Studies on the separation of dissolved uranium from alkaline carbonate leach slurries by resin-in-pulp process

Separation and Purification Technology 112 (2013) 54–60

Contents lists available at SciVerse ScienceDirect

Separation and Purification Technology journal homepage: www.elsevier.com/locate/seppur

Studies on the separation of dissolved uranium from alkaline carbonate leach slurries by resin-in-pulp process T. Sreenivas ⇑, K.C. Rajan Mineral Processing Division, Bhabha Atomic Research Centre, AMD Complex, Begumpet, Hyderabad 500 016, India

a r t i c l e

i n f o

Article history: Received 27 November 2012 Received in revised form 21 February 2013 Accepted 27 March 2013 Available online 6 April 2013 Keywords: Uranium Ion-exchange Resin-in-pulp Alkaline leaching

a b s t r a c t The increasing demand for uranium fuel for energy generation has triggered interest in the exploitation of various resources including lean tenor ores and tailing dumps which are otherwise considered uneconomical till recently. The process-schemes of new plants are incorporating several new technologies such that the process is economically attractive and eco-friendly. One such development catching the limelight in uranium ore processing industry everywhere is the resin-in-pulp (RIP) technology. This paper describes the results of laboratory studies on the recovery of dissolved uranium values from an alkaline leach slurry of a medium-grade uranium ore from Gogi (Karnataka, India) using the resin-in-pulp (RIP) process. Besides anionic carbonate complex of uranium – [UO2(CO3)3]4, the other major anionic constit3 2 3  , SO2 uents of the leach solution are CO2 3 , HCO 4 , Cl , PO4 and MoO4 . The total dissolved solutes (TDSs) are about 45 g/l. Various commercially available strong base anionic type resin-in-pulp (RIP) grade resins – both macro-porous and gel type, were studied with respect to their loading capacity. The gel type polystyrene based resins grafted with quaternary ammonium ion gave superior loading capacity in comparison to macro-porous resins. Parametric variation studies for optimizing the other process conditions, including adsorption kinetics, were then carried out on the short-listed resin. Results of the semi-continuous counter-current extraction and elution tests indicated that about 98% of the dissolved uranium values can be recovered during the loading process and practically the entire loaded uranium can be eluted using NaCl eluant. Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction The recognition of nuclear power as an energy source of minimal carbon foot-print led to increased demand for uranium metal in recent times [1]. This has resulted in revival of interest in exploitation of various uranium resources including low-grade ores (U3O8 < 0.1%) and uranium bearing tailing dumps, which were otherwise considered un-economical for extraction till recently [2,3]. The new uranium mills are embracing latest developments in the ore extraction technology such that the processing schemes are both economically attractive and environmentally benign [4,5]. The developments catching the attention also include the application of resin-in-pulp technology [6–8]. In the RIP process, the resin beads, generally coarse in size, are introduced directly into the leach slurry and equilibrated till the beads reach saturation loading with the desired ions in solution. The loaded resin is subsequently separated from the leach slurry using screens and the ions loaded on the resin are eluted in a normal static bed ion exchange column. Though the RIP technology is very old its use was discontinued, particularly in the uranium industry mainly due to high rate of re⇑ Corresponding author. Tel.: +91 4027763840; fax: +91 4027762940. E-mail address: [email protected] (T. Sreenivas). 1383-5866/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.seppur.2013.03.050

sin losses [9,10]. The advent of mechanically resilient resins with higher loading capacity and simultaneous development of specialized resin transfer pumps is responsible for the resurgence of the RIP technology in uranium industry [11–17]. Several successful pilot-plant trials were reported on uranium ores of South Africa, Canada, Niger and Australia [10,12,15,18,19]. India plans to meet about 30% of the overall energy requirements through nuclear power by 2032 which is close to 3% at present [20]. Accordingly, intense exploration and mining programs are in progress for harnessing the available indigenous uranium resources. The entire uranium resources discovered so far are low-grade and low- to medium-tonnage variety only [21–24]. About 45% of these resources are present in dolostone and limestone host rocks [22,23]. These deposits namely, the Tummalapalle in Andhra Pradesh and Gogi in Karnataka of southern India are of considerable significance to India as they are likely to emerge as main work-horses for future uranium requirements of nuclear power reactors [21–24]. The uranium resources present in the carbonate host rocks require alkaline processing technology for the recovery of uranium values [25]. The nature of uranium mineralization in both the Tummalapalle and Gogi deposits are very-fine grained and highly disseminated in the host matrix necessitating fine grinding for

55

T. Sreenivas, K.C. Rajan / Separation and Purification Technology 112 (2013) 54–60

100

61

60 40

8.5

6.7

1.8

-100 +150

-150 +210

21

20

-74 +100

Weight %

80

1 -210 +417

-37 +74

<37

0

Size in micrometres Fig. 1. Granulometry of leach residue: weight distribution in each size-class.

species is [UO2(CO3)3]4. NaHCO3 reagent is also added to alkaline leaching circuit to neutralize the NaOH formed as a reaction product of Na2CO3 with UO3 (Eqs. (3) and (4)). The gangue minerals present in the uranium ores, mainly sulfides, too react with the carbonate leachants resulting in formation of sulfates and NaHCO3 (Eqs. (5)–(7)). However, the siliceous and other minerals in the ore react only under aggressive leaching conditions viz. at high temperature and pressure. The uranium values dissolved during leaching of ore are separated out from the other anions by ion exchange or solvent extraction using strong base anion exchanger as depicted in Eq. (8). Oxidation of U4+ to U6+.

