Band 4, Heft 5 - WCCE8 2009

The 8th World Congress of Chemical Engineering (WCCE 8) Symposium on Process Design, held on August 23-27, 2009 in Montreal, Canada, offered a unique opportunity to focus on a comprehensive participation and learning experience over the full range and diversity of methods in the broad fields of process modeling and their use for design, simulation, optimization and control of chemical processes. The major themes of the symposium are categorized into the following themes of the congress:• Energy • Green Processing • New Materials • BiotechnologyAuthors of selected papers accepted for presentation in the Process Design symposium were invited to submit the full papers to be considered for publication in the journal of Chemical Product and Process Modeling (http://www.bepress.com/cppm). The papers that successfully passed the reviewing process of the journal are now published in the WCCE8 special issue of the journal.

This work presents an integrated reactive-distillation design based on a dividing-wall column (DWC) applied to an industrial case study within AkzoNobel. Remarkably, it is among the first industrial applications of a reactive DWC reported in literature. The benefits of this novel integrated design are the ability to overcome VLE and chemical equilibrium limitations, as well as to separate the main product as high purity side-stream. To solve the problem of a production shift required by market demand changes, we propose an innovative integrated design that combines reaction and separations into one reactive DWC that allows significant savings in capital and operating costs – up to 35% and 15%, respectively. Two scenarios are analyzed and the results of the rigorous AspenPlus simulations are provided.

This paper presents historical price data for two different crude oil types and examines the stationarity and inherent structure in oil price variation, applying many degrees of time resolution. Time Series Analysis results are then used to identify patterns and analyze the variation timescales. A specific goal of this study is to investigate and demonstrate the presence of fractal scaling. In particular, we postulate and prove that the mean size of the absolute values of price changes obeys a fractal scaling law (a power law) and can be expressed as a function of the analysis time interval (here, the latter is an independently varying parameter, ranging from a day up to a calendar year). The fractal structure of crude oil price variation is confirmed, the drift exponent is computed and the power scaling window of validity is depicted for both types, illustrating the interplay of both short- and long-term effects on the intrinsic structure of crude oil prices before and after 2008.

The effects of sound wave irradiation were investigated on the efficiency of chemical reaction and product distribution in plasma field generated by DC pulse discharge. Since the discharge plasma provides high energy locally, various chemical reactions ungoverned by thermodynamics can occur. However, the discharge space is extremely confined, which restricts the conversion of feed gas. In this work, the combination of discharge plasma and sound wave irradiation, which increases the vibration motion of the medium particle, was studied. An expansion of streamer into a fan-shape was observed with the irradiation of sound wave. The expansion of the discharge space was positively correlated with the vibration speed of the medium particle (Vs). Accordingly, the effects of sound wave irradiation on the methane conversion (XCH4) were evaluated. When methane diluted in argon (mole fraction of methane was 0.1) was used as the feed gas, methane conversion increased with Vs, and XCH4 at 3.0 m/s of Vs increased to approximately twice that for no sound wave irradiation. This increase in XCH4 resulted from the improvement in the reaction probability due to expansion of the discharge field and change of discharge properties. On the other hand, hydrogen formation and oily and deposit compounds formation were slightly increased. Therefore the enhancement of reaction efficiency and the possibility of product-control of plasma reaction by sound wave combined process were suggested.

Biopolymers are biodegradable polymers made from renewable resources. They can be made from plants or by microorganisms. Pseudomonas putida is one such microorganism. Under nutrient limitation and carbon excess it produces polyhydroxyalkanoates (PHA), which are linear polyesters, to store carbon as energy resource. PHAs are biodegradable and are used in the production of bioplastics.The production of biopolymers from microorganisms is currently not cost effective, because of expensive substrates, low yields and complex downstream processing. Therefore, intensification of PHA production to achieve higher yields in smaller, more efficient equipment at lower capital costs would be an advantage.Oscillatory baffled reactors (OBRs) are a form of plug flow reactor that can be used for long residence time reactions, as the plug flow is not dependent on achievement of a certain net flow velocity in the reactor, leading to much more compact designs. Fermentations, such as the production of PHA by P. putida, show characteristically long residence times. Therefore, they could be intensified in an OBR. Since one major advantage of the OBR is its highly uniform, controllable mixing, which can be used to enhance gas-to-liquid mass transfer, it should be suitable for biological reactions. Fed-batch or continuous fermentation is a second advantage, as it should minimize substrate and product inhibition. Lower, more uniform shear rates than in a stirred tank reactor would be an additional advantage for processing shear-sensitive cells. Previous research has shown that the OBR can provide higher biomass concentrations and greater gas-liquid mass transfer than conventional fermenters. In this work, the production of PHA by P. putida KT2442 was studied. The parameters investigated here were agitation rate and temperature. It was demonstrated that PHA can be produced in the OBR. The fermentation temperature and agitation rate resulting in the highest OD and PHA yield in this parameter space were determined as 30 ± 1°C and an oscillatory Reynolds number (Reo) of 300, respectively. It was also demonstrated that using an OBR can lead to higher OD (10.8 vs 7.26) and dry cell weight (3.75 g/L vs 2.4 g/L) than in a comparable conventional stirred tank reactor.

