CRE - 1

■ Thermodynamics, chemical kinetics, classification of reactions, variables affecting the

rate of reaction, definition of reaction rate
■ Kinetics of Homogeneous Reactions
– Concentration dependent term of a rate equation, single & multiple reactions,
elementary & non elementary reactions, kinetic view of equilibrium for elementary
reactions, molecularity & order of reaction, rate constant,
– Representation of a rate constant, kinetic models for non elementary reactions,
temperature dependence of rate equation, Arrhenius theory, collision theory &
transition state theory, comparisons of the theories
■ Interpretation of Batch Reactor Data
– constant volume batch reactor, integral method & differential method of analysis of
data, variable volume batch reactor, integral method of analysis, differential
method of analysis

1
CRE - 1
■ Introduction to Reactor Design
– material and energy balances
■ Single Ideal Reactors
– ideal batch reactor, space time & space velocity, steady state mixed flow reactor,
steady state plug flow reactor, holding time & space time for flow systems
■ Design for Single Reactions
– size comparison of single reactors, batch reactor, mixed reactor versus plug flow
reactors, first & second order reactions, variation of reaction rates for second
order reactions, general graphical comparison, multiple reactor system, plug flow
reactors in series and/or in parallel, equal sized mixed reactors in series, mixed
flow reactors of different sizes, reactors of different types in series, recycle reactor

2
CRE - II

3
Residence Time Distribution of Fluid in Vessels
■ E, the age distribution of fluid leaving vessel, the F curve, the C curve,
■ The mean & variance ways of using age distribution information
■ Relation among E, F & C curves & mean time for closed vessel
■ experimental methods and finding RTD by experiment, useful mathematical tools
■ linear & non-linear systems with flow models,
■ conversion directly from tracer information, model for non-ideal flow dispersion model,
■ fitting the dispersion model for small extents of dispersion & large extent of dispersion,
■ chemical reaction & dispersion, tanks in series model

4
Chemical Reactor Schematic
Input Ouput

Ideal

■ Kinetics ■ Contacting
– Chemical – Batch
– Physical – Continuous
■ Plug Flow
■ Mixed Flow

Reactor Performance = f (Input, kinetics, contacting)

5
 Ideal Reactor
1. Spatial distribution insignificant
A. Composition or pressure
B. Temperature
■ Contacting C. Velocity
– Batch

2. In batch, the composition changes only with time

■ Continuous
– Mixed Flow
3. In mixed flow reactor the concentration at exit is
– Plug Flow considered same as inside the reactor

4. In PFR, if the concentration changes along the length.

can it be considered ideal?

6
 non-ideal (Real) Reactor
Reaction Kinetics Effect of Mass
Rate law –rA =k 𝐶𝑟𝑛 Transfer

Reactor Design
and operation

Effect of Velocity distribution Heat Effects

(Flow pattern)

These effects increases the non-ideal behavior of the reactor

 CRE-II is to understand reactor which incorporates these complexities

 May be one of them or in different combination

7
Scenarios
■ We want to find ways of determining the dead zone volume and the fraction of the
volumetric flow rate bypassing/Shortcircuiting the system.
■ The three concepts
– RTD
– Mixing
– Model

8
 non-ideality-1
 Effect of velocity distribution
 Bypassing
 Inefficient mixing
 Presence of dead zones
 Heat Effects
 Non-isothermal
 Presence of hot spots
 Effect of mass transfer
 Both internal and external
 Example – Heterogeneous reactor adsorption and desorption

