Succinic Acid Production

2006:260 CIV

M A S T E R ' S T H E S I S

Simultaneous Fermentation and Crystallization in

Succinic Acid Production

Mario Winkler

Luleå University of Technology MSc Programmes in Engineering

Chemical Engineering

Department of Chemical Engineering and Geosciences Division of Biochemical and Chemical Engineering

2006:260 CIV - ISSN: 1402-1617 - ISRN: LTU-EX--06/260--SE

Simultaneous Fermentation and Crystalization in succinic acid

production

Mario Winkler

Lule˚ a University of Technology

Div. Biochemical and Chemical Process Engineering and

Technical University Bergakademie Freiberg

Dept. of Energy Process Engineering and Chemical Engineering

June 21, 2006

Supervisors:

Prof. Thomas Dimmig, Prof. Kris A. Berglund, Dr. Kuchling, Christian Andersson

A ^BSTRACT

Succinc acid is a high potential candidate as a building block chemical, as an origin for many other chemicals made from renewable resources. It has been successfully produced by fermentation and can be processed into a great variety of products. Therefore, it is an object of interest for the growing biobased industry.

This paper deals with the succinate fermentation of an E. coli -strain, called AFP184 and the influence of the succinate concentration on the productivity in this fermenta- tion. The fermentations are run in two parts; an aerobic growth phase where cells are produced, and an anaerobic production phase in which succinate is produced. It has been investigated if the NH

OH used as a base chemical in the aerobic phase has any effect on the growth characteristics and if an inhibiting effect of the growing succinate concentration could be countered by precipitating the succinate with Ca(OH)

.

It was determined that the growth rate was highest when NH

OH was used as base and further fermentations used NH

OH during the aerobic phase and NH

OH, Ca(OH)

or mixtures thereof during the anaerobic phase. It was found that the best results were obtained if Ca(OH)

was added to a fermentation ran with NH

OH as base. This lead to a high productivity that could be kept up over a total succinate concentration of 50 g/l, which in previous studies had been the maximum concentration achieved.

The fermentations were done at the Division of Biochemical and chemical process en- gineering at Lule˚ a university of technology.

ii

A CKNOWLEDGEMENT

First I want to thank Mrs. Popp for making this internship possible in the first place by arranging the exchange with Lule˚ a university.

I want to thank Prof. Dimmig and Dr. Kuchling at the TU Freiberg for supervising this work.

I want to thank Prof. Kris Berglund for giving me the opportunity to work at the Division of Biochemical and Chemical Process Engineering in Lule˚ a university.

I want to thank my supervisor Christian Andersson for his advices and his help during my work. He always supported and encouraged me a lot to to advance with my report and my experiments. I also want to thank him for being a good friend.

I also want to thank Stephan Herz for being a friend and dialogue partner during the times of my nightshifts and beyond.

iii

Die Wissenschaft, sie ist und bleibt, was einer ab vom andern schreibt.

