VIETNAM NATIONAL UNIVERSITY

HO CHI MINH UNIVERSITY OF TECHNOLOGY


Faculty of Chemical Engineering

REPORT: LABORATORY OF UNIT OPERATIONS

RETENTION TIME
Course ID: CH3015 – Class: CC01 – Group: 1
Lecturer: Dr. Phạm Hoàng Huy Phước Lợi

Members ID

Ho Chi Minh City, 2025

 OUTLINE

1. ABSTRACT: ...................................................................................................................... 3
1.1. Experimental purpose: ................................................................................................. 3
1.2. Experimental results: ................................................................................................... 3
2. EXPERIMENTAL THEORIES : .................................................................................... 4
2.1. Basic concept: .............................................................................................................. 4
2.2. Methods of studying retention time: ............................................................................ 4
2.3. Retention time distribution function of CSTRs’s model:.............................................. 6
3. EXPERIMENTAL EQUIPMENT: ................................................................................. 7
4. EXPERIMENTAL METHODS: ..................................................................................... 7
5. EXPERIMENT RESULTS:............................................................................................. 8
6. DISCUSSION: ............................................................... Error! Bookmark not defined.
7. CALCULATION:............................................................................................................ 12
8. REFERRENCES: ........................................................................................................... 16

2
1. ABSTRACT:
1.1. Experimental purpose:
- Survey retention time of a system of continuous stirred-tank reactors (CSTRs).
- Determine real retention time distribution function of and compare with theoretical
retention time distribution function.
1.2. Experimental results:
Absorbance for 1 reactor are counted: D = 0.040; 0.038; 0.030
Maximum absorbance (for 1 reactor): Do = 0,040
Reactor diameter: d = 14 (cm)
Water level in the CSTR: H = 12 (cm)
Flowrate of water: v = 0,4 (litter/min)
Table 1: Raw results
t D
(min) 1 tank system 2 tank system 3 tank system
0 0 0 0
1 0.025 0.002 0.000
2 0.022 0.002 0.001
3 0.023 0.006 0.001
4 0.039 0.010 0.000
5 0.009 0.010 0.002
6 0.008 0.008 0.006
7 0.010 0.008 0.002
8 0.006 0.011 0.011
9 0.003 0.009 0.004
10 0.005 0.009 0.000
11 0.004 0.005 0.005
12 0.004 0.009 0.004
13 0.004 0.008 0.002
14 0.002 0.005 0.006
15 0.002 0.001 0.008
16 0.004 0.004
17 0.008 0.005
18 0.003 0.002
19 0.008 0.003
20 0.002 0.002
21 0.005 0.000
22 0.003 0.004
23 0.000 0.000
24 0.000 0.000
25 0.000 0.004
26 0.007
27 0.002
28 0.003
29 0.002
30 0.001

3
 31 0.006
32 0.003
33 0.000
34 0.000
35 0.000

2. EXPERIMENTAL THEORIES:
2.1. Basic concept:
In the equipment system, different fluid elements will follow different paths. Based on
defined retention time distribution function, we can evaluate correlatively about the flows in
the equipments, drawbacks when designing such as some areas where propeller cannot reach,...
and find ways to overcome thanks to this evaluation.
Study of retention time is a necessary method for comparing equipment based on material
flow, from which we can improve, set up the optimal model.
Also, based on retention time distribution function, we can operate optimally and thereby
set parameters, control methods as well as optimize the process in the equipment.
The retention time of an element in the system is the time that the element stays in the
reactor, or in any equipment to be surveyed. Retention time of equipment is a probability
quantity. So, all retention times are around the average retention time, therefore determining
the average retention time is particularly significant.
1 n
t V =  t Vi (1)
n i =1
Where: tVi is retention time of an element i.
With the definition of retention time distribution function F (tV) = E (tV):

t V =  f (t V )t V .dt V = E(t V ) (2)
0


c
A
I (t V )t V .dt V
Or: t V = 0

(3)
c
A
I (t V ).dt V
0

With point functions we have:

K

c t i i
t= i =1
K
(4)
c
i =1
i

K is equal split.
Average retention time according to volume:
V
= R =t (5)
VM
With VR: Volume of fluid in the reactor, litter.
VM: Flowrate of the input, litter/sec.
If the indicator does not reach the ideal correlation, the above equation is not satisfied (If
  t , the indicator may be absorbed into the wall or some additional details of reactor).
2.2. Methods of studying retention time:
The retention time study can be carried out according to the following methods:

4
 1) Determine the composition of components at time t (or ) of leaving the equipment,
determine function F(t) or F().
2) Determine the composition of components at time t (or ) of still remaining in the
equipment, function I(t) or I().
3) Determine the composition of components at time t (or ) of being out of the equipment,
function f(t) or f().
To survey the performance of a real reactor, we often use the stimulus – response method
(marking method). The types of input stimuli and output responses are shown in Figure 1.

Input signal Output signal

(Stimuli) Reactor (Response)

Any input signal

Concentration of indicator

Output

Time
Concentration of indicator

Circulated input signal

Output

Time
Concentration of indicator

Step signal

Output

Time
ncentration of indicator

Pulse signal

5
 Figure 1: Frequently-used stimulus – response signals

So, marked components must have characteristics that does not influence and distinct from
the components that make up the correlation in the system.
Indicators for liquid environment may be colored solution, radioactive substances, stable
radioactive isotopes, glowed solid particles... Indicators appropriate to the nature of the
components in the system must have appropriate density, viscosity, diffusion coefficient.
When there are appropriate indicators, we can put them into the system in two types of
signals: random signal (Stochas) and determined signal (Determinis). Determined signal can be
divided into two types: circulated signal and uncirculated signal.
To survey equipment, people use uncirculated determined signal. This type of signal can be
created thanks to:
1) Marked by collision.
2) Marked by continuous input of a specified amount.
3) Marked by occupying the whole place in the system.
Because it is convenient for the use and uniformity of the stimulus signal in the form of step
and pulse. In this experiment, we use marked by collision type. This type is the implementation
of the technical condition of Dirac’s function (Dirac’s momentum function), also known as
Delta’s function.
0, t  0
(t ) = 
, t = 0


 (t ).dt = 1
−
This marked type is suitable for colored solution.

2.3. Retention time distribution function of CSTRs’s model:

Most real equipment usually has the distribution function of the CSTRs’s model.
One ideal reactor with the type of collision mark must satisfy: Volume VR in the reactor is
constant over time, in the reactor there is a complete agitation of a component in the system
homogeneously at all positions in volume VR. So, in the reactor there is a surge of input flow.
Average retention time according to volume:
V
= R (6)
VM
C Ira  t
= 1 − exp − 
CIvaø o
 
When connecting ideal reactors to each other, we have CSTRs’s model:
The general form of distribution function:
n −1
 t  t
 −  . exp − 
  
f (t ) = 1 /  =  (7)
(n − 1)!
When n = 1, wa have an ideal CSTRs’s model. When n = , we have an ideal push model.
Assume that initially there was no indicator in the input flow, and then an impulse signal is
applied by putting a certain amount of indicator to the fluid flow for a short period of time.
The curve showing the concentration of indicator over time in the output flow
corresponding to the pulse stimulation signal at the input is called the curve C. Initial
concentration of indicator is Co. With the area below the curve is equal to 1, we have:

6
 t t t
Dimensionless compact time:  = = = (8)
t  V
With t is a time when any fluid component passes through the equipment.
t and  are determined according to (4) and (5).
   
C 1
0 C.d = 0 C 0 d = 1 or C 0 = 0 C.d = t 0 Cdt (9)

2,0
D Small dispersion
Tube =0
uL D
= 0,002
uL

1,5

Stirrer Medium dispersion

D D
= = 0,025
uL uL

1,0

Large dispersion
D
= 0,02
uL
0,5

0 0,5 1 1,5

Figure 2: The curve C showing the response at output flow for

the pulse signal at the input flow.

