Introduction

8-Hydroxyquinoline and its substituted derivatives have attracted considerable interest [1,2,3,4,5,6] due to their remarkable coordinating abilities and have found their way into various applications. 8-Hydroxyquinolines have been, for example, applied for analytical [7] and separation purposes as well as for the formation of luminescent coordination compounds [8,9,10,11]. The 8-hydroxyquinoline unit has also proved to be a valuable building block for the construction of different artificial receptors (hosts) [12, 13], self-assembled systems and other supramolecular architectures [14]. In addition, representatives of this class of compounds have been recognized as substances with interesting biological activities, including antibacterial, antiviral, anticancer, anti-inflammatory, antidiabetic and other activities [15,16,17,18,19]. The syntheses of a wide range of 8-hydroxyquinoline derivatives and the examinations of their chemical and biological properties have been the subject of intensive research over a long period of time and the achieved results are the focus of review articles and books.

Due to the above-mentioned remarkable complexation properties of 8-hydroxyquinolines, their application as extractants for various metal ions in solvent extraction processes has attracted great attention (for examples, see [20,21,22,23,24,25,26,27,28]). Solvent extraction is a well-established method for concentrating and purifying metal ions from dilute or multi-component aqueous media and represents, among other things, a crucial step of hydrometallurgical processes [29,30,31,32,33,34]. In the current context of the need to find new sources for elements of strategic importance, such as indium [35,36,37,38], the recovery of such valuable metals from low-concentrated resources becomes more and more important (for a review on recovery of indium from aqueous solutions by solvent extraction, see [39]).

The aim of the present study was the evaluation of the potential of selected 8-hydroxyquinoline derivatives to extract indium ions in the presence of other metal ions, such as Fe(III) and Zn(II), from aqueous sulfuric acid media with high selectivity. Such media are provided, for instance, by bioleaching processes of sulfide minerals, such as sphalerite, and contain besides the major elements, such as iron and zinc ions, also several minor constituents, including also the strategic element indium. It should be stressed at this point that the selective extraction of indium from solutions containing only low concentration of this element still represents a particular challenge. Recent advances in indium (bio)hydrometallurgy, including among other things a discussion on leaching with sulfuric acid, bioleaching and solvent extraction with different extractants, are summarized in [40].

Results and discussion

Structures of the selected compounds (selection criteria)

As candidates for our study, we have selected the compounds 1–11 (Fig. 1) bearing alkanoyl or alkyl groups of different chain lengths. Depending on the nature of the substituents and the position of their connection with the aromatic ring, chelators with different complexation properties were expected.

Fig. 1
figure 1

Structures of the 8-hydroxyquinoline derivatives 1–11 considered in this study

Full size image

Various factors influence the ability of organic compounds to act as extracting agents and such factors have been intensively discussed in the literature (for an example, see [41]). The knowledge of the acid dissociation constants is, among other things, of high importance for the prediction of the metal chelating properties and for the selection of suitable candidates. In the case of 8-hydroxyquinolines, alkyl substitution in the 2-, 5-, or 7-position increases the pKa1 (pKNH, see Table 1) and pKa2 (pKOH) values, as expected from the substituent effects. In contrast, the presence of an alkanoyl group in the 5-position decreases the basicity of the nitrogen atom and increases the acidity of the OH group (decrease in the pKa1 and pKa2 values). Table 1 includes some experimental pKa values from different sources [25, 41,42,43,44]. Their comparison with the predicted/calculated pKa values [45], which are summarized in Table S1, shows a good agreement between the experimental and predicted data. Therefore, in the absence of experimental data, the consideration of the calculated pKa values (see Table S1) represents a good alternative and provides valuable information for the design of new chelators.

