Introduction

The separation of uranium (actinide) from its associated elements, such as lanthanum and praseodymium (lanthanides), is desired in view of its increased demand in nuclear industries. The conventional methods for the separation of lanthanides and actinides include adsorption, chemical precipitation, membrane filtration, coagulation and flocculation, electrochemical treatment, and solvent extraction [1–9]. However, most of these methods suffer from technical, economic and health problems related to low selectivity, long time of extraction, and large quantity of hazardous materials used. Sorption by functionalized resin provides an environmentally and economically favorable method for removing and separation of soluble lanthanides and actinides from sources such as nuclear and other industrial wastewater [10, 11].

An especially interesting family of materials used for the separation of uranium, lanthanum and praseodymium are those containing phosphate and amino groups [12–14]. These groups can both chelate cationic and adsorb anionic metal species through electrostatic interactions or hydrogen bonding [15, 16]. In this context, phosphonated polyethylenimine resin arises as an attractive material possessing a large number of secondary amine, ternary amine and phosphonate groups.

This study was performed to develop an effective and practical sorbent suitable for removal and isolation of actinides and lanthanides. We have grafted a polyethylenimine methylene phosphonic acid group on a Merifield resin to obtain a new ions exchange resin. The reaction was performed in distilled water under classic heat. This reaction conducts to a highly functionalized polymer possessing phosphonic acid and amine moieties as chelating groups. This resin, in sodium form, was applied as a new sorption material for uranium(VI), lanthanum(III) and praseodymium(III) ions, from nitrate medium, and in batch process. The effects, on sorption process, of some analytical parameters such as shaking speed, time contact, aqueous pH, initial concentration of U(VI), La(III) and Pr(III), temperature and salts addition were studied in detail, while for resin regeneration, three different acids at variables concentrations were tested.

Experimental

Reagents

All solutions were prepared from analytical grade chemicals and distilled water. Commercially Merrifield resin (Chloromethyl styrene–divinylbenzene copolymer, S-3% DVB) and phosphorus acid (H3PO3) were purchased from Rohm and Haas Company. Polyethylenimine (Lupasol® G20) is commercially available from BASF with ratio of primary:secondary:tertiary amines (1:0.91:0.64) with an average molecular weight (M w) of 1300. UO2(NO3)2·6H2O, La(NO3)3·6H2O, Pr(NO3)3·6H2O, Arsenazo III and hydrochloride acid (37%) was provided from Sigma-Aldrich. Nitric acid (70%)), sodium hydroxide (≥97.0%, pellets)), potassium chloride, thiocyanate, and cyanide (KCl, KSCN, KCN) and sodium chloride, nitrate, sulphate and dithionite (NaCl, NaNO3, Na2SO4, Na2S2O4) were purchased from Merck. A stock solutions of 1 × 10−2 mol/L of UO2(NO3)2, La(NO3)3 and Pr(NO3)3 were prepared by dissolving 5.0213, 4.3301 and 4.3501 g respectively in 1 L of distilled water. The diluted solutions of uranyl, lanthanum and praseodymium ions were prepared by appropriate dilution of the stock solutions. The initials pH of the sample solutions were adjusted in the desired range by adding dilutes HNO3 or NaOH solutions.

Apparatus

A FT-IR (Perkin-Elmer 684 IR spectrophotometer) was used for the confirmation of the presence of phosphoanated polyethylenimine on Merifield resin. FT-IR spectra were obtained with KBr pellets and the spectrum was taken from 4000 to 400 cm−1. All pH measurements were performed with a WTW 3310 Set 2 digital pH meter. The sorption of uranium(VI), lanthanum(III) and Praseodymium(III) on functionalized resin was studied by batch technique. A shaker (Haier model) was used for sorption experiments except for temperature effect where a magnetic stirrer (RCT Basic IKAMAG Stirrer with ETS-D5 Temperature Controller) was used. Specord® 210 Plus model analytic Jena UV–Vis spectrophotometer was used to determine U(VI), La(III) and Pr(III) as Mn+–Arsenazo III complex in aqueous phase [17–19].

Sorption studies

In aqueous phase, the uranium(VI), lanthanum(III) and praseodymium(III) concentration was determined spectrophotometrically with Arsenazo III. In a test tube, which contained 2 mL of ammonium acetate/HCl buffer (pH 3.3), were added 100 µL of U(VI), La(III) or Pr(III) solutions to be analyzed and 100 µL of Arsenazo III solution (10−3 mol/L). Arsenazo III reacts with U(VI), La(III) and Pr(III) to form a blue complexes which can be estimated at λ max = 651, 660 and 578 nm respectively [17–19].

