Chemical and Biomolecular Engineering - Theses
Permanent URI for this collection
Search Results
1 - 8 of 8
-
ItemStudies on modeling and scale-up of ceramic hybrid pulsed column( 2018)Extraction of lithium from chloride brine in salt lakes is a separation process of great importance. It is also a promising industrial field where solvent extraction columns are likely to be used. However, the extraction solution is corrosive to traditional stainless steel column because of its high chloride content, therefore a type of novel anti-corrosive ceramic hybrid pulsed column is designed and tested in order to be considered for industrial applications, as well as a ceramic standard pulsed sieve plate column for comparison. The hydrodynamics of the two columns are tested under pilot plant conditions. For each column, effects of pulsation intensity and two phase velocities on holdup, characteristic velocity and Sauter mean diameter are investigated. Holdup of ceramic hybrid pulsed column is higher by around 50%. Correlations are proposed to predict holdup for both columns with ARD of 5.9% and 9.3% respectively. Characteristic velocity, which is key parameter in calculating column throughput, is investigated and modelled. Results show that characteristic velocities of ceramic pulsed sieve plate column are larger. Sauter mean diameter of ceramic hybrid pulsed column is smaller by around 30%. Correlations are proposed to predict the Sauter mean diameter, d_32, with ARD of 13.6% and 4.2% respectively. The higher holdup and smaller d_32 in the ceramic hybrid pulsed column contribute to larger mass transfer area and hence better mass transfer efficiency. Axial dispersion and mass transfer parameters, which are important to determine column height, are tested with a standard liquid-liquid system with medium interfacial tension, 30% TBP in Shellsol 2046–water with acetic acid as solute. Axial dispersion coefficients, E_c, of the ceramic hybrid pulsed column are less by around 50%. Effects of pulsation and two phase velocities on E_c are investigated. Correlations are proposed to predict E_c with ARD of 6.0% and 6.9% respectively. The height of overall mass transfer unit, H_ocp, of the ceramic hybrid pulsed column is less by around 40%, indicating better mass transfer efficiency. Effects of pulsation and two phase velocities on H_ocp are investigated. Volumetric mass transfer coefficients, K_oc a, for both columns are calculated. K_oc a of the ceramic hybrid pulsed column is higher by around 50%. Correlations are proposed to predict K_ox a with ARD of 12.0% and 7.0% respectively. Under proper operating condition, height of transfer unit of ceramic hybrid pulsed column can reach as low as 0.2m, showing very good efficiency. Two-phase computational fluid dynamics (CFD) models for the two columns are developed with commercial software ANSYS FLUENT. Hydrodynamic performance including two-phase distribution and velocity fields are generated. Holdup and axial dispersion coefficients are studied systematically, and CFD successfully predicts the higher holdup and lower axial dispersion coefficients for the ceramic hybrid pulsed column in experiments. CFD gives reasonable predictions for the trends of holdup and axial dispersion coefficients with pulsation intensity and two phase velocities. Predicted axial dispersion coefficients with this method are accurate. However the breakage and coalescence phenomenon of drop swarms is so complicated that the development of more accurate prediction method for holdup based on population balance model (PBM) still needs further research. A pilot ceramic hybrid pulsed column for lithium extraction from salt lake brine is designed for industrial applications. Physical properties are tested for the real system from Qinghai province, China. A MATLAB program is coded under the framework of design from first principles for the demonstration column, and its applicability is validated with experimental data. Program simulations are performed to investigate the effects of key operating parameters including two phase flow ratio and pulsation intensity. With considerations of volumetric efficiency and solvent reprocessing cost, these two parameters are determined to be V_c:V_d=1:1 and Af=1.0cm/s respectively. The demonstration column diameter and height are determined with the MATLAB program, under the operating conditions determined previously. To reach an annual yield of 10 tons/year and 99% lithium recovery in the extraction section, the demonstration column needs to have an effective height of 1.8m and an internal diameter of 0.36m.
-
ItemA fundamental study of emulsions formed in a hexane-based lipid extraction from slurries of ruptured microalgae( 2018)Efficient lipid recovery is a major barrier to economical production of algal biofuels. The use of non-polar solvents is promising, as they can be recovered via centrifugation, avoiding energy-intensive evaporation of the water phase. Emulsions are central to this process. However, there is currently little understanding of the emulsion properties and how they relate to extraction and separation. This thesis investigated the fundamental physical mechanisms underlying the formation and subsequent destablisation of these emulsions in a hexane-based lipid extraction process.
