Amino-functionalised cross-linked polyacrylamide for the adsorption of U(VI) ions from contaminated aqueous solutions

ABSTRACT The present work discusses the adsorption of U(VI) ions from aqueous solutions using amino-functionalised cross-linked polyacrylamide adsorbent. Functional groups identification and surface morphology of the prepared polymeric adsorbent were explored by Fourier Transform Infrared (FTIR), elemental analysis and Scanning Electron Microscope (SEM) techniques. The adsorption of U(VI) ions from aqueous solutions using the prepared polymeric adsorbent was studied and the optimum adsorption conditions were addressed. The results showed that a maximum adsorption capacity of 68.6 mg/g could be achieved at room temperature under the following conditions; solution pH 4.0, contact time 240 min and adsorbent dose 0.7 g/ L. The adsorption process was found to follow Langmuir isotherm model with pseudo-second order kinetic mechanism. Moreover, the thermodynamic studies suggested that the adsorption process is spontaneous and exothermic. Experiments on the adsorbent regeneration and reusability were carried out and the results were promising. As a final point, the prepared polymeric adsorbent was applied for the removal of U(VI) ions from real contaminated sample and about 96% of U(VI) ions were removed in a single run.


Introduction
Uranium is a radioactive element that naturally exists in our planet especially in granites and sedimentary rocks [1].Because of its high solubility in water, uranium salts could be leached from their rocks by the action of ground water passage over uranium-containing rocks.Moreover, the utilisation of uranium in several industries, such as mining, ore processing, fossil fuel extraction, commercial aviation, and use of uranium-containing phosphate fertilisers, increases its concentration in the environment [2,3].Such increased amount should be controlled before waste disposal because high levels of uranium badly affect the ground and stream water [4,5].Because uranium decays by alpha particles, external exposure to uranium is not as dangerous as exposure to other radioactive elements because the skin will block the alpha particles.Ingestion of high concentrations of uranium, however, can cause severe health effects, such as cancer of the bone or liver.Inhaling large concentrations of uranium can cause lung cancer.Uranium is also a toxic chemical, meaning that ingestion of uranium can cause kidney damage from its chemical properties much sooner than its radioactive properties would cause cancers of the bone or liver.Therefore, the World Health Organization (WHO) has set a maximum contaminant level for uranium in drinking water of 9.0 mg/L [6][7][8].To stick to this maximum level and keep our environment protected, industrial wastes containing uranium must be treated before discharging; to avoid the contamination of our environment with high doses of uranium.
The efficiency of the adsorption process depends mainly on the chemical and physical characteristic and surface functionality of the used adsorbents.Therefore, research is ongoing to synthesise new adsorbents that exhibit high adsorption capacity with fast adsorption kinetics [21][22][23].Hydrogels are cross-linked hydrophilic polymers those act as supports for metal ions including uranyl ions either by complexation with metal ions, by acting as ion exchange resins or by acting as selective adsorbents.These polymeric ligands are specifically tailor-synthesised for better selectivity and efficiency.Due to a number of advantages associated with these hydrogels, they have attracted considerable attention for the removal of uranyl ions (UO 2 2+ ) from seawater and wastewater.The advantages associated with the hydrogels include their selective nature, reusability, high efficiency, low cost, the ease of handling, eco-friendly nature, easy availability and easy syntheses.The chelating polymers are synthesised either by polymerisation of functionalised monomers or by incorporation of chelating groups into the polymer matrix by stepwise functionalization of existing polymers.Cross-linked polymeric hydrogels with diverse chelating groups are extensively being used as potential adsorbing materials for recovery of uranium ions from wastewaters [25,26].
In this work, an amino-functionalised solid-phase adsorbent based on cross-linked polyacrylamide was prepared, characterised and evaluated as U(VI) ions adsorbent.The complexation ability of the prepared adsorbent relies on the presence of nitrogen and oxygen atoms as donor sites.Investigations on the parameters that influence the adsorption process were carried out and the best adsorption conditions were defined.Furthermore, isotherm, kinetic and thermodynamic studies were highlighted.Adsorbent recyclability as well as real sample applications was considered during the current work.