The general method of sorption, used for this study, is described as follows: 0.05 g (w) of polyethylenimine methylene sodium phosphonate grafted Merrifield resin (PEIPR-Na) was equilibrated with 5 mL (V) of metal ions solution of known concentration in a stoppered Pyrex glass flask at an ambient temperature (22 ± 1 °C) in a shaker for predetermined time. The resin was separated by filtration and the filtrate was analyzed by UV–Vis spectrophotometer in presence of Arsenazo III for metal ions content [17–19]. The sorption of U(VI), La(III) and Pr(III) ions on the resin at three different temperatures 22, 27 and 35 °C was investigated. For temperature effect, a magnetic stirrer was used and stirring speed was 1000 round per minute to maintain resin particles in suspension. All data reported are based on the average of three replicate measurements.

The percentage of metal ions that is sorbed on the resin (Removal yield) was determined by comparing its concentrations before and after sorption (Eq. 1).

$${\text{Removal}}\ {\text{yield}}\left( {{\% }} \right) = \frac{{C_{i} - C_{\text{e}} }}{{C_{i} }} \times 100$$
(1)

The amount of metal uptakes at time t, q t (mg/g), was calculated by Eq. (2):

$$q_{t} \left( {{\text{mg}}/{\text{g}}} \right) = \frac{{\left( {C_{i} - C_{t} } \right)}}{w} \times V$$
(2)

The distribution coefficient (D) of the metal ions between the aqueous solution and the solid resin is also calculated from Eq. (3):

$$D({\text{mL}}/{\text{g}}) = \frac{[{\text{M}}]_{\text{resin}}}{[{\text{M}}]_{\text{aq}}} = \frac{(C_{i}-C_{\text{e}})}{w} \times \frac{V}{C_{\text{e}}}$$
(3)

where [M]resin and [M]aq is La(III),Pr(III) or U(VI) ions concentration in resin phase (mg/g) and in aqueous solution (mol/L) respectively. C i , C t and C e are respectively the initial, time t and equilibrium La(III), Pr(III) or U(VI) concentration (g/L). V is the volume of the solution (5 mL). w is the mass of the functionalized resin used (0.05 g).

Desorption studies

Desorption of U(VI), La(III) and Pr(III) was performed by mixing 1.000 g of loaded resin with U(VI), La(III) or Pr(III) with 5.0 mL of acid solutions (HCl, HNO3 and H2SO4) at known concentration, and shaken at 310 rpm for 4 h at ambient temperature (22 ± 1 °C). The final U(VI), La(III) and Pr(III) concentrations in the aqueous phase were analyzed by UV–Vis spectrophotometer. All data reported are based on the average of three replicate measurements.

Results and discussion

Ion exchange resin synthesis

Polyethylenimine methylene phosphonic acid grafted Merrifield resin (PEIPR) was originally designed for removing of metal ions, taking advantage of high affinity of phosphate for metals [20]. In order to use this polymer in SPE technique, we prepared a polystyrene–divinyl benzene resin that contained a maximum of polyethylenimine methylene phosphonic group. This fully phosphonated resin (PEIPR) was prepared in two steps (Fig. 1). In first step, after washing with absolute ethanol, the vacuum dried Merrifield resin 3% was reacted with polyethylenimine Lupasol® G20. The reaction mixture was refluxed in toluene for 24 h. The polyethylenimine grafted polystyrene resin beads obtained were purified from the excess of reactants by washing repeatedly with water and ethanol. In the second step, the polyethylenimine grafted polystyrene resin beads was functionalized by phosphonic acid group using the Moedritzer–Irani reaction [21], a mixture of polyethyleneimine grafted polystyrene resin, phosphorous acid, hydrochloric acid–water (1:1) solution and formaldehyde aqueous solution was vigorously stirred and refluxed for 24 h in toluene. The resulting polymer was filtered and washed repeatedly with distilled water and ethanol. Preliminary sorption tests show that PEIPR beads (in acidic form) are not a good extractant for La(III), Pr(III) and U(VI). Similar results were observed with other phosphate resins [22]. The presence of H+ saturates the active cites (hetero-atoms: O and N) on the resin and gives it a positive charge which prevents the approach of the cations to be extracted. For this reason, the resin (PEIPR) was treated, several times, with a solution of 1 M sodium hydroxide in order to replace the H+ on phosphonic acid group by Na+ (Fig. 1). The resulting sodium phosphate polymer (PEIPR-Na) was filtered and washed repeatedly with distilled water and dried in vacuum.

