Introduction

Gadolinium is usually considered as one of the lanthanide series in the periodic table, and it has a lot of uses in the field of structural components, fluorescent materials, mechanical devices, electronic industry, and nuclear industry (Singha et al. 2014). Gadolinium is frequently found in nature in two alternative types of ores such as bastnasite and monazite (Zamani et al. 2012). In nuclear field, gadolinium is used as a neutron absorber concerning the control level regulations of nuclear reactors (Rufus et al. 2018). Gd-153 isotope is utilized in X-ray fluorescence in addition to osteoporosis screening. Gadolinium is a gamma-emitter including half-life time about 8 months, so it is applicable for medical disciplines (Othersen et al. 2007). As consequence of utilizing radiotracers, they may leak into the encompassing environment causing poisonous impacts such as radioactive contamination. In addition, radiotracers may lead to a great damage to the body organs as a result of direct contact during treatment or in other methods applied to human beings (Singha et al. 2014).

Europium radioisotopes are used as burn-up monitors to assess the performance of reactors fuels (Kazakov et al. 2018). Europium is used in several fields such as material science and electronics. The struggle between growing requirement in a variety of industries and also the restricted amount of Gd(III) and Eu(III) resulted in an excessive growth in the demand for the recovery of both elements. Therefore, the determination, separation, and recovery of Eu(III) and Gd(III) are very critical due to their existence at very trace concentrations (Aghamohammadhasan et al. 2017).

There are many analytical instruments presented for the determination of gadolinium and europium in different industrial, geological, and environmental samples such as inductively coupled plasma optical emission spectrometry (ICP-OES) (Li and Hu 2010), electron spin resonance (ESR) (Moyer and McCarthy 1969), spectrofluorometry (Aghamohammadhasan et al. 2017), and neutron activation analysis (Bartolini et al. 2003). Despite the exceptional development of instrumental strength for detection and determination of trace elements, the determination of gadolinium and europium in trace concentrations by recent analytical techniques is, in many cases, difficult (Shizhong et al. 2007). This is not solely as a result of the deficient sensitivity of the methods, but additionally due to the matrix interference (Kanna et al. 2005). So as to conquer these troubles, a lot of separation or preconcentration methods, for example, ion exchange, co-precipitation, solvent extraction, solid-phase extraction, resin chelation, etc., are still required preceding the analysis (Liang and Chen 2005).

Solid-phase extraction has grown to be the most widely recognized method for the determination of trace concentrations of REEs in diverse matrices as a result of its extraordinary recovery, little extraction time, and high enrichment factor as compared to liquid–liquid extraction (Liang et al. 2005). Many adsorbent materials, such as chelating resin (Person et al. 2011), modified Amberlite XAD-4 (Karadas et al. 2011), graphene oxide-TiO2 composite (Zhang et al. 2015), and modified polystyrene (Fedyunina et al. 2012) have been proposed for separation and preconcentration of REEs. Among these adsorbents, nanomaterials are applied as adsorbents, with crystallite size less than 100 nm, because of their distinct mechanical, electronic, thermal, and biological properties (Pyrzynska 2013). Among these nanomaterials, trimanganese tetraoxide, Mn3O4 (hausmannite), is considered to be the maximum stable manganese oxide at high temperatures in contrast of other manganese oxides, for example, MnO2 and Mn2O3 (Mansournia et al. 2015). Hausmannite is a brown to black metallic mineral (Bastami and Entezari 2010). At room temperature, it has spinel structure and a tetragonal phase, wherever the Mn3+ and Mn2+ ions occupy octahedral and tetrahedral positions within the spinel structure, respectively (Vázquez-Olmos et al. 2005; Belkhedkar and Ubale 2014). Furthermore, Mn3O4 has been applied for manufacturing of some magnetic materials like manganese zinc ferrite (Chen et al. 2014). Lately, a lot of interest has been devoted for the synthesis of manganese oxides nanoparticles. Mn3O4 NPs are widely applied in different applications such as catalysis (Lei et al. 2007), Ion exchange, molecular adsorption (Zhang et al. 2010) due to their structural flexibility and unique properties. Manganese oxide (Mn3O4 NPs) is an important scavenger for heavy metal removal from soil, sediments, and rock because of its principal sorptive properties such as high porosity, large surface area, as well as the presence of more active sites for binding with metal ions (Liu et al. 2015).