2UO2 þ O2 ! 2UO3 Dissolution and complexation of U

adequate exposure for hydrometallurgical extraction [26]. The fine size range of the particulate matter and high viscosity of the alkaline leach solution made the filtration of the leach slurry an arduous task. Studies on filtration of leach slurry on horizontal vacuum belt filter showed need for high dosage of chemical flocculants and filtration under hot conditions (55–60 °C) (Fig. 1) for the separation of pregnant leach solution from the leach slurry [25]. Even after adopting these conditions the rate of filtration was about 500 kg/h m2 only. Hence, there is a strong necessity to look into the alternatives to the cost-intensive filtration and clarification process. This paper describes the results of the applied research on the recovery of dissolved uranium values from the alkaline leach slurry of an ore using the RIP process, which is considered as a suitable alternative to conventional process of ‘filtration – clarification – static bed ion exchange’ route. Though several studies were reported on the application of RIP for capture of uranium from sulfuric acid leach slurries [7,10,11,13–19] published information on carbonate based slurries is very scanty [8,24,27]. Detailed investigations on the applicability of RIP process with new generation RIP-grade resins for uranium bearing carbonate slurries are very much imperative for uranium ore processing in general and Indian in particular.

ð1Þ 6+

with carbonate ions.

2UO3 þ 2Na2 CO3 þ 4NaHCO3 ! 2Na4 UO2 ðCO3 Þ3 þ 2H2 O

ð2Þ

Prevention of re-precipitation of dissolved uranium upon reaction with sodium bicarbonate.

UO3 þ 3Na2 CO3 þ H2 O ! Na4 UO2 ðCO3 Þ3 þ 2NaOH

ð3Þ

NaHCO3 þ NaOH ! Na2 CO3 þ H2 O

ð4Þ

Dissolution of gangue minerals like pyrite, quartz and alumina.

2FeS2 þ 7O2 þ 8Na2 CO3 þ 6H2 O ! 2FeðOHÞ2 þ 4Na2 SO4 þ 8NaHCO3

ð5Þ

SiO2 þ 2Na2 CO3 þ H2 O ! Na2 SiO3 þ 2NaHCO3

ð6Þ

Al2 O3  3H2 O þ 2Na2 CO3 ! 2NaAlO3 þ 2NaHCO3 þ 2H2 O

ð7Þ

The ion-exchange process using strong base anion exchanger. 

4R4 NCl þ ½UO2 ðCO3 Þ3 4 $ ðR4 NÞ4 UO2 ðCO3 Þ3 þ 4Cl

ð8Þ

3. Materials and methods

2. Process chemistry

3.1. Leach slurry

Recovery of uranium values hosted in carbonate gangue of an ore using alkaline process generally involve a series of hydrometallurgical operations comprising of comminution for adequate liberation of the desired mineral phase, oxidative dissolution of uranium minerals under atmospheric or elevated temperature and pressure with leachants, filtration and clarification of the uranium laden leach solution, purification and concentration of the uranium values by ion exchange followed by precipitation of dissolved uranium [25]. Few variants in the process scheme also exist, particularly with ores of high tenor [8]. In such cases the leach solution goes for product precipitation directly. Similarly if RIP process is adopted, the stages of filtration of leach slurry and clarification of leach liquor prior to ion exchange operation can be eliminated. The process chemistry of alkaline leaching of uranium ores with Na2CO3 and NaHCO3 reagent combine, which are the common leachants used industrially, under oxidizing conditions is explained in the chemical reactions shown in Eqs. (1)–(7). It essentially involves oxidation of the insoluble U4+, which is the common oxidation state of uranium in majority of economically significant minerals, to soluble U6+ (Eq. (1)). The U6+ complexes with CO2 present in 3 the aqueous phase to form uranyl carbonate ions (Eq. (2)). Depending on the pH of the leach slurry the ionic charge of the uranyl carbonate complex varies. While neutral complexes are predominant at pH 5–6.5 the divalent species – [UO2(CO3)3]2 is present in the between pH 6.5 and 7.6. In the pH range of 7.6–12 the dominant