Fig. 1
figure 1

Schematic illustration of the synthesis process of PEIPR-Na from Merrifield resin and polyethylenimine Lupasol® G20

Full size image

Ion exchange resin characterisation

The presence of amine, on the polyethylenimine (PEI) grafted resin, was confirmed by the presence of absorption at 2910 and 1602 cm−1 (NH). The presence of phosphonic acid on PEIPR was confirmed by the appearance of absorption at 1110 cm−1 (P=O), 1030 (P–O), 2372 and at 2340 cm−1 (P–OH) with the disappearance of NH2 band at 2910 cm−1 [23–25].

The experimental CHNP analysis data (%) of the PEIPR (in acidic form) was: C: 57.44; H: 6.339; N: 7.00; P: 4.95. This analysis confirm the presence of the two groups, polyethylenimine and phosphonic acid, onto the Merrifield resin.

Uranium(VI), Lanthanum(III) and praseodymium(III) sorption

Effect of shaking speed

The shaking speed (θ) showed a medium effect on U(VI) sorption, yield sorption increases from 94.4 to 97.1% when θ increases from 0 to 310 round per minute (rpm) (Fig. 2).

Fig. 2
figure 2

Removal yield of La(III), Pr(III) and U(VI) by PEIPR-Na as a function of shaking speed, [La(III)] i  = [Pr(III)] i  = [U(VI)] i  = 0.5 × 10−3 mol/L, V = 5.0 mL, w = 0.05 g, t = 4 h and T = 22 ± 1 °C

Full size image

Shaking speed effect was more important with La(III) and Pr(III) ions. La(III) and Pr(III) ions recovery were respectively only 17.9 and 30.8% without shaking but at shaking speed near 310 rpm, yield recovery becomes 44.3 and 58.6% respectively (Fig. 2). The effect of increasing agitation is to decrease the film resistance to mass transfer surrounding the sodium phosphonate resin beads (PEIPR-Na), which would be beneficial to the U(VI), La(III) and Pr(III) ions sorption onto the resin. Therefore, a shaking speed of 310 rpm was maintained for the subsequent experiments to ensure the invariance of this parameter.

Effect of time contact

The equilibrium sorption time of La(III), Pr(III) and U(VI) on investigated resin (PEIPR-Na) was studied. As shown in Fig. 3, the period of 30 min can be considered as the optimum time for La(III) sorption, while for Pr(III) and U(VI), 250 min is necessary to attain the equilibrium states. Also, high sorption rate and capacity were observed towards U(VI) ions than Pr(III) and La(III). The sorption capacities of PEIPR-Na reached in this study were 1.0, 6.2 and 18.2 mg of La(III), Pr(III) and U(VI) respectively per one gram of resin. Therefore, a time contact of 250 min was maintained for the subsequent experiments to ensure the invariance of this parameter. Contact times, required to reach maximum sorption on PEIPR-Na, were lower than those of many sorbents cited in literature [1, 2, 20, 26].

Fig. 3
figure 3

Removal of La(III), Pr(III) and U(VI) by PEIPR-Na as a function of time, [La(III)] i  = [Pr(III)] i  = [U(VI)] i  = 0.5 × 10−3 mol/L, V = 5.0 mL, w = 0.05 g, θ = 310 rpm and T = 22 ± 1 °C

Full size image

The results of the contact time studies on sorption of La(III), Pr(III) and U(VI) on PEIPR-Na resin (Fig. 3) are further analyzed using the standard kinetic models, like the Lagergren first order equation and pseudo-second order type, to understand the rate controlling step in the sorption process [26]. The linearized integrated form of the first-order rate equation for the boundary conditions t = 0 and t = t is given as (Eq. 4):

$${\text{Ln}}(q_{\text{e}} - q_{t}) = {\text{Ln}}q_{\text{e}} - k_{1} t$$
(4)

where q e is the amount of La(III), Pr(III) and U(VI) sorbed (mg/g) on the functionalized resin at equilibrium and K 1 is the rate constant of first order adsorption (1/min). Similarly, the relation for linearized integrated form of pseudo-second order kinetic model is given in Eq. (5).