Eugen Roth, german lyricist

iv

C ONTENTS

Chapter 1: Introduction 2

1.1 Biological fuel and chemicals production . . . . 2

1.1.1 Biotechnology . . . . 2

1.1.2 Biomass . . . . 2

1.1.3 Biorefineries . . . . 3

1.2 Succinic acid . . . . 4

1.2.1 Succinic acid . . . . 4

1.2.2 Downstream processing . . . . 5

1.3 Succinate producing organisms . . . . 5

1.3.1 AFP111 and AFP184 . . . . 6

1.3.2 Problems . . . . 8

Chapter 2: Experimental 9 2.1 Materials and method . . . . 9

2.1.1 Density measurement . . . . 10

2.1.2 Optical density measurements . . . . 10

2.1.3 Colony forming units . . . . 11

2.1.4 Acid and sugar analysis . . . . 11

2.1.5 Calculations . . . . 11

2.2 Fermentations . . . . 12

2.2.1 Fermentation 34 to 36 . . . . 12

2.2.2 Fermentation 37 . . . . 13

2.2.3 Fermentation 39 . . . . 13

2.2.4 Fermentation 41 . . . . 13

2.2.5 Fermentation 42 . . . . 14

2.2.6 Fermentation 43 . . . . 14

2.2.7 Fermentation 44 . . . . 14

2.2.8 Fermentation 61 . . . . 14

Chapter 3: Results and Discussion 15

Chapter 4: Conclusion 26

Chapter 5: Future work 27

Appendix A:Data from fermentations 28

Appendix B:Declaration 36

vi

List of Figures

1.1 Mixed acid fermentation . . . . 6

1.2 Mutated mixed acid fermentation pathway . . . . 7

3.1 Fermentation 34 to 36 . . . . 15

3.2 Fermentation 37 . . . . 16

3.3 Fermentation 39 . . . . 17

3.4 Fermentation 41 . . . . 18

3.5 Fermentation 42 . . . . 19

3.6 Fermentation 43 . . . . 20

3.7 Fermentation 44 . . . . 21

3.8 Fermentation 61 . . . . 22

3.9 Fermentation 61 sugar consumption and CFU . . . . 24 3.10 sugar consumption and CFU for fermentation F42 and fermentation F43 25

vii

List of Tables

1 Nutrients for fermentation media . . . . 10

1 productivity and concentration of succinic acid based on calculated suc- cinic acid amounts . . . . 23

A.1 Fermentation 34 to 36 . . . . 28

A.2 Fermentation 37 . . . . 29

A.3 Fermentation 39 . . . . 30

A.4 Fermentation 41 . . . . 31

A.5 Fermentation 42 . . . . 32

A.6 Fermentation 43 . . . . 33

A.7 Fermentation 44 . . . . 34

A.8 Fermentation 61 . . . . 35

1

C ^HAPTER 1 Introduction

1.1 Biological fuel and chemicals production

1.1.1 Biotechnology

Biotechnology is an engineering science where substances are produced by microorganisms.

Today biotechnology is becoming more and more important because many microorgan- isms can produce substances which today are mainly produced from fossil fuels and in particular oil, of which we inevitably will run out. Therefore, biotechnology is one of the most important alternatives to petroleum based chemicals production and more and more products made from petrochemicals could be replaced with similar products made from renewable resources.

Biotechnology that is used for industrial production, is called white biotechnology. There also exists red, green, grey and blue biotechnology. Red biotechnology deals with medical applications, such as the production of antiobiotics by fungi. Green biotechnology is used in agricultural processes, for example crop improvement through genetic manipulation.

Grey biotechnolgy is used in environmental engeneering. The term blue biotechnology is used to describe the marine and aquatic applications of biotechnology.

This work deals with production of succinic acid from renewable resources by E. coli fermentation and is an example of white biotechnology.

1.1.2 Biomass

As carbon sources, such as petroleum and natural gas, exist in limited supplies, they should be replaced by renewable raw materials: such as biomass. Biomass is defined by The Biomass Advisory Committee (BTAC) of the United States as follows (U.S. congress 2000):”The term biomass means any organic matter that is available on a renewable or

2

1.1. Biological fuel and chemicals production 3 recurring basis (excluding oldgrowth timber), including dedicated energy crops and trees, agricultural food and feed crop residues, aquatic plants, wood and wood residues, animal wastes, and other waste materials.” [5]

Like crude oil, biomass has a complex composition. After separation into the main groups of substances subsequent treatment and processing of those substances lead to a whole palette of products.

Enough waste plant biomass is available in the United States (at least 250 billion kg) each year to supply all of the organic chemicals that can readily be made from (or be re- placed by) biomass [3]. Although this example speaks just for the United States, biomass is available almost everywhere around the world. Biomass can be obtained for example from waste products from the agricultural and forest industries, e.g. molasses, corn stover and sawdust.

1.1.3 Biorefineries

Just like petrochemical-based products, the price of biomass-derived products depends highly on the costs of the production and of the separation, but different to petrochemical- based products low-cost material can be used as raw material [3]. Therefore, it should be a future goal to design biorefineries, which also have the same principle as common petrolchemical-refineries, where several products could be produced in the same plant and thus decreasing the production costs for each product.