In the previous studies, numerous materials have been used as solid-phase extractants for the preconcentration, recovery, and sorption of gadolinium such as modified silica (El-Sofany 2008), Amberlite (Gad et al. 2017), and gibbsite (Huittinen et al. 2009). Several adsorbents have been applied for the sorption and separation of Eu(III) such as imprinted polymer (Alizadeh and Amjadi 2013), 1-(2-pyridylazo)-2-naphthol (Mallah et al. 2008), di(2-ethylhexyl)phosphoric acid and Triton X-100 (Ohashi et al. 2007), and graphene oxide nanosheet (Sun et al. 2012). Till now, no research has so far been published regarding the use of Mn3O4 nanoparticles (Mn3O4 NPs) as sorbent for europium and gadolinium ions.

This work is focusing on the possible use of nano Mn3O4 as solid-phase extractant for the sorption and preconcentration of europium and gadolinium ions from aqueous solutions. The parameters affecting the sorption of gadolinium and europium like pH, shaking time, metal concentration, and temperature were examined. The obtained results are discussed according to different kinetic and isotherm models. Furthermore, the possible use of Mn3O4 for preconcentration of Gd and Eu from aqueous solutions was carried out and discussed.

Experimental

Synthesis of nanosized Mn3O4

The synthesis of Mn3O4 nanoparticles was performed using the co-precipitation method (Shrividhya et al. 2014). First, 1.0 M of MnSO4. H2O was dissolved in certain volume of de-ionized water, and then, 2.0 M of NaOH was added drop wise to the previous solution. Ammonia was added to adjust pH at 11 ± 0.2. To precipitate Mn3O4 nanoparticles, the solution should be stirred for 2 h at certain temperature (60 °C). The precipitated nanoparticles were gathered, washed with de-ionized water and ethanol 2–3 times, and dried in hot air oven at 100 °C for 12 h. Finally, Mn3O4 NPs were annealed for 2 h at 300 °C.

The formation of Mn3O4 nanoparticles in the alkaline aqueous solution may be described by Eqs. (1–3) (Dhaouadi et al. 2012):

$${\text{Mn}}^{2 + } + 2{\text{OH}}^{ - } \to {\text{Mn}}\left( {\text{OH}} \right)_{2} ,$$
(1)
$${\text{Mn}}\left( {{\text{OH}}} \right)_{{\text{2}}} \xrightarrow{{100\,\,{}^{ \circ }{\text{C}}}}{\text{MnO }} + {\text{ H}}_{{\text{2}}} {\text{O}},$$
(2)
$${\text{3 MnO }} + {\text{ }}0.{\text{5 O}}_{{\text{2}}} \xrightarrow{{300\,\,{}^{ \circ }{\text{C}}}}{\text{Mn}}_{{\text{3}}} {\text{O}}_{{\text{4}}} .$$
(3)

First, manganese ions are reduced in alkaline medium to Mn(OH)2, secondly Mn(OH)2 decomposed into MnO at 100 °C for 12 h, and finally, in the presence of atmospheric oxygen at muffle furnace, MnO is oxidized to Mn3O4 at 300 °C.

Reagents

Manganese sulfate monohydrate (Sigma-Aldrich, USA) and sodium hydroxide (Sigma-Aldrich, USA) were used without purification. Standard stock solution of Gd (1000 mg/L) was used. Eu standard solution (1000 mg/L) was prepared by dissolving certain weight of Eu2O3 (Sigma-Aldrich, USA) in concentrated HNO3 and diluting to the required volume with 18 MΩ de-ionized water.

Instrumentation

The characterization of nanosized Mn3O4 was conducted with several experimental techniques: X-ray diffraction (XRD) patterns of nanosized Mn3O4 were analyzed using BRUKER X-Ray Diffractometer (Germany) of type AXS D8 ADVANCE (Cu target, λ = 1.540 Å) in the range of 2θ = 10°–80° at 40 kV potential and 40 Ma current. Elemental composition was detected by an Oxford energy-dispersive X-ray (EDX) spectrometer (Oxford Link ISIS) attached to the SEM (Model Jeol JSM-5600 LV) using a Si/Li detector with Be window. Transmission electron microscope (TEM) images of samples were obtained using a Hitachi HT-7700 microscope at voltage 100 kV. Samples were prepared by depositing a drop of colloidal suspension of sorbent in ethanol onto a carbon-coated copper TEM grid. The morphology of Mn3O4 NPs was examined using a Hitachi, SU8000-Field Emission Scanning Electron Microscope (FE-SEM). Fourier Transform Infrared (FTIR) spectra were obtained by a spectrometer (NICOLET iS10 model) within the wave number from 400 to 4000 cm−1 on KBr pellets. The initial concentration and final concentration of Gd(III) and Eu(III) were measured by Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES), Prodigy, USA. Quantachrome, Nova 1100e series, version 2.1, USA, was utilized for determination of the surface area and porosity of Mn3O4 NPs. pH meter (Cyperscan 500 pH, USA) which was preliminary calibrated using suitable buffer solutions was used for measuring the pH value of aqueous solutions.

