Amine extraction of chromium(VI) from aqueous acidic solutions

Amine extraction of chromium(VI) from aqueous acidic solutions

Separation and Purification Technology 36 (2004) 63–75 Amine extraction of chromium(VI) from aqueous acidic solutions Aynur Senol∗ Department of Chem...

174KB Sizes 2 Downloads 20 Views

Separation and Purification Technology 36 (2004) 63–75

Amine extraction of chromium(VI) from aqueous acidic solutions Aynur Senol∗ Department of Chemical Engineering, Faculty of Engineering, Istanbul University, 34850 Avcilar, Istanbul, Turkey Received 8 November 2002; received in revised form 25 April 2003; accepted 25 April 2003

Abstract Extraction of chromium (Cr(VI)) from the aqueous acidic solution into a co-existing organic phase containing Alamine 336 (C8 –C10 tertiary amine mixture) and diluent (xylene) was studied at isothermal condition (298.2 K). The distribution ratio was measured for two initial aqueous concentrations of Cr(VI), i.e. 0.019 mol/dm3 (1 mg/ml) and 0.192 mol/dm3 (10 mg/ml) against various molarities of the aqueous phase sulphuric acid in the range from 0.05 to 4 mol/dm3 . The overall loading factor of amine (Zt ) exhibits a maximum restricted between the aqueous phase acidity of 0.75–1 mol/dm3 H2 SO4 for a lower amine concentration ◦ related to the ratio of initial concentrations of amine to Cr(VI), C◦NH3 /C Cr(VI) ≈ 0.75 or less than this limit. In the range above this value a continual decreasing of Zt has been observed. A comparative study of the extraction degrees of Cr(VI) by Alamine 336 and Aliquat 336 extractants for the pH range of 1–7 has been evaluated. The results were correlated using various versions of the mass action law, i.e. a chemodel approach and a modified version of the Langmuir equilibrium model comprising the formation of only one type Cr(VI)–amine aggregated structure. © 2003 Elsevier B.V. All rights reserved. Keywords: Liquid–liquid extraction; Chromium(VI); Alamine 336; Modeling

1. Introduction Chromium (CrVI) enters the environment through its use as a corrosion inhibitor, especially the chemical passivating component of stainless steels. It also emerges from various plant operations such as the chromic acid anhydride production, the chromium electroplating, cooling towers, and leather tanning wastes. These effluents can contain significant quantities of Cr(VI) endangering the quality of surface water. The presence of Cr(VI) even in trace quantities, known to be a strong oxidant, a mutagen and a ∗

Tel./fax: +90-212-591-1997. E-mail address: [email protected] (A. Senol).

carcinogen on human health, produces a toxic influence on aquatic life. Thus the removing of Cr(VI) from industrial wastewater to its permissible limit before the latter has been drained into surface waters is a deep concern, as well as a challenging problem in the industry. The total chromium maximum contaminant level (MCL) and the World Health Organization (WHO) level represent a guideline value that can be found in “Guidelines for Drinking-Water Quality, Vol. 1, pp. 80–85, WHO, Geneva, 1984” yielding a limit of 10−6 mol/dm3 Cr(VI) in drinking water [1]. However, according to the US EPA standards Cr(VI) contents in fresh water should not exceed 4×10−4 mol/dm3 required for the protection of aquatic life [1]. Actually, Sadaoui et al. [2] define the waste-

1383-5866/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/S1383-5866(03)00153-9

64

A. Senol / Separation and Purification Technology 36 (2004) 63–75

Nomenclature A a, b CA Cacid CCr(VI) CCr(VI) ◦ C Cr(VI) CCr2 O7 C◦NH3 C(NH3 )2 A C(NH3 )2 SO4 C((NR3 H)2 )q (Cr2 O7 )p CSO4 Cr2 O2− 7 2− Cr3 O2− 10 ; Cr4 O13 D E e¯ f H2 A H2 CrO4 H2 SO4 KL N NR3 (NR3 H)2 A (NR3 H)2 CrO4 (NR3 H)2 Cr 2 O7 ((NR3 H)2 )q (Cr 2 O7 )p (NR3 H)2 SO4 SO4 p, q Ve Zt z (overbar) Greek letters βpq σ Subscripts max obs

acid anion coefficients in Eq. (12) concentration of acid anion in the aqueous phase (mol/dm3 ) overall concentration of acid in the aqueous phase (mol/dm3 ) concentration of uncomplexed Cr(VI) in aqueous phase (mol/dm3 ) overall concentration of complexed Cr(VI) (mol/dm3 ) initial concentration of Cr(VI) (mol/dm3 ) concentration of dichromate ion in the aqueous phase (mol/dm3 ) initial concentration of amine in solvent mixture (mol/dm3 ) concentration of free (uncomplexed) amine (mol/dm3 ) concentration of noncomplexed amine in organic-phase (mol/dm3 ) overall concentration of complex in the organic phase (mol/dm3 ) concentration of sulfate ion in the aqueous phase (mol/dm3 ) dichromate ion polychromate ions distribution ratio of Cr(VI) referred to the amine mixture degree of extraction, extracted Cr(VI)/initial Cr(VI) (%)  relative mean error, e¯ = (100/N) N |(Eobs −Emod )/Eobs | (%) function symbol undissociated inorganic acid in the aqueous phase chromic acid sulphuric acid Langmuir extraction constant (per (mol/dm3 )) number of observation tertiary amine quaternary ammonium salt of acid amine–Cr(VI) complex amine–Cr(VI) complex amine–Cr(VI) complex ammonium sulfate sulfate ion number of chromate ion and amine involved in complex, respectively initial amine (extractant) vol.% in solvent mixture (vol.%) overall loading factor of amine associated number species in the organic phase apparent equilibrium extraction constant ((mol/dm3 )1−p −q ) root-mean-square deviation maximum observed