The leach slurry used in the experimental work was obtained by subjecting a medium grade uranium ore (U3O8 0.18%) from Gogi (Karnataka, India) to atmospheric alkaline leaching. Sodium carbonate and sodium bicarbonate were used as lixiviants. The mineralogical composition of the ore and the partial chemical composition of the leach slurry obtained under optimized conditions for maximum leachability of uranium are given in Tables

Table 1a Mineralogical composition of Gogi uranium ore. Mineral Calcite Quartz + chert Feldspars Micaceous minerals (chlorite, biotite and clay) Ferromagnesian minerals (mainly hornblende with minor epidote) Barite Zircon Oxides (magnetite, hematite and goethite) Sulfides (pyrite, marcasite, chalcopyrite and traces of galena) Radioactive minerals (coffinite, pitchblende and adsorbed uranium in association with carbonaceous matter and goethite) Others (determined by difference) Total

Weight (%) 61.6 13.1 1.6 5.9 0.4 0.6 0.08 0.45 6.1 0.9

9.3 100.00

56

T. Sreenivas, K.C. Rajan / Separation and Purification Technology 112 (2013) 54–60

Table 1b Partial chemical composition of leach solution generated during alkaline leaching of Gogi uranium ore. Analyte

Concentration (g/l)

U3O8 Na2CO3 NaHCO3 Na2SO4 [PO4]3 [Cl]1

0.734 36.5 14.5 2.0 0.003 0.7 0.001

MoO2 4 Total dissolved solute (TDS) pH

The gel type resins investigated include PFA460/4783 & PFA600/ 4740 (Purolite, France), Ambersep(™) 400 SO4 (Rohm and Haas, UK) and Indion ARU 103 (Ion Exchange India) while the macro-porous resins studied are 920USO4, 920UHCSO4, 920UCl, Amberlite IRA910UCl (Rohm and Haas, UK) and A500/2788 (Purolite, France). The chemical and particulate characteristics of the resins are given in Tables 2a and 2b [28–30]. 3.3. Other reagents All the other chemicals used in the test work are of A.R. Grade. Only de-mineralized (D.M.) water was used in the tests.

46.5 9.5

3.4. Resin in solution/pulp test work

U3O8 g/l of wet settled resin

A

100

The resins obtained from various manufacturers were used in the experiments without any chemical pre-treatment except for soaking the beads in de-mineralized water overnight at insitu pH. Known volume of the wet settled resin (w.s.r.) was taken-out from the stock regularly and added to the leach solution or pulp (solids concentration 30% by weight) as the case may be. The experimental studies were carried out by screening various ion-exchange resins for their uranium loading capacity by resin-insolution (RIS) technique. The short-listed resin was then used for elucidating other operating conditions for maximal loading. The optimum conditions arrived from the RIS experiments was used for the resin-in-pulp (RIP) test work. The mixing in ‘resin-in-solution’ tests was carried out by shake-flask technique while overhead stirrer was used for mixing in the ‘resin-in-pulp’ experiments. After the completion of requisite contact time the resin was separated from the solution or the slurry on a nylon screen. The loaded resin was washed with D.M. water to remove adhered ore particles and transferred to an elution column. The elution tests were carried out by downward flow of the eluant at a pre-defined flow-rate in a 15 mm diameter and 400 mm height glass column using NaCl as eluant. The eluant solution was collected periodically and analyzed for uranium and other constituent ions. All the experiments were conducted at ambient temperature, about 25 °C.

80 60 40 20 0 ARU 103

PFA4740

PFA4783

400SO4

Gel Resin 100

U3O8 g/l of wet settled resin

B

80 60 40 20 0 920 USO4 920UHCSO4

920UCl

IRA910U

A500/2788

Macroporous Resin

4. Chemical and data analysis

Fig. 2. Screening of resins based on loading capacity of uranium from alkaline leach solution. (A) Gel type and (B) macro-porous.

1a and 1b. The size-wise distribution of solids in the leach slurry is given in Fig. 1. 3.2. Resin Commercially available strong base anion exchange resins of both macro-porous and gel type were chosen for their suitability for recovery of uranium from the specific leach slurry under study.

The metallurgical accounting in all the experiments is based on wet chemical analysis of uranium and other species in various streams. The uranium concentration in the aqueous phase was determined by liquid fluorimetry (ELICO Model SL 174). The relative standard deviation (RSD) in the uranium analysis was ±2%. 1 The CO2 , PO3 and Cl1 concentration was estimated by 3 , HCO 4 2 titrimetry. The SO4 content was analyzed by gravimetric analysis. The loading of uranium on the resin phase is expressed as grams of U3O8 per liter of wet settled resin (w.s.r.) and calculated as given in the following equation:

Table 2a Characteristics of gel type resins used in RIP studies. Name of the resin Manufacturer

PFA460/4783 Purolite, France

PFA600/4740

Ambersep 400 SO4 Rohm and Haas, UK

Indion ARU 103 Ion exchange, India

d50 (lm)

550

600

650

550

q (g/l) (based on dry resin > 600 lm)

