CSTRs parameter modeling

The value of n is the only parameter in this equation. Several procedures can be used to find its value when the RTD is known experiment or calculation from the variance, as in /i = 1/C (t ) = 1/ t C t), or from a suitable loglog plot or the peak of the curve as explained for the CSTR battery model. The Peclet number for dispersion is also related to n, and may be obtainable from correlations of operating variables. 

The sum of squares as defined by Equation 7.8 is the general form for the objective function in nonlinear regression. Measurements are made. Models are postulated. Optimization techniques are used to adjust the model parameters so that the sum-of-squares is minimized. There is no requirement that the model represent a simple reactor such as a CSTR or isothermal PER. If necessary, the model could represent a nonisothermal PFR with variable physical properties. It could be one of the distributed parameter models in Chapters 8 or 9. The model... 

Many working groups have modeled the performance of diesel particulate traps during the past few decades. Concentrated parameter models (CSTR assumption) have been applied for the evaluation of formal kinetic models and model parameters. The formal kinetic parameters lump the heat and mass transfer effects with the reaction kinetics of the complicated reaction network of diesel soot combustion. Those models and model parameters were used for the characterization of the performance of different filter geometries and filter materials, as well as of the performance of a variety of catalytically active coatings and fuel additives [58],... 

A simple dynamic model is discussed as a first attempt to explain the experimentally observed oscillations in the rate of propylene oxide oxidation on porous silver films in a CSTR. The model assumes that the periodic phenomena originate from formation and fast combustion of surface polymeric structures of propylene oxide. The numerical simulations are generally in qualitative agreement with the experimental results. However, this is a zeroth order model and further experimental and theoretical work is required to improve the understanding of this complex system. The in situ use of IR Spectroscopy could elucidate some of the underlying chemistry on the catalyst surface and provide useful information about surface coverages. This information could then be used to either extract some of the surface kinetic parameters of... 

Here we use a single parameter to account for the nonideality of our reactor. This parameter is most always evaluated by analyzing the RTD determined from a tracer test. Examples of one-parameter models for a nonideal CSTR include the reactor dead volume V, where no reaction takes place, or the fraction / of fluid bypassing the reactor, thereby exiting unreacted. Examples of one-parameter models for tubular reactors include the tanks-in-series model and the dispersion model. For the tanks-in-series model, the parameter is the number of tanks, n, and for the dispersion model, it is the dispersion coefficient D,. Knowing the parameter values, we then proceed to determine the conversion and/or effluent concentrations for the reactor. 

In addition to the one-parameter models of tanks-in-series and dispersion, many other one-parameter models exist when a combination of ideal reactors is to model the real reactor. For example, if the real reactor were modeled as a PFR and CSTR in series, the parameter would be the fi action,/, of the total reactor volume that behaves as a CSTR Another one-parameter model would be the fi action of fluid that bypasses the ideal reactor. We can dream up many other situations which would alter the behavior of ideal reactors in a way that adequately describes a real reactor. However, it m be that one parameter is not sufficient to yield an adequate comparison between theoiy... 

Comparison of conversions for a PFR and CSTR with the zero-parameter and two-parameter models. symbolizes the conversion... 

The steady states which are unstable using the static analysis discussed above are always unstable. However, steady states that are stable from a static point of view may prove to be unstable when the full dynamic analysis is performed. That is to say simply that branch 2 in Figure 4.8 is always unstable, while branches 1,3 in Figure 4.8 and branch 4 in Figure 4.8 can be stable or unstable depending upon the dynamic stability analysis of the system. As mentioned earlier, the analysis for the CSTR presented here is mathematically equivalent to that of a catalyst pellet using lumped parameter models or a distributed parameter model made discrete by a technique such as the orthogonal collocation technique. However, in the latter case, the system dimensionality will increase considerably, with n dimensions for each state variable, where n is the number of internal collocation points. 

This work was based on the developmoit of a steatfy-state model which can simulate both polymerization rate and effluent PSD for a seed-fed CSTR. The model has only one adjustable parameter the coefficient for radical desorption. Figures 8.2 and 8.3 illustrate the fitting of PSD data for a typical experimoiL Figure 8.2 shows PSD histograms of the seed and effluent latexes. The PSD data... 

This case study is a reactor-separator-recycle system to produce monochlorobenzene. The operating parameters and sizes for one of the synthesis alternatives are optimised using the detailed models and the costing information provided. Each unit has a capital cost, Cc, and an operating cost, Q, which is incorporated into the objective function through a pay out time of 2.5 years. The principal units are a CSTR and two separation columns. The models have been reformulated in terms of component flowrates, Fsj. The reactor is a continuous stirred tank reactor (CSTR) which models the reaction between chlorine and benzene (A) to produce monochlorobenzene ( B) and dichlorobenzene (C) at a constant temperature. The maximum (global) profit is 2081/day. 

It has been proposed to represent the TF with 10 CSTR staged units 0.3 ft in length. Develop solutions to this problem using a finite difference method of solving an ordinary differential equation and a lumped parameter model employing a method of solution of simultaneous linear algebraic equations. 

The competition between reaction and diffusion can be represented by the lEM-model. t is identified with a diffusion time t = yL /33 (see Sec. 3.2 above) where different diffusivities 3j and hence different micromixing times t- may be used for each species. This simple lumped parameter model gives results comparable to those of more sophisticated distributed models, at least for reaction systems which are not too "stiff". An interesting property is revealed by numerical simulations. The simplest way to represent partial segregation in a fluid is to consider that it consists of a mixture of macrofluid (fraction 3) and microfluid (fraction 1-3). It turns out that the ratio (1 - 3)/3 is always close to that of two characteristic times [28 QlSj. In the case of erosive mixing of two reactants in a CSTR (erosion controls mixing and the product of erosion is a microfluid), one finds... 

