Showing papers in "Chemical Engineering Research & Design in 1994"

Review of Potential Technologies for the Removal of Dissolved Components from Produced Water

Abstract: A wide range of technologies was assessed for both oil (relatively high produced water flowrates) and gas fields (relatively low produced water flowrates). Technologies that were felt to offer the greatest potential (based on current stages of development) were found to be ion exchange for heavy metal removal, and air (or gas/vapour) stripping, activated carbon adsorption (with regeneration of the carbon by wet air oxidation) and biological treatment for dissolved organic removal. In gas fields, wet air oxidation could be used without the activated carbon adsorption step. However, none of the systems evaluated would be capable of removing all of the groups of dissolved components from produced water alone. This implies that, if removal of all dissolved components is required, 2 or more systems would have to be used in series. In practice, technology selection could be based on the toxicity reduction achieved and/or the organic loading reduction achieved (of the technologies assessed, only biological oxidation and wet air oxidation achieved significant organic loading reduction. (author)

A computational model for the gas-liquid flow in stirred reactors

Abstract: A model is described for calculating the gas-liquid flow in stirred vessels. The general fluid flow code FLUENT is used for calculating the single-phase flow pattern. This flow pattern is used as input for an in-house code named GHOST! which calculates the distribution of the gas over the vessel on the basis of balance equations. A mathematical model for bubble break-up and bubble-coalescence, based on local turbulence intensity and local energy dissipation rate, is incorporated in this code. Details regarding modelling the impeller, bubble coalescence and bubble break-up are given. The GHOST! code is capable of calculating local void fraction, local bubble size, local interfacial area and local mass transfer. These local values can be integrated to yield the overall gas holdup and the overall mass transfer rate. There is a good agreement between computational results and measurements. Based on the simulations, it is concluded that full homogeneity of the gas-liquid mixture will never be achieved. This knowledge should be used in the optimization process

Terminal bubble rise velocity in liquids

Abstract: We present new experimental results of terminal bubble rise velocity for a variety of fluids and a model which accurately predicts the bubble rise velocity over a wide range of bubble sizes. A comparison is made between experimental results obtained in the investigation as well as by other authors and the predictions of the correlations of Clift et al. (1978), Wallis (1974) and the present authors

Mass exchange networks for waste minimization : a simultaneous approach : Process design

Abstract: The synthesis problem of mass-exchange networks is addressed, where a number of process streams, rich in terms of certain components (typically pollutants) are integrated with lean process or utility streams in order to meet process specifications (e.g. environmental regulations) on their final compositions. A mathematical programming approach is adopted, based on a hyperstructure representation of the mass exchange network. The synthesis problem is then formulated as a mixed integer nonlinear programming optimization problem (MINLP), where both network operating and investment cost are optimized simultaneously. In contrast to all previously published work that simplifies the problem by assuming decomposition based on the concept of pinch, we treat the synthesis of mass exchange networks without decomposition. In the first part, the case of a single component is examined, whereas mass exchange in multiple components can be handled in a straightforward manner. In the second part the approach is applied to reactive mass exchange networks, where chemical sorption takes place. In the third part, the mass exchange hyperstructure is extended to include regeneration of the lean streams. A number of examples illustrate the applicability of the proposed approach to several problems of waste minimization and demonstrate the impact of simultaneously minimizing operating and investment cost

The Effects of Air Inlet Geometry and Spray Cone Angle on the Wall Deposition Rate in Spray Dryers