675

675

730

680

Theoretical capacity (eq/l)

1.60

1.30

1.4

1.30

Moisture retention capacity (%)

40–45

47–54

40–47

47–52

Maximum reversible swelling Cl to OH (%)

20

20

Matrix

Gel polystyrene crosslink with DVB

Functional group

Quaternary ammonium ion

Gel styrene EDMA copolymer, iso-porous

57

T. Sreenivas, K.C. Rajan / Separation and Purification Technology 112 (2013) 54–60 Table 2b Characteristics of macro-porous type resins used in RIP studies. Name of the resin

Purolite A500/2788

AmbersepTM920 UCl

AmbersepTM920 UHCSO4

AmbersepTM920 USO4

Amberlite IRA910 UCl

Manufacturer

Purolite, France

Rohm and Haas, UK

d50 (lm)

900

850

850

850

700–900

q (g/l) (based on dry resin > 600 lm)

675

700

693

735

700

Theoretical capacity (eq/l)

1.15

1.0

1.0

1.0

1.0

Moisture retention capacity (%)

53–58

48–60

48–60

48–60

54–61

Maximum reversible swelling Cl to OH (%)

15

20

–

5

15

Matrix

Macro-porous polystyrene crosslink with divinyl benezene

Macro-reticular cross-linked with polystyrene

Functional group

Quaternary ammonium ions

60

0.7

20

5.75

40

52.1

5

6

0 1

5.1. Characterization of leach slurry and resins

69.4

1.15

5. Results and discussion

68.7

67.55 61.8

9.7

where CF and CB are concentration of U3O8 in g/l, VR and VL are volume of wet settled resin and feed solution respectively in liters. All the experiments were carried out in triplicate and only mean values are reported. The maximum standard deviation observed between the reported experimental values never exceeded ±3%.

80

21.3

ð9Þ

30.8

CF  CB  VL VR

U3O8 g/l of wet settled resin

Loading Capacity ¼

Dimethly ethanol ammonium chloride

2

3

4

Stage Number

The partial chemical composition of the leach slurry (Table 1b) shows U3O8 concentration of about 0.734 g/l and total dissolved solutes (TDSs) at 47 g/l, contributed mainly by Na2CO3 and NaHCO3, being 36.5 g/l and 14.5 g/l respectively. The other anionic 1 species of interest in the slurry like SO2 are present to the 4 and Cl extent of 2 g/l and 0.7 g/l respectively, while the concentration of 3 MoO2 4 and PO4 is negligible. The granulometry of the leach residue given in Fig. 1 indicates presence of almost 60% by weight of the material in finer than 37 lm and the d80 size is 75 lm. The functional group for majority of both the gel and macro-porous type anion exchange resins is quaternary ammonium ion only (Tables 2a and 2b) with DVB cross-linkage in gel type and polystyrene linkage in macro-porous resin. The diameter of the resin beads in both the gel and macro-porous resins is significantly more than the d80 size of particulate material in the leach slurry.

Fig. 3. Stage-wise loading of uranium on strong base anionic type resin PFA 4783.

5.2. Screening of resins

5.3.1. Loading capacity of resin The loading capacity of uranium on PFA460/4783 resin was determined by equilibration of 1 ml of the resin with 50 ml of leach solution at ambient temperature for 120 min contact time per stage. This was repeated six times by contacting the same resin with fresh leach solution. The differential and cumulative loading capacity of uranium at each stage is shown in Fig. 3. U3O8 loading of about 31 g/l of w.s.r. was achieved in the first cycle itself and it progressively increased to about 68 g/l of wet settled resin in the fourth cycle. Subsequent stages of equilibration did not show any significant additional loading. Though the concentration of U3O8 is relatively lower in the leach solution, about 730 mg/l in compar 2 ison to other competing ions particularly CO2 3 , HCO3 and SO4 , (Table 1b), its affinity for the exchangeable sites on the strong base anion exchange resin is much higher than other competing ions. The trend agrees well with the findings of McGarvey and Ungar [31] which reported the affinity order as [UO2(CO3)3]4 > 2  2  MoO2 > SO2 for a strong 4 > [UO2(CO3)2] 4 > CO3 > Cl > HCO3

Studies on loading capacity of uranium for various commercially available resins listed in Tables 2a and 2b was carried out by equilibrating 1 ml of the respective resin with 100 ml of leach solution for a total period of 12 h spread over six stages of equal time-interval at ambient temperature. At the end of each stage of equilibration the resin was separated out and contacted again with 100 ml of fresh leach solution. The performance was evaluated on the basis of quantity of U3O8 loaded per liter of wet settled resin (w.s.r.) only. Results of the experiments are given in Fig. 2a and b for the gel and macro-porous type resins respectively. The loading of uranium was more on gel-type resins in comparison to that obtained on macro-porous resins. The loading capacity of U3O8 was about 65–70 g/l of wet settled resin (w.s.r.) for almost all the gel type resins tested excepting the ARU 103 resin which was about 55 g U3O8/l of w.s.r. The macro-porous resins gave maximum loading of 40–55 g U3O8/l of w.s.r. The lower loading capac-

ity of macro-porous resins is mainly attributed to the lesser number of functional groups present in the polymeric resin in comparison to the gel type variety for a given unit weight [30]. This is reflected in the theoretical exchange capacity of the resins too (Tables 2a and 2b), which is about 1.3 and 1–1.1 eq/l for gel type and macro-porous type respectively. Since both PFA600/4740 and PFA 460/4783 resins have similar physical properties and uranium loading capacity, one of them, viz. PFA460/4783 was taken for detailed parametric variation study.

