Inorganic ligand-modified, colloid-enhanced ultrafiltration: A novel method for removing uranium from aqueous solution

water research 43 (2009) 4751–4759

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Inorganic ligand-modified, colloid-enhanced ultrafiltration: A novel method for removing uranium from aqueous solution Jim D. Roach a,*, Jesus H. Zapien b a b

Pre-Medical Education Unit, Weill Cornell Medical College-Qatar, PO Box 24144, Doha, Qatar Chemistry Department, Emporia State University, Emporia, Kansas 66801, USA

article info

abstract

Article history:

Inorganic ligand-modified, colloid-enhanced ultrafiltration (ILM-CEUF) is as a novel

Received 20 April 2009

membrane-based separation method for selectively removing target ions from aqueous

Received in revised form

solution. Traditional colloid-enhanced ultrafiltration (CEUF) is a well-established

30 July 2009

membrane-based separation technique that can be used to separate metal ions from other

Accepted 3 August 2009

aqueous solution components. Ligand-modified, colloid-enhanced ultrafiltration (LM-

Published online 11 August 2009

CEUF) uses organic ligands that selectively complex target ions and also associate with a water-soluble colloid, such as a surfactant micelle or polyelectrolyte. The colloid, asso-

Keywords:

ciated -ligand, and target ion are then concentrated using an ultrafilter, producing a filtrate

Uranium

with a low concentration of the target ion. While traditional LM-CEUF techniques are able

Colloid-enhanced ultrafiltration

to provide quantitative separations of a variety of ionic pollutants, the high costs of the

Ultrafiltration

chelating agents make such techniques nonviable in most remediation schemes. This

Equilibrium dialysis

study investigated the replacement of organic ligands with carbonate for the selective

Poly(diallyldimethylammonium)

removal of U(VI) from aqueous solution. In slightly to moderately basic solutions containing carbonate, UO2(CO3)4 3 can be made

chloride Polyelectrolyte

to dominate the U(VI) speciation. Using poly(diallyldimethylammonium) chloride, the

Carbonate

effectiveness and efficiency of ILM-CEUF for removing U(VI) from other aqueous solution components was investigated as a function of carbonate concentration, pH, and ionic strength. Uranium separations of greater than 99.6% were achieved; even in the presence of large excesses of competing ions. The specific separation of U(VI) from Sr2þ was also examined. ª 2009 Elsevier Ltd. All rights reserved.

1.

Introduction

Improved techniques for removing uranium from aqueous solution are needed because of growing concern over uranium levels in natural waters and also because uranium must be isolated from other spent nuclear fuel components for proper storage and disposal. At sites throughout the world, groundwater supplies are threatened by advancing plumes containing uranium. These plumes are the result of uranium release from natural mineralization, mine tailings, phosphate

fertilizers, and even munitions containing depleted uranium (Kratz and Schnug, 2006). In samples taken from test wells at a site near Fry Canyon, Utah, USA, uranium concentrations as high as 16.3 mg L1 were reported (Naftz et al., 2004). The United States Environmental Protection Agency recently established a maximum contaminant level (MCL) for uranium in drinking water of just 0.030 mg L1 (U.S. EPA, 2000). Separation of uranium from other spent nuclear fuel metals such as strontium and cesium greatly increases the efficiency by which repository space can be utilized. Presently, there is

* Corresponding author. Tel.: þ974 492 8240; fax: þ974 492 8111. E-mail address: [email protected] (J.D. Roach). 0043-1354/$ – see front matter ª 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2009.08.007

4752

water research 43 (2009) 4751–4759

approximately 54,000 tons of spent nuclear fuel in interim storage at operational and decommissioned power plants throughout the United States (Ahearne et al., 2007). Achieving the goal of a consolidated permanent disposal facility (i.e. Yucca Mountain, Nevada) will be more easily attained with improved uranium separation technologies. Uranium separation from other aqueous solution components has generally involved either solvent extraction (Kolarik, 1981) or ion-exchange (Zhang and Clifford, 1994). However, several less traditional schemes have been investigated in recent years. Removal of uranium from aqueous solution through coagulation with iron and aluminum was studied by Sorg (1988) and Gafvert et al. (2002). Separation schemes involving uranium adsorption onto various media, including modified granulated activated carbon (Coleman et al., 2003), apatite (Bostick et al., 1999), and chitosan (Gerente et al., 1999), have also been investigated. Use of a nanofiltration process for separation of uranium complexes in water was studied by both Raff and Wilken (1999) and FavreRe´guillon et al. (2005). A remediation technique utilizing permanent reactive barriers composed of zero valent iron to reduce water-soluble uranium species was field-tested by the U.S. EPA (Naftz et al., 2004; Feltcorn and Breeden, 1997). Microorganisms have even been examined as possible remediating agents (Tsuruta, 2007). In addition, ultrafiltration

based separation techniques have been studied using micellesolubilized chelating agents like trioctylphosphine oxide (Pramauro et al., 1992) and 4-aminosalicylic acid (Pramauro et al., 1996). As illustrated in Fig. 1, the technique described in this study utilizes a water-soluble polyelectrolyte, an inorganic ligand, and ultrafiltration in a novel way. Colloid-enhanced ultrafiltration (CEUF) is a well-established membrane-based separation technique that can be used to remove ions from aqueous solution (Scamehorn et al., 1986; Christian et al., 1995; Dunaway et al., 1998; Tondre, 2000). General CEUF techniques can use charged macromolecular species such as surfactant micelles, in a process called micelle enhanced ultrafiltration (MEUF), or polyelectrolytes, in a process called polyelectrolyte-enhanced ultrafiltration (PEUF), to electrostatically bind oppositely charged ions. The colloid and target ion are then concentrated using an ultrafilter, producing a filtrate with a lower concentration of the target ion. The larger pore diameters of UF membranes can provide higher permeability and permeate flux values than those associated with nanofiltration (NF) membranes. The removal of several multivalent species, both positively and negatively charged, through CEUF techniques is well documented. The MEUF process has been investigated using hexadecylpyridinium chloride for the removal of chromate (Christian et al., 1988) and using tetradecyl trimethyl