Chemicals and instruments
The chemicals and reagents used during this study were of analytical grade and used without any further purification.N,N methylene bis-acrylamide (NMBA) and Acrylamide (AAm) monomers and uranyl nitrate hexa-hydrate (UO 2 (NO 3 ) 2 .6H 2 O) were purchased from Merk.Uranium stock solution of 1000 ppm was prepared by dissolving accurate weight of uranyl nitrate hexa-hydrate (UO 2 (NO 3 ) 2 .6H 2 O) in distilled water.The working standard solutions of uranium (VI) ions were prepared by appropriate dilution of the stock solution immediately before use.
The pH measurements were obtained with a Jenway (Felsted Dunmow, England) model 3310 pH metre.Infrared spectra (FT-IR) were examined using FT-IR spectrometer (Bomen, Hartman & Braun, and model MB-157, Canada) in the region from 4000 to 400 cm −1 under ambient air condition using KBr as a diluent.The elemental analysis was obtained with a PerkinElmer 2400 CHNS analyser.Morphology of the prepared polymeric adsorbent before and after adsorption was examined by scanning electron microscope (SEM) using QUANTA 200 with ≥10 kV accelerated voltage.A double-beam spectrophotometer of high resolution power (Shimadzu UV-Visible recording spectrophotometer type UV-60A) and ICP-OES (Optima 2100DV, Perkin-Elmer, USA) were used for the determination of uranium concentration.

Preparation method
The cross-linked polyacrylamide (CPAAm) was prepared according to Sharaf et. al., 2007 [27].In brief, 10 g of AAm and 1.14 g NMBA cross-linker were dissolved in 250 mL ethanol.62 mg K 2 S 2 O 8 was added to this solution and stirred for 6 hours at 70°C in a water bath.The resultant white powder of the crosslinked polyacrylamide was collected by filtration, washed several times with water, ethanol, and acetone, and dried at 80°C under reduced pressure.Introduction of the amino-group into the crosslinked polyacrylamide matrix was done as described by Sharaf et.al. [28].For about 9 hours, a mixture of CPAAm (10 g) and 100 ml ethylendiamine was refluxed at 100-110°C.Afterwards, the mixture was decanted slowly into a huge volume of cold water.The resulting amino-functionalised polymer was filtered off and washed with NaCl solution to remove the excess of ethylendiamine.Finally, the prepared amino-functionalised cross-linked polyacrylamide was washed with distilled water, methanol and acetone and dried at 80°C under reduced pressure.

Adsorption experiments
Batch adsorption studies are often performed to reveal the influence of solution pH, contact time, temperature, metal ion initial concentration, and dose of the adsorbent on the adsorption behaviour of an adsorbent and consequently the optimal conditions are addressed.The batch adsorption experiments for studying the influence of solution pH, in the range from 1.0 to 7.0, were executed in a thermo shaking water bath (scientific precision SWB 27-27 L) by shaking 0.01 g of the prepared amino-functionalised adsorbent, at 200 rpm for 24.0 h, with U(VI) solutions (25.0 ml, 50.0 mg/L).Afterwards, the adsorbent was isolated and the concentration of U(VI) ions in the filtrate was measured using a spectrophotometric approach employing the Arsenazo III indicator [29] and ICP-OES.To minimise errors in measurements, adsorption experiments were carried out three times and average values were presented.The adsorption capacity (Q, in mg/g) was calculated as given in Equation ( 1): where C i and C f are the initial and final concentration of U(VI) ions (mg/L), respectively, V is the volume of the aqueous phase (L) and m is the mass of the adsorbent (0.01 g).The adsorption efficiency (R%) was calculated using Equation (2): After addressing the optimum pH, the influence of other adsorption parameters were investigated in the indicated intervals; those were contact time (5.0-600.0min), adsorbent amount (0.3-1.0 g/L), initial concentration of U(VI) ions (20.0-300 mg/L) and temperature (25-60°C).

