Mixer-settler Research Articles

Continuous enzymatic penicillin G hydrolysis in countercurrent water–butyl acetate biphasic systems

Penicillin G (Pen G) was continuously hydrolyzed into phenylacetic acid (PAA) and aminopenicillanic acid (APA) by penicillin acylase in countercurrently contacted water–butyl acetate biphasic systems. A set of three mixer–settlers was used to establish countercurrent contact of the two immiscible liquid phases. The enzymatic reaction was catalyzed by immobilized penicillin acylase. Pen G was fed in the central mixer. APA continuously left the system with the aqueous exit stream and PAA with the organic exit stream. The conversion in the countercurrently contacted biphasic system was significantly higher than the conversion that can be achieved in an equivalent batch system. A mathematical model, which is based on partition and reaction equilibria, was used to predict the concentrations and pH along the countercurrent reactor. Provided that no APA crystallization occurs, the mathematical model predicted the conversion and the pH profile in the mixer–settler cascade accurately.

Scaling-up experience of actinides partitioning from intermediate level high salted alkaline waste streams of Purex process using Versatic-10 extractant

Partitioning of minor alpha-emitting actinides, especially U, Pu and Am from medium active alkaline waste is possible from intermediate level liquid wastes (ILLW) produced during spent fuel reprocessing following Purex process. This paper deals with the efficient removal of alpha-activity from ILLW by solvent extraction process. Counter current batch extraction with O/A ratio 2:1 as well as multistage mixer settler has demonstrated that most of the alpha-activity was removed from the alkaline effluents using 20% Versatic-10 (V-10) in dodecane after giving 3 to 4 contacts, thus converting alkaline waste as non-alpha waste. Under the optimum conditions (pH 9.0-9.5 and VA-10), both Pu(IV) and Am(III) are highly extractable whereas U(VI) is relatively poorly extracted. To assess the applicability of this process for regular treatment of the waste, a feasibility study on pilot plant scale using six stage mixer settler was operated to treat the ILLW. The results indicated that almost >99.90% alpha-emitting actinides are removed. Dilute nitric acid (0.5M HNO3) served as the most efficient strippant for all these actinides. This facilitate an easy regeneration of the extractant which can be recycled. This method is useful in obtaining alpha-free wastes and had positive impact on ease and safety aspects during subsequent waste treatment and long term storage.

Counter-current extraction process for recovery of U(IV) from phosphoric acid using octylphenyl acid phosphate (OPAP) extractant

The extraction of U(IV) from phosphoric acid by octylphenyl acid phosphate (OPAP) in kerosene was investigated. Parameters affecting the extraction of U(IV) from phosphoric acid were investigated. The effects of H3PO4, H2SO4, H2O2, Na2CO3 and (NH4)2CO3 concentrations, phase ratio and temperature on the stripping of uranium from the organic phase were studied. Based on the bench-scale results, a continuous counter-current extraction flow sheet was developed and tested using a 16-stage horizontal type mixer settler. The continuous extraction scrubbing stripping showed that the extraction efficiency of the developed process is 99%, whereas the stripping efficiency is 97%.

Extraction of rare earth metals with a multistage mixer-settler extraction column

Extraction behavior of rare earth metals within a mixer–settler extraction column has been analyzed with the stage efficiency calculated from mass transfer coefficients and interfacial area. The mass transfer coefficient within the dispersed drops is determined from a rigid sphere model by taking into account the residence time distribution of drops, and the coefficient around the drops is calculated by Ranz–Marshall's correlation with the terminal settling velocity of a rigid sphere. The interfacial area of dispersed drops is calculated by the use of correlations for the drop diameter and the holdup of dispersed phase in the mixer–settler extraction column. The calculated results for the separation of samarium and gadolinium with a five-stage mixer–settler extraction column are compared with the experimental results at various agitation speeds and flow ratios between two phases. The extraction behavior in the multistage column is explained by a model based on the hydrodynamics and the mass transfer within the mixer. Effects of the pH value in aqueous phase, the extractant concentration in organic phase and the number of stages on the extraction behavior in the mixer–settler column are also shown.

