Online-analytical characterization of 2,3-butanediol … · Os resultados em RAMOS foram suficientemente satisfatórios para realizar fermentações num biorreactor de 3L. As análises

Online-analytical characterization of 2,3-butanediol production by Bacillus licheniformis DSM 8785

Elsa Catarina Policarpo Requeixa

Thesis to obtain the Master of Science Degree in

Biological Engineering

Supervisors:

Dr. rer. nat. Tino Schleptz

Dr. Carla da Conceio Caramujo Rocha de Carvalho

Examination Committee

Chairperson: Dr. Helena Maria Rodrigues Vasconcelos Pinheiro

Supervisor: Dr. Carla da Conceio Caramujo Rocha de Carvalho

Members of the Committee: Dr. Maria Catarina Marques Dias de Almeida

November 2015

II

To my parents

III

Acknowledgements

The present thesis was completed at the Institute for Biochemical Engineering (AVT.BioVT) of the RWTH

Aachen University titled Online-analytical characterization of 2,3-butanediol production

by Bacillus licheniformis DSM 8785. I am grateful to everyone who made this possible.

First and foremost, Id like to thank the wonderful people without whom none of this could have been done:

my incredibly supporting parents Maria and Romo, my brother Tiago and my family. Thank you for

everything.

Dr. rer. nat Tino Schleptz for being my supervisor and of this thesis. He gave me the opportunity to work

in an international environment at AVT-BioVT for my dissertation. Thank you for all the inspiration, support,

motivation and the productive discussions we had throughout the entire project. And for all the challenges

and opportunities which allowed me to go further in my career.

Dr. Carla Carvalho for all the support, guidance and constructive feedback. For always being available to

give a helping hand and for believing that I could do the job.

The best friends a person could want: Ana Rita Santiago, Joana Marques, Hlder Baptista, Nuno Santos,

Nuno Salgueiro, Marta Vaz, Vanessa Paquete, Diana Jorge, Ana Raquel Bragana, Ricardo Ribero, Sara

Mendes, Joo Mateus, Ins Graa, Bruno Oliveira, Raquel Pires, Diogo Sebastio and all my fellows

already-graduated-or-soon-to-be Biological Engineers.

Steffen Eich, who was always there for me. He always tried to cheer me up and made my life easier. I thank

him for reading this manuscript and all his patience with me. My deepest gratitude for all the moments and

the important lessons learnt while in Aachen.

Evi Breugelmans, for the companionship away from home, for not letting me forget how important it is never

to give up and for providing relief under the most stressful situations while in Aachen.

Bio-AVT group for all the contributions, good working environment and their friendly aid. A special thanks

to David Flitsch, Tobias Habicher, Lena Meiner, Andreas Schulte and Dirk Kreyenschulte for all the help

and support during my experiments and the students: Bertram Geinitz, Benedikt Heyman, Diana Noffke,

Christina Kavelage, Kyra Hoffmann, Lena Altenhoff, Pia Hndel, Sarah Stachurski and Simon Seidl.

Without them the work would not have been that much fun.

IV

Abstract

Online-analytical characterization of 2,3-butanediol production by

Bacillus licheniformis DSM 8785

Microbial production of 2,3-butanediol (2,3-BD) has a history of more than 100 years. By developing an

efficient bio-based process for the microbial production of 2,3-BD from renewable resources, fossil fuel

supplies can be preserved and environmental benefits can be obtained.

Bacillus licheniformis conducts the metabolic pathway of 2,3-BD fermentation under microaerophilic

conditions and is, thus, an effective producer of 2,3-BD. Oxygen is a limiting substrate with regard to growth

and an inhibitor with regard to the specific metabolite productivity. The study of 2,3-BD production by

B. licheniformis DSM 8785 was carried out during batch cultivations and bioreactor scale in consideration

of medium composition (carbon source and concentration) and cultivation parameters (e.g. temperature,

stirring speed) by applying different online analytical techniques to develop an improved fermentation

process.

Shake flasks in a Respiration Activity Monitoring System (RAMOS) were used to characterize the microbial

respiration activity, oxygen transfer rate (OTR), and to study the metabolism and growth of the bacteria.

The initial substrate concentration and temperature effected the yield of 2,3-BD production and the OTR

profile. From 180 g/L of glucose at 30C and 200 rpm, 26 g/L 2,3-BD were produced.

The results from shake flask scale provided data to perform fermentations in a 3 L-bioreactor. Online-

analyzes of DOT (Dissolved oxygen tension), redox potential, pH and stirred rate affect the 2,3-BD

productivity. Oxygen transfer coefficients (kLa) of 58.65 h1 and 53.72 h1 were found to be optimal for

conversion of 60 g/L and 180 g/L of glucose, respectively. 180 g/L of initial glucose concentration, 30C,

400 rpm and an aeration rate of 0.5 vvm were found to be the conditions for a successful scale-up by

B. licheniformis DSM 8785. In conclusion, fermentations in RAMOS-device can provide a platform for

high-throughput studies of 2,3-BD production.

Key-words: 2,3-butanediol; Oxygen limitation; RAMOS; Oxygen Transfer Rate (OTR);

Bacillus licheniformis

V

Caracterizao e anlise online da produo do 2,3butanediol por

Bacillus licheniformis DSM8785

A produo microbiana de 2,3-butanediol (2,3-BD) tem uma histria com mais de 100 anos. Atravs do

desenvolvimento de um processo eficiente de base biolgica para uma produo microbiana de 2,3-BD, a

partir de recursos renovveis, fontes de combustveis fsseis e pode ser preservada obtendo, assim,

benefcios ambientais.

Bacillus licheniformis realiza a via metablica fermentativa do 2,3-BD, sob condies de microaerofilia, e

assim, um produtor eficaz de 2,3-BD. O oxignio um substrato limitante em termos de crescimento e um

inibidor no que se refere produtividade especfica dos metabolitos. Estes resultados so discutidos nesta

tese. O estudo da produo de 2,3-BD por B. licheniformis DSM 8785 foi realizado durante uma cultura

batch e em escala de bioreactor, tendo em considerao a composio do meio (fonte de carbono e

concentrao) e os parmetros de cultura (temperatura, taxa de agitao) atravs de diferentes tcnicas

de anlise online para o desenvolvimento de um processo de fermentao melhorada.

Frascos de agitao foram utilizados no dispositivo RAMOS (Sistema de Monitoramento de Actividade de

Oxignio) onde foram utilizados para caracterizar a actividade respiratria microbiana, a taxa de

transferncia de oxignio (OTR), o metabolismo e crescimento das bactrias. A concentrao inicial de

substrato e temperatura tm efeitos no rendimento da produo de 2,3-BDl e do perfil de OTR. A partir de

180 g/L de glucose, a 30 C e 200 rpm, 26 g/L de 2,3-BD foram produzidos.

Os resultados em RAMOS foram suficientemente satisfatrios para realizar fermentaes num biorreactor

de 3L. As anlises online de DOT (tenso de oxignio dissolvido), potencial redox, pH e a da taxa de

agitao afectam a produtividade do 2,3-BD. Foram encontrados os coeficientes ptimos de transferncia

de oxignio (kLa), 58.65 h-1 e 53.72 h-1 para 60 g/L e 180 g/L, respectivamente. Concentrao inicial 180 g/L

de glucose, a 30 C, 400 rpm e uma taxa de arejamento de 0,5 vvm foram detectadas como as condies

ideais para um aumento de escala bem-sucedida por B. licheniformis DSM 8785. Pode-se, por isso,

concluir que as fermentaes em RAMOS podem ser uma boa plataforma para estudos de alto rendimento

de produo de 2,3-BD.