Table 1 Experimental pKa values [25, 41,42,43,44] and logKd [25, 41] of selected 8-hydroxyquinolines
Full size table

When looking at the logKD data given in Table 1, it can be seen that methyl substitution in the 2- and 5-position causes about four- to fivefold increase in the distribution constant KD (distribution between the organic phase and the aqueous phase), whereas for the substitution in 7-position even an eightfold increase in KD in comparison to the unsubstituted 8-hydroxyquinoline is observed [25]. An extension of the substituent in the 7-position by only one CH2-unit (from methyl to ethyl group) leads to a further marked increase in KD (almost eightfold increase in KD for 7-ethyl-8-hydroxyquinoline in comparison to the 7-methyl-substituted analog) [25]. The incorporation of a longer alkyl chain into the quinoline ring, resulting in increased organic phase solubility, represents a valuable strategy for developing chelators with improved extraction properties [24, 27]. Steric hindrance (and other factors), however, may be responsible for unfavorable effects, as shown for some 2- and 7-substituted 8-hydroxyquinolines (even for methyl-substituted derivatives) in studies on extraction of aluminum and other metal ions [41, 46].

Among the 8-hydroxyquinoline derivatives 1–11, compounds 1–3, containing an alkanoyl group in the 5-position, were expected to be the best extractors for indium ions under the chosen extraction conditions as a result of the favorable influence of the electron-withdrawing group on the electronic situation in the heterocyclic ring. In contrast, the formation of an intramolecular C=O···HO hydrogen bond between the alkanoyl substituent and the OH group (in a hydrogen bonded six-membered ring), as in the case of the 7-substituted derivatives 5–7, was supposed to have an unfavorable influence on the extraction behavior of the 8-hydroxyquinolines. It should be noted at this stage that the existence of an intramolecular hydrogen bond between the OH group and the quinoline nitrogen atom has also been discussed in the literature for 8-hydroxyquinolines. For example, the comparison of the properties of 8-hydroxyquinoline with those of 5-, 6-, and 7-hydroxyquinolines revealed that the properties are fully consistent with the existence of a weak intramolecular hydrogen bond in a five-membered ring (for a discussion, see [47]).

In comparison to the 5-alkanoyl derivatives, lower extractability towards indium ions had to be assumed for compounds 8–11, bearing an alkyl group. Furthermore, less favorable extractability of indium by compounds 4 and 10 (compared to 1–3 and 8/9, respectively) was expected as a result of the steric hindrance of the methyl group, as observed previously for the formation of aluminum(III) complexes [46]. We were further interested to see how the different chain lengths affect the extractability of the 8-hydroxyquinolines towards indium ions and the kinetics of the extraction process.

It should be pointed out here that the extraction studies have confirmed our expectations and shown the favorable complexation properties of monosubstituted 8-hydroxyquinolines bearing an alkanoyl group in the 5-position.

Synthesis of compounds 1–11

The syntheses of the target compounds 1–11 started from 8-hydroxyquinoline (in the case of 1–3, 5, 6, 8, 9, and 11) or 2-methyl-8-quinolinol (in the case of 4, 7, and 10). The 5- and 7-substituted alkanoyl derivatives 1–7 were prepared by the Friedel–Crafts acylation of the above-mentioned heterocyclic educts with hexanoyl chloride, nonanoyl chloride, or dodecanoyl chloride according to the procedure reported in [21, 48, 49]. The alkanoyl derivatives 2, 3, and 5 were further converted into the alkyl derivatives 8, 9, and 11 by a Huang-Minlon reduction, as described in [27, 28, 48, 49]. The basis for the synthesis of compound 10 was 5-dodecanoyl-2-methyl-8-quinolinol, which was prepared by a treatment of 2-methyl-8-quinolinol with dodecanoyl chloride [27, 28, 50].

Extraction behavior of compounds 1–11

The extraction of an aqueous solution containing the metal ions (indium, zink, and iron ions) with the chloroform solution of the corresponding 8-hydroxyquinoline derivative was performed at 23 °C by means of mechanical shaking (see “Experimental”). The phase ratio V(org):V(w) was 1:1 (15 cm3 each) and the pH of the aqueous solution was adjusted to 2. The extractability of indium ions by the tested compounds was studied as a function of the metal concentration (2.61 × 10−5 mol/dm3 [3 mg/dm3], 6.10 × 10−5 mol/dm3 [7 mg/dm3] or 7.84 × 10−5 mol/dm3 [9 mg/dm3] indium), concentration of the chelator (c = 2.8, 5, or 10 mM solutions) and as a function of time. The molar ratio of In(III):Zn(II):Fe(III) in the aqueous solutions was 1:1:0, 1:1:1, 1:10:10, and 1:100:100. The extractability of the tested compounds was analyzed after the shaking time of 5, 10, 15, 30, 60, and 120 min. Re-extraction was performed with H2SO4 (c = 1.5 M). The determination of the metal concentration in the aqueous phase was carried out on the basis of ICP-OES measurements.

