforementioned, the ternary complex of Th(IV) containing 250 ng mL−1 of Th(IV) and different concentrations of inter-
was hydrophobic and could extract from aqueous solution to the small fering ions were prepared and analyzed by the proposed method. The
volume of DES as an extractor phase. Thus, volume of DES affects the highest concentration of the added interfering ions that caused a varia-
concentration of the extracted complex. Fig. 5C shows the interaction tion of greater than ±5% on the absorbance of the analyte is defined as
of DES volume and CTAB concentration on the response. According to the tolerance limit. The results are available in Table 2. It can be ob-
Fig. 5C, with increasing the concentration of CTAB from 30 to 50 μM, served that a large excess of most cations and anions tested have no sig-
the absorbance raised up due to the complete formation of the ternary nificant effects on the determination of Th(IV). Zr4+ and Cu2+ ions
complex of Th(IV)-thorin-CTAB and more extraction to the DES phase didn't interfere in the extraction and determination of Th(IV) up to
in the volume range of 100–170 μL. At higher DES volumes over than 500 and 300 fold excess by using masking agents of tartaric acid and
170 μL, a slight decrease in the absorbance was observed because of rais- thiourea at 2 × 10−3 M, respectively. Interference of Fe3+ ions was elim-
ing the sedimented phase volume. Thereby, the optimum DES volume inated using ascorbic acid (5 × 10−3 M) in the recommended proce-
of 170 μL was set for the rest of experiments. Fig. 5D demonstrated dure. Uranium didn't interfere in the microextraction of Th(IV) up to
that increasing the concentration of thorin from 25 to 60 μM and 17,500 ng mL−1.
DES volume from 100 to 170 was necessary for the complete extrac-
tion of the ternary complex into the extractor phase. Based on the 3.6. Analysis of real samples
CCD results, the optimum conditions were: sample pH of 3.2, thorin
concentration 60 μM, concentration of CTAB 50 μM and DES volume To assess the applicability of the developed method, monitoring of
170 μL. The predicted response with the optimum conditions of var- Th(IV) in river and sea waters, under optimum conditions were evalu-
iables was 0.537. In order to confirm the predicted response, exper- ated. Since there was no Th(IV) detected in the samples, the waters
iments were repeated five times at the optimum conditions. The were spiked with Th(IV) at four concentration levels of 20, 50, 100
average response of Th(IV) with the confidence level of 95% was and 300 ng mL−1 and the analyses were performed in three replicates.
found to be 0.549 ± 0.006, that was in good agreement with the pre- The recoveries of Th(IV) ions in water samples were ranged in 92.5%–
dicted response. 109.4% (Table 3), indicating that the matrices of the water samples
had a negligible effect on the performance of the developed method.
3.4. Analytical figures of merit Also, the developed method was applied to find the amount of Th(IV)
in rock sample for triplicate analyses and assessed with the analysis by
The analytical performance of the established method was evaluated ICP-AES. The statistical significance of difference in the obtained results
in terms of the linearity, limit of detection (LOD) and precision. Under were evaluated by Fisher F-test and student t-test (Table 4). The com-
the optimal conditions, the calibration curve for the determination of parison of the precisions achieved by F-test indicated to not differ in
Th(IV) was linear in the concentration range from 10 to 600 ng mL−1 those obtained in a statistically significant manner. In addition, the
with the correlation coefficient of 0.9992 and LOD 2.1 ng mL−1 which
was calculated based on 3Sb/m (where Sb and m are the standard devi-
ation of the blank and the slope of calibration curve, respectively). Also,
Table 3
the precision of the method described as relative standard deviation Analytical results for Th(IV) monitoring in water sample.
(RSD%) for five replicate measurements of Th(IV) at 100 ng mL−1 was
Sample Added (ng mL−1) Founded (ng mL−1) Recovery %
1.7%. The preconcentration factor which was calculated from the ratio
of initial Th(IV) sample volume (10 mL) and the sedimented DES vol- River water 20 18.5 ± 1.5 92.5
50 50.3 ± 1.6 100.6
ume (170 μL) obtained 59.
100 95.3 ± 2.1 95.3
300 311.5 ± 2.9 103.8
3.5. Interference studies Seawater 20 20.1 ± 0.9 100.5
50 54.7 ± 2.3 109.4
The influence of possible interferent cations and anions on the 100 106.2 ± 2.8 106.2
300 320.0 ± 3.0 106.7
microextraction process was also investigated. Binary solutions
 S. Sadeghi, A. Davami / Journal of Molecular Liquids 291 (2019) 111242 7

Table 4 Appendix A. Supplementary data

Determination of Th(IV) in rock sample by the proposed method and comparison with the
ICP-AES method.
Supplementary data to this article can be found online at https://doi.
Sample Th(IV) found (μg g−1) Relative error (%) org/10.1016/j.molliq.2019.111242.
Proposed method ICP-AES method

