Recovery of Uranium and Rare Earth Elements from Western Desert Phosphate Rocks with EDTA and Nitric Acid Solutions

Dissolution of uranium and rare earth elements from the Abu-Tartur phosphate rocks, Egypt using a mixed EDTA + nitric acid solution was studied. The effect of factors such as particle size, EDTA and nitric acid concentrations, agitation time, solid to liquid ratio, and dissolution temperature was studied, and the dissolution conditions were optimized. The optimized dissolution conditions were applied to a representative sample of the weathered rock assaying 1067 ppm rare earth elements. Before being processed, the representative sample was completely characterized using mineralogical and chemical analyses. From the obtained leach liquor, uranium was recovered by anion exchange with Duolite A 101D resin, and after uranium adsorption the rare elements were recovered by cation exchange with Dowex 50W-X8 resin.


INTRODUCTION
The Abu-Tartur phosphate deposit (Western desert, Egypt) is one of the largest phosphate deposits in the world. It is geomorphologically localized in the southeastern portion of the Abu-Tartur plateau (600 km from Cairo) and is the largest source of uranium and rare earth elements (REEs) in Egypt [1]. Abu-Tartur plateau has a huge reserve of phosphate ore reaching up to billion tons [2]. Most of the world's phosphate rocks are of sedimentary origin and are primarily composed of the apatite group in association with a wide assortment of some accessory minerals, fl uorides, carbonates, clays, quartz, silicates, and metal oxides [3][4][5]. In general, the phosphate rock (PR) found in the earth's crust contains relatively high levels of environmentally toxic elements as well as precious elements having diverse applications in green and sustainable industrial energy production. Uranium is one of such elements found in PR, usually at levels between 50 and 200 ppm, but some deposits contain signifi cantly higher levels of uranium (>800 ppm) [6]. Uranium contained in PR is dispersed during the use of fertilizers on agricultural soils; it contaminates soil and water and poses hazard to health via food chain [7]. Uranium is an essential element in the atomic energy program. In spite of its low concentration in phosphate ores, its recovery is signifi cant from environmental and economical viewpoints [8]. Along with environmental impact minimization and human health protection, extraction of uranium from PR can reduce mine waste and extend resource use [9]. The previously used procedure for processing of phosphate ores involved hydrochloric acid leaching to obtain monocalcium phosphate (MCP) in the presence of a reductant to keep U(IV) in the insoluble form [10][11][12][13][14]. Recovery of U from Egyptian phosphorites is conducted only on an experimental scale by the Nuclear Materials Authority of the Egyptian government [15]. There are several techniques for recovering uranium from its leach liquors, including ion exchange [16,17].
Phosphate rock associated with trace amount of REEs was considered as one of the potential new resources for REEs due to its huge reservation, about 300 billion tons in total [18,19]. The recovery of REEs from phosphate rock attracted much attention recently [20].
The diff usion-controlled process of REEs recovery is characterized by the activation energy of 11.76 kJ mol -1 [21]. The kinetics of REEs leaching from phosphate ore using HNO 3 were described [22].
On the other hand, polyamino polycarboxylates such as EDTA, TETA, or DOTA form strong complexes with alkaline earth elements, trivalent lanthanides, and actinides in aqueous solution [23,24]. The dissolution of the phosphate rock is favored by the formation of stable Ca(EDTA) 2− complexes in the solution [25,26]. The ion exchange sorption of REE complexes with EDTA on an anion exchanger in the H 2 EDTA 2form can be performed.
This study is aimed at analyzing and optimizing the conditions for uranium and REEs dissolution from Abu-Tartur phosphate rocks using a binary EDTA + nitric acid solution.

Materials and Chemicals
The phosphate samples were obtained from Abu-Tartur (Egypt); samples with diff erent particle size were taken. The Duolite A 101D strongly basic anionexchange resin (Auchtel Products, Mumbai, India) was used in the chloride form. The Dowex 50W-X8 cationexchange resin (Merck) was used in the hydrogen form. Ethylenediaminetetraacetic acid disodium salt (Na 2 EDTA), hydrochloric acid (37%), sulfuric acid (98%), and nitric acid (60%) were purchased from Merck. First, the calcareous Abu-Tartur phosphate sample was completely analyzed for determine major oxides and trace elements by X-ray fl uorescence (XRF) technique with a PW 1510 Philips experimental unit equipped with an automatic sample changer (JEOL-JSX-3222 Element Analyzer; Atomic Energy Authority, Cairo, Egypt). The device was interfaced with a computer system; X-40 program for spectroscopy was used. The detection limit was 5.0 ppm. Atomic spectrophotometry was used for identifi cation and confi rmation of trace elements including uranium and REEs [27]. The results obtained are given in Tables 1  and 2; as can be seen, P 2 O 5 assays only 21.8%, the CaO content reaches 45.6%, and the loss in ignition amounts to 9.2%; U assays only 30 ppm, and the REEs content reaches 1067 ppm.