Palavras-chave: 2,3-butanediol; limitao de oxignio; RAMOS; Taxa de Transferncia de Oxignio

(OTR); Bacillus licheniformis

VI

Table of Contents

Acknowledgements ...................................................................................................................................... III

Abstract ........................................................................................................................................................ IV

Table of Contents ......................................................................................................................................... VI

Tables list ..................................................................................................................................................... IX

Figures list ..................................................................................................................................................... X

Nomenclature ............................................................................................................................................ XIV

1. Introduction ............................................................................................................................................ 1

1.1. 2,3-Butanediol ............................................................................................................................... 1

1.1.1. Chemical properties .............................................................................................................. 1

1.1.2. Importance of 2,3-Butanediol ................................................................................................ 2

1.2. Microbial 2,3-BD producers ........................................................................................................... 3

1.3. Metabolic Pathway ........................................................................................................................ 5

1.4. Substrates ..................................................................................................................................... 7

1.5. Factors affecting the 2,3-BD production ....................................................................................... 8

1.5.1. pH .......................................................................................................................................... 9

1.5.2. Temperature .......................................................................................................................... 9

1.5.3. Aeration ............................................................................................................................... 10

1.5.4. Agitation .............................................................................................................................. 11

1.5.5. Medium Composition .......................................................................................................... 11

1.5.6. Substrate concentration ...................................................................................................... 12

1.5.7. Water activity ....................................................................................................................... 13

1.6. Gas-liquid mass transfer ............................................................................................................. 13

VII

1.7. Reactor Operation Mode ............................................................................................................. 18

2. Theoretical Background .......................................................................................................................... 20

2.1. The RAMOS device .................................................................................................................... 20

2.2. Scale-up from shake flasks to bioreactors .................................................................................. 22

3. Purpose ............................................................................................................................................... 26

4. Materials and Methods ........................................................................................................................ 27

4.1. Chemicals and reagents ............................................................................................................. 27

4.2. Culture Media .............................................................................................................................. 27

4.2.1. Medium for agar plates ....................................................................................................... 27

4.2.2. Medium for shake flask and bioreactor cultivations ............................................................ 28

4.3. Bacterial Strain ............................................................................................................................ 29

4.3.1. Microorganism ..................................................................................................................... 29

4.3.2. Culture preservation ............................................................................................................ 29

4.4. Culture conditions ....................................................................................................................... 29

4.4.1. Precultures .......................................................................................................................... 29

4.4.2. RAMOS cultivation with special shake flasks ..................................................................... 30

4.4.3. RAMOS cultivation .............................................................................................................. 30

4.4.4. Cultivations in the 3 L bioreactor scale ............................................................................... 31

4.5. Analytical procedures .................................................................................................................. 32

4.5.1. Optical density ..................................................................................................................... 32

4.5.2. Cell Dry Weight (CDW) ....................................................................................................... 32

4.5.3. pH measurement ................................................................................................................. 32

4.5.4. High performance liquid chromatography (HPLC) .............................................................. 32

VIII

5. Results ................................................................................................................................................ 34

5.1. 2,3-BD production by Bacillus licheniformis DSM 8785 in special shake flasks ......................... 34

5.2. 2,3-BD production by Bacillus licheniformis DSM 8785 with RAMOS flask experiments ........... 37

5.2.1. Influence of initial sugar concentration on 2,3-Butanediol production ................................ 38

5.2.2. Influence of initial temperature on 2,3-Butanediol production ................................................... 46

5.2.3. Influence of maximum OTR on 2,3-Butanediol production ....................................................... 51

5.3. Scale-up of 2,3-BD with Bacillus licheniformis DSM 8785 to 3L-bioreactor scale ...................... 53

5.3.1. Influence of initial glucose concentration in 2,3-BD production .......................................... 53

5.3.2. Correlation of the stirring rate and the oxygen mass transfer coefficient ................................. 55

5.3.3. 2,3-BD production in the light of oxygen limitation .................................................................... 56

6. Discussion ........................................................................................................................................... 61

6.1. 2,3-BD production in special shake flasks ....................................................................................... 61

6.2. 2,3-BD with Bacillus licheniformis DSM 8785 with RAMOS flask experiments ............................... 62

6.3. 2,3-BD with Bacillus licheniformis DSM 8785 with scale-up to 3L bioreactor scale ........................ 64

7. Conclusions and prospects ..................................................................................................................... 67

8. References .............................................................................................................................................. 68

IX

Tables list

Table 1 Stereoisomeric forms of 2,3-butanediol produced by bacterial fermentation. ................................. 4

Table 2 2,3-BD productions with different microorganisms and raw materials (A. Singh, 1995). ................ 7

Table 3 Typical Respiratory Quotient values for various substrates (Klaus, 1980) .................................... 17

Table 4 Comparison of the 2,3-BD production using different microorganisms ......................................... 19

Table 5 Values of v, w, x and y with z=0 for the different shake flasks. ..................................................... 23

Table 6 Composition of the Agar medium .................................................................................................. 27

Table 7 Basal medium composition (with 180 g/L) (Jurchescu I.-M., 2013) .............................................. 28

Table 8 Main Characteristics used during the HPLC measurement .......................................................... 33

Table 9 Comparison of yield, productivity, residual glucose, ethanol and acetoin concentrations during

cultivation with B. licheniformis DSM 8785 using different initial glucose concentrations and filling volumes.

.................................................................................................................................................................... 41

Table 10 Average kLa values for each speed control ................................................................................. 56

X

Figures list

Figure 1 Stereoisomers of 2,3-BD (Ji X.-J. , 2011) ...................................................................................... 2

Figure 2 Applications of 2,3-butanediol. (Afschar, 1993) ............................................................................. 3

Figure 3 Mixed-acid-2,3-butanediol pathway [adptaded from Henriksen & Nilsson (2001), (Jurchescu I.-M.

, 2013)] .......................................................................................................................................................... 6

Figure 4 The effect of relative oxygen availability on fractional product yields in B. polymyxa ................. 10

Figure 5 Mass transfer phenomena which occur during the transport of oxygen from the gas phase to the

cells in a bioreactor (Bchs, 2014) .............................................................................................................. 14

Figure 6 Graphical representation of the two film theory. A substance is transported from phase 1 to phase

2, passing through two stagnant films and an interface. ............................................................................. 15

Figure 7 a) RAMOS shake flask; b) RAMOS device at the chair of Biochemical Engineering at the RWTH

Aachen University, equipped with 8 shake flasks. ...................................................................................... 20

Figure 8 General set-up of a RAMOS device as introduced by Anderlei et al. (Sven Hansen, 2012) ....... 21

Figure 9 OTR profiles for typical metabolic phenomena determined with a RAMOS device (Tibor Anderlei,

2000) ........................................................................................................................................................... 22

Figure 10 Determination of OTR, CTR and RQ by exhaust gas analysis in a bioreactor (Bchs, 2014) .. 24

Figure 11 Characterization of the bioreactor by determination of kLa value (Bchs, 2014) ....................... 25

Figure 12 a) A specially configuration of shake flask; b) Shake Bioreactor device from the Biochemical

Engineering at the RWTH Aachen University, equiped with 8 shake flasks with 100 mL; c) View of the

configuration in the shake flask. .................................................................................................................. 30

Figure 13 Front view of the 3L-bioreactor with the specific electrodes from the Biochemical Engineering at

the RWTH Aachen University. .................................................................................................................... 31

Figure 14 Effect of flask configuration in OTR, CTR and RQ on 2,3-BD production with

B. licheniformis DSM 8785. Conditions: 500 mL shake flasks, 100 mL medium containing 180 g/L glucose,

T=30 C, N=100 rpm, d0= 5 cm and initial pH 6.5. ..................................................................................... 35

file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189300file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189303file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189304file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189304file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189309file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189311file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189311file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189311file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189312file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189312file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189313file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189313file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189313

XI

Figure 15 Effect of flask configuration in OTR integral on 2,3-BD production with B. licheniformis

DSM 8785. Conditions: 500 mL shake flasks, 100 mL medium containing 180 g/L glucose, T=30 C, N=100

rpm, d0= 5 cm and initial pH 6.5. ................................................................................................................ 36

Figure 16 Effect of flask configuration in fermentation end-products and 2,3-BD production with

B. licheniformis DSM 8785. Conditions: 500 mL shake flasks, 100 mL medium containing 180 g/L glucose,

T=30 C, N=100 rpm, d0= 5 cm and initial pH 6.5. ..................................................................................... 37

Figure 17 Effect of the initial glucose concentration on 2,3-BD production with

B. licheniformis DSM 8785. Conditions: two shake flasks with 20 mL and two shake flasks with 40mL

with medium containing 180g/L - pH=6.64 (left side) and 250g/L - pH=6.59 (right side) of initial glucose

concentration, T=30 C, N=200 rpm and d0= 5 cm .................................................................................... 38

Figure 18 Effect of initial glucose concentration on 2,3-BD production with

B. licheniformis DSM 8785. Conditions: two shake flasks with 20 mL and two shake flasks with 40mL with

medium containing 180g/L - pH=6.64 (left side) and 250g/L - pH=6.59 (right side) of initial glucose

concentration, T=30 C, N=200 rpm and d0= 5 cm .................................................................................... 39

Figure 19 Comparison of final products concentrations during cultivation with

B. licheniformis DSM 8785 using different initial glucose concentrations and filling volumes.

Conditions: two shake flasks with 20 mL and two shake flasks with 40mL with medium containing 180g/L -

pH=6.64 (on top) and 250g/L - pH=6.59 (below) of initial glucose concentration, T=30 C, N=200 rpm and

d0= 5 cm. .................................................................................................................................................... 40

