Adsorption of U(VI) from acid solution on a low-cost sorbent:equilibrium, kinetic, and thermodynamic assessments

Mohamed F. Cheira · Hamed I. Mira · Ahmed K. Sakr · Saad A. Mohamed

Abstract In this paper, waste clay was cured with ethyl acetate to obtain treated clay (TC), which was modified with gallic acid to obtain a low-cost sorbent that was characterized by EDX, SEM, and FTIR analysis. Uranium(VI) adsorption was achieved using the batch adsorption method on the TC and gallic acid-modified treated clay (GMTC). The maximum uptakes of U(VI) on TC and GMTC were 37.2 and 193.0 mg/g, respectively.The U(VI) adsorption kinetics on the TC and GMTC sorbents were well-fitted by the pseudo-second-order mechanism, and the adsorption equilibrium followed the Langmuir model.The optimum parameters were applied to El Sela leach solution for uranium recovery.

Keywords Waste clay · Gallic acid · Modification ·Uranium adsorption · Equilibrium kinetics

1 Introduction

Clay is utilized in the edible oil refining process to remove colored materials, pigments, soap, gums, trace metal ion impurities, and oxidizing composites, as well as other unwanted deposits, leaving the purified oil. It can be used in the refining process not only for the superficial removal of color, but also as a tool to remove unwanted toxic wastes like heavy metal ions and pesticides [1].Waste clay has a brownish color, and contains about 20-37% oil deposits. It is noted that waste clay not only presents a fire hazard (spontaneous auto-ignition), but also has an unpleasant odor. Accordingly, the landfilling of waste clay may lead to severe management and packing problems [2].

Uranium is one of the crucial elements.It represents the cornerstone in the scope of nuclear energy, and is utilized in nuclear power plants as a fuel[3].The pre-concentration and separation of uranium on different adsorbents are effective alternative methods for its recovery from pregnant solutions.Adsorption techniques have been applied to the separation of uranium in dilute solutions. Many sophisticated synthetic adsorbent materials have recently been developed for the adsorption of uranium from aqueous solutions.Tributylphosphate-coated magnetic Pst-DVB particles were prepared for uranium separation [4]. Amidoxime-functionalized ultra-high molecular weight polyethylene fibers were used to absorb uranium U(VI) from nitrate solutions [5].

An amidoxime-based fibrous adsorbent denoted as PE/PP-g-(PAAc-co-PAO) was prepared by pre-irradiation grafting of acrylic acid and acrylonitrile onto polyethylenecoated polypropylene skin-core (PE/PP) fibers using60Co γ-ray irradiation, followed by amidoximation. The prepared amidoxime-based PE/PP fibers were applied for the extraction of uranium from seawater [6]. Amidoximebased adsorbents prepared by co-grafting acrylic acid with acrylonitrile onto HDPE fibers were also used for the recovery of uranium from seawater [7]. The adsorption of uranium(VI) ions from aqueous solutions by diethylethanolammonium organovolcanics was studied under different experimental conditions [8].

In recent times, clay has become a vital mineral to be considered in uranium recovery due to the proposed utilization of bentonite or clay minerals as adsorbent materials in wastewater. Thermal- and chemical-modified bentonite(TCMB)was applied for uranium adsorption from aqueous waste solution, and the theoretical capacity of TCMB was about 29 mg/g [9]. The retention of uranyl ions on clay modified with titanium oxide was investigated, and it was established that the retention increased with an increase in contact time,and the uranium sorption reached equilibrium after 5 days of contact time [10]. Kaolin was modified by calcination followed with acid activation and applied for U(VI) adsorption [11].

The sorption of uranium on hexadecyltrimethylammonium-modified bromide bentonite was studied, and increased uranium sorption was obtained in the pH range 3-6 [12]. An aminopropyl-modified mesoporous material(NH2-MCM-41) was used for uranium adsorption from solutions at pH 2-4 and 60 °C with 173 min of shaking time [13]. U(VI) sorption was achieved from polluted groundwater within 1 h and 20 h by applying nanoporous and non-nanoporous Al2O3, respectively, and the sorption capacities of the nanoporous and non-nanoporous Al2O3were 28 mmol/g and 18 mmol/g, respectively [14].

Various considerations were examined in the surface speciation of adsorbed U(VI)on montmorillonite[15].The adsorption of uranyl ions on humic acid, montmorillonite,kaolinite, and composite clay materials (both humic acid and clays)was investigated by gauging the response of the system to clay suspensions [16]. The U(VI) sorption mechanism and surface complexation scheme on montmorillonite were appraised in the absence and presence of carbonate in macroscopic sorption tests [17]. The adsorption of uranium(VI)using humic and kaolinite associates as the adsorbent in batch procedures was studied [18]. U(VI)was adsorbed on clay in the absence and presence of humic acid [19]. U(VI) extraction using natural kaolinite and N-[3-(trimethoxysilyl)-propyl] diethylenetriamine-modified kaolinite was also examined [20].