Fig. 1 – U(VI) removal using ILM-PEUF.

water research 43 (2009) 4751–4759

4753

ammonium bromide for the removal of perchlorate (Yoon et al., 2003). The PEUF process has been investigated using sodium poly(styrenesulfonate) for the removal of Cu2þ (Sasaki et al., 1989), using sodium poly(styrenesulfonate) for the removal of hardness ions (Tabatabai et al., 1995), using poly(diallyldimethylammonium) chloride for the removal of chromate (Sriratana et al., 1996), using poly (diallyldimethylammonium) chloride for the removal of arsenic (Pookrod et al., 2004; Gallo et al., 2006), and using

poly(diallyldimethylammonium) chloride for the removal of perchlorate (Huq et al., 2007; Roach and Tush, 2008). While near quantitative separations are achievable with many colloid/contaminant systems, an inherent problem with using conventional CEUF for the removal of ions, especially cationic metals, is that there is no selectivity in the process, except on the basis of the charge of the target ion. In addition for MEUF processes, because the critical micelle concentration (CMC) of surfactant monomers is not in micelles, not all surfactant molecules are retained in the retentate. Hence, while target ions are removed using MEUF, other contaminants, the surfactant molecules themselves, are introduced. Specificity for cation removal has been achieved through ligand-modified, CEUF (LM-CEUF) which uses derivatized chelating agents that selectively bind a target metal ion, and then solubilize in, or electrostatically bind to, a colloidal pseudophase. Ligand- modified, micellar-enhanced ultrafiltration (LM-MEUF) requires a ligand that consists of a chelating group and a hydrophobic moiety. Such ligands are able to bind a target metal ion and then solubilize in surfactant micelles. The LM-MEUF process has been investigated using iminodiacetic acid derivatives for the removal of Cu2þ by Klepac et al. (1991) and using nitrilotriacetic acid derivatives for the removal of Pb2þ by Roach et al. (2003). Near quantitative separations were achieved with each of the aforementioned ligand/metal systems. Ligand-modified, polyelectrolyte-enhanced ultrafiltration (LM-PEUF) utilizes multivalent anionic ligand-metal complexes that electrostatically bind to cationic polyelectrolytes. Through use of high-molecular weight polyelectrolytes, polymer does not enter the permeate stream,

Fig. 2 – Distribution diagram of the U(VI)-hydroxide system at 25 8C in the range 3.00
Fig. 3 – Distribution diagram of the U(VI)-hydroxidecarbonate system at 25 8C in the range 3.00
Table 1 – Uranium (VI) speciation in aqueous solutions containing carbonate. U(VI)-Complex

Complex formation reaction

UO2OHþ UO2(OH)2 UO2(OH) 3 UO2(OH)2 4 (UO2)2OH3þ (UO2)2(OH)2þ 2 (UO2)3(OH)2þ 4 (UO2)3(OH)þ 5 (UO2)3(OH) 7 (UO2)4(OH)þ 7 UO2CO3 2 UO2(CO3)2 UO2(CO3)4 3 (UO2)3 (CO3)6 6 (UO2)2 CO3(OH) 3

 þ UO2þ 2 þ OH % UO2OH  UO2þ þ 2 OH % UO (OH) 2 2 2   UO2þ 2 þ 3 OH % UO2(OH)3 2  UO2þ þ 4 OH % UO (OH) 2 2 4  3þ 2 UO2þ 2 þ OH % (UO2)2OH 2þ  2 UO2þ þ 2 OH % (UO ) (OH) 2 2 2 2 2þ  3 UO2þ 2 þ 4 OH % (UO2)3(OH)4 þ  3 UO2þ þ 5 OH % (UO ) (OH) 2 2 3 5   3 UO2þ 2 þ 7 OH % (UO2)3(OH)7 2þ  4 UO2 þ 7 OH % (UO2)4(OH)þ 7 2 UO2þ 2 þ CO3 % UO2CO3 2þ 2 2 UO2 þ 2 CO3 % UO2(CO3)2 2 4 UO2þ 2 þ 3 CO3 % UO2(CO3)3 2 6 3 UO2þ þ 6 CO % (UO ) (CO 2 3 2 3 3) 6 2þ 2  2 UO2 þ CO3 þ 3 OH % (UO2)2 CO3(OH) 3 2  11 UO2þ 2 þ 6 CO3 þ 12 OH % (UO2)11(CO3)6(OH)2 12

(UO2)11(CO3)6(OH)2 12

log10Ka 8.8 17.7 22.8 23 11.3 22.38 44.1 54.45 67 76.1 9.68 16.9 21.60 54.0 41.2 204

a Values taken from Grenthe et al. (2004).

