Pulsed Column Research Articles

A CFD‐PBM‐ANN framework to simulate the liquid–liquid two‐phase flow in a pulsed column

AbstractCFD‐PBM numerical simulation is a powerful tool in the research of droplet swarm behavior. In this work, an artificial neural network (ANN) based droplet breakage frequency function is established based on the directly measured data from our previous studies. Then, the weights and biases of ANN are embedded into the CFD‐PBM code in the form of matrices and vectors. For the first time, a CFD‐PBM‐ANN simulation framework is established. Simulation results are in good agreement with the experimental data under different operation conditions. The cumulative droplet size distribution decreases with the increase of interfacial tension and pulse intensity. It is also found by the simulation that the droplet breakage frequency is relatively high at the edge of disc and doughnut plate, which is accordant with the distribution of turbulent energy dissipation and velocity gradient.

Interfacial relaxation of droplets caused by mass transfer in a pulsed column

AbstractThe method developed in our previous work (Wang et al., AIChE J. 2020; 66(8): e16257), was used to study the influence of the mass‐transfer direction and flux on the dynamic interfacial tension (DIFT) in a pulsed column. The breakup frequency was the highest near the entrance of the dispersed droplet, corresponding to the lowest interfacial tension and highest mass transfer flux. The decrease in the interfacial tension was proportional to the square root of mass‐transfer flux, and this difference is defined as a new parameter, named interfacial relaxation. When the mass‐transfer direction was changed from the dispersed phase to continuous phase to the opposite direction, the proportion of binary breakage was decreased from ~80% to ~40% and the interfacial relaxation was more sensitive to the mass‐transfer flux. A DIFT model was finally established based on the interfacial relaxation parameter, which predicted the experimental results with an error of ±10%.

Hydrodynamic features of pulsed solvent extractor for separation of two metals by using the antagonistic effect of solvents

Separating copper and cobalt ions is crucial due to the industry’s strategic reliance on both these elements. When the extraction process is able to significantly increase the separation factor, it becomes favorable to separate two ions. However, the presence of Cu(II) ions together with Co(II) hinders the achievement of optimum efficiency when using commonly available extractants. This study conducted the separation of the two elements using both batch and continuous methods in a pilot plant pulsed column equipped with a disc and doughnut structure. The initial step involved optimizing the key variables to maximize the separation factor using the central composite design procedure. The optimization of Cyanex272, Cyphos IL 101 concentrations, and the pH value of the aqueous phase were all adjusted to 0.024 M, 0.046 M, and 7.3, correspondingly. In the following step, the hydrodynamic characteristics and extraction performance were examined in the pulsed column of the pilot plant. The findings indicated that the presence of Cyphos IL 101 resulted in an increased separation factor and efficiency within the column. As a result, the ionic liquid enhances performance without encountering any operational issues. This additive is considered an environmentally friendly solvent and does not cause any negative impacts. Consequently, it is suggested for utilization in continuous industrial processes.

Experimental investigation, modelling, and order of magnitude analysis of oxygen mass transfer in pulsed plate column with α‐Fe2O3 nanofluid

AbstractVolumetric oxygen mass transfer coefficient (kLa) is an important parameter in the design of various reactors and bioreactors. In the present work, the influence of α‐Fe2O3 nanofluid on the enhancement of kLa is studied in a pulsed plate column (PPC). An enhancement factor of greater than one showed that the nanofluid is favourable in enhancing the mass transfer rate. The effect of pulsing velocity on kLa is observed to fall under two regimes: the dispersion regime and emulsion regime. The kLa enhancement factor is found to be higher in TiO2 nanofluid than in α‐Fe2O3 nanofluid, indicating that the type of nanofluid influences the enhancement factor. The order of magnitude analysis showed that localized convection triggered by the Brownian motion of nanoparticles is the phenomenon responsible for kLa enhancement. A dimensionless multiple regression analysis (MRA) model was developed to predict kLa in the nanoparticle loading range of 0.003–0.019 (v/v%), relating the Sherwood number with oscillating Reynolds number (1200 ≤ Reo ≤ 20,000), gas flow Reynolds number (0.135 ≤ Reg ≤0.370), Schmidt number (1300 ≤ Sc ≤2700), and Brownian Reynolds number (2.81 × 10−4 ≤ ReB ≤5 × 10−4). The pseudo‐homogeneous model could accurately predict the enhancement until critical loading conditions.