Gallic acid (3,4,5-trihydroxy benzoic acid) was used to improve the uranium adsorption capacity of the treated clay because it has three hydroxyl groups as well as one carboxyl group, and thus, it has an immense ability to bond with cations.This study aims to investigate U(VI)sorption after modification of the treated clay by gallic acid. Ethyl acetate was utilized to reduce the oil residues in the waste clay.The treated clay and gallic acid-modified treated clay were used to extract U(VI)from a sulfate solution.The pH,contact time,uranium concentration,adsorbent dosage,and temperature were investigated. Additionally, the kinetics,equilibrium, and thermodynamics of the adsorption were evaluated, along with the optimum conditions for uranium extraction from El Sela leach solution.

2 Materials and methods

2.1 Materials

A synthetic uranium solution of 1000 mg/L U(VI) was prepared by melting 1.782 g uranyl acetate (BDH Chem.,England) in 1.0 L distilled water acidified by 10 mL concentrated H2SO4. This solution was primarily utilized to measure the associated parameters of uranium adsorption.

X-Ray diffraction (XRD) patterns of the treated waste clay were recorded using a Philips instrument (PW 3710/31), with a diffractometer (PW 1775, 21) position,scintillation counter, and Cu-target tube. The morphological structures of all the types of clay were examined using an environmental scanning electron microscope (SEM,Philips XL 30). Furthermore, Fourier transform infrared spectroscopy (FTIR, Shimadzu IR Prestige-21) controlled by IR resolution software was employed using the KBr tablet method to evaluate the functional groups of all the clay types.

2.2 Waste clay treatment

To obtain the treated clay, various experiments were performed individually using different concentrations of ethanol or ethyl acetate for removing the oil content from the studied waste clay, which was supplied by Alexandria Co. for oils and soap, Egypt. The experiments were performed on a particular weight of waste clay using mixtures of ethanol and water with different ethanol concentrations(10-90%),or a blend of ethyl acetate and distilled water at an agitation speed of 50-300 rpm for contact times varying from 15 to 120 min. These parameters are explained in detail in the Supplementary Materials (Figure S1). The removal efficiency of the impurities was calculated using the following equation:

where Wiand Wfare the original and final weight of the waste and treated clay (mg), respectively.

2.3 Impregnation of treated clay

A dry system was operated for the modified process. A series of experiments were performed to optimize the modified conditions. 5 g of the treated clay was suspended in 10 mL deionized water, then mixed well with different amounts (0.1-1 g) of gallic acid, and then dissolved in 50 ethanol with stirring for a fixed contact time(10 min)until complete homogenization occurred. The slurry was then allowed to stand until the solution evaporated. The modified treated clay was then dried (50 °C) to complete dryness.The powder form of gallic acid-modified treated clay(GMTC) was obtained using an optimized mass ratio of 0.4 g gallic acid to 5 g treated clay.

2.4 Adsorption experiments

In the present work, the treated clay (TC) and gallic acid-impregnated treated clay (GMTC) were examined individually for uranium sorption from standard and leach solutions using the batch method. A series of U(VI) sorption experiments were carried out to obtain the optimum conditions for the adsorbent dose, pH, initial uranium concentration, and contact time. The sorption experiments were carried out by mixing 50 mg of either TC or GMTC with 50 mL standard solutions of different initial uranium concentrations in 200 mL beakers using a mechanical shaker at 200 rpm and contact times varying from 5 to 180 min at different temperatures. The pH was studied over the range from 1 to 5.5,and the pH was adjusted using 1 M NaOH or 1 M H2SO4solutions. The adsorbent dose was also studied in the 10-400 mg range.Additionally,the uranium equilibrium isotherms, sorption kinetics, and thermodynamic parameters were obtained from the corresponding experiments.

The sorption capacity (qe,mg/g), adsorption efficiency(E, %), and distribution coefficient (Kd) were evaluated from the following equations:

where Coand Ce(mg/L) are the original and equilibrium U(VI)concentrations,respectively,and V(L)and m(g)are the solution volume and dry adsorbent weight,respectively.

2.5 Desorption studies

The uranium elution was performed from the uraniumloaded adsorbent obtained as previously mentioned. The desorption procedure was carried out using different amounts of the uranium-loaded adsorbent and different concentrations of the eluting agents at room temperature for different contact times (15-120 min); the solution was then filtered, and the concentration of the uranium in the eluate was analyzed with the appropriate technique.

3 Results and discussion

3.1 Description of the studied clay

3.1.1 Clay chemical analysis

The used waste clay was supplied by Alexandria Co.for oils and soap. The analytical constitutions of natural Chinese clay used in oil refining in Egypt, waste clay, and treated clay,as well as that of GMTC,are given in Table 1.The high concentration of silicon oxide and aluminum oxide confirmed that the clay was an alumino-silicate material; however, the high-silica montmorillonite was enriched with different alkali and transition metals. The isomorphic replacement of Mg2+to Al3+and Al3+to Si4+gave a negative charge at the clay surface. Based on the datasheet of natural Chinese clay, its pH is 3.5, indicating that it is considered an acidic adsorbent [21].

Structurally, clay is constructed of two basic structural blocks, i.e., octahedral alumina and tetrahedral silica sheets. A single cell contains one Al2O3octahedral sheet packed between two SiO2tetrahedral sheets. The silicate sheets have a slightly negative charge that is compensated by the complementary cations in the middle layers [21].The charge is so weak that the cations (Na+, Mg2+, or Ca2+ions) can be adsorbed with an associated hydration shell.The cations held in this pattern on the clay can easily be replaced by ion exchange.The existence of MgO,Na2O,CaO, and K2O indicated that Mg2+, Na+, Ca2+, and K+were the principal exchangeable cations during the activation process. Additionally, the gallic acid-modified treated clay (GMTC) had increased amounts of hydroxyl and carbonyl groups, which combined with uranium ions.