4754

water research 43 (2009) 4751–4759

Table 2 – Predominance of U(VI) species as function of CO2L concentration at [U(VI)]total [ 5 mM and pH [ 8.0. 3

Table 4 – Predominance of U(VI) species as function of CO2L concentration at [U(VI)]total [ 5 mM and pH [ 9.0. 3

UO2 UO2(CO3)2 (UO2)2CO3 UO2 UO2 [CO2 3 ]total 2 (mM) (OH)2 (OH) (%) (CO3)4 (OH) 3 3 3 (%) (%) (%) (%)

UO2 UO2 UO2 (UO2)2CO3 [CO2 3 ]total UO2 2 4  (mM) (OH)2 (OH) 3 (CO3)2 (%) (CO3)3 (%) (OH)3 (%) (%) (%)

0.10 0.20 0.25 0.50 0.67 1.00 2.00 2.50 5.00

0.10 0.20 0.25 0.50 0.67 1.00 2.00 2.50 5.00

42.43 32.00 28.80 18.10 13.42 7.14 1.37 0.75 0.10

6.40 4.82 4.32 2.72 2.02 1.08 0.21 0.11 0.02

1.78 5.42 7.60 19.16 25.20 30.20 23.40 19.94 11.20

0.28 1.70 2.98 15.06 26.40 47.40 73.80 78.80 88.60

48.83 55.60 56.00 44.80 32.64 13.84 1.03 0.38 0.01

thus eliminating a drawback of LM-MEUF. The effectiveness of LM-PEUF is well-established. Tuncay et al. investigated the application of two commercially available polyprotic ligands, 4,5-dihydroxy-1,3-benzenedisulfonic acid (Tiron) (Tuncay et al., 1994a) and diethylenetriaminepentaacetic acid (DTPA) (Tuncay et al., 1994b), for the removal of Cu2þ and Pb2þ using LM-PEUF. Roach et al. (2003) studied an NTA derivative for the selective removal of Pb2þ in the presence of Ca2þ using LM-PEUF. While both LM-MEUF and LM-PEUF processes can provide quantitative separation of solute species, their reliance on costly organic chelating molecules severely limits their economic viability. Replacing the costly organic ligands with inorganic ligands, e.g. carbonate, holds great promise for improving the economic viability of LMCEUF processes. Many inorganic ligands form a myriad of complexes with metals in aqueous solution. Of particular interest for inorganic ligand-modified, colloid-enhanced ultrafiltration (ILM-CEUF) investigations involving uranium are multivalent anionic U(VI)-carbonate complexes. In solution these anionic complexes are expected to partition in cationic colloidal pseudophases, and thus be retained during ultrafiltration. Uncomplexed metal ions would be expelled as a result of repulsion between cationic colloid and cationic metal solution components (Klepac et al., 1991; Tuncay et al., 1994a; Christian et al., 1989; Krehbiel et al., 1992). This partitioning and expulsion provide the basis for a separation process.

29.60 23.20 20.20 8.70 4.66 1.58 0.21 0.11 0.01

44.60 35.00 30.60 13.14 7.02 2.38 0.32 0.16 0.02

1.13 3.52 4.80 8.16 7.78 6.00 3.22 2.58 1.31

2.

Materials and methods

2.1.

Chemicals

1.69 10.50 17.82 60.20 76.81 89.40 96.20 97.20 98.60

22.72 27.72 26.44 9.72 3.71 0.65 0.02 0.01 0.00

Poly(diallyldimethylammonium) chloride (PDADMAC), 20% in water with an average molecular weight between 400,000 and 500,000 Da, was obtained from Aldrich Chemical Co. The molar concentration of PDADMAC solutions are reported in terms of the monomeric unit. Reagent grade uranyl nitrate hexahydrate was obtained from Mallinkrodt. Solutions were prepared using reagent grade sodium chloride, hydrochloric acid, sodium hydroxide, and sodium carbonate obtained from Fisher Scientific. The water used in this study was purified by a Labconco WaterPro RO/PS system.

2.2.

PDADMAC purification

PDADMAC was purified by removing low molecular weight fragments through ultrafiltration using an Amicon model 8400 stirred batch cell equipped with an Amicon Diaflo10,000 Da MWCO regenerated cellulose ultrafilter. Retentate PDADMAC was washed with deionized water until no polymer was detected in the permeate through Total Organic Carbon analysis as performed on a Dohrman Model DC-180 total organic carbon (TOC) analyzer. This was done to ensure that PDADMAC fragments would not be able to pass through dialysis membranes.

Table 3 – Predominance of U(VI) species as function of CO2L concentration at [U(VI)]total [ 5 mM and pH [ 8.5. 3

Table 5 – Predominance of U(VI) species as function of CO2L concentration at [U(VI)]total [ 5 mM and pH [ 9.5. 3

UO2 UO2 (UO2)2CO3 UO2 UO2 [CO2 3 ]total (mM) (OH)2 (%) (OH) (CO3)2 (CO3)4 (OH) 3 2 3 3 (%) (%) (%) (%)

[CO2 3 ]total (mM)

UO2 (OH)2 (%)

UO2 (OH) 3 (%)

UO2 (CO3)2 2 (%)

UO2 (CO3)4 3 (%)

(UO2)2CO3 (OH) 3 (%)

0.10 0.20 0.25 0.50 0.67 1.00 2.00 2.50 5.00

0.10 0.20 0.25 0.50 0.67 1.00 2.00 2.50 5.00

15.92 13.46 11.86 4.70 2.42 0.80 0.10 0.05 0.01

75.80 64.20 56.60 22.40 11.52 3.80 0.49 0.25 0.03

0.46 1.54 2.12 3.32 3.04 2.28 1.19 0.96 0.48

1.91 12.64 21.60 67.00 82.20 93.00 98.20 98.80 99.40

5.72 8.12 7.88 2.46 0.87 0.14 0.00 0.00 0.00

38.60 29.40 26.20 14.28 9.08 3.68 0.54 0.28 0.04

18.42 14.00 12.44 6.82 4.34 1.75 0.26 0.13 0.02

1.61 4.92 6.82 14.82 16.74 15.32 9.06 7.42 3.86

0.79 4.86 8.40 36.40 55.00 75.60 90.00 92.20 96.00

40.40 46.80 46.00 27.52 14.84 3.65 0.16 0.05 0.00

4755

water research 43 (2009) 4751–4759

Table 6 – Predominance of U(VI) species as function of CO2L concentration at [U(VI)]total [ 5 mM and pH [ 10. 3

Table 7 – Predominance of U(VI) species as function of CO2L concentration at [U(VI)]total [ 5 mM and pH [ 10.5. 3

[CO2 3 ]total (mM)

[CO2 3 ]total (mM)

0.10 0.20 0.25 0.50 0.67 1.00 2.00 2.50 5.00

2.3.

