List of possibilities

*


  1. *Ideal reactor: Assume an ideal reactor type, solve the material balance equation. Take preferably the 'steady state' for the computation of continuous reactors. Remember: for the CSTR the differential equation changes to a simple algebraic equation when omitting the instationary term dc/dt. Don't forget: the ideal batch reactor and the stationary ideal tubular flow reactor have the same solution of the kinetic equation and are 'in analogy' by the transformation from time to length or vice versa. Try to understand and apply: A very clear graphical and numerical visualization and calculation of the space time can be achieved by choosing the the function of 1/r (r = reaction rate) versus the conversion U. A plot of this function allows to give an understandable explanation of the fact that in case of 'simple' reactions the totally back-mixed CSTR is always 'worse' than TFR or STR (areas under the curves/rightangles). By the way, can you formulate a general form of reactor material balance (not containing 'exotic' mass transfer effects)? Hope Yes, control?

    back to top

  2. *Reactor combination: When the residence time distribution (RTD) is known (experimentally) and it can be seen that the 'real' RTD-behaviour of the chosen reactor can be described by a combination of ideal reactors (for example: modelling a tubular flow reactor - with e.g. an axial dispersion - by a series of CSTRs, see further information, or the 'bypass-model' !!). In this case the material balance for the reactor combination can be defined and solved ( possibly by taking partial flows for the different branches and making balances for splitting and joining nodes)

    back to top

  3. *'Area-Method': When the RTD of a real reactor is experimentally known and modelling by a reactor combination is too intricate the 'Area-Evaluation-Method' can be used. The method needs the RTD-information and the reaction kinetics as 'inputs'. The method bases on the formula:
    Ci = Integral(Ci(t)*w(t)dt).
    The output concentration can be evaluated graphically or numerically as area in a plot Ci(t) versus W(t). This method describes the 'reality' of the RTD in a very precise manner. But the disadvantage of the method is that it is strictly valid only for segregated fluids and not for 'molecular' dispersed fluids. The mistake which arises from this difference is by far less important than the mistake from choosing a 'wrong' RTD. For reactions of 1st Order the difference can not be realized,- only for reactions of higher or lower order there is a remarkable difference. For reactions with higher Orders the output concentration is higher for segregated fluids (see the question on '2nd order reactions and segregation' as an example). This behaviour and the validity of the formula can be explained with the assumption that a segregated fluid corresponds to a 'infinitesimal crowd' of 'micro-batch-reactors' (= segregated fluid!) . For molecular dispersed fluids the formula can not be explained (lack of 'individuals'). Look to 2 informative     sketches !!

    back to top

take your browser back for previous text or:

Back to main test page