Procedures for sorption of Gd and Eu

The standard solutions of Gd or Eu ions (20–400 mg/l) were pipetted into different bottles. The solutions were adjusted in different pH values (3.0–10.0) using 0.01–0.1 M HCl and/or NaOH solutions. After that, the new sorbents Mn3O4 NPs were added with dose (0.01–0.25 g/20 ml) to Gd or Eu solution at room temperature of 25 °C. The mixture was vigorously shacked for prefixed time (5–300 min). After centrifugation, the residual gadolinium or europium ion concentration in the filtrate was analyzed by ICP-OES. The removal or uptake percentage (R %) can be calculated from the following equation:

$$R\% = \left( {\frac{{C_{\text{o}} - C_{\text{e}} }}{{C_{\text{o}} }}} \right) \times 100.$$
(4)

The sorption capacity of Gd(III) or Eu(III) ions by Mn3O4 NPs was calculated from the following equation:

$$q_{\text{e}} = \left( {C_{\text{o}} - C_{\text{e}} } \right) \times \frac{V}{m},$$
(5)

where qe (mg/g) is the amount of Gd(III) or Eu(III) adsorbed per unit mass of Mn3O4 NPs, Co is the initial Gd(III) or Eu(III) concentration (mg/l), Ce is the equilibrium concentrations of Gd(III) or Eu(III) (mg/l), V (liters) is volume of Gd(III) or Eu(III) solution, and m (g) is the weight of sorbent.

Desorption procedures

Mn3O4 NPs were tested for their re-usability after the removal europium and gadolinium ions from aqueous solutions. After the adsorption experiment was carried out as previously mentioned in “Procedures for sorption of Gd and Eu”, the metal-loaded Mn3O4 NPs were rinsed twice in de-ionized water, and then, different concentrations of hydrochloric acid, nitric acid, and EDTA were added as desorbent agents. The sample was shacked for specific time. Finally, the sample was centrifuged for 20 min at 6000 rpm and the filtrate was analyzed for the released concentration of Eu(III) or Gd(III) using ICP-OES.

The desorption efficiency was calculated using the following equation:

$${\text{Desorption efficiency }}\left( \% \right) \, = \frac{\text{release metal concentration}}{\text{initially sorbed metal concetration}} \times 100.$$
(6)

Results and discussion

Characterization of Mn3O4 nanoparticles

The nanosized Mn3O4 used as sorbents was characterized in terms of chemical analysis and structural composition.

Chemical analysis

Energy-dispersive X-ray spectroscopy (EDX)

The composition of synthesized Mn3O4 NPs was analyzed using energy-dispersive X-ray (EDX) spectrometer as shown in Fig. 1. From the results given in Fig. 1, it was concluded that the synthesized Mn3O4 was composed only from Mn and O. No other peaks for any elements as impurities have been found in the EDX spectrum, which demonstrates that the Mn3O4 NPs are of high purity.

Fig. 1
figure 1

EDX spectrum of Mn3O4 NPs

Full size image

Structural analysis

X-ray diffraction

All the peaks of XRD pattern of Mn3O4 NPs (shown in Fig. 2) can be perfectly matched with the standard diffraction data of hausmannite structure of Mn3O4 [JCPDS Card 01-1127 (D)] and also matched well with literature data (Liu et al. 2015). In addition, no any peaks corresponding to impurity phases have been detected, indicating the high purity and good crystallinity of Mn3O4 NPs. Furthermore, the average crystallite size was measured to be 30–35 nm (as compared to 50 nm which was obtained by Dhaouadi et al. 2012) in relation to the five main peaks (112), (103), (211), (224), and (321) using the Debye–Scherrer equation, which is described as follows:

$$E = 0.9 \, \lambda /\beta \, \cos \theta .$$

where E is the size of crystallite in nanometers, β (FWHM, radian) is the full width at half maximum, θ is the Bragg angle (°), and λ is the wave length of X-ray (λ = 1.5418 Å) (Dhaouadi et al. 2012).