3. EXPERIMENTAL EQUIPMENT:
Test tube, stopwatch, pipet, colored subtances and light transmittance (absorbance) meter.

4. EXPERIMENTAL METHODS:
Proceed in the following order:
1) Open the valve for water to flow into the high-positioned tank until there is water in the
overflowing tube.
2) Open the valve for water to flow through flow meter into the CSTRs system and adjust
the flowrate.
3) System of one reactor: When the system is stable, put colored subtance into reactor 1.
Click the stopwatch (At the same time with the time when we put the colored subtance

7
 into the reactor), take root time. Use test tubes to take samples from time to time, then
compare the color.
4) System of two, three reactors: Do the same like one reactor, put the colored subtance
into reactor 1 and take samples at the last reactor (from the last interconnected pipe).

5. EXPERIMENT RESULTS:
Table 2: 1 tank system
t Experiment Theory
D
(min) tiDi θ Eθ θ Eθ
0 0 0.000 0.000 0.000 0.000 1.000
1 0.025 0.025 0.212 0.625 0.216 0.806
2 0.022 0.044 0.423 0.550 0.432 0.649
3 0.023 0.069 0.635 0.575 0.649 0.523
4 0.039 0.156 0.847 0.975 0.865 0.421
5 0.009 0.045 1.059 0.225 1.081 0.339
6 0.008 0.048 1.270 0.200 1.297 0.273
7 0.010 0.070 1.482 0.250 1.514 0.220
8 0.006 0.048 1.694 0.150 1.730 0.177
9 0.003 0.027 1.906 0.075 1.946 0.143
10 0.005 0.050 2.117 0.125 2.162 0.115
11 0.004 0.044 2.329 0.100 2.378 0.093
12 0.004 0.048 2.541 0.100 2.595 0.075
13 0.004 0.052 2.752 0.100 2.811 0.060
14 0.002 0.028 2.964 0.050 3.027 0.048
15 0.002 0.030 3.176 0.050 3.243 0.039
Total 0.166 0.784
τ (min) 4.625
t (avg) 4.723

The values are needed to calculated:

D0 = 0.040 (m)
tavg = 4.723 (min)
v = 0.0004 (m3/min)
V = 0.00185 (m3)
 =  (min)

8
 𝐷
Figure 3: The graph shows the relation 𝐷 = 𝑓(𝜃) for the theoretical and empirical case of a
0
1-tank system

Table 3: 2 tank system

t Experiment Theory
D
(min) tiDi θ Eθ θ Eθ
0 0 0 0.000 0.000 0.000 0.000
1 0.002 0.002 0.094 0.100 0.108 0.348
2 0.002 0.004 0.188 0.100 0.216 0.561
3 0.006 0.018 0.282 0.300 0.324 0.678
4 0.010 0.040 0.376 0.500 0.432 0.728
5 0.010 0.050 0.470 0.500 0.541 0.733
6 0.008 0.048 0.564 0.400 0.649 0.709
7 0.008 0.056 0.658 0.400 0.757 0.666
8 0.011 0.088 0.752 0.550 0.865 0.613
9 0.009 0.081 0.846 0.450 0.973 0.556
10 0.009 0.090 0.941 0.450 1.081 0.498
11 0.005 0.055 1.035 0.250 1.189 0.441
12 0.009 0.108 1.129 0.450 1.297 0.388
13 0.008 0.104 1.223 0.400 1.405 0.338
14 0.005 0.070 1.317 0.250 1.514 0.293
15 0.001 0.015 1.411 0.050 1.622 0.253
16 0.004 0.064 1.505 0.200 1.730 0.218
17 0.008 0.136 1.599 0.400 1.838 0.186
18 0.003 0.054 1.693 0.150 1.946 0.159
19 0.008 0.152 1.787 0.400 2.054 0.135
20 0.002 0.040 1.881 0.100 2.162 0.115
21 0.005 0.105 1.975 0.250 2.270 0.097
22 0.003 0.066 2.069 0.150 2.378 0.082
23 0.000 0.000 2.163 0.000 2.486 0.069
24 0.000 0.000 2.257 0.000 2.595 0.058

9
 25 0.000 0.000 2.351 0.000 2.703 0.049
Total 0.136 1.446
τ (min) 9.250
t (avg) 10.632

The values are needed to calculated:

D0 = 0.020 (m)
ttb = 10.632 (min)
v = 0.0004 (m3/min)
V = 0.0370 (m3)
 =  (min)