Abstract: The use of a numerical simulation to minimise the wall deposition rate in a spray dryer is demonstrated, with the numerical simulation solving the equation of continuity and the Navier-Stokes equations inside the dryer using the k-e model for turbulence. To validate the model, a solution of sodium chloride containing 20% by mass of the salt has been sprayed at the rate of 0.0012 kg/s from a two-fluid nozzle into a 0.935 m diameter, 1.69 m high cylinder-on-cone chamber. The simulation has predicted the wall deposition rate (measured as 0.000044 kg/s) within 16% at an inlet air temperature of 245°C when turbulence constants in the k-e model for recirculating flows as recommended by Abujelala and Lilley (1984) have been employed. This numerical simulation has been used to explore methods for decreasing the wall deposition rate, including simple modifications to the air inlet geometry (to eliminate swirl in the inlet air) and a reduction in the spray cone angle from 60 to 45°. Within the constraints imposed by the experimental equipment, we have suggested that the maximum spray cone angle (60°) and the maximum amount of swirl in the inlet air (62°) tend to minimise the wall deposition rate. The measured trends in the wall deposition rate caused by decreasing the amount of swirl in the inlet air (0.000093 kg/s measured, an increase) and the included angle of the spray cone (0.000099 kg/s measured, an increase) have been predicted by the simulation, which suggested deposition rates of 0.000053 kg/s for the no-swirl case and 0.000056 kg/s for the reduction in the spray cone angle respectively. We demonstrate the potential for using this type of numerical simulation to refine the operation of spray dryers

Artificial Neural Networks - Studies in-Process Modeling and Control

Abstract: In the mid 1980s wide spread interest in Artificial Neural Network research re-emerged following a period of reduced research funding. The much wider availability and power of computing systems, together with new research studies, resulted in a far greater market for the technology. The seeds were sown for claims by some that the technique provided much sought after pragmatic solutions, and by others that it provided a panacea to all complex modelling problems. Unlike ARMA, NARMA and multivariate statistical modelling approaches the methodology has been attributed the potential of accurately describing the behaviour of extremely complex systems. But is the approach so different? Should we not consider the concept of Neural Networks as being an integral part of system representation, modelling and identification? In this respect perhaps we really do have an established, but still developing, theory and technology represented within a new framework. Indeed, selling the technology in anything but this manner might discredit what could well prove to be a valuable engineering tool. We examine the contribution that various network methodologies can make to the process modelling and control toolbox. Feedforward networks with sigmoidal activation functions, radial basis function networks and autoassociative networks are reviewed and studied using data from industrial processes. Finally, the concept of dynamic networks is introduced with an example of nonlinear predictive control

Single-phase flow in stirred reactors

Abstract: Single-phase flow patterns in stirred reactors were investigated both computational, with FLUENT, and experimental, by Laser Doppler Velocimetry (LDV). This was done for three different impeller types, viz. two axial flow impellers and a disc-turbine. The effect of geometrical modifications to the vessel geometry was studied as well. In particular, the exact baffle arrangement has a strong impact on the flow pattern. It will be shown that the single-phase flow patterns exhibit unexpected features not recognized before. An accurate numerical computation requires full three dimensional grids. Further, anisotropic turbulence models, in this case the Algebraic Stress Model, give better predictions than the k-e turbulence model. When these requirements are met, the predicted flow patterns compare very well with the experimental data. Thus, a quick assessment of the influence of geometrical variables can be made just using numerical simulations

Hydrate equilibrium data of methyl cyclo-pentane with methane or nitrogen

Abstract: It has been demonstrated experimentally that methyl cyclo-pentane can form gas hydrates. The first hydrate equilibrium data on the newly discovered hydrate structure have been determined over the seabed temperature range (274-288 K). In addition to methane, nitrogen For the first time has been used as help gas to fill small cavities of the above structure. The new structure is more stable than those formed by the light gases, and hence reduces their hydrate free zones

A Froth Based Flotation Kinetic-Model

Abstract: A kinetic model for the prediction of the performance of a froth flotation cell is proposed. The model is based on the assumption of a free-flowing concentrate froth. The flux of bubble surface is estimated from the cell aeration rate and the specific surface of the bubbles in the froth. The specific surface in the froth is based on the mean bubble size which is estimated by off-line image processing of photographs taken of the froth during laboratory scale batch tests. A sequence of batch flotation tests in which a low-rank UK coal was demineralized, by varying the initial addition of Sodium Dodecyl Sulphate (SDS), was used to test the consistency of the model predictions with experimental data. Although all the model parameters have a physical reality, not all were able to be measured at this stage. Nevertheless, the consistency of the predictions with the experimental data was encouraging and the potential value of extended on-line image processing as a tool for improved experimentation was apparent