Kinetic studies
Kinetic studies are performed to collect more information about the mechanism of an adsorption process and reveal the nature of adsorbate/adsorbent interactions.Several kinetic models were reported in literatures, of which, pseudo-first-order (PFO) and pseudo-second-order (PSO) [30,31] are widely applied to describe the mechanism of an adsorption process.The equations of these two models are presented in Table S1.According to Ho and Mckay, pseudo-first-order model proposes physical adsorption mechanism while pseudo-second-order model proposes chemical adsorption one.[32].
Covering a time range from 5.0 to 600.0 min, 10.0 mg of the prepared aminofunctionalised polymer was shaken, separately at constant speed of 200 rpm, with 25.0 ml of 50.0 mg/L U(VI) ions solution adjusted at pH 4.0.After each time period, the adsorbent was isolated by means of filtration and the concentration of U(VI) ions in the filtrate was determined.

Isotherm studies
Adsorption isotherm studies are performed to calculate the maximum theoretical capacity of an adsorbent towards a certain adsorbate.Moreover, information about the adsorption mechanism could be clarified [22].In this regard, Langmuir and Freundlich isotherm models are usually used.On the basis of Langmuir model, the maximum adsorption capacity is obtained as a result of saturated monolayer adsorption of the adsorbate molecules at uniform binding sites on the adsorbent surface [33].On the contrary, Freundlich model supposes that the maximum theoretical capacity is due to multi-layer adsorption of the adsorbed molecules on non-uniform surface of the adsorbent [33,34].The linear equations of the two models are given in Table S1.
Covering a concentration range from 20 to 300 mg/L, 10.0 mg of the prepared aminofunctionalised adsorbent was shaken, separately, with 25 ml of each U(VI) ions concentration adjusted at pH 4.0 for 240 min at constant speed of 200 rpm.After equilibrium, the adsorbent was filtered and the concentration of U(VI) ions was determined.

Thermodynamic studies
The required information for doing thermodynamic studies is acquired by exploring the dependence of adsorption behaviour of an adsorbent towards a certain adsorbat as a function of temperature.After that, the parameters needed to evaluate the thermodynamic nature are calculated using equations listed in Table 2 [35,36].These parameters are Gibbs free energy change (ΔG o ), standard entropy change (ΔS o ) and standard enthalpy change (ΔH o ).Data required for thermodynamic studies was collected over a temperature range from 303 to 343 K by shaking 10.0 mg of the prepared amino-functionalised polymer with 25 mL of 50 ppm U(VI) ions solution adjusted at pH 4.0 for 240 min at 200 rpm.The amount of U(VI) ions in the filtrate was measured.

Adsorbent regeneration and reusability
Adsorbent reusability studies are as important as adsorption optimisation studies because the capability of reuse adsorbents several times is economically advisable.Consequently, after finishing an adsorption run, it is preferred to regenerate the adsorbent by desorbing the adsorbed substance and reuse it.In this concern, citric acid, nitric acid, sulphuric acid and hydrochloric acid were examined as desorbing agents.The uranium-loaded adsorbent was shaken, separately, with 0.5 M acid solution for 24 h.The concentration of the desorbed U(VI) ions was determined and the acid that give the maximum desorption amount was identified.Afterwards, a solution of 0.1 M NaOH was applied to remove the excess of the acid and to activate the adsorbent.Finally, the regenerated adsorbent was washed with deionised water till being neutral and reused for another adsorption batch.

Error function analysis
Two error functions, namely the residual root mean square error (RMSE) and the average relative error (ARE), are used to evaluate the fitness of the studied isotherm and kinetic models, respectively.The equations are given below [37]: where q e;exp: and q e;exp: are the experimental adsorption capacity (mg/g) at equilibrium and at time t, respectively, q e;cal: q t;cal: are the calculated adsorption capacity by the studied model at equilibrium and at time t, respectively, C e is the equilibrium concentration (mg/L) and 'n' is the number of observations.Smaller values of RMSE and APE mean a good fitting of a model with the experimental data [38].