Carbon deposition during hydrocarbon steam reforming is often a major cause of pathological reactor performance and catalyst deactivation. In this paper, we report the effect of forced periodic cycling between propane-steam reforming feed and a carbon gasifying agent (CO2) as a novel reactor strategy to both improve product yield (H2 and CO) and catalyst longevity. Experiments were carried out over Co-Ni/Al2O3 catalyst in a fluidized bed reactor. Cycle period, ?, was varied between 5 to 20 mins at 5 different cycle splits, (0.1 ? s ? 0.9). Both H2 and CO formation rates were higher (up to 5-fold and more than 10-fold, respectively) than that attainable under steady-state operation at all periods investigated. In particular, the time-average H2:CO ratio was lower (< 3.0) than the steady-state equivalent for the pure propane steam reforming (14.0), although it increased monotonically with cycle split. Composition cycling with CO2 also improved catalyst stability and longevity compared to steady-state performance at the cycle periods examined. This strategic reactor operation is therefore a potentially useful key to green process engineering in the overall petrochemical plant design to effect greenhouse gas emission reduction.

Kernel PCA, as a multivariate statistical process monitoring (MSPM) tool, is a powerful technique capable of coping with non linear relations between variables, thus outperforming classical linear techniques when non linearities are present in data. In real industrial chemical processes, multiple plant operating modes often lead to multiple nominal operation regions, and MSPM techniques that do not take account of this fact show increased false alarm and missing alarm rates. The existence of multiple operation modes is often more frequent than clearly expressed strong non linear relations between the variables involved. Non linear relations do exist, but the small variability allowed in key variables during normal plant operation prevents these non linear relations from being expressed in the data. In this work, a fault detection tool based on Kernel PCA is tested in such multiple operation modes environments, with final objective of implementing the tool in a real industrial installation. Robustness of the tool for coping with a certain percentage of outliers is particularly examined. The tool is applied to three case studies: (i) a two dimensional toy example, (ii) a realistic simulation usually used as a bench-mark example, known as the Tennessee Eastman Process, (iii) real data from a methanol plant. Hence, the feasibility of implementing fault detection tools based on this technique in the field of chemical industrial processes is discussed.

The adsorption of 4-chlorophenol on an expanded commercial clay used in water purification has been studied. Experiments using 4-chlorophenol concentrations of 50, 100, 250, 500, 750, 1000, 1250 and 1500 mg L-1 were carried out in column reactors at 25ºC.A new adsorption kinetics equation has been developed based on the equation for the variation in solute concentration in the liquid phase, representing the difference between the adsorption rate minus the desorption rate that at equilibrium represents the Langmuir isotherm. This was used to derive a new finite equation for the development of the adsorption curve.

In this paper, we proposed a uniformly bounded error estimator for a common class of bioreactor models. The biomass and other products are estimated by means of substrate concentration measurements employing the characteristics of the unstructured cell growth models, which are linearly dependent. The estimation methodology is based on a suitable change of variable which allows generating artificial variables to infer the remaining mass concentrations constructing a differential-algebraic structure. The proposed methodology is applied to a class of Monod unstructured kinetic model with success. Some remarks about the convergence characteristics of the proposed estimator are given and numerical simulations show its satisfactory performance.

