Introduction

In the context of the worldwide effort to reduce the ecological impact of human activities, ionic liquids (ILs) could be a suitable substitute for volatile organic compounds (VOCs) currently used in many separation procedures [1]. Application of ionic liquids in solvent extraction is being investigated by a great number of scientific groups [1,2,3,4]. Even though the ionic liquids were found to enhance the extraction process, the separation efficiency (expressed by distribution ratios) also depends on the properties of the extracted metal. Strontium, which occurs in aqueous media preferably as the solvated strontium cation or in complex with some anions [5], is difficult to transfer to the organic phase without the addition of a suitable extractant. The most widely used and studied extractants for the separation of strontium from the aqueous to the organic phase are crown-ethers, especially dicyclohexano-18-crown-6 (DCH18C6) or di-tert-butyl-dicyclohexano-18-crown-6 (DTBDCH18C6) [6].

Although strontium extraction using crown-ethers has been studied worldwide [3, 4, 6,7,8,9], the full understanding of the process mechanism was not sufficiently explained, namely regarding the role of the solvent. When VOCs like chloroform or 1-octanol are used as solvents, the transfer of the extracted metal occurs only in the form of an electroneutral molecule. In the case of crown-ethers as extractants, extracted ion associate consists of strontium cation, crown-ether (CE), and charge compensating anions [10,11,12,13] as per Eq. (1).

$${{\text{Sr}}}^{2+}+{\text{CE}}+2 {{\text{NO}}}_{3}^{-}\leftrightarrow {\left[{\text{Sr}}\cdot {\text{CE}}\cdot {\left({{\text{NO}}}_{3}\right)}_{2}\right]}_{{\text{org}}}$$
(1)

The amount of extracted strontium is determined by the solubility of the ion associate in each phase and can be increased by changing the parameters of both phases including the properties of organic solvent. Involving ionic liquids into strontium extraction using crown-ethers, high distribution ratios can be achieved [13,14,15]. Moreover, efficient strontium extraction into imidazolium based ionic liquids was observed even from water, e.g., without the transfer of the charge compensating anion [3].

The possible explanation can be found in presumption of charge compensation by ionic liquid anion [12] (Eq. (2))

$${{\text{Sr}}}^{2+}+{\text{CE}}+2 {\left[{{\text{NTf}}}_{2}\right]}^{-}\leftrightarrow {\left[{\text{Sr}}\cdot {\text{CE}}\cdot {\left({{\text{NTf}}}_{2}\right)}_{2}\right]}_{{\text{org}}}$$
(2)

or in the mechanism of ion exchange of strontium-crown-ether complex with ionic liquid cation (Eq. (3)) that was suggested by Dietz et al. [12] and successfully confirmed by Xu et al. [8] and Jensen et al. [16]

$${\left[{\text{Sr}}\cdot {\text{CE}}\right]}_{{\text{aq}}}^{2+}+2{\left[{{\text{C}}}_{{\text{n}}}{\text{mim}}\right]}_{{\text{org}}}^{+}\leftrightarrow {\left[{\text{Sr}}\cdot {\text{CE}}\right]}_{{\text{org}}}^{2+}+2{\left[{{\text{C}}}_{{\text{n}}}{\text{mim}}\right]}_{{\text{aq}}}^{+}$$
(3)

However, the experiments of the same groups indicated that the extraction does not proceed only by one mechanism [8, 16]. The shift of the prevailing mechanism from ion exchange to the transfer of the uncharged ion associate proceeds with increasing chain length of the [Cnmim]+ cation and is much more significant for alkaline earth metals compared to alkali metals [15].