A. Senol / Separation and Purification Technology 36 (2004) 63–75

water norm of Cr(VI) to be restricted strongly about (1.9×10−6 mol/dm3 ) that will become more severe in the near future. Excessive concentrations of chromium and other heavy metals in soluble form are thought of as a toxicant group which strictly reduces the activity of microorganisms. It is well known that the reduced oxidation state of chromium, Cr(III), is much less toxic and soluble at neutral pH values than its oxidized state of Cr(VI) [3,4]. Studies on biological systems by Veenstra et al. [3], and Cheremisinoff and Habib [4] have shown that the oxidized state is particularly worrying, because chromium is basely present in the Cr(VI) form in aerobic biological systems. The deleterious effects of toxic Cr(VI) compounds on biological processes are complex, and generally depend on the influent matrix such as pH, the concentration of toxicant, solubility, the toxicity composition limit of other incorporated ions or molecules present in the bulk [5]. Because of high toxicity of Cr(VI), emphasis in recent years has been on chromium removing, recovery and reuse. Liquid–liquid extraction is the most effective conventional method to be imperative to significantly reduce the Cr(VI) discharge levels and to promote recycling and reuse. Within the limited number of potential extractants, tributyl phosphate (TBP), tertiary (Alamine) or quaternary (Aliquat 336) amines, alkylphosphoric acid, and phosphine oxide (Cyanex 921, 923) and oxime derivatives (LIX 84) are of critical importance due to their favorable qualities such as coordination ability and stability of the complex strength [6–12]. Recently, liquid membrane techniques that combine the amine extraction and stripping operations in a single process are gaining considerable attention for separating and recovering Cr(VI) compounds from aqueous solutions [10–16]. A project of extensive equilibrium studies by Cr(VI) based on a nondispersive technique in hollow-fiber modules has been fulfilled by Ortiz and co-workers [15–18], Sirkar and co-workers [10,11], and Yang et al. [13]. At last, significant progress has been made in the use of neutral crown ethers, not only for the extraction of a metallic cation and its salts, but also for the extraction of more complex ion pairs appearing as a complex anion [19–21]. Another conventional method of treating wastewater containing Cr(VI) includes its chemical reduction to convert Cr(VI) into the Cr(III) cations by SO2 , Na2 SO3 , or

65

a ferrous ion compound (FeSO4 ) as reducing agents [22,23], and the following precipitation of cations as Cr(OH)3 . Permeable-reactive redox walls (reactive barriers) placed below the ground surface in the path of flowing groundwater could provide an alternative remediation approach for removing Cr(VI) chemicals from contaminated waters [24]. However, particularly frustrating aspects of the reduction method are the significant sludge production and the long term environmental consequences. The micellar enhanced ultrafiltration method has been applied to bind a heavy metal ion, especially Cr(VI) to a large species which is easily retained by an ultrafiltration membrane [2]. Recently, various natural and biological adsorbent materials, such as activated carbon, leaf mould, peat moss, silicagel, a natural clinoptilolite, and green algae [25–34] have been reported to be potentially useful in the removal of Cr(VI) and heavy metals from aqueous solution through adsorption. A comparison of low-cost sorbents and their applications have well documented by Polland et al. [35]. However, the adsorption process is generally costly. Microorganisms are also effective at removing many heavy metals from waters including chromium. Cr(VI) is an essential cellular micronutrient, and can be actively and temporarily accumulated in the bacterial cells, as long as toxic concentrations are not reached [3,36]. Additionally, many types of bacteria manufacture extracellular polymers, which also act to bind metals. Currently, to detoxify or remove various heavy metals activated sludge or other suspended growth processes have been used [3,36]. Nevertheless, among the possible alternatives to the treatment of industrial chromium plating baths or contaminated wastewaters, solvent extraction combining with membrane technology offers advantages [37–41]. However, the literature revealed very little insight relating to the validity of a generalized method for the treatment of Cr(VI) effluents at high concentration levels (over than 1.92 × 10−3 mol/dm3 ) and the evaluation of the effectiveness regeneration capabilities for recycling or recovery purposes [25,39–41]. In this context, the objective of this study has been to evaluate the distribution data for liquid–liquid extraction of Cr(VI) between water and Alamine 336 or Aliquate 336 used as a carrier. Process considerations dealing with the competition between various solvent extraction methods for Cr(VI), such as an amine, TBP, ethers

66

A. Senol / Separation and Purification Technology 36 (2004) 63–75

or alkylphosphoric acid solvents still remain a challenging problem since such systems show extremely nonideal behavior of a complex aggregation. Distribution of Cr(VI) between water and amine has been studied at isotermal conditions (298 K). This article will also discuss the effect of acid, amine and Cr(VI) concentrations on the extraction power of the solvent, as well as the competition between chemical interaction capabilities relative to Alamine 336 and Aliquat 336 regarding the pH range used. A graphical interpretation of optimum conditions has also been evaluated. Results were correlated in terms of a chemodel approach and modified Langmuir equilibrium model.

2. Theoretical Traditionally, characterization of the overall extraction equilibrium of Cr(VI)/amine/diluent system is evaluated due to simultaneous reactions (1) and (2) using a chemical modeling approach of King and co-workers [42,43] attributed to the amine system regarding as an ion exchanger. 2NR3 + H2 A = (NR3 H)2 A

(1)

q(NR3 H)2 A + pCr 2 O2− 7 = ((NR3 H)2 )q (Cr 2 O7 )p + qA2−

(2)

where H2 A, Cr2 O2− 7 , and NR3 represent the undissociated inorganic acid and dichromate ions in the aqueous phase, and tertiary amine, respectively. (NR3 H)2 A and ((NR3 H)2 )q (Cr 2 O7 )p denote quaternary ammonium salt of acid and the amine–Cr(VI) complex. Overbar is attributed to the species in the organic phase. The “conditioned” overall extraction constant (βpq ) of equilibrium in terms of Eq. (2) including the activity coefficients of species is defined in molarity scale (mol/dm3 )1−p−q , as: q

βpq =

C((NR3 H)2 )q (Cr2 O7 )p CA p

CCr2 O7 C(NR3 H)2 A

q

,

p = 1, k; q = 1, l (3)

where CCr2 O7 , CA , C(NR3 H)2 A , and C((NR3 H)2 )q (Cr2 O7 )p designate the equilibrium concentrations of noncomplexed dichromate ion and inorganic acid anion in

the aqueous phase, free (noncomplexed) amine and Cr(VI)–amine (p,q) complex, respectively. At a given temperature βpq is expected to depend on the ionic strength of the transferred metallic ion and the solvation efficiency of amine system used. The total equilibrium content of complexed Cr(VI), CCr(VI) = pC((NR3 H)2 )q (Cr2 O7 )p , is the sum of contributions of the individual complexes. CCr(VI) =

l k   p=1 q=1

p

q

q

pβpq CCr2 O7 C(NR3 )2 A /CA

(4)