3.1.2 Surface area

The specific surface areas of the natural Chinese clay,waste clay, and treated waste clay were evaluated by application of the BET equation to N2adsorption isotherms obtained using a sorptiometer (Quantachrome Co., NOVA 2000). The natural Chinese clay had a surface area of 9.83 m2/g, whereas the surface area of the waste clay and treated clay was found to be 2.85 m2/g and 5.91 m2/g,respectively. It appeared that the treatment process cleared the pores of the waste clay particles of impurities, thereby increasing its pore size and surface area. Furthermore, the surface area according to the BET equation for the treated clay modified with gallic acid was 8.21 m2/g. The impregnation process increased the surface area of the treated waste clay;thus,it increased its uranium adsorption capacity.

Table 1 Analysis of natural Chinese, waste, and treated clay, as well as GMTC

3.1.3 XRD analysis

The XRD pattern of the treated clay is demonstrated in Fig. 1. The obtained results revealed that the treated clay was composed mainly of montmorillonite (Na,Ca)0.3(-Al,Mg)2Si4O10(OH)2·n(H2O) with small quantities of quartz SiO2and cristobalite SiO2. The XRD spectrum of the TC shows the peaks of montmorillonite (2θ = 8.85,d = 9.98 A˚;2θ = 19.73°d = 4.49 A˚;2θ = 26.67°d = 3.34 A˚)[22].It can be deduced that the removal of the edible oil during the refining process and the treatment of the waste clay with ethyl acetate did not change the principal structure of the treated and natural Chinese clay.

3.1.4 SEM analysis

Fig. 1 XRD configuration of the treated clay

The SEM images of the natural Chinese, waste, and treated clay, as well as that of the gallic acid-modified treated clay and uranium-loaded gallic acid-modified treated clay, are presented in Fig. 2. As shown in Fig. 2a,the natural Chinese clay layers were separated,and it had a large facade area, more irregular structure, and more porous nature. The SEM image in Fig. 2b demonstrates that the external surface of the waste clay particles was covered by oil-like materials,and was almost non-porous.The SEM image in Fig. 2c shows the roughness and irregular layer structure of the clay treated with ethyl acetate, whose surface structure resembles that of the natural Chinese clay surface.

Fig. 2 SEM images of the surfaces of the a natural Chinese clay,b waste clay,c treated clay,d gallic acid-modified treated clay,and e uraniumloaded gallic acid-modified treated clay

Moreover,the surface of the gallic acid-modified treated clay was observed using SEM to demonstrate the change in its surface features after modification and loading with uranium(VI).The SEM images in Fig. 2d of the surface of the gallic acid-modified treated clay show that the rough surface became relatively bright after modification with gallic acid, which filled in most of the vacant pores. Uranium(VI) appears as brilliant spots on the surface of the modified treated clay in Fig. 2e, which confirms its adsorption on the surface of GMTC.

3.1.5 FTIR investigation

FTIR is a vital analysis technique that reveals the various characteristic functional groups existing in the studied adsorbents. The FTIR spectra of the natural Chinese clay,waste clay, and treated clay were recorded to acquire information regarding the wavenumber changes corresponding to the functional groups.Generally,the spectrum of clay shows two important groups (OH and Si-O). The spectra of all the samples(Fig. 3)show three intense peaks,one at 1672 cm-1,which was associated with the presence of structural water within the montmorillonite matrix, and two at 1072 and 790 cm-1, which were attributed to Si-O stretching vibrations, indicating the presence of silicate.The peaks at 612 and 470 cm-1corresponded to Al-O-Si and Si-O-Si bending vibrations. In addition, the band observed at 3433 cm-1was assigned as the OH stretching band [23].

Furthermore, the spectrum in Fig. 3b illustrates the presence of unsaturated residual oil and organic impurities in the waste clay,exhibiting peaks at 1631 and 1458 cm-1assigned to C=O and C=C or OH bonds. Moreover, a strong OH stretching band appeared at 3432 cm-1. The spectrum in Fig. 3c exhibits an intense band at 1065 cm-1,which was attributed to the Si-O stretching vibration, and two bands were observed at 790 and 470 cm-1, corresponding to amorphous silica [24]. The spectrum of the treated clay was very similar to the spectrum of the natural Chinese clay, which confirmed that the chemical structure of both the treated clay and natural Chinese clay was very similar and indicating that the organic impurities were almost complete removed.

The spectrum of the treated clay modified with gallic acid is given in Fig. 3d.The main difference observed was the appearance of two peaks at 1650 and 1455 cm-1,which were assigned as C=O and C=C bands. The obtained data in Fig. 3e showed that the main differences between the spectra of the treated clay modified with gallic acid in the absence of uranium ions and the presence of uranium ions were the appearance of the U=O stretching band at ～924 cm-1and also two weak U-O bands at ～475 and ～415 cm-1[25].