$$\frac{t}{{q_{t}}} = \frac{1}{{k_{2} q_{\text{e}}^{2}}} + \frac{1}{{q_{\text{e}} }}t$$
(5)

The units of k 2 in the pseudo-second order model is g/mg min [26]. The correlation constant (r) computed from the trend line drawn from data of Fig. 3 were 0.834, 0.969 and 0.969 for first order and 0.959, 0.996 and 0.999 for the pseudo-second order kinetic model for La(III), Pr(III) and U(VI) sorption respectively (Table 1). The value of correlation constant clearly indicates that the sorption of La(III), Pr(III) and U(VI) species on PEIPR-Na resin follow the pseudo-second order kinetics better than the first order model implying that the rate-controlling step in the exchange process is the chemisorptive forces between the metal nitrate ions in the solution phase and the active cites (N, O and P) on the resin [27].

Table 1 Kinetic model parameters for La(III), Pr(III) and U(VI) sorption by PEIPR-Na
Full size table

Effect of pH

The pH value is one of the most important factors that affected the metal ions sorption process from aqueous systems by using batch method. The pH of solution controls the surface charge of the sorbent and ionization of the sorbate in solution. The influence of pH on the sorption of La(III), Pr(III) and U(VI), from nitrate solution, was investigated in the range of 0.7–6.8 as is shown in Fig. 4. The results indicate that sorption capacities are lowest at low pH (pH < 2). Highest sorption capacities are observed at pH > 4.3 for La(III) (≈1 mg/g), pH near 4.2 for Pr(III) (≈6 mg/g) and pH between 2.9 and 5.7 for U(VI) (≈19 mg/g). Similar pH intervals were obtained with other sorbent in La(III), Pr(III) and U(VI) sorption [1, 6, 28].

Fig. 4
figure 4

Removal of La(III), Pr(III) and U(VI) by PEIPR-Na as a function of pH, [La(III)] i  = [Pr(III)] i  = [U(VI)] i  = 0.5 × 10−3 mol/L, w = 0.05 g, V = 5 mL, θ = 310 rpm and T = 22 ± 1 °C

Full size image

Uptake of La(III), Pr(III) and U(VI) ions at slightly acidic conditions may be explained to proceed via complex formation between these ions and the active sites on the resin. This sorption trend can likely be ascribed to the effect of competitive binding between La(III), Pr(III) or U(VI) and hydrogen ions for the binding sides on the surface of the resins. The low sorption capacity at low pH is due to the competition between the excess of H3O+ ions in the medium and positively charged cationic species (mainly presented as [Ln(H2O)n]3+ and UO2 2+ present in solution, where Ln = La(III) or Pr(III) [1, 29].

The literature shows that the pKa value of phosphonic acid is about 2 and 5 [30]. Thus, slightly acidic conditions favoured sorption of La(III), Pr(III) and U(VI) ions onto functionalized resin as shown by earlier researchers [1, 6, 31]. This is due to the presence of negatively charged functional groups (phosphonate) on the resin in a slightly acidic medium, more negative sites are becoming available for the sorption of La(III), Pr(III) and U(VI), which leads to the uptake increases. At higher pH, La(III), Pr(III) and U(VI) uptake decrease because ions lead to more stable hydroxides complexes in aqueous phase like [Ln(OH)2]+, [Ln(OH)]2+, [Ln(OH)4]−, Ln(OH)3, [UO2(OH)]+, [(UO2)2(OH)2]2+, and [(UO2)3(OH)5]+. For example, formation of hydroxocomplexes of U(VI) could be explained by the following equilibriums (Eqs. 6, 7 and 8) [32–34].

$${\text{UO}}_{2}^{2 + } + 2{\text{H}}_{2} {\text{O}} \leftrightarrow [({\text{UO}}_{2} )({\text{OH}})]^{ + } + {\text{H}}_{3} {\text{O}}^{ + }$$
(6)
$$2{\text{UO}}_{2}^{2 + } + 4{\text{H}}_{2} {\text{O}} \leftrightarrow [({\text{UO}}_{2} )_{2} ({\text{OH}})_{2} ]^{2 + } + 2{\text{H}}_{3} {\text{O}}^{ + }$$
(7)
$$3{\text{UO}}_{2}^{2 + } + 10{\text{H}}_{2} {\text{O}} \leftrightarrow [({\text{UO}}_{2} )_{3} ({\text{OH}})_{5} ]^{ + } + 5{\text{H}}_{3} {\text{O}}^{ + }$$
(8)

