Distributions of

Residence Times for

Chemical Reactors

Chapter 4
BKF3472
Objectives
 Define a residence time distribution RTD [E(t), F(t)] and the
mean residence time.

 Determine E(t) form tracer data.

 Write the RTD functions (E(t), F(t), I(t)) for ideal CSTRs, PFRs,
and laminar flow reactors.

 Predict conversions from RTD data using the segregation and

maximum mixedness models.

 Predict effluent concentrations for multiple reactions using the

segregation and maximum mixedness models.
Possible Deviations from ideality:

Short Circuiting or By-Pass – Reactant flows into the tank through the
inlet and then directly goes out through the outlet without reacting if the
inlet and outlet are close by or if there exists an easy route between the
two.

3
1. Dead Zone 2. Short Circuiting

4
 Topics to cover
 General characteristics

 Measurement of the RTD

 Characteristics of the RTD

 RTD in Ideal Reactor

 Reactor modeling with the RTD

 Zero parameter models

 RTD and multiple reactions

 General Characteristics

 Analyze non ideal reactor system

 3 concepts to describe non-ideal reactor

 Residence Time Distribution
 Mixing quality
 Model used to describe the system
 Residence Time Distribution
(RTD)
 Residence time
 time of atoms have spent in the reactor

 Residence Time Distribution (RTD)

 Characteristic of the mixing that occurs in the chemical
reactor. i.e. plug flow and thoroughly-mixed flow show
different RTD

 RTD purposes
 To diagnose problems of reactors in operation
 To predict conversion or effluent concentrations in
existing/available reactors when a new reaction is used
 RTD determination via
 Inject tracer, colored or radioactive material,
 Time tracer have spent I the reactor is recorded

 2 method of injection
 Pulse input
 Step input
Tracer:
•non-reactive species
•easily detectable
•similar physical properties to those of the reacting mixture
•completely soluble in the mixture
•not adsorb on the walls/other surfaces in the reactor
•colored and radioactive materials are most common types of
8

tracers
Measurements of the
RTD
 Pulse Input

N  C (t )vt
Integrating of dN  vC (t )dt

resulting to N o   vC (t )dt
0
 
resulting to N o   vC (t )dt
0

C (t )
Residence-time distribution function, E (t )  

 C (t )dt
0

 Fraction of matrial leaving 

the reactor that has resided  t 2
   E (t )dt
in the reactor for times  t1
 
between t1 and t 2 
See example 13-1!
 Step Input

 Drawbacks
 Difficult to maintain a constant tracer
concentration in the feed
 Large error

 expensive
Characteristics of the
RTD

E(t) - exit age distribution function

 Characteristics
 Integral relationship
 Mean Residence Time
 Variance
 Skewness
 Normalized RTD function
 Internal Age Distribution
Integral Relationship
 fraction of effluent 
 
t  spending in reactor for 
0 E (t )dt  less than time t @ from   F (t )
 
t  0 to t  t  Cumulative
distribution
Graph plotting function

F (t)

t (min)
Mean Residence Time tm
 when v=v0 and in the absence of dispersion effect, =tm
(proof is at pg 880-881 4th edition and only for closed
system).  = space time
 Give the average time the effluent molecules spent in the
reactor
 Commonly used in analysis. Identified also as “first moment”

t m   tE (t )dt
0
Variance
 Also known as square of standard deviation
 To indicate the ‘spread’ of the distribution. The
magnitude measures how spread is the data.
 Commonly use in analysis
 Also identified as the 2nd moment


 2
  (t  tm) 2 E (t )dt
0
Skewness
 To measure the extent that a distribution is
skewed to one direction or another in reference to
the mean.
 Also identified as Third-moment

1 
s   (t  t m ) 3 E (t )dt
3
3 0
 2
Normalized RTD Function
 To compare directly the flow performance inside reactors
of different sizes.
 Example, CSTR:

1
E (t )  e t /

t
E ()   E (t )  e 
where  

Internal Age Distribution
 To identify the reaction mixture condition in the reactor
 Internal Age Distribution,           
 I(α )d α = Fraction of molecules inside the reactor that
have been inside the reactor between a time (α) and
(α+d α)
 Example
From our experimental data of the exit tracer concentration from pulse trace test

t(min) : 0 1 2 3 4 5 6
C(mg/m3) : 0 0 0.1 0.2 0.3 0.1 0

→ → → →
RTD in Ideal Reactor (Pulse Inj.)
RTD for Ideal PFR
PFR- Inject a pulse at t=0