Figure 20 Comparison of OTR, CTR and RQ during cultivation with B. licheniformis DSM 8785 using

different initial glucose concentrations and filling volumes. Conditions: two shake flasks with 10, 15,

20 and 25 mL with medium containing 50g/L - pH=6.69 (left side) and 180g/L - pH=6.69 (right side) of initial

glucose concentration, T=37 C, N=200 rpm and d0= 5 cm ....................................................................... 42

Figure 21 Comparison of OTR integral during cultivation with B. licheniformis DSM 8785 using



glucose concentrtion, T=37 C, N=200 rpm and d0= 5 cm ......................................................................... 44

file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189314file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189314file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189314file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189319file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189319file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189319file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189319file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189320file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189320file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189320file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189320

XII










different initial temperatures and filling volumes. Conditions: two shake flasks with 10, 15, 20 and 25

mL at 37 C - pH=6.60 (left side) and 30 C - pH=6.69 (right side), 50 g/L of initial glucose concentration,

N=200 rpm and d0= 5 cm ........................................................................................................................... 48


different initial temperatures and filling volumes. Conditions: two shake flasks with 10, 15, 20 and 25 mL

at 37 C - pH=6.60 (left side) and 30 C - pH=6.69 (right side), 50 g/L of initial glucose concentration, N=200

rpm and d0= 5 cm ....................................................................................................................................... 49



at 30 C - pH=6.69 (left side) and 37 C - pH=6.40 (right side), 180 g/L of initial glucose concentration,

N=200 rpm and d0= 5 cm ........................................................................................................................... 50



at 30 C - pH=6.69 (left side) and 37 C - pH=6.40 (right side), 180 g/L of initial glucose concentration,

N=200 rpm and d0= 5 cm ........................................................................................................................... 51

Figure 28 Comparison of maximum OTR during cultivation with B. licheniformis DSM 8785 using

different initial glucose concentration. Conditions: two shake flasks with 10, 15, 20 and 25 mL with

medium containing 50 g/L and 180g/L (down) and two shake flasks with 20 and 40 mL with medium

containing 180g/L and 250 g/L (up) of initial glucose concentration at 30 C, 200 rpm. ............................ 52

file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189321file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189321file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189321file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189321file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189322file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189322file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189322file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189322file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189323file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189323file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189323file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189323file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189324file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189324file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189324file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189324file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189325file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189325file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189325file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189325file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189326file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189326file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189326file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189326file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189327file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189327file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189327file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189327

XIII

Figure 29 The time-course of fermentation by B. licheniformis DSM 8785 using 60 g/L of initial glucose

concentration as substrate. The fermentation was performed at 30 C in a 3-L bioreactor with stirring at

400 rpm and airflow at 0.5 vvm. .................................................................................................................. 54

Figure 30 The time-course of fermentation by B. licheniformis DSM 8785 using 180 g/L of initial glucose

concentration as substrate. The fermentation was performed at 30 C in a 3-L bioreactor with stirring at

400 rpm and airflow at 0.5 vvm. .................................................................................................................. 55

Figure 31 Average kLa value calculated by changing agitation speed as a control strategy. .................... 56

Figure 32 Influence of the dissolved oxygen tension in the profile of potential redox, OTR, CTR and RQ.

Batch fermentation of 2,3-BD from 60 g/L of initial glucose concentration by B. licheniformis DSM 8785.

Fermentation was performed at 30 C in a 3-L bioreactor with stirring at 400 rpm and airflow at 0.5 vvm..

.................................................................................................................................................................... 57

Figure 33 Influence of the dissolved oxygen tension in the profile of potential redox, OTR, CTR and RQ.

Batch fermentation of 2,3-BD from 180 g/L of initial glucose concentration by B. licheniformis DSM 8785.

Fermentation was performed at 30 C in a 3-L bioreactor with stirring at 400 rpm and airflow at 0.5 vvm..

.................................................................................................................................................................... 59

file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189328file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189328file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189328file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189329file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189329file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189329file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189330file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189331file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189331file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189331file:///C:/Users/Elsa/Downloads/Elsa%20thesis_CC%206Dez.docx%23_Toc437189331

XIV

Nomenclature

2,3-BD 2,3-butanediol

a Ratio between the interfacial area and the liquid volume

ATCC American Type Culture Collection

ATP Adenosine-5'-triphosphate

CDW Cell dry weight

CG,i Gas concentration in the gas side of an interface

CL Gas concentration in the liquid

CL,i Gas concentration in the liquid side of an interface

CTR Carbon Dioxide Transfer Rate

d Flask diameter

d0 Shaking diameter

DSMZ Deutsche Sammlung fr Mikroorganismen und Zellkulturen

GRAS Generally Regarded As Safe

h Henrys law coefficient

HPLC High performance liquid chromatography

IUPAC International Union of Pure and Applied Chemistry

KG Global mass transfer coefficient on the gas side

KL Global mass transfer coefficient on the liquid side

kLa Volumetric oxygen transfer coefficient

NAD+/NADH Nicotinamide adenine dinucleotide

NB Nutrient broth

OD Optical density

OTR Oxygen Transfer Rate

OUR Oxygen uptake rate

Pg Partial Pressure of a Gas

pO2 Oxygen partial pressure

PUMAs Polyurethane-melamides

R Universal gas constant

RAMOS Respiratory Activity Monitoring System

rpm Revolutions per minute

RQ Respiratory quotient

T Temperature

VG Headspace volume of the flask

VL Volume of liquid in the flask

1

1. Introduction

Several processes for the conversion of biomass in biorefineries for energy, chemistry-derived

products and some fuels production have been improved worldwide (Ji X.-J. , 2011).

2,3-butanediol (2,3-BD) is a bulk-chemical with a large number of industrial applications. The

heating value of 2,3-BD of 27 198 J/g is comparable to ethanol (29 055 J/g) and methanol

(22 081 J/g). Thus, 2,3-BD or its derivatives might be easily used as a liquid fuel. The production

of 2,3-butanediol assumed particular importance after the Second World War due to the need for

synthetic rubber. By developing an efficient bio-based process for the microbial production of

2,3-BD from renewable resources, fossil fuel supplies can be preserved and environmental

benefits can be obtained (Celinska, 2009).

1.1. 2,3-Butanediol

1.1.1. Chemical properties

2,3-BD is also known as 2,3-butylene glycol, dimethylene glycol and dimethylethylene glycol. The

IUPAC name is butane-2,3-diol and its molecular formula is CH3CH(OH)CH(OH)CH3. The

molecular weight is 90.12 g/mol and in physical aspects, this compound is colorless, may be

presented as an odorless liquid or in crystalline form, is hygroscopic and soluble in water (Syu,

2001).

Due to the presence of two chiral carbon centers, the 2,3-BD has three stereoisomers and the

boiling points of the three isomers range between 177C and 182C. In the Figure 1 the three

isomers of 2,3-BD are depicted: the optically inactive isomer (R,S)-2,3-BD, also known as

meso-2,3-BD; the optically active forms are (2R,3R)-BD or D-(-)-2,3-BD and (2S,3S)-BD

or L-(+)-2,3-BD. The levo-isomer has a low freezing point (-60C) which represents the basis for

the commercial interest of 2,3-BD as an antifreeze agent (Celinska, 2009).

2

1.1.2. Importance of 2,3-Butanediol

The production of 2,3-BD is of great interest because of the various applications of 2,3-BD. It can

be used in the manufacture of printing inks, perfumes, chemicals, foods, fumigants, moistening

agents, fuels, explosives, plasticizers, pharmaceuticals and aeronautical products. 2,3-BD can be

derived from the bioconversion of natural resources (Ji X.-J. e., 2011; Lan Ge, 2011).

Methyl ethyl ketone (MEK), the dehydration product of 2,3-BD, can be furthermore used for resins,

paints and other solvents. This product is an effective fuel additive with a higher heat of

combustion value than ethanol. MEK can be used to produce high quality aviation fuels due to

the subsequent large formation of octane isomers. Furthermore, 2,3-BD can be converted into

1,3-butadiene (1,3-BD), which can be used to produce synthetic rubber, polyester and

polyurethane synthesis (Lan Ge, 2011). 2,3-BD can also be used for the production of oxides,

nitrogen, ether and ketone derivatives as well in halogenated substitutes, esters of monobasic

and dibasic acids (Liebmann, 1945).

After catalytic dehydrogenation diacetyl forms of 2,3-BD, can be used as food additives. A good

example of that is the improvement of butter taste and the wine flavor (Lan Ge, 2011).

Polyurethane-melamides (PUMAs) are synthesized by esterification of 2,3-BD with maleic acid.