5.3. Parametric variation studies with PFA460/4783 for resin-insolution tests

58

T. Sreenivas, K.C. Rajan / Separation and Purification Technology 112 (2013) 54–60

U3O8 g/l wet settled resin

100 80 60 40 20 0 0

60

120

180

240

300

360

Contact time (min) Fig. 4. Kinetics of uranium sorption on strong base anionic resin PFA 4783.

base anion exchange resin in the pH range of 9–10.0. The species distribution diagram of uranium–carbonate system at different pH conditions indicate predominant presence of [UO2(CO3)3]4 at pH higher than 7.5 [32]. As the electrostatic attraction between the ions in solution and the resin is proportional to the ionic charge on the ion in the solution phase, the affinity of [UO2(CO3)3]4 is much higher than any other competing ions inspite of being present at much lower concentration levels [33]. 5.3.2. Effect of contact time The kinetics of uranium loading on PFA 460/4783 resin was determined by equilibrating 2 ml of resin with 500 ml of leach solution at ambient temperature with periodic drawl of solution samples from the reaction vessel for estimating the progress of loading or adsorption. The results are illustrated in Fig. 4. The kinetics of sorption of uranium was significantly fast in the initial phase, up to about 45 min and remained slow thereafter. About 52 g U3O8/l of w.s.r. was loaded in the initial 45 min of time and reached equilibrium in about 120 min where a loading capacity of about 57 g of U3O8 g/l of wet settled resin was achieved. The results of the contact time studies on loading of uranium on PFA 460/4783 resin (Fig. 4) are further analyzed using the standard kinetic models [34,35], like the Lagergren first order equation and pseudo-second order type, to understand the rate controlling step in the sorption process. The linearized integrated form of the first-order rate equation for the boundary conditions t = 0 and t = t is given as:

LnðQ e  Q t Þ ¼ LnðQ e Þ  kt

ð10Þ

where Qe and Qt are the amount of uranium adsorbed (mg/g) on the resin at equilibrium and time ‘t’ respectively, and ‘k’ is the rate constant of first order adsorption (min1). The slope of the straight line

3.5 3

t/Qt

2.5

R2 = 0.9884

2 1.5 1 0.5

graph obtained upon plotting of Ln(Qe  Qt) versus ‘t’ gives the value of rate constant of the sorption process. Similarly, the relation for linearized integrated form of pseudo-second order kinetic model is given in Eq. (11).

t 1 1 ¼ þ t Q t kQ 2e Q e

ð11Þ

A plot of Qtt versus ‘t’ gives a straight line with the value for Q1e and kQ1 2 from the slope and intercept respectively. The units of ‘k’ e in the pseudo-second order model is g mg1 min1. The correlation constant ‘R2’ computed from the trend line drawn for various experimental data points was 0.8 for first order and 0.988 for the pseudo-second order kinetic model (Fig. 5). The theoretically computed equilibrium adsorption capacity in the pseudo-second order case is 56 g of U3O8/l of wet settled resin against the experimentally obtained value of about 57 g of U3O8/l of w.s.r and the rate constant is 0.00138 g mg1 min1. The value of correlation constant clearly indicates that the sorption of uranium species on PFA 460/4783 resin follow the pseudo-second order kinetics better than the first order model implying that the rate-controlling step in the exchange process is the chemisorptive forces between the uranyl carbonate ions in the solution phase and the quaternary ammonium ions on the resin [35]. 5.3.3. Resin to leach solution ratio The optimum quantity of resin required for maximum loading of desired species from the leach solution is an important parameter in the overall process economics. The objective being to achieve maximum loading or recovery with minimum resin inventory. The optimal ratio was elucidated by varying the volume of resin, PFA 4783, for fixed quantity of leach solution, which was studied with a resin volume of 0.5–15 ml keeping the volume of leach solution constant at 100 ml (U3O8 700 mg/l). The contact time in all the single-stage loading experiments was 100 min and the reaction was carried out at ambient temperature. The extent of depletion of uranium in the leach solution in single-stage equilibration at various resins to leach solution volume ratios is shown in Fig. 6. About 60% of the uranium values in the leach solution were adsorbed on the resin, when the resin to leach solution volume ratio was 1:100. The uptake increased to 80% when the ratio was1:50 and the maximum uptake was in the range of 90–95% achieved in the ratio range of 1:33–1:6.5. Results of the single-stage extraction tests indicate resin to leach solution volume ratio of 1:50–1:33 as ideal choice for maximum loading with minimum resin inventory.