Incorporating Eq. (4) into the balance equation for Cr(VI), the equilibrium model is derived. Interpretation of the equilibrium results of investigated amine systems by many authors [6,7] revealed that all possible Cr(VI)–amine (p,q) combinations for p = 1−k and q = 1−l should not be explicitly evaluated. In the prediction of equilibrium different sets of the appropriate amine–polychromate, i.e. one amine per multiple chromates (amine1 –Cr(VI)x ) aggregated structure combinations have been selected for Cr(VI), regarding the overall loading region and the maximum loading value, i.e. the plateau of the loading curve. Accordingly, aggregation of simple complexes into larger adducts has also been assumed, i.e. amineq –Cr(VI)p (q,p) complex formation of types (2, 3) and (3, 4). Cr(VI) may exist in the aqueous phase in different ionic forms with regard to the total amount of chromium and the pH variables dictating which particular chromate species will predominate [44,45]. H2 CrO4 ⇔ H+ + HCrO− 4 2− + HCrO− 4 ⇔ H + CrO4 2− 2HCrO− 4 ⇔ Cr 2 O7 + H2 O

(log K = −0.8) (log K = −6.5)

(5) (6)

(log K = 1.52) (7)

Higher forms of chromate polymers have also been reported by Sengupta and Clifford [44,45] in strong acidic solutions at high concentration of Cr(VI) larger than 0.1 mol/dm3 . − 2− + Cr2 O2− 7 + H + HCrO4 ⇔ Cr 3 O10 + H2 O

(8)

− 2− + Cr3 O2− 10 + H + HCrO4 ⇔ Cr 4 O13 + H2 O

(9)

This would call for assumption of a more complex aggregation between amine and polychromate species. The same remarks of transferring polymeric chromium

A. Senol / Separation and Purification Technology 36 (2004) 63–75

species through a membrane have been observed by Vallejo et al. [38]. Another questionable aspect of a preliminary analysis including Cr(VI)–amine systems [44,48] is the assumption that Cr 2 O2− 7 is the only counterion exchanged at acidic pH, as indicated in the following equation given for a weakly basic resin. 2− (NR3 H)2 SO4 + Cr 2 O2− 7 = (NR3 H)2 Cr 2 O7 + SO4

(10) Alternatively, Yang et al. [11] proposed an overall mechanism of the Cr(VI) extraction with an alkylamine through ion-pair formation of (NR3 H)2 CrO4 in contrast to the spectroscopic findings of Deptula [46] favoring (NR3 H)2 Cr 2 O7 aggregation from a sulphuric acid solution. The sorption of ionic species from aqueous solutions onto a solid could be basely described by the Langmuir isotherm. The assumptions of Langmuir type sorption through ion-exchange may be completely satisfied and adapted to amine–Cr(VI) system using the modified version of Li and Bowman [30], where the amine and Cr(VI) are regarded as adsorbent and adsorbat, respectively. CCr(VI) =

KL CCr(VI),max CCr(VI) 1 + KL CCr(VI)

(11)

where CCr(VI) , CCr(VI) and CCr(VI),max represent the Cr(VI) amount in aqueous phase, the overall complexed Cr(VI) by amine at equilibria and the maximum possible extraction capacity of amine referred to the plateau in the loading curve, respectively. The Langmuir extraction constant, KL , in per (mol/dm3 ) is attributed to the overall reaction in terms of Eq. (10), supposing the formation of only one type aggregated structure.

3. Experimental 3.1. Materials The commercial extractant Alamine 336 (Henkel Co.), a C8 –C10 saturated straight-chain tertiary amine mixture, is a pale yellow liquid practically insoluble in water (<5 ppm) with an average molecular weight of 392 g/mol, a density of 0.81 g/cm3 , and a viscosity of 11 mPa s (30 ◦ C). Analytical grade reagents

67

K2 Cr2 O7 (Merck, 99.5%) and H2 SO4 (Merck, 98%) were used to prepare acidic aqueous Cr(VI) solutions. Xylene used as a nonpolar diluent for amine system in the organic phase (≥ 99.5%, GC) was furnished from Fluka. All the chemicals were used without further purification. Deionized and redistilled water was used in aqueous solutions. 3.2. Experimental procedure The extraction experiments were performed using equilibrium glass cells, each equipped with a magnetic stirrer and thermostated at (298 ± 0.2) K. The equal volumes (10 cm3 ) of initial aqueous and organic phases were agitated for 2 h and then left to settle for about 18–20 h at a fixed temperature (298 K) and pressure (101.2 kPa). The effective separation of the phases was ensured by centrifugation. Aqueous-phase Cr(VI) concentration was determined spectrophotometrically using 1,5-diphenylcarbazide indicator [46] as well as a Shimadzu UV-160 A visible UV-spectrophotometer (540 nm). The Cr(VI) content in the organic phase was determined by mass balance. Because the third phase formation was observed in preliminary experiments with amine/xylene/aqueous Cr(VI) system for aqueous sulphuric acid concentrations varying above 4 mol/dm3 , the initial acid concentrations were restricted in the range 0.05–4 mol/dm3 . Tests covering the influence of amine concentrations on the extraction degree of Cr(VI) were performed using the initial amine concentrations in inert (xylene) diluent (C◦NH3 ) fixed in the range 0.01–0.145 and 0.041–0.620 mol/dm3 ◦ for the initial aqueous Cr(VI) concentrations (C Cr(VI) ) of 0.019 mol/dm3 (1 mg/ml) and 0.192 mol/dm3 (10 mg/ml), respectively. The aqueous phase H2 SO4 molarities of 0.05, 0.10, 0.50, 0.75, 1, 2, 3, and 4 mol/dm3 , as well as the effect of aqueous Cr(VI) concentration have been tested. Additionally, a comparative study of the extraction degrees of Cr(VI) by Alamine 336 and Aliquat 336 (Henkel Co.) extractants for the pH range of 1–7 has been evaluated. In these experiments, the Cr(VI) concentration of the simulated rinse-water was 5.8×10−3 mol/dm3 ; the extractant content in the organic phase was selected at 0.015 mol/dm3 . A 1 mol/dm3 H2 SO4 solution was used for pH adjustment of the aqueous phase to desired pH value. To avoid changes in the aqueous phase acid composition, before extraction a preliminary saturation of the organic