PUMAs can have a huge impact in cardiovascular applications since this process in the polymer

chains can activate double bonds and when introduced, it may undergo further modification via

specific grafting, thus improving the tissue compatibility (Celinska, 2009; Petrini, 1999). In the

Figure 2 it is represented the most important uses of 2,3-BD (Celinska, 2009).

Figure 1 Stereoisomers of 2,3-BD (Ji X.-J. , 2011)

3

Figure 2 Applications of 2,3-butanediol. (Afschar, 1993)

The chemical synthesis of 2,3-BD is based on the breaking bonds of the hydrocarbon fraction

(C4), crack gases, and after butadiene and isobutene have been removed. This fraction contains

up to 77% butenes, while the residual 23% correspond to a mixture of butane and isobutene

(Heinz Grfje, 2000). The chemical synthesis requires high pressure, high temperature,

expensive catalysts, release of toxic intermediates and dependence on non-renewable materials,

which result in complex process and low yield.

2,3-BD production is growing at an annual rate of 4-7%. Since 2,3-BD has a unique structure and

its chemical synthesis is pricey, large scale chemical synthesis has not been established.

Considering the limitations of the chemical industry to produce pure 2,3-BD most of the 2,3-BD is

not sold as a separate product, but mixed with other products (Lan Ge, 2011; Syu, 2001).

1.2. Microbial 2,3-BD producers

Compared to the chemical synthesis, the microbial 2,3-BD production is cheaper. In microbial

fermentation processes, typically carbon sources like glucose and sucrose from renewable

feedstocks are used. Furthermore, no toxic products are generated making the microbial

fermentation process quite appealing (Jiang & Liu, 2014).

4

Microbial 2,3-BD production dates back to 1906, when Harden and Walpole reported for the first

time results about the 2,3-BD synthesis with Klebsiella pneumoniae. Industrial-scale production

of 2,3-BD by fermentation was first proposed in 1933 by Fulmer EI et al (1933).

2,3-BD fermentation regained interest in the last decade mostly because the fermenting sugars

and lignocellulosic hydrolysates are cheap substrates. Since that, numerous efforts have been

made in improving fermentation processes. In the production of a particular isomer, the

stereoisomeric forms are dependent on the microorganism used. Different microorganisms are

able to synthesize different stereosiomers of 2,3-BD. Yet, most strains form a mixture of two

isomers. Strains of genera Klebsiella and Enterobacter produce L-(+)- and meso-2,3-BD, strains

of the genus Bacillus produce D-(-)- and meso-2,3-BD, strains of the genus Serratia form

meso-2,3-BD and B. polymyxa yields pure D-(-)-2,3-BD (Ji X.-J. , 2011). Table 1 summarizes the

nature of 2,3-BD produced by different microbial species (A. Singh, 1995).

Table 1 Stereoisomeric forms of 2,3-butanediol produced by bacterial fermentation.

Microorganism Stereoisomeric

Pseudomonas hydrophila 50% racemic; 48% meso; 2% levo

Bacillus polymyxa D(-) levo

Bacillus subtilis 65% levo, 35% meso

Klebsiela pneumoniae (Aerobacter

aerogenes) 5%-14% dextro, 86%-95% meso

Serratia spp. Mainly meso

Many bacterial strains are able to synthesize 2,3-BD from pyruvate. However, only a few are able

to produce 2,3-BD in significant quantities. Although 2,3-BD formation has been observed in

several yeasts (Kloeckera apiculate) and filamentous fungi (Rhizopus nigricans), the conversion

efficiency was extremely low (approximately 0.003 g/g glucose) in these species. Thus, bacteria

represent the only organisms of industrial importance in 2,3-BD production. They can convert

xylose to a variety of products in the absence of oxygen. The rate, yield and products formed

depend not only on the diverse metabolic pathways operating during anaerobic fermentation but

also on the species, strains, substrates and culture conditions used. (A. Singh, 1995)

Bacterial species considered to be of industrial importance in 2,3-BD production belong to the

genera Klebsiella, Enterobacter, Bacillus and Serratia (Ji X.-J. , 2011). Some investigations were

carried using native producers, such as Klebsiella pneumoniae, Klebsiella oxytoca, Serratia

marcescens and Enterobacter aerogenes (Jiang Y., 2014). Since these microorganisms belong

to risk group 2 (pathogenic) they are not favorable for industrial-scale fermentation and 2,3-BD

5

producers belonging to group 1 (safe) would be highly preferred, as Bacillus species. The purpose

was to find the non-pathogenic microorganisms for the production of 2,3-BD at developed

bioprocesses in large scale, which is possibly less expensive.

1.3. Metabolic Pathway

During bacterial metabolism, monosaccharides or other carbon sources can be converted to

2,3 - BD via pyruvate as a central metabolic intermediate. Pyruvate is formed from hexoses (e.g.

glucose) in the Embden-Meyerhof pathway (EMP pathway or gycolysis) or from pentoses (e.g.

xylose) in a combination of the pentose phosphate pathway and Embden-Meyerhof pathway.

Pyruvate is the branch point where the catabolic reactions diverge into two different

energy-production pathways. Either pyruvate is channeled via acetyl-CoA into the tricarboxylic

acid cycle under aerobic conditions or it is subject to the mixed-acid-2,3-BD fermentation under

anaerobic conditions in three steps. In the branch leading to 2,3-BD, the first step is the

decarboxylation of pyruvate to -acetolactate in the presence of the enzyme -acetolactate

synthase. In turn, -acetolactate is converted into acetoin (acetyl methyl carbinol) by the enzyme

-acetolactate decarboxylase. In a third step, acetoin is reversibly reduced to 2,3-BD by the

enzyme acetoin reductase (2,3-BD dehydrogenase) (Ji X.-J. , 2011). This reaction helps to

maintain the NAD+/ NADH2 balance inside the cell by the anaerobic reactions in absence of

oxygen (Jansen, 2005). The enzymes required for 2,3-BD production are expressed during the

late log and stationary phase when oxygen limitation exists. (Celinska, 2009). Figure 3 illustrates

the mixed-acid-2,3-BD fermentation pathway. The main products are glycerol, ethanol, acetate,

succinate, lactate and formate.

The net reaction of 2,3-BD production from glucose is summed up in equation (1), with a

maximum yield of 0.5 g/g. The theoretical maximum molar yield of 2,3-BD from glucose is 1.0 and

from xylose is 0.83 (A. Singh, 1995).

(1)

6

Figure 3 Mixed-acid-2,3-butanediol pathway [adptaded from Henriksen & Nilsson (2001), (Jurchescu I.-M. ,

2013)

2,3-BD might play a role in preventing intracellular acidification by changing the metabolism from

acid production to formation of neutral compounds like 2,3-BD. Consequently, accumulation of

acidic products in the medium causes a decrease in extracellular pH media, which results in

intracellular acid accumulation due to the transmembrane pH gradient (Van Houdt, 2007; Jiang

Y., 2014; Blomqvist, 1993).

The reversible reaction between acetoin and 2,3-BD can influence the maintenance of the

intracellular NADH/NAD+ balance, where a conversion of NAD+ to NADH occurs. (Blomqvist,

1993) NAD+ can be regenerated by the reduction of acetoin to 2,3-BD. After the glucose is

exhausted and the production of NADH stops, the reverse reaction could occur.

7

1.4. Substrates

For the development of an economical fermentation process of 2,3-BD, the use of low-priced

carbohydrate raw materials is essential, since the major cost of the process is mostly affected by

the substrate cost (Celinska, 2009). Microorganisms are able to ferment a broad variety of

substrates to 2,3-BD, including pure sugars, noncellulosic and lignocellulosic substrates (Ji X.-

J. et al., 2011). The majority of substrates for 2,3-BD production are divided in pure sugars,

non-cellulosic substrates and lignocellulosic substrates. Table 2 shows bacterial species capable

of 2,3-BD production with different feedstock.

Table 2 2,3-BD productions with different microorganisms and raw materials (A. Singh, 1995).

Microorganism Raw Materials

Bacillus polymyxa

Molasses

Bacillus subtillis

Klebsiella pneumoniae

Aeromonas hydrophila

Serratia marcescens

B. polymyxa Corn strach, acid hydrolyzed wheat, whole

wheat, barley Aerobacter aerogenes

K. pneumoniae

Wood hydrolyzed, starch

A. hydrophila

B. polymyxa

Sulfite waste liquor

B. subtilis

A. hydrophila

A. Aerogenes

Serratia species Sucrose

A. Aerogenes

Pentoses K. pneumoniae

Klebsiella oxytoca

B. polymyxa

8

Glucose is commonly used in fermentation processes. Escherichia coli are able to grow on

glycerol and a variety of pentoses and hexoses (Jiang & Liu, 2014). Bacillus licheniformis,

Paenibacillus polymyxa and Klebsiella oxytoca can growth too, including on mannose, galactose,

xylose, sucrose and lactose (Jiang & Liu, 2014; Champluvier B., 1989; de Mas, 1988; Nilegaonkar

S. S., 1996). B. licheniformis and P. polymyxa can additional ferment fructose, cellobiose (two

molecules of glucose) and glucose (Nilegaonkar S. S., 1996; de Mas, 1988). Some studies have

been made with different substrates and in the same microorganisms. The productivity of 2,3-BD

is maximum with glucose followed by cellobiose, fructose, sucrose, starch and mannose with

B. licheniformis. K. pneumoniae showed the best result with glucose followed by mannose and

cellobiose. In the case of P. polymyxa, the best results for 2,3-BD production were attained with

glycerol and starch (Nilegaonkar S. S., 1996).