Understanding flux decline in crossflow microfiltration. Part 2 - Effects of process parameters

Abstract: Further results from an experimental study of membrane fouling and permeate flux decline during crossflow microfiltration are presented. A computer controlled microfilter and a variety of well characterised particulate solids and polymeric membranes were used to acquire a range of data over typical operating conditions. Example data highlight influences of the process parameters filtration pressure, crossflow velocity, suspension concentration, and particle surface charge, and demonstrate the interdependence of the process operating conditions with particle size, size distribution and shape. Many of the results obtained are discussed with respect to existing literature data which are apparently contradictory, but the current data provides explanations for these contradictions and enable conclusions to be drawn.

Estimation of void fraction and pressure drop for two-phase flow in fine passages

Abstract: The void fractions and frictional pressure drops occurring in two-phase air/liquid flows in smooth-walled, circular-section ducts with diameters from 0.74 to 3.07 mm have been measured for vertical and horizontal flows over a wide range of conditions: 0.05 1000 and it is concluded that fine passage dimensions are do not of themselves present any problems to these correlations under turbulent conditions. For Re L <1000, a modification of the Beattie correlation is recommended. Both the Lockhart-Martinelli and the more complex CISE correlations for void fraction describe the measured void fractions satisfactorily over the whole range of conditions studied

Simulation of Non-Newtonian Flow in a Hydrocyclone

Abstract: The results of a numerical simulation of a non-Newtonian flow of a shear thinning fluid inside a hydrocyclone are presented. The fluids used were carboxymethylcellulose polymer solutions whose rheological properties were described using a power law shear-rate dependence. This system is a model analogue for drilling fluids. An axisymmetrical, laminar swirl flow was analysed by solving a set of conservation equations in which the air core radius was explicitly included, employing a surface-tension force balance equation. The results predicted a distinct difference in velocity distribution within the hydrocyclone between a shear-thinning fluid compared to a simple Newtonian fluid. The predicted non-Newtonian fluid velocity agreed well with independent flow and velocity measurements obtained from Laser Doppler Anemometry. The thickness of the vortex finder is shown to have a significant effect on the flow pattern in the hydrocyclone, and this is likely to affect the classification efficiency. This result has important implications for the design and selection of hydrocyclones for handling non-Newtonian mixtures

Gas-liquid contacting with axial flow impellers

Abstract: We focuse on the gas-dispersion characteristics of axial flow impellers. Various impellers are compared: an inclined blade impeller with flat blades, an A315 impeller and a Leeuwrik impeller. The goal is to gain insight in the gas-dispersion properties of such axial flow impellers. Therefore, the effects of impeller position and sparger geometry on flow pattern, power consumption and mass transfer are studied in a vessel 0.444 m in diameter, using liquids of different viscosity. The power consumption is strongly related to the different flow patterns and different types of cavities. Under certain conditions time-dependent, asymmetrical flow patterns can occur. The sparger type influences both the power needed to prevent flooding and the overall mass transfer coefficient k l a. The liquid viscosity does not affect the basic hydrodynamic phenomena in the vessel for viscosities not exceeding 80 m Pa s

Heat transfer across the wall of dividing wall columns

Abstract: We address heat transfer across the wall of the dividing wall column. Heat flows through the column are analysed and used to identify regions of the wall at which the horizontal heat transfer has beneficial and detrimental effects. The results of a case study show that energy savings of the order of 10% are possible by installing insulation at the appropriate part of the dividing wall

Flow generated by a disc turbine. IV: Multiple impellers

Abstract: The three dimensional turbulent flow generated by multiple impellers was measured using a laser Doppler anemometer. The tank diameter was 300 mm with a flat bottom and provided with four T/10 baffles. Two impellers were mounted on the same shaft. A standard Rushton disc turbine (DT) and a pitched blade downflow turbine (PTD) were used as impellers. Two impeller combinations were studied: (1) DT-DT and (2) DT-PTD. The impeller diameter was 100 mm and the impeller speed was 5 r/s in all the cases. The impeller clearance from the bottom and the distance between the two impellers were varied over a wide range. The effect of these variables on the flow pattern has been presented. Energy balance has been established and the values of hydraulic efficiency and pumping effectiveness have been reported.

