PDMC Coursework Sem1 2023 24 1
PDMC Coursework Sem1 2023 24 1
PDMC Coursework Sem1 2023 24 1
Use of generative AI: This is a Type B coursework meaning that the use of generative AI tools is
permitted as an assistive tool for e.g. proof reading and generating code snippets. Remember that as
described in the University of Bath’ integrity statement, that the final output must remain your own
work. You are required to add a text section at the end of your LiveScript in which you acknowledge the
use of GenAI and provide a brief description of the context in which the tool was used. We require you
to include the following:
• Name and version of the generative AI system used. For example: ChatGPT-3.5
• Publisher (= the company that made the AI system). For example OpenAI
• URL of the AI system
• For each instance in which you used the tool, a brief description of the context in which the tool
was used. For example: to proofread my final draft only
• For each instance in which you used the tool, the URL of the shared link to the conversation with
ChatGPT or equivalent
• Confirmation that the work is your own
If you did not use GenAI tools for the coursework please say so explicitly in this textbox.
Question 1 20 %
Question 2 60 %
• Results a – c 30 %
• Discussion a – c 15 %
• Safety suggestion d 15 %
1
CE20236 Process Dynamics, Modelling & Control Coursework Semester 1 2023/24
Submitting your coursework
Please note: The PDMC coursework is part of a University trial to use Crowdmark to make it easier to
provide personalised and more detailed feedback. The link to Crowdmark is available on Moodle where
you can also find instructions how to upload your solution.
2
CE20236 Process Dynamics, Modelling & Control Coursework Semester 1 2023/24
Question 1
1,3-propanediol (PDO) is used in a wide variety of products including personal care products, food and
flavours, pharmaceuticals and dietary supplements, laundry and cleaning products, and also serves as
a heat transfer fluid and as a building block for polyurethane and polyester resins. Traditionally it is
synthesised from fossil fuels e.g. in the Shell process. In 2007, Dupont and Tate & Lyle opened a
100 million lb/ yr plant to produce PDO from plant derived glucose. The fermentation process uses an
engineered strain of E. coli which was modified with genes from baker's yeast and Klebsiella
pneumoniae. (Tullo, 2007). The downstream processing of the process includes micro- and
ultrafiltration to remove cells and proteins from the fermentation broth, ion exchange to remove
impurities, flash evaporation to remove excess water and finally distillation.
For this coursework we will have a closer look at the flash evaporation step. The feed to the flash unit
consists of 2 mol% glycerol, 3 mol% PDO and you can assume that the rest is water. The flash
evaporator is operated at 2. 5 bar and 165 °C. The vapour pressure for all components can be
described by the Antoine equation given in the following form
𝐵 (1)
𝑙𝑜𝑔10 (𝑃∗ ) = 𝐴 −
𝑇+𝐶
Where P* is the vapour pressure in bar and T is the temperature in K. For PDO and water you can use
the coefficients provided in Table 1.
Table 1 Parameters for the Antoine equation (NIST Chemistry Webbook, 2020)
A B C
1,3 propanediol 6.29523 3105.018 6.101
Water 3.55959 643.748 -198.043
b) Calculate the percentage of the total feed that is evaporated and the corresponding mole
fractions of the vapour and the liquid phase. (The derivation of the equations necessary to solve
this problem is available on the pre-recorded material on Moodle).
3
CE20236 Process Dynamics, Modelling & Control Coursework Semester 1 2023/24
Question 2 Explosion at T2 Laboratories Explosion
(adapted from Fogler, 2011)
On the 19th of December 2007, a powerful explosion shook Jacksonville, Florida, when a reactor
ruptured and caught fire at T2 Laboratories. Four people lost their lives. Thirty-two people within the
vicinity suffered injuries. The power of the shock wave was felt 15 miles away.
What happened
On the fatal day, a reactor operator contacted one of the plant owners informing him that there was
no cooling water flow to the reactor. The temperature in the reactor continued to rise. The pressure
also increased as hydrogen was produced at an increased rate to the point where the pressure control
valve system on the 1-inch diameter hydrogen venting stream could no longer maintain the operating
pressure at 4.4 atm absolute. Unknown to the plant employees, a second, stronger, exothermic
reaction kicked in, the decomposition reaction of diglyme either catalysed or enhanced by the
presence of sodium:
4
CE20236 Process Dynamics, Modelling & Control Coursework Semester 1 2023/24
4-inch vent pipe
and rupture disk
Solid sodium Pressure control valve
metal loaded
to reactor
Hot oil in
1-inch hydrogen
vent pipe
Hot oil out
Cooling jacket vent
to atmosphere
This rapid decomposition led to the creation of even more hydrogen causing the pressure to rise even
faster eventually causing the 4-inch rupture disk to break at 28.4 atm absolute. Even with the relief
line open, the rate of production of hydrogen was now much greater than the rate of venting
eventually causing the reactor to rupture in a horrific explosion.
Simplified model
The runaway reaction can be approximately modelled by two reactions:
A + B → C + ½ D (gas) (2)
and
(5)
𝑟2𝑆 = −𝑘𝑆 𝐶𝑆
where Ci is the concentration of component i and ki is the rate constant. The temperature
dependence of the rate constants is given by Arrhenius law:
𝐸𝑎 (6)
𝑘(𝑇) = 𝑘0 𝑒𝑥𝑝 (− )
𝑅𝑇
where k0 is the pre-exponential factor, Ea the activation energy and R the ideal gas constant. The
reaction characteristics for the two exothermic reactions are summarised in Table 2. The heat of
reactions can be assumed to be constant.