Non-cellulosic substrates have the potential for being a low cost source for 2,3-BD production.

One of the most promise alternatives is the food industry residue. It presents a high sugar content

like starch hydrolysate derived from corn transformation, raw and decolored molasses from sugar

beet extraction and whey from cheese manufacture. The molasses, a residue from sugar cane

juice, was studied too. A little quantity of nutrient supplementation is required for the conversion

of molasses. 2,3-BD concentrations above 100 g/L with K. oxytoca was achieved in batch

fermentation (Ji X.-J. , 2011).

Lignocellulosic is being considered as the largest renewable biological resource and as an

alternative approach in the conversion of biomass substrates to liquid fuels and chemical

feedstocks. The production of 2,3-BD from lignocellulosic wastes includes various agricultural

residues (straws, hulls, stems, stalks), several types of woods and municipal waste, but the

compositions of these materials vary (Saha, 2003; Lo, 2009). Wood hydrolysate is an example of

a low cost lignocellulosic substrate and widely available agricultural residue. In the begin of 1980s,

Yu et al. (1982) realized that K. pneumoniae is adequate to produce 2,3-BD from enzymatically

hydrolysed wood hemicellulose and acid. B. polymyxa produced 2,3-BD when grown on

Flavourzyme-hydrolysed (HF) aspen wood while a thermophilic B, licheniformis strain X10 is a

candidate for the development of efficient industrial production of 2,3-BD from corn stover

hydrolysate (Celinska, 2009; Ji X.-J., 2011; Li L., 2014).

1.5. Factors affecting the 2,3-BD production

Many cultural, environment and nutritional factors can affect the 2,3-BD fermentation due to the

metabolism of 2,3-BD. In this chapter some parameters will be introduced.

9

1.5.1. pH

In general, the pH is a fundamental parameter in the regulation of bacterial metabolism. The

influence is especially important in the processes involving multiple end-product formation like in

mixed acid 2,3-BD fermentation. The pH of the fermentation medium affects the biomass

composition and the metabolic pathway that the bacteria take. In mixed acid 2,3-BD fermentation,

alkaline conditions with initial pH values above 6.3-6.5 favor, the production of organic acids and

consequently 2,3-BD yield decreases. In contrast, acidic conditions with pH values below 6.3 led

to more than 10-fold reduction of organic acid synthesis and 3- to 7-fold increase in 2,3-BD

formation (Garg & Jain, 1995).

Some bacteria counteract acidification by switching the metabolism from acid production to

synthesis of neutral compounds like alcohols or 2,3-BD (Houdt, 2007). The ratio of 2,3-BD to the

precursor acetoin might be close to 25 between pH-values of 5.2 to 6.0. Above a pH of 7.0 2,3-BD

is not anymore produced, but formic acid concentration rises and CO2 level falls suggesting that

the intracellular NAD/NADH balance is maintained by reduction of CO2 to formic acid under these

conditions. A pH above 6 causes a sharp decrease in the activity of -acetolactate synthase one

of the key enzymes in the 2,3-BD pathway (Jansen N. B., 2005). Production of lactic acid and

acetic acid is minimal at pH-values below 5 and rapidly increases at pH-values over 6 (A. Singh,

1995). However, the optimum pH-value for 2,3-BD production depends on the microorganism and

substrate employed (Celinska, 2009).

According to Grover et al (1990), the optimum pH value is in the range of 6.0-6.2 for most of the

substrates (including wood hydrolysates). With B. licheniformis and B. polymyxa maximum

2,3-BD production from glucose was obtained at a pH of 6.0 (Raspoet, 1991) while for

B. amyloliquefaciens an optimum pH was found to be 6.5. For K. pneumoniae, pH values in the

range of 5.2-5.6 or 5.0-6.0 were best when glucose or xylose and sucrose were chosen as

substrates, respectively. (Yang, 2011). Using E. aerogenes a pH of 6 was found to be best for

2,3-BD formation (Celinska, 2009).

1.5.2. Temperature

The efficiencies of bioprocesses are strictly temperature-dependent due to the strong

dependence of enzymatic activity and cellular maintenance upon temperature. Since 2,3-BD

synthesis is a growth-associated phenomenon, the optimum temperature for product formation

should be similar to the optimum temperature for maximum biomass yield (Garg & Jain, 1995).

Again, the optimum temperature for 2,3-BD production depends on the strain and substrate used

for cultivation. For that reason, the optimal value should be determined individually for each case

(Celinska, 2009). Li et al. (2013) reported that the highest concentration of 2,3-BD was obtained

10

when the temperature for the cultivation of a thermophilic B. licheniformis strain was 50C. In case

of K. pneumoniae, temperatures in the range of 35 to 37C were considered as best for maximum

2,3-BD production (Ji X.-J. , 2009). A temperature of 37C was determined as optimal for batch

cultivation using B. amyloliquefaciens (Yang, 2011). In contrast, Jurchescu et al. in a fed-batch

cultivation has found an optimum value of 30C for B. licheniformis (Jurchescu, 2013).

1.5.3. Aeration

One of the most important parameter for 2,3-BD production is considered to be the oxygen

availability. 2,3-BD production is a product of anaerobic fermentation and is formed under oxygen

limited or microaerophilic conditions. Nevertheless, aeration was shown to enhance 2,3-BD

synthesis, particularly at high substrate concentrations or during the fermentation of pentose

sugars (Celinska, 2009; A. Singh, 1995). However, too high oxygen supply prevents 2,3-BD

production by a rapid and irreversible inactivation of the -acetolactate synthase. Higher oxygen

supply favors the production of cell mass at the expense of 2,3-BD. A reduced oxygen supply

increases the 2,3-BD yield but decreases the conversion rate due to a lower cell concentration

(Garg & Jain, 1995; Jansen N. B., 1984).

The effect of relative oxygen availability on the fractional yield of 2,3-BD and other anaerobic

metabolites is shown in Figure 4 at the example of B. polymyxa (Celinska, 2009).

Under aerobic conditions, NADH from glycolysis is regenerated via respiration. In contrast, under

anaerobic conditions NADH is regenerated in fermentation pathways yielding 2,3-BD, acetoin,

lactate, acetate, formate, succinate or ethanol. If aeration is controlled properly, ethanol and

formate synthesis will be widely prevented and glucose can be almost entirely converted to

Figure 4 The effect of relative oxygen availability on fractional product yields in B. polymyxa

11

2,3- BD. Increasing the oxygen supply results in acetic acid production and, if the oxygen supply

overcomes the oxygen demand, only biomass and CO2 will be produced. Moes et al. (1985)

reported an increased acetoin formation at high oxygen supply rates for B. subtilis. Dissolved

oxygen levels above 100 ppb led to acetoin excretion, while 2,3-BD production was induced at

levels below 100 ppb (Celinska, 2009). Yu and Saddler (1982) have reported that the

microorganism K. pneumoniae can metabolize glucose anaerobically, but requires some air to

metabolize xylose. Alam et al (1990) studied the effect of aeration on 2,3-BD production by

B. amyloliquefaciens. They showed that at low aeration rates the 2,3-BD to biomass ratio

increased with high glucose concentrations.

1.5.4. Agitation

Agitation has not only a strong impact on mixing but also on the oxygen supply in fermentation

processes. Therefore, agitation has likewise an effect on 2,3-BD fermentation (Celinska, 2009).

In experiments with B. amyloliquefaciens good results were obtained using stirring rates of

300 rpm the maximum 2,3-BD production was found close to 100 g/L with a 2,3-BD productivity

of 1.02 g/(L.h).

Using a two-stage agitation speed control with 300 rpm in the first 15 hours and thereafter 200 rpm

resulted in higher 2,3-BD concentrations and yields with K. oxytoca than fermentation at one fixed

agitation speed, between 100 and 400 rpm (Ji X.-J. , 2009). Likewise, better results were obtained

with two-stage agitation speed control in fed-batch fermentation with B. licheniformis (Li L., 2013).