Distribution Behavior of U(VI), Pu(IV), Am(III), and Zr(IV) with N,N-Dihexyl Octanamide under Uranium-Loading Conditions

While the tri-n-butyl phosphate (TBP)-based PUREX process has been the workhorse of the nuclear fuel reprocessing industry for the last four and a half decades, a few drawbacks associated with the use of TBP have caused concern to the separation scientists and technologists. These shortcomings may pose a serious challenge particularly during the reprocessing of (a) short cooled thermal reactor fuels, (b) fast reactor fuels with the larger Pu content and significantly higher burn up, and (c) while treating various waste streams for their disposal to the environment. The N,N-dialkyl aliphatic amides have received particular attention as alternate potential extractants for the reprocessing of spent nuclear fuels in view of (a) the innocuous nature of their degradation products, namely, carboxylic acids/amines and (b) the possibility to incinerate the used solvent leading to reduced volume of secondary waste. The physical and chemical properties of these amides are influenced strongly by the nature of alkyl groups. The extractant N,N-dihexyl octanamide (DHOA) was found to be a promising candidate among a large number of extractants studied. Laboratory batch studies as well as mixer settler studies were performed under process conditions with DHOA and compared with those of TBP. DHOA was found to extract Pu(IV) more efficiently than TBP, both at trace-level concentration as well as under uranium loading conditions. In addition, the extraction behavior of Am(III) and Zr(IV) was studied at varying nitric acid concentrations (1 to 6 M). Extraction behavior of uranium at macroconcentrations (9.9 to 157.7 g/l) was carried out at different temperatures, and it was observed that DU decreased with the increase in U loading as well as with the increase of temperature (in the range 25 to 45°C) and that the two-phase reaction was exothermic in nature. Mixer settler studies on U(VI) revealed that DHOA is similar to TBP during the extraction cycle but better than TBP during the stripping cycle.

Utilization of Micromixers for Extraction Processes

The aim of the investigation was to evaluate the extraction performance of a mixer settler set-up for miniplant technology using static micromixers as an alternative to conventional stirring apparatuses. A comprehensive experimental study was conducted at BASF AG to broaden the technology base for the “extraction” unit operation which is well established for miniplants in order to be utilized for microplant systems. The work proved that micromixers, or micromixer arrays, are highly efficient apparatuses for extraction purposes. The extraction efficiency was found to be a function of volume flow, which could be explained in light of the volume flow dependence of the mean droplet size and, hence, the specific surface area of the dispersions intermediately formed. At optimal flow conditions, one practical mixer settler stage was found to yield one theoretical plate for most systems investigated.

Uranium extraction from purified wet process Jordanian phosphoric acid: a development study

A study on the physical properties, liquid–liquid equilibrium and the performance of a pulsed sieve plate column was conducted using the system: Jordanian purified filter phosphoric acid (28% P 2O 5)–uranium–0.5 M DEPA–0.125 M TOPO diluted in kerosene. Density and surface tension were found to follow simple negative linear relationship with temperature while an Arrhenius type model was adequate to fit the viscosity dependence on temperature. The effect of electromotive force (EMF) on the uranium distribution coefficient was studied and a practical limit of oxidation of the uranium was found to be in the range 400–410 mV. Vant Hoff equation was found to adequately describe the dependence of uranium distribution coefficient on temperature. Uranium distribution coefficients obtained for the Jordanian acid were higher than reported data for other acids, with distribution coefficients as high as 6.0 obtained at about 45°C. A preliminary evaluation of the performance of a pulsed sieve plate column was conducted and yielded promising results for the application of such contactors in this field. Recoveries of 90% were obtained at moderate pulsation intensities. Solvent entrainment results indicate that about 1900 ppm of the solvent is lost to the acid, which is higher than that for mixer settlers.

Simultaneous purification of dysprosium and terbium from dysprosium concentrate using 2-ethyl hexyl phosphonic acid mono-2-ethyl hexyl ester as an extractant