Dirac Delta Function                                    

F(t)
RTD in Ideal Reactor (Pulse Inj.)
Ideal CSTR Co
t
C (t )  Co e 
C(t)

e t 
E (t ) 


t
F (t )  1  e 
RTD for Diagnose
Faulty Operation
 
 (Ideal CSTR)  

Model as PFR with recycle and some dispersion

Modelling Using
Ideal RTD
Perfect Operation

Passing (BP)

Dead Volume
 Tubular Reactor

Perfect Operation
of PFR (P)

PFR with Channeling

(Bypassing, BP)

PFR with Dead

Volume (DV)
Diagnose the problem
occurred the plug flow reactor.
A liquid fluid reacts according to
A→R

as the reactant A flows through a 60m3 plug flow reactor with the flow rate of
1m3 s-1 at the initial concentration of 0.05 mol L-1. The reaction is zero order with
the rate constant of 0.045 mol L-1 min-1. The tracer test was performed before
the plug flow reactor was first used and after it was used for 2 years. The
results are shown in Figure 1(a) and Figure 1(b).
30 30 Figure 1. (a) Tracer data
C (x10-6molcm-3)

C (x10-6molcm-3)

20 20 of the reactor before it

10 10 was first used. (b) Tracer
data of the reactor after it
0 0
was used for 2 years.
13 14 15 16 17 9 10 11 12 13
t (s) t (s)

(a) (b)
 Model to Calculate the
Exit Concentrations and
Conversions
Combination of Ideal and Real Reactor involves:

 Zero Parameter Model

 Segregation model
 Maximum Mixedness Model

 One Parameter Models

 Tanks-in series model
 Dispersion model

 Two Parameter Models

 Bypassing
 Dead Space
 Recycling
 Segregation Model

 Mixing of the globules of different ages occurs here

Little batch reactors (globules) inside a CSTR.

X3>X2>X1

Mixing occurs at the latest possible moment. Each little batch reactor (globule) exiting
the mean conversion for the segregation model is
 PFR
 Detail description of segregation

A → B
Apply the segregation model to an ideal PFR
Solve for X(t) for a first order reaction in a batch reactor. The
conversion-time relationship is

                                           
Calculate the mean conversion

                                      
 LFR
For a Laminar flow reactor the RTD function is
                            
           

The mean conversion is

                                                      
  
The last integral is the exponential integral and can be
evaluated from Tabulated values         
Fortunately, Hilder developed an approximate formula (   =Da).
 CSTR

Mean Conversion of CSTR

Mean Conversion of Multiple
Reactions
For multiple reactions use an ODE solver to couple the mole
balance equations, dCi/dt=ri, with the segregation model
equations: d    /dt=Ci(t)*E(t), where C i is the concentration of
i in the batch reactor at time t and    is the concentration of i
after mixing the batch reactors at the exit.

                                                              
 Example:
The second order reaction
            
occurs in the liquid phase. The RTD function for the reactor in which it is to be carried
out is given by
E(t)=0 for 0 < t < 10
E(t) = 0.01 (t–10) for 10 < t < 20
E(t) = 0.01 (30–t) for 20 < t < 30
E(t) = 0 for t > 30
                                                         
This RTD function is the same one we previously studied in these lectures. The
entering concentration is 2 molar and the specific reaction rate is 0.06 dm 3/mol•s.
(a) What is the conversion after 30 seconds in a batch reactor?
Solution
(b) What conversion would be achieved in a PFR with the same mean residence time?
Solution
(c) What conversion would be achieved in a CSTR with the same mean residence
time?
Solution
(d) What is the conversion predicted by the segregation model?
Solution
Continue Lec
 Maximum Mixedness Model
Mixing occurs at the earliest possible moment.

                                                                                                          

 
Note E()=E(t)   
              
E()d  =Fraction of molecules that have a life expectancy between
 +d  and  .
  

                                                                                                          
                              
Modeling maximum mixedness as a plug flow reactor with side
entrances
Dividing by  and taking the limit as  goes to zero

Substitute

Differentiating the first term and recalling

We obtain
We need to integrate backwards from = (the entrance) to  = 0
(the exit). In real systems we have some maximum value of  (say 
 = 200 minutes) rather than = minutes.
Consequently we integrate backward from  = 200. However,
because most ODE packages will not integrate backwards, we have
to use the transfer
z = T -  to integrate forward

Thus,

In terms of conversion
The End