For E. aerogenes and K. pneumoniae an agitation speed of 220 rpm was found to be optimal

leading to a 5-fold and 15-fold enhancement of 2,3-BD production from glucose and lactose,

respectively (Barret EL, 1983).

1.5.5. Medium Composition

A culture medium must contain all the essential nutrients that the particular microorganism needs

for growth and maintenance. Nevertheless, for an efficient 2,3-BD synthesis some additives like

vitamins and trace elements have to be supplemented (Garg & Jain, 1995; Ji X.-J., 2011).

Yeast extract (YE), urea, ammonium salts and trace elements are important for protein synthesis

and to improve 2,3-BD yields (Ji X.-J. , 2011). Nilegaonkar et al. (1992) have reported that the

maximum yield of 2,3-BD production was obtained with addition of peptone/beef extract as

medium supplement in fermentations with B. licheniformis on glucose. Laube et al (1984)

investigated the effect of yeast extract on 2,3-BD production from glucose by B. polymyxa and a

level of 1.5% (w/v) was found to be optimal. However, with a level of 0.5% of yeast similar results

12

could be achieved due to the increasing phosphate, iron and manganese concentration added to

the culture medium (Ji X.-J. , 2011).

Prelaminar studies for B. amyloliquefaciens with a culture medium containing soybean meal,

which is rich in protein, showed that 2,3-BD production was improved with addition of 20 g/L

soybean meal achieved concentration of 60.1 g/L. The 2,3-BD productivity was much lower at low

concentrations or in the absence of soybean meal (Yang, 2011). Medium containing glucose and

ammonium phosphate was constructed based on response surface experiments with

K. pneumoniae, resulted in concentrations of 2,3-BD above 90 g/L in fed-batch and 50 g/L in

batch cultivations. K. oxytoca with a culture medium with glucose or xylose with a supplement of

urea, corn steep liquor and mineral salts showed high 2,3-BD production (Ji X.-J., 2011). On the

other hand, a medium with acetate, succinate, pyruvate or propionate supplementation 2,3-BD

production was enhanced using B. polymyxa. Acetate was proven to be the best inducer, leading

to the highest yield and product concentration. The addition of butyrate, valerate, malate, formate

and lactate did not indicate an effect on 2,3-BD production (Celinska, 2009).

1.5.6. Substrate concentration

In most studies of 2,3-BD, the most frequently applied initial sugar concentrations range between

5-10% (Garg & Jain, 1995). 2,3-BD yield and production rate often depended on the particular

raw material used and the initial sugar concentration. Some studies suggested that when the

sugar concentration in the raw material is increased, the level of toxicity also increases resulting

in poor substrate utilization (Jansen N. B., 2005). Therefore, in industrial-scale fermentations

substrates are frequently diluted to lower sugar concentrations (Voloch M, 1985).

2,3-BD fermentation with pure sugars does not contain other inhibitory compounds. E. aerogenes

is a facultative anaerobe that depending on the strain and the microaerophilic conditions can

produce 2,3-BD. Converti et al. (2002) studied increasing initial glucose concentrations from 9.0

to 72 g/L in batch culture. Turnovers of 35% of initial sugar source to 2,3-BD could be observed

whereby higher 2,3-BD yields were obtained with lower sugar concentrations.

The specific growth rate of K. oxytoca was revealed to decrease with increasing initial xylose

concentration which might be due to the decreasing water activity. However, maximum values for

2,3-BD productivity were obtained at initial sugar concentrations of around 100 g/L xylose (Jansen

N. B., 1984). Similar results were reported, a few years later, by the same investigator but with

K. pneumoniae growing on xylose (Jansen N. B., 2005).

Research studies carried out with B. amyloliquefaciens revealed a maximum productivity with

glucose concentration of 120 g/L. and it had been shown that a low sugar concentrations the

13

fermentation proceeded faster. B. polymyxa showed similar results with glucose as substrate

(Laube, 1984). During cultivations performed with B. licheniformis, the best results regarding

2,3-BD productivity were reached with an initial glucose concentration of 20 g/L containing

peptone 1% (w/v) and beef extract 1% (w/v). In addition, the yield corresponded to 94% of the

theoretical yield (Nilegaonkar S., 1992). A newly isolated thermophilic B. licheniformis strain was

tested with different glucose concentration between 64 to 180 g/L. At glucose concentrations

higher than 152 g/L the conversion was inhibited and only a small amount of 2,3-BD was

produced. Consequently, initial glucose concentrations between 64 and 125 g/L were applied in

successive investigations (Li et al, 2013). 180 g/L of initial glucose concentration, 30C, 400 rpm

and an aeration rate of 1.2 L/min were reported as optimal conditions for a successful scale-up

by B. licheniformis DSM 8785 in a 3.5 L-bioreactor (Jurchescu I.-M., 2013).

1.5.7. Water activity

Another important parameter which affects 2,3-BD production is water activity (aw). Water activity

is related to osmotic pressure and inversely correlates with total solute concentration. Strictly

speaking, it depends on the molar concentration and activity coefficient of each solute. A number

of key kinetic and bioenergetics parameters are influenced by the water activity. The most

important are the duration of culture log phase, the maximum specific growth rate, the

thermodynamic efficiency, the maintenance energy coefficient and the biomass yield (Garg &

Jain, 1995). Klebsiella species are not as osmotolerant as other microorganisms and that is the

reason with very high sugar concentrations in 2,3-BD process with Klebsiella species are not

suitable. With a water activity of 0.985 the growth rate of K. pneumoniae was found to be 50%

optimal and became lesser than 10% optimal at water activities below 0.975 (A. Singh, 1995).

1.6. Gas-liquid mass transfer

Anaerobic fermentations take place in the absence of oxygen; in these processes, multiple

compounds can act as oxidizing agents, such as sulfate or nitrate. On the other hand, in aerobic

processes, oxygen is used as the final electron acceptor, emitting carbon dioxide and water

(among other byproducts) as a result. Thus, it is important to ensure an adequate delivery of

oxygen from a gas stream to the culture broth. As consequence, a precise estimation of the

oxygen transfer rate (OTR) at different scales and different operational conditions has a relevant

role for the prediction of the metabolic pathway and a crucial importance for the selection, design

and scale-up of bioreactors. The OTR is influenced by physicochemical parameters of the gas

and liquid as well as operational conditions and geometry of the bioreactor. Likewise, the

presence of cells and their consumption of oxygen lead to an enhanced oxygen transfer since the

dissolved oxygen tension decreases with increasing biomass (Suresh S, 2009). Figure 5 is a

14

schematic view of the different phenomena that occur when oxygen passes from the gas phase

into the fermentation media and diffuses to the microbial cell.

The transport of substrates to cells occurs at a rate considerably higher than the rate of the

metabolic biochemical reactions. However, if mass transfer rate is lower than reaction rate,

transport rate can be the step controlling the overall process rate. Furthermore, the mass transfer

rate may be influenced by the chemical rate of the bioprocess. For instance, oxygen is consumed

by the suspended microorganism, and therefore an enhancement of the OTR takes place. The

OTR is one of the most important factors in 2,3-BD fermentation (Suresh S, 2009). Carbon dioxide

is a respiratory byproduct of aerobic as well as anaerobic processes. Its accumulation in culture

media can have inhibitory effects on microbial growth, enzyme activity and in extreme cases leads

to cell lysis (Chester et al, 1983). Proper removal of CO2 from media is thus of paramount

importance for fermentation processes. CTR is the rate of exchange of CO2 between the liquid

and the gas phases. These rates can be readily calculated from in- and outlet gas stream analysis

(Syu, 2001).

The two film theory of gas absorption is a useful model to describe mass transfer between two

different phases (Suresh S, 2009). This theory states that when two phases are in contact, an

interface is formed, a substance that is being transported from one phase (having a bulk

concentration C1, in mol m-3) to the other phase (with a smaller concentration C2) has to be

transported from the first phase through the interface, and from the interface to the second phase

as Figure 6 shows.

Figure 5 Mass transfer phenomena which occur during the transport of oxygen from the gas phase to the cells in a bioreactor (Bchs, 2014)

15

Figure 6 Graphical representation of the two film theory. A substance is transported from phase 1 to

phase 2, passing through two stagnant films and an interface.

The theory assumes that both phases are perfectly mixed, thus having a homogeneous

concentration of the present substances where a stagnant film of fluid is formed on each side of

the interface. It is assumed then that there is no transfer resistance in the bulk of the fluids and in

the interface, and that all of the resistance is in the transport through the films. In the particular

case of O2, phase 2 is the liquid phase and phase 1 is the gas phase. The mass transfer flow rate

can be expressed by equation (2).

