Step 2 AR Construction

The direction of travel for PFR concentrations is toward an increasing concentration of B. The PFR trajectory appears to be fairly linear (the trajectory does not appear to contain concavities), and thus mixing in this instance will not allow for the achievement of many new concentrations. Note that the equilibrium concentration for Cg is small ( 0.011 mol/L) compared to the initial concentration of A (1.0 mol/L), indicating that the majority of A has been converted to C instead of B. Next, the CSTR locus from the feed is determined. 

2 CSTR Locus from the Feed To plot the CSTR locus from the feed point, the system of CSTR equations must be solved (simultaneously) over a range of residence times. In comparison to the Van de Vusse system, the kinetics for this system is more difficult to solve analytically. Numerical solution of the CSTR locus is therefore required. 

As a first attempt at solving the CSTR locus, a standard numerical root finding routine will be employed. MAT-LAB s built-in solver routine, fsolveO, is used here for convenience, although many other numerical routines may be equally appropriate. A residence time and feed concentration to the CSTR must be supplied in order to 

a smooth locus of points for the feasible CSTR effluent concentrations is generated, which are represented by x s in the figure. The results indicate that a higher concentration of B is achieved in a CSTR, in the range 0.85 mol/L Ca 0.10 mol/L. For values of c outside of this range, larger Cg values are achieved in a PFR. 

4 CSTR Locus with Different Initial Guesses In Section 5.4.1, we indicated that the kinetics supplied for this system is known to exhibit multiple steady states however, no such concentrations have been identified in the computation of the current candidate region. A much larger set of achievable concentrations is in fact achievable. To see this, consider solving for the CSTR locus from the feed using a different initial guess. 

---