UO2 (OH)2 (%)

UO2 (OH) 3 (%)

UO2 (CO3)2 2 (%)

6.12 5.76 5.42 3.06 1.83 0.68 0.09 0.07 0.01

92.40 86.80 81.80 46.20 27.60 10.32 1.41 1.11 0.09

0.09 0.33 0.49 1.08 1.14 0.97 0.54 0.50 0.22

UO2 (UO2)2CO3 (CO3)4 (OH) 3 3 (%) (%) 0.82 6.06 11.10 49.00 69.00 88.00 98.00 98.40 99.60

0.60 1.05 1.16 0.74 0.35 0.07 0.00 0.00 0.00

0.10 0.20 0.25 0.50 0.67 1.00 2.00 2.50 5.00

Equilibrium dialysis

Equilibrium dialysis (ED) experiments were performed using dialysis cells from Fisher Scientific, having 5 mL compartments separated by 6000 Da MWCO regenerated cellulose membranes, also obtained from Fisher Scientific. Test solutions were placed in one compartment, called the retentate, and deionized water was placed in the other compartment, called the permeate. The cells were equilibrated at room temperature for four days before retentate and permeate solutions were removed. Previous studies have shown that equilibrium is typically achieved in such experiments within forty-eight hours (Christian et al., 1994). Separation results obtained using equilibrium dialysis techniques typically agree well with those obtained using ultrafiltration processes (Roberts et al., 2000; Roach and Tush, 2008).

2.4.

pH adjustments

Solution pH was adjusted by adding acid or base dropwise to test solutions. The pH was determined using a pH meter that was standardized in solutions containing calibration buffers and PDADMAC. Because ionic strength affects the binding of anions to PDADMAC, buffers were not used in the ED studies. Instead, solution pH was adjusted through addition of 0.1 M hydrochloric acid and 0.1 M sodium hydroxide to the desired value.

2.5.

Analytic methods

Uranium concentrations were determined spectrophotometrically on a Cary 50 Bio UV–vis spectrophotometer using arsenazo III as described previously (Savvin, 1961; Kressin, 1984). The limit of detection for uranium using this method was found to be 0.355 mM. Strontium concentrations were determined by flame AA analysis using a Perkin Elemer AAnalyst 100 and the method of standard additions. For ED experiments, the effectiveness of U(VI) retention was defined in terms of the uranium concentration in the permeate, [U]permeate, and its initial concentration, [U]initial.    %Retention ¼ ð100%Þ 1  ½Upermeate =½Uinitial

(1)

Equation (2) was used to quantify the extent of strontium expulsion from retentate to permeate using strontium

UO2 (OH)2 (%)

UO2 (OH) 3 (%)

UO2 (CO3)2 2 (%)

UO2 (CO3)4 3 (%)

(UO2)2CO3 (OH) 3 (%)

2.04 2.04 2.02 1.78 1.52 0.95 0.20 0.11 0.01

97.80 96.80 96.00 85.00 72.40 45.00 9.44 5.06 0.66

0.01 0.03 0.05 0.17 0.26 0.36 0.30 0.25 0.13

0.12 0.94 1.82 12.86 25.60 53.60 90.00 94.60 99.20

0.03 0.07 0.08 0.13 0.13 0.07 0.00 0.00 0.00

concentrations in both permeate ([Sr]permeate) and retentate ([Sr]retentate) after equilibration.    %Expulsion ¼ ð100%Þ 1  ½Srretentate =½Srpermeate

2.6.

(2)

Speciation calculations

Distributions of U(VI) species were calculated as a function of pH using Wolfram Mathematica 6.0. The U(VI) species included in speciation calculations are listed in Table 1. All calculations suppressed precipitate formation. Activity coefficients were calculated using the Debye–Huckel Extended  Law at 25 C. pffiffi I pffiffi log10 gj ¼ z2j ð0:5091Þ 1 þ Baj I

(3)

In Equation (3), the activity coefficient gj of ion j with charge zj is obtained for a solution of molar ionic strength I. As proposed by Scatchard (1976), the term Baj was assigned a value of 1.5 for all activity coefficient calculations. The thermodynamic data associated with the formation of these species were taken from the review by Grenthe et al. (2004). Because several U(VI)-complexes contain more than one mole of U(VI) per mole of complex, speciation is a function of total U(VI) concentration. Calculations were performed using total U(VI) concentrations of between 5 mM and 100 mM in order to best match the solution conditions associated with the dialysis studies. While this concentration range is greater than typical contaminated surface and ground waters, higher concentrations were needed to effectively quantify the extent of U(VI) separations using the arsenazo III method. Additionally, groundwater samples collected in April 1997 from Utah’s Fry Canyon site contained U(VI) at levels as high as 68.5 mM (Naftz et al., 2004).

3.

Results and discussion

3.1.