0 = . ( 1) = . (2 ) 0 = . (

) = . ( ) (2)

where J0 is the molar flux of oxygen (mol s-1 m-2) through the gasliquid interface; kG and kL are

the local mass transfer coefficients; pG is the oxygen partial pressure in the gas bubble; and CL,

the dissolved oxygen concentration in the bulk liquid, but the interfacial concentrations are not

directly measurable, KL, global mass transfer coefficient on the liquid side and KG, global mass

transfer coefficient on the liquid side and considering the overall mass transfer coefficient, the

equation can be rewritten. p is the oxygen pressure in equilibrium with liquid phase and C is the

oxygen saturation concentration in the bulk liquid in equilibrium to the bulk gas phase.

The solubility of a gas in equilibrium with a liquid can be calculated by the Henrys law, depending

on the temperature and pressure of the gas/liquid system as equation (3) shows.

= . () (3)

is the gas solubility, is the gas partial pressure and () is the temperature-dependent

Henrys law coefficient for that gas. Henrys law is applicable as long as the concentration of

dissolved gas is small and the temperature and pressure are far from the critical values.

The oxygen mass transfer rate per unit of reactor volume, 2, is obtained by multiplying the

overall flux by the gasliquid interfacial area per unit of liquid volume, a as equation (4) present.

16

2 = 0 = (

) (4)

The determination of volumetric oxygen mass transfer coefficient (kLa) in bioreactors is essential

in order to establish aeration efficiency and to quantify the effects of the operating variables on

the provision of dissolved oxygen. Some developments have been made to improve the oxygen

transfer rate in bioreactors. These improvements comprise the aeration and homogenization

systems, the bioreactor type and its mechanical design as well as the composition of the

fermentation medium (Suresh S, 2009).

The rate at which oxygen is consumed by the microbial culture is called the Oxygen Uptake Rate

(OUR) and, for most cases, its value can be approximated by the OTR due to the limited solubility

of oxygen. Hence, the differential term in equation (5) can be neglected (Royce, 1992).

OUR = OTR d[O2]

dt OTR

OTR = kLa (C CL)

(5)

The oxygen supply is a critical factor affecting 2,3-BD production, productivity and yield, as

previously mentioned. When oxygen is relatively limited, acetic acid can be produced, while when

it is certainly limited, acetoin, 2,3-BD, ethanol and lactic acid are then produced. Without oxygen,

equal molar amounts of 2,3-BD and ethanol are formed (Syu, 2001). The measuring of the

respiratory activity, or respirometry, is a powerful tool for monitoring and controlling industrial

fermentation processes.

The most commonly measured variables in respirometry are the OTR and CTR. The ratio of the

net molar quantity of CO2 evolved by a microorganism (CTR) and the molar quantity of oxygen

accordingly consumed (OTR) is called the Respiratory Quotient (RQ). The RQ provides precious

information about the state of a culture, such as the substrate on which the microbe is growing.

Substrates that are highly reduced, like alkanes, will naturally require more oxygen per carbon

atom to be completely oxidized than highly oxidized substrates (Anderlei, Zang, Papaspyrou, &

Bchs, 2004). Table 3 shows the typical respiratory quotients for several different substrates.

17

Table 3 Typical Respiratory Quotient values for various substrates (Klaus, 1980)

Substrate RQ

Carbohydrate 1.00

Glycerol 0.68

Formate 0.50

Lactate 0.99

Acetate 1.01 0.04

Succinate 1.23

Ethanol 0.67

Furthermore, the effect of volumetric oxygen mass transfer coefficient was studied to understand

the mechanism in different oxygen supply conditions. For 2,3-BD production with P. polymyxa, a

programmed variation of the kLa was employed. The kLa was set at three different levels: 40 h-1

(0-19 h) with 450 rpm; 21 h-1 (19-41 h) with 350 rpm and 8 h-1 (41-55 h) with 350 rpm. In the end

of fermentation, the production of 2,3-BD was 44 g/L with a productivity of 0.79 g/(L.h) (Fages,

1986). Zeng et al. (1990) determined OUR rates and observed that the levels varied according to

the dilution rate (D) in continuous cultivations with Enterobacter aerogenes. Different OUR values

were obtained by changing the impeller speed using a constant aeration rate. As the dilution rate

increased, the yield and product concentration decreased. A suitable control strategy was

presented in the study of Zhang et al. (2010) by combining RQ control with a constant residual

sucrose concentration fed-batch cultivation with S. marcescens. Based on stoichiometric

calculations, the RQ value was set to 1.0-1.5 for cell growth and 1.8-2.0 for 2,3-BD production.

Nevertheless, OTR, kLa, OUR and RQ control is not easy to implement but in a study with a simple

two-stage agitation speed control strategy promising results were already achieved using

K. oxytoca (Ji X.-J., 2009).

18

1.7. Reactor Operation Mode

In an industrial context, fermentation can be broadly defined as the transformation of matter

through the deliberate cultivation of microorganisms. Humanity has made use of fermentation

since ancient times, producing many different commodities that heavily influenced the shape of

todays society (Vasic-Racki & D., 2006) Nowadays, fermentation is employed in a vast array of

applications, such as health-care products (production of antibiotics, vaccines, monoclonal

antibodies and other therapeutic molecules), production of food additives, microbial enzymes

(particularly hydrolytic enzymes), production of industrial platform chemicals and fuel (various

alcohols, solvents, polymers for bioplastics, lipids, organic acids and polysaccharides) and

wastewater treatment and soil bioremediation (Waites, Morgan, & Rockey, 2001).

The effects of reactor operation mode on 2,3-BD formation is important to establish an optimal

process design. Different operation modes were tested including batch, fed-batch, continuous

culture, cell recycle and immobilized cell systems. In order to recover 2,3-BD from the

fermentation broth product concentrations above 80 g/L are required (Ji X.-J., 2011).

Batch cultivations were studied with different microorganisms. During batch cultivation on

glucose, a 2,3-BD concentration of 95.5 g/L was obtained using Klebsiella oxytoca (Ji X.-J. e.,

2009). B. licheniformis DSM 8785 was studied by Jurchescu et al. (2013) in a 3.5 L-bioreactor

and a yield of 0.42 g 2,3-BD / g glucose was obtained.

Using the fed-batch cultivation mode and increasing stirring rates high final 2,3-BD concentrations

could be obtained. Yang et al. (2011) reported for B. amyloliquefaciens a productivity of

2.22 g/(L.h). Yu and Saddler (1983) developed a double-fed batch technique for K. pneumoniae

and increased the sugar content by 20 g/L daily, leading to over 100 g/L 2,3-BD from glucose and

over 80 g/L from xylose.

Using a continuous cultivation mode, 2,3-BD productivity was increased, due to the fact that the

reactor was operated at steady state near maximum reaction rate. Successful enhancement of

2,3-BD productivity was reported by Zeng et al. (1990) and Lee and Maddox (1986).

Ramachandran and Goma (1988) employed a continuous cell recycle system based on

ultrafiltration for efficient 2,3-BD production with K. pneumoniae. Higher 2,3-BD productivities

were obtained compared to batch and continuous fermentation systems. A portion of the

ultrafiltrate was removed from the system to maintain the dilution rate, while another portion was

recycled back to the reactor together with the cells by the use of an ultrafiltration unit. Furthermore,

a cell recycle system with a microfiltration module was developed by Zeng et al. (1991). The

system was employed during cultivation with E. aerogenes, which resulted in a three-fold increase

of 2,3-BD productivity, up to 14.6 g/(L.h).

19

In further studies for 2,3-BD production immobilized cells were employed. K. pneumoniae cells

immobilized in calcium alginate were added to a packed column reactor and efficient 2,3-BD

production from whey permeate was achieved by Lee and Maddox (1986). A summary of the

highest 2,3-BD concentrations, yields and productivities reported until now in literature is given in

Table 4.

Table 4 Comparison of the 2,3-BD production using different microorganisms

Microorganism Substrate Max.

2,3-BD [g/L]

Yield [g/g]*1

Productivity [g/(L.h)]

Methods Reference

Bacillus subtilis

AJ1992 Glucose 2.5 0.38 0.33 Batch Moes et al.

(1985)

Bacillus amyloliquefaciens

B10-127 Glucose 92,3 0.15 0.96 Batch Yang et al.

(2011)

B10-127 Glucose 66.5 ~0.42 2.22 Fed-batch Yang et al.