A closed form solution has been obtained for the release kinetics of a solute from a spherical drug matrix into a finite volume of liquid, taking into account the effect of rate of absorption. The proposed model results clearly show the effect of the absorption rate constant on the rate of drug release. The obtained results are compared with the experimental data and diffusion-only model results. There is a significant difference in the release profile when the rate of absorption of drug is slow. The most important feature of the mathematical relationship between the liquid concentration verses time is its ability to predict change in the performance of the drug by manipulating the parameters of the equation. These parameters include the initial concentration of the drug, the radius of the drug and diffusivity of the drug in the solid to name a few. Therefore, a substantial number of experiments can be eliminated when the optimal performance of a drug is sought after.

The closed form solution to the conjugated boundary value problem posed by a counter current hemodialyzer facilitates the estimation of the overall mass transfer coefficient. Comparison of the proposed model results with published experimental data shows good agreement for Urea and Creatinine clearances over a published range of blood and dialyzate flow rates. This model predicts clearances with a maximum error of less than 4% for both Urea and Creatinine when blood flow is 75% of the dialyzate flow. However, when both blood and dialyzate flows are identical the model over predicts the experimental data by 1.47% in the case of Urea and 4.75 for Creatinine flows of 300 ml/min. Although the concentration profile is an infinite series involving confluent hypergeometric functions, 2 terms of the series were sufficient (Mathematica notebook program) to produce these results. Overall mass transfer coefficients can now be deduced from the Sherwood numbers and provide possible improvement over currently used area coefficients.

Methyl ester sulfonation with sulfur trioxide derived from oleum is possible under special conditions on a pilot plant scale. Quantum chemical calculations were used to study the relative stability between intermediates in the proposed mechanism. In this report an analysis of a mathematical model for a falling film sulfonation reactor is presented. It aims to estimate the temperature and conversion profiles of the thin film. The model treats the heat released and the trioxide sulfur dissolution assuming that the film theory is applicable. The temperature and concentration gradients can exist across the film during the chemical reaction. The equations were solved using finite differences. The most important data obtained by the mathematical model for a subsequent correlation with the properties of the reagents and product are the conversion, the density and viscosity of the sulfonic product. The results indicate an increase in the axial temperature of the liquid film and the conversion from the top reactor in accordance with the experimental results. It considers that at the upper section of the reactor the reaction is controlled by means of mass transfer due to gas phase turbulence. A mild conversion on the reactor bottom means that the liquid phase controls the mass transfer due to the amounts of sulfur trioxide transferred into the film which produces changes in the film composition.

To realize process intensification, active species like radicals and their mass transfer need to be considered. A two-layered catalyst bed reactor was used to investigate the participation of radicals in two different reaction processes. Two-layered catalyst bed (2LCB), mixed catalyst beds (MCB) and separated catalyst beds (SCB) were compared in oxidative coupling of propane. The dependency on the contact time for the concentration of propylene in the effluent was almost constant at experimental conditions. However, that of benzene with 2LCB and MCB was higher than SCB, especially at a lower flow rate. The quartz sand layer for SCB between the catalyst beds prevented the highly active intermediate generated in the first layer from being transferred in the gas phase to the second layer. This result supports the idea of the participation of the active intermediate in the direct coupling to benzene.

A new method for the synthesis of mass exchanger networks for multi-period operations is presented in this paper. The new technique which is called multi-period interval based MINLP superstructure (MIBMS) for mass exchange network synthesis (MENS) is adapted from the interval based MINLP superstructure for MENs presented by Isafiade and Fraser (2008). Process parameters such as supply and target compositions and flowrates of streams can vary over a specified range due to environmental or economic reasons. In this paper, the IBMS model for MENS is modified to handle variations in the aforementioned set of parameters by including the index ‘p' in the IBMS model and using the maximum area approach in the objective function as presented by Verheyen and Zhang (2006) for multi-period heat exchanger networks synthesis (HENS). The index ‘p' represents each period of operation which in this paper can be unequal. The maximum area approach ensures that each mass exchanger connecting the same pair of streams in more than one period is able to transfer mass in such streams for all the periods. It should be known that this technique applies to multi-period problems where the process parameters for each period of operation as well as the duration are specified. The method does not handle scenarios where the process parameters are uncertain. The new technique is applied to three examples.