68

A. Senol / Separation and Purification Technology 36 (2004) 63–75

phase with sulphuric acid solution was carried out. However, a similar extraction degree was observed by the amine unsaturated with acid due to the high concentrations of acid used in the aqueous phase. 4. Results and discussion 4.1. Criterion of extraction degree The results were interpreted in terms of distribution ratio (D = CCr(VI) /CCr(VI) , the ratio of the overall extracted Cr(VI) in the organic phase to total aqueous-phase Cr(VI) concentration at equilibria), degree of extraction (E (%) = 100D/(1+D)), and overall (total) loading factor (Zt ). The overall loading factor of amine (Zt ) is the ratio of total amount of Cr(VI) extracted to total amount of amine in the organic phase, CCr(VI) /C◦NH3 . 4.2. Evaluation of results: factors influencing extraction power 4.2.1. Effect of amine and acid amounts Study of the extraction system in Figs. 1 and 2 con◦ taining C Cr(VI) = 0.019 mol/dm3 (1 mg/ml) Cr(VI) in

Fig. 1. Variation of extraction degree against amine concentration for eight initial sulphuric acid concentrations; Alamine 336/xylene ◦ system; C Cr(VI) =1 mg/ml=0.019 mol/dm3 .

Fig. 2. Comparison of extraction isotherms in terms of distribution ratio (D) depending on the initial amine and acid concentrations; ◦ Alamine 336/xylene system; C Cr(VI) =1 mg/ml = 0.019 mol/dm3 .

the aqueous phase solution, tested for eight different acidities, reveals that the extraction degree (%E) and the distribution ratio (D) of Cr(VI) are remarkably high varying against the Alamine 336 concentration asymptotically towards a maximum limit for an initial amine content (C◦NH3 ) larger than 1.5 vol.% (0.031 mol/dm3 ). It is apparent from Tables 1 and 2 that the same remarks hold for the initial Cr(VI) concentration of 0.192 mol/dm3 (10 mg/ml) favoring the extraction with a %E value of about 90% for C◦NH3 ≈ 0.124 mol/dm3 . Since the nonpolar Alamine 336 by itself is a relatively poor solvating medium for the polar complexes the rate of increasing the slope of the %E curve above the mentioned C◦NH3 limits is rather low by reason of the enhanced amount of amine reducing the solvation degree. Fig. 2 and Table 2 illustrate the influence of the acid and carrier concentrations on D which decreases reasonably with increasing the acid content in the aqueous phase. It can be recognized depending on the repartition diagram of Cr(VI) species that according to pH and chromium concentrations several and different in nature species of Cr(VI) may be present at a given pH [20]. Increasing acidity favours acidic species of chromate probably comprising complex formation of lower solvation degree.

A. Senol / Separation and Purification Technology 36 (2004) 63–75

69

Table 1 ◦ Extraction degree of Cr(VI) for Alamine 336/xylene system at 298 K (C Cr(VI) =10 mg/ml=0.192 mol/dm3 ) Alm336 concentration, C◦NH3 vol.%

2 4 6 7 8 9 10 12 15 20 30

%E

mol/dm3

Acid concentration in the aqueous phase, Cacid (mol/dm3 )

0.041 0.083 0.124 0.145 0.165 0.186 0.207 0.248 0.310 0.413 0.620

0.05

0.1

0.5

0.75

1

2

3

4

34.7 65.5 89.8 95.6 98.8 99.3 99.6 99.8 99.93 99.98 99.99

35.8 66.5 90.7 96.0 98.7 99.2 99.6 99.8 99.91 99.95 99.99

36.8 69.0 92.0 96.3 98.3 98.9 99.3 99.5 99.8 99.93 99.98

36.0 70.6 91.7 95.8 98.0 98.8 99.1 99.5 99.7 99.9 99.97

35.2 69.2 91.5 95.6 97.8 98.7 98.9 99.4 99.7 99.9 99.96

33.5 67.4 90.6 95.3 97.5 98.5 98.8 99.2 99.6 99.8 99.95

31.8 67.1 89.9 95.1 97.0 98.0 98.6 99.0 99.4 99.7 99.93

30.2 66.5 89.0 94.9 96.7 97.5 98.2 98.8 99.2 99.6 99.91

Accordingly, some authors assume that even in acidic media the dominating factor is the chromate extraction [6,20,48]. The noticeably lower D values at a high acidity range may also be attributed to the formation of intramolecular bonded species between H2 SO4 and Cr(VI) species resulting in a complex ion presumably capable of lower solvation affinity to amine. However, Sengupta and Clliford [44] has found that at high concentration of bisulfate (HSO− 4 ) Cr(VI) may form mononuclear complexes of type CrSO2− 7 that is indicative for decreasing D. In addition to the above Deptula [48] revealed that besides extraction of chromate ions, extraction of isopoly-acids must also be taken

into account, especially in concentrated solutions of Cr2 SO2− 7 and H2 SO4 . Conversely, the amine/diluent system favors the formation of overloaded polar Cr(VI)p –amineq structures (p ≥ q) corresponding to Zt >1. Study of Fig. 3 and Table 3 reveals that the overall loading factor (Zt ) and D exhibit a maximum restricted in the aqueous phase acidity range varying between 0.5 and 1 mol/dm3 for a lower amine loading relative to the ratio of initial con◦ centrations, C◦NH3 /C Cr(VI) ≤ 0.75. In the range above this value a continual decreasing of D (or Zt , Table 3) has been observed, as depicted in Fig. 3. To estimate the strength of the amine–Cr(VI) complexation