In the case of ILs with more hydrophobic cation, the situation becomes more complicated. On the basis of experimental observations, Dietz et al. [4] suggested a third extraction mechanism based on the extraction of undissociated acid together with several water molecules in complex with CE resulting in [H3O·CE]+org complex followed by an exchange between H3O+ and the metal cation as per Eq. (4).

$${{\text{Sr}}}_{{\text{aq}}}^{2+}+2{\left[{{\text{H}}}_{3}{\text{O}}\cdot {\text{CE}}\right]}_{{\text{org}}}^{+}\leftrightarrow 2{{\text{H}}}_{3}{{\text{O}}}_{{\text{aq}}}^{+}+{\left[{\text{Sr}}\cdot {\text{CE}}\right]}_{{\text{org}}}^{2+}+{{\text{CE}}}_{{\text{org}}}$$
(4)

The shift of the prevailing mechanism can be assigned also to the composition of aqueous phase. It was found that the distribution ratio of strontium from low-concentration HNO3 solutions (c ~ 10−3 mol L−1) usually reaches relatively high values, and its value decreases rapidly with increasing acidity [3, 8, 9, 13, 17]. This unexpected trend, which is inconsistent with the observed increase in the strontium distribution ratio with acid concentration for VOCs, Xu et al. [8] explains by the competition between Sr2+ and H3O+ cations, the concentration of which increases with acid addition. At the same time, the concentration of the charge compensating anions (e.g. NO3−) increases and eventually becomes high enough to allow extraction of strontium in the form of the uncharged ion associate [Sr·CE·(NO3)2] [17].

In addition to the acid concentration, its anion also plays a role. In a study [17], a comparison of the effect of HNO3 and HCl on the extraction of strontium into ILs using DCH18C6 can be found. If the acid concentrations are within the range of 10−2 to 5·10−1 mol L−1, the difference in the strontium distribution ratios between anions of the acids is negligible, whereas at higher concentrations, the separation efficiency of strontium clearly depends on the acid anion. Compared to the extraction mechanism, Garvey et al. [17] suggested that acid anion significantly affects only the extraction of the uncharged ion associate, which is the prevailing mechanism at higher acidities.

This paper is a continuation of previous research in the field of the application of ionic liquids as solvent for the extraction of strontium from acidic solutions using crown ethers [18]. Promising results of strontium extraction, even from oxalic and citric acid environments, have revealed the potential application of ionic liquids for the recovery (regeneration) of spent decontamination solutions. The direct involvement of ionic liquids in the extraction mechanism significantly improves the strontium separation efficiency compared to VOCs systems, especially from weakly acidic solutions. However, a deeper understanding of the extraction mechanism requires a more detailed study of selected parameters, which is the focus of this paper.

Experimental

Materials

The ionic liquids, 1-alkyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl]imides, [Cnmim][NTf2], n = 2, 4, 6, 7, 10, were purchased from Iolitec, Germany, with a purity of more than 99%. Chloroform (with a purity of more than 99%) and benzene (p.a.) were procured from VWR Chemicals, USA, and Lachema Brno, Czechia, respectively. DCH18C6 (with a purity of 98%) was obtained from Sigma Aldrich as a mixture of isomers. All other chemicals were purchased with p.a. purity grade—SrCl2·6H2O from Lachema Brno, Czechia, HNO3 and HCl from Lach-Ner, Neratovice, Czechia, and HClO4 from Sigma Aldrich. A 85Sr tracer stock solution was prepared from commercially available 5·10−3 M SrCl2 (containing 85Sr, reference activity 138 MBq mL−1) in 0.1 M HCl distributed by Polatom.

Methods

The organic phase was prepared as 1·10−3 M or 10−1 M solution of DCH18C6 in [Cnmim][NTf2], n = 2, 4, 6, 7, 10 or chloroform or benzene. In the study of the extracting agent concentration dependence, DCH18C6 solutions in concentrations ranging from 10−5 to 10−1 mol L−1 were used. Aqueous phase consisted of:

  • HNO3 or HCl or HClO4 in concentration range from 10−3 to 101 mol L−1 or

  • SrCl2 (2·10−5–6·10−1 mol L−1) solution in HNO3.

Aqueous phase was spiked with 85Sr tracer. Total aqueous concentration of carrying strontium before extraction was ~ 10−6 mol L−1, unless stated otherwise.