Uranium speciation in aqueous solution

In aqueous solutions containing carbonate the speciation of U(VI) consists of a plethora of hydroxide and carbonate complexes. Table 1 provides formation equilibria and corresponding formation constant values for several

4756

water research 43 (2009) 4751–4759

Table 8 – Predominance of U(VI) species as function of CO2L concentration at [U(VI)]total [ 5 mM and pH [ 11. 3

Table 9 – Predominance of U(VI) species as function of pH at [U(VI)]total [ 0.1mM and [CO2L 3 ] [ 5 mM.

[CO2 3 ]total (mM)

pH

0.10 0.20 0.25 0.50 0.67 1.00 2.00 2.50 5.00

UO2 (OH)2 (%)

UO2 (OH) 3 (%)

UO2 (CO3)2 2 (%)

UO2 (CO3)4 3 (%)

(UO2)2CO3 (OH) 3 (%)

0.66 0.66 0.66 0.65 0.64 0.61 0.41 0.31 0.06

99.00 99.00 99.00 98.14 97.00 92.20 62.20 46.00 9.68

0.00 0.00 0.00 0.01 0.02 0.04 0.10 0.11 0.10

0.01 0.06 0.12 0.92 2.16 6.90 37.00 53.40 90.20

0.00 0.00 0.00 0.01 0.01 0.01 0.01 0.01 0.00

U(VI)-complexes. Speciation calculations were performed using the first ten reactions in Table 1 at [U(VI)]total ¼ 0.01 mM. The results of these calculations are illustrated in Fig. 2 and reveal that the dominant species were UO2þ 2 at pH values less than 5; UO2(OH)2 (aq) at pH values between 5 and 9; and UO2(OH) 3 at pH values between 9 and 11. Speciation calculations were also performed using all species in Table 1 at [U(VI)]total ¼ 0.1 mM and [CO2 3 ]total ¼ 5 mM. Fig. 3 illustrates the results of these calculations and reveals that the dominant species were UO2þ 2 at pH values less than 5; UO2CO3 (aq) at pH values between 5 and 6;  UO2CO3, UO2(CO3)2 2 , and (UO2)2CO3(OH)3 at a pH of 6; 2 UO2(CO3)2 at pH values between 6 and 8; and UO2(CO3)4 3 at pH values between 8 and 11. The PEUF process is most effective when solution conditions provide near quantitative partitioning of target ions in polyelectrolyte pseudophases. The optimal conditions for ILM-PEUF based separation of U(VI) from other aqueous solution components are expected to be those in which dominates the speciation because of its strong UO2(CO3)4 3 electrostatic attraction to cationic polyelectrolytes. Such conditions prevail in a pH range between 8 and 11 because the  activity of CO2 3 is sufficiently high while OH activity remains low enough to prevent significant formation of U(VI)-hydroxide 4 complexes like UO2(OH) 3 . A maximum in UO2(CO3)3 predominance exists between pH values of 9.5 and 10. The proportion of U(VI) in the form of UO2(CO3)4 3 at a given pH is enhanced by increasing the total concentration of carbonate. Tables 2–8 illustrate the relationship between speciation and total carbonate concentration at a given pH in the range pH ¼ 8 to pH ¼ 11. For effective U(VI) separations using ILM-PEUF, total carbonate concentrations must be high enough to allow predominance of UO2(CO3)4 3 in the speciation, yet low enough to prevent significant carbonate-polyelectrolyte interactions. Electrostatic attraction of free CO2 3 to the polyelectrolyte diminishes the effective polymeric charge thus reducing the partitioning of U(VI)-complexes in the colloidal pseudophase. For solution conditions in which [U(VI)]total ¼ 5 mM, pH values are between 8.0 and 9.0, and [CO2 3 ]total < 1 mM, appreciable percentages of U(VI) exist as UO2(OH)2 and UO2(OH) 3 . These complexes, being neutral and monovalent, are not expected to show significant polyelectrolyte partitioning. As pH values increase in the range from 8.0 to 11.0, the proportion of U(VI) in the form of UO2(OH) 3 at a given carbonate concentration also increases. At pH values between 9.5 and 10.5 and total carbonate

UO2 UO2 UO2 (UO2)2CO3 UO2 2 4  (OH)2 (%) (OH) 3 (%) (CO3)2 (%) (CO3)3 (%) (OH)3 (%)

7.0 7.5 8.0 8.5 9.0 9.5

0.69 0.32 0.12 0.04 0.02 <0.01

0.01 0.02 0.02 0.02 0.02 0.04

52.37 29.27 11.72 4.07 1.39 0.51

34.93 67.63 87.74 95.81 98.56 99.44

10.92 2.59 0.39 0.05 <0.01 <0.01

concentrations between 2.5 mM and 5 mM, more than 99% of U(VI) is in the form of UO2(CO3)4 3 . Table 9 provides speciation results using parameters similar to the experimental solution conditions described in the Equilibrium Dialysis sections that follow. For solutions at pH ¼ 9.5, with [[U(VI)]total ¼ 0.1 mM and with [CO2 3 ]total ¼ 5 mM, monovalent and neutral U(VI)complexes account for less than 0.05% of the speciation.

3.2.

Equilibrium dialysis-effect of pH

The effect of solution pH, in the range 7 - 11, on U(VI) retention was investigated through ED. Initial retentate PDADMAC, total carbonate, and total uranium concentrations were 50, 5, and 0.1 mM respectively. As Table 10 indicates, at pH values between 8 and 11 permeate U(VI) concentrations were below the 0.355 mM limit of detection. For pH values of 6.99 and 7.47, permeate concentrations of U(VI) were 2.01 and 0.506 mM respectively. Lower retentions in that pH range are a consequence of increased levels of monovalent, divalent and 2 neutral U(VI) complexes, specifically UO2(OH) 3 , UO2(CO3) ,  and (UO2)2CO3(OH)3 , in the speciation. Table 9 provides the results of speciation calculations performed using approximate ED solution conditions.