Forecasting Sauter mean droplet size and examining the range of droplet sizes in a Tenova liquid–liquid extraction column

A new type of Tenova pulsed extraction column was introduced in 2017. It is the newest generation of pulsed columns. Due to the internal equipment of this column and the lack of moving parts and the simplicity and speed of repairs and maintenance, it has been the focus of researchers in recent years. No correlations for predicting the mean drop size and drop size distribution of the Tenova column have been reported. The Sauter mean drop diameter and drop size distribution are investigated for a Tenova pulsed column with a diameter and an active height of 7.4 and 73 cm, respectively. Three standard chemical systems of isobutyl acetate-water, isobutanol-water, and toluene-water have been used. The effects of pulse intensity, dispersed and continuous phase flow rates have been taken into account. In each experiment, 200–300 drops have been analyzed in a total of 10,000 drops. The investigation covered a spectrum of physical properties, notably surface tension (within a range of 1.75–36 mN/m). Operating conditions including pulse intensity (in the range of 0.2–2 cm/s) and the flow rate of continuous and dispersed phases (in the range of 8–30 L/h) have been investigated. Methods based on artificial intelligence (AI) such as multilayer perceptron neural networks and gene expression programming were combined with a dimensional analysis approach to provide a new approach to estimating the mean drop diameter (d32). Experimental results have been compared with the equations found by other researchers in similar columns. The variation of drop size distribution has also been experimentally obtained.Methods based on artificial intelligence (AI) such as multilayer perceptron neural networks and gene expression programming were combined with a dimensional analysis approach to provide a new approach to estimating the mean drop diameter (d32). Experimental results have been compared with the equations found by other researchers in similar columns. The variation of drop size distribution has also been experimentally obtained.

CFD-PB studies on effect of internals on specific interfacial area in a pulsed sieve plate columns

A CFD model coupled with population balance equations is developed to systematically study the effects of column internals on specific interfacial area generated in a pulsed sieve plate column (PSPC). The model has been validated against experimental data on holdup and Sauter mean drop diameter in laboratory scale PSPCs of different geometries. The column internals comprised of sieve plates with varying hole diameter, fractional open area and interplate spacing. Typically the standard sieve plate cartridge is characterised by sieve plates (3.2 mm diameter sieve holes with an fractional open area of 23%) placed 50 mm apart. A systematic response surface methodology has been adopted to determine the relative importance of these parameters on specific interfacial area. 3 factor 3 level Box Behnken design has been adopted. Specific interfacial area decreases with an increase in sieve hole diameter, interplate spacing and open area. Sieve hole diameter is found to be the most significant parameter affecting specific interfacial area. Important hydrodynamic variables like flow velocities in continuous and dispersed phases and how column internals affect them are also discussed.

A data-driven model of drop size prediction based on artificial neural networks using small-scale data sets

An artificial neural network (ANN) method is introduced to predict drop size in two kinds of pulsed columns with small-scale data sets. After training, the deviation between calculate and experimental results are 3.8% and 9.3%, respectively. Through ANN model, the influence of interfacial tension and pulsation intensity on the droplet diameter has been developed. Droplet size gradually increases with the increase of interfacial tension, and decreases with the increase of pulse intensity. It can be seen that the accuracy of ANN model in predicting droplet size outside the training set range is reach the same level as the accuracy of correlation obtained based on experiments within this range. For two kinds of columns, the drop size prediction deviations of ANN model are 9.6% and 18.5% and the deviations in correlations are 11% and 15%.