The CIS model and the Dispersion model are referred to as one-parameter models, since only a single parameter, D/uLot N, is used to characterize mixing. When there is very little mixing in the direction of flow, i.e., when A is large otD/uL is small, the physical basis of the Dispersion model is stronger than that of the CIS model. Moreover, when the number of CSTRs in series is large, it can be tedious to calculate the performance of the series of reactors. On the other hand, it is relatively straightforward to use even the most complex rate equations in the CIS model. It is not necessary to restrict the analyses to first-order rate equations. 

Real reactors deviate more or less from these ideal behaviors. Deviations may be detected with re.sidence time distributions (RTD) obtained with the aid of tracer tests. In other cases a mechanism may be postulated and its parameters checked against test data. The commonest models are combinations of CSTRs and PFRs in series and/or parallel. Thus, a stirred tank may be assumed completely mixed in the vicinity of the impeller and in plug flow near the outlet. 

Distribution models are curvefits of empirical RTDs. The Gaussian distribution is a one-parameter function based on the statistical rule with that name. The Erlang and gamma models are based on the concept of the multistage CSTR. RTD curves often can be well fitted by ratios of polynomials of the time. 

The described experimental rig for the anionic polymerisation of dienes has been shown to behave as an ideal CSTR. The mathematical model developed allows the prediction of the MWD at future points in the reactor history, once suitable kinetic parameters have been estimated. 

The goal is to determine a functional form for (a, b,. .., T) that can be used to design reactors. The simplest case is to suppose that the reaction rate has been measured at various values a,b,..., T. A CSTR can be used for these measurements as discussed in Section 7.1.2. Suppose J data points have been measured. The jXh point in the data is denoted as S/t-data aj,bj,..., Tj) where Uj, bj,..., 7 are experimentally observed values. Corresponding to this measured reaction rate will be a predicted rate, modeii p bj,7 ). The predicted rate depends on the parameters of the model e.g., on k,m,n,r,s,... in Equation (7.4) and these parameters are chosen to obtain the best fit of the experimental... 

The next two steps after the development of a mathematical process model and before its implementation to "real life" applications, are to handle the numerical solution of the model s ode s and to estimate some unknown parameters. The computer program which handles the numerical solution of the present model has been written in a very general way. After inputing concentrations, flowrate data and reaction operating conditions, the user has the options to select from a variety of different modes of reactor operation (batch, semi-batch, single continuous, continuous train, CSTR-tube) or reactor startup conditions (seeded, unseeded, full or half-full of water or emulsion recipe and empty). Then, IMSL subroutine DCEAR handles the numerical integration of the ode s. Parameter estimation of the only two unknown parameters e and Dw has been described and is further discussed in (32). 

Different reactor networks can give rise to the same residence time distribution function. For example, a CSTR characterized by a space time Tj followed by a PFR characterized by a space time t2 has an F(t) curve that is identical to that of these two reactors operated in the reverse order. Consequently, the F(t) curve alone is not sufficient, in general, to permit one to determine the conversion in a nonideal reactor. As a result, several mathematical models of reactor performance have been developed to provide estimates of the conversion levels in nonideal reactors. These models vary in their degree of complexity and range of applicability. In this textbook we will confine the discussion to models in which a single parameter is used to characterize the nonideal flow pattern. Multiparameter models have been developed for handling more complex situations (e.g., that which prevails in a fluidized bed reactor), but these are beyond the scope of this textbook. [See Levenspiel (2) and Himmelblau and Bischoff (4).]... 

In the previous section we indicated how various mathematical models may be used to simulate the performance of a reactor in which the flow patterns do not fit the ideal CSTR or PFR conditions. The models treated represent only a small fraction of the large number that have been proposed by various authors. However, they are among the simplest and most widely used models, and they permit one to bracket the expected performance of an isothermal reactor. However, small variations in temperature can lead to much more significant changes in the reactor performance than do reasonably large deviations inflow patterns from idealized conditions. Because the rate constant depends exponentially on temperature, uncertainties in this parameter can lead to design uncertainties that will make any quantitative analysis of performance in terms of the residence time distribution function little more than an academic exercise. Nonetheless, there are many situations where such analyses are useful. 

Stirred tank performance often is nearly ideal CSTR or the model may need to take into account bypassing, stagnant zones or other parameters associated with the geometry and operation of the vessel and the agitator. Sometimes the vessel can be visualized as a zone of complete mixing in the vicinity of the impellers followed by a plug flow zone elsewhere, thus a CSTR followed by a PFR. 

The CRE approach for modeling chemical reactors is based on mole and energy balances, chemical rate laws, and idealized flow models.2 The latter are usually constructed (Wen and Fan 1975) using some combination of plug-flow reactors (PFRs) and continuous-stirred-tank reactors (CSTRs). (We review both types of reactors below.) The CRE approach thus avoids solving a detailed flow model based on the momentum balance equation. However, this simplification comes at the cost of introducing unknown model parameters to describe the flow rates between various sub-regions inside the reactor. The choice of a particular model is far from unique,3 but can result in very different predictions for product yields with complex chemistry. 

---