Temperature dependence of sorption

Temperature affects directly the removal yield of La(III),Pr(III) and U(VI) ions onto functionalized resin. In the present investigation the sorption experiments were performed at three different temperatures 22, 27 and 35 °C. It was found that the removal yields of La(III), Pr(III) and U(VI) ions increased from 13.7, 49.2 and 41.0% to 49.2, 60.6 and 81.8% respectively with increasing temperature from 22 to 35 °C (Fig. 5). This result is expected; at higher temperature, ions are moving faster, which retards specific or electrostatic interactions that become weaker. The ions become smaller because solvation is reduced. The breaking of bonds on the resin surface and increase in the diffusion rate of La(III), Pr(III) and U(VI) ions across the external boundary layer and in the internal pores of the sorbent. The NH, N and –P(O)(O−Na+)2 groups of PEIPR-Na are partially protonated at low temperatures but deprotonation degrees increase at high temperature resulting in an increase metal sorption capacities [35–37].

Fig. 5
figure 5

Yield removal of La(III), Pr(III) and U(VI) by PEIPR-Na as a function of temperature, [La(III)] i  = [Pr(III)] i  = [U(VI)] i  = 0.5 × 10−3 mol/L, w = 0.05 g, V = 5.0 mL and θ = 310 rpm

Full size image

Thermodynamic parameters

Assuming that the activity coefficients are unity at low concentrations, thermodynamic parameters (\(\Delta G^{ \circ } , \Delta H^{ \circ }\) and \(\Delta S^{ \circ } )\) were calculated using Eq. (3) and the following relations (Eqs. 9, 10 and 11) [38]:

$$\Delta G^{ \circ } = \Delta H^{ \circ } - T\Delta S^{ \circ }$$
(9)
$$\Delta G^{ 0} = - RT\,{\text{Ln }}D$$
(10)
$${\text{Ln}}\,D = \frac{{ - \Delta H^{ \circ } }}{R} \times \frac{1}{T} + \frac{{\Delta S^{ \circ } }}{R}$$
(11)

where R is the universal gas constant (8.314 J/K mol), T is the absolute temperature (K), \(\Delta G^{ \circ } , \Delta H^{ \circ } {\text{and }}\Delta S^{ \circ }\) are changes in standard free energy, standard enthalpy, and standard entropy, respectively. \(\Delta H^{ \circ }\) and \(\Delta S^{ \circ }\) were obtained from the linear plot of Ln D versus 1/T (Fig. 6) and presented in Table 2.

Fig. 6
figure 6

Ln D of La(III), Pr(III) and U(VI) removal by PEIPR-Na as a function of 1/T, [La(III)] i  = [Pr(III)] i  = [U(VI)] i  = 0.5 × 10−3 mol/L, w = 0.05 g, V = 5 mL and θ = 310 rpm

Full size image
Table 2 Thermodynamic parameters for the sorption of La(III), Pr(III) and U(VI) on PEIPR-Na
Full size table

The positive values of ΔS° reflect the big affinity of PEIPR-Na resin for La(III), Pr(III) and U(VI) and suggests increased randomness at the solid-solution interface during sorption. The positive values of ΔH° show endothermic nature of sorption process. The Gibbs free enthalpy change ΔG° was less than 0 indicating that the adsorption process was spontaneous. On the other hand, the observed decrease in the negative values of ΔG° with elevated temperature indicates that the adsorption is more favorable at higher temperatures (see Table 1) [39–41].

Effect of initial metal concentration

The sorption equilibrium of metals ions between aqueous solution and the resin can be described by a sorption isotherm. The sorption experiments were performed using different initials concentrations of La(III), Pr(III) and U(VI) at 22 ± 1 °C. La(III), Pr(III) and U(VI) sorption isotherms on functionalized resin are presented in Fig. 7 as a function of the initial concentration of metals ions in the aqueous medium. The amounts of La(III), Pr(III) and U(VI) sorbed per unit mass of the resin increased with the initial metals concentrations as expected, to reach the plateau values that represent saturation of the active groups which are available for interaction with La(III), Pr(III) and U(VI). In other terms, to obtain the maximum sorption capacities for the interested metal ions, the initial concentration were increased from 1.0 × 10−5 to 1.2 × 10−3 mol/L for La(III), from 5.0 × 10−6 to 1.2 × 10−3 mol/L for Pr(III) and from 5.0 × 10−6 to 1.0 × 10−2 mol/L for U(VI). For both metals, resin were saturated at relatively low concentrations (1.0 × 10−3 mol/L with La(III), 0.5 × 10−3 mol/L with Pr(III)) and 5.0 × 10−3 mol/L with U(VI) indicating strong binding for these metal ions. Experimental sorption capacities (q m(exp.)) are 1.06, 6.23 and 47.30 mg/g resin for La(III), Pr(III) and U(VI), respectively. Pr(III) and U (VI) are quantitatively removed at a metal concentration equal or less than 0.5 × 10−4 and 0.1 × 10−4 mol/L, respectively, this behavior indicates that PEIPR-Na resin is very effective in Pr(III) and U(VI) sorption from dilute solutions unlike La (III), even at a concentration equal to 0.5 × 10−4 mol/L the sorption is not complete. The curves show that the uranyl ions are more extracted than lanthanum and praseodymium, in the considered concentration range.