As expected, among the tested 8-hydroxyquinolines 1–11, compounds 1–3, bearing an alkanoyl group in the 5-position of the quinoline ring, were shown to be the most effective extractors for indium(III) under the chosen extraction conditions. Compound 2, bearing n-nonanoyl group, was proven to be a particularly powerful extractor for indium ions, showing not only the best selectivity for indium in the presence of iron and zinc ions, but also the most favorable extraction kinetics. The methyl group in the 2-position, as in 4, causes some steric hindrance, which prevents the effective complexation of indium(III) from the acidic aqueous medium. The extractability of indium ions by the 5-alkanoyl derivatives 1–4 decreases in the sequence: 2 > 3 ≥ 1 ≫ 4.

In the presence of zinc ions, the 8-hydroxyquinolines 2 and 3 (chloroform, c = 10 mM) are capable of extracting indium(III) into the chloroform phase with 99.9 and 97.6% yield, respectively (see Fig. 2 and Table S2). With more dilute chelator solutions (c = 2.8 mM) the extraction yields amount to 97.8 and 76.7% for 2 and 3, respectively (see Fig. S1 and Table S2). Thus, it appears that under the chosen extraction conditions the extractability of indium by compound 2 is only slightly dependent on the concentration of the chelator (in contrast to compound 3). Experiments with aqueous solutions containing equimolar amounts of indium, zinc and iron ions (In(III):Zn(II):Fe(III) = 1:1:1) revealed that indium(III) can be extracted with 77.5, 98.6, and 93% by 1, 2 and 3, respectively (10 mM chloroform solutions of 1–3; for details, see Fig. 3 and Tables S3 and S4). Furthermore, the experiments showed that the extraction efficiency of 1–3 for indium(III) decreases as the content of iron and zinc ions in the mixture increases (see Table S5a and S5b); however, the observed extraction yields are still significant and provide further indication of the favorable extraction properties of these compounds, in particular of the derivative 2 (98.6, 72.2, and 33.3% extraction yield of indium for the molar ratio of In(III):Zn(II):Fe(III) amounted to 1:1:1, 1:10:10, and 1:100:100, respectively). It should be noted that the last-mentioned compound was found to display the best selectivity (Fig. 4) for indium over zinc and iron ions as well as the best extraction kinetics (see Tables S5 and S6), indium ions can be extracted by compound 2 with about 43, 56, and 88% after 5, 15, and 30 min shaking time, respectively (the other tested compounds require longer time, see Table S6). The extracted indium can be stripped into aqueous phase (H2SO4, c = 1.5 M) with good yields (see Tables S2–S4 and S6 given in the Supporting Information) and without significant loss of the chelator.

Fig. 2
figure 2

Extractability of indium ions by 8-hydroxyquinolines 2–10 from acidic aqueous media containing equimolar amounts of zinc ions (pH 2; In(III):Zn(II) = 1:1; indium, c = 7.84 × 10−5 mol/dm3 [9 mg/dm3]; chelator, c = 10 mM, for further details, see Table S2)

Full size image
Fig. 3
figure 3

Extractability of indium ions by selected 8-hydroxyquinolines (compounds 1–3, 5, and 8) from acidic aqueous media containing equimolar amounts of zinc and iron ions (pH 2; molar ratio of In(III):Zn(II):Fe(III), 1:1:1; indium, c = 6.10 × 10−5 mol/dm3 [7 mg/dm3]; chelator, c = 10 mM, for further details, see Table S3)

Full size image
Fig. 4
figure 4

Extractability of indium, zinc, and iron ions by 8-hydroxyquinolines 1–10 from acidic aqueous media containing equimolar amounts of indium, zinc, and iron ions (pH 2; molar ratio of In(III):Zn(II):Fe(III), 1:1:1; indium, c = 6.10 × 10−5 mol/dm3 [7 mg/dm3]; chelator, c = 10 mM for further details, see Tables S3 and S4)