9
 non-ideality-2
■ Ideal Reactors
– PFR – No mixing
– CSTR – Complete mixing
■ Non ideal reactors are in between so always deviates from ideality.
– the reactant concentration is relatively high at the point where the feed enters and low in
the stagnant regions that develop in corners and behind baffles
– in a real plug flow reactor, fluid velocity profiles, turbulent mixing, and molecular
diffusion cause molecules to move with changing speeds and in different directions.
■ The deviations from ideal reactor conditions pose several problems in the design and analysis
of reactors. How to account for this? - Three interrelated factors needs to be known
– Residence Time Distribution (RTD)
– State of aggregation
– Earliness or lateness of mixing
■ Whether these factors can be ignored or are crucial depends on relative values of the following
– The time for reaction 𝑡𝑟𝑥𝑛
– The time for mixing 𝑡𝑚𝑖𝑥
10
 Earliness or lateness of mixing
■ Once the fluid elements enter they can mix
– As soon as they enter i.e. near inlet
– As late as after entry i.e. near outlet
■ Important when multiple reactant streams are entering
■ Which one is desirable?
 State of aggregation
■ Extremes these states can be called microfluids and macrofluids
■ Single phase close to these extremes
■ Multiple phase
– Solids always behaves as Macrofluids
– Gas and liquids depends on process

What exactly is needed

for efficient operation?
 Residence time distribution (RTD)
■ Cause of RTD (deviation from ideal flow pattern)
– Channeling
– Recycling
– Creation of stagnant region
■ Fluid elements taking different routes will spent different lengths of time
– This causes non-consistent, non-uniform conversion or product distribution
RTD
■ The age of a fluid element is defined as the time it has resided within the
reactor.
■ A fluid element is a small volume of fluid relative to the size of the
reactor yet sufficiently large to exhibit continuous properties such as
density and concentration
■ For characterization of non-ideal flow reactors we use
– The residence time distribution function E(t)
– the cumulative distribution function F(t)
– Integral age distribution I(t)
– the mean residence time tm
– the variance, σ2
■ Same is evaluated for ideal reactors as well to know how far our real
(non-ideal) reactor is off the norm from an ideal reactor
14
C-Curve
■ Measurement of concentration at outlet

15
E-Curve, Age Distribution -1
■ Elements of fluids taking different routes take different lengths of time to
pass through reactor and reach exit
■ The distribution of these times of stream leaving the vessel is called Exit
Age Distribution or Residence Time Distribution given by E-curve
■ This is defined as the fraction of material in the outlet stream that has
been in the system for the period between t and t + dt, and is equal to
E(t)dt
𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑜𝑓 𝑓𝑙𝑢𝑖𝑑 𝑙𝑒𝑎𝑣𝑖𝑛𝑔 𝑡ℎ𝑒 𝑣𝑒𝑠𝑠𝑒𝑙 𝑡ℎ𝑎𝑡
𝐸 𝑡 . 𝑑𝑡 =
ℎ𝑎𝑠 𝑟𝑒𝑠𝑖𝑑𝑒𝑛𝑐𝑒 𝑡𝑖𝑚𝑒 𝑜𝑓 (𝑡, 𝑡 + 𝑑𝑡)
■ In short, RTD shows the status of mixing
– for no mixing in plug flow
– Complete mixing in CSTR
– Any other reactor is in between.
16
E-Curve, Age Distribution -2
■ E(t) has the unit of inverse time (e.g. min–1)
■ In probability terms, E(t) is a density rather than a distribution.
■ It is a fundamental indicator of the flow and mixing pattern in a chemical
reactor
■ A tracer is introduced into a reactor the concentration of the tracer is
monitored at the exit stream, it is possible to ascertain the age distribution
of the fluid elements within the reactor.
■ The residence time distribution function must satisfy the following

E(t) = 0 for t < 0 since no fluid can exit before it enters

E(t) ≥ 0 for t > 0 since mass fractions are always positive

The E-curve is just the C-curve divided by the area under the C-curve 17
 E-Curve, Age Distribution -3
■ Fluid enters and leaves the reactor only one time – closed boundary
■ When fluid enters multiple time – open boundary
– Exit age estimated as fraction of exit stream between time t and t+dt

𝑡1
■ Fraction of stream younger than t1 𝐸 𝑑𝑡
0
∞ 𝑡1
■ Fraction of stream older than t1 𝐸 𝑑𝑡 = 1 − 𝐸 𝑑𝑡
𝑡1 0 18
F-Curve, Cumulative RTD
■ This function is the volume fraction of material in the outlet stream,
which has been in the system for times less than t is equal to F(t).
■ If “ages” are assigned to the different fluid elements leaving the system,
F(t) would be the volume fraction of the outlet stream having an age less
than t.
𝑡
𝑡 𝑑𝐹(𝑡) 0
𝐶 𝑡 𝑣 𝑑𝑡
𝐹 𝑡 = 𝐸(𝑡) 𝑑𝑡 = 𝐸(𝑡) 𝐹 𝑡 = ∞
0
𝑑𝑡 𝐶 𝑡 𝑣 𝑑𝑡
0