2-tank system
Comparision: Experimental and Theoretical
0.800

0.600

Eθ 0.400
0.200

0.000
0.000 0.500 1.000 1.500 2.000 2.500 3.000
θ

Experimental data Theoretical data

𝐷
Figure 4: The graph shows the relation 𝐷 = 𝑓(𝜃) for the theoretical and empirical case of a
0
2-tank system

Table 4: 3 tank system

t Experiment Theory
D
(min) tiDi θ Eθ θ Eθ
0 0 0.000 0.000 0.000 0.000 0.000
1 0.000 0.000 0.059 0.000 0.072 0.056
2 0.001 0.002 0.119 0.075 0.144 0.182
3 0.001 0.003 0.178 0.075 0.216 0.330
4 0.000 0.000 0.238 0.000 0.288 0.472
5 0.002 0.010 0.297 0.150 0.360 0.595
6 0.006 0.036 0.357 0.450 0.432 0.690
7 0.002 0.014 0.416 0.150 0.505 0.756
8 0.011 0.088 0.476 0.825 0.577 0.796
9 0.004 0.036 0.535 0.300 0.649 0.811
10 0.000 0.000 0.595 0.000 0.721 0.807
11 0.005 0.055 0.654 0.375 0.793 0.787

10
 12 0.004 0.048 0.713 0.300 0.865 0.754
13 0.002 0.026 0.773 0.150 0.937 0.713
14 0.006 0.084 0.832 0.450 1.009 0.666
15 0.008 0.120 0.892 0.600 1.081 0.616
16 0.004 0.064 0.951 0.300 1.153 0.565
17 0.005 0.085 1.011 0.375 1.225 0.513
18 0.002 0.036 1.070 0.150 1.297 0.464
19 0.003 0.057 1.130 0.225 1.369 0.416
20 0.002 0.040 1.189 0.150 1.441 0.371
21 0.000 0.000 1.249 0.000 1.514 0.330
22 0.004 0.088 1.308 0.300 1.586 0.292
23 0.000 0.000 1.367 0.000 1.658 0.257
24 0.000 0.000 1.427 0.000 1.730 0.225
25 0.004 0.100 1.486 0.300 1.802 0.197
26 0.007 0.182 1.546 0.525 1.874 0.172
27 0.002 0.054 1.605 0.150 1.946 0.149
28 0.003 0.084 1.665 0.225 2.018 0.129
29 0.002 0.058 1.724 0.150 2.090 0.112
30 0.001 0.030 1.784 0.075 2.162 0.096
31 0.006 0.186 1.843 0.450 2.234 0.083
32 0.003 0.096 1.902 0.225 2.306 0.071
33 0.000 0.000 1.962 0.000 2.378 0.061
34 0.000 0.000 2.021 0.000 2.450 0.052
35 0.000 0.000 2.081 0.000 2.523 0.044
Total 0.1 1.682
τ (min) 13.875
t (avg) 16.820

The values are needed to calculated:

D0 = 0.040/3 (m)
tavg = 16.820 (min)
v = 0.0004 (m3/min)
V = 0.00555 (m3)
 =  (min)

11
 𝐷
Figure 5: The graph shows the relation 𝐷 = 𝑓(𝜃) for the theoretical and empirical case of a
0
3-tank system

6. CALCULATION:
Calculated values for 1 tank system
6.1.1. Experiment:
Maximum absorption: Do = 0.04

Average retention time: t =

 Dit i
 Di
t
Compact time:  =
t
D
Measurement of retention time’s distribution: E =
Do
6.1.2. Theory:
Water flowrate input to the tank: v = 0.4 (l/min) = 0.0004 (m3/min)
Tank volume: V = (/4). 0,142. 0,12 = 0.00185 (m3)
V
Volume time:  = = 4.625
v
t
Compact time:  =

Measurement of retention time’s distribution: E = e −

Calculated values for 2 tank system

6.1.3. Experiment:
Maximum absorption: Do = 0.04/2 = 0.02

Average retention time: t =

 Dit i
 Di
t
Compact time:  =
t
D
Measurement of retention time’s distribution: E =
Do
12
 6.1.4. Theory:
Water flowrate input to the tank: v = 0.4 (l/min) = 0.0004 (m3/min)
Tank volume: V = 2. (/4). 0.142. 0.12 = 0.00370 (m3)
V
Volume time:  = = 9.250
v
t
Compact time:  =

N(N) N −1 − N
Measurement of retention time’s distribution: E = e = 2.(2)e −2.
(N − 1)!