Rock 55.51 ± 1.41 57.70 ± 0.90 3.79

References
t Crit. = 2.78; t Calc. = 2.26.
FCrit. = 39.0; FCalc. = 2.45. [1] A. Safavi, M. Sadeghi, Design and evaluation of a thorium (IV) selective optode, Anal.
Chim. Acta 567 (2006) 184–188, https://doi.org/10.1016/j.aca.2006.03.027.
[2] S. Rastegarzadeh, N. Pourreza, I. Saeedi, An optical chemical sensor for thorium (IV)
determination based on thorin, J. Hazard. Mater. 173 (2010) 110–114, https://doi.
org/10.1016/j. jhazmat.2009.08.055.
mean concentration of Th(IV) in rock sample achieved by the two [3] Z.S. Moghaddam, M. Kaykhaii, M. Khajeh, A.R. Oveisi, Synthesis of UiO-66-OH zirco-
methods was compared by t-test. The calculated t-value at 95% confi- nium metal-organic framework and its application for selective extraction and trace
dence level, 2.45, was lower than the critical t-value (Table 4), meaning determination of thorium in water samples by spectrophotometry, Spectrochim.
Acta A Mol. Biomol. Spectrosc. 194 (2018) 76–82, https://doi.org/10.1016/j.saa.
that there was no significant priority between the two available
2018.01.010.
methods. Thus, the proposed method was suitable for determination [4] P.K. Tarafder, S.K. Pradhan, R.K. Mondal, Separation of thorium by facile liquid–
of Th(IV) in rock sample. liquid extraction, and its rapid spectrophotometric and ICP-AES determination in
rocks and minerals, J. Radioanal. Nucl. Chem. 309 (2016) 1021–1028, https://doi.
org/10.1007/s10967-016-4738-0.
3.7. Comparison with other methods [5] S. Veyseh, A. Niazi, Flotation-assisted homogeneous liquid-liquid microextraction
for determination of thorium in water samples by inductively coupled plasma
The suggested method was compared with some reported mass spectrometry and Box-Behnken design, Sep. Sci. Technol. 52 (2017)
1289–1297, https://doi.org/10.1080/01496 395.2017.1287200.
extraction methods of Th(IV) based on UV–Vis. spectrophotometric [6] B.F. Liu, L.B. Liu, J.K. Cheng, Separation and determination of thorium, uranium and
determination [13,30–33] and the results are represented in Table 5. It mixed rare-earth elements as their UV/Vis absorbing complexes by capillary zone
is obvious that the developed method has a wide linear range, electrophoresis, Talanta 47 (1998) 291–299, https://doi.org/10.1016/S0039-9140
(98)00137-4.
comparatively low detection limits, high precision, and suitable [7] V.V. Raut, S. Jeyakumar, M.K. Das, A. Chandane, B.S. Tomar, Separation and determi-
preconcentration factor. As well, this method is a rapid, simple, selec- nation of trace thorium in uranium matrix using chelation ion chromatography, Sep.
tive, sensitive and green microextraction method that doesn't use Sci. Technol. 52 (2017) 2356–2362, https://doi.org/10.1080/01496395.2016.
1262420.
heating or ultrasonic treatment to accelerate and shorten the extraction [8] N.L. Misra, S. Dhara, V.C. Adya, S.V. Godbole, K.D. Singh Mudher, S.K. Aggarwal, Trace
process which distinguishes it from other microextraction methods. element determination in thorium oxide using total reflection X-ray fluorescence
spectrometry, Spectrochim. Acta B 63 (2008) 81–85, https://doi.org/10.1016/j.sab.