-
ItemSynergistic solvent extraction of rare earth elements( 2017)The separation of the rare earth elements is a difficult task owing to the chemical similarity of the lanthanides. The method most commonly utilised industrially is solvent extraction using EHEHPA (2-ethylhexyl phosphonic acid, mono-2-ethylhexyl ester). Due to the relatively small separation factors between adjacent lanthanides, a large number of equilibrium stages are required for the separation and a considerable amount of acid and base is consumed. In addition, recovery of the heavier rare earths from the solvent at the end of the process (stripping) requires quite high acid concentrations. In this work, mixed or synergistic extractants were examined to attempt to improve upon the conventional process using EHEHPA alone. Three classes of mixtures were studied using a combination of slope analysis and spectroscopy at process relevant extractant and metal concentrations. The first two systems, a mixture of EHEHPA with Cyanex 272 (bis-2,4,4-trimethylpentyl phosphinic acid) and the quaternary ammonium phosphonate ionic liquid R4N+EHEHP-, revealed antagonistic interactions that result in a lower stripping acidity, but also reduced loading capacity. 31P1H NMR spectroscopy showed that a mixed complex is formed with EHEHPA and Cyanex 272, but antagonistic interactions between the extractants are more significant for extraction results. For R4N+EHEHP-, the spectroscopic studies indicated that the extracted complex is exactly consistent with the corresponding EHEHPA-rare earth species. The quaternary ammonium ion moderates the rare earth extraction behaviour of EHEHPA by ion pairing with the phosphonate ion in a pH dependent interaction. The extraction results and spectroscopy for R4N+EHEHP- are consistent with a simple mixture of the quaternary ammonium chloride and EHEHPA once the organic phases are equilibrated at the same pH, in contrast with some reports in the literature. For the third and final class, two carboxylic acid containing mixtures were examined: CA-12 (sec-octylphenoxyacetic acid) with Cyanex 272, and a novel mixture of naphthenic acid and CA-12. The separation factors are largely determined by those of the components of the solvent mixture, and can be tailored somewhat by changing their proportion in the organic phase. The naphthenic acid/CA-12 solvent was therefore found to exhibit a promising set of separation factors for the extraction of lanthanides from yttrium, as yttrium is among the least extracted rare earth by both solvents. However, at elevated loading the separation factors revert to those of naphthenic acid alone, due to its considerably larger concentration. An analysis of the predicted capital and operating costs for several industrially relevant separations demonstrated that for most separations, the saving in stripping acid for the phosphonic/phosphinic acid mixture and R4N+EHEHP- is outweighed by an increase in equipment size due to the reduced loading capacity. However, for the separation of the heaviest elements, Lu and Yb, R4N+EHEHP- was found to be the best performing option due to the large saving in stripping acid consumption. The work shows that separation factors for the mixed extractants can largely be predicted from those of the individual components, while stripping acidity cost savings are only significant for the heaviest rare earths. New extractants are therefore required to make a significant improvement upon the conventional process, and this is suggested as the way forward for future work.
-
ItemMulti-scale interface-capturing methods for thin-film coalescence( 2016)Solvent extraction units are core separation systems in the metals, pharmaceutical and nuclear reprocessing industries. In Australia, extraction units are widely used for the separation of base metals and rare-earth elements from their leached ores. A throughput-limiting process is the efficient separation of raffinate and loaded organic phases via coalescence of dispersed droplets. New research findings are needed in order to quantify the physical and chemical regimes that favour coalescence, which will aid in the design and optimisation of hydrometallurgical processing equipment; with both economic and environmental benefits. Coalescence remains difficult to model due to the disparate length scales involved: a successful model must capture both droplet and film-scale dynamics. While multiphase flows have been well simulated by interface-capturing methods, the detail of thin-film drainage is difficult to simultaneously resolve using these techniques alone. Conversely, existing planar-interface drainage models are not accurate when applied in emulsion settings. This thesis demonstrates the predictive power of multi-scale simulation methodologies, whereby droplet-scale interface-capturing techniques are coupled to a lubrication analysis of thin-film drainage. The developed multi-scale interface-capturing (MSIC) method is applied to droplet/wall and droplet/droplet interactions in emulsion systems. The thesis details the implementation of appropriate interface-capturing methods, together with coupled film equations, using an open-source finite-volume solver. Level-set, volume-of-fluid (VOF) and high-order VOF techniques are detailed for droplet-scale interface motion; while a new method is presented for the automated discretisation of the required finite-difference drainage relations. The approach is modular, in that alternative interface-capturing techniques can be coupled to the same base film equations. For validation purposes, MSIC model results were compared with existing Stokes–Reynolds–Young–Laplace theory in the low-inertia limit. Quantitative agreement was achieved for both static and dynamic response scenarios. The model was then used to simulate free-droplet collisions in emulsion systems under moderate-inertia conditions. Results are consistent with existing experimental data for droplet/wall interactions, while outcome predictions for binary droplet collisions agree with existing high-speed video sequences in the presence of continuous-phase electrolytes. Modelling of emulsion interactions requires an accurate description of repulsive electrical-double-layer (EDL) forces; as such, a multiphase ion-transport model was used to confirm the validity of existing disjoining pressure treatments under low-inertia conditions. In the studied case, however, it is the action of interface immobilisation which promotes a bouncing outcome, and not the presence of EDL repulsion. The techniques presented are important for the development of coalescence kernels in macro-scale population balance equation models. Physically accurate coalescence models can be used to generate comprehensive outcome regime maps, with accompanying quantitative drainage-time data. The MSIC model can be used to discern the influence of continuous-phase chemistry on outcome regime boundaries, which is yet to be studied experimentally. Though the present implementation is restricted to axisymmetric collisions, the model can be generalised for the future study of three-dimensional interactions.