Fig. 3 FTIR spectrums of the a natural Chinese clay, b waste clay,c treated clay,d gallic acid modified treated clay, and e loaded uranium on gallic acid modified treated clay

Finally, the suggested mechanisms of the treatment and modification of the clay as well as that of the adsorption of uranium ions on the gallic acid-modified treated clay are illustrated in Scheme 1.

3.2 Adsorption experiments

3.2.1 pH

The influence of pH on the U(VI) adsorption onto TC and GMTC was studied at different pH values from 1 to 5.5 at fixed experimental conditions of 50 mg of the adsorbent and 50 mL of synthetic solution assaying 200 mg/L of U(VI) for a contact time of 60 min at 25 °C. From the obtained data in Fig. 4a, the uranium adsorption efficiencies increased from 5.44 and 30.23 to 18.6 and 96.95%with increasing the pH value of the solution from 1.0 to 4.5 for TC and GMTC, respectively. Subsequently, increasing the range of pH to 5.5 led to a decrease in the uranium adsorption efficiencies to 6.75 and 49.35% for TC and GMTC, respectively. Thus, the maximum uranium adsorption efficiency was attained in a sulfate solution of pH 4.5.

At high acidity(pH ＜4.5),the clay surface will become covered with H3O+ions, and the active oxygen atoms of the carbonyl and hydroxyl groups of gallic acid,as well as the oxygen atoms of the aluminol and silanol groups on the GMTC and TC adsorbent surfaces, will be partially protonated. The favorable electrostatic effect between the active sites with positive hydrogen ions and the complex anionic species [UO2(SO4)22-and UO2(SO4)34-] was slightly reduced because of the bisulfate ions (HSO4-) that competed with the uranium anionic species and some neutral complexes formed by the reaction of some positive hydrogen ions with anionic uranium species [26].

In the solution at pH 4.5, the uranium existed in hydrolyzed form, and the following cationic species were identified: UO22+, the [(UO2)2(OH)2]2+dimer, and the[(UO2)3(OH)5]+trimer. The active sites at the montmorillonite edges were recognized to be silanol (Si-OH) and aluminol (Al-OH) sites, which form the chief components of this type of clay (SiO2and Al2O3). Each crystal of treated clay had a substantial negative net charge due to the release of the hydrogen cation of the hydroxyl group [27].Also,the deprotonation of the oxygen atoms of GMTC and the formation of various uranium hydrolysis products led to high adsorption due to the electrostatic attraction and complexation system. This led to an increase in the uranium adsorption efficiency to the maximum value.

In contrast,the adsorption uptake at pH＞4.5 decreased gradually as a result of the gradual precipitation of the uranyl hydroxide species UO2(OH)2, [UO2(OH)3]-,[UO2(OH)4]2-, [(UO2)3(OH)7]-, [(UO2)3(OH)8]2-, and[(UO2)3(OH)11]5-[28]. Therefore, it was not possible to perform the pH effect experiment at pH values above 5.5.

3.2.2 Contact time

The adsorption efficiencies of U(VI)were established on the two adsorbents (TC and GMTC) for different contact times (5-120 min) with experimental settings of pH 4.5,50 mL of 200 mg/L U(VI) aqueous solution, and a 50 mg dose at room temperature. As presented in Fig. 4b, the adsorption efficiencies of uranium increased to 18.6 and 96.75% for the TC and GMTC adsorbents, respectively,with increase in the contact time from 5 min until equilibrium was reached after about 60 min.Further,increasing the contact time to 120 min had no remarkable effect on the U(VI) adsorption efficiencies. Given these results, a contact time of 60 min was used for further optimizing the conditions for uranium ion adsorption using the TC and GMTC sorbents.

Scheme 1 Proposed mechanism of uranium ion adsorption on gallic acid-modified treated clay

3.2.3 Adsorbent dose

Figure 4c depicts the influence of adsorbent dosage on the uranium adsorption efficiency. In this study, the effect of the adsorbent dose on uranium adsorption was studied using the sorbents TC and GMTC in the dose range from 10 to 400 mg using a fixed pH of 4.5, 60 min of contact time, and 50 mL of solution assaying 200 mg/L of U(VI)ions at room temperature. The obtained results revealed that the U(VI) adsorption efficiencies increased with an increase in the amount of the TC and GMTC sorbents. At higher doses, more active sites were available for ion exchange due the increased surface area. Hence, the adsorption efficiencies of U(VI) increased from 3.7 and 30.3% to 99.10 and 99.6% with an increase in the adsorbent doses from 10 to 300 mg for treated clay (TC)and from 10 to 75 mg for gallic acid-modified treated clay(GMTC), respectively. However, above these concentrations, the adsorption efficiencies remained constant up to 400 mg. The corresponding maximum uptakes of the TC and GMTC adsorbents decreased from 37.2 and 193.0 mg/g to 22.3 and 145.2 mg/g at 300 and 75 mg and to 18.98 and 24.98 mg/g at 400 mg adsorbent doses, respectively.Thus,the 50 mg adsorbent dose was selected for use in the subsequent tests.

