Column pulsed extraction

Nagel, V. Gilmore, M. Scott, S. Bateman pulsed column pilot-plant campaign to extract cobalt from the nickel electrolyte stream at Anglo Platinum s base metal refinery. International Solvent Extraction Conference, Cape Town, South Africa, Mar. 17-21, 2002, pp 970-975. 

A recent and extremely important development lies in the application of the technique of liquid extraction to metallurgical processes. The successful development of methods for the purification of uranium fuel and for the recovery of spent fuel elements in the nuclear power industry by extraction methods, mainly based on packed, including pulsed, columns as discussed in Section 13.5 has led to their application to other metallurgical processes. Of these, the recovery of copper from acid leach liquors and subsequent electro-winning from these liquors is the most extensive, although further applications to nickel and other metals are being developed. In many of these processes, some form of chemical complex is formed between the solute and the solvent so that the kinetics of the process become important. The extraction operation may be either a physical operation, as discussed previously, or a chemical operation. Chemical operations have been classified by Hanson(1) as follows ... 

THORNTON, J. D. Trans. Inst. Chem. Eng. 35 (1957) 316. Liquid-liquid extraction. Part XIII. The effect of pulse wave-form and plate geometry on the performance and throughput of a pulsed column. 

The need to use multiple extraction to achieve efficient extraction required the development of new types of continuously working extractors, especially mixer-settlers and pulsed columns, which were suitable for remotely controlled operations. These new extractors could be built for continuous flow and in multiple stages, allowing very efficient isolation of substances in high yield. A good example is the production of rare earth elements in >99.999% purity in ton amounts by mixer-settler batteries containing hundreds of stages. These topics will be further developed in Chapters 6 and 7. 

Irradiated UO2 is dissolved in nitric acid, resulting in a dissolver solution with the approximate composition listed in Table 12.7. This is treated by the Purex process. The main steps in the conventional Purex process are shown schematically in Fig. 12.5. All existing plants listed in Table 12.8 use some variation of the Purex process. Typically, the extractant composition (percentage TBP, diluent) and the extraction equipment (i.e., pulse columns, mixer-settlers, etc.), vary from plant to plant. However, the upper concentration limit is 30% TBP to prevent a phase reversal due to the increased density of the fully loaded solvent phase. 

A pilot plant, containing two pulsed columns, one for extraction and one for stripping, and batchwise evaporation was in operation in Sweden during 1981. Pilot plant operations have also been performed in Holland (Fig. 14.6) and Germany. The experience from these tests shows that the process concept is technically practicable and well proven. The economics of the process, however, are strongly dependent on the cost for disposal of spent pickling liquors. 

Kinetics of phase transfer should be sufficiently fast, both at the extraction and stripping steps, to allow short-residence time contactors to be employed. In pulsed columns, the contacting time averages a few minutes, while in centrifugal contactors, this contacting time might shorten to only a few seconds. 

In the tests carried out in centrifugal extractors, the extraction and recovery of cesium higher than 99.99% were obtained on simulated effluents, with a very good coherence between calculated flowsheets and experimental results.102 Tests confirmed the feasibility of the implementation of the cesium process in pulsed columns, the latter representing the most adapted contactors for the industrial implementation to overcome the drawback due to the presence of solid matter in waste to be treated. 

For cesium extraction, Couette columns have been used. The operations in such a contactor can be easily extrapolated to a pulsed column, and the quantity of implemented solvent is less important. The hydrodynamic behavior is satisfactory for the two calixarene systems studied in spite of a more emulsion-prone behavior with system 2. The selectivity obtained compared to other elements is excellent. 

The uranium and plutonium are recovered for further use by first dissolving the spent fuel in nitric acid and subjecting the resulting solution to a solvent extraction process. Several different processes exist, the best known being the Purex process (Fig. 18), in which tributyl phosphate (TBP) (30% solution in kerosene) is the extractant. Extraction is carried out in compact mixer-settlers or air-pulsed columns fabricated of stainless steel, with about 99.9% removal of uranium and plutonium in the extract. 

Nickel and cobalt often occur with copper, and must be separated in pure form from hydrometallurgical leach liquors. Organic acid extractants can quite readily separate copper from cobalt and nickel, but the separation of cobalt from nickel is rather difficult. In one Ni/Co separation process, di-2-ethyl hexyl phosphoric acid (D2EHPA) is used as extractant, with strict control of the pH of the aqueous phase to take full advantage of the slightly different equilibrium constants for the Co and Ni reactions. Pulsed column contactors are used rather than mixer-settlers, and nickel impurity is removed from the loaded organic phase by scrubbing it with a cobalt-rich phase. 