Fig. 7
figure 7

Sorption isotherm of La(III), Pr(III) and U(VI) on PEIPR-Na as a function of initial La(III), Pr(III) and U(VI) concentrations, V = 5 mL, w = 0.05 g, θ = 310 rpm, T = 22 ± 1 °C

Full size image

The sorption data were applied tomodels to understand the sorption mechanism. The Langmuir and Freundlich models are expressed as Eqs. (12) and (13), respectively [38]:

$$\frac{{C_{\text{e}} }}{{q_{\text{e}} }} = \frac{1}{{K_{\text{L}} q_{\text{m}} }} + \frac{{C_{\text{e}} }}{{q_{\text{m}} }}$$
(12)
$${\text{Ln}}q_{\rm{e}} = {\text{Ln}}K_{\rm{F}} + \frac{1}{n}{\text{Ln}}C_{\rm{e}}$$
(13)

where q m (mg/g) is the maximum specific uptake corresponding to the site saturation, q e (mg/g) and C e (mg/L) are the amount of sorbed metal on the modified resin and the concentration of metal in the solution at equilibrium time, respectively.

The K L (L/mg) and K F are the Langmuir and Freundlich sorption isotherm model coefficient. The K L is the constant related to the free energy of adsorption and represents the affinity between the metal and functionalized resin, and the K F is the constant indicative of the relative adsorption capacity of the absorbent (mg/g), respectively. The magnitude of 1/n gives a measure of favorability of adsorption and the value of 1/n less than 1 represent a favorable sorption. The isotherm parameters calculated from the fitting are given in Table 3.

Table 3 Isotherm model parameters for La(III), Pr(III) and U(VI) sorption by PEIPR-Na
Full size table

Is concluded from the higher correlation coefficients (r = 1 and 0.999 respectively) that the Langmuir equation fits the data better than the Freundlich model for both Pr(III) and U(VI) sorption from which the theoretical maximum sorption capacities (q m) are very close to the experimental values (see Table 2). The Langmuir model indicates that the sorbed Pr(III) and U(VI) is uniformly distributed in a monolayer coverage of the surface of the sorbent [38]. While for La(III), the Freundlich equation fits the data better than the Langmuir model (r = 0.992) indicating a multilayer adsorption nature of this metal ions on PEIPR-Na resin [38]. For the study the value of 1/n = 0.391 exhibits the same trend of a beneficial sorption (Table 3). It is noteworthy that this resins shows significantly similar or higher sorption capacities, towards La(III), Pr(III) and U(VI) as compared with other sorbents as seen in Table 4.

Table 4 A comparison of the present method with other solid phase sorption and separation chromatographic method for the sorption of La(III), Pr(III) and U(VI)
Full size table

Effect of salts addition

Hydrometallurgical and nuclear effluents often contain a considerable amount of ions especially sodium, potassium, nitrate, sulphate and others which affect the affinity between the sorbent and the sorbate. Therefore, investigation of the ions effect on the La(III), Pr(III) and U(VI) sorption onto PEIPR-Na sorbent is important. To investigate the effect, different sodium and/or potassium salts (chloride, nitrate, cyanide, thiocyanate, sulfate and dithionite) are added in the aqueous phase at the same concentration (1.0 mol/L). La(III), Pr(III) and U(VI) concentrations were fixed at 0.5 × 10−3 mol/L.