Calculated values for 3 tank system

6.1.5. Experiment:
Maximum absorption: Do = 0.04/3 =0.013

Average retention time: t =

 Dit i
 Di
t
Compact time:  =
t
D
Measurement of retention time’s distribution: E =
Do
6.1.6. Theory:
Water flowrate input to the tank: v = 0.4 (l/min) = 0.0004 (m3/min)
Tank volume: V = 3. (/4). 0.142. 0.12 = 0.00555 (m3)
V
Volume time:  = = 13.875
v
t
Compact time:  =

N(N) N −1 −N 3(3) 2 −3
Measurement of retention time’s distribution: E = e = e
(N − 1)! 2
7. DISCUSSION:
Problem 1: Compare the retention time distribution function of theory and experiment.

 1 tank system: As shown in Figure 3, the retention time distribution (RTD) function of the
experimental data exhibits noticeable differences compared to the theoretical curve. In general,
the experimental Eθ values are higher than those predicted by theory at several points,
particularly in the early stage, while the theoretical curve shows a smoother and more gradual
decline. Additionally, the theoretical curve extends over larger dimensionless time values (θ),
indicating a longer predicted residence time compared to the experimental results.

These discrepancies suggest that the actual flow behavior in the tank deviates from the ideal
single-tank model. The theoretical model assumes perfect mixing and uniform flow patterns,
while in practice, slight irregularities such as dead zones or non-ideal circulation may occur.
Although the system does not contain multiple pipes that could cause flow bypassing, there
may still be obtuse or stagnant regions within the tank where the agitator’s effect is limited,
leading to deviations in the experimental RTD curve.

13
 2 tank system: As illustrated in Figure 4, the experimental retention time distribution (RTD)
function shows a general trend similar to the theoretical curve, where the Eθ values increase to
a maximum and then gradually decrease. This pattern indicates that the flow behavior of the 2-
tank system roughly follows the expected theoretical model. However, the experimental curve
exhibits significant fluctuations instead of the smooth rise and decline observed in the
theoretical data.

These irregular variations in the experimental results suggest non-ideal flow conditions,
possibly due to imperfect mixing or slight variations in fluid velocity between the two tanks.
Furthermore, the theoretical Eθ values are generally higher than the experimental ones,
indicating that the theoretical model assumes a more efficient mixing performance than what
actually occurs in practice. Overall, while the 2-tank system behaves more closely to the
theoretical model than the 1-tank system, minor deviations still arise due to experimental
limitations and non-ideal hydrodynamic effects.

 3 tank system: As shown in Figure 5, the experimental retention time distribution (RTD)
function differs significantly from the theoretical curve. The experimental Eθ values fluctuate
strongly throughout the entire range of θ, whereas the theoretical data display a smooth and
continuous profile that rises to a maximum and then gradually decreases.

This high degree of fluctuation in the experimental results indicates pronounced non-ideal flow
behavior, which may be caused by uneven mixing, variations in flow velocity between tanks,
or disturbances during measurement. In theory, the 3-tank system should provide a more stable
and uniform flow pattern than systems with fewer tanks; however, the experimental data
suggest that the mixing efficiency or interconnection between the tanks was not perfectly
achieved in practice. These discrepancies highlight the influence of real operational conditions,
such as imperfect agitation and flow instability, which lead to deviations from the ideal
theoretical model.

Problem 2: The phenomenon of the process and the equipment incures the unsteady

 During the experiment, the system exhibited several unsteady phenomena that affected the
accuracy and stability of the results. One major cause of unsteadiness in the system is the
formation of dead or obtuse zones inside the tank. This occurs when the agitator speed is too
low, resulting in areas that cannot be fully stirred. As a result, mixing becomes uneven and
certain parts of the fluid remain stagnant. Increasing the agitation speed can help eliminate these
zones and promote better mixing.