There exists no definition of the term biorefinery that is not controversial. One descrip- tion is given by the United States Department of Energy, in its energy, environmental and economics handbook:”A biorefinery is an overall concept of a processing plant where biomass feedstocks are converted and extracted into a spectrum of valuable products, based on the petrochemical refinery.” [12]

But there is more or less agreement about the goal, which is briefly defined as: devel- oped biorefineries, so-called phase III biorefineries, start with a biomass feedstock-mix to produce a multiplicity of products by a technology-mix.

An example of the phase I biorefinery is a dry-milling ethanol plant. It uses grain as a feedstock, has a fixed processing capability and produces a fixed amount of ethanol, feed co-products and carbon dioxide. It has almost no flexibility in processing. Therefore, this type can only be used for comparable purposes.

An example of the phase II biorefinery is the current wet-milling technology. This tech-

nology uses grain-feedstocks, yet has the capability of producing various end-products,

depending on product demand. Such products include starch, high frucose corn syrup,

ethanol, corn oil and corn gluten feed and meal. This type opens numerous possibilities

to connect industrial product lines with existing agricultural production units. Thus, the

integrated production of biodegradable plastics, such as poly-3-hydroxybutyrate, sugar

and ethanol in a conventional sugar plant is an example of a phase II biorefinery.

1.2. Succinic acid 4 A phase III biorefinery is not only able to produce a variety of chemicals, fuels and in- termediates or end-products, but can also use various types of feedstocks and processing methods to produce products for the industrial market. The flexibility of its feedstock use is the factor of first priority for adaptability towards changes in demand and supply for feed, food and industrial commodities.

The main-purpose for biorefineries is that for one process alone it is difficult to achieve a price for the end-product that is economically competitive to the petrochemically pro- duced product. The only advantage a green process has is that the raw materials are mostly waste products in other industries and therefore, cost only very little, like corn steep liquor (CSL), yeast extract (YE) or molasses for example.

Currently, three biorefinery systems are favored in research and development[5]. First, the whole-crop biorefinery, which uses raw materilas such as cereals or maize. Second, the green biorefinery, which uses naturally wet biomass, such as green grass, lucerne, clover, or immature cereal. Third, the lignocellulose feedstock (LCF) biorefinery, which uses naturally dry raw materials such as cellulose-containing biomass and wastes.

1.2 Succinic acid

1.2.1 Succinic acid

Succinic acid was first discovered by Georgius Agricola in 1546 by distilling amber. At room temperature, pure succinic acid is a solid that forms colorless, odorless prisms. It has a melting point of 185 ℃ and a boiling point of 235 ℃. The anion, succinate, is a component of the citric acid cycle,the esters of succinic acid are called dialkyl suc- cinates [1]. It is found to be a high potential candidate for being a building block chemical [14], as an origin for, e.g.: 1,4-butanediol, tetrahydrofuran, n-butyrolactone, adipic acid, n-methylpyrrolidone and linear aliphatic esters [16].

These, and others, are chemicals for industries producing food and pharmaceutical prod- ucts, surfactants and detergents, green solvents, biodegradable plastics and ingredients to stimulate animal and plant growth.

Succinic acid is produced today mainly from butane through maleic anhydride. It’s cur-

rently a low volume chemical, but if it’s possible to invent a green process to produce

succinic acid using biomass instead of petrochemicals and making this process economi-

cally competitive with petrochemical-based succinic acid, fermentation-derived succinate

has the potential to supply over 5.4*10

kg industrial products per year [16].

1.3. Succinate producing organisms 5

1.2.2 Downstream processing

A major problem of biological succinc acid production is the cost for the separation of the product from the fermentation liquid.

Energy consuming separation operations increase the cost for producing chemicals from biomass. In order to keep the cost to a minimum it is necessary to have an effective separation process and to keep the number of unit operations in the separation train as low as possible.

When using microorganisms for production, the products are either found inside the cells (intracellular) or outside (extracellular).