3.3.

Equilibrium dialysis-effect of competing ions

The effects of competing-ion concentrations were investigated using added NaCl. A chloride salt was selected for use because aqueous U(VI)-chloride complexes are very weak with log bo

Table 10 – Effect of pH on U(VI) retention at [CO2L 3 ]total [ 5 mM. pH

[U(VI)] (mM)

permeate

6.99 7.47 8.07 8.48 9.01 9.53 9.93 10.50 11.08

2.01 .506 BLD BLD BLD BLD BLD BLD BLD

[U(VI)]initial ¼ 0.100 mM; [PDADMAC] ¼ 50 mM. BLD ¼ below limit of detection (<0.355 mM).

U(VI) retention 98.0 99.5 >99.6% >99.6% >99.6% >99.6% >99.6% >99.6% >99.6%

4757

water research 43 (2009) 4751–4759

Table 11 – Effect of competing-ion concentration on U(VI) retention at pH [ 9.5 and [CO2L 3 ]total [ 5 mM. [NaCl] (mM)

[U(VI)]permeate (mM)

U(VI) retention

BLD BLD BLD BLD BLD 0.602

>99.6% >99.6% >99.6% >99.6% >99.6% 99.4%

1.02 5.01 10.0 20.0 50.0 100.

[U(VI)]initial ¼ 0.100 mM; [PDADMAC] ¼ 50 mM. BLD ¼ below limit of detection (<0.355 mM>).

values of 0.17 and 1.1 for UO2Clþ and UO2Cl2 respectively (Grenthe et al., 2004). Solution pH values were adjusted to 9.5 and initial retentate PDADMAC, total carbonate, and total uranium concentrations were 50, 5, and 0.1 mM respectively. Sodium chloride concentrations were varied from 1 to 100 mM.  In addition to Cl, other anions, specifically free CO2 3 and HCO3 , also compete with U(VI)-complexes for PDADMC binding sites. Calculations reveal that the concentrations of free CO2 3 and 4 HCO 3 are 0.97 and 3.8 mM respectively and that UO2(CO3)3 accounts for well over 99% of the U(VI) speciation under these solution conditions. As Table 11 indicates, permeate concentrations of U(VI) were below the limit of detection at NaCl concentrations less than that of added polyelectrolyte. These results illustrate the strong affinity of UO2(CO3)4 3 for PDADMAC.

3.4.

Equilibrium dialysis-expulsion of strontium

Ion-expulsion ultrafiltration (IEUF) utilizes charged colloidal pseudophases to separate like-charged ions from other solute species. Using PDADMAC, cations, particularly multivalent cations like Sr2þ, are expelled from retentate to permeate as a result of Donnan exclusion interactions. To demonstrate the effectiveness of ILM-CEUF in separating U(VI) from Sr, dialysis experiments were performed at pH ¼ 8.0. Because SrCO3 has a Ksp value of only 9  1010, a lower solution pH was employed to reduce the level of free carbonate in solution (Brown et al., 2009). At pH ¼ 8.0, no precipitate was observed in solutions containing 0.05 M PDADMAC, 0.1 mM total U(VI), 1.0 mM Sr, and 5.0 mM total carbonate. The results of these dialysis experiments are provided in Table 12. The fact that the permeate concentration of Sr at times exceeds that initially in the retentate results from the

transfer of water from permeate to retentate because of osmotic pressure effects. The expulsion of Sr2þ by 0.05 M PDADMAC becomes greater as the concentration of Sr2þ decreases. This trend is expected and has been observed and explained previously (Christian et al., 1989). It is worth noting that during remediation of typical ground waters, other divalent cations (e.g. Ca2þ and Mg2þ) are likely present at levels far exceeding that of the Sr2þ in this study. Although previous studies have illustrated that Ca2þ in solution is effectively expelled from retentate to permeate, as Sr2þ was here, it must of course first be in solution (Tuncay et al., 1994a). Although both CaCO3 and MgCO3 are more soluble than SrCO3, at the carbonate concentrations used in this study, precipitation would likely occur in ground waters with high levels of Ca2þ and Mg2þ.

4.

Conclusions

Uranium (VI) removal from aqueous solution can be achieved using PDADMAC and carbonate in an ILM-PEUF process. demonstrates extensive partitioning Tetravalent UO2(CO3)4 3 in PDADMAC psuedophases even in the presence of large excesses of competing ions like carbonate and chloride. The highest U(VI) retention values were obtained at pH values for which UO2(CO3)4 3 dominates the speciation. Dialysis experiments also demonstrated the potential use of ILM-CEUF for separating U(VI) from Sr in aqueous solution. Future studies will compare U(VI) removal efficiencies obtained using PDADMAC to those obtained using other cationic colloidal systems, both polyelectrolyte- and surfactant-based. Other systems may provide means whereby colloid can be recovered for reuse. Lack of an effective colloid regeneration scheme is a major obstacle to the economic viability of CEUF processes. Although, as it relates to U(VI) and Sr separation, PDADMAC ‘recovery’ wouldn’t necessarily be required since Sr could be expelled to permeate at higher pH, is bound to PDADMAC, and then the pH when UO2(CO3)4 3 lowered to the point that UO2þ 2 dominates the U(VI) speciation. At his lower pH, U(VI) could be expelled to the permeate, leaving a retentate of PDADMAC for reuse. In addition, more precise analytical methods will be employed in the future to better quantify permeate U(VI) concentrations. Through speciation calculations and experiment, solution parameters such as colloid concentration, carbonate concentration, hardness ion concentration, and pH can be adjusted to