Experimental Procedures
The working sample was collected from an experimental mine of Abu Tartur plateau located in the Western Desert of Egypt. Five diff erent size fractions (700, 500, 150, 74, and 63 μm) were obtained by sieving through ASTM standard sieves.
Dissolution of U and REEs. In each experiment, a weighed portion of the representative sample was agitated in a certain volume of the solution for a defi nite time period at defi nite S/L ratio. In these experiments, a 40-g portion of the fi ne Abu-Tartur representative sample was put in a 250-mL beaker. A number of experiments were carried out with diff erent values of EDTA concentration, reaction time, solid/liquid ratio, and temperature. The reaction mixture was agitated using a mechanical stirrer. After that, the mixture was washed several times with distilled water to remove any soluble materials. The EDTA concentration ensuring the maximum dissolution of U and REEs was subsequently used for optimizing the other parameters.  Recovery of U and REEs. Recovery of uranium. Duolite A 101D anion exchange resin (2 mL) in the form of wet settled resin (w.s.r.) was packed in a small glass column 2 cm in diameter, the leach liquor of U/ REEs was passed through it at a rate of 2 mL min -1 , and 100-mL effl uent samples were collected for uranium analysis. In so doing, uranium was adsorbed on the resin and was absent in the effl uent solution, whereas REEs remained in the effl uent solution and were absent on the resin. After the adsorption, uranium was eluted with a 1 M NaCl solution acidifi ed with 0.15 N H 2 SO 4 at a fl ow rate of 3 mL min -1 , and 10-mL eluate samples were collected for uranium analysis.
Recovery of REEs. The effl uent solution after the uranium adsorption, which contained the leached REEs, was passed through the Dowex 50W-X8 cation exchange resin (w.s.r.). 5 mL of the resin was fi xed in a glass column 2 cm in diameter, and the REE solution was passed through it at a fl ow rate of 0.5 mL min -1 . 100-mL effl uent samples were taken for REEs analysis. After complete adsorption, the resin was washed with 2 M HCl to remove all impurities except REEs; then, the REEs were eluted with 4 M HCl at a fl ow rate of 2 mL min -1 , and 10-mL eluate samples were collected for REEs analysis.

Analytical Control
The uranium content was determined by oxidimetric titration with ammonium metavanadate. For this purpose, uranium was preliminarily reduced with (NH 4 ) 2 Fe(SO 4 ) 2 in the presence of diphenylamine sulfonate indicator until it changed color to slightly violet [28]. Also, the uranium content was determined spectrophotometrically with Arsenazo III (λ = 655 nm). The REEs content was determined spectrophotometrically with Arsenazo III [29].

Analysis of the Product
Finally, the precipitated sodium diuranate and rare earth oxide products were analyzed in the NMA Labs with an XL30 Philips type scanning electron microscope (SEM); the results were confi rmed by ICP-OES [30].

Optimization of Dissolution Conditions
Eff ect of grain size. The infl uence of phosphate rock fi neness on the dissolution of U and REEs from phosphate rock was studied using fractions of 700, 500, 150, 74, and 63 μm size. The other parameters were fi xed: 0.3 M EDTA, S/L = 1/3, room temperature. The experimental results are shown in Fig. 1 as the dependence of the U and REEs dissolution effi ciency on the particle size. As can be seen, the dissolution effi ciency increases with a decrease in the particle size from 700 to 74 μm, which may be due to an increase in the active surface area, and does not further change in going to 63 μm; thus, 74 μm is the optimum phosphate particle size, and it was chosen for the other experiments on the U and REEs dissolution.
Eff ect of EDTA concentration. The eff ect of EDTA concentration on the U and REE dissolution effi ciency was studied at the EDTA concentration in the range from 0.01 to 0.4 M. The other dissolution conditions were kept constant: sample weight 40 g, ore particle size  74 μm, solid/liquid ratio 1/3, leaching time 30 min, and temperature 25°C. The results are shown in Fig. 2. As can be seen, with an increase in the EDTA concentration from 0.01 to 0.4 M the U dissolution effi ciency increased from 19 to 35% and the REE dissolution effi ciency increased from 28 to 47 %. Similar result was obtained already with 0.3 M EDTA, and this concentration was chosen as optimum. The residue mainly consists of SiO 2 , CaO, Fe 2 O 3 .
Eff ect of dissolution time. To evaluate the infl uence of time on the dissolution of U and REEs from Abu-Tartur phosphate rock, the time was varied from 30 to 150 min. The other conditions were kept constant: sample weight 40 g, EDTA concentration 0.3 M, S/L ratio 1/3, and room temperature. Figure 3 shows that the U and REEs dissolution effi ciency increases to 51 and 66%, respectively, at 90 min and does not appreciably increase at longer contact times; therefore, we used the agitation time of 90 min in all the experiments.
Eff ect of solid/liquid ratio. To evaluate the eff ect of the ore weight to solution volume ratio on the U and REE dissolution effi ciency, we kept the ore sample weight constant (40 g) and varied the solution volume so that the S/L ratio ranged from 1/3 to 1/6. The other variables were kept constant: EDTA concentration 0.3 M, dissolution time 90 min, and room temperature. Figure 4 shows that the best S/L ratio is 1/4; at this ratio, the U and REE dissolution effi ciency reaches 62 and 79%, respectively.
Eff ect of temperature. To determine the optimum temperature, we performed experiments at 25, 40, 60, 80, and 100°C, keeping the other factors on the optimum level (EDTA concentration 0.3 M, dissolution time 90 min, S/L ratio 1/4). Figure 5 shows that the best results are reached at room temperature. Probably, at elevated temperatures the EDTA complexes of U and REEs are unstable. Therefore, all the experiments were made at room temperature.