Fig. 4 Effect of the a pH,b contact time,c adsorbent dose,and d initial uranium concentration on the uranium adsorption efficiencies using TC and GMTC as adsorbents

3.2.4 Initial uranium ion concentration

Several batch experiments were performed to study the effect of varying the initial U(IV) concentration on the adsorption efficiency of U(IV) using the TC and GMTC sorbents. These experiments were carried out by shaking 50 mL of a standard solution of uranium ions with concentrations in the range of 25-800 mg/L at pH 4.5 with 50 mg of either TC or GMTC for a contact time of 60 min at 25 °C. The results plotted in Fig. 4d reveal that as the initial concentration of uranium ions increased, the uranium uptake (qe, mg/g) increased, and the maximum loading was obtained at a 200 mg/L initial uranium concentration. The maximum adsorption efficiencies at 200 mg/L initial uranium concentration for the TC and GMTC sorbents were 18.60 and 96.50%, respectively.Therefore,the maximum uranium loading capacities on TC and GMTC were 37.2 and 193.0 mg/g,respectively.Above 200 mg/L, the amount of loaded U(VI)remained constant,indicating that the two sorbents had reached their maximum uptake capacities (saturation capacity) because the mobility of the uranyl ions in the solutions was the highest,and the active sites were filled and blocked by uranyl ions from the solution.

3.2.5 Temperature

The adsorption efficiency and uptake of uranium ions was studied on the TC and GMTC sorbents within the range 25-55 °C while other experimental factors were kept constant, with a solution pH of 4.5, 50 mg of the two sorbents, 50 mL of aqueous solution containing 200 mg/L of U(VI), and 60 min of contact time. As shown in the results displayed in Table 2, the U(VI) adsorption efficiency decreased slightly from 18.60 and 96.50% to 13.48 and 92.20% as the temperature was increased from 25 to 55 °C, and the uptake capacities of the TC and GMTC sorbents decreased from 37.2 and 193.0 mg/g at 25 °C to 26.9 and 184.4 mg/g at 55 °C, respectively. This was seemingly due to the nature of the adsorbents.As they were previously treated and modified by ethyl acetate and gallic acid, the energy led to the decomposition of any organic materials on clay that might associate with the uranium ions into solution and decreased the surface activity.Astonishingly, this result implied that the reaction was not related to energy, and was an exothermic reaction.Accordingly, room temperature was the optimum temperature for the extraction of uranium ions using the low-cost sorbents TC and GMTC.

Table 2 Effect of temperature on uranium adsorption efficiency and uptake using TC and GMTC as sorbents

3.2.6 Adsorption kinetics

The adsorption kinetics are standard critical data for determining the kinetics model of the extraction and appraising the action of the adsorbents. Different kinetics models, including Lagergren’s pseudo-first-order, pseudosecond-order, Elovich, and intraparticle models, were applied to the empirical data to fit the uranium adsorption kinetics on treated clay and treated clay modified with gallic acid.The expression for the pseudo-first-order model is given as [28]:

where qeand qtare the U(VI) uptake at equilibrium and time t (mg/g), and the pseudo-first-order constant is k1(min-1). The relationship of log (qe-qt) against t can be used to evaluate the applicability of the kinetic model and to evaluate the rate constant k1and qe(Fig. 5a and Table 3). The obtained data demonstrated that the adsorption mechanism of the two adsorbents did not fit the pseudo-first-order model.

The pseudo-second-order model is given in the following equation [29]:

where k2(g/mg min) is the rate constant, qt(mg/g) is the amount of uranium ion uptake at time t(min),and qeis the U(VI)uptake at equilibrium(mg/g).The plots of t/qtversus time should provide straight lines(Fig. 5b).From the data,the plots of the adsorbents TC and GMTC were straight lines with high correlation coefficients (0.995 and 0.996,respectively) that were close to unity (Table 3). Additionally, the values of the equilibrium adsorption amount obtained from this model (qcal) were close to those obtained from the experiments (qexp). These findings implied that the adsorption of uranium ions on the adsorbents TC and GMTC followed pseudo-second-order kinetics well.

The Elovich model represents adsorption of a chemical nature, and is generally given as [30]:

where α is the initial rate of adsorption(mg/g min)and β is the Elovich constant (g/mg). The plot of qtversus ln(t) should yield a linear relationship with a slope of 1/β and an intercept of 1/βlnαβ (Fig. 5c). The Elovich kinetic parameters α and β and the correlation coefficient are also given in Table 3.The coefficients α and β are related to the chemisorption rate and surface coverage,respectively.The obtained data confirmed that the Elovich model did not fit the experimental data well.

The intraparticle model depicts a linear correlation among the adsorbed amount of uranium ions at time t (qt)versus the square root of time(t0.5),and expressed as[31]:

where qtis the amount of uranium ions adsorbed at time t(mg/g),Kidis the intraparticle diffusion rate constant(mg/g min0.5), and t and I are the adsorption time (min) and boundary layer thickness. Figure 5d demonstrates that the plots of qeversus t0.5did not pass through the origin.If this plot passes through the origin,intraparticle diffusion is the kinetic rate-determining process.The estimated parameters of the particle and pore diffusion model are given in Table 3. They show that the process of intraparticle diffusion was not the kinetic rate-determining process. The high Kidvalue indicates an improvement in the adsorption rate and an enhanced adsorption mechanism, which correlated to improved bonding between uranium ions and the adsorbent.