The properties of F(t) are:

• F(t) = 0 when t < 0
• F(t) ≥ 0 when t ≥ 0
• F(∞) = 1
• dF(t)/dt = E(t) > 0
The cumulative exit age distribution is a non-
negative, monotone non-decreasing function
19
INTERNAL AGE DISTRIBUTION FUNCTION I(t)
■ The fraction of the material within the system for times between t and t + dt is equal to
I(t)dt, where t is the age or length of time a fluid element has been in the vessel.
■ I(t) has properties similar to E(t) for a perfect mixer.

∞ 𝑡 𝑡
𝑣
𝐼(𝑡) 𝑑𝑡 = 1 𝐼 𝑡 = 1 − 𝐸(𝑡) 𝑑𝑡 𝑡𝐼 𝑡 = 1 − 𝐸(𝑡) 𝑑𝑡
0 𝑉 0 0

𝑣
𝐼 𝑡 = 1 − 𝐹(𝑡)
𝑉

1 −𝑡
■ Perfectly mixed CSTR 𝐼 𝑡 = 𝐸 𝑡 = 𝑒 𝑡
𝑡

For Perfectly mixed CSTR the exit age and internal age are same
20
Intensity function 𝛬(𝑡)
■ Fraction of fluid in the vessel of age t that will leave at a time between t and t+dt. The
intensity function Λ(t) is related to the E(t) and I(t) functions

𝑣 𝐸(𝑡)
𝛬 𝑡 =
𝑉 𝐼(𝑡)

■ The intensity function Λ(θ) is useful in detecting the existence of dead space and
bypassing, and it allows some insight into the basic mechanism of the flow processes

21
Intensity function 𝛬(𝑡)
■ The intensity function of a completely mixed vessel is a constant, Λ(θ) = 1, a straight
line parallel to the abscissa.
■ The intensity function of two or more identical, ideally mixed vessels in series starts
from the origin and increases monotonically to an asymptotic value (Λ → n), which is
equal to the number of vessels.
■ There are no stagnant zones in the system.

22
■ Various curves representing non-
ideal flow pattern

23
Moments of RTD -1
■ A recommended method of characterizing the RTD in flow systems is by using their
moments. They are –
– Mean (1st Moment)
– Variance (2nd Moment)
– Skewness (3rd Moment)
■ The mean value or the centroid of distribution for a concentration versus time curve is -

∞ 𝑡=∞
0
𝑡𝐶 𝑡 𝑑𝑡 𝑡=0 𝑡𝑖𝐶(𝑡𝑖)∆𝑡𝑖
𝜇=𝑡= ∞
𝜇=𝑡= 𝑡=∞
𝐶 𝑡 𝑑𝑡 𝑡=0 𝐶(𝑡𝑖)∆𝑡𝑖
0

∞ 𝑡=∞
0
𝑡𝐸 𝑡 𝑑𝑡 𝑡=0 𝑡𝑖𝐸(𝑡𝑖)∆𝑡𝑖
𝜇=𝑡= ∞ 𝜇=𝑡= 𝑡=∞
0
𝐸 𝑡 𝑑𝑡 𝑡=0 𝐸(𝑡𝑖)∆𝑡𝑖

24
Mean residence time-1
■ The average time spent by material flowing at volumetric rate v through a volume V is
known as the mean residence time 𝑡
𝑉
𝑡=
𝑣
■ The mean residence time 𝑡 requires two conditions
– There must not be a density change in the flowing stream as it passes through the
system secondly,
– There must be no backmixing of the material in the system either upstream past the
feed point, or of the product stream back into the system.