Fig. 2
figure 2

XRD pattern of Mn3O4 NPs

Full size image
Fourier transform infrared (FTIR) analysis

The prepared Mn3O4 NPs were characterized by FTIR analysis, as shown in Fig. 3. Three bands were detected at 415,622 and 514 cm−1, associated with the coupling between Mn–O stretching modes of tetrahedral and octahedral sites (Dubal et al. 2011), and this approves the formation of Mn3O4 NPs. In addition, the bands from 3500 to 1500 cm−1 m can be correlated to the vibration of O–H in the weakly bonded physisorbed water molecules. A wide band at around 3419 cm−1 was related to the stretching vibrations of the O–H bond. It was found that the other two weak peaks at 2344 cm−1 and 1638 cm−1 were totally assigned to the bending vibrations of O–H bond combined with Mn atoms. The peaks at 1117 cm−1 were due to the O–H-bending modes of γ–OH (Sherin et al. 2014). Consequently, the FTIR spectrum confirms the formation of Mn3O4 NPs and this outcome is in a good agreement with data obtained from X-ray diffraction. The IR spectra after the adsorption processes of Gd(III) and Eu(III) ions onto Mn3O4 NPs are also shown in Fig. 3. The FTIR spectrum explained that the peaks were slightly shifted towards higher value. This shift in peak values may be owing to the formation of chemical bond between functional groups present on Mn3O4 NPs and Eu(III) and Gd(III) (Sartape et al. 2017). The band shifting from 1638 up to 1654 cm−1, from 1638 to 1634 cm−1, and from 1117 to 1122 cm−1 and 1124 cm−1 may be due to the interaction between the O–H group of Mn3O4 NPs and Eu(III) and Gd(III). The appearance of the new sharp band around 1383 cm−1 can be apportioned to the O–H···O in-plane bending. The peaks around 410 cm−1, 530 cm−1 and 630 cm−1 correspond to Mn–O–Eu as well as Mn–O–Gd bond, respectively (Al Lafi and Al Abdullah 2015).

Fig. 3
figure 3

FTIR spectrum of a Mn3O4 NPs before adsorption, b after adsorption of Eu(III), and c after adsorption of Gd(III)

Full size image

Morphological study

FE-SEM and HR-TEM analysis

The surface morphology of Mn3O4 NPs was studied using FE-SEM and HR-TEM. From SEM image given in Fig. 4a, it can be observed that the aggregated Mn3O4 NPs have nearly cubic like structure. From HR-TEM image in Fig. 4b, it was shown that the average diameter of Mn3O4 NPs was ranged from 30 to 32 nm. This value is in a good agreement with the data that was obtained from XRD.

Fig. 4
figure 4

a FE-SEM image and b HR-TEM image of the as-prepared Mn3O4 NPs

Full size image

Physical properties of Mn3O4 nanoparticles

The N2 adsorption and desorption isotherm of synthesized Mn3O4 nanoparticles is given in Table S1 (See supplementary materials). The BET surface area of Mn3O4 was calculated to be 13.4 m2/g. The pore diameter was detected in the range of 9.12 nm.

Point of zero charges (pHpzc) of Mn3O4 nanoparticles

pHpzc is the value of pH at which the net surface charge of Mn3O4 nanoparticles in aqueous solutions is zero. The procedures of pHpzc were determined according to the previously reported method by Dalvand et al. (2016). First, six bottles of volume of 20 mL of NaCl (0.01 M) were adjusted at different pH values (2, 4, 6, 8, 10, and 12), using 0.1 M HCl or 0.1 M NaOH. Then, 0.05 g Mn3O4 nanoparticles was added to each bottle and shacked for 2 days. As illustrated in Fig. 5, the intersection point of the initial pH and final pH is called the pHpzc. The pHpzc for Mn3O4 NPs is 4.6, which confirmed the results reported by Durmus et al. (2010). At a pH higher than 4.6, the surface of Mn3O4 NPs becomes negatively charged which gives high preference for the binding between Mn3O4 NPs and the elements under study [Eu(III) and Gd(III)].