Hybrid modeling of drop breakage in pulsed sieve tray extraction columns

Accurate models for pulsed sieve tray extraction columns (PSEs) depend on the correct prediction of the drop diameter to estimate extractive mass transfer across the phase boundary. Phenomenologically, the drop diameter is determined by a balance of drop breakage and coalescence. While for most industrial solvent systems, coalescence plays a minor role; breakage is mostly the dominant phenomenon determining the drop diameter. However, most modeling approaches for drop breakage in PSEs are characterized by a trade-off between a broad validity range and good prediction accuracy. To overcome this limitation, we developed a hybrid breakage model for drop breakage in PSEs in which a physical-empirical model basis is enhanced by data-driven parameter estimator models (PEMs). The hybrid model is based on a revised form of Garthe’s breakage model, for which we developed a linear PEM for the model parameters and two data-driven PEMs for dstab and d100, respectively. The hybrid breakage model was validated on 743 experimental data sets and evaluated based on the pull metric. In a sensitivity analysis, the model correctly predicted the breakage probability over a wide range of solvent properties, operating conditions, and sieve tray geometries. In future studies, the hybrid breakage model can be incorporated into extraction column models without an initial parametrization.

Experimental Studies on Two-Phase Flow with a Novel Slotted Plate Internal in an Air Pulsed Column

ABSTRACT A novel air pulsed column plate internal featuring plates having concentric circular slots of fixed width (3 mm) is conceptualized, tested, and compared against standard sieve plate internal for counter-current two-phase flow. This new plate design (slotted plates) offers significant intensification in terms of specific interfacial area by as much as ~ 50% vis-à-vis standard sieve plates. The studies are carried out in a 3 inch diameter air pulsed column with tap water as the continuous phase and 30% (v/v) tributyl phosphate in dodecane as the dispersed phase. A high-speed imaging system is used to quantify the state of dispersion and obtain drop size distribution along with Sauter mean drop diameter. The effects of pulsing velocity, dispersed phase velocity and continuous phase velocity on dispersed phase holdup, drop size and consequentially specific interfacial area have been systematically studied for both internals. In all cases, slotted plates are characterized by generation of smaller drops and a higher holdup leading to significant improvement in specific interfacial area. Previously reported correlations for estimating dispersed phase holdup and Sauter mean diameter in pulsed sieve plate columns are found to be inadequate for slotted plate internals. Therefore, new correlations for prediction of holdup and drop diameter have been proposed for the new plate design.

Investigation of dispersed phase holdup in a Tenova pulsed liquid-liquid extraction column

BackgroundPulsed extraction columns have been designed and studied in different types of internal equipment such as packed, tray, and doughnut discs. In recent years (2017), researchers have turned to research on Tenova columns, which is the newest generation of pulsed columns. Correlations for predicting the hydrodynamics of the Tenova column have not been reported. MethodA pulsed Tenova column with a 7.4 cm diameter and 130 cm height including 30 pairs of modified doughnuts and discs was fabricated and tested. Standard chemical systems with low, medium, and high interfacial tension approved by the European Federation of Liquid-Liquid Extraction are used. Holdup as one of the most important hydrodynamic parameters in designing is studied. The effects of dispersed and continuous phase velocity and pulsation intensity have been taken into account. Significant findingsDimensional analysis used for predicting hold-up and an error about 8% have been obtained. Other experimental relations proposed by researchers in pulsed columns have been studied and the error of about 16% found. First and second critical pulsation intensity, which are the commencement and the end of breaking of drops has been seen in these experiments. The recommended operating pulsation intensity is between these two critical values.