Table 12 – Separation of U(VI) from Sr at pH [ 8.0 and [CO2L 3 ]total [ 5 mM. [Sr]initial (mM) 0.497 0.750 0.998 1.50 2.02

[Sr]retentate (mM)

[Sr]permeate (mM)

[U(VI)]permeate (mM)

Sr expulsion

U(VI) retention

0.0399 0.0769 0.102 0.170 0.269

0.512 0.733 0.939 1.42 1.91

BLD BLD BLD BLD BLD

92.2 89.5 89.1 88.0 85.9

>99.6% >99.6% >99.6% >99.6% >99.6%

[U(VI)]initial ¼ 0.100 mM; [PDADMAC] ¼ 50 mM. BLD ¼ below limit of detection (<0.355 mM).

4758

water research 43 (2009) 4751–4759

maximize U(VI) separation efficiencies. The ILM-CEUF process holds great promise as a viable alternative to traditional U(VI) separation methods.

references

Ahearne, J., Budnitz, R.J., Chu, M., Crowley, K., Meserve, R., Hagengruber, R., Moniz, E.J., Slakey, F., (2007) Consolidated interim storage of commercial spent nuclear fuel: a technical and programmatic assessment. Discussion paper of the American Physical Society, February APS Panel on Public Affairs. Bostick, W.D., Jarabek, R.J., Bostick, D.A., Conca, J., 1999. Phosphate induced metal stabilization: use of apatite and bone char for the removal of soluble radionuclides in authentic and simulated DOE groundwaters. Advances in Environmental Research 3 (4), 488–498. Brown, T.L., LeMay, H.E., Bursten, B.E., Murphy, C.J., 2009. Chemistry the Central Science, 11th ed. Pearson Education, Inc., Upper Saddle River, New Jersey, p. 1116. Christian, S.D., Bhat, S.N., Tucker, E.E., Scamehorn, J.F., El-Sayed, D.A., 1988. Micellar-enhanced ultrafiltration of chromate anion from aqueous streams. AIChE Journal 34 (2), 189–194. Christian, S.D., Tucker, E.E., Scamehorn, J.F., Lee, B., Sasaki, K.J., 1989. Ion-expulsion ultrafiltration–-a new method for purifying aqueous streams. Langmuir 5, 876–879. Christian, S.D., Tucker, E.E., Scamehorn, J.F., Uchiyama, H., 1994. On the interpretation of solubilization results obtained from semiequilibrium dialysis experiments. Colloid and Polymer Science 272, 745–754. Christian, S.D., Tucker, E.E., Scamehorn, J.F., 1995. Colloidenhanced ultrafiltration in remediating groundwater. Specialty Chemical 15 (3), 148–151. Coleman, C.R., Coronado, P.R., Maxwell, R.S., Reynolds, J.G., 2003. Granulated activated carbon modified with hydrophobic silica aerogel-potential composite materials for the removal of uranium from aqueous solutions. Environmental Science and Technology 37 (10), 2286–2290. Dunaway, C.S., Christian, S.D., Tucker, E.E., Scamehorn, J.F., 1998. Study of the binding of anions by a cationic polyelectrolyte using equilibrium dialysis. 1. Chromate anions. Langmuir 14, 1002–1012. Favre-Re´guillon, A., Lebuzit, G., Foos, J., Guy, A., Sorin, A., Lemaire, M., Draye, M., 2005. Selective rejection of dissolved uranium carbonate from seawater using cross-flow filtration technology. Separation Science and Technology 40, 623–631. Feltcorn, E., Breeden, R., 1997. Reactive barriers for uranium removal. U.S. Environmental Protection Agency Ground Water Currents Issue No 26, 1–2. Gafvert, T., Ellmark, C., Holm, E., 2002. Removal of radionuclides at a waterworks. Journal of Environmental Radioactivity 63, 105–115. Gallo, D., Acosta, E.J., Scamehorn, J.F., Sabatini, D.A., 2006. Pilot scale study of polyelectrolyte enhanced ultrafiltration for arsenic removal. Journal of American Water Works Association 98 (1), 106–116. Gerente, C., Andres, Y., Le Cloirec, P., 1999. Uranium removal onto chitosan: competition with organic substances. Environmental Technology 20, 515–521. Grenthe, I., Fuger, J., Konings, R.J.M., Lemire, R.J., Muller, A.B., Nguyen-Trung, C., Wanner, H., 2004. In: Wanner, H., Forest, I. (Eds.), Chemical thermodynamics of uranium. North Holland Elsevier, Amsterdam. Huq, H.P., Yang, J.S., Yang, J.W., 2007. Removal of perchlorate from groundwater by the polyelectrolyte-enhanced ultrafiltration process. Desalination 204, 335–343.