0 0

60

120

180

240

300

360

Time (min.) Fig. 5. Pseudo second order kinetic plot of uranium sorption on PFA 4783 resin.

5.3.4. Continuous counter-current resin-in-solution tests The number of theoretical mass-transfer steps necessary for maximum loading of uranyl carbonate anions on the resin was

59

T. Sreenivas, K.C. Rajan / Separation and Purification Technology 112 (2013) 54–60

U3O8 attained maximum level at 4th stage, about 99.3% and remained practically invariant subsequently (Fig. 8) showing good compliance with the McCabe Plot given in Fig. 7.

%U3O8 depletion from leach solution

100 80 60

5.4. Resin-in-pulp extraction with PFA460/4783 for alkaline leach slurries

40 20 0

Volume Ratio of Resin to Leach solution Fig. 6. Sorption of uranium on PFA 4783 resin – effect of variation of volume ratio of resin to leach solution.

80

U3O8 g/l of wet settled resin

70 60 50 40 30 20 10

Based on the extensive ‘‘resin-in-solution’’ test work data generated with PFA460/4783 resin, the necessary experimental conditions for carrying out semi-continuous counter-current ‘‘resin-in-pulp’’ experiments were developed. The optimum values and other conditions developed from the ‘‘resin-in-solution’’ studies are given in Table 3. The alkaline leach slurry contained about 30% by weight of leached solids. The resin-in-pulp tests were carried out at pH of about 9.5 at ambient temperature. Five reaction vessels containing 58 ml of the leach slurry and 1 ml of resin in each one of them were agitated gently using an overhead paddle type stirrer for 20 min at a peripheral speed of 0.75 m/s. The reactor contents were then separated manually using a screen of 35# size (Tyler). The screen over size (resin) and the undersize (pulp) were moved in counter-current directions in carousel mode. After each stage of equilibration a small volume of leach solution was drawn and analyzed for extent of depletion of U3O8 concentration. About 80% of the uranium values were extracted in the first stage and it attained about 96.3% in stage 4 (Fig. 8). Continuing equilibration for two more stages improved the extraction efficiency to 97% only. The system attained steady state after four cycles and the average U3O8 content in the barren solution was about 18 mg/l which implies an extraction efficiency of about 98%. 5.5. Resin-in-pulp studies with PFA460/4783 for alkaline leach slurries – Elution

0 0

100

200

300

400

500

600

700

800

U3O8 mg/l in barren liquor

Elution experiments were carried out in a water-jacketed glass column at ambient temperature. The column was filled with 10 mL

determined from the McCabe Thiele plot constructed using the single-stage equilibrium isotherm generated from the data obtained in Fig. 6. The working line in McCabe Thiele plot (Fig. 7) was obtained by joining the co-ordinates corresponding to 80% loading capacity of the uranium species on the resin and the point that represent the practical U3O8 concentration in the exit stream [15]. The results of the earlier experiments showed maximum loading capacity of 70 g U3O8 per liter of w.s.r. indicating the 80% loading capacity as 56 g of U3O8/l of w.s.r. The practical U3O8 concentration in the exit stream is about 15 mg/l. Using these data points for working line, the McCabe Thiele plot suggests necessity of four stages of equilibration for attaining overall extraction efficiency of about 98%, during the counter-current operation. The counter-current extraction studies were performed experimentally using four contactors (shake flasks) containing 100 ml of leach solution and 2 ml of resin. The reactor contents were kept under agitation for 25 min at ambient temperature in each stage. At the end of this contact period, the reactor contents were discharged manually on a nylon screen and both the fractions were moved in counter current directions. The uranium concentration in the solution depletes after each stage of contact while the resin gets progressively loaded with uranium values from the solution. After completion of all the stages of contacts, the loaded resin was washed and charged into the elution column for the recovery of loaded uranium values. The cumulative extraction efficiency of

% Extraction of U3O8

Fig. 7. Mc Cabe Thiele plot of uranium sorption on PFA 4783 resin in ‘resin-insolution’ experiments.

100 80 60 40 20 0 1

2

3

4

5

6

Stage Number Resin-in-solution

Resin-in-pulp

Fig. 8. Stage-wise loading of uranium on PFA 4783 resin in ‘resin-in-solution’ and ‘resin-in-pulp’ experiments on an alkaline leach slurry.