An Integrated Model Combining Mass Transfer and Chemical Reaction for Co-Decontamination Extraction Step of PUREX Process in a Pulsed Extraction Column

ABSTRACT The accurate prediction of the extraction behaviors of various solutes in PUREX reprocessing process is crucial for the operation and implementation of the actual process. In this paper, an integrated model is developed to predict the extraction behaviors of U, Np, Pu, and HNO3 in the co-decontamination step (1A extraction step) in a pulsed extraction column. The model couples several physical and chemical processes, such as countercurrent flow, mass transfer, and chemical reaction. The mass transfer coefficients and the distribution ratios of U(VI), Pu(IV), Np(IV), Np(V), Np(VI), HNO3 and HNO2 can be obtained using this model. In particular, the redox and disproportion reactions of Np are considered in the model, and the flow direction of Np can be judged under various process conditions and the corresponding influential factors can be analyzed. The judgment is based on the yield calculated from the relative concentration profiles of Np(VI), Np(V), and Np(IV) in the two phases. For neptunium to inter 1AP step, it is necessary to select high nitric acid concentration, low nitrite concentration and especially high flow ratio. For neptunium to inter 1AW step, low nitric acid concentration, high nitrite concentration and low flow ratio are needed. Compared with the experimental data, the relative errors of the distribution ratios of various solutes are less than 30%, and the relative errors of the concentrations of U(VI) and Pu(IV) at the outlet of the organic phase are less than 10%, and our model is indicated to be reliable and applicable for co-decontamination step in the PUREX reprocessing process.

Evaluation of the operating parameters impact on hydrodynamic parameters of horizontal pulsed sieve-plate column using response surface methodology

The uranium extraction process from leach liquor solution of Bandar Abbas ore with Alamine 336 in a horizontal pulsed sieve-plate column has been evaluated numerically. In this study; Alamine 336 with a concentration of 0.125 mol L−1, isodecanol with a concentration of 5% (v/v), and kerosene were used as the organic phase. The aqueous phase was a leach liquor solution containing 0.25 g L−1 uranium and 0.22 mol L−1 sulfate. In this study, the response surface methodology (RSM) was used to investigate the effect of three effective operating parameters in the pulsed column as pulse intensity and the continuous and dispersed phases flow rates on the hydrodynamic parameters such as dispersed phase holdup, slip and characteristics velocities. The obtained results show that the hydrodynamic parameters are strongly dependent on the pulse intensity and turbulence degree of the system. In addition, three new practical correlations based on operating parameters of a column are presented to predict the dispersed phase holdup, slip, and characteristics velocities by the RSM. The correlations presented by the response surface method showed a good prediction of the experimental data with a maximum R2 value of 0.98.

Comparison of dispersed phase holdup and slip velocity in chemical systems with different interfacial tension in an L-shaped pulsed sieve-plate extraction column

The behavior of dispersed phase holdup (DPH) and slip velocity (SV) was studied in a new type of pulsed extraction columns, named “L-shaped pulsed sieve-plate column (LPSPC)” for five chemical systems, including water-kerosene, nitric acid-5% (v/v) TBP/kerosene, nitric acid-15% (v/v) TBP/kerosene, nitric acid-30% (v/v) TBP/kerosene, and uranyl nitrate-30% (v/v) TBP/kerosene. Then the impact of pulsation intensity, interfacial tension, and flow rate of dispersed and continuous phases, on DPH and SV was investigated. The results revealed that both DPH and SV were greatly influenced by the pulsation intensity, interfacial tension and flow rate of dispersed phase though the flow rate of continuous phase had a weaker effect. Finally, using the response surface method (RSM) and dimensional analysis procedure, new semi-experimental correlations were proposed to predict DPH and SV, which corresponded satisfactorily to the experimental data. The average absolute relative error (AARE) values of the correlations obtained by dimensional analysis for DPH and SV were about (6.56, 8.68%) and (5.52, 8.40%) for the horizontal and vertical sections, respectively.