Fig. 5
figure 5

pHpzc of Mn3O4 nanoparticles

Full size image

Batch sorption investigations

The sorption of gadolinium and europium ions from aqueous solution with Mn3O4 nanoparticles was affected by different factors such as pH, shaking time, initial metal concentration, and temperature. These factors were studied in a batch adsorption technique.

Effect of hydrogen ion concentration (pH)

The uptake of Gd(III) and Eu(III) ions from aqueous solutions using Mn3O4 NPs was assessed in the ranges of pH from 2.0 to 10.0, which is shown in Fig. 6. It was indicated that the adsorption of Gd(III) and Eu(III) ions on Mn3O4 NPs is increasing by increasing the pH. The maximum removal percentages were achieved at pH 5.0 for Gd(III) and Eu(III). The adsorption of Gd(III) and Eu(III) above pH 5.0 decreased due to precipitation of both elements.

Fig. 6
figure 6

Effect of pH on the sorption of Gd(III) and Eu(III) ions by Mn3O4 NPs (Co = 100 mg/l, V = 20 ml, W = 0.1 g, time = 24 h, T = 25 °C, and shaking speed = 300 rpm)

Full size image

Metal ion concentration effect

The sorption capacity of Mn3O4 NPs for the uptake of Gd(III) and Eu(III) from aqueous solution with initial concentration from 20 to 800 mg/L at pH value 5.0 for Gd(III) and Eu(III) is shown in Fig. 7. The sorption capacity Mn3O4 NPs was noticed to increase to a definite value as a result of the increase in the metal concentration from 20 to 800 mg/L, and this behavior can be explained by the progressive increase in the electrostatic interaction between Gd(III) and Eu(III) ions and Mn3O4 NPs active sites (Laraous et al. 2005). It is obvious that the maximum adsorption capacity of Mn3O4 NPs is acquired at concentration of 300 mg/L for both Gd(III) and Eu(III). Slow stability in the capacity was observed with the increase in Gd(III) and Eu(III) concentration, because all the active sites on the surface of Mn3O4 NPs were saturated, and thus, no Gd(III) and Eu(III) could be adsorbed (Sayed et al. 2017). The sorption capacity of Mn3O4 NPs for Gd(III) and Eu(III) was 11.7 and 25.3 mg/g, respectively.

Fig. 7
figure 7

Effect of initial Gd(III) and Eu(III) concentration on the sorption capacity of Mn3O4 NPs [W = 0.1 g, time = 24 h, V = 20 ml, T = 25 °C, pH = 5.0 (Gd) and pH = 5.0 (Eu), and shaking speed = 300 rpm]

Full size image

Shaking time effect

The sorption of Gd(III) and Eu(III) by Mn3O4 NPs was investigated within shaking time from 30 to 600 min as it is indicated in Fig. 8. The results showed that the uptake of the Gd(III) and Eu(III) includes two successive stages; the first stage includes fast uptake throughout the first 300 min because of the presence of many active sites on Mn3O4 NPs surface. The second stage indicates the equilibrium of sorption process at about 300 min. These two discrete steps were also identified by Burke et al. (2013). The equilibrium time for sorption of Gd(III) and Eu(III) was optimized at 300 and 420 min, respectively.

Fig. 8
figure 8

Shaking time effect on the sorption of Gd(III) and Eu(III) ions by Mn3O4 NPs [W = 0. 1 g, Co = 300 ppm, V = 20 ml, T = 25 °C, pH = 5.0 (Eu) and pH = 5.0 (Gd), and shaking speed = 300 rpm]

Full size image
Kinetic modeling

The mechanism of adsorption with time was discussed through pseudo-first-order model, pseudo-second-order model, and intraparticle diffusion model.

Pseudo-first-order model The Legergren pseudo-first-order kinetic model can be expressed by the following equation (Kowanga et al. 2016):

$${\text{Log}}(q_{\text{e}} - q_{\text{t}} ) = { \log }q_{\text{e}} {-}k_{ 1} t \, / 2. 30 3 ,$$
(7)

where qe is the amount of Gd(III) or Eu(III) adsorbed at equilibrium per unit mass of Mn3O4 NPs (mg/g), k1 is the rate constant of pseudo-first-order equation (min−1), and qt is the amount adsorbed of Gd(III) or Eu(III) per unit mass of Mn3O4 NPs (mg/g) at any time t. We can plot log (qe − qt) versus t, as shown in Fig. S1, from the slope and intercept; qe and t can be calculated.