Extracellular products can be processed directly, but for an intracellular product one has first to get the product out of the cells. This is accomplished through destruction of the cell wall and cell membrane after which the intracellular products are released into the fermentation broth. After that the product can be processed like an extracellular product. In the case of succinic acid production the product is excreted out of the cell and can be processed directly.

Succinic acid production by E. coli works best at neutral pH and in order to balance the pH reduction imposed by the excreted acid a base chemical like sodium or ammonium hydroxide must be added. The acid and base will form a salt, which later needs to be further purified. The salt can either dissolve in the fermentation liquid or form a solid precipitate. If the salt dissolves it needs to be separated from the fermentation broth.

This can be done in a number of ways including evaporation, crystallization and reactive distillation. If the salt precipitates, this could greatly help the following separations, since a more or less pure solid material can be separated from the broth.

In order to produce crystals of the free acid a possibility could be to filter the solids out of the media and acidify the solids to get the acidified product that is soluble in water and/or alcohol. This is normally possible with strong acids like hydrochloric acid or sulfuric acid, which would create the acidified product and a salt. For example calciumsuccinate would react with sulfuric acid to succinic acid and calciumsulfate.

C

H

(COO)

Ca + 2H

SO

−→ C

H

(COOH)

+ 2CaSO

(1.1) After the product is dissolved in a solution and separated from the solids of the media follows mostly a combination of distillations, to recycle used chemicals, and cleaning pro- cesses, to clean the solution of other dissolved products.

1.3 Succinate producing organisms

The anion of succinic acid, succinate, is a component of the citric acid cycle, which is a

metabolic pathway that forms part of the break down of carbohydrates, fats and proteins

into carbon dioxide and water in order to generate energy in all living cells that utilize

1.3. Succinate producing organisms 6 oxygen as part of cellular respiration.

E. coli is a facultative anaerobe, which means that it is able to adjust to anaerobic as well as aerobic environments. Under anaerobic conditions E. coli produces a mixture of organic acids and ethanol. This provides another possibility to produce succinate, called the mixed acid fermentation (shown in figure 1.1).

Figure 1.1: Schematic sketch of the mixed acid fermentation pathway in Escherica coli. Only reaction directions to products are shown.

At the moment there are several bacteria known that can produce succinate in a significant amount, e.g. Actinobacillus succinogenes [16],

Anaerobiospirillum succiniciproducens [10] and several E. coli strains [9, 4].

1.3.1 AFP111 and AFP184

The E. coli strain AFP111 originates from a wild type E. coli and has three spontaneous mutations. A mutation in the glucose specific phosphotransferase system, enables the phosphorylation of glucose only with the aid of the enzyme glucokinase [13]. A muta- tion in the pyruvate formate lyase gene prevents the formation of acetyl coenzyme and formate. An alteration in the lactate dehydrogenase reduces lactate production.

These three mutations lead to a higher succinate production, by deactivating the for-

mation of three products (ethanol, formate and lactate), through deactivation of the

1.3. Succinate producing organisms 7 necessary enzyme systems. Phosphoenolpyruvate, which is an intermediate in the mixed acid fermentation to produce succinate, is used to phosphorylate the glucose, but because of the mutation in the glucose specific phosphotransferase system the phosphorylation of glucose can only be done by the enzyme glucokinase, leaving all the phosphoenolpyruvate for the succinate and acetate producing pathways.

The E. coli strain AFP184 originates from the near wild type C 600 strain. It was metabolically engineered to obtain the same three gene mutations as in the AFP111 but taking advantage of the C600 strains xylose fermenting capabilities and strong growth characteristics [15]

A schematic sketch of the mutated mixed acid fermentation pathway is shown in figure 1.2.

Figure 1.2: Schematic sketch of the mutated mixed acid fermentation pathway in Escherica coli strain AFP111 and AFP184. Only reaction directions to products are shown. Grey pathways are ”switched off ”

Experiments with AFP184 show that it is not necessary to run the fermentation in fed-

batch mode. Instead the fermentation can be run in batch mode with a high initial sugar

concentration (approximately 100 g/L) without any increase in by-product formation [8].