Characterisation study
The amino-functionalised cross-linked polyacrylamide was prepared as shown in Scheme 1.The chemical structure and surface morphology were studied using FTIR, elemental analysis and SEM techniques.According to FTIR studies, Figure 1(a) has characteristic peaks at 3429 and 1650 cm -1 .These peaks belong to the N-H stretching vibration in -NHgroup of N,N-methylene bis(acrylamide) or -CONH 2 groups of acrylamide.Additionally, the peak at 2930 cm -1 is due to symmetric or asymmetric stretching vibration of the -CH 2 groups of acrylamide or N,N′-methylene bis (acrylamide).The successful introduction of the amino-groups into the polymer surface was confirmed by the appearance of a new peak at 3656 cm −1 , which is related to the amino-group ((Figure 1(b)).The elemental analysis results confirm the successful functionalization of the cross-linked polyacrylamide because of the good agreement between the calculated and experimental values, see Table S2 [39].The functional group capacity of the amino-functionalised polymer was determined by indirect neutralisation titration and was 8.2 mmol per gram of the polymer.After U(VI) ions loading (Figure 1(c)).The FTIR spectrum of the U(VI)-loaded polymer shows a peak at about 895 cm −1 which could be assigned to the asymmetrical vibration of the uranyl ion [40].The peaks at 1384 and 1557 cm −1 belong to the uranyl ion [40].Moreover, the deformation of NH 2 and the shift in the carbonyl group of amide (from 1650 to 1644 cm −1 .These changes in the IR spectrum of the chelating polymer after the adsorption of uranyl ions confirms the complexation between the functional groups on the polymer surface and the U(VI) ions.In addition, SEM images shown in Figure 2 give further evidence on the fixation of U(VI) ions on the surface of the prepared polymeric adsorbent because a change in the adsorbent surface morphology was detected after the adsorption.Scheme 1.Synthetic pathway of the amino-functionalized cross-linked polyacrylamide adsorbent.

Factors affecting the adsorption process 3.2.1.1. Effect of pH.
The pH of an adsorbate solution has a great effect on the adsorp- tion affinity of an adsorbent because the pH of a solution might changes the surface chemistry of the polymeric adsorbent and the aqueous chemistry of the adsorbate.Therefore, it is necessary to study how the pH of the U(VI) ions solution influences on the adsorption affinity of the prepared polymeric adsorbent.Uranyl nitrate working solutions were adjusted to desired pH with either 0.1 M HNO 3 or 0.1 M NaOH solution.The theoretical distribution of uranium (1.20 mM) aqueous speciation as a function of pH in 0.1 M HNO 3 was identified by Medusa/Hydra programand presented in Figure S1 [41].As shown in Figure S1, (UO 2 NO 3  +) and (UO 2
In Figure 3(a), gradual increase in the adsorption capacity was observed with an increase in pH of the U(VI) ions solution till reaching a critical pH value of 4.0 and then went down.In acidic solutions, amino groups on the polymer surface tend to adsorb hydrogen protons from the solution and the adsorbent surface becomes protonated.Repulsion between the positively charged adsorbent and (UO 2 NO 3  +) and (UO 2 2+ ) makes the adsorption of uranyl ions not favourable [29].Gradual increase in the solution pH increases the affinity of uranyl ions adsorption because the concentration of hydrogen ions decreases.The adsorption capacity of the prepared polymeric adsorbent started to increase by increasing the solution pH until reaching critical value at pH 4.0.Afterwards, the adsorption capacity decreases because of the formation of stable uranium complexes, such as and [(UO 2 ) 4 (OH)] 7+ , with lower adsorption affinities [19].Consequently, the working solutions were adjusted to pH 4.0 during the rest of this study.