The extraction experiments were performed by contacting equal volumes of both phases (750 μl) and shaking them at vortex at 3000 rpm for 2 min (except for time of contact dependence experiments) at laboratory temperature (~ 25 °C). After that, the phases were separated by centrifugation (1 min at 1900 rcf). Aliquots of both phases (600 μl) were then taken for γ counting of 85Sr at well-type 2′′ NaI:Tl scintillation detector connected to TESLA 3102 single channel counter. After the extraction, pH value of the aqueous phase (abbreviated as pHeq) was measured by pH meter PHM240 (MeterLab) equipped with a combination Red-Rod glass electrode.

The extraction efficiency of strontium was evaluated by its distribution ratios calculated from measured net count rates of both phases using Eq. 5,

$${\text{D}} \, \text{[-] }\text{=} \, \frac{{c}_{\text{org}}}{{c}_{\text{aq}}}\approx \frac{{I}_{\text{org}}}{{I}_{\text{aq}}},$$
(5)

where corg and caq are concentrations of strontium in organic and aqueous phase, respectively and Iorg and Iaq are net count rates of 85Sr in the respective phases. Minimum (Dmin) and maximum (Dmax) obtainable values of distribution ratios were determined using detection limits [19]. The uncertainties of distribution ratios were quantified as a combined uncertainty of the statistics of the measurement and pipetting.

Results and discussions

To get a better view on the mechanism of strontium extraction using crown-ethers, considerable number of experiments were performed—the following dependencies were studied:

  • Type of organic solvent (IL, chloroform, benzene)

  • Nitric, hydrochloric or perchloric acid concentration

  • Chain length of imidazolium based ionic liquids

  • Time of contact

  • Strontium concentration

  • Crown-ether concentration

The strontium extraction using crown-ether is highly dependent on the acid concentration. Using traditional volatile organic solvents (VOCs)—chloroform and benzene—the distribution ratio is negligible at low acid concentrations and sharply increases with increasing acid concentration (Fig. 1). This trend is in agreement with supposed mechanism when in the case of these VOCs, the strontium extraction is supposed to carry out via uncharged ion associate consisting of strontium cation, crown-ether and anion compensating the charge. Compared chloroform and benzene as solvent, strontium extraction is more efficient into chloroform. This observation agrees with the polarity of each solvent that can be expressed by their dielectric constant (2.3 for benzene and 4.8 for chloroform [20, 21]).

Fig. 1
figure 1

The dependence of distribution ratio D of strontium extraction on the concentration of nitric (solid line), hydrochloric (dash-and-dot line) or perchloric (dashed line) acid into 10−1 M DCH18C6 in chloroform (filed circle) or benzene (filed diamond) (Vaq:Vorg = 1, c(Sr)aq,init ~ 10−6 mol L−1, T ~ 25 °C)

Full size image

The trends of dependence of strontium extraction on nitric acid concentration for chloroform and benzene are comparable to published results using similar system [10] and to system using 1-octanol [10, 22] as the organic phase. An unexpected change of the trend—a sudden drop in the distribution ratio—was observed in the system with chloroform at concentrations of nitric acid higher than 3 mol L−1.

The effect of acid anion character on strontium extraction is significant. Except for the case discussed above, the trends of distribution ratio dependence on acid concentration are similar for all tested acids. Nevertheless, more efficient extraction from nitric acid than from perchloric and hydrochloric acid indicates that nitrates form the ion associate with higher stability or solubility in chloroform and benzene in comparison to hydrochloric and perchloric anions. The strontium extraction from hydrochloric acid using benzene as organic solvent was observed to be inefficient.

When using ionic liquids as a solvent, significant differences in strontium extraction compared to VOCs were observed (Fig. 2). In agreement with the published results [8, 9, 13, 15, 17], the strontium extraction at low nitric acid concentrations is relatively high and significantly dependent on the chain length of IL cation. Dietz et al. [12] suggested the mechanism of ion exchange of strontium-crown-ether complex with ionic liquid cation. Regarding this mechanism, the extraction efficiency depends on the solubility of the IL cation in the aqueous phase (i.e. on its hydrophobicity), which in the case of imidazolium based cations is mainly determined by the length of the side chain. As the side chain length of this cation increases, its hydrophobicity increases, thus decreasing the importance of strontium extraction by ion exchange mechanism.