Pulsed columns are generally not used for column diameters greater than 180 cm because of problems with mechanical support and pulse generation. The great amount of backmixing also causes uncertainties in column performance. Small pulsed columns have been successfully used for the extraction of metals from slurries (Warwick, 1973). 

Centrifugal contraction or pulsed columns are used for these solvent extraction operations in preference to longer packed columns because the latter would complicate the... 

Extraction columns can be further sub-divided according to the method used to promote contact between the phases packed, plate, mechanically agitated, or pulsed columns. Various types of proprietary centrifugal extractors are also used. 

Carlson and Nielsen (C3) described the pilot and full-scale plant separation of an ore containing more than 30% combined columbium and tantalum oxide using a sulfuric-hydrofluoric acid leach and methyl isobutyl ketone (MIBK) as solvent in pulsed columns. The —200 mesh columbite-tantalite ore was digested with 70% HF until the combined (Ca + Ta)20s in the leach liquid reached 3 Ib/gal at which time it was diluted to 15N free acid and clarified by filtration. This solution was contacted countercurrently in the pulsed column where Ta and Cb were extracted by MIBK. Columbium was stripped from the organic with demineralized water which diluted the free acid in the solvent, making possible the transfer of all the Cb the Ta-loaded solvent was then stripped with demineralized water causing the transfer of Ta to the aqueous phase. The oxides were then precipitated with 28% ammonium hydroxide solution. Conversion to the respective oxides was by calcination of the precipitates. 

Uranyl nitrate was extracted with tributyl phosphate in a 3-in. diameter pulse column with a perforated-plate section height of approximately 9 ft. Plate-free end sections 3-4 in. in diameter and up to 6 ft in length were incorporated in the design to give several minutes of holdup time for phase separation. The pulse was applied by means of reciprocating stainless-steel bellows or a reciprocating piston. 

Conventional solvent extraction contactors, mixer-settlers or pulsed columns, have been used exclusively until now for the Pu-U partitioning step. Centrifugal contactors have been considered, but there has been some concern about the compatibility of short residence time with the kinetics of plutonium reduction. 

Extraction tests with U showed that the extraction efficiency is very little affected by internal electrodes. For example an HTS value of 0,8 m was obtained for the U extraction. This value is similar to that of normal pulsed columns of these dimensions. 

Flowsheet testing and data collection were performed using the General Atomic Company solvent extraction pilot plant equipment shown in Figure 1. Included in this equipment are several 5.1 to 7.6 cm (2 to 3 in.) diameter cylindrical glass pulse columns, a 15-2 cm (6 in.) diameter annular pulse column, a centrifugal contactor (Robatel Co.) and associated tanks, feed systems and concentrators. 

The exhaustive U, Pu extraction step is best achieved in pulsed columns rather than in mixer settlers in order to keep the contact time lower at the very high radiation level. This operation is expected to produce as its main product a HAW raffinate that is virtually free of Pu (and Np, U) a scrub appears therefore superfluous and would dilute only the HAW. In order to attain still a good extraction efficiency within a few stages, A/O phase ratios < 3 should be avoided. This ratio corresponds to a TBP saturation by heavy metals of about 18%. Considerable amounts of Zr are therefore co-extracted, being however present essentially as inactive isotope. Either a complexing or a reductive stripping is advisable in order to keep the aqueous flow small and the acidity sufficiently high to avoid hydrolysis of Zr. 

The following extraction-stripping operation provides for complete Am and Cm extraction, together however with the RE fraction of the FP and substantial amounts of Ru. It acts in the 1st option as a back-up cycle for Pu and Np separation, otherwise these elements too must be separated here completely (option 2). The utilisation of pulsed columns remains the first choice also for this operation and the aqueous raffinate must meet the final specifications for "alpha free" HAW. Again it is therefore of highest importance to run the extraction section at convenient conditions, i.e. at a sufficient stage number and phase ratio. To an A/O ratio of 0.33 corresponds 33% TBP saturation (RE). Only experimental counter-current tests may be able to show if this phase ratio can be increased and if perhaps a scrub could provide for a better separation mainly from Ru, without compromising the complete Am-Cm extraction. 

---