During the solvent extraction fractionation of rare earth chloride obtained from monazite, a heavy rare earth fraction assaying 60% Y 2O 3 is produced. This is purified further to 93% Y 2O 3 by another solvent extraction cycle. During this step, most of the Dy and Tb are separated to yield a concentrate assaying >50% Dy 2O 3, 14% Tb 4O 7, 10% Gd 2O 3, 2.4% Ho 2O 3, and 21% Y 2O 3. An attempt has been made to process this Dy-rich concentrate to obtain a high grade Dy 2O 3 by solvent extraction using 2-ethyl hexyl phosphonic acid mono-2-ethyl hexyl ester (PC 88A) in paraffinic kerosene as an extractant. The distribution ratio ( D) of Dy was determined as a function of initial concentration of HCl for different initial concentrations of Dy. The separation factors of Dy/Gd, Dy/Tb, Y/Dy, Ho/Dy, and Er/Dy were determined experimentally. Using the experimentally determined D values of Dy and the separation factors, the distribution data for Gd, Tb, Y, Ho, and Er were derived mathematically. Based on these data, empirical mathematical models have been made to predict the concentration of these metal ions in organic and aqueous phases at various initial acidities and metal concentrations. Using these models, a computer program was developed to calculate the concentration of the metal ions in organic and aqueous phases at various stages of extraction and scrubbing in a counter-current solvent extraction cascade. Using this program, the parameters of the process, such as the initial acidity of the feed, the acidities of scrubbing and stripping solutions, the phase ratio, the number of stages in extraction, and the scrubbing sections have been optimised to obtain >97% purity of Dy. Under these optimised conditions, counter-current extraction and scrubbing tests were carried out using mixer settlers of 50-ml mixer capacity. Mixer settler runs yielded four different products from its four exit points. The raffinate coming from the first exit was essentially Gd-rich solution containing other light rare earths (LRE). The fourth and final exit point was for the stripped solution that contained 85% Y 2O 3 with 95% recovery. The remaining two exit points, involving scrubbing stages, yielded an 83% pure Tb 4O 7 concentrate and a 97% pure Dy 2O 3 concentrate.

Dynamic behaviour of the mixer–settler cascade. Extractive separation of the rare earths

Dynamic modelling of a countercurrent multistage process is important not only for solving start-up and control problems but also for its possible use for performance analysis, specifically in a multicomponent non-linear extraction process. However, the respective differential equations are often stiff, complicated by non-linear equilibrium relations, and their numerical solution is very difficult. The problem is overcome by the proposed pulsed-flow model. In the model, each extraction stage is considered as a set of ideal mixers and the continuous process is represented by repeated sequences of pulsed flow and a batch mass transfer. Moreover, it is possible to introduce extraction efficiency into the pulsed-flow model. A typical example of the process with the above features is the extractive separation of the rare earth elements in a mixer–settler cascade. Adequacy of the pulsed-flow model was tested and its parameters were determined by dynamic experiments on a 6-stage cascade. The pulsed-flow model proved to be very useful for finding the steady-state operating conditions ensuring the required product purity in the rare earths separation using bis(2-ethylhexyl)phosphoric acid (DEHPA) in a 30-stage extraction–stripping cascade with extract and raffinate refluxes. The model was also used for the prediction of responses to some disturbances and for the analysis of control configurations.

Back extraction of lactic acid with microporous hollow fiber membrane

In this work, back extraction was considered a promising alternative to regenerate carboxylic acids of low-volatility into appropriate aqueous solution from organic phase. Identification of an efficient back extraction system for lactic acid recovery was focused on. Screening of back extraction reagents was first carried out by employing a wide variety of inorganic compounds. From the viewpoints of high stripping power, low cost and capability of simultaneously regenerating the extractant, aqueous sodium chloride solution was selected as the most suitable one for recovery of lactic acid from lactate–TOMAC (tri- n-octylmethylammonium chloride) complex. However, TOMAC and oleyl alcohol are likely to pose the troubling emulsion problem in conventional mixer–settler system. Therefore, non-dispersive back extraction of lactic acid from organic phase was attempted in a microporous hollow fiber (MHF) membrane device by aqueous NaCl solution. A satisfactory recovery was accomplished, signifying the great potential of integrating membrane back extraction with extractive fermentation process for lactic acid production.

Recovery of titanium from the leach liquors of titaniferous magnetites by solvent extraction: Part 3. Continuous mini-plant trials

The recovery of titanium from the sulphuric acid leach of a titaniferous magnetite ore by solvent extraction has been investigated in two mini-plant campaigns. The leach liquor contained approximately 5 g l −1 titanium, 1 g l −1 vanadium, and 55 g l −1 iron in 75 g l −1 sulphuric acid. The objective was to achieve the maximum possible titanium recovery with maximum rejection of iron and vanadium. The organic phase comprised 10 vol.% tri- n-octylphosphine oxide (TOPO) in an aliphatic diluent. Stripping was effected using sulphuric acid. The first continuous counter-current mixer–settler mini-plant employed mixers of 100 ml capacity and was operated for 180 h. A larger mini-plant of 500 ml mixer capacity was subsequently commissioned and operated for 102 h. The degree of titanium extraction varied between 76 and 88%. Vanadium co-extraction was negligible (less than 1%), while the iron concentration in the loaded strip liquor was reduced to less than 2 g l −1. Approximately 60% of titanium loaded onto the organic phase was stripped, while 97% of the iron and 76% of the vanadium were stripped.

