Instrumentation and Process

Control
Lecture 7
Mathematical Models of Chemical Processes
Lecture 7

1. Stirred-Tank Heating Process: Constant Holdup

2. Stirred-Tank Heating Process: Variable Holdup
3. Electrically Heated Stirred Tank: Constant Holdup
4. Steam-Heated Stirred Tank: Constant Holdup
5. Liquid Storage Systems
6. Continuous Stirred-Tank Reactor (CSTR)
7. Staged Systems (a Three-Stage Absorber)
8. Fed-Batch Bioreactor

1
 Electrically Heated Stirred Tank:
Constant Holdup
Lecture 7

Figure 2.3 Stirred-tank heating process with constant holdup, V.

 Electrically Heated Stirred Tank:
Constant Holdup
Assumptions:

1. The metal heating element has a significant thermal

Lecture 7

capacitance and that the electrical heating rate Q directly

affects the temperature of the element rather than the
liquid contents.
2. Neglect the temperature gradients in the heating element
(for simplicity) that result from heat conduction and
assume that the element has a uniform temperature (Te).
This temperature can be interpreted as the average
temperature for the heating element.
 Model Development - I
• The unsteady-state energy balances for the tank:

where m = Vρ and mC is the thermal capacitances of the tank contents

Lecture 7

• The unsteady-state energy balances for the heating

element:

meCe is the product of the mass of metal in the heating element and its
specific heat (thermal capacitances of the heating element)
heAe is the product of the heat transfer coefficient and area available for
heat transfer
Q is an input variable: electrical power dissipation in the heating
element
 • The model given by Eqs. 2-47 and 2-48 in suitable form for
calculation of the unknown output variables Te and T?
• Two output variables and two differential equations
• All of the other quantities must be either model
parameters (constants) or inputs (known functions of
Lecture 7

time)
• For a specific process, m, C, me, Ce, he, and Ae are known
parameters determined by the design of the process,
(materials of construction, and its operating conditions)
• Input variables w, Ti, and Q must be specified as
functions of time for the model to be completely
• The dynamic model can then be solved for T and Te as
functions of time by integration after initial conditions are
specified for T and Te
 Model Development - III
• If flow rate w is constant, Eqs. 2-47 and 2-48 can be
converted into a single second-order differential equation.
• Solve Eq. 2-47 for Te and then differentiate to find dTe/dt
• Substituting the expressions for Te and dTe/dt into Eq. 2-48
yields
Lecture 7

• The model in Eq. 2-49 can be simplified when meCe (the

thermal capacitance of the heating element) is very small
compared to mC
• When meCe = 0, Eq. 2-49 reverts to the first-order model,
Eq. 2-36, which was derived for the case where the heating
7
 Model Development - IV
• The model of Eq. 2-49 consists of only a single equation
and a single output variable (T).
• The intermediate variable (Te), is less important than T
and has been eliminated from the earlier model (Eqs. 2-47
Lecture 7

and 2-48).
• To integrate Eq. 2-49, we require initial conditions for both
T and dT/dt at t = 0, because it is a second-order
differential equation. The initial condition for dT/dt can be
found by evaluating the right side of Eq. 2-47 when t = 0,
using the values of Te(0) and T(0). For both models, the
inputs (w, Ti, Q) must be specified as functions of time
 Example 2.4 - I
An electrically heated stirred-tank process can be modeled
by Eqs. 2-47 and 2-48 or, equivalently, by Eq. 2-49 alone.
Process design and operating conditions are characterized
by the following four parameter groups:
Lecture 7
 Example 2.4 - II
a) Calculate the nominal steady-state temperature (T).
b) Assume that the process is initially at the steady state
determined in part (a). Calculate the response, T(t), to a
sudden change in Q from 5000 to 5400 kcal/min using
Lecture 7

Eq. 2-49. Plot the temperature response.

c) Suppose that it can be assumed that the term meCe/heAe
is small relative to other terms in Eq. 2-49. Calculate the
response T(t) for the conditions of part (b), using a first-
order differential equation approximation to Eq. 2-49.
Plot T(t) on the graph for part (b).
d) What can we conclude about the accuracy of the
approximation for part (c)?
 Example 2.4 - III
a) The steady-state form of Eq. 2-49 is

Substituting parameter values gives Ť= 350 °C.

Lecture 7

b) Substitution of the parameter values in Eq. 2-49 gives

The following solution can be derived using standard

solution methods (Kreyszig, 2011):

This response is plotted in Fig. 2.4 as the dashed curve (a).

Lecture 7 Example 2.4 - IV

Figure 2.4 Responses of an electrically heated stirred-tank process to a sudden change in the
heater input.
 Example 2.4 - V
c) If we assume that meCe is small relative to other terms,
then Eq. 2-49 can be approximated by the first-order
differential equation:
Lecture 7

The solution is
 Example 2.4 - VI
d) Figure 2.4 shows that the approximate solution (b) is
quite good, matching the exact solution very well over
the entire response. For purposes of process control,
this approximate model is likely to be as useful as the
Lecture 7

more complicated, exact model.

 Steam Heated Stirred Tank: Constant
Holdup
• Steam (or some other heating medium) can be
condensed within a coil or jacket to heat liquid in a
stirred tank, and the inlet steam pressure can be varied
Lecture 7

by adjusting a control valve.