Full size image

The ability to extract indium ions by 8-hydroxyquinolines 8 and 9, bearing a long-chain alkyl group in 5-position, was shown to be lower than those of the 5-alkanoyl analogs (see Fig. 4 and Tables S2–S6). As observed in the case of the 5-alkanoyl derivatives, the introduction of the methyl group at the 2-position decreases the ability of the 5-alkyl derivatives to extract indium ions. This tendency is clearly visible by the comparison of the extraction behavior of 2-methyl-5-dodecyl-8-hydroxyquinoline (10) with that of 5-dodecyl-8-hydroxyquinoline (9). Furthermore, the results given in Table S4 show a clear evidence for a stronger complexation of indium ions by 8 and 9 in comparison to 11, consisting of n-nonyl group at 7-position. The less favorable extractability of indium(III) by 11 seems to be the result of the steric hindrance by the nonyl group, as observed in the case of the 2-methyl derivatives.

In comparison to compounds 1–3, a strong reduction of the extraction efficiency toward indium(III) has also been observed for 8-hydroxyquinolines bearing an 7-alkanoyl group (compounds 5 and 6). As already mentioned above, the formation of intramolecular OH···O=C hydrogen bonds (Fig. 5), seems to be responsible for the relative weak extractability of indium ions by 5 and 6. The presence of the methyl group at 2-position causes further weakening of the extractability, as is clearly visible by the extraction behavior of compound 7 (see Fig. 4, and Tables S2 and S4).

Fig. 5
figure 5

Partial 1H NMR spectra of 2 and 5 (500 MHz, CDCl3) indicating the formation of an intramolecular OH∙∙∙O hydrogen bond in the case of 5

Full size image

For all tested 8-hydroxyquinolines, the ability of extracting indium ions from acidic aqueous phase into chloroform phase decreases in the sequence: 2 > 3 > 1 > 9 > 6, 5 > 8 > 10, 7, 4 (see Tables in Supporting Information). Thus, the performed experiments revealed that the presence of an alkanoyl group in the 5-position of the quinoline ring (especially nonanoyl group in the case of the tested compounds) seems to be a particularly suitable architecture for the effective extraction of indium ions from aqueous media at pH value of 2. The strong influence of the 5-alkanoyl substituent on the ability of extracting indium in the presence of zinc and iron ions is clearly visible in the direct comparison of the properties of compounds 2 and 3 with those of the unsubstituted 8-hydroxyquinoline (see Fig. 6, and Figs. S3 and S4 in Supporting Information). In contrast to the effective chelators 2 and 3, the unsubstituted 8-hydroxyquinoline displays no selectivity for indium in the presence of iron and zinc ions under the chosen extraction conditions.

Fig. 6
figure 6

Extractability of indium, zinc and iron ions by 5-nonanoyl-8-hydroxyquinoline (2), 5-dodecanoyl-8-hydroxyquinoline (3), and 8-hydroxyquinoline (8-OHQ) from acidic aqueous media (pH 2) containing a equimolar amounts of indium, zinc and iron ions [molar ratio of In(III):Zn(II):Fe(III), 1:1:1] and b tenfold excess of zinc and iron ions [molar ratio of In(III):Zn(II):Fe(III), 1:10:10]; indium, c = 6.10 × 10−5 mol/dm3 [7 mg/dm3]; chelator, c = 10 mM

Full size image

Conclusion

The aim of this work was to study the extraction efficiency of selected 8-hydroxyquinolines (compounds 1–11), bearing alkanoyl or alkyl groups of different chain lengths, toward indium ions from acidic aqueous media into chloroform. Studies on extraction of indium ions by the tested compounds have confirmed our expectations and allowed the identification of interesting structure–activity relationships. The extraction properties of the 8-hydroxyquinolines differ in a characteristic manner on their structure and the observed tendencies can be summarized as follows: (a) compounds bearing an 5-alkanoyl group are more effective indium extractors than the 5-alkyl-, 7-alkanoyl-, and 7-alkyl-bearing analogs under the chosen extraction conditions, (b) among the tested 5-alkanoyl derivatives, compound 2, consisting of nonanoyl group, was shown to have the best extraction properties, (c) the introduction of the methyl group at the 2-position of the quinoline ring (as in 4, 7, and 10) leads to steric hindrance, which prevents the effective complexation of indium(III), (d) the formation of an intramolecular C=O···HO hydrogen bond between the alkanoyl substituent and the quinoline OH group, as in the case of the 7-substituted derivatives 5–7, seems to be responsible for the relatively low extraction efficiency toward the indium ions, (e) the presence of a long-chain alkyl group in 7-position, as in the case of compound 11, leads to some steric bulk, which unfavorably influences the extractability of indium.