The mixer–settler principle as a separation unit in supercritical fluid processes

Processes with supercritical fluids are known for both solid and liquid materials. Liquids are typically brought into contact with the supercritical fluid in countercurrent columns. In recent years, high-pressure spraying processes such as the spray extraction of viscous media have been investigated. The central idea is to improve mass transfer by dispersing the liquid phase in the supercritical phase. A mixer–settler system is investigated for the regeneration of caffeine-loaded CO 2 at various conditions. Water is sprayed into the CO 2 flow and the resulting two-phase flow passes through a static mixer to intensify phase contact and prevent coalescence of the water phase. The system shows an efficiency of 80–90% in comparison with the theoretical maximum based on ternary equilibrium data from the literature. The pressure loss of the static mixer (Kenics) is determined experimentally for various conditions. The pressure drop of the two-phase flow depends on CO 2 velocity and is significantly determined by the mass ratio. It does not exceed 2500 Pa in the present investigations, and can be neglected in supercritical fluid plant design.

The kinetics of extraction in a novel solvent extraction process with bottom gas injection without moving parts

The extraction kinetics in a novel solvent extraction process have been investigated using the CuSO 4–H 2SO 4–H 2O–LIX 860–kerosene system. In this process, the liquid–liquid emulsion is generated by bottom gas injection, rather than mechanical stirring. This process has a number of advantages over the mixer–settler unit or the spray column in terms of simple equipment configuration and the ease of cleaning and process control, while providing a sufficiently large interfacial area for mass transfer. The overall rate constant increased with increasing gas injection rate and height of the aqueous phase but decreased with increasing injector diameter. The interfacial area per unit volume of the aqueous phase was estimated from the mean drop size and the dispersed-phase holdup computed from previously established correlations. The variation of the overall rate constant was correlated against a set of dimensionless numbers, representing the operating conditions. Extraction rates and efficiencies comparable to those in a mechanically generated emulsion were attained by this new method.

Residence time distributions in SX–EW equipment

Residence time distribution functions were determined and modelled for a pilot plant mixer–settler and an electrowinning cell. The mixer RTD is modelled by a small plug flow reactor, a small perfect mixer and a large size one. The settler RTD is modelled by Cholette-Cloutier units in parallel with a series of small perfect mixers, although frequently the settler is assumed to behave as a plug flow. Finally the electrowinning cell RTD is very close to being a perfect mixer.

A novel solvent extraction process with bottom gas injection without moving parts

A novel solvent extraction process in which the emulsion is generated by bottom gas injection rather than mechanical stirring has been developed. This process has a number of advantages over the mixer–settler unit or the spray column in terms of simple equipment configuration and the ease of cleaning and process control, while providing a sufficiently large interfacial area for mass transfer. The equipment consists of a horizontal cylindrical vessel in which the two immiscible liquids flow countercurrently. High-strength gas jets are injected from the bottom at certain intervals along the length of the vessel. The gas jet creates a plume zone consisting of an emulsion of the two liquids which contains a large interfacial area for rapid mass transfer. The two liquids then disengage and flow in the opposite directions before entering another plume zone. Thus, the process combines the simplicity of a cylindrical vessel, having no moving parts, with the contacting efficiency of a mixer–settler. The gas can be recycled in a closed loop, thus eliminating mist and other emission problems. By using an inert gas, the oxidation degradation of the extractants and/or solutes can be prevented.