• The condensation pressure (Ps) then fixes the steam
temperature (Ts) through an appropriate thermodynamic
relation or from tabular information such as the steam
tables.
 Steam Heated Stirred Tank: Constant
Holdup
• Consider the stirred-tank heating system of Section 2.4.1
with constant holdup and a steam heating coil.
• We assume that the thermal capacitance of the liquid
Lecture 7

condensate is negligible compared to the thermal

capacitances of the tank liquid and the wall of the
heating coil.
• This assumption is reasonable when a steam trap is
used to remove the condensate from the coil as it is
produced.
 Steam Heated Stirred Tank: Constant
Holdup
• The dynamic model of energy balances on the liquid:
Lecture 7

• The dynamic model of energy balances on the heating

coil wall:

• where the subscripts w, s, and p refer, respectively, to the

wall of the heating coil and to its steam and process
sides.
• Note that these energy balances are similar to Eqs. 2-47
and 2-48 for the electrically heated example.
 Continuous Stirred-Tank Reactor
(CSTR)
• Continuous stirred-tank reactors (CSTR) have
widespread application in industry
• CSTR models tend to be simpler than models for other
Lecture 7

types of continuous reactors such as tubular reactors and

packed-bed reactors
• Consider a simple liquid-phase, irreversible chemical
reaction where chemical species A reacts to form species
B. The reaction can be written as A → B.
 Continuous Stirred-Tank Reactor
(CSTR)
• Assume that the rate of reaction is first-order with respect
to component A:
Lecture 7

– where r is the rate of reaction of A per unit volume, k is the

reaction rate constant (with units of reciprocal time), and c A is
the molar concentration of species A.
• For single-phase reactions, the rate constant is typically a
strong function of reaction temperature given by the
Arrhenius relation:

– where k0 is the frequency factor, E is the activation energy, and

R is the gas constant.
 Continuous Stirred-Tank Reactor
(CSTR)
• Initial CSTR model development is based on three
assumptions:
1. The CSTR is perfectly mixed.
Lecture 7

2. The mass densities of the feed and product streams are

equal and constant. They are denoted by ρ.
3. The liquid volume V in the reactor is kept constant by an
overflow line.
 Unsteady-state mass and component
balances for species A
The unsteady-state mass balance for the CSTR is
Lecture 7

Because V and ρ are constant, Eq. 2-64 reduces to

The unsteady-state component balances for species A

(in molar concentration units) is
 Unsteady-state energy balance for
CSTR
Assumptions:
4.The thermal capacitances of the coolant and the cooling coil wall are negligible
compared to the thermal capacitance of the liquid in the tank.
5.All of the coolant is at a uniform temperature, Tc. (That is, the increase in
Lecture 7

coolant temperature as the coolant passes through the coil is neglected.)

6.The rate of heat transfer from the reactor contents to the coolant is given by
4. Q = UA(Tc − T) (2-67)
5. where U is the overall heat transfer coefficient and A is the heat transfer area. Both of
these model parameters are assumed to be constant.
7.The enthalpy change associated with the mixing of the feed and the liquid in
the tank is negligible compared with the enthalpy change for the chemical
reaction. In other words, the heat of mixing is negligible compared to the heat of
reaction.
8.Shaft work and heat losses to the ambient can be neglected.
 Unsteady-state energy balance for
CSTR
CSTR energy balance is convenient for analysis and can
be derived from Eqs. 2-62 and 2-63 and Assumptions 1–
8
Lecture 7

where ΔHR is the heat of reaction per mole of A that is

reacted.
 EXAMPLE 2.5
To illustrate how the CSTR can exhibit nonlinear dynamic behavior, we
simulate the effect of a step change in the coolant temperature T c in
positive and negative directions. Table 2.3 shows the parameters
and nominal operating condition for the CSTR based on Eqs. 2-66
and 2-68 for the exothermic, irreversible first-order reaction A → B.
Lecture 7

The two state variables of the ODEs are the concentration of A (cA)
and the reactor temperature T. The manipulated input variable is the
jacket water temperature (Tc). Two cases are simulated, one based
on increased cooling by changing Tc from 300 to 290 K and one
reducing the cooling rate by increasing Tc from 300 to 305 K. These
model equations are solved in MATLAB with a numerical integrator
(ode15s) over a 10-min horizon. The decrease in T c results in an
increase in cA. The results are displayed in two plots of the
temperature and reactor concentration as a function of time (Figs.
2.7 and 2.8).
 Lecture 7

25
 Figure 2.7 Reactor temperature variation with step
changes in cooling water temperature from 300 to 305
K and from 300 to 290 K.

At a jacket temperature of 305 K, the

reactor model has an oscillatory
response. The oscillations are
characterized by apparent reaction
run-away with a temperature spike.
Lecture 7

However, when the concentration

drops to a low value, the reactor then
cools until the concentration builds,
then there is another temperature
rise. It is not unusual for chemical
reactors to exhibit such widely
different behaviors for different
directional changes in the operating
conditions.

Figure 2.8 Reactant A concentration variation with step

changes in cooling water temperature to 305 and 290 K.
26