Professional Documents
Culture Documents
UNIT-3: Department of Chemical Engineering Sathyabama University
UNIT-3: Department of Chemical Engineering Sathyabama University
UNIT-3
INTRODUCTION
There are many ways of processing a fluid: in a single batch or flow reactor, in a chain of
reactors possibly with interstage feed injection or heating and in a reactor with recycle of the
product stream using various feed ratios and conditions. The reactor system selected will
influence the economics of the process by dictating the size of the units needed and by fixing the
ratio of products formed. The first factor, reactor size, may well vary a hundredfold among
competing designs while the second factor, product distribution, is usually of prime
consideration where it can be varied and controlled. For single reactions product distribution is
fixed; hence, the important factor in comparing designs is the reactor size. The size comparison
of various single and multiple ideal reactor systems is considered.
Batch Reactor
The batch reactor has the advantage of small instrumentation cost and flexibility of operation
(may be shut down easily and quickly). It has the disadvantage of high labor and handling cost,
often considerable shutdown time to empty, clean out, and refill, and poorer quality control of
the product. Hence it may be generalized to state that the batch reactor is well suited to produce
small amounts of material and to produce many different products from one piece of equipment.
On the other hand, for the chemical treatment of materials in large amounts the continuous
process is nearly always found to be more economical.
- rA = - 1 dNA
----------- = kCAn
V dt
To provide a quick comparison of the performance of plug flow with mixed flow reactors,the
performance graph is used. For identical feed composition CAo and flow rate FA, the ordinate of
this figure gives directly the volume ratio required for any specified conversion.
1. For any particular duty and for all positive reaction orders the mixed reactor is always larger
than the plug flow reactor. The ratio of volumes increases with reaction order.
2. When conversion is small, the reactor performance is only slightly affected by flow type. The
performance ratio increases very rapidly at high conversion; consequently, a proper
representation of the flow becomes very important in this range of conversion.
3. Density variation during reaction affects design; however, it is normally of secondary
importance compared to the difference in flow type. Dashed lines represent fixed values of the
dimensionless reaction rate group, defined as
For reactions with arbitrary but known rate the performance capabilities of mixed and plug flow
reactors are best illustrated in above figure. The ratio of shaded and of hatched areas gives the
ratio of space-times needed in these two reactors. The rate curve drawn in above figure is typical
of the large class of reactions whose rate decreases continually on approach to equilibrium (this
includes all nth-order reactions, n > 0). For such reactions it can be seen that mixed flow always
needs a larger volume than does plug flow for any given duty.
Consider N plug flow reactors connected in series, and let X1, X2, . . . , XN, be the fractional
conversion of component A leaving reactor 1, 2, . . . , N. Basing the material balance on the feed
rate of A to the first reactor, we find for the
ith reactor
V/ FO =Ʃ Vi / FO = ∫ dX / ( -r)
Hence, N plug flow reactors in series with a total volume V gives the same conversion as a
single plug flow reactor of volume V.
Branch D consists of two reactors in series; hence, it may be considered to be a single reactor of
volume
VD = 50 + 30 = 80 liters
Now for reactors in parallel V/F must be identical if the conversion is to be the same in each
branch. Therefore, two-thirds of the feed must be fed to branch D.
In plug flow, the concentration of reactant decreases progressively through the system; in mixed
flow, the concentration drops immediately to a low value. Because of this fact, a plug flow
reactor is more efficient than a mixed flow reactor for reactions whose rates increase with
reactant concentration, such as nth-order irreversible reactions, n > 0. Consider a system of N
mixed flow reactors connected in series. Though the concentration is uniform in each reactor,
there is, nevertheless, a change in concentration as fluid moves from reactor to reactor. This
stepwise drop in concentration, illustrated in above figure, suggests that the larger the number of
units in series, the closer should the behavior of the system approach plug flow. This will be
shown to be so. Quantitatively evaluate the behavior of a series of N equal-size mixed flow
reactors. Density changes will be assumed to be negligible; hence ρ = 0 . As a rule, with mixed
flow reactors it is more convenient to develop the necessary equations in terms of concentrations
rather than fractional conversions .
(a) Find the conversion for the same treatment rate. For the single reactor at 90% conversion
we have from chart
kCoτ = 90
For the two reactors the space-time or holding time is doubled; hence, the operation will be
represented by the dashed line of chart where
kCoτ= 180
This line cuts the N = 2 line at a conversion X = 97.4%, point a.