Model-based fuzzy controller

Abstract: Most fuzzy controllers developed to date have been of the rule-based type, where the rules in the controller attempt to model the operators response to particular process situations. These controllers require considerable 'knowledge engineering' in that someone has to gather a collection of rules from knowledgeable operators and then condense them into a consistent rule-base for the controller. An alternative approach to using fuzzy logic in a controller is described in this paper. Instead of attempting to model the operator's decision making process, this controller design uses a fuzzy model of the process itself and imbeds this in a relatively conventional model-based controller. The paper also describes two tests of the controller design. The first is a simple level control simulation, and the second is the temperature control of a laboratory heat exchanger. The results of this work indicate that the fuzzy-model-based controller described here can equal or even exceed the performance of more traditional control techniques, even on quite simple processes.

The use of a partially simulated exothermic (parsex) reactor for experimental testing of control algorithms : Process operation and control

Abstract: A configuration has been devised which allows for the testing of estimation and control algorithms in an experimental system which realistically simulates an exothermic reaction taking place in a batch or continuous stirred reactor. In the PARSEX (Partially Simulated Exothermic) reactor, flows, levels and temperatures are measured experimentally and live steam is injected at a rate calculated to simulate the exothermicity of the reaction. Temperature control of a batch reactor has been studied using Generic Model Control (GMC). Since, in most industrial applications, the state variables required for controller implementation are not all measurable or, not with sufficient accuracy for control purposes, a state estimation technique has been applied as well. The amount of heat released by the reactions has been estimated online using an extended Kalman filter, and incorporated into the GMC algorithm. Simulation results had shown that the Kalman filter gave an accurate estimate of the amount of heat released and together with the GMC controller, gave reliable robust control. An experimental extension of the work using the PARSEX reactor shows that the extended Kalman filter is rather more sensitive to plant/model mismatch than would have been predicted from simulations alone; these directly affect the performance of the GMC controller

Sauter Mean and Maximum Bubble Diameters in Aerated Stirred Vessels

Abstract: A semi-empirical correlation, based on the concepts of a critical Weber number and Kolmogoroff's theory of isotropic turbulence, was suggested by Parthasarathy et al. (1991) to estimate the bubble Sauter mean diameter d 32 in aerated stirred vessels, in a non-coalescing environment. The correlation was shown to be sufficiently reliable using experimental bubble size data for two-radial flow impellers in a 0.195 m diameter tank. The validity of the correlation is corroborated further, using a range of impellers, in standard stirred tanks of two sizes. The maximum bubble diameter d max , which is defined as the diameter that is larger than 99% of all the diameters in the cumulative number distribution of bubbles, is found to be linearly related to d 32 . The d 32 /d max ratio is shown to have an average value of 0.785, and is independent of the level of agitation

Prediction of Drop Size, Dispersed-Phase Holdup, Slip Velocity, and Limiting Throughputs in Packed Extraction Columns

Abstract: Empirical correlations for the prediction of drop size, dispersed-phase holdup, slip velocity, and maximum throughputs in packed columns are presented. Published experimental results obtained with both random and ordered packings are considered. The drop-size correlation, which is based on measurements from 376 runs with and without mass transfer from nine different sources for conditions when the continuous phase wets the packing, reproduces the data with an average absolute value of the relative deviation of 15.7%. A large bank data (2023 points with and without mass transfer from 10 different groups of investigators; continuous-phase wetting; d p >d p,cr for random packings; Φ<Φ f ) has been used to develop a correlation for the prediction of dispersed-phase holdup. The same body of data has also been used to derive a correlation for slip velocity without using holdup. By using the correlation for dispersed-phase holdup, the average absolute values of the relative deviation in holdup and slip velocity are 18.7 and 16.4%, respectively. The corresponding figures for the slip-velocity correlation are 20.1 and 15.6%, respectively. On the basis of data from 845 measurements with both continuous-phase and dispersed-phase wetting, an equation for maximum throughputs is derived which reproduces the data with an average absolute value of the relative deviation of 19.5%

