Department of Chemical Engineering

Indian Institute of Technology (ISM), Dhanbad

End-Semester Examination (Winter, 2021-22)
CHC-307: Process Modelling and Simulation
Time: 3 Hrs Attempt all questions Marks: 48% of 100

1. Define the following with a suitable example (Need not to present detailed
Mathematical Equations/Derivation)
a. Deterministic Model
b. Empirical Model
c. Distributed Parameter Model
d. Accuracy & Precession [5x4]

Answer:
a. A deterministic model uses rigorous mathematical equations based on a solid
theoretical basis that provides a single solution for the modeled variables. Because this
type of model does not explicitly simulate the effects of data uncertainty or variability,
changes in model outputs are solely due to changes in model components.
Examples of deterministic models are the Navier-stokes equation, equation of
continuity, Molecular simulation, etc.

b. An empirical model is one where the structure is determined by the observed statistical
relationship among experimental data. These models can be used to develop useful
relationships for forecasting and describing trends in behavior, but they are not
necessarily mechanistically relevant. They don’t explain the real causes and
mechanisms for the relationships.
Examples are the Arrhenius equation, Newton’s law of cooling, and Weather prediction
models.

c. Distributed parameter models are those in which the unknown (dependent) variable is
a function of a spatial coordinate and the time. Such systems are described by partial
differential equations.
Examples: Unsteady state heat conduction, Shrinking core model.

d. Accuracy is the closeness of the results (computed or modeled) to their “true” value.
The “true” value is the value it would have if we had perfect information. Precision is
a measure of how close your results agree with each other. The closer each result is to
the other results, the more precise your computation.
Examples: Computer programs are normally precise but may or may not be accurate.
Empirical models like Dittus–Boelter equation and Arrhenius equation are precise but
not accurate.
 2. The well-known Arrhenius rate expression
/
= .
is used to find rate constants as a function of temperature (where the universal gas
constant, = 8.3145 J/mol.K). For the following experimental data, find the
coefficients and by linear regression.
k (1/s) 0.002 0.004 0.02 0.04
T (K) 300 320 340 360
∑ .∑( ) ∑( ).∑( . )
For linear regression, intercept = and
.∑ (∑ )
∑( . ) .∑
slope = , where N = number of data points. [20i]
∑
Solution: (Following two alternatives are identical, equal morks for both of them)
/
= . is linearized by taking log on both sides
ln = ln − which is of the form = + where = ln , = 1/ , = − / , and
= ln
k (1/s) T (K) = 1/ = ln ∗
0.002 300 0.0033 -6.21 1.09E-05 -0.0205
0.004 320 0.0031 -5.52 9.61E-06 -0.0171
0.02 340 0.0029 -3.91 8.41E-06 -0.0113
0.04 360 0.0028 -3.21 7.84E-06 -0.009

0.0121 -18.85 3.676E-05 -0.0579

. × ×( . ) . ×( . )
= × . × ( . )
= 12.165 Therefore, = 1.91951 × 10 (1/s)
. . × .
= . ×
= −5579,

Therefore, for =− / ; = 5579 ∗ 8.3145 = 46386 (J/mol)

Alternatively:

log = log − ∗ log which is of the form = + where = ln ,

= 1/ , = − ∗ log , and = log
k (1/s) T (K) = 1/ = log ∗
0.002 300 0.0033 -2.699 1.089E-05 -0.00891
0.004 320 0.0031 -2.398 9.610E-06 -0.00743
0.02 340 0.0029 -1.699 8.410E-06 -0.00493
0.04 360 0.0028 -1.398 7.840E-06 -0.00391

0.0121 -8.194 3.675E-05 -0.02518

. × ( . ) . ×( . )
= ( . )
= 6.0144 and = 10 = 1.033 × 10 (1/s)
× . ×

. . × .
slope = = −2665.42,
. ×
× . ∗ .
=− =− = 51030
.
Comments on the Two Alternative solutions
Estimated values by two methods:
Natural Log, ln( ) log ( )
A 1.919 × 10 10.33 × 10
E 46386 51030

There is a significant difference in Frequency Factor and Activation Energy, but

the final results are very similar (see the graph given below). This behavior is
known as the kinetic compensation effect.
0.045
0.04
0.035
0.03
0.025
k. 1/s

0.02
0.015
0.01
0.005
0
290 300 310 320 330 340 350 360 370
T, K
 3. Two binary mixtures M and N of the same
components having known composition,
temperature, pressure, and molar flowrate are
mixed adiabatically to get a product P. The
pressure of each stream and the mixer are the
same.
(a) Develop the model equations (analogous to
the MESH equation for an equilibrium stage), and
(b) determine the degree of freedom. [15+5]
Solution:
(First Part)
Two feeds, M and N, coming into the mixer are known, and product P is unknown.
Let the known composition (mole fraction) of M be and that of N is where i = 1, 2
for binary mixture of same component (if binary mixture of different compositions are
assumed. i = 1, 2, 3, 4).
Temperatures of M and N are and , respectively.
The pressure of all the streams are same, i.e., and
Molar flowrates of M is (mole/s) and that of N is (mole/s).
Since temperature, pressure, and composition of M and N are known, their enthalpies
and are also known. Variables of the unknown stream to be determined are composition
( ), temperature (1 ), molar flowrate (mole/s) (1 ), and
the vapor fraction = (1 ).
The MESH equations for this system are as follows:

Material Balance:
= + (C number)
Equilibrium:
= (This equation is not applicable since no vapor and liquid are in equilibrium)
Summation
∑ =1 (1 equation)
If overall material balance = + is considered, ∑ = 1 no longer
remains an independent equation
Proof: Summing all the component balance equations, we have
∑ = ∑ + ∑ = ∑ = +
which, along with + = leads to ∑ = 1.
Enthalpy
. . ( , )+ (1 − )ℎ ( , )= . + . (1 equation)
We have total ( + 3) of variables (Unknowns), viz. , , and and ( +
2) of equations.
Thus the DOF is ( + 3) − ( + 2) = 1.
This DOF can be visualized by considering the system. When two streams are mixed, the
product will attain certain temperature and pressure. The vapor fraction will adjust naturally
(freely) at this temperature and pressure. This is why DOF appears to be 1. However, knowing
the concentration, temperature, and pressure, the vapor fraction can be modeled as follows:
(Second Part):
In addition to the above variables, if mole fraction of vapor and liquid fractions of product P,
( and ), and molar flow rates of vapor and liquid phases V and L are also added (i.e.,
2C+2 new variables added), than following additional equations apply

= (1 equation)

= + (1 equation)
= + (C equations)
= ( , , , ). (C equations)
∑ −∑ =0 (1 equation)
(We have only one summation equation since we have applied overall and all the components
material balance)
. . ( , )+ (1 − )ℎ ( , )= . + .
This enthalpy equation is not new, it has already appeared in first part.
Therefore 2C+3 new equations are added. Thus total number of equations becomes
( + 2) + (2 + 3) = 3 + 5
And total number of variables becomes
( + 3) + (2 + 2) = 3 + 5
Number of variables and number of equations are equal: DOF = 0
This can be visualized by considering the system when two streams are mixed, the product will
attain certain temperature at the same pressure. At this temperature and pressure, the vapor
fraction will adjust naturally according to its enthalpy content.

Merging both the above mentioned parts we have:

Material Balance:
+ = + (C number)
Equilibrium:
= ( , , , ). (C equations)
Summation
∑ =1 (1 equation)
∑ =1 (1 equation)
Enthalpy
. ( , )+ .ℎ ( , )= . + . (1 equation)
We have total (2 + 3) of variables (Unknowns), viz. , , , , and ; and same
(2 + 3) of equations. Thus the DOF is (3 + 5) − (3 + 5) = 0.
 4. What are (a) initial value problems and (b) boundary value problems?
A 0.5 m long cylindrical metallic rod of 10 cm diameter is exposed to ambient air at
30°C. The rod is being heated from one end by a heat source at 100°C supplying heat
to the rod such that the temperature gradient at the hot end is -76.4 °C/m. Determine
the temperature profile of the rod. The metal’s thermal conductivity (k), and heat
transfer coefficient (h) are 250 W/m.K. and 1.5 W/m2.K, respectively. Assume k and h
are constant all along the length. The recursive equation for one-dimensional heat
conduction with heat loss comes to be [20]
− (2 + ) + =− ∞

where = .
= .
, =2 Δ , and =
Take the length of each volume element, = . ( )
Solution:
When more than one conditions are required to completely define a problem, two situations
may arise
(a) All the conditions are defined at the same value of an independent variable (same
point), the problem is called initial value problem.
(b) At least one condition is defined at different value of independent variable (at a
different point), the problem is called boundary value problem.

The given numerical problem is an initial value problem as both the conditions are specified
at x = 0.
From the definition of the problem, we have
= −76.4 (°C/m), = 100 °C, = 30°C, = 250 (W/m.°C), ℎ = 1.5 (W/m2.°C),
Δ = 0.1 (m), = 0.05 (m).

With these, circumferential area =2 Δ = 2 ∗ 3.141 ∗ 0.05 ∗ 0.1 = 0.03141

Cross-sectional area = = 0.00785
∗ . ∗( . )
And = .
= ∗ .
= 0.0024.
Temperature of the first volume element can be estimated by the definition of derivative
( )
≈ /

Giving:
(−76.4)0.1
= + or = 100 + = 96.18

With the give temperature gradient at = 0, we can the estimate rate of heat coming in as follows
=− = −250 ∗ 0.00785 ∗ (−76.4) = 150 ( / )
=0
Applying Energy balance over the first volume element we get
= +
( 1− 2)
Where =ℎ ( − ) and =− =
Δ
=Δ
( 1− 2)
Which gives − = +ℎ ( − )
=0 Δ
 Δ
Someone may use following alternative:
Or − Δ = 1 − 2 + ( − ) ( − )
=0 = 1 0 Δ + (1 + ) 1 −
Δx/2
, giving
Or = Δ + (1 + ) 1 − , giving = (3 + ) −2 −
=0