1.3. Succinate producing organisms 8

1.3.2 Problems

In fermentations with AFP184, succinate production ceases at an acid concentration

around 50 g/l. In this work we investigated if the acid concentration was responsible for

the production inhibition and if we could counter that by adding Ca(OH)

, precipitating

calciumsuccinate and thus reducing the acid concentration in solution.

C HAPTER 2 Experimental

2.1 Materials and method

The bioreactor used for the fermentations is a 12 l fermenter, empty weight 74,15 kg, from Belach Bioteknik AB, controlled with the software Bio-Phantom 2000. A more detailed description of the system is presented by Andersson [2].

The fermentations were run in dual-phase mode with an aerobic growth phase and an anaerobic production phase. To avoid oxygen limitation during the aerobic growth period the dissolved oxygen (DO) was not allowed to fall below 30% saturation. The DO was regulated with a constant flow of compressed air at a flow rate of 10 l/min and variation in stirrer speed. The initial speed used was 500 rpm and led to a DO of 100%. As DO dropped due to bacterial growth the stirrer speed was increased in intervals to a maximum of 1000 rpm. At higher rotations the cell walls might rupture due to shear stress. Carbon dioxide at a flow rate of 3 l/min was used to achieve anaerobic conditions and push the mixed acid fermentation pathway towards the production of succinate. The fermentation temperature was 37 ℃ and the pressure was 0.1 bar above atmospheric to assure that no unfiltered air could leak into the reactor. The pH-value was allowed to vary between 6.60 and 6.70 with a 2 M H

SO

and a 15% NH

solution as pH controlling agents.

The starting volume of the fermentations was 8 l including an inoculum volume of 0.5 l.

The media used was based on corn steep liquor (CSL) and a minimal addition of inorganic nutrients in the form of salts. The composition was chosen so that the media is low cost and industrially adaptable. Nutrients according to table 1 were mixed together with 2 l of 400 g/l glucose solution and water to a volume of 7.5 l. The added sugar corresponded to an initial concentration of 100 g/l.

9

2.1. Materials and method 10 Substance Weight [g]

Corn Steep Liquor(CSL) 266

K

HPO

11.2

KH

PO

4.8

(NH

)

SO

26.7

MgSO

1.6

Table 1: Nutrients for fermentation media

The media, except the sugar solution, was sterilised in the reactor at 121 ℃ for 20 minutes. The sugar solution was filter sterilised through a 0.22 µm membrane filter. The inoculum (0.5 l) was grown in four Erlenmeyer flasks with Tryptic soy Broth (TSB) on a shake table at 37 ℃ for 16 hours. Before inoculation, the flasks with TSB were autoclaved at 125 ℃ for 15 minutes. A peristaltic pump with a sterile hose and cannula was used to pump the sugar solution and inoculum into the reactor. The fermentations were started after the sugar solution and inoculum had been added. The aerobic phase lasted for 8 hours or until the optical density at 550 nm (OD

) reached a value of 35. Samples with a volume of 15 ml were taken during the growth and fermentation period (the indication of the sample-identification gives the time in hours, starting with T0). The samples were placed in sterile test tubes and stored in a refrigerator for later analysis.

After termination of the fermentation an antibacterial solution was added. The reactor was then emptied, filled with water and sterilised. After cooling the reactor was cleaned.

2.1.1 Density measurement

The weight of an 1,5 mL microtube was measured with a lab precision balance. Then 1 ml sample from the fermenter was added to the microtube and the weight of the full microtube was then divided by 1 mL giving the density in g/mL or kg/L.

2.1.2 Optical density measurements

Optical density gives an approximate value of the cell mass and therefore the cell growth can be estimated. The drawback is that this method measures all solid material includ- ing dead cells. A Genesys 10 UV scanning spectrophotometer from Thermo Spectronic was used to determine the optical density at 550 nm. The samples were diluted with phosphate buffered saline (PBS) solution at a pH of 7.4 so that the measured OD did not exceed 0.5 in absorbance, in order to keep the OD within the linear response range of the spectrophotometer.