Pseudo-second-order model The linear form of pseudo-second-order equation can be expressed by the following equation (Plazinski et al. 2013):

$$t/q_{\text{t}} = \, 1/k_{2} q^{2}_{\text{e}} + \, t/q_{\text{e}} ,$$
(8)

where k2 is rate constant of the pseudo-second-order equation (g/mg/min); similarly, qe and qt as defined in case of pseudo-first-order. Plotting of t/qt against t gives the parameters of qe and k2, as illustrated in Fig. S2. Referring to Table S2, it is observed that the correlation coefficients (R2) of Gd(III) and Eu(III) were in most cases greater than that for the pseudo-first-order model. Accordingly, the experimental data of the adsorption kinetics were matched well with the pseudo-second-order kinetic model. Besides, the theoretical sorption capacity (qe, calculated) and the experimental sorption capacity (qe, experimental) values were in good match for second-order model, while for first order, they are different. These results recommended that the adsorption of Gd3+ and Eu3+ ions on Mn3O4 NPs follows the kinetic model of the second-order type.

Intraparticle diffusion model The metal ion transfers from the liquid phase to the solid phase within the sorption process in many steps; film diffusion, pore diffusion, and particle diffusion (Srivastava et al. 2006).

The intraparticle diffusion model (IPD) is expressed by the following equation:

$$q_{\text{t}} = k_{\text{id}} t^{0.5} + \, C,$$
(9)

where C = thickness of the boundary layer, kid = the constant of intraparticle diffusion [mg/(g min1/2)], and qt = the amount of Gd(III) or Eu(III) adsorbed on mass unit of Mn3O4 NPs (mg/g) at time t. From the plot of qt against t0.5 as illustrated in Fig. S3, kid was calculated from the slope (Shrihari et al. 2005).

The IPD kinetics model was applied in predicting the rate-limiting step which is film diffusion or intraparticle diffusion. The rate-controlling step will be intraparticle diffusion, if the plot of qt as opposed to t0.5 passes through the point of origin (Özcan and Özcan 2005). From the data revealed in Table S2, it is observed that the plots in this study show a multi-linearity; first, the plots show an initial rapid portion and then horizontal linear portion. The first sharp step of the plots is as a result of the external surface sorption of Gd(III) or Eu(III) on Mn3O4 NPs. The second part of the plot recommends the gradual adsorption step with controlled intraparticle diffusion. As the lines do not pass through the point of origin, we can conclude that the intraparticle diffusion is not the only rate-regulating step (Mall et al. 2005). This trend of IDP model was also similarly reported by Liu et al. (2013) and Fierro et al. (2008).

Adsorption isotherms

The adsorption isotherm models illustrate the relation between the amount adsorbed by a unit mass of sorbent and the amount of metal ion remaining in the solution at equilibrium. The sorption isotherm of Mn3O4 NPs was studied in terms of the two sorption isotherms: Freundlich and Langmuir models. These models are applied to determine the maximum capacity of Gd(III) and Eu(III) sorption by Mn3O4 NPs. Fitting of the isotherm to the experimental data was evaluated related to the correlation coefficient R2, i.e., the isotherm which gives R2 value close to one is considered to give the best fit. The adsorption isotherm studies were investigated at adsorbent weight (0.1 g of Mn3O4 NPs), temperature (25 °C), varying Gd(III) and Eu(III) concentrations (20–400 mg/l) and pH = 5.0 (Eu) and pH = 4.0 (Gd) (Fierro et al. 2008; Galhoum et al. 2017).

Freundlich isotherm The Freundlich isotherm model is widely applied to illustrate the multilayer sorption on heterogeneous surfaces (Gad et al. 2014).

The Freundlich isotherm is expressed using the following equation:

$$\log \, q_{\text{e}} = \log \, K_{\text{F}} + \frac{1}{n}\log \, C_{\text{e}} ,$$
(10)

where Ce is the concentration of Gd(III) or Eu(III) at equilibrium (mg/l), qe is the amount of Gd(III) or Eu(III) adsorbed using 1 g of Mn3O4 NPs (mg/g), KF is the Freundlich constant (mg/g), and 1/n is the adsorption intensity (g/l).