A new process for oxygen generation step for the hydrogen producing sulphur-iodine thermochemical cycle

Abstract: This study presents a new design, and thermodynamic and engineering analyses of the H 2 SO 4 decomposition section of the sulphur-iodine thermochemical cycle for producing hydrogen. Excess oxygen is used as an energy vector in various direct contact adiabatic equipment and shell and tube heat exchangers are eliminated as much as possible. Thermodynamic (energy and exergy) and cost analyses have been carried out. The results show that energetic and exergetic efficiencies are 64.2% and 64.0% respectively and typical cost is 3.3$(1990) per kmol SO 2 for 4$US(1990)/GJ nuclear heat cost

Visualisation of cake formation in crossflow microfiltration : Separation processes

Abstract: A high speed video camera was used to record the behaviour of micron-sized particles in a crossflow microfiltration unit. The visualisations were carried out at conditions representative of industrial crossflow microfilters. The dynamics of single particles and of suspensions accumulating at the membrane surface, and the subsequent behaviour of the cake layer formed, were observed. The effect of parameters such as particle size, feed suspension concentration, pH, transmembrane pressure and crossflow velocity on the development of the particle layer and the corresponding permeate flux was investigated. It is shown that, for feed suspensions containing only particulate matter, cake formation is substantially avoided when the shear stress in the crossflow stream is greater than about 10 -6 times the pressure loss across the cake which would otherwise have formed

Process Intensification - Polymer Film Compact Heat-Exchanger (Pfche)

Abstract: During the last decade there have been significant advances in the development and application of metal versions of compact heat exchangers. In parallel with this trend, high performance polymer films have become commercially available. These materials (eg, polyether ether ketone and polyimide) have working temperatures in excess of 250 o C and exhibit excellent resistance to most process fluids. It was therefore recognised that this new generation of polymers could play a key role in fostering the development of a new class of compact heat exchangers which could be cheaper, lighter and more corrosion resistant than their metal equivalents

Drop-size distribution and average drop size in a Wirz extraction column : Separation processes

Abstract: Drop-size distributions and volume-surface average diameters have been measured in a Wirz-II agitated extraction column for six different liquid-liquid systems as a function of phase flow rates and intensity of agitation at three positions along the column axis. Experimental drop-size distributions were compared with three two-parameter theoretical distribution functions, namely, the log-normal, Gamma, and Weibull. The Gamma function was found to be the most suitable. An empirical correlation for a dimensionless width of the drop-size distribution is presented in terms of the rate of agitation and interfacial tension. The Sauter mean diameter was found to depend on the speed of agitation, physical properties of liquid systems, and phase flow rates. It could be correlated using an additive model involving the ratio of buoyancy to interfacial forces at low agitation, and a turbulent breakup mechanism at high agitation, the effect of flow rates being expressed in terms of the dispersed-phase holdup

The design of multiproduct batch plants with process performance models

Abstract: A model is presented for the optimal design of multiproduct batch plants, which uses process performance information. The process description resembles the familiar polynomial model for sizes and cycle times. While the former approaches considered constant values for the size and cycle time factors, they are considered as functions of a set of process variables. The results obtained show significantly improved solutions with this model, when typical chemical engineering unit operations are involved. Operating costs are naturally included

A comparison of the kinetics of flash calcination of kaolinite in different calciners

Abstract: We have determined reaction rates and activation energies for the dehydroxylation of kaolinite by flash calcination using two laboratory and two industrial pilot flash calciners under various conditions. The kinetics of dehydroxylation/water loss can be fitted to a 3-dimensional diffusion model, but changes in mechanism occur at high dehydroxylation levels and lead to deviations from that model. The effects of furnace gas and particle size are discussed.