 Another issue is the circulation of fluid caused by non-uniform mixing. Uneven flow may
result from the properties of the liquid or insufficient agitation. This leads to irregular residence
times and unstable flow paths. To minimize this effect, the speed of the agitator should be
optimized to ensure a more homogeneous fluid movement throughout the tank.

 The shape of the equipment can also create unstable flow regions. Irregular surfaces of the
tank, agitator, or piping may cause swirl zones, bypass flow, or stagnant areas. These regions
can temporarily or persistently distort the flow pattern. Proper equipment design, including the
addition of baffles, can help reduce these unwanted areas.

 The rotational instability of the agitator is another factor contributing to unsteady flow.
Variations in motor speed or shaft friction can change the mixing intensity over time. Using a

14
stable power supply and monitoring the agitator speed with control equipment such as a VFD
can improve consistency.

 The adsorption of the indicator on the equipment surface can gradually decrease its
concentration, affecting measurement accuracy. Choosing a less adsorptive tracer or pre-
treating the tank surface can help minimize this effect.

 Lastly, temperature fluctuations during the experiment alter the fluid’s viscosity and density,
leading to inconsistent flow behavior. Maintaining constant environmental and operating
temperatures is essential to ensure steady mixing conditions.

 In conclusion, unsteady behavior mainly arises from insufficient agitation, poor flow design,
motor instability, and temperature variation. Improving mixing efficiency, stabilizing the
agitator, optimizing tank design, and controlling experimental conditions can significantly
enhance system stability and data reliability.

Problem 3: Evaluate the errors

a) Errors Due to the Equipment
 Photometer:
- The photometer is the major source of error in the experiment.
- Dirty or unclean cuvettes and samples cause light absorption by impurities, leading to
inaccurate readings.
- The instrument is highly sensitive; even small dust particles or fingerprints on the
cuvette can affect absorbance.
- Different cuvette orientations in the holder produce variations due to the uneven position
of dirt or scratches.
- The old condition of the photometer may reduce accuracy.
- Unstable electric supply to the instrument also contributes to inconsistent results.

 Agitators:
- The swirling motion is uneven because of unstable voltage or inconsistent motor speed.
- Low agitation speed leads to poor mixing, creating heterogeneous concentration
distribution in the tank.
- Unsteady stirring affects the homogeneity of the tracer, leading to fluctuation in
experimental data.

 Piping System:
- Variations in pipe and valve resistance make it difficult to maintain stable inlet and
outlet flow rates.
- This problem becomes more severe in multi-tank systems, where the flow balance
between tanks is hard to control.
- During experiments, manual valve adjustments are needed to maintain tank volume,
which disturbs the steady-state condition.
- As a result, fluid level and residence time distribution are not constant, causing
deviations from the theoretical model.

b) Operational Errors
- Errors may occur due to inaccurate timing when collecting samples.

15
 - Cuvettes may not be completely clean or dry before measurement, causing
contamination from the previous sample.
- Lack of drying equipment or spare cuvettes increases the risk of measurement errors.
- Unstable flow rate control during operation also leads to inconsistency between runs.
- These operational factors together reduce the precision and repeatability of the
experiment.

c) Errors Caused by the Initial Moment and Pulse Signal

- The theoretical Dirac delta function assumes an instantaneous injection of the tracer.
- In practice, using a pipette introduces a time delay of about five seconds, so the input
pulse is not truly instantaneous.
- This time difference changes the shape of the pulse signal, creating a gap between
theoretical and experimental RTD curves.
- Consequently, the experimental data deviate from the ideal model based on the Dirac
function.

8. REFERRENCES:
[1]. Vuõ Baù Minh, “Quaù trình vaø Thieát bò trong Coâng Ngheä Hoùa Hoïc & Thöïc phaåm – Taäp 4:
Kyõ thuaät phaûn öùng”, Nhaø xuaát baûn Ñaïi hoïc Quoác gia TpHCM, 2004, 380tr.
[2]. Levenspiel Octave, “Chemical reaction Engineering”, John Willey and Sons,
Bookcomp, New York, 1999

16
 Retention time

17