1

Any change in these procedures will be named specifically for each fermentation.

2.1. Materials and method 11

2.1.3 Colony forming units

In order to determine the actual number of living cells in the fermenter, an aliquot of the 15 ml samples was diluted with sterile TSB solution. 100 µl of the diluted sample was used as inoculum for agar plates which were incubated at 37 ℃ for 16 hours. The colonies formed were then counted and with the dilution ratio the number of living cells in the original sample could be established.

2.1.4 Acid and sugar analysis

The concentration of acids and sugars were measured with a High Performance Liquid Chromatograph system (HPLC). The software TotalChrome from Perkin Elmer was used for controlling the system and for integration of the peak areas of the chromatograms.

The areas were then compared with standard curves for the respective substance. The HPLC system is described in more detail by Andersson [2].

A Perkin Elmer series 200 UV/VIS detector was used together with a C-18 column from Waters for the acid analysis. The mobile phase was 98 % 50 mM KH

PO

buffer and 2 % acetonitrile at a pH of 2.5. The flow rate through the column was 0.35 ml/min. The detector used a wavelength of 210 nm and an analysis time of 20 minutes.

The sugars were analysed with a series 200 refractive index detector from Perkin Elmer together with a carbohydrate analysis column, Aminex HPX-87P, from Biorad. Water at a flow rate of 0.6 ml/min was used as mobile phase. The column was kept at 85 ℃ and the analysis time was 20 minutes.

From each of the samples taken during the fermentations 2 ml were transferred to new test tubes and centrifuged for 10 minutes at 10 000 rpm and 4 ℃. The supernatant was seperated by pipetting. The diluent used for acid analysis was the running mobile phase and for sugar analysis water.

2.1.5 Calculations

Several fermentations were done to investigate the effect of Ca(OH)

usage as pH con- trolling agent instead of NH

OH. Therefore several parameters were changed for each fermentation. Which parameter that was varied depended on the results of previous fer- mentations.

The total amount of sugar that was consumed is calculated as follows

m

= m

− (V

∗ c

) (2.1)

V

= m

− m

ρ (2.2)

m

= m

− ( m

− m

ρ ∗ c

) (2.3)

2.2. Fermentations 12 With m

being the amount of consumed sugar in g, m

the amount of sugar added in g, m

the weight of the reactor at the end of the fermentation in kg, m

the weight of the empty reactor in, which is 74.15 kg, V

the volume of the reactor at the end of the fermentation in L, ρ the density of the fermentation media at the end of the fermentation in kg/L and c

the concentration of sugar at the end of the fermentation g/L.

In the fermentations where sugar was added during fermentation the equation has to be complemented:

m

= m

− ( m

− m

ρ

∗ c

) − ( m

− m

ρ

∗ c

) (2.4) With m

being the total amount of the added sugar in g, m

the reactor weight at the last measurement before sugar addition in kg, ρ

the density of the media at the last measurement before sugar addition in kg/L, c

the sugar concentration at the last measurement before sugar addition, m

the weight of the reactor at the end of fermen- tation in kg, ρ

the density of the media at the end of the fermentation in kg/L and c

the sugar concentration at the end of the fermentation.

Based on the yield achieved by Lennartsson in similar fermentations [8], which was 0.88 g succinic acid per g glucose, the calculated theoretical succinic acid production in the fermentations with Ca(OH)

according to the amount of sugar that was consumed in the fermentation without the sugar that was consumed in the aerobic phase.

m

= (m

− m

) ∗ 0.88g/g (2.5) m

= 800g − ( m

− m

ρ

∗ c

) (2.6)

With m

being the amount of succinic acid produced during anaerobic phase, m

the amount of sugar consumed during aerobic phase (at T8), ρ

, c

and m

the density, the sugar concentration and the weight of the reactor after the aerobic phase (at T8).

For fermentations with NH

OH used as base the amount of produced succinic acid can be calculated as follows:

m

= ( m

− m

ρ ∗ c

) (2.7)

With c

being the concentration of succinic acid.