25
Mean residence time-2
∞ ∞
■ Mathematically 𝑉= 𝑢𝑑𝑡 𝐸 𝑠 𝑑𝑠
0 𝑡
∞ ∞
■ Constant volumetric flow 𝑉
𝑡= 𝐸 𝑠 𝑑𝑠 𝑑𝑡
𝑣 0 𝑡

∞
𝑉
𝑡= = 𝑡𝐸 𝑡 𝑑𝑡
𝑣 0

The right-hand side is just the mean residence time and we see that the mean
residence time is just the space time 

26
Moments of RTD -2
■ The second moment is taken about the mean and is referred to as the variance or square
of the standard deviation

∞ 𝑡=∞
𝑡 − 𝑡 2 𝐶 𝑡 𝑑𝑡 𝑡=0 𝑡 − 𝑡 2 𝐶(𝑡𝑖)∆𝑡𝑖
2
𝜎 = 0
∞
𝜎2 = 𝑡=∞
𝐶 𝑡 𝑑𝑡 𝑡=0 𝐶(𝑡𝑖)∆𝑡𝑖
0
∞ 𝑡=∞
𝑡=0 𝑡𝑖𝐸(𝑡𝑖)∆𝑡𝑖
𝜎2 = 𝑡 − 𝑡 2 𝐸 𝑡 𝑑𝑡 𝜇=𝑡= 𝑡=∞
0 𝑡=0 𝐸(𝑡𝑖)∆𝑡𝑖

■ The magnitude of the variance σ represents the square of the distribution spread and has
the units of (time)2.
■ The greater the value of this moment, the greater the spread of the RTD.
■ The variance is particularly useful for matching experimental curves to one of a family
of theoretical curves.
27
Moments of RTD -3
■ The third moment is taken about the mean and is known as the skewness, γ

∞
𝛾= 𝑡 − 𝑡 3 𝐸 𝑡 𝑑𝑡
0

■ The magnitude of this moment measures the extent that the distribution is skewed in one
direction or another in reference to the mean

28
Mean Mean

Variance

Skewness

29
Normalized Distribution-1
■ It is more convenient to use dimensionless time θ and corresponding RTD

𝑡 1
𝜃= 𝑑𝜃 = 𝑑𝑡
𝑡 𝑡
■ E(t) and E(θ) define same physical entity

𝐸 𝜃 𝑑𝜃 = 𝐸 𝑡 𝑑𝑡 𝐸 𝜃 𝑑𝑡 = 𝑡𝐸 𝑡 𝑑𝑡

■ For ideal CSTR, mass balance is given by -

𝑑𝑐 𝑑𝑐
𝑣𝑐𝑜 = 𝑣𝑐 + 𝑉 𝑡 + 𝑐 = 𝑐0
𝑑𝑡 𝑑𝑡

■ Apply Laplace transform and solving

30
Normalized Distribution-2
■ It is more convenient to use dimensionless time θ and corresponding RTD

𝐶(𝑡) −𝑡
=1−𝑒 𝑡 𝐹 𝜃
𝐶0
■ You can find out E(θ), I(θ) and other dimensionless terms easily

■ Find out the terms for plug flow reactor

■ Why you want to calculate in terms of normalized function ??

– it is possible to directly compare the flow performance inside different reactors.
– If the normalized function E(θ) is used, all perfectly mixed CSTRs have numerically
the same RTD. If E(t) is used, its numerical values can change for different CSTRs.

31
 RTD PFR
Output
Input

■ How long fluid elements reside in the

Concentration

Concentration
reactor

■ How to determine that? t0

Time
tt t0
Time
tt

– Pulse tracer input

CSTR Non-ideal Reactor
Input Output Input Output
Concentration

Concentration
Concentration

Concentration
t0 tt t0 tt t0 tt t0 tt
Time Time Time Time
32
Experimental methods of finding RTD
■ Tracer input and measure tracer concentration at exit
■ Non-reactive tracer, No chemical reaction