Table 2 ◦ Distribution ratio of Cr(VI) for Alamine 336/xylene system at 298 K (C Cr(VI) =10 mg/ml=0.192 mol/dm3 ) Alm336 concentration, C◦NH3 Vol.%

2 4 6 7 8 9 10 12 15 20 30

mol/dm3

0.041 0.083 0.124 0.145 0.165 0.186 0.207 0.248 0.310 0.413 0.620

D Acid concentration in the aqueous phase, Cacid (mol/dm3 ) 0.05

0.1

0.5

0.75

1

2

3

4

0.53 1.90 8.76 22.2 80.0 132.3 258.7 516.0 1351 3999 9999

0.56 1.99 9.81 23.7 76.3 126.4 237.1 433.8 1162 1922 8332

0.58 2.23 11.5 25.7 54.3 91.2 136.9 211.8 507.9 1448 4544

0.56 2.40 11.1 22.5 48.4 82.3 116.0 186.8 343.8 876.2 2940

0.54 2.25 10.7 21.6 44.4 75.9 94.7 169.9 316.5 713.3 2499

0.50 2.07 9.70 20.4 38.4 60.5 83.0 121.7 228.9 499.0 1817

0.47 2.04 8.88 19.5 32.3 50.0 70.4 96.1 167.1 321.6 1428

0.43 1.99 8.10 18.5 28.9 39.3 53.1 82.3 118.1 255.4 1162

70

A. Senol / Separation and Purification Technology 36 (2004) 63–75

Fig. 3. Variation of distribution ratio (D) with aqueous-phase sulphuric acid concentration for Alamine 336/xylene system; ◦ C Cr(VI) =1 mg/ml=0.019 mol/dm3 .

depending on the chromium concentration runs were performed using various Cr(VI) solutions. Referring to Fig. 4, it can be argued that the highest strength of the complex solvation yields the sulphuric acid concentration of 1 mol/dm3 promoting probably a complex formation of largest Zt ≈ 1.75 corresponding to the plateau in the loading curve. These concepts can be confirmed by the results from Table 3 and Fig. 4 manifesting that the controlling factor for chemical interaction is hydrophobicity, polarity, and ionizing strength of the formed polychromate structures or mononuclear complex ions distinguishing the divergent behaviors relative to the species at overly high concentration levels in the bulk. Besides this fact, a simultaneous co-extraction of bichromate (dimmer) and chromate species deservedly may occur. These findings, among other factors are comprehensively supported by the results from Figs. 1–5 presuming that different mechanisms control one or simultaneously at least two (amine per multiple chromates) complex formation depending on the solvation degree of diluent.

Table 3 Variation of loading factor (Zt ) with the initial amine, acid and Cr(VI) concentrations for Alamine 336/xylene system at 298 K Alm336 concentration, C◦NH3 Vol.%

C◦ 2 4 6 7 8 9 10 12 15 20 30

Cr(VI)

Zt

mol/dm3

= 10 mg/ml = 0.192 0.041 0.083 0.124 0.145 0.165 0.186 0.207 0.248 0.310 0.413 0.620

Acid concentration in the aqueous phase, Cacid (mol/dm3 ) 0.05

0.1

0.5

0.75

1

2

3

4

1.628 1.518 1.393 1.268 1.152 1.027 0.925 0.774 0.620 0.466 0.310

1.679 1.541 1.407 1.273 1.150 1.026 0.925 0.774 0.620 0.465 0.310

1.726 1.599 1.427 1.277 1.146 1.023 0.923 0.772 0.619 0.465 0.310

1.699 1.636 1.422 1.271 1.142 1.022 0.921 0.772 0.619 0.465 0.310

1.651 1.604 1.419 1.268 1.140 1.021 0.919 0.771 0.619 0.465 0.310

1.572 1.562 1.406 1.264 1.137 1.018 0.918 0.769 0.618 0.465 0.310

1.492 1.554 1.394 1.261 1.131 1.013 0.916 0.768 0.617 0.464 0.310

1.417 1.541 1.381 1.259 1.127 1.008 0.912 0.766 0.616 0.464 0.310

1.294 0.911 0.620 0.469 0.370 0.310 0.232 0.154 0.133

1.325 0.910 0.619 0.469 0.370 0.310 0.232 0.154 0.133

1.346 0.888 0.617 0.469 0.370 0.310 0.232 0.154 0.133

1.313 0.873 0.614 0.468 0.370 0.310 0.232 0.154 0.133

1.273 0.860 0.611 0.468 0.369 0.310 0.232 0.154 0.133

1.178 0.811 0.599 0.465 0.367 0.309 0.231 0.154 0.133

1.108 0.794 0.585 0.459 0.363 0.306 0.230 0.153 0.132

1.043 0.775 0.566 0.451 0.357 0.302 0.229 0.153 0.131

mol/dm3

C◦ Cr(VI) = 1 mg/ml = 0.019 mol/dm3 0.5 0.010 1 0.021 1.5 0.031 2 0.041 2.5 0.052 3 0.062 4 0.083 6 0.124 7 0.145

A. Senol / Separation and Purification Technology 36 (2004) 63–75

Fig. 4. Variation of extractability variables, loading factor (Zt ) and distribution ratio (D) of Cr(VI) with aqueous-phase Cr(VI) concentration for Alamine 336/xylene system; Cacid = 1 mol/dm3 ; C◦NH3 = 0.0413 mol/dm3 .

Consequently, it is expected the polarity and the ionizing strength of the formed chromium species at highly acidic media to control the complex formation of structures with different polarity that may influence the solvation degree. 4.2.2. Comparison of the extraction degrees for Alamine 336 and Aliquat 336 The percentages of extraction of Cr(VI) through two carriers, i.e. Alamine 336 and Aliquat 336, obtained for the identical solute and diluent conditions against various equilibrium pH values of the aqueous solutions are illustrated in Fig. 5. Referring to that figure it can be concluded that the extraction degree of Cr(VI) by Aliquat 336 increases in the pH range varying from 1 to 2. It remains almost 100% between pH at 2 and 4.5, and then declines for the pH values closely approximating neutral zone. The pH dependence of %E for Alamine 336 is very much different, as compared with Aliquat 336, reflecting a continual decreasing for the whole pH range. This trend of changes can be attributed to a conversion of NR3 H+ to neutral NR3 molecules, as pH approaches 7, presumably being inactive to extract Cr(VI) through chemical interaction. The physical extraction through dipole–dipole interaction appearing at a highly low level is negligible in nature. The results in Fig. 5 suggest that a moderate extraction degree of Cr(VI) by

71

Fig. 5. Comparison of extraction isotherms of Alamine 336 and Aliquat 336 systems depending on the aqueous phase pH range; ◦ C Cr(VI) = 5.8×10−3 mol/dm3 ; C◦NH3 = 0.015 mol/dm3 .