2.2 Fermentations

2.2.1 Fermentation 34 to 36

These fermentations were only run in aerobic phase with Ca(OH)

as base and different

amounts of (NH

)

SO

in the media to get a better idea about the influence of ammonia

on the growth rate.

2.2. Fermentations 13

2.2.2 Fermentation 37

Changes made for this fermentation:

• succinic acid solution added for raising the acid concentration up to 50 g/l after T8 The succinic acid solution was neutralized with ammonium hydroxide.

By adding succinic acid solution after the growth phase we tried to exclude the possibility that the absence of any other substrate is the reason for the productivity loss.

If the AFP184 would have produced around 50 g/l of succinic acid additionally after 18 hours of production phase we would have known that the bacteria cease to produce because they run out of another substrate and that the succinic acid concentration is not responsible for the problem.

2.2.3 Fermentation 39

Changes for this fermentation:

• 250 g/l Ca(OH)

-solution used as base during anaerobic phase

• 1 l of an 400 g/l glucose solution added after T24

We did not use Ca(OH)

as base during the growth phase, because previous fermen- tations have shown that we do not have the same growth rate with that base due to nitrogen limitation. We would also not have the possibility to measure the optical den- sity to determine approximately if we reached the necessary cell density since the calcium hydroxide is added as a slurry and would interfere with the optical density measurement, causing an increase.

To check if the production can kept on for longer than 18 hours, we had to add more sugar after 18 hours of production phase, because at that point nearly all of the sugar has been metabolised.

2.2.4 Fermentation 41

Changes for this fermentation:

• 250 g/l Ca(OH)

-solution used as base during anaerobic phase

• 1 l of an 400 g/l glucose solution added at T17

This fermentation is a replica of fermentation 39, but this time we also took samples

between 8 and 24 hours to see how the bacterias behave during the production phase. In

this way we would obtain a better understanding of what is responsible for the production

loss and also when it starts to occur.

2.2. Fermentations 14

2.2.5 Fermentation 42

This fermentation was performed without any changes in the receipe described above but we have taken samples in the production phase so that we could compare the ”original”

fermentation with the fermentations run with Ca(OH)

.

2.2.6 Fermentation 43

Changes for this fermentation:

• at T10 500 ml of 250g/l Ca(OH)

-solution added and after T16 additional 20 g of the solution was added every hour

• at T15 1 l of a 400 g/l glucose-solution was added

In this fermentation we used NH

OH as base also during the production phase but we added Ca(OH)

every hour so that the concentration of succinate was kept low enough to not exceed 50 g/l.

2.2.7 Fermentation 44

Changes for this fermentation:

• 250 g/l Ca(OH)

-solution used as base during anaerobic phase

• 2 l of 400 g/l glucose-solution added between T16 and T18

• between T22 and T24 switched to aerobic phase by airating with pressurized air at 10 l/h, NH

OH used as base

• at T24 switched back to anaerobic conditions and Ca(OH)

In this fermentation we tried to add another growth phase into the fermentation by switching over to aerobic conditions. If the bacterias would divide, we would have more viable cells in the media and thus giving us a high producitivity again.

2.2.8 Fermentation 61

Changes for this fermentation:

• 250 g/l Ca(OH)

-solution used as base during anerobic phase

• 1 l of 400 g/l glucose solution added at T22

• NH

OH-solution added randomly

In this fermentation we tried approximately the same as in fermentation 43, but here we

mainly used Ca(OH)

as base and added NH

OH randomly to adjust pH and to supply

ammonia.

C HAPTER 3 Results and Discussion

Fermentation 34 to 36

As seen in fig.3.1 an increase in the ammonium sulfate concentration increases the cell density. The increase from 40 to 50 gram ammonium sulfate did not seem to have a major effect, which indicates that a larger additon might be necessary to further increase the cell density.

Figure 3.1: OD550 vs. time for different initial amounts of ammoniumsulfate

15

16

Fermentation 37

Fig.3.2 shows that after 18 hours of production phase there is still enough sugar left for production but the viability of the cells already decreases.