97/02574 Factors influencing the moisture content of coarse coal after centrifugation

It is a well-known fact that a typical engineering design problem usually deals with more than one design criterion. If each design criterion is stated as an objective function to be optimised, then the engineering design problem becomes a multicriterion optimisation problem, requiring the simultaneous optimisation of more than one objective function.In this paper, it is shown how the design of solvent extraction flow-sheets can be stated as a multicriterion optimisation problem, using the positive weighted sum approach. This is used not only to obtain parametric optimisation (i.e., the best operating conditions: agitation speed, residence time and phase flow ratio) but also to help in structural optimisation (i.e., to synthesise the best process flow-sheet: number of stages, flow structure and phase recycle ratio). We demonstrate this over a case study, namely, the selective separation of two chemically akin and hard to separate metals, zinc and cadmium, commonly found together in the leaching liquor of complex ores.With this case study, it is shown that the design solutions are richer and more wide-ranging when put together from the vantage point of multicriterion optimization, whereas they become narrow-minded and/or biased if the starting point is a single criterion point of view. Three other conclusions of less general validity were also obtained: (i) the opposite effects of feed phase flow-rates on recovery and purity; (ii) the high sensitivity of short optimum residence times to variations in agitation speed; (iii) the ability of counter-flow associations of a variable number of mixer–settler units to accommodate changes in metal purity and overall recovery in response to drivers in market prices and environmental policies.

Design of a Mixer Settler Plant for Isotope Separation by Chemical Exchange

The methodology necessary to design a solvent extraction plant for isotope separation by chemical exchange is outlined. This process involves the calculation of the number of stages used in banks of mixer settlers used in such a plant. The feed is introduced at both ends of the plant, and the product is withdrawn at one end and the tailings at another. The resulting analysis, illustrated for the case of uranium enrichment, produces a fairly simple algorithm, and then more complicated configurations are examined. The method of design can be used for other systems where the isotope separation factors are more favorable

Fault Tree Analysis of System Anomaly Leading to Red Oil Explosion in Plutonium Evaporator

Results of scenario identification, preparation of reliability data, fault tree construction and its analysis, are rendered for a system anomaly considered liable to lead to a red oil explosion in a Pu evaporator of a typical model of reprocessing facility. Japan Atomic Energy Research Institute has introduced the model plant data, the basic failure frequency data together with the fault tree analysis code FTL from NUKEM GmbH, Germany, and used to execute the above analyses. The frequency of occurrence of the system anomaly initiated by failures of the Pu purification process such as a pulse column failure or a mixer settler failure has been evaluated to be 2.6×10-3/yr for the solvent carry over into the evaporator for the German model reprocessing facility. In addition to this event, if an abnormal state such as continuously heating the solution to over 150°C inside the evaporator would happen with the analyzed occurrence probability of 1.0×10−8, it would be the more possible to result in the red oil explosion. The occurrence frequency for the combined events has been calculated to be 2.6×10−11/yr, so small compared with the public acceptance limit which might be set as 1×10-6/yr.

Fault Tree Analysis of System Anomaly Leading to Red Oil Explosion in Plutonium Evaporator.

Results of scenario identification, preparation of reliability data, fault tree construction and its analysis, are rendered for a system anomaly considered liable to lead to a red oil explosion in a Pu evaporator of a typical model of reprocessing facility. Japan Atomic Energy Research Institute has introduced the model plant data, the basic failure frequency data together with the fault tree analysis code FTL from NUKEM GmbH, Germany, and used to execute the above analyses. The frequency of occurrence of the system anomaly initiated by failures of the Pu purification process such as a pulse column failure or a mixer settler failure has been evaluated to be 2.6×10-3/yr for the solvent carry over into the evaporator for the German model reprocessing facility. In addition to this event, if an abnormal state such as continuously heating the solution to over 150°C inside the evaporator would happen with the analyzed occurrence probability of 1.0×10−8, it would be the more possible to result in the red oil explosion. The occurrence frequency for the combined events has been calculated to be 2.6×10−11/yr, so small compared with the public acceptance limit which might be set as 1×10-6/yr.

Analysis of a Single-Stage Mixer–Settler Performance

Abstract A new design of a single-stage mixer–settler is presented. The mixer, a vertical cylindrical vessel 5.5 cm in diameter and 18.0 cm long, is used for mixing the phases in its lower part. Complete solute transfer occurs in the mixer since coalescence of the dispersed phase droplets takes place on their way up via the mixer outlet to the settler. The settler, a vertical tower 3.4 cm in diameter and 40.0 cm long, serves only as a flash separator. The contactor has been tested by the toluene-acetic acid–water system in both directions of mass transfer. The contactor operates as an ideal stage when tested with the acid system. The hydrodynamics and the dynamic behavior of the contactor have been studied. A model approach has been adopted for the mixer. Agreement between the model output data and the experimental results was excellent.