Fig. 2
figure 2

The dependence of distribution ratio D of strontium extraction on the concentration of nitric acid into 10−1 M DCH18C6 in various imidazolium based ionic liquids (Vaq:Vorg = 1, c(Sr)aq,init ~ 10−6 mol L−1, T ~ 25 °C)

Full size image

Experiments aimed at studying the effect of acid concentration in aqueous phase (Fig. 2) also showed that the strontium distribution ratio decreases with increasing nitric acid concentration up to ca 3 mol L−1 and then increases with increasing nitric acid concentration for ionic liquids with shorter chain ([Cnmim][NTf2], n = 2 or 4). The sharp decrease of strontium extraction corresponds to the explanation of Xu et al. [8] when competition between Sr2+ and H3O+ cations occurs.

The concentration of H3O+ is connected to the concentration of acid anions that can compensate the charge in forming ion associate. Thus, at sufficiently high concentration of nitric acid, strontium is extracted mainly in the form of the uncharged ion associate, which changes the trend of strontium extraction dependence on acid concentration.

For ionic liquids with longer chain (n > 6), the trend is similar at low and high nitric acid concentration, however, in the region of 0.05-5 M nitric acid, the trend of distribution ratio dependence on nitric acid concentration changes several times. This phenomenon can be explained by Dietz et al. [4] suggestion of the third two-steps mechanism when nitric acid is extracted with crown-ether and then strontium cation is exchanged with H3O+ in the complex with CE. In the case of this mechanism, the role of acid is crucial, therefore the effect of various mineral acids on strontium extraction was studied.

As shown in Fig. 3, strontium extraction from hydrochloric acid media into ionic liquid with shorter chain ([C4mim][NTf2]) is similar to the extraction from nitric acid. The extraction into more hydrophobic ionic liquid ([C10mim][NTf2]) seems to be more complicated. At low acid concentration, when ion exchange mechanism prevails, no difference between extraction from hydrochloric and nitric media was observed—ion exchange seems to be independent on type of acid anion. However, with increasing acid concentration, dependence for hydrochloric acid is completely different from nitric acid. With the presumption of ion-exchange between Sr2+-CE and H3O+-CE discussed for the case of nitric acid, the result at Fig. 3 indicates that hydrochloric acid does not allow the extraction via this third mechanism. At acid concentrations higher than 3 mol L−1, where the extraction via uncharged ion associate is supposed to prevail, strontium extraction from nitric acid into hydrophobic [C10mim][NTf2] is more efficient than from hydrochloric acid. The same result was observed for extraction into chloroform and benzene (Fig. 1). On the other hand, in the case of [C4mim][NTf2] as less hydrophobic ionic liquid than [C10mim][NTf2], the strontium extraction from 5 M or more concentrated hydrochloric acid seems to be more efficient than from nitric acid. Results from Figs. 1, 2 and 3 showed that the concentration of acid is one of the crucial parameters for strontium extraction. Therefore, further experiments studying the effects of time of contact, concentration of strontium and concentration of crown-ether were performed at different nitric acid concentrations.

Fig. 3
figure 3

The dependence of distribution ratio D of strontium extraction on the concentration of nitric (solid line), hydrochloric (dash-and-dot line) or perchloric (dash line) acid into 10−1 M DCH18C6 in [C4mim][NTf2] (filed triangle) or [C10mim][NTf2] (X symbol) (Vaq:Vorg = 1, c(Sr)aq,init ~ 10−6 mol L−1, T ~ 25 °C)

Full size image

To distinguish between processes directly related to strontium and those affecting its extraction indirectly (such as spontaneous IL cation transfer into aqueous phase), the dependencies of strontium extraction on the time of contact of phases (Fig. 4) were compared to the same experiments with pre-equilibrated system (both phases without strontium were contacted for 4 h). Except for a very short time of contact (less than 30 s), no significant difference between pre-equilibrated system and system without pre-equilibration was observed.

