Simulation Tasks

3-weeks course 27032: Microbiology and Fermentation Technology

Simulation Exercise: General information, outline and task-list

Simulation Exercise: General information, outline and task-list....................................................... 1 General instructions .................................................................................................................... 3 Guidelines for report writing ....................................................................................................... 3 Task 1. Getting started................................................................................................................. 4 Task 2. Substrate inhibition kinetics. ........................................................................................... 5 Task 3. Antibiotic Production...................................................................................................... 6 Task 4. Designing a chemostat operation..................................................................................... 8 Task 5. Fed-batch process by using yeast. ................................................................................. 10 Task 5-A. Chemostat process for protein production ............................................................. 10 Task 5-B. Fed-batch process for baker’s yeast production ..................................................... 10 Task 6. Mystery of a "continuous stirred overflow tank reactor"................................................ 12

Availability of assistance Teaching assistance will be provided from Monday to Thursday from 9:00 to 12:00. You are welcome to contact Kiran, Louise or Ana Rita between 13:00-17:00, either by email or by an appointment (fix by phone/email). Kiran Raosaheb Patil Email: [email protected] Phone: 4525 2703/2677 (O), 26204296 (M) Office: Room 226/126, Building 223. Louise Mølgaard Email: [email protected] Phone: 45252699(O) Office: Room 218, Building 223. Ana Rita Brochado Email: [email protected] Phone: 45252997(O) Office: Room 213, Building 223.

General instructions 1. You can work either in groups or individually. However, the report must be handed in on individual basis. 2. Monday 16th June, before you start your experiments, will be the deadline for receiving reports. Hand them to your teacher in-charge of the experiments. 3. Note that you will use some of the results from the simulation exercise for you laboratory work with yeast fermentations.

Guidelines for report writing 1. Please do not write a very long report. An ideal report should not be more than 15 A4 pages. 2. Use tables, figures etc. instead of long descriptions to increase the clarity and to save space. 3. Whenever necessary, write down the appropriate equations used in the simulations. 4. Provide justifications for the choices made. For example, if you suggest that a continuous fermentation is better than batch fermentation, provide the reasons you used to arrive at this recommendation. 5. It is not necessary, and is discouraged, to submit the MATLAB programs (either provided to you or written/modified by you) with the report. 6. Make a separate sub-section in your report for each task. 7. Submit reports in print, please do not email them.

Task 1. Getting started. Briefly list different industrial/natural processes/products where design of bio-reactors and optimization of their operation is of relevance (Minimum 3 and maximum 5 processes). Take the following points into consideration: i) Is the product a high-value or low-value product? Depending on this, the design objective may be maximization of yield, conversion of substrate or volumetric productivity of the product. ii) In what mode (batch, fed-batch, continuous) is the process operated? Why is this mode preferred for this process?

Task 2. Substrate inhibition kinetics. Plot a curve showing relationship between substrate concentration ‘s’ and growth rate ‘µ’ for a) monod kinetics; and b) substrate inhibition kinetics; both described in the following equations. Discuss the difference between the natures of these two curves. What is the possible implication of this difference on the design of a bio-process?

µ max ⋅ S

µ=

Simple Monod growth kinetics

(S + Ks ) µ max ⋅ S

µ= 

 





 



S2 S + Ks + Ki

 



Monod kinetics with substrate inhibition













Task 3. Antibiotic Production Since Alexander Fleming discovered penicillin in 1928 this compound has revolutionized the treatment of infectious disease. A pharmaceutical company wants to produce penicillin from the fungus Penicillium chrysogenum. The market demand is a fermentor capable of producing 10.000.000 kg/year. In order to access the productivity of the strain prior to large scale production a small scale batch fermentation was run and the concentration of the product was measured by taking samples from the fermentation. These were the initial small scale results. time (h)

Product (g/L)

Biomass (g/L) Glucose (g/L)

0

0,0

0,1

50,0

5

0,0

0,2

49,9

10

0,0

0,4

49,8

15

0,0

0,8

49,6

20

0,0

1,6

49,2

24

0,0

2,9

48,6

28

0,0

5,0

47,5

35

0,0

13,4

43,3

40

0,0

27,0

36,5

50

0,1

27,1

36,2

60

0,2

27,2

34,7

80

0,5

27,5

25,7

100

0,8

27,6

16,7

120

1,1

27,8

7,7

130

1,2

27,9

3,2

A. What is the volumetric productivity (qp)? Calculate both the overall volumetric productivity and the volumetric productivity in the “production phase”. B. Discuss what you should take into consideration when up-scaling a fermentation process from lab-scale to production scale? Assume that you can run a batch fermentation in one week, where the bioreactor is running 6 days in total with 2 days of biomass formation, 4 days of maximum productivity and one day doing cleaning. Furthermore due to holidays the fermentation plant is shut down 3 weeks a year.