= −76.4 ∗ 0.1 + (1 + 0.0024) ∗ 96.18 − 0.0024 ∗ 30 = 88.7 °

With the known values of and , the recursive equation can be applied to estimate
temperature of subsequent volume elements.
= (2 + ) − − ∞

= (2.0024) − − 0.072

Therefore
= 2.0024 ∗ − − 0.072 = 2.0024 ∗ 88.7 − 96.18 − 0.072 = 81.36
= 2.0024 ∗ − − 0.072 = 2.0024 ∗ 81.36 − 88.7 − 0.072 = 74.14
= 2.0024 ∗ − − 0.072 = 2.0024 ∗ 74.14 − 81.36 − 0.072 = 67.02
N=5 is the last volume element
5. Derive the model equations (partial differential equations) and the recursive equations
for component A and reactant R by applying material balance over the shaded volume
element (shown in the figure given below) in the case of Mass Transfer of A from a
Gas into a Reacting Falling Liquid Film. The liquid feed consists of solvent B and a
reactant R. Component A is absorbed at the gas-liquid interface, and then it diffuses in
the falling liquid film, where it reactants with R in the liquid phase. The first-order
reaction + → take place in the liquid phase, in which k is the first-order reaction
rate constant (− = ), independent of the concentration of R. Assume R and P
are nonvolatile and remain in the liquid phase. Inlet concentrations of A and R in the
liquid feed from the top are and , respectively, and the concentration of A at
the gas-liquid interface is , . The velocity in the y-direction (u) inside the falling
liquid film is given by
3
= 1−
2
where is the average velocity of the liquid, and δ is the film thickness. The
diffusion of R in the liquid film can be neglected since R is nonvolatile.
[20]
CA0 , CR0 x

1
y
2

CA,eq

j–1

j Gas Phase
j+1 containing A

M–1

N N–1 i+1 i i–1 2 1

δ
Molar fluxes of component A across the shaded volume
element ( , ) given in the question is shown in Figure 1.
For the material balance in the steady-state, we have
− , −
Rate of moles of Rate of moles of Rate of moles of
= +
A in A out A consumed

Which gives:
− Δ . +( . )| Δ .
Figure 1
= − Δ . +( . )| Δ . + (Δ Δ )

After rearranging and dividing throughout by Δ . Δ . we get

( . )| ( . )|
+ − . =0 ( . )| ( . )|
lim + lim − . =0
→ →
In terms of finite-difference notations, we have Gives + − =0
, , , , . .
, ,
+ − . =0 (1)

After rearrangement, the recursive equation in the x-direction comes to be

,
− (2 + + ) ,
+ ,
=− ,
Where = and = (2)

For = (BC-1)
, ,
− ≈−
/
Substituting in equation (1)
, , , ,
. .
, ,
+ − . =0
−(3 + + ). + ,
=− ,
+2 , (3)

For = (BC-2)
− =0
Then equation (1) becomes
, , . .
, ,
+ − . =0

− , ,
(1 + + ) = − ,
(4)
Equations (2), (3) and (4) forms the following recursive matrix equation in y-direction (for All j = 2, 3, …)

−(3 + + )
1 0 0 0 , = − ,
+2 ,
−(2 + + ) −
1 1 0 , ,
−(2 + + ) −
0 1 1 0 , ,

−(1 + + ) −
0 0 0 1 , ,

(5)
For = ( − ) the matrix equation for all i = 1, 2, …, N is
−(3 + + ) − +2
1 0 0 0 , = ,
−(2 + + ) −
1 1 0 ,
−(2 + + ) −
0 1 1 0 ,

−(1 + + ) −
0 0 0 1 ,

(6)
Molar fluxes of the component R across the shaded volume
element ( , ) given in the question is shown in Figure 2.
For the material balance in the steady-state, we have
,

Rate of moles of Rate of moles of Rate of moles of

= –
R in R out R formed

Which gives:
( . )| Δ . = ( . )| Δ . − (Δ Δ ) Figure 2
After rearranging and dividing throughout by Δ . Δ . we get
( . )| ( . )|
+ . =0 ( . )| ( . )|
lim − . =0
→
In terms of finite-difference notations, we have Gives − =0
. .
, ,
+ . =0 (7)

After rearrangement, the recursive equation in the x-direction comes to be

,
= ,
+ ,
Where = (8)

For = (BC – 4)
,
=
Substituting in equation (8)
,
= + ,
(9)
Where , is evaluated by solving the matrix equation (6). Thus the sequence of calculation (Algorithm) would
be
1. Solve the matrix Equation (5) for = 1
2. Find , for all i from equation (9)
3. Solve the matrix Equation (4) for next j
4. Find , for all i from equation (8)
5. Repeat steps 3. And 4. Up to the end of the falling film.