Figure 8 shows the La(III), Pr(III) and U(VI) sorption efficiency with various salts in aqueous solutions. From this figure, we note that both La(III), Pr(III) and U(VI) sorption efficiency increased with addition of KCl, NaNO3, NaCl and KSCN and decreased when we add Na2SO4, KCN and Na2S2O4 salts in the aqueous solution.

Fig. 8
figure 8

Sodium and potassium salts addition effect on La(III), Pr(III) and U(VI) sorption by PEIPR-Na, [Salt] = 1.0 mol/L, w = 0.05 g, V = 5.0 mL, [La(III)] i  = [Pr(III)] i  = [U(VI)] i  = 0.5 × 10−3 mol/L, θ = 310 rpm, T = 22 ± 1 °C

Full size image

Generally, this phenomenon might be attributed to the following reasons. One is that the formation of nonextractable anionic La(III), Pr(III) and U(VI) species at higher free anionic concentrations (NO3 −, Cl−, SCN−, CN− SO4 2− and S2O4 2−). The other is that ionic strength of solution can influence the activities coefficients of La(III), Pr(III) and U(VI) which limits their transfer to the sorbent.

The order of the increases effect is given as:

KCl < NaNO3 < NaCl < KSCN for Pr(III), KSCN < NaCl < KCl < NaNO3 for La(III) and KSCN < NaCl = KCl = NaNO3 for U(VI), while the order of the decreases effect is given as: Na2SO4 < KCN < Na2S2O4 for Pr(III), Na2S2O4 < Na2SO4 < KCN for La(III) and Na2S2O4 < Na2SO4 < KCN for U(VI).

It’s important to note that: (i) Addition of Na2S2O4, in aqueous solution, prevents Pr(III) sorption (Removal yield = 0%), (ii) Addition of NaCl, KCl and NaNO3 in aqueous solution makes U(VI) sorption quantitative (Removal yield = 100%).

Effect of salts concentration

From precedent figure (Fig. 8), it can be seen that NaNO3 have most positive effect on the sorption process of La(III) and U(VI), while KSCN makes the same effect on Pr(III) sorption. Therefore, investigation of the NaNO3 and KSCN salts concentration as a criterion of the ionic strengths on the La(III), U(VI) and Pr(III) sorption onto PEIPR-Na is important. Figure 9 shows respectively the effect of sodium nitrate and potassium thiocyanate concentrations on the removal yield of La(III), U(VI) and Pr(III), onto synthesised resin at the initials La(III), U(VI) and Pr(III) concentrations of 0.5 × 10−3 mol/L. As it can be seen from this figure, increasing salts concentration from 0 to 1.0 mol/L has positive effect on the sorption percentage of both La(III), U(VI) and Pr(III) ions, whereas, at salts concentration between 1.0 and 2.0 mol/L, the removal yield sorption does not change significantly for both ions.

Fig. 9
figure 9

Effect of NaNO3 and KSCN salts concentration on La(III), Pr(III) and U(VI) sorption on PEIPR-Na resin, w = 0.05 g; V = 5.0 mL, [La(III)] = [La(III)] i  = [Pr(III)] i  = [U(VI)] i  = 0.5 × 10−3 mol/L, θ = 310 rpm, T = 22 ± 1 °C

Full size image

Uranium(VI), lanthanum(III) and praseodymium(III) desorption

Effect of eluent nature

Desorption is crucial to reuse functionalized resin and it will decrease the process cost. Figure 4 has actually illustrated that low La(III) Pr(III) and U(VI) sorption was observed at low pH values; a matter which implies that desorption from the loaded PEIPR-Na would occur with changing the nature of acid and increasing her concentration [12, 26]. For this purpose, three mineral acids, HCl, HNO3 and H2SO4, at the same concentration (1.0 mol/L), were examined as eventual eluents by batch method in one cycle and at room temperature using 5 mL acid volume for 0.05 g loaded PEIPR-Na. Results are summarised in Fig. 10. From this figure, we can note that:

Fig. 10
figure 10

Effect of eluent nature on La(III), Pr(III) and U(VI) elution, w = 0.05 g, v = 5.0 mL, [Acid] = 1.0 mol/L, t = 4 h, θ = 310 rpm, T = 22 ± 1 °C

Full size image

  • The stripping percentage of La(III) using sulphuric acid, at 1.0 mol/L was very good (87.5%),

  • The stripping percentage of Pr(III) using both hydrochloric, sulphuric and nitric acids, at 1.0 mol/L, was medium and do not exceed 65% with HCl solution,

  • HCl solution (1 M) was the better eluent of U(VI) from PEIPR-Na resin (73.1%),

  • The order of the increases in the stripping percentage of La(III), Pr(III) and U(VI) are given as:Hydrochloric acid (57.8%) <nitric acid (85.6%) sulphuric acid <(87.5%) for La(III), sulphuric acid (55.6%) <nitric acid (61.9%) <hydrochloric acid (64.2%) for Pr(III) and sulphuric acid (24.5%) <nitric acid (59.9%) <hydrochloric acid (73.1%) for U(VI).