Fig. 5 Evaluation of the U(VI)adsorption kinetics.a Pseudo-first-order,b pseudo-second-order,c Elovich kinetic,and d intraparticle models on the TC and GMTC adsorbents

Table 3 Kinetic parameters of U(VI)adsorption on the TC and GMTC adsorbents

3.2.7 Adsorption isotherms

Adsorption isotherms are essential to discover the details of the adsorption mechanism, surface properties,and other properties affecting the adsorption procedures.The isotherm parameters of the Langmuir, Freundlich,Dubinin and Radushkevich (D-R), and Temkin models were evaluated to determine their capability to interpret experimental results.

According to the Langmuir model, U(VI) uptake happens on a homogeneous surface via a saturated monolayer of U(VI) on the surface of the TC or GMTC adsorbents with a fixed adsorption energy. Moreover, there is no movement of the uranium ions within the surface plane[32]. The following equation expresses the Langmuir isotherm:

Fig. 6 Fitting of the U(VI)adsorption isotherms using the a Langmuir,b Freundlich,c Dubinin and Radushkevich(D-R),and d Temkin models for the TC and GMTC adsorbents

Table 4 Isotherm parameters of different models for U(VI) adsorption on the TC and GMTC adsorbents

where Ce(mg/L) is the concentration of the solution at equilibrium, qe,and qmax(mg/g) are the uranium uptake at equilibrium and the maximum uptake capacity,and b is the affinity constant (L/mg).

The resulting linear plots of Ce/qeagainst Ceare shown in Fig. 6a.As recorded in Table 4,the maximum capacities(38.91 mg/g and 196.08 mg/g) were very close to the experimental capacities (37.20 mg/g and 193.00 mg/g),and the correlation coefficient values(R2)approached unity for the TC and GMTC adsorbents. These results demonstrated that the adsorption of the uranium ions followed the Langmuir model, which implies that the TC and GMTC adsorbents were homogeneous in the aqueous state and followed a monolayer adsorption process.

The Freundlich isotherm was also used to model the adsorption of the uranium ions on the adsorbent surface.This model is commonly applied to determine the heterogeneity and surface energies. It assumes that metal adsorption occurs with a heterogeneous energetic distribution of the active sites, accompanied by interactions between the adsorbed solutes[33].The following equation describes the Freundlich model:

where Kfis a constant linked to the overall uranium(VI)adsorption capacity (mg/g), 1/n is a constant related to surface heterogeneity, and qeand Ceare the U(VI) uptake(mg/g) and U(VI) concentration at equilibrium (mg/L).Plotting logqeversus logCeyields a regression line that permits the determination of the 1/n and Kfvalues. The values of the Freundlich constants 1/n and Kfwere estimated from the slope and intercept(Fig. 6b),and are listed in Table 4. The Kf(mg/g) values of the TC and GMTC adsorbents were smaller than the experimental capacities of U(VI). The R2values of the TC and GMTC adsorbents were 0.599 and 0.674. These data indicated that the two adsorbents did not have heterogeneous coverage. Therefore, the Freundlich isotherm did not agree with the experimental data.

The Dubinin-Radushkevich (D-R) model assumes a Gaussian distribution of energy sites and discriminates among chemical and physical adsorption as a function of the adsorption heterogeneity [34]. It is employed to appraise the heterogeneity of the surface energies. The DR isotherm is expressed linearly as follows:

where qDis the D-R model monolayer capacity(mg/g),BD(mol2/kJ2) is a constant relevant to the adsorption energy,qe(mg/g) is the amount of U(VI) adsorbed at equilibrium,and the Polanyi potential ε is related to the equilibrium as follows:

where R and T are the gas constant and absolute temperature, and Ce(mg/L) is the U(VI) concentration at equilibrium. The constant BDgives the mean adsorption free energy (E) per mole of U(VI) as it migrates to the surface of the studied adsorbent and is transferred to the solid surface from infinity in the solution, and can be computed using the following relationship:

The D-R constants qDand BDwere estimated from the linear curves of lnqeversus ε2using the intercepts and slopes, which are shown in Fig. 6c and Table 4. If the quantity of E is within 8-16 kJ/mol, the adsorption is assumed to progress through chemisorption, while if E is smaller than 8 kJ/mol, the adsorption mode is physisorption [35]. From the obtained data listed in Table 4, the quantities of E were smaller than 1 kJ/mol for U(VI)adsorption on the TC and GMTC adsorbents, indicating that the adsorption mode of U(VI) was physisorption.Additionally, for the TC and GMTC adsorbents, the quantities of qDwere 35.51 and 171.29 mg/g, and the correlation constants (R2) were 0.735 and 0.908, respectively. The D-R model was not in agreement with the experimental data of the TC and GMTC adsorbents.

The Temkin isotherm predicts a linear range of adsorption energies. The adsorption heat of the molecules decreases linearly with increasing coverage, based on sorbent-sorbate interactions, and the sorption process is characterized by a uniform binding energy distribution[35]. The linear formula of the Temkin isotherm is given as:

where Ce(mg/L)and qe(mg/g)are the U(VI)concentration and the amount of U(VI) adsorbed at equilibrium, respectively. bT(kJ/mol) is the Temkin constant related to the heat of adsorption, and KT(L/g) is the binding constant at equilibrium corresponding to the maximum binding energy.The Temkin constants were estimated from the plot of qeagainst lnCeas shown in Fig. 6d, and are listed in Table 4.The data indicated that the magnitudes of R2were 0.607 and 0.789 for the TC and GMTC adsorbents,respectively.This outcome suggested that the experimental data were not fitted well by the Temkin isotherm.