Modeling of mass transfer coefficient using plug, axial dispersion and forward mixing models for extraction of uranium from uranyl nitrate solution by TBP/Kerosene in a horizontal-vertical pulsed sieve-plate extraction column

The behavior of hydrodynamic parameters and mass transfer of an L-shaped pulsed sieve-plate extraction column (LPSPC) in the process of extracting uranium from uranyl nitrate solution by (30% (v/v)) tri-butyl phosphate (TBP)-Kerosene was investigated. The overall volumetric coefficient of mass transfer was subjected to sensitivity analysis by changing the operating parameters such as pulsation intensity and the volumetric flow rate of dispersed and continuous phases. Experimental observations showed that the pulsation intensity had the most significant impact on the behavior of the studied parameters so that its increase led to an increase in the intensity of mass transfer while reducing the droplet size and dispersed phase holdup (DPH) to 55 and 47%, respectively. The mass transfer coefficient was determined by plug flow model (PFM), axial dispersion model (ADM), and forward mixing model (FMM). The absolute average relative error (AARE) of the prediction of the concentration distribution of the transferred component along the length of the column for ADM and FMM was found to be less than 10%, while for PFM, it was less than 30%.

Study on a new four‐fiber sensor signal processing algorithm of hydrodynamic characteristics in liquid–liquid two‐phase flow

AbstractIn the process of two‐phase countercurrent flow in the extraction column, the study of the hydrodynamic characteristics of dispersed phase droplets is crucial for the design and scaling up of the extraction column. In a pulsed extraction column with a liquid–liquid system, the four‐sensor optical fiber probe was used to assess the hydrodynamic characteristics of the dispersed phase droplets, including size, mean velocity, and holdup. 1 M HNO3 and 30% TBP/kerosene were used as continuous phase and dispersed phase, respectively. The effects of two‐phase superficial velocities and pulsation intensity were investigated on droplet size, mean velocity, and holdup. The findings show that aside from the local holdup, the pulsation intensity has the greatest impact on the hydrodynamic characteristics. A new and concise algorithm was used to deal with the light signals to determine the droplet size and mean velocity. This algorithm created two Cartesian coordinate systems, each with a separate z‐axis using the leading probe as the origin of the coordinate instead of using the vector approach. Compared to the previous algorithm, both the needed number of the time interval between the voltage peaks and that of the direction angles decrease. The relative error of the droplet size obtained by this improved algorithm is within 7%, and droplet velocity is almost the same, compared to the previous algorithm, which indicates that the new algorithm of the four‐sensor optical fiber signal processing is reliable to the measurement of the hydrodynamic characteristics.

Time-resolved fourier transform infrared emission spectroscopy of NH radical in the X3Σ− ground state

The NH radical is an extremely important specie in nitrogen chemical reaction networks, in the interstellar medium and atmospheric chemistry. Time resolved Fourier transform spectroscopy technique in the frequency range 10–13 µm has been applied for the measurement of a pure rotational spectrum of the NH free radical (NH) in the ground X3Σ− electronic state. Twelve high N (26–29) triplet-resolved pure rotation lines of NH were experimentally observed in the laboratory and compared with satellite ACE (Atmospheric Chemistry Experiment) solar data. In addition, discharge-generated vibration-rotation NH radical bands in the spectral range 1923–3571 cm−1 have been measured with a microsecond time resolution and spectral resolution of 0.02 cm−1. The spectra of the NH radical have been studied in two experimental arrangements. Firstly, in a pulsed positive column discharge of pure hydrogen-nitrogen mixture and secondly, in a discharge of nitrogen-ammonia mixture in the presence of argon buffer gas. Both production methods are described and compared. The population analysis of the experimental spectra was performed via modelling using the accurate MoLLIST line lists for NH. It was shown that laboratory data can be well reproduced by use of a mixture of the local-thermal-equilibrium (LTE) and non-LTE models corresponding to high temperatures (up to 8000 K and up to 6000 K rotational).