Thus, H2SO4 solution was selected as eluent for La(III) and HCl for Pr(III) and U(VI) in order to study the optimum eluent concentrations.

Effect of eluent concentration on U(VI), La(III) and Pr(III) elution

The recovery of La(III), Pr(III) and U(VI) from loaded PEIPR-Na was studied using, respectively, different concentrations of sulphuric (for La(III)) and hydrochloric acids (for Pr(III) and U(VI)) in one cycle and in the range 0.01–5.0 mol/L. The results obtained are presented in Fig. 11. From this figure we note that recovery of both La(III), Pr(III) and U(VI) increased with increasing acid concentration and reached 94.3% at a concentration of 5.0 mol/L using H2SO4 for La(III). For Pr(III) and U(VI), elution yields reached 72.5 and 84.3% respectively using HCl 5.0 mol/L. Analyzing the results obtained, it can be observed that for the recovery of lanthanum ions a 2 M H2SO4 solution can be used and for the recovery of praseodymium and uranyl ions from the same sorbent it is appropriate to use a solution of 2 mol/LHCl.

Fig. 11
figure 11

Effect of leachant concentration on La(III), Pr(III) and U(VI) elution, w = 0.05 g, v = 5.0 mL, t = 4 h, θ = 310 rpm, T = 22 ± 1 °C

Full size image

Conclusions

In this study, the sorption capacities of La(III), Pr(III) and U(VI) on a novel polyethylenimine sodium phosphonate resin (PEIPR-Na) were studied by batch tests conducted under various experimental conditions such as shaking speed, contact time, pH, initial La(III), Pr(III) and U(VI) content, temperature and presence of another salt in aqueous solution. Based on the results, the following conclusions are summarized as below:

  • The sorption of the investigated metal ions increases by increasing the shaking speed and contact time.

  • The equilibrium batch experiment data demonstrate that PEIPR-Na is most selective towards U(VI) than Pr(III) and La(III). The maximum sorption capacities were 47.30 mg g−1 for U(VI), 6.23 mg/g for Pr(III) and only 1.06 mg/g for La(III) under the given experimental conditions.

  • The sorption of La(III), Pr(III) and U(VI) on the resin is strongly dependent on pH medium. Sorption of both La(III), Pr(III) and U(VI) are low in more acidic medium.

  • The calculated thermodynamic parameters showed the feasibility, endothermic and spontaneous nature of the sorption of both La(III), Pr(III) and U(VI) ions.

  • La(III), Pr(III) and U(VI) sorption efficiency increased with addition of KCl, NaNO3, NaCl, and KSCN and decreased when we add Na2SO4, KCN and Na2S2O4 salts in the aqueous solution. Presence of Na2S2O4, in aqueous solution, prevents Pr(III) sorption (Removal yield = 0%) while presence of KCl, NaNO3 and NaCl makes uranyl sorption quantitative (100%). This enables the possibility of separation of Pr(III) from aqueous solutions containing La(III) and U(VI).

  • The results of La(III), Pr(III) and U(VI) ions desorption reported that: (i) At low acid concentration (≤0.1 mol/L) Pr(III) and U(VI) were most desorbed (≥58.0%) while desorption of La(III) do not exceed 47.0%. (ii) At high acid concentration (≥0.5 mol/L), results were inversed, and La(III) and U(VI) were most desorbed (94.3% for La(III) and 84.3% for U(VI)) while desorption of Pr(III) do not exceed 72.5%. This enables the possibility of separation of La(III), Pr(III) and U(VI) from aqueous solutions.

  • Quantitative sorption of Pr(III) and U(VI), from dilute solutions (concentration ≤10−5 mol/L), suggests the possible application of the resin in the pre-concentration of minor Pr(III) and U(VI) from environmental/laboratory waste samples.

  • The results obtained in this study make PEIPR-Na as promising candidate for sorption, immobilization and pre-concentration of U(VI) ions from large volume of solutions containing La(III) and Pr(III).