Easy Difficult

■ These are based on how the tracer is injected

– Pulse (C-Curve)
– Step (F-Curve)
– Periodic
– Random

33
Pulse Injection Experiment-1
Feed Outlet

Injection, N0 (kg or moles) Detection

■ A quantity m of tracer is injected at the inlet of a system during a period of time, which
is very short compared to the mean residence time, t . The concentration of this tracer
material is measured in the outlet stream as a function of time. Assume the following
conditions:
1. Constant flowrate u(l/min) and fluid density ρ(g/l).
2. Only one flowing phase.
3. Closed system input and output by bulk flow only (i.e., no diffusion takes place
across the system boundaries).
4. Flat velocity profiles at the inlet and outlet.
5. Linearity with respect to the tracer analysis, that is, the magnitude of the response
at the outlet is directly proportional to the amount of tracer injected.
6. The tracer is completely conserved within the system and is identical to the
process fluid in its flow and mixing behavior.
Pulse Injection Experiment
■ On what basis the tracer is chosen?

– It should not affect the flow.

– It should be injected in a short time and be conveniently analyzed by a suitable
method.
– It should not react with or be absorbed by the surface of the system.
– Its molecular diffusivity should be low and it should be conserved (i.e., a mass
balance on it must be possible).
– The tracer must be uniformly distributed in the inlet fluid, and the analysis of the
exit should give a proper average concentration in the outlet fluid.
Pulse Injection Experiment-2
Feed Outlet

Injection, N0 (kg or moles) Detection

■ Tracer of amount N0 is injected in one shot

■ Outlet concentration of tracer is measured with respect to time
■ The time Δt is chosen so small that the C(t) and C(t+ Δt) is essentially the same

The amount of tracer leaving at any time can be defined as ∆𝑁 = 𝐶 𝑡 𝑣∆𝑡 …1

Mass or mole / Volume Volumetric Flow rate,

mass or moles / sec
∆𝑁 𝐶 𝑡 𝑣
Dividing by Total Amount injected = ∆𝑡
𝑁0 𝑁0
…2
Represents fraction of material having RT between t and t+ Δt
36
Pulse Injection Experiment-3
𝐶 𝑡 𝑣
For pulse injection 𝐸(𝑡) = …3
𝑁0

∆𝑁
From Eq. 1, 2 & 3 = 𝐸(𝑡)∆𝑡 …4
𝑁0

■ E(t) is RTD function, defines how much time different fluid elements have spent in reactor
■ E(t)dt is the fraction of fluid exiting the reactor that spent time between t to t+dt

Feed Outlet

Detection
Injection, N0 (kg or moles)

37
Pulse Injection Experiment-4
𝑁0
Area = 𝑣

0
𝑣
E = C. 𝑁
C

■ To find the E curve from the Cpulse curve simply change the concentration scale such that
the area under the curve is unity. Divide the concentration readings by N0/v
■ What if N0 is not known??
– Sum up all the exit concentration to get N0

∞ 𝐶(𝑡)
𝑑𝑁 = 𝑣𝐶 𝑡 𝑑𝑡 𝑁0 = 𝑣𝐶 𝑡 𝑑𝑡 𝐸(𝑡) = ∞
0 0
𝐶 𝑡 𝑑𝑡
38
Step Injection Experiment-1
Feed Outlet

Injection, N0 (kg or moles) Detection

■ Feed to a system is switched instantaneously from one supply to another

1. The second input fluid must be clearly distinguishable from first
2. There should be no change in fluid propertied which affect flow like density and
viscosity
3. Second fluid, usually a dye is used to make distinction

Cout
Cin

t t
■ Consider v m3/s of fluid flowing through a vessel of volume V. Now at time t = 0 switch
from ordinary fluid to fluid with tracer of concentration
Ques : Till what time this change is maintained?
Ans : So that you observe same concentration at outlet
which was at inlet

■ A material balance relates the different measured quantities of the output curve of a
step input
40
Shaded Area