In summary, the 5-alkanoyl derivatives 1–3 are capable of extracting indium ions with good yields under the chosen extraction conditions. The selectivity of 1–3 for indium over zinc ions was shown to be high, whereas the selectivity for indium over iron ions was particularly favorable in the case of the compound 2, bearing n-nonanoyl group. Compound 2 was proven to be a particularly powerful extractor for indium ions, showing not only the best selectivity for indium in the presence of iron and zinc ions, but also the most favorable extraction kinetics. Furthermore, the extracted indium ions can be stripped into aqueous phase (1.5 M H2SO4) with relatively good yields.

Such chelators are valuable candidates for the extraction of indium ions from diluted solutions by a multistage extraction process, in which the initially very low concentration of indium can be significantly increased by several successive extraction and re-extraction steps (for an example of a multistage extraction process using di-(2-ethylhexyl)phosphoric acid (DEHPA), cyanex 272, and cyanex 923 as chelators, see [53]).

Due to the problems with the exact prediction of the complexation and extraction properties of a chelator, the identification of characteristic structural features, having the ability to influence the properties in a predictable way, is of high importance. The results obtained with the selected 8-hydroxyquinolines represent a valuable base for further developments in this field.

Experimental

All reagents were purified and dried according to general procedures. Indium(III) and zinc(II) sulfate were purchased from Alfa Aeser, whereas iron(III) sulfate from MP Biomedicals, LLC. The extraction procedure was performed with the help of an IKA HS 250 shaker. The pH value of the aqueous solution was measured with a measuring device OPR 70 from VWR. The ICP-OES measurements were performed on a Thermo Scientific iCAP 6500 Series ICP Spectrometer. 1H NMR and 13C NMR spectra were recorded using a Bruker AVANCE 500 spectrometer at 293 K (1H NMR: 500 MHz, 13C NMR: 125 MHz). The chemical shifts are reported relative to tetramethylsilane.

Extraction procedure

An aliquot of the organic phase (15 cm3) containing the corresponding 8-hydroxyquinoline derivative (chloroform solution, c = 2.8, 5, or 10 mM) and an aliquot of the aqueous solution (15 cm3) containing indium(III) sulfate (c = 2.61 × 10−5 mol/dm3 [3 mg/dm3], 6.10 × 10−5 mol/dm3 [7 mg/dm3], or 7.84 × 10−5 mol/dm3 [9 mg/dm3] indium), zinc(II) sulfate and iron(III) sulfate [In(III):Zn(II):Fe(III): 1:1:0, 1:1:1, 1:10:10, and 1:100:100, mole ratio] at pH value 2 was shaken for 5, 10, 15, 30, 60, or 120 min at 23 °C in Erlenmeyer flasks on an IKA shaker with 300 rds min−1. After phase separation, the organic phase was stripped by re-extraction with an equivalent volume of 1.5 M sulfuric acid. The concentration of the metal ions in the aqueous phase of each extractions step was determined by ICP-OES measurements.

Preparation of compounds 1–11

Compounds 1–11 were prepared according to the known procedures, as described above (see “Synthesis of compounds 1–11”). In the case of compounds 1, 3, 6, 10, and 11, the 13C NMR data have not previously been reported (see [28, 51, 52]) and are given below.

5-Hexanoyl-8-hydroxyquinoline (1)

1H NMR (500 MHz, CDCl3): δ = 9.46 (dd, J = 8.8, 1.6, 1H), 8.79 (dd, J = 4.2, 1.6 Hz, 1H), 8.12 (d, J = 8.1 Hz, 1H), 7.57 (dd, J = 8.8, 4.2 Hz, 1H), 7.14 (d, J = 8.1 Hz, 1H), 3.03 (t, J = 7.5 Hz, 2H), 1.77 (p, J = 7.4 Hz, 2H), 1.48-1.27 (m, 4H), 0.91 (t, J = 7.1 Hz, 3H) ppm; 13C NMR (125 MHz, CDCl3): δ = 201.92, 156.33, 148.15, 138.19, 136.16, 132.66, 127.24, 125.15, 124.00, 108.14, 40.47, 31.70, 24.89, 22.66, 14.08 ppm.