(b) Find the treatment rate for the same conversion. Staying on the 90% conversion
line, we find for N = 2 that
kCoτ = 27.5, point b
Comparing the value of the reaction rate group for N = 1 and N = 2, we find
Since V2 = 2V1 the ratio of flow rates becomes
Thus, the treatment rate can be raised to 6.6 times the original.
For arbitrary kinetics in mixed flow reactors of different size, two types of questions may be
asked: how to find the outlet conversion from a given reactor system, and the inverse question,
how to find the best setup to achieve a given conversion. Different procedures are used for these
two problems.
A graphical procedure for finding the outlet composition from a series of mixed flow reactors of
various sizes for reactions with negligible density change has been presented by Jones (1951).
All that is needed is an r versus C curve for component A to represent the reaction rate at various
concentrations. It is illustrated by the use of this method by considering three mixed flow
reactors in series with volumes, feed rates, concentrations, space-times (equal to residence times
because ρ = O), and volumetric flow rates as shown in figure below. Now from equation noting
that ρ = 0, we may write for component A in the first reactor
Plot the C versus r curve for component A and it is as shown in figure above. To find the
conditions in the first reactor note that the inlet concentration Co is known (point L), that C, and
( - r ) , correspond to a point on the curve to be found (point M), and that the slope of the line
LM = MNINL = (-r),l (C, - Co) = -(l/rl) from equation. Hence, from Co draw a line of slope -
(l/r,) until it cuts the rate curve; this gives C1. Similarly, we find from equation, that a line of
slope -(1/r2) from point N cuts the curve at P, giving the concentration C2 of material leaving the
second reactor. This procedure is then repeated as many times as needed. With slight
modification this graphical method can be extended to reactions in which density changes are
appreciable.
To find the minimum size of two mixed flow reactors in series to achieve a specified conversion
of feed which reacts with arbitrary but known kinetics, the following procedure is followed. The
basic performance expressions, from equations, give, in turn, for the first reactor
τ 1/ Co = X 1 / (-r1)
and for the second reactor
τ 2 / Co = X 2- X 1 / (-r2)
as the intermediate conversion XI changes, so does the size ratio of the units (represented by the
two shaded areas) as well as the total volume of the two vessels required (the total area shaded).
Figure shows that the total reactor volume is as small as possible (total shaded area is minimized)
when the rectangle KLMN is as large as possible. This brings us to the problem of choosing XI
(or point M on the curve) so as to maximize the area of this rectangle. This is done according to
Maximization of Rectangles concept.In the below figure, construct a rectangle between the x-y
axes and touching the arbitrary curve at point M(x, y). The area of the rectangle is then
A =xy
This area is maximized when
da = 0 =xdy+ydx
or when –dy/dx =y/x
In words, this condition means that the area is maximized when M is at that point where the
slope of the curve equals the slope of the diagonal NL of the rectangle. Depending on the shape
of the curve, there may be more than one or there may be no "best" point. However, for nth-order
kinetics, n > 0, there always is just one "best" point. The optimum size ratio of the two reactors is
achieved where the slope of the rate curve at M equals the diagonal NL. The best value of M is
shown in figure, and this determines the intermediate conversion X, as well as the size of units
needed. The optimum size ratio for two mixed flow reactors in series is found in general to be
dependent on the kinetics of the reaction and on the conversion level. For the special case of
first-order reactions equal-size reactors are best; for reaction orders n > 1 the smaller reactor
should come first; for n < 1 the larger should come first. However, Szepe and Levenspiel (1964)
show that the advantage of the minimum size system over the equal-size system is quite small,
only a few percent at most. Hence, overall economic consideration would nearly always
recommend using equal-size units.
For the most effective use of a given set of ideal reactors we have the following general rules:
1. For a reaction whose rate-concentration curve rises monotonically (any nth-order reaction, n >
0) the reactors should be connected in series. They should be ordered so as to keep the
concentration of reactant as high as possible if the rate-concentration curve is concave (n > I),
and as low as possible if the curve is convex (n < 1). As an example, for the case of above figure
the ordering of units should be plug, small mixed, large mixed, for n > 1; the reverse order
should be used when n < 1.
2. For reactions where the rate-concentration curve passes through a maximum or minimum the
arrangement of units depends on the actual shape of curve, the conversion level desired, and the
units available. No simple rules can be suggested.
3. Whatever may be the kinetics and the reactor system, an examination of the l/(-r) vs. CA curve
is a good way to find the best arrangement of units.