-
ItemInfluence of plate wettability on the performance of a pulsed disc and doughnut column( 2015)The pulsed disc and doughnut column (PDDC) can be seen as a variation of the pulsed perforated plate column (PPPC). It has gained attention due to its simple design, compact nature and lack of internal moving parts leading to safety and economic benefits. The PPPC hydrodynamics and mass transfer performance has been widely studied and reported and due to its similarity, applied to PDDC in a few studies. A few investigators have reported the effect of plate wettability on the performance of PPPC with contradictory results. In this study a tri-n-octylamine – sulphuric acid – water system is used in the study of the hydrodynamics and mass transfer performance of a pilot plant PDDC with mass transfer from dispersed phase to continuous phase and with both aqueous and organic dispersions at low, intermediate and high agitation intensities. To simulate different plate wettability three materials with distinct wetting properties has been selected, namely: Teflon (hydrophobic), Nylon (intermediate) and Stainless Steel (hydrophilic). It was found that Teflon behaves as a super-hydrophobic material in the liquid-liquid system used and may cause phase inversion for organic dispersion with insufficient pulsation. It was found that the magnitude of the effect of agitation intensity on holdup is influenced by the plate wettability. Holdup, measured in terms of characteristic velocity, was lower for Teflon and higher for Stainless Steel for aqueous dispersion. For organic dispersion Nylon showed the lowest characteristic velocity and Stainless Steel the highest. The effect of agitation intensity on the characteristic velocity from the low to intermediate agitation intensity was similar for intermediate and non-wetting plates in both organic and aqueous dispersions, but has a little effect on the wetting plate, i.e. Teflon with dispersed organic and Stainless Steel with dispersed aqueous. Despite the changes in the holdup, the Sauter mean diameter was independent of the plate wettability for dispersed aqueous. However, the distribution of droplet size changed with the operating parameters and three distribution functions were fitted to experimental data, namely: Lognormal, Gamma and Weibull (also known as Rosin-Rammer). The later fitted the raw and grouped data the best. For dispersed organic, the drop size in Nylon plates was significantly smaller than that of Teflon plates. The measured holdup and droplet size was compared with existing models and their accuracy assessed. Two holdup models were modified to incorporate the wetting characteristic of the plate materials. The holdup and droplet size models, suitable for this system, were used to calculate the mass transfer coefficient based on the axial dispersion model which was also compared with existing models and a new correlation developed.