Alamine 336 at high acidic media accompanying by an appropriate range of stripping with alkalis may be performed through Alamine solutions, while Aliquat 336 promotes both extraction and stripping operations simultaneously at a wide range of pH values. However, the analysis for an effective stripping of Cr(VI) from the Aliquat 336 and Alamine 336 solutions, both containing equal amount of 1.92×10−3 mol/dm3 Cr(VI) by a 0.025 mol/dm3 NaOH solution results in 37 and 92% stripping degrees, respectively. Consequently, this property is an advantage for the treatment of Cr(VI)-contaminated wastewaters by Alamine 336 system only at high acidic region. The situation being reached at moderate and high values of pH would call for assumption that Aliquat 336 is the most valuable extractant comprising simultaneously both effective extraction and stripping for the overall pH range. This is in accordance with the findings of Ortiz and co-workers [15–18]. 4.3. Model estimates and reliability analysis. Graphical interpretation of optimum conditions The optimum conditions for amine extraction of Cr(VI) may be obtained through the graphical

72

A. Senol / Separation and Purification Technology 36 (2004) 63–75

using a log-basis linear homogeneous differential equation which explicit form results in the following function between %E and its dependent variables, Ve and Cacid , with the coefficients a and b depending on the initial Cr(VI) concentration. %E = %Emax {1 − exp(aVe + bCacid )}

Fig. 6. Estimation of optimum extraction conditions through graphical interpolation of equilibrium curves related to CCr(VI) = 1 mg/ml = 0.019 mol/dm3 and CCr(VI) = 10 mg/ml = 0.192 mol/dm3 concentrations for Alamine 336/diluent system; Cacid = 1 mol/dm3 .

interpretation of the equilibrium curve, %E = f(C◦NH3 )), as depicted in Fig. 6. The regions reflecting the asymptotically linear sections k and k for two initial Cr(VI) concentrations are attributed to an extraction power of amine at a highly low degree. Conversely, the corresponding l and l linear sections reproduce the extraction behaviors of amine at extremely small loading factors by reason of an excess amount of amine presumably favoring an undesirable emulsifying regime of third phase formation. So, the conditions attributed to these linear regions are practically not appropriate for an effective extraction process. The intersection of the asymptotical drawing lines to these sections at the intercepting points m and m for each curve, respectively, should provide the conditions of optimum extraction in terms of %E, Zt , and C◦NH3 variables, regarding the initial concentrations in the aqueous phase. A graphical interpolation of conditions attributed to m and m points results in the following optimum ranges: %E ≈ 95–99%, Zt ≈ 0.80–0.50 and C◦NH3 ≈ 0.021–0.038 mol/dm3 ◦ for C Cr(VI) = 0.019 mol/dm3 , and %E ≈ 91.5–95.6%, Zt ≈ 1.42–1.27 and C◦NH3 ≈ 0.124–0.145 mol/dm3 for ◦ C Cr(VI) = 0.192 mol/dm3 . The equilibrium data given in Fig. 1 and Table 1 for extraction degree (%E), initial amine concentrations (Ve (vol.%)) and acid molarities (Cacid ) were fitted

(12)

where %Emax is the maximum possible limit of %E for the system studied; Ve is the extactant (Alamine 336) vol.% in the solvent mixture; and Cacid is the molarity of sulphuric acid in aqueous solution (mol/dm3 ). Estimates were performed using the coef◦ ficients a = −1.679 and b = 0.231 for C Cr(VI) = 0.019 mol/dm3 , and a = −0.341 and b = −0.0055 for 0.192 mol/dm3 initial aqueous phase Cr(VI) content with regard to the selected eight acid molarities. Consequently, the proposed model, Eq. (12), proved to be reasonably accurate, yielding mean relative errors (¯e (%)) of 9.31% for 0.019 mol/dm3 initial Cr(VI) concentration and 6.96% for 0.192 mol/dm3 Cr(VI) concentration, respectively. The equilibrium data from Figs. 1–3 and Table 3 were interpreted in terms of Eq. (13) using the concepts of chemical modeling defined by Eqs. (3) and (4). The assumption inherent in this approach is attributed to a total concentration of complexed Cr(VI) (CCr(VI) = pC((NR3 H)2 )q (Cr2 O7 )p ) evaluated from Eq. (4) and the aqueous sulfate (SO2− 4 ) concentration equal to the initial acid molarity. Zt = =

CCr(VI) C◦NH3 l k p=1

q p q = 1 pβpq CCr(VI) C(NR3 H)2 SO4 q C◦NH3 CSO4

(13)

However, the chemodel presumes the formation of aggregated complexes. Estimates were performed using the multivariable procedures of linpack algorithm [47] for 1 and 2 selected appropriate complex combinations regarding Zt . The best fits display the approach comprising the formation of only one associated Cr(VI)–amine (p,q) structure of different stoichiometry depending on the acid molarity used, i.e. (2,1), (3,1), (3,2) and (4,3). Table 4 presents a quantitative assessment of the predicted equilibrium constants (βpq ) for selected individual complexes in terms of

A. Senol / Separation and Purification Technology 36 (2004) 63–75

73

Table 4 Extraction constants of Eqs. (13) and (14) and root-mean-square-deviation (σ) of model estimates for Cr(VI)–amine complexation in terms of acid concentration ◦



Cacid a (mol/dm3 )

C Cr(VI) = 0.019 mol/dm3 KL ; (σ(Zt )b ) (per (mol/dm3 ))