In conclusion, the data above showed that the parameters of the Langmuir model corresponded better to the experimental value than those obtained using the other studied models.

3.2.8 Thermodynamic adsorption studies

Thermodynamic measurements were also used to determine the adsorption mechanism via the variation in the Gibbs free energy (ΔG°), enthalpy (ΔH°), and entropy(ΔS°). The U(VI) thermodynamic sorption parameters of the TC and GMTC adsorbents were estimated using the Van’t Hoff equations [36]:

where Kd(L/g) is the adsorption coefficient, ΔG° is the Gibbs free energy (kJ/mol), ΔH° is the variation in enthalpy (kJ/mol), ΔS° is the variation in entropy(J/mol K), R is the universal gas constant, and T is the absolute temperature(K).The thermodynamic variations of TC and GMTC were estimated by graphing logKdversus 1/T(Fig. 7).The plot produced a straight line from which the magnitudes of ΔH°and ΔS°were evaluated using the slope and intercept,respectively.The quantities of ΔH°,ΔS°,and ΔG° are listed in Table 5.

From the extended results, the positive value of ΔG°on the TC adsorbent indicated that the adsorption mode was non-spontaneous. The increase in ΔG° with increased temperature demonstrated that the adsorption reaction became unfavorable as the temperature was raised. However, the negative values of ΔG° for GMTC indicated that the adsorption of uranium ions on the GMTC adsorbent was feasible and spontaneous.Additionally,the Gibbs free energy showed that the adsorption steps were favorable for the electrostatic adsorption reactions between U(VI) and the GMTC adsorbent. The negative value of ΔH° may suggest an exothermic process of adsorption. The negative value of ΔS° confirmed the probability of adsorption and the randomness on the adsorbent/solution interface through the U(VI) adsorption mode on the TC and GMTC adsorbents.

Fig. 7 Plot of log Kd versus 1/T for U(VI) adsorption on the TC and GMTC adsorbents

3.3 Elution assessments

3.3.1 Eluting agent type

In this experiment, different types of eluting agents,namely, HNO3, HCl, H2SO4, NaCl, Na2CO3, and (NH4)2-CO3, were utilized for the desorption of uranium from the loaded adsorbents UO2-TC and UO2-GMTC, while the other desorption parameters were kept constant at a 1 M eluting agent concentration, 1:30 (S:L) phase ratio (30 mL of eluent and 1 g of adsorbent),and 60 min contact time at 25 °C. As illustrated in Fig. 8a, it was obvious that the maximum elution efficiencies of uranium(VI) from the loaded TC and GMTC adsorbents were obtained using 1 M sulfuric acid, with desorption efficiencies of 80 and 85%,respectively.Hence,sulfuric acid was used as eluent for the following experiments.

3.3.2 Sulfuric acid concentration

Various H2SO4concentrations from 0.2 to 2 M were used to investigate the optimum concentration for the elution process of the two uranium-loaded adsorbents(UO2-TC and UO2-GMTC).The other experimental factors were set constant at 60 min contact time,an S/L phase ratio of 1:30,and 25 °C.From the obtained results in Fig. 8b,it was clear that the desorption efficiencies of uranium(VI)from the loaded TC and GMTC adsorbents increased with an increase in H2SO4concentration until reaching the maximum efficiencies of 80 and 85% at 1 M H2SO4,respectively. Accordingly, 1 M H2SO4was the best concentration for uranium desorption.

Table 5 Thermodynamic parameters of U(VI) adsorption on the TC and GMTC adsorbents

Fig. 8 Influence of the a eluting agent,b H2SO4 acid concentration,c S/L phase ratio,and d contact time on the uranium desorption efficiencies for the TC and GMTC adsorbents

3.3.3 Solid/liquid phase ratio

To increase the desorption efficiency of uranium, the effect of the S/L phase ratio was studied in the range from 1:10 to 1:70, while the other factors were fixed at a 1.0 M sulfuric acid and 60 min contact time at room temperature(Fig. 8c). The obtained data clearly demonstrated that the uranium desorption efficiencies gradually increased with an increase in S/L phase ratio until the attainment of the highest efficiencies (90 and 94%) at a 1:50 S/L ratio, and remained constant at phase ratios from 1:50 to 1:70 S:L for the uranium-loaded TC and GMTC adsorbents. Therefore,the optimized 1:50 S/L phase ratio was used for further experiments.

3.3.4 Contact time

The effect of contact time on the desorption of uranium from the uranium-loaded TC and GMTC adsorbents was investigated using different contact times (15-120 min),while the other factors were kept constant at 1.0 M H2SO4,a 1:50 S/L phase ratio,and room temperature.As shown in the obtained data in Fig. 8d, the uranium desorption efficiencies rose from 51 and 60% to 96 and 99% for the TC and GMTC adsorbents, respectively, as the contact time was increased from 15 to 75 min, and the desorption efficiencies remained constant from 75 to 120 min. Hence, a 75 min contact time was used in the following experiments.