-
ItemThermodynamic modelling of liquid–liquid equilibria using the nonrandom two-liquid model and its applications( 2015)Solvent extraction is a separation technique widely used in a variety of industrial applications. The basis of separation by this technique is the distribution of a solute between two immiscible solvents, which fundamentally is a phenomenon of thermodynamic phase equilibrium. As a result, the thermodynamic modelling of liquid–liquid equilibria (LLE) is a significant problem of solvent extraction. In general, there are two approaches to calculate phase equilibrium: minimizing the Gibbs free energy combined with the Tangent Plane Distance (TPD) criterion for stability test and solving the isoactivity equations. Compared with the first approach, the second is easier, however, it strongly depends on initial estimation and may lead to erroneous results which correspond to maxima, local minima and saddle points of the Gibbs free energy. Therefore, the primary aim of this thesis is to understand the solution structure of the isoactivity equations of LLE and develop a procedure to determine the correct, physically realistic solution. This thesis has three parts: firstly, understanding the isoactivity equations of LLE using the nonrandom two-liquid (NRTL) model, the most popular thermodynamic model; secondly, regression of NRTL parameters using particle swarm optimization (PSO) and insights into the model’s capabilities in correlating LLE data; thirdly, application of the symmetric eNRTL model and the developed PSO method to the modelling of phenol extraction. The solution structure of the isoactivity equations for ternary and quaternary LLE systems using the NRTL model under two types of mass balance constraints were investigated. The first constraint specifies the concentration of components (one component in a ternary system and two components in a quaternary system) in one phase. In this case, the three isoactivity equations of a ternary LLE system were presented in a three dimensional space as three surfaces with their intersection lines extracted. Three types of solutions were revealed, namely exact solutions, symmetric solutions and approximate solutions. These analyses were called Solution Structure Categorization (SSC). Results yielded by SSC further led to development of a procedure to identify the correct solution of LLE for ternary and quaternary systems. The second constraint specifies the total amount of each component in a system. In this case, the SSC method was again applied and it was found that all solutions of isoactivity equations can be categorized into two types when converted into mole fractions: one correct solution and a number of symmetric solutions representing a homogeneous phase. A procedure based on solving isoactivity equations to determine the correct solution was also proposed, which was shown to be simple and effective for a number of ternary and quaternary LLE systems from a wide range of literature sources. The new procedure is recommended to be used as a parallel procedure to minimization of Gibbs free energy combined with stability test by the TPD criterion. The NRTL model has binary interaction parameters and non-randomness parameters that need to be regressed before the model can be used. The particle swarm optimization (PSO) method was successfully used to regress the NRTL parameters from liquid–liquid equilibria (LLE) data and the resulting parameters showed smaller root-mean square deviations (RMSD) compared with literature values. Analysis of the results revealed that multiple groups of parameters with sufficiently small RMSDs can be found for the same set of LLE data. The activities calculated using these parameters and their corresponding predicted mole fractions can be far beyond the reasonable range of activity, demonstrating that the NRTL model does not always represent the intrinsic activities of components with these parameters. Finally, extraction of phenol by toluene in the presence of sodium hydroxide was investigated with varying pH and varying concentration of sodium hydroxide to mimic extraction of alkaloids as acidity constant of phenol is close to that of many alkaloids, for example morphine. The phase equilibrium was modelled by the symmetric eNRTL model using the developed PSO method and the correlation agreed well with the experimental results.
-
ItemThe recovery of zinc salts from hot-dip galvanizing effluent streams( 2012)Hot-dip galvanizing is used to coat a range of steel products with a protective layer of zinc. There is a growing interest in the recovery of metals and hydrochloric acid recycling from spent pickling liquor of these hot-dip galvanizing plants. This research aims to explore and develop a process for zinc recovery from such waste using solvent extraction with tri-n-butyl phosphate (TBP) and di-2-ethylhexyl phosphoric acid as extractants (D2EHPA). The proposed recovery of zinc from spent pickling liquor involves two solvent extraction stages: first to selectively recover zinc chloride from the waste using TBP, and second to extract zinc cations into a sulphate media using D2EHPA. The ultimate aim is to produce a zinc sulphate solution that meets the requirements for sale to electrowinning plants. HCl and ZnCl2 extraction from model solutions (0–120 g/L ZnCl2, 0.5–2 mol/L HCl, 5 mol/L total Cl) using TBP diluted in ShellSol 2046 (30–70 vol% TBP) was investigated. Experimental results showed that HCl and ZnCl2 are extracted simultaneously by TBP; in the presence of zinc, HCl extraction increased, likely to be due to the extraction of acido-metal complexes (HZnCl3 or H2ZnCl4). However, zinc is still predominantly extracted as ZnCl2. The effect of dilute acid or alkali in the stripping solution was found to have a negligible effect on zinc and HCl back-extraction from loaded TBP; the presence of chloride ion in the stripping solution yielded a greater effect on the stripping of zinc and HCl. The production of zinc sulphate solution with low chloride content can be achieved using the proposed process provided that the entrained aqueous phase is scrubbed from the organic phase in the second stage solvent extraction. A model of the HCl-ZnCl2-TBP-diluent system is presented using the intrinsic equilibrium constant of each reaction. The Pitzer and Hildebrand-Scott models were used to estimate the aqueous and organic phase activity coefficients respectively. The equilibrium constants and solubility parameters of the extracted species were then determined from experimental and literature data. The model was able to predict co-extraction of water, HCl and ZnCl2 by TBP diluted in ShellSol 2046 up to 2 mol/L Zn (130 g/L) and 1 mol/L HCl. A model for the Zn-D2EHPA system reported in the literature was examined and found to agree with experimental results available. The model developed was tested against HCl and ZnCl2 extraction by TBP from spent pickling liquor obtained from a hot-dip galvanizing plant. Modelling results obtained were promising given that the Pitzer mixing parameters for Zn-Fe(II) were not available and had to be estimated. The extraction of HCl was found to be enhanced due to the high Fe(II) concentration in the spent pickling liquor. This increase can be attributed to an increase in acido-metal extraction. Simulation of the two stage extraction process using the model indicated that approximately 80 g/L Zn can be extracted from a feed solution containing 120 g/L Zn and 1 mol/L HCl. The modelling was also able to predict both ZnCl2 and HCl extraction using a hollow fiber membrane contactor; no additional fitted parameters were used in this modelling. The results of this simulation indicate that a contactor of 3.6 m length is required to achieve 99 % zinc extraction.