41
Step Injection Experiment-4
𝑡 𝑡
𝐶𝑜𝑢𝑡 = 𝐶0 𝐸 𝑡 𝑑𝑡 𝐶𝑜𝑢𝑡 = 𝐶0 𝐸 𝑡 𝑑𝑡
0 0

0 𝑡<0 𝐶𝑜𝑢𝑡 𝑡
𝐶0 𝑡 = = 𝐸 𝑡 𝑑𝑡 = 𝐹(𝑡)
𝐶0 𝑡≥0
𝐶0 0

𝐶𝑜𝑢𝑡 𝑑 𝐶(𝑡)
𝐹 𝑡 = 𝐸 𝑡 =
𝐶0 𝑠𝑡𝑒𝑝
𝑑𝑡 𝐶0 𝑠𝑡𝑒𝑝

■ Like EAD function E(t) can be directly estimated from pulse input the CAD function
F(t) can be estimated directly from Step input function.
■ Consider a step change at time t = 0 and change is given by following function
Relation Between E and F curve
■ To relate E with F imagine a steady flow of white fluid. Then at time t = 0 switch to
red and measure the rising concentration of red fluid in the exit stream i.e. F-curve
■ At any time t > 0 red fluid and only red fluid in the exit stream is younger than age t

Using either step or pulse inputs can

conveniently give the RTD and mean flow rate
of fluid in the vessel but these relationships only
hold for closed vessels

43
44
Comparison Pulse and Step Input
■ The step is usually easier to carry out experimentally than the pulse test
■ Advantage of step is that the total amount of tracer in the feed over the period of the
test does not have to be known
■ Sometimes difficult to maintain a constant tracer concentration in the feed.
■ Key is the differentiation of data in the step change when the change at outlet is
observed otherwise it may lead to large error
■ Large amount of tracer required for step test, so if the tracer is very expensive, a
pulse test is almost always used to minimize the cost
Pulse Injection Example -1
■ A sample of the tracer A at 320 K was injected as a pulse to a reactor, and the effluent
concentration was measured as a function of time.

t (min) 0 1 2 3 4 5 6 7 8 9 10 12 14
C (g/m3) 0 1 5 8 10 8 6 4 3 2.2 1.5 0.6 0
12

10 ■ Determine the fraction of material that has spent

8 – 3 to 6 min
– 7.75 to 8.25 min
C, G/M3

2
■ Fraction of material that has spent less than 3 min
in the reactor
0
0 2 4 6 8 10 12 14 16
T, MIN

Fogler Example 13.1 46

■ Since the time steps are different the range is divided into two parts from
– 0 to 10
– 10 to 14
■ Apply any formula to integrate….use quadrature formula

47
Pulse Injection Example -3
■ A large tank (860 liters) is used as a gas-liquid contactor. Gas bubbles up through the vessel
and out the top, liquid flows in at one part and out the other at 5 liters. To get an idea of the
flow pattern of liquid in this tank a pulse of tracer (150 gm) is injected at the liquid inlet
and measured at the outlet, as shown in Fig.
A. Is this a properly done experiment?
B. If so, find the liquid fraction in the vessel.
C. Determine the E curve for the liquid.
D. Qualitatively what do you think is happening in the vessel?

Levenspiel Example 11.2 48

Pulse Injection Example -2
■ A pipeline (10 cm I.D., 19.1 m long)
simultaneously transports gas and liquid
from here to there. The volumetric flow rate
of gas and liquid are 60,000 cm3/s and 300
cm3/s, respectively. Pulse tracer tests on the
fluids flowing through the pipe give results
as shown in Fig. What fraction of the pipe is
occupied by gas and what fraction by
liquid? Levenspiel Exercise 11.6

49
Conversion in Non-ideal reactors
■ To Predict conversion in real reactor following models were proposed
■ Zero adjustable Parameter Model
– Segregation Model
– Maximum Mixedness Model
■ One Adjustable Parameter Model
– Tanks in series Model
– Dispersion Model
■ Two adjustable Parameter Model
– Real Reactors are modeled as combination of ideal reactors
■ For the zero adjustable parameter models, we do not need to make any intermediate
calculations; we use the E- and F-Curves directly to predict the conversion given the kinetic
parameters.
■ For the one-parameter models, we use the RTD to calculate mean residence time, tm, and
variance, 2, which we can then use
– to find the number of tanks in series necessary to accurately model a nonideal CSTR
– calculate the Peclet number, Pe, to find the conversion in a tubular flow reactor using the
dispersion model