Modern fluid separation processes occur under very complex conditions. As a result, their modeling is usually based on severe assumptions and experimentally estimated gross parameters. On the other hand, the process optimization can only be achieved with reasonable model accuracy, whereas the process rates should be considered in a rigorous way, with respect to both transport phenomena and chemistry.Depending on the complexity of process fluid dynamics, different modeling ways can be applied, their rigor and complexity stretching over a wide range. Among these ways are a direct application of the equations of fluid dynamics, the hydrodynamic analogy method and the rate-based approach. These methods are complementary in the sense that, together, they are able to govern both simple and very complex process fluid dynamic conditions. Moreover, there is also another complementarity between the different approaches, namely, they can be applied in combination, for instance, by estimating process parameters by a more rigorous method and delivering them to the less rigorous one. In this contribution, the complementary modeling is discussed in detail and illustrated with case studies.

The industrial production of refined sodium bicarbonate is realized in large-scale bubble columns, called the BIR columns. Several complex phenomena exist in these columns, such as three-phase flow, gas-liquid mass transfer of CO2, and crystallization. This work deals with the development of an operational BIR column model using the compartmental modeling approach. It is fed by previous works and current studies. A parametric sensitivity analysis is realized to identify the more important modeling parameters and to point the future research that have to be performed. Some simulations results are presented for typical operating conditions of an industrial column.

The supercritical antisolvent (SAS) process has been widely used to obtain many types of crystalline and amorphous particles of polymers and drugs. In this process, solute particles precipitate from an organic solution when it is sprayed into a compressed antisolvent continuum. The antisolvent is miscible with the organic solvent, but immiscible with the solute. The objective of this work was to investigate the application of supercritical carbon dioxide as antisolvent for the recrystallization of caffeine; a compound largely employed in pharmaceutical and food industries. The carbon dioxide was pressurized at 100bar and the solution was injected into a precipitation chamber, thereby inducing its supersaturation and particle precipitation. The effect of process parameters such as initial concentration of caffeine in the organic solution (3, 5 and 10mg/mL), organic solution flow rate (3mL/min), concentration of CO2 (96 and 98mol%), drying flow rate (990mL/min) and pressure gradient between the precipitation chamber and the exit of the sprayer tube (110 and 220bar) on the solid state properties of caffeine was investigated. The pressure (100bar) and temperature (60oC) of the precipitation chamber were kept constant. The results showed a great difference between the processed and unprocessed samples. In all experimental conditions a reduction in caffeine mean particle size was observed, with length ranging from 0.6 to 168.3?m and width from 0.1 to 12.7?m, compared to the mean size of unprocessed caffeine particles (16.6 to 214.4?m in length and 7.1 to 131.2?m in width). This fact was confirmed by differential scanning calorimetry (DSC), where the precipitated particles presented lower fusion enthalpy than the unprocessed ones. Smaller crystal needles (0.3?m to 0.4?m) were obtained with crystallization at the lowest caffeine concentration in the presence of 98mol% CO2.

Effect of changing the liquid level in an airlift membrane bioreactor of 0.7 m height, 0.24 m width and 0.18 m depth was studied both experimentally and by simulation. Three-dimensional simulations of the airlift membrane bioreactor were carried out at two different liquid levels above the membrane. The simulations were based on the two-fluid model with the standard k–? model for the turbulence. The results showed that by lowering the liquid level, the quality of mixing and uniformity of the velocity distribution of liquid phase in the riser would be improved while the shear stress on the membrane surface would be reduced. Higher shear stress on the membrane surface at high levels of liquid minimizes the extent to which particles settle on the membrane, thus, fouling will be reduced and flux of liquid through membrane will be enhanced. Moreover, it was shown that by lowering the liquid level, the fraction of air in the downcomers becomes lower.

This paper illustrates the operational difficulties arising from simultaneously performing exothermic and endothermic reactions, and demonstrates that a plant can be built and safely operated by integrating the design and plantwide control issues. The behaviour of reactor – separation – recycle systems carrying the coupled reactions A ?P + Q (endo) and B + Q ? R (exo) is investigated. Irrespective of the control structure, state multiplicity cannot be removed if the intermediate component Q is recycled. Therefore, the chemical reactor should be designed such that the recycle of Q can be avoided without economic penalty. The theoretical findings are applied to the design and control of a plant coupling ethylbenzene dehydrogenation and nitrobenzene hydrogenation for simultaneous production of styrene and aniline. After plant design, a rigorous dynamic model is developed using AspenDynamics. A plantwide control structure is implemented and shown to be able to achieve stable operation. Production rate changes of reasonable magnitude can be achieved by changing the reactor-inlet flow rates or bed-inlet temperature.