Fig. 4
figure 4

The dependence of distribution ratio D of strontium extraction on the time of contact from 1 mM (filed circle), 1 M (filed diamond) or 7 M (filed triangle) nitric acid: non-pre-equilibrated system (solid line), pre-equilibrated system (dashed line) (Vaq:Vorg = 1, c(Sr)aq,init ~ 10−6 mol L−1, T ~ 25 °C)

Full size image

The contact time is more influential at low concentration of nitric acid (Fig. 4). The distribution ratio was found to be maximal for time of contact in range 30 s—1 min and then slowly drops by about 15% in 30 min. With increasing concentration of nitric acid, the influence of the contact time becomes negligible.

The effect of strontium concentration on distribution ratio is shown at Fig. 5. When the concentration of strontium is much lower than concentrations of crown-ether or nitric acid, the distribution ratio is independent on strontium concentration. The presumed extraction mechanisms can be expressed by following equations:

Fig. 5
figure 5

The dependence of distribution ratio D of strontium extraction on strontium concentration from 1 mM (filed circle), 1 M (filed diamond) or 7 M (filed triangle) nitric acid into 10−1 M (solid or dashed line) or 1 mM (dash-and-dot line) DCH18C6 in [C4mim][NTf2] (Vaq:Vorg = 1, T ~ 25 °C)

Full size image
$$m {{\text{Sr}}}_{{\text{aq}}}^{2+}+n{\mathrm{ CE}}_{{\text{org}}}+2{{m ({\text{NO}}}_{3}^{-})}_{{\text{aq}}}\leftrightarrow {\left[{{\text{Sr}}}_{m}\cdot {\left({\text{CE}}\right)}_{n}\cdot {\left({{\text{NO}}}_{3}\right)}_{2m}\right]}_{{\text{org}}}$$
(6)
$$m{\mathrm{ Sr}}_{{\text{aq}}}^{2+}+{n\mathrm{ CE}}_{{\text{org}}}+2m {\left({{\text{C}}}_{{\text{n}}}{{\text{mim}}}^{+}\right)}_{{\text{org}}}\leftrightarrow {\left[ {{\text{Sr}}}_{m}\cdot {\left({\text{CE}}\right)}_{n}\right]}_{{\text{org}}}^{2{\text{m}}+}+2m {\left({{\text{C}}}_{{\text{n}}}{{\text{mim}}}^{+}\right)}_{{\text{aq}}}$$
(7)

The extraction of strontium can be in both cases mathematically expressed using their respective extraction constants (Eq. 8 for Eq. 6 and Eq. 9 for Eq. 7) defined by equilibrium concentrations in the respective phases as

$${K}_{ex}=\frac{c{(\left[{{\text{Sr}}}_{m}\cdot {\left({\text{CE}}\right)}_{n}\cdot {\left({{\text{NO}}}_{3}\right)}_{2m}\right])}_{{\text{org}}}}{c{\left({{\text{Sr}}}^{2+}\right)}_{{\text{aq}}}^{m} c{\left({\text{CE}}\right)}_{{\text{org}}}^{n} c{\left({{\text{NO}}}_{3}^{-}\right)}_{{\text{aq}}}^{2}}{K}_{\gamma }$$
(8)
$${K}_{ex}=\frac{c{\left({\left[{{\text{Sr}}}_{m}\cdot {{\text{CE}}}_{n}\right]}^{2{\text{m}}+}\right)}_{{\text{org}}} c{\left({{\text{C}}}_{{\text{n}}}{{\text{mim}}}^{+}\right)}_{{\text{aq}}}^{2m}}{c{\left({{\text{Sr}}}^{2+}\right)}_{{\text{aq}}}^{m} c{\left({\text{CE}}\right)}_{{\text{org}}}^{n} c{\left({{\text{C}}}_{{\text{n}}}{{\text{mim}}}^{+}\right)}_{{\text{org}}}^{2m}}{K}_{\gamma }$$
(9)