Effect of contact time and kinetic studies. Contact time and kinetic studies are
important to find out information about how fast an adsorption process takes place and what is the adsorption mechanism [7,8].Plotting the contact time versus the adsorption efficiency showed that the adsorption of U(VI) ions by the prepared polymeric adsorbent is fast and more than 50% of uranyl ions in the solution was removed within 15 min, see Figure 3(b).This suggests that the binding sites on the polymer surface are large and accessible.After the first 15 min, there are still free active sites ready to accommodate more uranyl ions and about 93% of uranyl ions was removed after 240 min.Afterwards, the adsorption rate increased slowly till reaching a steady state.Based on this result, a contact time of 240 min was established throughout the rest of this study.
For kinetic studies, pseudo-first order and pseudo-second order kinetic models were investigated.The plots as well as the kinetic constants obtained for each model are shown in Figure 4 and Table 1, respectively.Based on the regression coefficients (R 2 ) and the error functions (RMSE and APE) values, we concluded that the adsorption of uranyl ions by the prepared polymeric adsorbent follows a second-order mechanism because of the large value of R 2 and small values of RMSE and APE [37].Moreover, the closeness of the calculated adsorption capacity by PSO model to that obtained experimentally, 70 vs 68.6 mg/g, suggests the suitability of PSO model to describe the current adsorption process.It is worth mentioning that the adsorption of U(VI) ions by other adsorbents followed pseudo-second order kinetic model [15][16][17][18].Fitting of the experimental data by pseudo-second-order model implies that the adsorption mechanism is chemisorption; the rate determining step depends on amount of uranyl ions in the solution and the number of active sites on the adsorbent surface.
where K d is the distribution coefficient, qe (mg g −1 ) is the uptake capacity at equilibrium and Ce (mg L −1 ) is the equilibrium concentration of the adsorbate and ΔG o (kJ mol −1 ) is the change in Gibb's free energy, ΔS o (J/mol K −1 ) is the change in standard entropy, ΔH o (KJ/mol) is the change in standard enthalpy, T (K) is the kelvin temperature and R (8.314 J mol −1 K −1 ) is the general gas constant.

Effect of uranium initial concentration and isotherm studies.
The influence of uranyl ions initial concentration on the adsorption capacity of the prepared polymeric adsorbent was investigated by shaking 0.01 g of the adsorbent with different 25 mL aliquots of uranyl ions solutions over a concentration range from 20.0 to 300 mg/L.The remaining amount of uranyl ions in each aliquot was measured.The results showed that the sorption efficiency decreases with increasing the initial concentration of U(VI) ions, see Figure 3(c).This was attributed to the limited number of amino groups on the adsorbent compared to the U(VI) ions concentrations.Langmuir and Freundlich isotherm models were used to explore more information about the nature of the adsorption process.The plots and constants of the two models were given in Figure 5 and Table 1, respectively.As mentioned in the kinetic study section, we will rely on the correlation coefficients and error functions values to decide which isotherm model is the best to describe the adsorption of U(VI) ions by the prepared polymeric adsorbent.Based on the R 2 , RMSE and APE values as well as the calculated capacity (q max ), the current adsorption process follows Langmuir isotherm model.This assumes homogeneous distribution of the binding functional groups on the adsorbent surface and monolayer adsorption of U(VI) ions on the adsorbent surface.It is noteworthy that similar behaviour was reported for the adsorption of U(VI) by other adsorbents [18][19][20][21][22].Moreover, the prepared polymeric adsorbent showed competitive adsorption capacity [42,43], see Table 3.

Effect of adsorbent dosage.
Sorption efficiency as a function of polymer amount (0.3-1.0 g/L) was executed on uranium ions solution of 50 ppm concentration and the experimental data was shown in Figure 3(d).Indeed, the adsorbent dose influences the sorption efficiency.As the amount of the adsorbent increases, the sorption percentage increases.For a fixed amount of U(VI) ions (50 ppm), an adsorption percentage of about 95% could be achieved with 0.7 g/ L in 240 min.

Thermodynamic studies
Thermodynamic studies revealed that the adsorption of U(VI) ions by the prepared polymeric adsorbent is inversely related to the temperature; the adsorption capacity decreases as the temperature increases (Figure 3(e)).Moreover, with a close look to the calculated thermodynamic parameters, ΔG o , ΔH o and ΔS o , see Table 2, it is clear that the current adsorption process is energetically favoured and takes place spontaneously because of the negative value of ΔG o that obtained at room temperature.The negative value of ΔH o , obtained from the slope of the straight line obtained by plotting log k d versus (1/T), reflects the exothermic nature of the adsorption process (Figure S2).Finally, the positive value obtained for ΔS o suggests the irreversibility of the adsorption process and the high stability of the complex formed between the U(VI) ions and amino-groups on the adsorbent surface.