Cfd and Experimental Studies of Fluid and Particle Flow in Horizontal Primary Separators

Abstract: The flow field of liquid in two dimensional and three dimensional models of horizontal primary oil/water separators has been studied experimentally using LDA and PDA measurement techniques. A CFD simulation has also been applied to these configurations to assess its suitability for further modelling work. It was found that the CFD results were in reasonable agreement with velocity measurements in a small two dimensional, rectangular model separator, but there were substantial differences between measured flow patterns and CFD predictions in a larger three dimensional cylindrical separator. Measurements were also made of particle size distribution at locations near the inlet of the small two dimensional model separator. These were found to be consistent with size dependent particle trajectories predicted by the CFD simulation. It was concluded that the CFD simulation may be used to investigate design developments of the small two dimensional separator but the simulation as used in the article is not suitable for use in the larger three dimensional separator. Further work is required to establish the cause of this limitation. It appears to be due to the limit on the number of computational cells permitted in the CFD simulation which in turn restricts the minimum volume of cells used in the CFD simulation

One-dimensional, non-adiabatic, microscopic model of electrothermal desorption process dynamics

Abstract: A new thermal desorption process in which the heat for desorption is generated inside the adsorbent particles by passing an electric current through them (Joule's heat), called electrothermal desorption, is modelled on a column scale. A one-dimensional, non-adiabatic, microscopic model describing the process dynamics is presented. The model was obtained as a set of nonlinear partial differential equations with three independent variables (time, column axial coordinate and particle radial coordinate), which was solved using the double orthogonal collocation method. On the basis of this model, time and space concentration and temperature profiles were simulated. The influence of different process parameters was investigated, and the advantage of desorption with internal sorbent heating was proven

The high-temperature drying of softwood boards: a kiln-wide model

Abstract: Characteristic drying curves are normally determined from laboratory experimental data to specify the drying behaviour in a commercial-scale dryer. By contrast in this paper, we obtain the drying kinetics from the numerical results of a more rigorous mathematical model. In our example, we use a model developed for the high-temperature drying of a single softwood board to describe the drying behaviour of a stack of boards in a kiln. Analysis of the mechanism of moisture movement has led us to identify three stages in the drying of a sapwood board and two for a heartwood board. The drying kinetics can be represented by a dual characteristic curve covering the two falling-rate periods, to yield simplified expressions for the drying kinetics. These expressions can be coupled with mass and energy balances over a control volume to give equations which can be solved numerically to determine the humidity, temperature and moisture-content profiles in the airflow direction within a timber kiln. This two-step modelling procedure provides a computationally efficient way of analysing conditions in any batch dryer

CHiPS : a process synthesis package : Process design

Abstract: CHiPS is a package for the synthesis of complete heat-integrated processes. Based on a fully user extensible unit model library, using discrete programming techniques, the package can handle a large variety of problems and will generate processes using distillation units, reactors, flash units, absorbers, etc. Flowsheets generated can include recycle and makeup streams. The program has been written in the C and Fortran languages and is portable across a variety of machines: CHiPs runs on single workstations, on workstation clusters, and on distributed memory multicomputers. We outline the main features of the package, including the following: (1) single pass generation of heat integrated processes, (2) processes are generated allowing for the effect of unit operating conditions on down-stream units, (3) the N best solutions are found, instead of just the optimal flowsheet, for any value of N, (4) a unit model interface is defined, which allows a user to write new unit models and easily add them to the basic package, and (4) the epee interface is available for use within a distributed process engineering environment. We describe briefly some of the critical implementation issues relating to memory and computer time restrictions imposed by the requirements of an interactive computer aided design environment. We discuss how sparse memory techniques enable the package to run in low memory conditions whilst still allowing efficient re-use of repeated computations, and how parallel implementations allow us to increase the complexity of the models used whilst generating results in reasonable amounts of time