The effect of phase contact time on the extraction of iron(III) with D2EHPA in decane is shown in Tables 1 and 2. As seen, the extraction of iron from acidic solutions is slow. Thus, it takes at least 120 min to reach equilibrium at 25°С. As reported by Karpacheva and Ilozheva (1969), the slow extraction is due to the slow rate of formation in the aqueous phase of the iron(III) complex with the D2EHPA anion (FeR2+) caused by the low concentrations of the initial reagents (Fe3+ and R–) in the aqueous phase. However, a decrease in acidity results in an increase in the R– concentration in the aqueous phase and in an increase in the extraction rate. Thus, when extracting Fe3+ from slightly acidic sulfate solutions (pH = 1.7), the system reaches equilibrium after 3–6 min (Kozlov et al. 2008).
Table 1
Effect of phase contact time (τ) on the extraction of iron(III) with D2EHPA in decane. Aqueous phase: 0.93 g/L Fe(III) + 100 g/L H2SO4. Organic phase: 0.3 mol/L D2EHPA in decane. O:A = 1:1, Т = 25°С.
No. | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
τ, min | 2.5 | 5 | 10 | 15 | 30 | 60 | 120 | 180 | 240 |
Feaq, g/L | 0.616 | 0.42 | 0.21 | 0.115 | 0.042 | 0.036 | 0.029 | 0.029 | 0.029 |
Feorg, g/L | 0.314 | 0.51 | 0.72 | 0.815 | 0.888 | 0.894 | 0.901 | 0.901 | 0.901 |
DFe | 0.51 | 1.21 | 3.43 | 7.09 | 21.14 | 24.83 | 31.07 | 31.07 | 31.07 |
ε, % | 33.8 | 54.8 | 77.4 | 87.6 | 95.,5 | 96.12 | 96.9 | 96.9 | 96.9 |
Table 2
Effect of phase contact time (τ) on the extraction of iron(III) with D2EHPA in decane. Aqueous phase: 0.93 g/L Fe(III) + 100 g/L H2SO4. Organic phase: 0.3 mol/L D2EHPA in decane. O:A = 1:1, Т = 50°С.
No. | 1 | 2 | 3 | 4 | 5 | 6 |
τ, min | 2.5 | 5 | 10 | 15 | 30 | 60 |
Feaq, g/L | 0.012 | 0.010 | 0.009 | 0.009 | 0.007 | 0.007 |
Feorg, g/L | 0.948 | 0.950 | 0.951 | 0.951 | 0.953 | 0.953 |
DFe | 79.0 | 95.0 | 105.7 | 105.7 | 136.14 | 136.14 |
ε, % | 98.80 | 98.96 | 99.06 | 99.06 | 99.93 | 99.93 |
If necessary, the rate of iron extraction can be increased by increasing the temperature. As seen from Table 2, at a temperature of 50°C, almost complete extraction of iron can be achieved in this system after 5–10 minutes of phase contact. It should be noted that iron extraction can also be accelerated by the introduction of various catalysts. Thus, it has been reported Sinegribova and Muravyova 2000) that, when recovering iron(III) by D2EHPA in the presence of ethoxylated sodium lauryl sulfate, the system reached equilibrium after 1.5 min.
The effect of sulfuric acid concentration on the extraction of iron(III) with D2EHPA alone and in the presence of octyl alcohol was investigated and the results are shown in Fig. 1. As seen from the figure, the extraction efficiency of iron with the extractant (curve 1) is very high and the recovery of iron into the organic phase is close to 100% (97%) even at a sulfuric acid concentration of 1.0 mol/L. Therefore, stripping of iron from the loaded organic phase is difficult and requires a high concentration of H2SO4 (> 8.0 mol/L).
Figure 2 shows the effect of various proton-donor additives (HA) on iron extraction with D2EHPA. As seen, in the presence of all these additives, the iron extraction efficiency decreases which is due to the antagonistic effect caused by the interaction of D2EHPA and the additives resulting in the formation of stable associates. The formation of associates between HR and proton-donor additives (HA), such as АH···O = P(OR)2OH···OAH, with the coordination of HA both to the oxygen atoms of the POO group of HR and to the acidic proton of HR have been reported by Grigorieva et al. (2022). It is obvious that the strength and type of the interactions for different additives are different. Thus, according to 31P NMR data, in the case of octyl alcohol, a preferential interaction between the extractant and the alcohol through the oxygen of the octanol alcohol group and the hydroxy group of the phosphoric acid occurs, whereas, in the case of octanoic acid, its hydroxyl group preferentially interacts with the POO group of D2EHPA. In the D2EHPA and 4-tert-butylphenol mixtures, both types of interactions take place.