β; (p,q)c ; σ(Zt ) ((mol/dm3 )1−p−q )

KL ; (σ(Zt )b ) (per (mol/dm3 ))

β; (p,q)c ; σ(Zt ) ((mol/dm3 )1−p−q )

0.5

563.47; (0.337)

279.65; (0.311)

0.24449×103 0.40891×103 0.18058×104 0.20114×106

(2,1); (3,1); (4,1); (3,2);

0.800 0.856 0.873 0.874

1.0

423.21; (0.291)

176.15; (0.345)

0.35029×103 0.68174×103 0.30131×104 0.89845×106

(2,1); (3,1); (4,1); (3,2);

0.818 0.859 0.874 0.851

3.0

239.20; (0.186)

0.28405×105 (2,1); 0.435 0.20005×107 (3,1); 0.439 0.21795×109 (4,1); 0.440 0.64399×1010 (3,2); 0.434 0.87062×1017 (4,3); 0.353 0.39177×105 (2,1); 0.419 0.25444×107 (3,1); 0.431 0.24879×109 (4,1); 0.432 0.47190×1010 (3,2); 0.414 0.40936×1016 (4,3); 0.338 0.57269×105 (2,1); 0.376 0.31689×107 (3,1); 0.403 0.25129×109 (4,1); 0.409 0.26331×1010 (3,2); 0.367 0.26559×1015 (4,3); 0.327

65.53; (0.492)

0.59300×103 0.14279×104 0.62652×104 0.60582×108

(2,1); (3,1); (4,1); (3,2);

0.822 0.849 0.861 0.793

a b c

C Cr(VI) = 0.192 mol/dm3

Initial aqueous sulphuric acid concentration. Root-mean-square-deviation (RMSD) in terms of Zt . Stoichiometric ratio Cr(VI)–amine in the complex formation.

root-mean-square-deviation (σ) of Zt factor, regarding three molarities of acid used, i.e. 0.5, 1 and 3 mol/dm3 . The data from Table 3 were also correlated with respect to the modified Langmuir approach of Li and Bowman [30] defined by Eq. (11) assuming that z = Zt,max = CCr(VI),max /C◦NH3 . The model was rearranged through incorporating CCr(VI) from Eq. (11) into Eq. (14) to give a sentence structure including the Zt and KL variables. Zt =

CCr(VI) C◦NH3

=

zKL CCr(VI) 1 + KL CCr(VI)

(14)

The estimated KL values in per (mol/dm3 ) by Eq. (14) depending on three acid concentrations of 0.5, 1 and 3 M are given in Table 4. The maximum loading value (z = Zt,max ) of 1.75 was performed due to Fig. 4. The reliability analysis of Eq. (14) results in σ(Zt ) of 0.271 and 0.383 for 0.019 and 0.192 mol/dm3 initial Cr(VI) concentrations, respectively. The model is expected to be an improvement in data fit for the associated Cr(VI)–amine systems. Table 4 illustrates the consistency of predictions achieved for both the chemodel and the modified Langmuir approach defined by Eqs. (13) and (14). Consequently,

both approaches proved to be reasonably accurate, yielding σ(Zt ) = 0.327 for Eq. (14) and σ(Zt ) = 0.583 for Eq. (13) considering all the systems studied.

5. Conclusions The isothermal distribution of Cr(VI) onto aqueous/organic two-phase system containing Alamine 336 (or Aliquat 336) as a reactive carrier has been elucidated by simultaneous effects of chemical and physical interactions closely related to the nature of amine and the anionic species transferred. The acidity and the chromium concentration affecting the distribution of the anionic species in the aqueous phase appear to be the main variables in the extraction reagent affinity for Cr(VI). Characterization of Cr(VI)–amine complexation is intimately connected to the solvation efficiency of amine, the acidity, and the affinity of Cr(VI) to polychromate formation in the aqueous phase. Simultaneously, the effect of additional controlling factors, such as the swing effect of a mixed diluent and the third phase formation can also modify the reversible complexation stage. The way to formulate a design strategy for Cr(VI)–amine

74

A. Senol / Separation and Purification Technology 36 (2004) 63–75

complexation including chemical interaction variables along with contribution of overloaded (p ≥ q) structures has been discussed. The largest strength of the complex solvation, promoting probably a simultaneous co-extraction of dichromate (dimmer) species and chromate (monomer) species responsible for various Cr(VI)–amine complex formation referred to Zt,max ≈ 1.75 has been found for 1 mol/dm3 acid so◦ lution and the optimum ratio of C◦NH3 /C Cr(VI) ≈ 0.75. Chemodel presumes the formation of aggregated Cr(VI)–amine structures of type (2, 1), (3, 1), (3, 2) and (4, 3). The extraction results were also correlated in terms of a modified Langmuir model and a linear homogeneous differential approach for the %E variable. It is also expected a version of the Markus et al. linear solvation energy relation [49] including the solvatochromic parameters of dipole–dipole interactions of ionic species to reflect the influence of the factors mentioned. Practical applications of the solvent extraction process by a tertiary or quaternary amine system have been extensively studied for the removal of Cr(VI) from washing operations of final metallic products but until not there has not been any distribution data of the Alamine 336 extraction of Cr(VI) from water effluents at highly acidic aqueous media including larger amount of chromium(VI). Much research in these phenomena remains to be done, above all at different isothermal conditions with mixed amine carriers to estimate the factors modifying the regeneration stage.

Acknowledgements The author thanks Henkel Co. for providing Alamine 336 and Aliquat 336 solvents. The author is also grateful to Professor Dr Emin Ulusoy and Research Fund of the Istanbul University for the technical support of this study.

References [1] J.A. Salvato, P.E. Dee, Environmental Engineering and Sanitation fourth ed (1992) WileyNew York. [2] Z. Sadaoui, C. Azoug, G. Charbit, F. Charbit, J. Chem. Eng. Jpn. 30 (1997) 799–805. [3] J.N. Veenstra, D.A. Sanders, A. Seyoung, J. Environ. Eng. June (1999) 522–531.