Predicting Mass Transfer in Liquid–Liquid Extraction Columns

In this work, the GEneralised Multifluid Modelling Approach (GEMMA) is applied to the simulation of liquid–liquid extraction in a Rotating Disc Column (RDC) and a Pulsed Sieve-plate Extraction Column (PSEC). A mass transfer modelling methodology is developed, in which the multiphase flows, droplet size distribution and dispersed phase holdup predicted with computational fluid dynamics are coupled to mass transfer correlations to predict the overall mass transfer. The numerical results for the stage-averaged dispersed phase holdup, Sauter mean droplet diameter and axial solute concentration in the RDC and PSEC agree with experimental observations. The proposed modelling method provides an accurate predictive tool for complex multiphase flows, such as those observed in intensified liquid–liquid extraction, and provides an alternative approach to column design using empirical correlations or pilot plant study.

Modeling the axial distribution of mass transfer coefficient in an agitated‐pulsed extraction column

AbstractThe dispersed phase holdup and drop size in solvent extraction columns vary along the column height and this affects the mass transfer coefficient and interfacial area. In this article, mass transfer study was performed experimentally using a 25 mm diameter agitated pulsed column. The axial distribution of mass transfer coefficient was determined by coupling population balance equation and axial dispersion model by taking the longitudinal variation in hydrodynamic performance into consideration. Feasibility of different mass transfer models in predicting concentration profiles was evaluated and a novel correlation based on effective diffusivity was developed. The results showed that both overall and volumetric mass transfer coefficients have significant change along the column height and greatly depends on the agitation speed and pulsation intensity. Increasing dispersed phase velocity also augments the overall mass transfer coefficient. The maximum number of transfer unit was measured to be 10 m−1 at agitation speed of 1000 rpm.

Estimation of droplet size distribution by using maximum entropy programming and population balance equations in pulsed disc-doughnut column

The pulsed disc-doughnut column is used in various industrial applications. Knowing size distributions of the dispersed phase droplets through this column is necessary to management decisions for better extraction efficiency. To address this, we used various operational parameters, traditional sampling, characterizations, and maximum entropy modeling, and population balance equations to propose species distribution models for the size of droplets in the extraction zone. This paper is devoted studying the size distribution of the dispersed phase using three various liquid feeds in the absence of mass transfer at three operating variables, including the pulse intensity (Af) and flow rates of both liquid phases. The new correlations are proposed to describe Lagrange multipliers of the maximum entropy model regarding operating factors and the physical properties of phases. The population balance model was used to describe the distribution of droplets with main parameters in the breakage and coalescence functions. The comparison appears excellent agreement between these models and the experimental data. The maximum entropy principle has been a straightforward strategy that has successfully described droplet sizes.

ColHySE: An advanced column hydrodynamic-based model for solvent extraction

An original one dimensional population balance model (PBM)-based model of liquid–liquid extraction columns is reported. Compared to existing simulators, ColHySE implements a more realistic description of the flow patterns in the contactor, and predicts its effect on the local droplet–droplet interactions (i.e. breakage and coalescence rates). Proper turbulent properties, extracted from single-phase flow CFD simulations, are used in the source terms of the PBM to evaluate locally the inhomogeneous breakage and coalescence rates, using the averaged Coulaloglou and Tavlarides kernels (Castellano et al., 2018). The sensitivity of the predicted droplets mean diameter, d32, and the holdup, φ, to the parameters of the used empirical and phenomenological models, on the one hand, and to the operating conditions of the column, on the other hand, was studied. Although some model parts must be refined, and an experimental validation remains necessary, the results confirm that the 1D-PBM methodology used in ColHySE is relevant for predicting the interfacial area in the pulsed column as a function of the operating conditions and geometry, hence highlighting its relevance to study the hydrodynamic stability and tendency to flooding. The sensitivity analysis has moreover highlighted the needs for an improved slip velocity model.