Table 3 Optimum conditions for RIP process. Resin type Porosity Solids specific gravity pH for adsorption Eh (versus saturated calomel electrode) Capacity of resin for uranium Resin to Leach solution volume ratio Density of leach solution Density of pulp Number of stages of contact Contact time per stage Total contact time

Strong base anionic 50% 2.8 g/cc 9.5–9.8 110–115 mV 70 g/l of w.s.r. 1:50 1.04 g/cc 1.24 g/cc 4–5 20 min 100 min

60

T. Sreenivas, K.C. Rajan / Separation and Purification Technology 112 (2013) 54–60 Table 4 Average chemical composition of elute. Analyte

g/l

U3O8 Na2CO3 NaHCO3 Na2SO4 TDS

8–10 3 25 Traces 70

of the loaded resin. Elution was carried out with 1.7 M NaCl solution. The pH of the eluate solution was 9. The flowrate of eluate was 2.5 BV/h (residence time 10 min). Eluate samples were taken for every 3–5 BV fraction and analyzed for uranium content. The partial chemical compositions of the eluted solution was U3O8 10 g/l, Na2CO3 3 g/l, NaHCO3 25 g/l, TDS 70 g/l and traces of Na2SO4 (Table 4). About 98% of the uranium values loaded on the resin was successfully recovered during elution. 6. Conclusion Resin-in-pulp technique was applied for purification and enrichment of uranium values from a finely ground uranium ore leach slurry of alkaline nature using commercially available strong base anion exchange resin. The chemical composition of the solution phase of the alkaline leach slurry (pH 9.5) contained total dissolved solutes concentration of about 40 g/l of total dissolved solutes (TDSs) composed predominantly with Na2CO3 and NaHCO3 and minor levels of Na2SO4. The uranium content was only 730 mg/l and d50 of solids was 34 lm. Studies have indicated better performance by gel type resins over the macro-porous resins. Amongst the various commercially available resins studied PFA 4740 and 4783 having quaternary ammonium ion on polystyrene crosslink with divinyl benzene (DVB) gave best performance. The maximum loading capacity achieved in the RIP studies was about 70 g of U3O8/l of wet settled resin amounting to 98% of loading. This has necessitated 4 stages of counter-current extraction with overall contact time of 100 min at a resin to leach slurry volume ratio of about 1:50. The single-stage loading tests were kinetically monitored and were found to follow pseudo-second order kinetics. Practically the entire uranium values loaded on the resin were eluted using NaCl. The RIP process was found quite efficient for uranium bearing alkaline leach slurries. Acknowledgments The authors are thankful to Dr. A.K. Suri, Director, Material Group, Bhabha Atomic Research Centre for his interest in the studies and encouragement. They also thank various resin manufacturers for generous supply of resin sample for the test work. References [1] Anon. Judge Nuclear on its Merits, 2012. http://www.iaea.org/OurWork/ST/NE/ judge-nuclear.html. (accessed 02.03.12). [2] M.J. Lottering, L. Lorenzen, N.S. Phala, J.S. Smit, G.A.C. Schalkwyk, Mineralogy and uranium leaching response of low grade South African ores, Minerals Engineering 21 (2008) 16–22. [3] P. Bartsch, Uranium potential supply and exploration outlook, Paper presented at IAEA Technical Meeting on low-grade uranium ores, Vienna, Austria, March 29–31, 2010. [4] A. Taylor, Innovations and trends in uranium ore treatment, Paper presented at ALTA 2007 Uranium Conference, Perth, Australia, 2007. [5] J. Carr, N. Zontov, S. Yamin, Meeting the future challenges of uranium industry, ALTA 2008 Uranium Conference, Perth, Australia, June 15–21, 2008. [6] V.V. Shatalov, D.I. Skorovarov, I.P. Smirnov, Development of advanced technology in the hydrometallurgy of uranium, Atomic Energy 6 (5) (1999) 339–349.