(2011)

Bacillus licheniformis

ATCC 9800 Glucose 8.7 0.12 0.47 Flasks

Nilegaonkar et al (1992)

DSM 8785 Glucose 72.6*2 0.42 0.86 Flasks Jurchescu

et al. (2013)

DSM 8785 Glucose 78.9 0.46 1.18 Batch Jurchescu

et al. (2013)

Enterobacter aerogenes

DSM 30053 Glucose 110.0 0.49 5.40 Fed-batch

Zeng et al. (1991)

Klebsiella pneumonia

SDM Glucose 150.0 0.43 4.21 Fed-batch

Ma et al. (2009)

Klebsiella oxytoca

ME-UD-3 Glucose 95.5 0.49 1.74 Batch Ji et al. (2009)

*1 2,3-BD yield is given in g/g substrate (glucose); the sugar contained in complex nutrients (e.g. yeast extract) are not considered *2 additional 10 g/L yeast extract and 10 g/L tryptone were used for cultivation

20

2. Theoretical Background

2.1. The RAMOS device

The RAMOS device (Respiratory Activity Monitoring System) was developed by Anderlei (2001)

at the chair of Biochemical Engineering at the RWTH Aachen University. RAMOS allows the

online measurement of the OTR and the CTR of bacterial cultures in shake flasks, bringing

together the simplicity and widespread use of these vessels with a better understanding of the

conditions and phenomena. This device has been used for several applications, like the

determination of oxygen limitation in shake flasks, process development and optimization

screening of microorganisms, optimization of the media, investigation of secondary substrate

limitations and monitoring of pre-cultures for fermentations in stirred reactors (Sven Hansen,

2012).

The initial phases of process development usually comprehend the screening and optimization of

producing strains, cultivation media and conditions. With so many different variables to test the

number of experiments that have to be performed will be very high. To minimize costs and

maximize the amount of information, shaken bioreactors with small volumes are employed. These

are the most widely used fermentation or bacterial culture systems in academic and industrial

research, ranging in volume from microtiter plates to a few hundred milliliter RAMOS flasks. Bchs

(2001) has estimated that 90% of bacterial culture experiments make use of shake flasks at some

point because it is an inexpensive and effective way of reproducibly performing several types of

industrially cell cultivations. Shake flasks are really usefulness in the development and scale-up

of processes, tasks such as drug discovery, elucidation of metabolic pathways, strain

development and optimization(Tibor Anderlei, 2001). Commercial versions of the RAMOS device

are distributed by Adolf Khner AG (Birsfelden, Swiss) and Hitec Zang GmbH (Herzogenrath,

Germany). In Figure 7 the typical in-house built RAMOS device used at the chair of Biochemical

Engineering at the RWTH Aachen University is depicted.

Figure 7 a) RAMOS shake flask; b) RAMOS device at the chair of Biochemical Engineering at the RWTH Aachen University, equipped with 8 shake flasks.

a) b)

21

A simplified scheme depicting the general set-up of the RAMOS device is presented in Figure 8.

In this system, up to eight modified Erlenmeyer flasks can be used. Each of these flasks is

adapted with a gas inlet, a gas outlet, an inoculation or sample port, and a modified top to mount

an electrochemical O2 sensor Figure 7a). The gas flow is adjusted by a thermal mass flow

controller (MFC), it is then distributed by the flasks via an eight way capillary splitter. All eight

capillaries have the same length to ensure equal pressure losses and thus avoiding preferential

flow. The gas in- and out-let are equipped with sterile cotton plugs. Behind the inlet valve there is

also a differential pressure sensor and the O2 sensor is separated from the gaseous phase by a

sterile membrane. The data from the sensors is continuously fed to a data processing unit, and

the OTR and CTR calculations are performed with specific software (Suresh S, 2009).

Figure 8 General set-up of a RAMOS device as introduced by Anderlei et al. (Sven Hansen, 2012)

During a RAMOS measurement, the gas flow is separated into three phases: a low flow rinsing

phase, a stop phase and a high flow rinsing phase. In the stop phase both valves are closed and

the respiration activity of the microorganisms subsequently leads to a decrease of the oxygen

partial pressure in the headspace of the measuring flask from which the computer calculates the

oxygen transfer rate (OTR). After the measuring phase the valves are opened again and air is

flushed into the flask - first with high and then with lower velocity - in such a manner that gas

concentrations inside the flask closely resemble those of a normal Erlenmeyer flask with a cotton

plug as sterile barrier. In this way, results obtained with the device can be transferred to normal

shake flasks (Tibor Anderlei, 2000). The OTR, CTR and RQ are calculated with equation (6), (7)

and (8). Here, VG is the headspace volume, VL is the liquid volume, R is the gas constant, T is the

temperature, pO2 is the oxygen partial pressure in the headspace and pCO2 is the carbon dioxide

partial pressure in the headspace. Hereby, pCO2 is obtained from the total pressure

measurement.

=2

.

(6)

22

=2

.

(7)

=

(8)

In Figure 9 some examples of typical metabolic phenomena that can be determined with a

RAMOS device are shown.

Figure 9 OTR profiles for typical metabolic phenomena determined with a RAMOS device (Tibor Anderlei,

2000)

2.2. Scale-up from shake flasks to bioreactors

Shaken systems are usually employed in one of the first stages of process development and if

wrong decisions regarding strains, media or culture conditions are made, it can be very costly and

could be almost impossible to overcome the negative results on more advanced stages of the

process development. Even if no problems are encountered, and since the growth conditions

during the screening phase can be very different from the conditions in the production phase,

insufficient knowledge about important scale-up parameters can also hide the potential of better

strain and media candidates (Bchs, 2001; Funke, 2010).

The limited volume/size of these fermentation systems can affect negatively the amount and

quality of information that can be obtained. For instance, in the case of the DOT usually

submerged electrodes are use in larger bioreactors. However, the size of the electrodes is not

imperceptible in comparison to the size of the shake flasks, and can affect the inherent

hydrodynamics of the process. In the RAMOS device the hydrodynamics compared to normal

23

shake flasks are not disturbed since the RAMOS flasks are only modified in the neck region and

the measurement takes place in the gas phase.

The prediction of the results to be obtained in an industrial scale, from the data collected in

laboratory or pilot plant scales, requires a careful analysis of the influence of the operational

conditions. As scale-up criterion from shake flasks to stirred tank bioreactors the maximal oxygen

transfer capacity or the kLa-value might be applied. According to Maier, the kLa depends on the

shake flask type and operating parameters. A correlation between mass transfer coefficient kLa

and relevant operating parameters in non-baffled shake flasks is given by equation (9).

= . . . 0

. (9)

where:

- VL is the liquid volume of the medium [mL]

- n is shaking frequency [rpm]

- d0 is the shaking diameter [cm]

- d is the diameter of the shake flask [cm]

- v, w, x, y and z are adimensional constants

The adimensional constants depend on the shake flask size. These values are represented in

Table 5.

Table 5 Values of v, w, x and y with z=0 for the different shake flasks.

Flask V w x y

50 0,000161 -0,87 1,15 0,42

100 0,000429 -0,86 1,02 0,45

250 0,000444 -0,85 1,15 0,38

500 0,000676 -0,81 1,11 0,37

1000 0,000247 -0,77 1.38 0,27 with z=0

To calculate from the kLa on the maximal oxygen transfer capacity (OTRmax, equation (10)) the

oxygen solubility LO2 needs to be known. The oxygen solubility can be determined based on the

concentrations of the components of the fermentation medium according to the literature (Wilhelm

E, 1977; Weisenberger & Schumpe, 1996; Rischbieter E, 1996).

= 2 2, (10)

where pabs is the absolute pressure in the system and 2, is the oxygen fraction in the gas phase.

Shaking bioreactors are equipped with or without baffles and are made of glass or plastic

materials. Baffles provide higher oxygen transfer at lower shaking frequency and the higher

24

hydromechanical stress could be an advantage. General geometry, vessel size, number and size

of baffles, surface properties of the inner wall of the vessel, shaking diameter, filling volume and

shaking conditions are variables to be considered with shake flasks. Due to the specific regular

rotating movement of liquid, a very well-defined gas/liquid mass transfer is obtained in unbaffled

shaken bioreactors compared to other bioreactors. In stirred bioreactor, for instance, the kinetics

and intensity of physical phenomena such as gas separation at the surface, foam generation,

bubble formation and stability of the medium have to be taken into account. All these complicating

phenomena are absolutely absent in the case of an unbaffled shaken bioreactor. The gas-liquid

mass transfer is, this way, much easier to determine and control (Anderlei T., 2001).

However, the determination of OTR, CTR, RQ and kLa can be easier in a larger stirred tank reactor

with the aid of an exhaust gas analysis. For kLa estimation three methods (the dynamic gassing

out method, by mass balance of the inle

Documents

Online-analytical characterization of 2,3-butanediol … · Os resultados em RAMOS foram suficientemente satisfatórios para realizar fermentações num biorreactor de 3L. As análises