Klepac, J., Simons, D.L., Tyler, R.W., Scamehorn, J.F., Christian, S.D., 1991. Use of ligand-modified micellar-enhanced ultrafiltration in the selective removal of metal ions from water. Separation Science and Technology 26 (2), 165–173. Kolarik, Z., 1981. In: Keim, R. (Ed.), Solvent extraction of uranium, eighth ed. Springer, Berlin, Heidelberg. Kratz, S., Schnug, E., 2006. Rock phosphates and P fertilizers as sources of U contamination in agricultural soils. In: Merkel, B.J., Hasche-Berger, A. (Eds.), Uranium in the Environment. Springer, Berlin, Heidelberg, pp. 57–68. Krehbiel, D.K., Scamehorn, J.F., Ritter, R., Christian, S.D., Tucker, E.E., 1992. Ion-expulsion ultrafiltration to remove chromate from wastewater. Separation Science and Technology 27 (13), 1775–1787. Kressin, I.K., 1984. Spectroscopic method for the determination of uranium in urine. Analytical Chemistry 56, 2269–2271. Naftz, D.L., Fuller, C.C., Davis, J.A., Piana, M.J., Morrison, S.J., Freethey, G.W., Rowland, R.C., 2004. Field demonstrations of permeable reactive barriers to control uranium contamination in groundwater. In: Wickramanayake, G.B., Gavaskar, A.R., Chen, S.C. (Eds.), Remediation of Chlorinated and Recalcitrant Compounds, vol. 6. Battelle Press, Columbus, Ohio, pp. 281–290. Pookrod, P., Haller, K.J., Scamehorn, J.F., 2004. Removal of arsenic anions from water using polyelectrolyte-enhanced ultrafiltration. Separation Science and Technology 39, 811–831. Pramauro, E., Prevot, A.B., Pelizzetti, E., Marchelli, R., Dossena, A., Biancardi, A., 1992. Quantitative removal of uranyl ions from aqueous solutions using micellar-enhanced ultrafiltration. Analytical Chimica Acta 264, 303–310. Pramauro, E., Prevot, A.B., Zelano, V., Gulmini, M., Viscardi, G., 1996. Selective recovery of uranium (VI) from aqueous acid solutions using micellar ultrafiltration. The Analyst 121, 1401–1405. Raff, O., Wilken, R.D., 1999. Removal of dissolved uranium by nanofiltration. Desalination 122, 147–150. Roach, J.D., Tush, D., 2008. Equilibrium dialysis and ultrafiltration investigations of perchlorate removal from aqueous solution using poly(diallyldimethylammonium) chloride. Water Research 42, 1204–1210. Roach, J.D., Christian, S.D., Scamehorn, J.F., Taylor, R.W., Tucker, E.E., 2003. Ligand modified-colloid enhanced ultrafiltration. Use of nitrilotriacetic acid derivatives For the selective removal of lead from aqueous solution. Separation Science and Technology 38 (9), 1925–1947. Roberts, B.L., Scamehorn, J.F., Christian, S.D., 2000. Micelle enhanced ultrafiltration in a spiral wound ultrafiltration module and comparison with stirred cell performance. ACS Symposium Series 740, 158–174. Sasaki, K.J., Burnett, S.L., Christian, S.D., Tucker, E.E., Scamehorn, J.F., 1989. Polyelectrolyte ultrafiltration of multivalent ions removal of copper (2þ) by sodium poly(styrenesulfonate). Langmuir 5, 363–369. Savvin, S.B., 1961. Analytical use of arsenazo III. Talanta 8, 673–685. Scamehorn, J.F., Ellington, R.T., Christian, S.D., Penney, B.W., Dunn, R.O., Bhat, S.N., 1986. Removal of multivalent metal cations from water using micellar-enhanced ultrafiltration. In: Li, N.N. (Ed.), Recent Advances in Separation Techniques-III. AICHE Symposium Series, vol. 250, p. 48. Scatchard, G., 1976. Equilibrium in Solution: Surface and Colloid Chemistry. Harvard University Press, Cambridge, Massachusetts, pp. 306. Sorg, T.J., 1988. Methods for removing uranium from drinking water. Journal of the American Water Works Association 80 (7), 105–111. Sriratana, S., Scamehorn, J.F., Chavadej, S., Saiwan, C., Haller, K.J., Christian, S.D., Tucker, E.E., 1996. Use of polyelectrolyte-enhanced ultrafiltration to remove chromate from water. Separation Science Technology 31 (18), 2493–2504.

water research 43 (2009) 4751–4759

Tabatabai, A., Scamehorn, J.F., Christian, S.D., 1995. Water softening using polyelectrolyte-enhanced ultrafiltration. Separation Science and Technology 30 (2), 211–224. Tondre, C., 2000. Surfactant-based colloidal particles as the extracting phase for the removal of metal ions from aqueous environments: kinetics and applied aspects. ACS Symposium Series 740, 139–157. Tsuruta, T., 2007. Removal and recovery of uranium using microorganisms isolated from North American uranium deposits. American Journal of Environmental Sciences 3 (2), 60–66. Tuncay, M., Christian, S.D., Tucker, E.E., Taylor, R.W., Scamehorn, J.F., 1994a. Ligand-modified polyelectrolyteenhanced ultrafiltration with electrostatic attachment of ligands 1. Removal of Cu(II) and Pb(II) with expulsion of Ca(II). Langmuir 10, 4688–4692.

4759

Tuncay, M., Christian, S.D., Tucker, E.E., Taylor, R.W., Scamehorn, J.F., 1994b. Ligand-modified polyelectrolyteenhanced ultrafiltration with electrostatic attachment of ligands. 2. Use of diethylenetriaminepentaacetic acid/cationic polyelectrolyte mixtures to remove both cations and anions from aqueous streams. Langmuir 10, 4693–4697. United States Environmental Protection Agency (2000)National Primary Drinking Water Regulations; Radionuclides; Final Rule, 65 Fed. Reg. 76708–76753. Yoon, J., Yoon, Y., Amy, G., Cho, J., Foss, D., Kim, T., 2003. Use of surfactant modified ultrafiltration for perchlorate (ClO 4) removal. Water Research 37, 2001–2012. Zhang, Z., Clifford, D., 1994. Exhausting and regenerating resins for uranium removal. Journal of the American Water Works Association 86 (4), 228–241.