[4] P.N. Cheremisinoff, Y.H. Habib, Water Sewage Works July (1972) 73–86. [5] Z. Lewandowski, Water Res. 19 (1985) 589–596. [6] Y.H. Huang, C.Y. Chen, J.F. Kuo, J. Chem. Eng. Jpn. 24 (1991) 149–154. [7] M.M. Hossain, T. Ru, J. Chem. Eng. Data 47 (2002) 203–208. [8] E. Salazar, M.I. Ortiz, A.M. Urtiaga, J.A. Irabien, Ind. Eng. Chem. Res. 31 (1992) 1516–1522. [9] B. Galán, A.M. Urtiaga, A.I. Alonso, J.A. Irabien, M.I. Ortiz, Ind. Eng. Chem. Res. 33 (1994) 1765–1770. [10] C.H. Yun, R. Prasad, A.K. Guha, K.K. Sirkar, Ind. Eng. Chem. Res. 32 (1993) 1186–1195. [11] Z.F. Yang, A.K. Guha, K.K. Sirkar, Ind. Eng. Chem. Res. 35 (1996) 4214–4220. [12] T. Vincent, E. Guibal, Ind. Eng. Chem. Res. 40 (2001) 1406– 1411. [13] X.Y. Yang, A.G. Fane, K. Soldenhoff, Ind. Eng. Chem. Res. 42 (2003) 392–403. [14] S.H. Lin, R.S. Juang, Ind. Eng. Chem. Res. 41 (2001) 2693– 2701. [15] M.I. Ortiz, B. Galán, J.A. Irabien, Ind. Eng. Chem. Res. 35 (1996) 1369–1377. [16] A.I. Alonso, A.M. Urtiaga, J.A. Irabien, M.I. Ortiz, Chem. Eng. Sci. 49 (1994) 901–909. [17] A.I. Alonso, B. Galán, M. Gonzalez, I. Ortiz, Ind. Eng. Chem. Res. 38 (1999) 1666–1675. [18] A.M. Eliceche, A.I. Alonso, I. Ortiz, Comput. Chem. Eng. 24 (2000) 2115–2123. [19] Y. Takada, Top. Curr. Chem. 121 (1984) 1–38. [20] C. Musikas, W.W. Schultz, in: J. Rydberg, C. Musikas, G. Choppin (Eds.), Principles and Practices of Solvent Extraction, Marcel Dekker, New York, 1992 (Chapter 11). [21] A.H. Bond, M.L. Dietz, B. Chiarizia, Ind. Eng. Chem. Res. 39 (2000) 3442–3464. [22] L.E. Eary, D. Rai, Environ. Sci. Technol. 22 (1988) 972–977. [23] M.R. Unnithan, T.S. Anirudhan, Ind. Eng. Chem. Res. 40 (2001) 2693–2701. [24] D.W. Blowes, C.J. Ptacek, J.L. Jambor, Environ. Sci. Technol. 31 (1997) 3348–3357. [25] S.K. Ouki, R.D. Neufeld, J. Chem. Technol. Biotechnol. 70 (1997) 3–8. [26] L. Monser, N. Adhoum, Sep. Purif. Technol. 26 (2002) 137– 146. [27] D.C. Sharma, C.F. Foster, Biores. Technol. 49 (1994) 31–40. [28] D.C. Sharma, C.F. Foster, Biores. Technol. 52 (1995) 261– 267. [29] K.S. Low, C.K. Lee, S.G. Tan, Environ. Technol. 18 (1997) 449–454. [30] Z. Li, R.S. Bowman, Environ. Sci. Technol. 31 (1997) 2407– 2412. [31] D. Roy, P.N. Greenlaw, B.S. Shane, J. Environ. Sci. Health A28 (1993) 37–50. [32] D. Gang, W. Hu, S.K. Banerji, T.E. Clevenger, Ind. Eng. Chem. Res. 40 (2001) 1200–1204. [33] Z. Hu, L.L. Yijiu, Y. Ni, Sep. Purif. Technol. 31 (2003) 13– 18.

A. Senol / Separation and Purification Technology 36 (2004) 63–75 [34] V.A. Tsiatouras, N. Evmiridis, Ind. Eng. Chem. Res. 42 (2003) 1137–1144. [35] S.J.J. Polland, G.D. Fowler, C.G. Sollars, R. Perry, Sci. Total Environ. 116 (1992) 31–52. [36] F.B. Dilek, C.F. Gokcay, U. Yetis, Water Res. 32 (1998) 303– 312. [37] R. Molinari, E. Drioli, G. Pantano, Sep. Sci. Technol. 24 (1989) 1015–1032. [38] M.E. Vallejo, F. Persin, C. Innocent, P. Sistat, G. Pourcelly, Sep. Purif. Technol. 21 (2000) 61–69. [39] B. Gupta, A. Deep, S.N. Tandon, Ind. Eng. Chem. Res. 41 (2002) 2948–2952. [40] F.J. Algucil, A.G. Coedo, M.T. Donato, Hydrometallurgy 57 (2000) 51–57. [41] F.J. Algucil, A.G. Coedo, M.T. Donato, A.M. Sastre, Hydrometallurgy 61 (2001) 13–19.

75

[42] A.S. Kertes, C.J. King, J. Biotechnol. Bioeng. 28 (1986) 269– 282. [43] J.A. Tamada, A.S. Kertes, C.J. King, Ind. Eng. Chem. Res. 29 (1990) 1319–1326. [44] A.K. Sengupta, D. Clifford, Ind. Eng. Chem. Fundam. 25 (1986) 249–258. [45] A.K. Sengupta, S. Subramanian, D. Clifford, J. Environ. Eng. 114 (1988) 137–153. [46] J. Bassett, R.C. Denney, G.H. Jeffery, J. Mendham, Vogel’s Textbook of Quantitative Inorganic Analysis, fifth ed, Longman, New York, 1978. [47] D.M. Himmelblau, Basic Principles and Calculations in Chemical Engineering, fifth ed, Prentice-Hall, Englewood Cliffs, NJ, 1989. [48] C. Deptula, J. Inorg. Nucl. Chem. 30 (1968) 1309–1316. [49] Y. Markus, M.J. Kamlet, R.W. Taft, J. Phys. Chem. 92 (1988) 3613–3622.