C. How big a fermentor would you need to meet market demands? (Assume that the usual size of a “large” fermentor is 500 m3) The cost of our substrate is 100 DKK/kg and Ysp=0,03 g penicillin / g substrate. The running cost of running the fermentation is 10 DKK/L/h. D. Calculate what price penicillin (per gram) should be sold at to break even with the running costs? E. If the pharmaceutical company wants to have a profit of at least 86% what would the price of the product be? F. Discuss different methods that could be used to increase profit?

Task 4. Designing a chemostat operation. Brian is a biological engineer and he works in a company for biomass production. A certain day, his boss comes to talk to him and gives him a new task: to produce a valuable microbial biomass (used as a starter culture for dairy products) using a chemostat mode of operation. To start, Brian’s boss supplies him some information about the fermentation conditions and what fermentor to use. It is a 10 L fermentor, the substrate concentration in the feed is 10 g/l and the yield of biomass growing in the given substrate is 0.5 g DW/ g subst. The mixing conditions and the growing temperature of the microorganism are already known, so Brian is ready to start. However, when he is in the plant to get everything ready and start the fermentation, he realizes that his boss totally forgot to give him one parameter… what is the flow rate he should use? Then, Brian decides to go and ask his boss this small number and he will know it by heart, as everything else! However, the boss says: “That is YOUR task!” a) After thinking for a while, he concludes that the best way to find the flow rate to use is to go to the plant and try different flow rates. At the same time, he collects some data. S is the substrate concentration inside the fermentor, F is the feed rate and X is the biomass concentration inside the fermentor. F (l/h)

S (g/l)

X (g DW/l)

0.05

0.01

4.99

0.5

0.13

4.94

2

0.83

4.59

2.5

1.30

4.35

3

2.13

3.94

3.5

3.87

3.07

4.2

10

0

His first thought is that a very low flow rate should be used. However, after some reflection he decides to check how much biomass in g DW he has in one hour for each flow rate he tested. What do you think he will conclude? b) Brian decides to study this situation more in detail. So, he guesses that this situation might be related with growth kinetics. Two different microbial growth models that may be valid in this situation are Monod and Contois. Determine the kinetic parameters, Ks and µ max, and wash-out

dilution rate for each model. What is the yield YXS? Does any of the two models seems to be more appropriate? Start by writing the mass balances.

µ = µ max

Cs Cs + K s

µ = µ max

Cs Cs + K s ⋅ X

c) Comment on differences between the kinetic models Contois and Monod, as well as their application.

Task 5. Fed-batch process by using yeast. The yeast Saccharomyces cerevisiae is widely used in different industrial applications ranging from baking, brewing to therapeutic protein production. A fed-batch production process is indispensable in applications where either high yield/productivity of biomass/proteins is desirable. One of the major reasons for this choice is so called crab-tree effect or overflow metabolism. In a simplified model the overflow metabolism can be seen as diversion of excess glucose-uptake to fermentative pathway (ethanol production) when yeast is exposed to high glucose levels. Thus, biomass yield drops dramatically at high glucose uptake rates due to ethanol production. This is of course undesirable when the primary objective is to produce biomass. A solution to this problem is to use fed-batch or chemostat mode of operation. Task 5-A. Chemostat process for protein production

The article by Carlsen et al. (Biotechnology & Bioengineering, 1996) presents a study on protein production by using chemostat operation. It uses a simple model for simulating over-flow metabolism. 1. Briefly describe the kinetic model used, especially in relation to overflow metabolism. What is Dcrit in this model, i.e. dilution rate at which ethanol formation is onset in an aerobic chemostat? What is Dmax? Verify your results by using figure 3. 2. Discuss why the optimal dilution rate for protein production equals the critical dilution rate (figure 6). What dilution rate you would recommend to use in a production process?