Langmuir isotherm The model of Langmuir isotherm is applicable in case of sorption on homogenous surfaces. This isotherm assumes that there are a certain number of active sites on the surface of the adsorbent, and thus, it includes formation of monolayer adsorption (Gad et al. 2014). The linear equation of this model is shown in Eq. (11):

$$\frac{{C_{\text{e}} }}{{q_{\text{e}} }} = \frac{1}{{Q_{\text{o}} b}} + \frac{{C_{\text{e}} }}{{Q_{\text{o}} }},$$
(11)

where qe is the amount of Gd(III) or Eu(III) adsorbed per mass unit of Mn3O4 NPs (mg/g), Ce is the Gd(III) or Eu(III) concentration at equilibrium (mg/L), b is the constant of Langmuir isotherm (L/mg), and Qo is the theoretical adsorption capacity (mg/g) (Sadia et al. 2013). Plotting of Ce/qe against Ce provides linear relation as indicated in Fig. S4; from the slope Qo, is obtained and b is obtained from intercept.

As shown in Table 1, Figs. S4 and S5, it is concluded that the adsorption pattern of Gd3+ and Eu3+ on Mn3O4 NPs was fitted well with the Langmuir model with R2 values of 0.996 and 0.997, which is higher than the case of Freundlich model, and it is seen that the adsorption of Gd(III) and Eu(III) is favorable. The high R2 values obtained from the Langmuir model revealed fitness of this model in expressing equilibrium sorption data, in addition to the adsorption capacity obtained from the Langmuir isotherm for Gd(III) is 12.6 mg/g which matched well with the experimental data (11.7 mg/g). Furthermore, the calculated sorption capacity from Langmuir model for Eu(III) is 26.8 mg/g, and this value is matched well with the experimental capacity (25.3 mg/g). In addition, the fitness of the model shows that Eu(III) and Gd(III) are adsorbed as monolayer onto Mn3O4 NPs surface with a distinct number of active sites, which are homogeneously distributed on the surface of Mn3O4 NPs (Mahmoud 2015). The values of KF and n show the increase of negative charge on the surface which enhances the electrostatic forces between Mn3O4 NPs and Gd ions and Eu ions leading to increasing in the adsorption of Gd3+ and Eu3+ (Chen 2015).

Table 1 Langmuir and Freundlich parameters for sorption of Eu(III) and Gd(III) ions onto Mn3O4 NPs
Full size table

Effect of temperature

The effect of temperature on the sorption of Eu3+ and Gd3+ from aqueous solutions was studied by varying the temperature between 25 and 65 °C. The data given in Fig. 9 showed that adsorption of these metal ions by Mn3O4 NPs increased with increase in temperature. The increase in adsorption with increasing temperature up to 65 °C is due to increased penetration of Eu3+ and Gd+ ions inside micropores at higher temperatures or the creation of new active sites (Al-Degs et al. 2008).

Fig. 9
figure 9

Effect temperature on the sorption of Gd(III) and Eu(III) ions by Mn3O4 NPs [W = 0. 1 g, Co = 300 ppm, V = 20 ml, pH = 5.0 (Eu) and (Gd), shaking time = 300 min for Eu(III) and 420 min for Gd(III), and shaking speed = 300 rpm]

Full size image
Thermodynamic studies

The sorption of Eu(III) and Gd(III) by Mn3O4 NPs increases by raising the temperature from 25 to 65 °C and this behavior approved the endothermic nature of the adsorption process. The changes in standard free energy (ΔG), enthalpy (ΔH), and entropy (ΔS) of adsorption were calculated from the following equations:

$$\Delta G \, = \, - RT \, \ln \, K_{\text{C}} ,$$
(12)

where R is the gas constant, KC the equilibrium constant, and T the temperature in K. The KC value is calculated by the following equation:

$$K_{\text{c}} = \frac{{C_{\text{A}} }}{{C_{\text{E}} }},$$
(13)

where CA is the equilibrium concentration of Eu(III) and Gd(III) on Mn3O4 NPs (mg/L) and CE is the equilibrium concentration of Eu(III) and Gd(III) in solution. The enthalpy changes (ΔH) for the process equilibrium and entropy (ΔS) can be calculated from van’t Hoff equation given by the following equation (Gad and El-Sayed 2009):

$$\ln \, K_{\text{C}} = \frac{ - \Delta H}{RT} + \frac{\Delta S}{R}.$$
(14)

By plotting ln Kc against 1/T, straight line is obtained as indicated in Fig. S6. The values of ΔS and ΔH can be obtained from the slope and intercept, respectively. As indicated in Table 2, the values of ΔH and ΔS for the sorption of Eu(III) and Gd(III) onto Mn3O4 NPs were found to be 23.1 kJ/mol and 65.1 J/mol k for Gd(III) and 30.9 kJ/mol and 100.9 J/mol K for Eu(III), respectively. The positive value of ΔH value shows the endothermic nature of the sorption process, while the positive value of ΔS illustrates the increase in the randomness of the system. The negative values of ΔG show the spontaneous nature of the sorption of Gd(III) and Eu(III) by Mn3O4 NPs.