2007.11.021.
3.8. Conclusions
[9] F.A. Aydin, M. Soylak, Solid phase extraction and preconcentration of uranium (VI)
and thorium(IV) on Duolite XAD761 prior to their inductively coupled plasma
For the first time, a combination of a selective and straightfor- mass spectrometric determination, Talanta 72 (2007) 187–192, https://doi.org/10.
ward DLLME method based on using hydrophobic DES for 1016/j.talanta.2006.10.013.
[10] L. He, Q. Jiang, Y. Jia, Y. Fang, S. Zou, Y. Yang, J. Liao, N. Liu, W. Feng, S. Luo, Y. Yang, L.
microextraction of Th(IV) from water and rock samples coupled Yang, L. Yuan, Solvent extraction of thorium(IV) and rare earth elements with novel
with UV–Vis. spectrophotometric detection was developed. The hy- polyaramide extractant containing preorganized chelating groups, J. Chem. Technol.
drophobic DES was synthesized from [HMIM]Br and SA that make Biotechnol. 88 (2013) 1930–1936, https://doi.org/10.1002/jctb.4055.
[11] N. Jalbani, M. Soylak, A.B. Munshi, T. Gul Kazi, Multivariate optimization of parame-
it potential alternative to replace with the chlorinated organic sol- ters for the determination of thorium in rock samples by cloud point extraction
vent as an extraction solvent. Surface response methodology based coupled to UV-Visible spectrophtotometry, Fresenius Environ. Bull. 23 (2014)
on CCD was successfully employed for the simultaneous investiga- 2304–2309.
[12] S. Shiri, A. Delpisheh, A. Haeri, A. Poornajaf, T. Khezeli, N. Badkiu, Floatation-
tion of effective variables on the response. The proposed method is spectrophotometric determination of thorium using complex formation with
rapid, easy, sensitive, cost effective and efficient for separation and eriochrome cyanine R, Anal. Chem. Insights 6 (2011) 1–6, https://doi.org/10.4137/
preconcentration of thorium. The performance of this method was ACI.S5949.
[13] A.A. Gouda, M.S. Elmasry, H. Hashem, H.M. El-Sayed, Eco-friendly environmental
tested for extraction and determination of Th(IV) in spiked water
trace analysis of thorium using a new supramolecular solvent-based liquid-liquid
and rock samples and the obtained results demonstrated the method microextraction combined with spectrophotometry, Microchem. J. 142 (2018)
is accurate, reliable and suitable for the determination of Th(IV) in 102–107, https://doi.org/10.1016/j.microc.2018.06.024.
[14] J. Gao, B.T. Manard, A. Castro, D.P. Montoya, N. Xu, R.M. Chamberlin, Solid-phase ex-
environmental samples at trace level.
traction microfluidic devices for matrix removal in trace element assay of actinide
materials, Talanta 167 (2017) 8–13, https://doi.org/10.1016/j.talanta.2017.01.080.
Acknowledgments [15] E. Yilmaz, M. Soylak, Latest trends, green aspects, and innovations in liquid-phase
based microextraction techniques: a review, Turk. J. Chem. 40 (2016) 868–893,
https://doi.org/10.3906/kim-1605-26.
Authors are grateful to the Research Council of the University of [16] M. Rezaee, Y. Assadi, M.R. Milani Hosseini, E. Aghaee, F. Ahmadi, S. Berijani, Determi-
Birjand for supporting this work. nation of organic compounds in water using dispersive liquid–liquid