Task 5-B. Fed-batch process for baker’s yeast production

It is desired to produce baker’s yeast using a fed-batch. The kinetics and mass balances for batch & fed-batch process are provided to you as MATLAB files. Use “Batch_saccharomyces.m” and “Fedbatch_saccharomyces.m” for simulating Batch and Fed-batch respectively. You can change certain conditions and parameters in the file. Please ask teaching assistants for help using these files for simulations. Refer to the article by Villadsen & Patil (Biotechnology & Bioengineering, 2007) for fed-batch equations. Note that you will run batch and fed-batch fermentations with yeast in this course. You should use these simulations to decide what should be the feeding profiles to obtain maximum biomass productivity (gDW/hr/L) in a fed-batch process.

1. First simulate a batch process with 10 g/L of initial glucose and 0.01 gDW/L of inoculum (initial biomass). Calculate yield of biomass and ethanol on glucose. How does the biomass yield compare with that in a chemostat process run below critical dilution rate? Comment on the difference. 2. Use the glucose, ethanol & biomass concentration at the end of batch phase to simulate fed-batch. Devise a feeding strategy (such as constant feed, linear feed, exponential feed, or combination of these) so as to obtain maximum biomass productivity. The feed glucose concentration is 100 g/L, while the initial volume (batch phase) is 3.5 L and the final volume will be 4.5 L. Consider two scenarios. A)

Assume that O2 transfer is not a limitation. Calculate the maximum value of Kla required in the process that you suggest.

B)

Now, assume that the maximum Kla value that can be achieved for the given reactor is 600 hr-1; and that the dissolved oxygen concentration needs to be at least 60% of the saturation value.

Plot glucose concentration, volume, biomass, ethanol, Kla and growth rate as function of time in all simulations that you try (report results for at least 5 different feeding schemes). Justify your choice of best feeding strategy in both cases. 3. Which situation (A or B) is industrially more relevant? Why? 4. Is fed-batch the best option in this case? Justify your answer.

Task 6. Mystery of a "continuous stirred overflow tank reactor". (Borrowed from “Chemical Reaction Engineering” by Octave Levenspiel)

Holmes: You say he was last seen tending1 this vat2 .... Sir Boss: You mean "continuous stirred overflow tank reactor" Mr. Holmes. Holmes: You must excuse my ignorance of your particular technical jargon, Sir Boss. Sir Boss: That's all right. However, you must find him, Mr. Holmes. Imhibit was a queer3 chap4; always staring into the reactor, taking deep breaths and licking his lips, but he was our very best operator. Why, since he left, our conversion of googliox has dropped from 80% to 75%. Holmes (tapping the side of the vat idly): By the way, what goes into the vat? Sir Boss: Just an elementary reaction between ethanol and googliox, if you know what I mean. Of course, we maintain a large excess of alcohol, about 100 to 1 and .... Holmes (interrupting): Intriguing, we checked every possible lead in town and found not a single clue. Sir Boss: We'll give the old chap a raise - about twopence per week - if only he'll come back. Watson: Pardon me, but may I ask a question? Holmes: Why certainly, Watson. Watson: What is the capacity of this vat, Sir Boss? Sir Boss: A hundred Imperial gallons5, and we always keep it filled to the brim. That is why we call it an overflow reactor. You see we are running at full capacity - profitable operation you know. Holmes: Well, my dear Watson, we must admit that we're stumped6, for without clues deductive powers are of no avail. Watson: Ahh, but that is where you are wrong, Holmes. (Turning to the manager) Imhibit was a largish fellow - say about 18 stone7- was he not? Sir Boss: Why yes, how did you know? Holmes (with awe): Amazing, my dear Watson! Watson (modestly): Why it's quite elementary, Holmes. We have all the clues necessary to deduce what happened to the happy fellow. With Sherlock Holmes and Sir Boss impatiently waiting, Dr. Watson casually leaned against the vat, slowly and carefully filled his pipe and - with the keen sense of the dramatic - lit it. There our story ends. (a) What happened to Imhibit? (b) How did Watson estimate his weight? What assumption did he make to arrive at that value?

1

Operating/Taking care of Tank 3 Strange 4 Man 5 1 Imperial gallon = 4.546 lit 6 Failed/Clueless 2

7

1 Stone = 6.4 Kg