Uranium elution and adsorbent reusability
A study on the desorption of U(VI) ions after loading on the adsorbent surface was carried out using different eluents; 0.5 M citric acid, 0.5 M nitric acid, 0.5 M hydrochloric acid, and 0.5 M sulphuric acid.The study was performed by shaking the loaded adsorbent with 25 mL of the eluent solution for 24 hrs.The elution efficiency was calculated and given in Table S3.Amongst all, nitric acid showed the highest elution power.Afterwards, the regenerated adsorbent was washed extensively by de-ionised water and was involved in another adsorption cycle under the optimum adsorption conditions.The reusability study confirmed that the prepared polymeric adsorbent is recyclable and could be used seven consecutive adsorption and desorption cycles with adsorption efficiency up to 90%, see Table S4.

Real sample treatment
After addressing the best adsorption condition using the batch adsorption method, the prepared polymeric adsorbent was used for real sample treatment to evaluate its efficiency as U(VI) ions adsorbent under field conditions.In this regard, the pH of the raffinate solution containing U(VI) ions was adjusted using either 0.1 M HNO 3 or 0.1 M NaOH solution.Afterwards, 0.1 mg of the prepared polymeric adsorbent was shaken with 100 ml of the raffinate solution for 4.0 hrs.Then, the adsorbent was filtered off and the concentration of U(VI) ions in the filtrate was determined.The results revealed that 96% of U(VI) ions were removed from the raffinate; indicating the effectiveness of the prepared polymeric adsorbent as U(VI) ions scavenger, see Table S5.

Conclusion
Amino-functionalised cross-linked polyacrylamide was prepared and investigated batchwise for uranium(VI) ions removal from aqueous solutions.The main variables affecting the removal process such as shaking time, U(VI) initial concentration, adsorbent dose, and reaction temperature were studied.Kinetic and isotherm investigations revealed that the adsorption process obeys pseudo-second order and Langmuir isotherm models, respectively.The maximum adsorption capacity was found to equal 68.6 mg/g; obtained at solution pH of 4.0, reaction time 240 min, adsorbent dose 0.7 g/ L, and room temperature.The prepared amino-functionalised polymeric adsorbent exhibited high stability for seven cycles of uranium adsorption and desorption processes; using nitric acid as eluent.Finally, application of the prepared polymeric adsorbent for U(VI) ions removal from real contaminated waste solution was performed, and about 96% of U(VI) ions were removed.

Figure 3 .
Figure 3.Effect of pH (a), contact time (b), initial concentration of U(VI) ions (c), adsorbent amount (d) and temperature (e) on the adsorption capacity of the prepared polymeric adsorbent towards U(VI) ions.Experimental conditions; U(VI) ions concentration 50 mg/L, adsorbent dose 0.7 g/L, contact time 24.0 h for (a), U(VI) ions concentration 50 mg/L, adsorbent dose 0.7 g/L, pH 4.0 for (b), U(VI) ions concentration 50 mg/L, pH 4.0 and contact time 240 min for (c), adsorbent dose 0.7 g/L, pH 4.0, contact time 240 min for (d), and adsorbent dose 0.7 g/L, pH 4.0, contact time 240 min for (e).All experiments were performed at temperature 298 K.

Figure 4 .
Figure 4. Pseudo-first order (a) and Pseudo-second order (b) plots for the adsorption of (VI) ions by the prepared polymeric adsorbent.

Figure 5 .
Figure 5. Langmuir (a) and Freundlich (b) plots for the adsorption of (VI) ions by the prepared polymeric adsorbent.

Table 1 .
Parameters of the Kinetic and isotherm models.

Table 2 .
Van't Hoff equation and thermodynamic parameters.

Table 3 .
Adsorption capacities towards U(VI) ions obtained by other adsorbents.