Presuming that strontium occurs in aqueous phase only as a solvated cation Sr2+ and in organic phase only in the complex with CE, the dependence of distribution ratio on strontium concentration can in both cases, be written as

$${\text{log}}\left(D\right)\approx {\text{log}}\left(\frac{c{\left({\text{Sr}}\right)}_{{\text{org}}}}{c({\text{S}}{{{\text{r}}}^{2+})}_{{\text{aq}}}}\right)=\left(m-1\right){\text{log}}\left(c{\left({{\text{Sr}}}^{2+}\right)}_{{\text{aq}}}\right)+C$$
(10)

From Eq. 10, the distribution ratio is independent on strontium concentration only when m = 1. Thus, it can be concluded that, when the concentration of strontium is much lower than concentrations of crown-ether or nitric acid, only one cation of strontium is extracted. under tested conditions.

Experiments aimed at the quantification of initial strontium concentration influence several sets of experiments were performed: extraction from 1 and 7M HNO3 with 0.1M CE in IL and extraction from 1mM HNO3 using 1mM CE in IL. With increasing strontium concentration, the distribution ratio starts to decrease monotonously when the concentration of strontium is comparable to crown-ether’s one due to lowering the amount of free crown-ether available for binding the metallic cation. This trend is noticeable for all tested nitric acid concentrations and no trend changes indicating dramatic shifts in extraction mechanism were observed.

Similarly, the effect of crown-ether concentration on the distribution ratio of strontium was studied (Fig. 6). The slope of the trendlines for all tested nitric acid concentrations was close to the value of 1 and no trend changes were observed. Summarizing the results presented in Fig. 5 and 6, the ratio of Sr2+:CE seems to be 1:1 under the conditions studied. This conclusion is in agreement with published results for strontium extraction using CE with different solvents [3, 11, 15].

Fig. 6
figure 6

The dependence of distribution ratio D of strontium extraction on DCH18C6 concentration from 1 mM (filed circle), 1 M (filed diamond) or 7 M (filed traingle) nitric acid into [C4mim][NTf2] (Vaq:Vorg = 1, c(Sr)aq,init ~ 10−6 mol L−1, T ~ 25 °C)

Full size image

Conclusions

This work focused on the extraction of strontium using dicyclohexano-18-crown-6 as the extracting agent and imidazolium-based ionic liquids, chloroform, or benzene as solvent. The effect of acid concentration, the type of mineral acid, time of contact and strontium or crown-ether concentration was studied, and results were discussed and compared with published data.

Strontium extraction into traditional volatile solvents using crown-ether is negligible at low acid concentration and increases with increasing concentration of mineral acid. This observation is in agreement with presumed uncharged ion associate extraction. However, extraction into chloroform was found to be more efficient than into benzene and it was found that the type of mineral acid plays key role. The distribution ratio of strontium extraction from nitric acid is much higher compared to that of hydrochloric acid.

When using ionic liquid as solvent, significant strontium extraction was observed even from low acid concentration, and the efficiency of the extraction decreases with hydrophobicity (the length of the side chain) of the ionic liquid. At low acid concentration, the extraction is presumed to carry out via ion exchange, independently on the type of mineral acid. With increasing acid concentration, the ion associate mechanism becomes more important as well as the type of mineral acid.

Regarding the results of extraction dependence on concentration of strontium or crown-ether, only one molecule of crown-ether is extracted with one strontium cation under all studied conditions. From the results obtained, it can be concluded that ionic liquids can be used as suitable solvent for efficient extraction of strontium using crown ether as the extractant. The quantitative extraction of strontium into ionic liquids even at low acid concentration is possible due to the direct involvement of the ionic liquids in the extraction mechanism. Their unique properties, especially low volatility and high thermal and radiation stability, offer a promising solution for the replacement of VOCs.