Key Findings

The following specific conclusions were derived from this research:

• the equilibrium stage model is a satisfactory basis for reactive distillation simulation;

• the response of hybrid columns to changes in operating conditions is often not intuitive and often does not follow the behaviour of non-reactive distillation columns (e.g. the effect of pressure and fractionation changes);

• the dynamic simulation of reactive distillation is possible without using a transformed composition space, provided that the process representation is complete and structured correctly;

• input multiplicity is usually present in hybrid reactive distillation, constraining the pairing of manipulated and controlled variables for linear control systems;

• output multiplicity is sometimes present in hybrid reactive distillation, resulting from at least four fundamental mechanisms (the fourth of which was identified in this work);

• pseudo-multiplicity (described in this work) is often present in reactive distillation but does not necessarily indicate a physically realisable output multiplicity;

• transitions between parallel steady states are possible following a range of common disturbances to the feed composition and rate and various perturbations in manipulated variables;

• output multiplicity compels the use of specific column start-up sequences and effectively excludes the application of many distillation control structures;

• the ether purity and the isobutene conversion can be controlled simultaneously with a simple linear control system, but only if the manipulated and controlled variables are chosen according to specific criteria (outlined here);

• novel packing arrangements were developed to support small catalyst particles within a pilot scale reactive distillation column, and to reduce the cost of non-reactive distillation stages at the laboratory and pilot scales.

13.1.3 Secondary Findings

A number of specific conclusions support the key findings:

• the commercial simulation packages, Pro/II™ and SpeedUp™, permitted accurate modelling of reactive distillation operations;

• the SpeedUp™ simulation environment is also effective for the implementation of dynamic simulation models of reactive distillation systems;

• the Pro/II™ simulation environment provided adequate physical property predictions for the successful design of pilot plant equipment (e.g. condenser, reboiler and product cooler) to be used for ETBE synthesis, when combined with conventional design methods;

• the operating pressure of a hybrid reactive distillation column must be optimised to balance the reaction and separation functions of the column;

• increasing fractionation in an hybrid column does not necessarily reduce the energy demand or increase the maximum product purity;

• the production of ETBE is most attractive for hydrocarbon feeds with a relatively low concentration of the reactive component, isobutene;

• a satisfactory design strategy for hybrid columns requires a priori study of residue curve diagrams and reactive residue curves, and a rigorous simulation study to produce an optimal design;

• the conversion of MTBE reactive distillation processes to ETBE can sometimes reduce the energy consumption and create equipment redundancies;

• the operation of hybrid columns is considerably more sensitive to the feed composition and the operating conditions compared with non-reactive distillation;

• input multiplicity commonly arises from competing effects that are present in the reactive distillation process;

• uniquely defined input conditions for reactive distillation do not preclude the possibility of more than one steady state operating point;

• singularities in the mass-molar flow relationships and inverse effects from the stage-to-stage energy balances, both known causes of output multiplicity in ideal binary distillation, can also cause output multiplicity in hybrid reactive distillation;

• multiple azeotropes in ternary and pseudo-ternary mixtures, a known cause of output multiplicity in azeotropic distillation, can also cause output multiplicity in hybrid reactive distillation;

• reaction hysteresis can arise from the interactions between the reactive and non-reactive sections of a hybrid reactive distillation column and, where present, can cause output multiplicity;

• many of the reported examples of output multiplicity in reactive distillation are not physically realisable, and only satisify the criteria for pseudo-multiplicity;

• control schemes which use the distillate rate as the primary manipulated variable are unsuitable for MTBE or ETBE columns due to poor dynamic responsiveness;

• material balance control schemes are unsuitable for MTBE or ETBE columns which exhibit multiple steady states for a constant reboiler duty or reflux rate due to the difficulties in achieving stable inventory control with linear controllers;

• an integrated control strategy can ensure all important operating parameters (i.e. the process performance) remains acceptable following process disturbances and set-point changes, and requires only simple linear controllers to implement;

• the synthesis of ETBE from ethanol and isobutene is feasible in a pilot scale hybrid reactive distillation column using novel packing arrangements for both the reactive and non-reactive column sections.

Was this article helpful?

0 0

Post a comment