5
CE20236 Process Dynamics, Modelling & Control Coursework Semester 1 2023/24
Table 2 Reaction characteristics for reactions 1 and 2.
Reaction 1 Reaction 2
k0 4 1014 L mol-1 h-1 1 1084 h-1
Ea 128,000 J mol-1 800,000 J mol-1
ΔHrxn -45,400 J mol-1 -3.2 105 J mol-1
Assuming that the volume of the batch does not change, the mole balances for the reaction liquid
yield:
𝑑𝐶𝐴 (7)
= 𝑟1𝐴 = −𝑘𝐴 𝐶𝐴 𝐶𝐵
𝑑𝑡
𝑑𝐶𝐵
= 𝑟1𝐵 = 𝑟1𝐴 = −𝑘𝐴 𝐶𝐴 𝐶𝐵 (8)
𝑑𝑡
𝑑𝐶𝑆
= 𝑟2𝑆 = −𝑘𝑆 𝐶𝑆 (9)
𝑑𝑡
If the gas in the head space behaves like an ideal gas, the mole balance for the head space yields:
𝑑𝑃 𝑅𝑇 ( 10 )
= (𝐹𝐷 − 𝐹𝑣𝑒𝑛𝑡 )
𝑑𝑡 𝑉𝐻
where VH is the volume of the head space in the reactor, Fvent is the molar flow rate out of the reactor
through one or both outlet lines, and FD is the molar flow rate of gas leaving the liquid of volume V0
and entering the head space:
Gas (hydrogen) exits the reactor through both the pressure control valve line and the rupture disk
line. At low gas production, the pressure control valve maintains set point pressure at the initial
pressure by venting all produced gas until the gas production reaches 11,400 mol/h:
When the pressure increases but is still below the rupture disk setting, the pressure control line vents
to the atmosphere (1 atm) according to
where P is the absolute pressure in the reactor (atm). The downstream pressure is 1 atm and the vent
rate, CV1, is 3,360 mol h-1 atm-1. If the pressure P in the reactor exceeds 28.2 atm, the relief line
activated by the rupture disk breaks and vents gas at a rate of CV2 = 53,600 mol h-1 atm-1 and the total
vent rate is given by
6
CE20236 Process Dynamics, Modelling & Control Coursework Semester 1 2023/24
The sum of the product of the number of moles of each species, Ni, and their heat capacities, CP,i, is
essentially constant at NiCPi = 1.26 107 J K-1.
Assumptions
You can assume that
• The liquid volume, V0, in the reactor remains constant at 4000 L.
• The vapour space in the head above the liquid, VH, occupies 5000 L.
• Any hydrogen produced (D) immediately appears as an input stream FD to the head space of the
reactor.
• Dissolved H2 and the vapour pressures for the liquid components in the reactor can be neglected.
• Any hydrogen generated obeys the ideal gas law.
• The temperature dependence of the heat of reaction can be neglected.
• The sum of the product of the number of moles of each species and their heat capacities is
constant.
• The reactor vessel will fail (explode) when the pressure exceeds 45 atm or 600 K.
Initial conditions
We start the modelling at the end of the heating period. Here, the initial temperature and initial
absolute pressure in the system are 422 K and 4.4 atm. The concentrations in the reactor at the end
of the heating period are CA0 = 4.3 mol / L, CB0 = 5.1 mol / L and CS0 = 3 mol / L.
Tasks:
a) Plot the reactor temperature, the head space pressure, and the concentrations of A, B, S, D as
function of time if the cooling system fails (UA = 0). Determine the time when the reactor
explodes. Discuss briefly what you observe for the diglyme concentration.
b) Model the reactor under normal operating conditions where the cooling water is turned on
(UA = 2.77 106 J/(h K)) once the temperature in the reactor reaches 455 K. Again, plot the reactor
temperature, the head space pressure, and the concentrations of A, B, S, D as function of time and
discuss your observations briefly.
c) What is the maximum time in minutes that the cooling can be lost (UA = 0) so that the reactor will
not reach the point of explosion? (In other words, how long did the operators have to reinstate
7
CE20236 Process Dynamics, Modelling & Control Coursework Semester 1 2023/24
the cooling before the reaction reached the point of no return and the reactor would have
exploded anyway?)
d) Briefly discuss three suggestions about how to change the process and/or plant to avoid this
accident taking into account principles of inherent safer design.
References
Fogler, H. S. “Elements of Chemical Reaction Engineering”, Prentice Hall, 2011.
NIST Chemistry WebBook, NIST Standard Reference Database Number 69, Eds. P.J. Linstrom and W.G.
Mallard, National Institute of Standards and Technology, Gaithersburg MD,
20899, https://doi.org/10.18434/T4D303, (retrieved November 27, 2021).
Richardson, A., LXXIII. Determinations of Vapour-Pressures of Alcohols and Organic Acids, and the
Relations Existing Between the Vapour-Pressures of the Alcohols and Organic Acids, J. Chem. Soc., 1886,
49, 761-776, https://doi.org/10.1039/ct8864900761