Figure 2 shows that, in the presence of octanoic acid (curve 1), the iron extraction efficiency decreases to a much lesser extent than in the presence of octyl alcohol and 4-tert-butylphenol (curves 2 and 3). This can be explained by the tendency of octanoic acid to self-associate and form dimers in the organic phase. It is the formation of stable dimers that makes these acids "inert" and to a significant extent prevents the formation of intermolecular associates. Octyl alcohol (curve 2) and 4-tert-butylphenol (curve 3), like octanoic acid, decreases iron extraction, however, unlike monocarboxylic acid, the effect of these additives on iron extraction is basically the same, despite the difference in their structure. The D2EHPA + octyl alcohol system is of great practical interest, therefore, this system has been studied in more detail.
In the presence of octyl alcohol, iron(III) extraction by the monomeric form of D2EHPA with a large excess of the extractant can be described by Eq. (1) (Stoyanov et al. 1984b):
Fe3+(aq) + (3 + s)HR(org) (FeR3·sHR)(org) + 3Н+(aq) (1)
with the extraction constant defined as follows:
K Fe−H = CFe(org)·C3H+(aq)·γe.c.(org)·γ3Н+(aq) /CFe(aq)·С3+sHR(org)·γ Fe(aq)·γ 3+sHR(org) (2)
where CFe(org), CFe(aq) and CH+(aq) are the analytical concentrations of iron in the organic and aqueous phases and hydrogen ions in the aqueous phase, CHR(org) is the concentration of D2EHPA monomers in the organic phase, s is the solvation number, i.e. the number of extractant molecules in the extracted iron complex, γe.c.(org) is the activity coefficient of the extracted compound, γHR(org) is the activity coefficient of monomeric D2EHPA, γH+(aq) and γFe(aq) are the activity coefficients of hydrogen and iron(III) ions in the aqueous phase respectively. At constant ionic strength, γFe(aq) and γН+(aq) are constant. Since the concentration of iron in the initial aqueous solution is low (5.0·10− 3 mol/L), the equilibrium concentrations of CHR(org) and CH+(aq) can be equated to their initial concentrations. In this case, the distribution ratio of iron (DFe) can be written as follows (Eq. 3):
D Fe = KIFe−H·С3+sHR(org)·γ3+sHR(org) /C3H+(aq)·γe,c,(org) (3)
where KIFe−H = KFe−H·γFe(aq)·γ3Н+(aq)
When octyl alcohol is the diluent, in which D2EHPA exists in the monomeric form, γHR(org) and γe.c.(org) are constant. Therefore, assuming that the activity of octyl alcohol is constant, we can write Eq. (3) as follows (Eq. 4):
logD Fe = logKIIFe−H + (3 + s)logСHR(org)) – 3logCH+(aq) (4)
According to Eq. (4), the values of the slope of the logDFe = f(logСHR(org)) and logDFe = f(logCH+(aq)) dependencies should be equal to 3 + s and − 3 respectively. As shown in Figs. 3 and 4, the slope of the plot of logDFe against pH (Fig. 3) is close to − 3 (–2.7 ± 0.2), while the slope of the plot of logDFe against logСHR(org) (Fig. 4) is equal to 4 (3.9 ± 0.2). The latter suggests that the solvation number (s) is equal to one and therefore, the extracted iron species in the D2EHPA and octyl alcohol mixture in excess extractant is [FeR3·HR] or [FeR2·HR2].
It should be added that, in the extraction systems with D2EHPA and three-charged elements, including iron (Baes and Bacer 1960), the solvation number is usually equal to 3 (Baes and Bacer 1960; Levin et al. 1972). Nevertheless, lower solvation numbers, 2, 1, and 0, can also be achieved in these systems for bismuth, thallium, and gallium respectively (Levin et al. 1972).