3.4 Reusability

To study the reuse of the TC and GMTC adsorbents,the uranium-loaded TC and GMTC adsorbents were regenerated using 1 M sulfuric acid, a 1:50 S/L phase ratio, and a contact time of 75 min at 25 °C. The adsorption-desorption processes were repeated several times until the desorption efficiencies decreased from 96 and 99% to 80 and 85%for the TC and GMTC adsorbents after six continuous cycles, indicating the excellent adsorption stability of the two adsorbents for uranium recovery.

3.5 Case study

3.5.1 Chemical analysis of El Sela ore material

Table 6 Chemical analysis of El Sela working sample

Recently, the low-grade uranium ore from El Sela area has been essentially comprised of potash-feldspar,plagioclase, muscovite, and biotite [37] and contains the secondary minerals sericite, kaolinite, and chlorite. Additionally, it contains opaques and garnet, which are accessory minerals. Uranophane and beta-uranophane[Ca.UO2(SiO3)2(OH)·5H2O] are the main uranium minerals identified in El Sela ore materials. B-Autunite (a uranium phosphate mineral, Ca(UO2)2(PO4)2·10-12H2O) was also observed[38].An El Sela sample was characterized to determine its major oxides and interesting trace elements using the wet chemical procedure.The obtained major and trace constituents of the studied sample are given in Table 6. SiO2assayed about 67.62% in the El Sela composite sample. Additionally, uranium and total rare earth elements assayed at 0.093 and 0.504%, respectively.

3.5.2 Uranium recovery

The uranium pregnant leach liquor was prepared by mixing 5 kg of the properly ground El Sela ore material(149-100 μm) with 20 L of 150 mg/L sulfuric acid solution at a stirring speed of 200 rpm for a leaching time of 6 h at 25 °C [39]. The insoluble gangue was then filtered,and the obtained leach liquor assayed 215 mg/L of uranium ions,representing a leaching efficiency of 93.61%,using an oxidometric technique.The associated ions in the pregnant solution were analyzed using the ICP-OES technique(Table 7).

The adsorption process was applied to the 20 L of pregnant solution, which contained 215 mg/L of uranium ions and some interfering ions,using 25 g of GMTC under the optimum conditions of pH 4.5, 60 min contact time,and room temperature. Based on the earlier results, the adsorption capacity of uranium(VI) on the GMTC was 193.0 mg U/g adsorbent. Finally, all the uranium ions in the leach liquor were adsorbed on the 25 g of GMTC adsorbent (4.2 g uranium ions/25 g adsorbent).

Table 7 Chemical analysis of the El Sela pregnant liquor

Fig. 9 XRF spectrum of the sodium diuranate product

Table 8 Chemical analysis of the sodium diuranate product

Elution of uranium from the loaded adsorbent (UO2-GMTC) was carried out using the conditions described above. The working uranium-loaded adsorbent containing ～4200 mg uranium was agitated using 1.25 L of 1 M(100 mg/L) sulfuric acid for 75 min of contact time at room temperature. The uranium(VI) in the aqueous phase was determined.It was found that the adsorption efficiency was 99%, and the uranium concentration was assayed at 3.36 g/L.

Precipitation of the uranium ions from the solution was carried out after the pre-concentration process. Sodium hydroxide was used to precipitate uranium ions as sodium diuranate, and finally, this precipitate was dried at 110 °C.The sodium diuranate ‘‘yellow cake’’ was prepared from the 1.25 L sulfate solution, which was subjected to concentration by evaporation to a volume of 300 mL of aqueous solution.The precipitation process was carried out by adjusting the pH to 7 using a 20% sodium hydroxide solution to precipitate the uranium ions as Na2U2O7·6H2O (5.92 g),which was dried at 110 °C.The final product was Na2U2O7and weighed 5.45 g.

The obtained XRF spectrum contained distinct peaks indicating that it was composed of sodium diuranate and some associated metal ions (Fig. 9 and Table 8). Furthermore, it was quantitatively analyzed by the ICP-OES technique to identify its chemical constituents (Table 8).The obtained data demonstrated that sodium diuranate assayed 95.54%.

4 Conclusion

Waste clay was treated by ethyl acetate.The treated clay was modified with gallic acid to obtain a low-cost adsorbent with high efficiency and the best performance for uranium ion adsorption. The treated clay and gallic acidmodified treated clay were utilized to extract uranium ions from a sulfate solution. The optimized adsorption conditions were pH 4.5, a contact time of 60 min, and ambient temperature. The maximum uranium adsorption capacities on TC and GMTC were 37.20 and 193.0 mg/g, respectively. The mode of uranium ion adsorption onto TC and GMTC followed the pseudo-second-order kinetic and Langmuir models well.The thermodynamic study revealed that the adsorption of the uranium ion on TC was nonspontaneous, while the adsorption process on GMTC was spontaneous. The adsorption reactions of TC and GMTC were exothermic due to the negative value of ΔH°. The negative ΔS° confirmed the probability of adsorption and the randomness on the adsorbent/solution interface of the U(VI) adsorption mode on the TC and GMTC adsorbents.Uranium ions were eluted from the uranium-loaded TC and GMTC adsorbents using 1 M sulfuric acid, a 1:50 S/L phase ratio, and a contact time of 75 min at 25 °C. The optimum conditions were applied to a pregnant solution of El Sela uranium solution to recover the uranium ions as sodium diuranate, which assayed 95.54%.