Eff ect of Mineral Acids Added to EDTA on the Dissolution of U and REEs
Diff erent mineral acids (HCl, H 2 SO 4 , and HNO 3 ) were added to 0.3 M EDTA. The other variables were fi xed: sample weight 40 g, ore particle size 74 μm, solid to liquid ratio 1/4, dissolution time 90 min, and 25°C. The best results were reached with HNO 3 (Figs. 6 and  7). With an increase in the nitric acid concentration from

Ion-Exchange Recovery of Uranium and REEs
The leach liquor was obtained under optimum conditions: PR particle size 74 μm, (0.3 M EDTA/0.05 M HNO 3 ), agitation time 90 min, S/L ratio 1/4, and room temperature. The leach liquor was analyzed colorimetrically to determine major oxides and by ICP-OES to determine trace elements. The results are given in Table 3. The leach liquor was passed through the anion-exchange resin to recover uranium and then through the cation-exchange resin to recover REEs.
Ion-exchange recovery of uranium. Uranium adsorption. After the dissolution of uranyl and REEs with EDTA + nitric acid, uranium was then adsorbed onto Duolite A 101D anion-exchange resin. The fi rst 100-mL portions of the effl uent contained virtually no   uranium; i.e., uranium was completely adsorbed on Duolite A 101D (Fig. 8). The uranium breakthrough became appreciable in the sixth portion of the effl uent sample and complete in the tenth portion.
Uranium elution. The uranium adsorption was followed by elution with a 1 M NaCl solution acidifi ed with 0.15 N H 2 SO 4 . The elution curve is shown in Fig. 9;  Fig. 11; it is bell-shaped with the maximum in the fi fth eluate portion. The REE elution was virtually complete.

Uranium and REE Precipitation
Uranium concentrate product. The eluate from Duolite A 101D was alkalized to pH 7 with sodium hydroxide to precipitate sodium diuranate (Na 2 U 2 O 7 ). The product was calcined at 750°C for 1 h, and the uranium content was determined chemically to be 69.0%. The product purity was 91.90%. The fi nal uranium product was identifi ed by scanning electron microscopy (SEM), Nuclear Material Authority, Cairo, Egypt (Fig. 12). The chemical composition of the uranium precipitate given by EDX analysis was follows: O 22.20%, Na 7.74%, P 0.65%, and U 69.41%. The results of ICP-OES analysis are given in Table 4.
Total rare earth oxides product (ΣREO). The total rare earth oxides (ΣREO) were prepared in the form of a light brownish powder from the uranium-free effl uent. The eluate from Dowex 50W-X8 was partially neutralized with ammonia to pH 1.5, and 30% oxalic acid was added to pH 0.8. In so doing, REE oxalates completely precipitated. The precipitation effi ciency reached 97%. The oxalate precipitate was fi ltered off , properly washed with distilled water, and calcined at 850°C for 1 h to obtain total rare earth it is bell-shaped with the maximum in the fi fth eluate portion. The uranium elution was almost complete.
Ion-exchange recovery of REEs. REE adsorption. The effl uent after Duolite A 101D anion-exchange resin was passed through the Dowex 50W-X8 cation-exchange resin for the REE adsorption. The effl uent was collected in 100-mL portions and analyzed for the REE content; the results are shown in Fig. 10. The REE breakthrough was observed starting from the fi fth portion, and starting from the tenth portion it became complete. REE elution. First, the resin was washed with 2 M HCl to remove possible impurities adsorbed jointly with REEs, and then it was eluted with     Fig. 13). The major components of the product were valuable light rare earths (lanthanum, cerium, praseodymium, neodymium, and samarium). The content of Ce 2 O 3 , La 2 O 3 , and Nd 2 O 3 was 39.9, 21.76, and 9.85%, respectively. Impurities of P 2 O 5 , CaO, MgO, and Fe 2 O 3 were also detected.

CONCLUSION
Our study of the dissolution behavior of uranium and REEs from the Abu-Tartur PR shows that the optimum conditions ensuring almost complete dissolution of both these two metal values are as follows: parical size 74 μm, 0.3 M EDTA + 0.05 M HNO 3 , agitation time 90 min, S/L ratio 1/4, and room temperature. From the leach liquor, uranium and REEs were successfully recovered by ion exchange. Duolite A 101D anion-exchange resin was used for uranium recovery, and Dowex 50W-X8 cation-exchange resin, for the REEs recovery from the uranium-free effl uent. The fl ow diagram of the process is shown in Fig. 14.