The electronic spectra of the extracted species of iron with D2EHPA in decane and octyl alcohol are given in Fig. 5. As seen, in decane, the position of the maximum is at 281 nm (curve 1) and this peak arises due to the electronic d–d transitions of Fe3+ in the octahedral environment (Cotton and Wilkinson 1969). The spectrum of the extracted species in octanol shows a maximum shifted to the long wavelength region (~ 300 nm), which indicates different structures of the complexes in decane and octanol.
The structures of the extracted iron species with an excess of the extractant in decane and in octyl alcohol are shown in Fig. 6. Obviously, the compositions of the extracted iron species in decane and octyl alcohol are different. Thus, the major extracted species are Fe(HR2)3 (Baes and Bacer 1960) and [FeR2·HR2] in decane and octyl alcohol respectively.
The antagonistic effect of alcohol on iron extraction is clearly shown when comparing the iron extraction with D2EHPA alone (Fig. 1, curve 1) and its mixture with octanol (Fig. 1, curves 2 and 3). As seen, in the presence of alcohol, the extraction efficiency of iron significantly decreases compared with using D2EHPA alone, which allows efficient stripping of iron from the loaded organic phase using 3.5–4.0 mol/L H2SO4. Figure 7 shows iron stripping isotherms for two stripping agents. It can be seen that, in both cases, quite efficient iron stripping is achieved. Also, stripping of the loaded Fe(III) with H2SO4 in the presence of HCl proceeds much more efficiently (curve 2). Thus, for the loaded organic phase containing 2.0 g/L Fe(III), almost complete stripping of iron can be achieved in 5–6 steps at an О:A ratio of 6:1, i.e. when concentrating the iron six times.
Iron stripping, like iron extraction, is slow, however an increase in temperature significantly accelerates the stripping process (Table 3). It should be noted that the temperature affects the extraction efficiency to a much greater extent than the stripping efficiency.
Table 3
Effect of phase contact time (τ) on the stripping of iron(III) from D2EHPA and octyl alcohol in decane at different temperatures. Organic phase: 0.3 mol/L D2EHPA + 1.2 mol/L octyl alcohol in decane, CFe(initial) = 2.80 g/L. Aqueous phase: 350 g/L H2SO4 + 36.5 g/L (1.0 mol/L) HCl. τ is variable, O:A = 5:1.
No. | 1 | 2 | 3 | 4 | 5 | 6 |
τ, min | 2,5 | 5 | 10 | 15 | 30 | 60 |
25°С |
Feaq, g/L | 8.61 | 8.87 | 9.12 | 9.37 | 9.49 | 9.58 |
Feorg, g/L | 1.06 | 1.02 | 0.97 | 0.93 | 0.89 | 0.87 |
DFe | 0.123 | 0.115 | 0.106 | 0.099 | 0.094 | 0.091 |
50°С |
Feaq, g/L | 10.60 | 10.71 | 11.02 | 11.24 | 11.20 | 11.41 |
Feorg, g/L | 0.66 | 0.63 | 0.61 | 0.56 | 0.56 | 0.53 |
DFe | 0.062 | 0.059 | 0.055 | 0.050 | 0.050 | 0.046 |
Figure 8 shows an iron extraction isotherm with a mixture of D2EHPA and octyl alcohol in decane together with an operating line (McCabe-Thiele diagram). It is seen that, in a five-stage extraction system, at an aqueous to organic phase ratio of 6:1, iron is recovered from the aqueous phase almost completely. In this case, iron is significantly concentrated in the organic phase and its further concentrating can be achieved at the stripping stage. Non-ferrous metals (in particular, Zn, Co, Ni) are almost not extracted under these conditions. Thus, during the extraction of zinc from sulfate solution containing 80 g/L Zn, the concentration of loaded Zn was 70–80 mg/L, i.e. ~0.1% of Zn is recovered with the mixture of D2EHPA and octyl alcohol. It can be added that, when producing sufficiently pure iron, it can be considered as a target component to be used for the production of ferrites, which are widely used as magnetic materials in radio engineering, electronics, etc.
It should also be added that, when the systems with D2EHPA are employed as extractants, the use of a rather expensive alkaline reagent (NaOH) is not required to neutralize the aqueous solution, which is necessary when iron is extracted with monocarboxylic acids (Van der Zeeuw 1976).