[7] J. Brown, C. Fleming, The re-emergence of resin in pulp with strong base resins as a low cost, technically viable process for the recovery of uranium, Paper # 20, 42nd Annual Meeting of the Canadian Mineral Processors, Ottawa, Ontario, Canada, January 19–21, 2010. [8] D. Van Tonder, Comparison of resin-in-pulp and direct precipitation for recovery of uranium in alkaline leach circuits, CIM 3 (2) (2012) 131–139. [9] Anon, Uranium extraction technology, IAEA Technical Report Series No. 359, IAEA, Vienna, Austria, 1993, pp. 157–226. [10] B. McKevvit, D. Dreisinger, Lessons from the Past: a literature review of resinin-pulp process in uranium industry, in: E.K. Lan, J.W. Rawson, E. Ozberk (Ed.), Proceedings of the 3rd International Conference on Uranium, vol. II, Metsoc, Saskatoon, Canada, 2010, pp. 3–14. [11] K. Mirjalili, M. Roshani, Resin-in-pulp method for uranium recovery from leached pulp of low-grade uranium ore, Hydrometallurgy 85 (2007) 103–109. [12] M. Mikhaylenko, J. Deventer, Notes on practical application of ion exchange resins in uranium extractive metallurgy, ALTA Uranium Conference, Perth, Australia, 2009. [13] V. Yahorava, J. Scheepers, M.H. Koteze, D. Auerswald, Evaluation of various durability tests to assess resins for in-pulp applications, Fifth Southern African Base Metals Conference Kasane, Botswana, The South African Institute of Mining and Metallurgy, July 2009, pp. 341–358. [14] V. Yahorava, J. Scheepers, M.H. Koteze, Comparison of various anion exchange resin for the recovery of uranium by means of RIP, ALTA Uranium Conference, Perth, Australia, 2009. [15] T. Udayar, M.H. Kotze, V. Yahorava, Recovery of uranium from dense slurries via resin-in-pulp, in: 6th Southern African Base Metals Conference 2011, The Southern African Institute of Mining and Metallurgy, 2011, pp. 49–64. [16] M. Kotze, The resurrection of hydrometallurgy’s own RIP van Winkle: Uranium, Presented at Mintek, 75, June 4–5, 2009. [17] T. Thulare, P. Vos, C. Phir, Developments in Resin In Pulp (RIP) for Uranium recovery in Africa, International Uranium Conference ALTA 2008, Perth, Australia, 2008. [18] D. Auerswarld, T. Udayar, M. Kotze, J. Scheepers, Operations of a resin-in-pulp (RIP) demonstration plant for recovering uranium from a South African gold pulp, in: Proceedings, ALTA Uranium Conference, Perth, Australia, 2011. [19] Y. Ling, N. Durupt, N. Banton, RIP studies at AREVA: R & D and applications for Niger and Canada projects, in: E.K. Lan, J.W. Rawson, E. Ozberk (Eds.), Proceedings of the 3rd International Conference on Uranium, vol. II, Metsoc, Saskatoon, Canada, 2010, pp. 27–38. [20] S.K. Jain, Inevitability of nuclear power in the asian region, Journal of Nuclear Power 24 (3–4) (2011) 55–65. [21] S. Banerjee, BARC Newsletter, vol. 32, September–October 2011, pp. 6–9. http://www.dae.gov.indin2010.html. (accessed 28.02.12). [22] R. Gupta, A.K. Sarangi, Overview of indian uranium production scenario in coming decades, Energy Procedia 7 (2011) 146–152. [23] A. Chaki, Uranium occurrences and exploration experience in India, in: E.K. Lan, J.W. Rawson, E. Ozberk (Eds.), Proceedings of the 3rd International Conference on Uranium, vol. I, Metsoc, Saskatoon, Canada, 2010, pp. 25–34. [24] A.K. Suri, T. Sreenivas, Developments in processing of conventional uranium ore resources of India, Paper presented at 60th International Mineral Processing Conference (IMPC), New Delhi, October, 2012. [25] R.C. Merritt, The Extractive Metallurgy of Uranium, Colorado School of Mines, Johnson Publishing Co., Boulder, Colorado, US, 1971. [26] A.K. Suri, N.P.H. Padmanabhan, T. Sreenivas, K. Anand Rao, A.K. Singh, K.T. Shenoy, T. Mishra, S.K. Ghosh, Process development studies for low-grade uranium deposit in alkaline host rocks of Tummalapalle. Presented at IAEA Technical Meeting on Low-grade Uranium Deposits at Vienna, 29–31st March 2010. [27] E. Panturu, Gh. Filip, St. Petrescu, D. Georgescu, F. Aurelian, R. Radulescu. Researches concerning the Purolite assimilation for use within the uranium separation-concentration resin in pulp process, in: Proceedings of the 9th International Conference on Tailings and Mine Waste, Fort Collins, Colorado, US, January 2002. [28] Retrieved from http://www.rohamhass.com. [29] Retrieved from http://www.purolite.com. [30] Retrieved from http://www.ionindia.com.; I.M. Abrams, J.R. Millar, A history of the origin and development of macroporous ion-exchange resins, Reactive and Functional Polymers 35 (1997) 7–22. [31] F.X. McGarvey, J. Ungar, The influence of resin functional group on the ionexchange recovery of uranium, Journal of the South African Institute of Mining and Metallurgy (April) (1981) 93–100. [32] M. Botha, L. Bester, E. Hardwick, Removal of uranium from mine water using ion exchange at Driefontein Mine, in: Proceedings of International Mine Water Conference, Pretoria, South Africa, 19th–23rd October 2009 (ISBN number: 978-0-9802623-5-3). [33] F. Helfferich, Ion Exchange, first ed., Dover Publications, Inc., New York, NY, 1995. [34] A. Rahmati, A. Ghaemi, Md. Samadfam, Kinetic and thermodynamic studies of uranium (VI) adsorption using amberlite IRA 910 resin, Annals of Nuclear Energy 39 (2012) 42–48. [35] G. Zhao, X. Wu, X. Tan, X. Wang, Sorption of heavy metal ions from aqueous solutions: a review, The Open Colloid Science Journal 4 (2011) 19–31.