Table 2 Thermodynamic parameters for sorption of Eu(III) and Gd(III) ions onto Mn3O4 NPs
Full size table

Possible preconcentration of Eu(III) and Gd(III)

Preconcentration of metal ion is frequently applied to any metal ion of low concentration before its measurement by any analytical technique. The preconcentration process means that increasing the concentration of analyte by transferring this analyte from large volume solution to small volume solution. The aim of this preconcentration process is increasing sensitivity, lowering detection limit of determination techniques, and removal of the interfering ions from matrix. There are many techniques used for preconcentration of analytes from aqueous solutions such as evaporation, solid-phase extraction, ion exchange, solvent extraction, and co-precipitation. SPE can be applied in different ways, one of these is applied here where the sorbent is added into the sample matrix and shaken for certain time, and then, it is separated from the solution by filtration or centrifugation. To determine the preconcentration factor of Mn3O4 NPs, the effect of eluent type and sample volume should be carried out. Preconcentration factor can be defined from the relation given in Eq. (15):

$${\text{Preconcentration factor }} = \frac{\text{highest sample volume}}{\text{lowest eluion volume}}.$$
(15)

Effect of eluent type

Desorption of the retained Eu(III) and Gd(III) from Mn3O4 NPs was experimented using different eluents like hydrochloric acid, nitric acid, and EDTA with different concentrations.

From the results given in Fig. S7, it can be concluded that 2.0 M HNO3 was found to be the best eluent compared to other solvents for desorption of Eu(III) and Gd(III) from the surface of Mn3O4 NPs.

Effect of the sample volume

It is important to optimize sample volume to acquire a high preconcentration factor to analyze real samples. The effect of solution volume was assessed by preconcentrating different volumes (50–1000 mL) of Gd(III) (20.0 ppm) and Eu(III) (5.0 ppm). Regarding to the results specified in Fig. 10, it can be concluded that when the aqueous solution volume of Gd(III) was up to 700 mL, the recovery percentage was above 95%. Thus, 700 mL was believed to be the highest enrichment sample volume. For Eu(III), at higher sample volumes more than 200 ml, the recovery % decreased gradually. Hence, the best sample volume for Eu(III) was found to be 200 mL. The preconcentration factor is determined as the ratio of the highest sample volume and the lowest final volume (Elvan et al. 2013). As the final volume of eluent for Gd(III) and Eu(III) was 10.0 mL, preconcentration factor of 70 was gained for Gd(III) and of 20 for Eu(III).

Fig. 10
figure 10

Sample volume effect on the uptake of Eu(III) and Gd(III) from aqueous solutions [W = 0. 1 g, Gd(III) concentration = 20 ppm and Eu(III) concentration = 5 ppm, pH = 5.0 (Eu) and pH = 5.0 (Gd), shaking speed = 300 rpm, and shaking time = 300 min for Eu(III) and 420 min for Gd(III)]

Full size image

Conclusion

In the present study, Mn3O4 NPs were successfully synthesized by simple co-precipitation method. The synthesized Mn3O4 NPs were used for the first time as a new sorbent for the sorption and preconcentration of Gd(III) and Eu(III) from aqueous solutions. The experimental data for the sorption of Gd(III) and Eu(III) using Mn3O4 NPs are more fitted to Langmuir isotherm model with maximum sorption capacity reached 26.8 and 12.6 mg/g for Eu(III) and Gd(III), respectively. The adsorption kinetics data were found to follow the pseudo-second-order model and the results obtained from intraparticle diffusion model exposed that the overall process was jointly controlled by external mass transfer and intraparticle diffusion. All the obtained results in this study suggest that nanosized Mn3O4 is a favorable sorbent for uptake of Gd(III) and Eu(III) from aqueous solutions. Thus, this paper shows that Mn3O4 NPs is a promising sorbent for a selective separation of Gd3+ and Eu3+ from aqueous solutions with preconcentration factor up to 70 for Gd(III) and up to 20 for Eu(III).