Table 5
A comparison between the analytical data of the suggested method with the other extraction methods based on UV–Vis spectrophotometry for the determination of Th(IV) ion.

Preconcentration technique Extraction solvent LOD Linear range RSD Extraction time (min) Ref.
(ng mL −1) (ng mL −1) (%)

SsLLMEa 1-decanol/THF 0.40 20–100 1.4 5 [13]

VALLMEb Chloroform 7.50 20–240 2.8 3 [30]
CPEc Triton X-114 – 10–1000 2-5 130 [31]
CPE Triton X-100 0.16 0.5–15 1.6 5 [32]
CPE Triton X-114 1.40 5–100 1.8 15 [33]
DES-DLLME [HMIM]-SA 2.10 10–600 1.7 1 This work
a
Supramolecular solvent-based liquid-liquid microextraction.
b
Vortex-assisted liquid–liquid microextraction.
c
Cloud point extraction.
8 S. Sadeghi, A. Davami / Journal of Molecular Liquids 291 (2019) 111242

microextraction, J. Chromatogr. A 1116 (2006) 1–9, https://doi.org/10.1016/j. [25] M.H. Habibollahi, K. Karimyan, H. Arfaeinia, N. Mirzaei, Y. Safari, R. Akramipour, H.
chroma.2006.03.007. Sharafi, N. Fattahi, Extraction and determination of heavy metals in soil and vegeta-
[17] E. Carasek, J. Merib, G. Mafra, D. Spudeit, A recent overview of the application of bles irrigated with treated municipal wastewater using new mode of dispersive
liquid-phase microextraction to the determination of organic micro-pollutants, liquid-liquid microextraction based on the solidified deep eutectic solvent followed
TrAC Trends Anal. Chem. 108 (2018) 203–209, https://doi.org/10.1016/j.trac.2018. by GFAAS, J. Sci. Food Agric. 99 (2019) 656–665, https://doi.org/10.1002/jsfa.9230.
09.002. [26] R. Akramipour, M.R. Golpayegani, S. Gheini, N. Fattahi, Speciation of organic/inor-
[18] F. Heydari, M. Ramezani, Micro-funnel magnetic stirring-assisted liquid–liquid ganic mercury and total mercury in blood samples using vortex assisted dispersive
microextraction technique combined with UV–Vis spectrophotometry for determi- liquid-liquid microextraction based on the freezing of deep eutectic solvent
nation of thorium in aqueous samples with the aid of experimental design, J. followed by GFAAS, Talanta 186 (2018) 17–23, https://doi.org/10.1016/j.talanta.
Radioanal. Nucl. Chem. 320 (2019) 27–36, https://doi.org/10.1007/s10967-019- 2018.04.042.
06452-6. [27] S. Zhu, J. Zhou, H. Jia, H. Zhang, Liquid–liquid microextraction of synthetic pigments
[19] I. Billard, A. Ouadi, C. Gaillard, Liquid–liquid extraction of actinides, lanthanides, and in beverages using a hydrophobic deep eutectic solvent, Food Chem. 243 (2018)
fission products by use of ionic liquids: from discovery to understanding, Anal. 351–356, https://doi.org/10.1016/j.foodchem.2017.09.141.
Bioanal. Chem. 400 (2011) 1555–1566, https://doi.org/10.1007/s00216-010-4478-x. [28] H. Ebrahimi-Najafabadi, R. Leardi, M. Jalali-Heravi, Experimental design in analytical
[20] M.M. Khataei, Y. Yamini, A. Nazaripour, M. Karimi, Novel generation of deep eutectic chemistry—part I: theory, J. AOAC Int. 97 (2014) 3–11, https://doi.org/10.5740/
solvent as an acceptor phase in three-phase hollow fiber liquid phase jaoacint.SGEE brahimi 1.
microextraction for extraction and preconcentration of steroidal hormones from bi- [29] S. Sadeghi, Z. Nasehi, Simultaneous determination of Brilliant Green and Crystal Vi-
ological fluids, Talanta 178 (2018) 473–480, https://doi.org/10.1016/j.talanta.2017. olet dyes in fish and water samples with dispersive liquid-liquid micro-extraction
09.068. using ionic liquid followed by zero crossing first derivative spectrophotometric
[21] D.J.G.P. Van Osch, L.F. Zubeir, A. Van den Bruinhorst, M.A.A. Rocha, M.C. Kroon, Hy- analysis method, Spectrochim. Acta A Mol. Biomol. Spectrosc. 201 (2018)
drophobic deep eutectic solvents as water-immiscible extractants, Green Chem. 134–142, https://doi.org/10.1016/j.saa.2018.04.061.
17 (2015) 4518–4521, https://doi.org/10.1039/C5GC01451D. [30] E. Zolfonoun, M. Salahinejad, Preconcentration procedure using vortex-assisted
[22] M.B. Arain, E. Yilmaz, M. Soylak, Deep eutectic solvent based ultrasonic assisted liq- liquid–liquid microextraction for the fast determination of trace levels of thorium
uid phase microextraction for the FAAS determination of cobalt, J. Mol. Liq. 224 in water samples, J. Radioanal. Nucl. Chem. 298 (2013) 1801–1807, https://doi.
(2016) 538–543, https://doi.org/10.1016/j.molliq.2016.10.005. org/10.1007/s10967-013-2684-7.
[23] R.A. Zounr, M. Tuzen, N. Deligonul, M.Y. Khuhawar, A highly selective and sensitive [31] A. Saha, S.B. Deb, A. Sarkar, M.K. Saxena, B.S. Tomar, Simultaneous preconcentration
ultrasonic assisted dispersive liquid phase microextraction based on deep eutectic of uranium and thorium in aqueous samples using cloud point extraction, RSC Adv.
solvent for determination of cadmium in food and water samples prior to electro- 6 (2016) 20109–20119, https://doi.org/10.1039/c5ra23734c.
thermal atomic absorption spectrometry, Food Chem. 253 (2018) 277–283, [32] A.S. Amin, M.A. Kassem, T.Y. Mohammed, Utilization of cloud-point extraction for
https://doi.org/10.1016/j.foodchem.2018.01.167. colorimetric determination of trace amounts of thorium(IV) in real sample, RSC
[24] R.A. Zounr, M. Tuzen, M.Y. Khuhawar, Ultrasound assisted deep eutectic solvent Adv. 5 (2015) 52095–52100, https://doi.org/10.1039/C5RA08806B.
based on dispersive liquid liquid microextraction of arsenic speciation in water [33] S.S.A. El-Hay, A.A. Gouda, Determination of thorium(IV) in real samples by spectro-
and environmental samples by electrothermal atomic absorption spectrometry, J. photometry after micelle-mediated cloud point extraction, J. Radioanal. Nucl. Chem.
Mol. Liq. 242 (2017) 441–446, https://doi.org/10.1016/j.molliq.2017.07.053. 310 (2016) 191–200, https://doi.org/10.1007/s10967-016-4780-y.