-
ItemThe use of a Y-Y shaped microfluidic device for the study of solvent extraction kinetics( 2012)Solvent extraction (SX) is usually defined as the process of transferring a substance from one liquid to another liquid phase for the purpose of purification and/or concentration. The commercial importance of SX processes is widely recognised in the metallurgical, pharmaceutical, “food and beverage”, petrochemical, nuclear, catalysis, polymer and material industry. SX is often considered to be the most efficient method of separating valuable products from complex liquid mixtures. Thus, SX is widely exploited in the production of precious metals and fine chemical products and constitutes one of the most important unit operations in the chemical industry. SX systems have been investigated under a variety of experimental conditions in both equilibrium and kinetic studies. Kinetics is important since it is intimately related to the mechanism of extraction, it has a major role in the choice of industrial operation conditions and it ultimately determines the size of the contacting equipment. Today a number of short residence time contactors are being proposed in which kinetics of extraction are becoming more important. However, due to the heterogeneous nature of these reactions, kinetic measurements are difficult. Therefore, mechanisms of extraction are not well understood. Indeed, a long term debate about the actual site of the extraction reaction seems still far from being concluded. Depending on the technique exploited for the kinetic study, different authors have often reported different mechanisms and/or contrasting conclusions. In order to thoroughly understand the mechanism of SX reactions, new tools to overcome limitations of traditional extraction kinetics techniques are necessary. Microfluidics is potentially one of these innovative tools. Microfluidics, the science and the technology that deals with the manipulation of small amount of fluids in microchannels, has proved to be a revolutionary tool in many fields of chemistry, physics and engineering. A few studies have already demonstrated the possibility to implement analytical techniques on microfluidic platforms. These include chromatography, electrophoresis as well as SX. The first report on SX procedures integrated on a microfluidic device was reported by Kitamori and co-workers in 2000. Since then, studies in this field have mushroomed in the literature. However, despite potential advantages that microfluidics could offer such as the high contact area per unit volume, the high throughput per total system volume and a significant reduction in the chemical quantities, some major complications also arise in the microenvironment. The fluid flow control is difficult and the full understanding of physical and chemical phenomena at this scale is still yet to be achieved. Consequently, very few studies on SX mechanistic have been attempted in a microdevice. In this thesis, the use of a Y-Y shaped microfluidic device for the study of solvent extraction kinetics relevant for the hydrometallurgical industry is demonstrated. The following contributions are discussed: i) a homogeneous (aq/aq) diffusion study of the Co(II) ion is carried out, ii) an improved stabilisation of a water/oil interface in the Y-Y shaped microchannel is achieved by coupling a guide structure with a glass wall surface treatment, iii) the heterogeneous diffusion of a probe molecule (8-hydroxyquinoline) is investigated and thoroughly explained in light of different mass transport models, each corresponding to different levels of approximation, iv) the extraction of Co(II) and Fe(III) by di (2-ethylhexyl) phosphoric acid (DEHPA) is performed in the microchannel and modelled; the extraction of Co(II) as well as that of Fe(III) are shown to occur in a regime controlled by the reaction kinetics. Consequences on mechanistic aspects are discussed and v) a prototype microfluidic device that integrated a UV-Vis detection system is also proposed. The functioning of the prototype is demonstrated and preliminary results on in situ detection of concentrated species are reported. Applications of the device to study concentrated SX systems such as those encountered in the hydrometallurgical industry are presented.