5-Dodecanoyl-8-hydroxyquinoline (3)

1H NMR (500 MHz, CDCl3) δ = 9.47 (dd, J = 8.7, 1.6 Hz, 1H), 8.80 (dd, J = 4.2, 1.6 Hz, 1H), 8.13 (d, J = 8.2 Hz, 1H), 7.58 (dd, J = 8.7, 4.2 Hz, 1H), 7.15 (d, J = 8.2 Hz, 1H), 3.04 (t, J = 7.5 Hz, 2H), 1.77 (p, J = 7.5 Hz, 2H), 1.48–1.19 (m, 16H), 0.87 (t, J = 7.0 Hz, 3H) ppm; 13C NMR (125 MHz, CDCl3) δ = 201.88, 156.24, 148.07, 138.13, 136.08, 132.56, 127.16, 125.11, 123.91, 108.05, 40.45, 31.92, 29.63, 29.52, 29.45, 29.35, 25.15, 22.70, 14.13 ppm.

7-Dodecanoyl-8-hydroxyquinoline (6)

1H NMR (500 MHz, CDCl3) δ = 8.99 (dd, J = 4.2, 1.7 Hz, 1H), 8.11 (dd, J = 8.3, 1.7 Hz, 1H), 7.82 (d, J = 8.8 Hz, 1H), 7.54 (dd, J = 8.3, 4.2 Hz, 1H), 7.27 (d, J = 8.8 Hz, 1H), 3.10 (t, J = 7.4 Hz, 2H), 1.81 (p, J = 7.4 Hz, 2H), 1.47–1.21 (m, 19H), 0.88 (t, J = 6.8 Hz, 3H) ppm; 13C NMR (125 MHz, CDCl3) δ = 206.48, 160.99, 149.84, 140.34, 135.63, 132.34, 125.50, 124.20, 117.29, 116.14, 39.36, 31.91, 29.62, 29.50, 29.47, 29.34, 24.46, 22.69, 14.13 ppm.

5-Dodecyl-2-methyl-8-hydroxyquinoline (10)

1H NMR (500 MHz, CDCl3): δ = 8.20 (d, J = 8.6 Hz, 1H), 7.31 (d, J = 8.6 Hz, 1H), 7.18 (d, J = 7.7 Hz, 1H), 7.05 (d, J = 7.7 Hz, 1H), 2.91 (t, J = 7.5 Hz, 2H), 2.72 (s, 3H), 1.65 (p, J = 7.7 Hz, 2H), 1.43-1.15 (m, 18H), 0.88 (t, J = 6.8 Hz, 3H) ppm; 13C NMR (125 MHz, CDCl3): δ = 156.13, 149.89, 138.02, 132.92, 129.17, 125.91, 125.06, 122.12, 109.13, 31.93, 31.83, 31.33, 29.68, 29.67, 29.64, 29.63, 29.55, 29.36, 24.78, 22.70, 14.13 ppm.

7-Nonyl-8-hydroxyquinoline (11)

1H NMR (500 MHz, CDCl3): δ = 8.90 (dd, J = 5.3, 1.5 Hz, 1H), 8.79 (dd, J = 8.4, 1.5 Hz, 1H), 7.79 (m, 1H), 7.70 (d, J = 8.3 Hz, 1H), 7.57 (d, J = 8.3 Hz, 1H), 2.95 (t, J = 7.5 Hz, 2H), 1.69 (p, J = 7.6 Hz, 2H), 1.40-1.19 (m, 12H), 0.86 (t, J = 7.0 Hz, 3H) ppm; 13C NMR (125 MHz, CDCl3): δ = 146.23, 146.01, 140.94, 136.70, 133.70, 129.20, 128.96, 119.26, 118.71, 31.87, 31.07, 29.61, 29.50, 29.31, 22.67, 14.11 ppm.