Figure 3.2: Fermentation 37

17

Fermentation 39

Fig.3.3 shows that Ca(OH)

and the added sugar after 18 hours of production phase had no positive effect on the viability of the cells.

Figure 3.3: Fermentation 39

18

Fermentation 41

As seen in fig.3.4 the viaibility of the cells already starts to decrease at the beginning of the production phase.

Figure 3.4: Fermentation 41

19

Fermentation 42

As seen in fig.3.5 the viability of the cells keeps almost constant in the production phase in a fermentation only run with NH

OH. It starts to decrease quickly after 12 hours of production phase.

Figure 3.5: Fermentation 42

20

Fermentation 43

As seen in fig.3.6 with the addition of Ca(OH)

the viability was kept constant also after 18 hours of production phase.

Figure 3.6: Fermentation 43

21

Fermentation 44

Fig.3.7 shows that aerating the system had no effect on the viability of the cells.

Figure 3.7: Fermentation 44

22

Fermentation 61

Fig.3.8 shows that the addition of NH

OH had a positive effect on the viability of the cells. The viability did not decrease that quickly as in previous fermentations.

Figure 3.8: Fermentation 61

23

Discussion

The results for each fermentation are shown in appendix A.

Fermentation 37 and preliminary shake flask experiments made at the division (results not shown here) showed that AFP184 is not able to keep up a high productivity in a medium with a succinic acid concentration of 50 g/l. If succinic acid would not inhibit the productivity the end concentration of succinic acid would have to be around 80 or 90 g/l in fermentation 37, but it was only around 60 g/l, leading to the conclusion that AFP184 is only able to produce succinic acid up to a concentration of approximately 50 g/l. The major part of the additional 10 g/l obtained were most likely produced dur- ing the addition of the acid solution. In order to counter this problem we tried to lower the acid concentration by adding Ca(OH)

that would lead succinic acid to precipitate as calciumsuccinate and thus reducing the concentration of the free acid.

They have all shown a decreased productivity (see table 1).

In a fermentation with NH

OH as base the sugar consumption and the CFU-values stay steady until around 14 to 16 hours in production phase, after that the viability of the culture decreases. When using Ca(OH)

as base during the production phase, the bacteria start to die immediately and thus logically decreasing the sugar consumption and productivity with it (see figure 3.9).

Based on the calculated amounts of succinic acid the corresponding average productiv- ity was calculated as follows:

P = m

t

(3.1)

With P being the average productivity, m

the calculated amount of succinic acid and t

the time in hours of the anaerobic phase.

Fermentation calc. amount of T

productivity V corresponding succinic acid [g] [hours] [g/h] [l] concentration [g/l]

39 752,33 40 18,81 12,7 59,26

41 497,44 24 20,73 14,24 34,94

43 707,69 30 23,59 11,31 62,55

44 366,15 32 11,44 13,27 27,58

61 647,3 37 12,89 12,46 38,27

42 402,94 16 25,18 10,37 38,84

Table 1: productivity and concentration of succinic acid based on calculated succinic acid

amounts

24

Figure 3.9: F61 sugar consumption and CFU. The sugar consumption at 24 hours was not taken in because it would have been negative, due to the addition of sugar after 22 hours.

Based on Lennartsson [8], the amount of succinic acid produced after 24 hours in a fermentation with NH

OH as base and 100 g/l glucose as starting sugar concentration is around 440 to 550 g, thus yielding a productivity of approximately 25 to 26 g/h. The only fermentation which was able to keep that productivity over 30 hours of production phase was fermentation F43 with 23,59 g/h. The calculations are based on the assumption that the same yield of 0,88 g

/g

is achieved. In that fermentation NH

OH was used as base and Ca(OH)

was added every hour in a small amount.

Comparing the CFU and the sugar consumption for a normal NH

OH-fermentation (F42) and fermentation F43, it seems that when Ca(OH)

is added in small amounts every hour it is giving us the expected upkeep of productivity (see figure 3.10).

Although fermentation F39 results in a low productivity, the calculated end concentra- tion of succinic acid was above 50 g/l and thus also indicating that Ca(OH)

is able to let the bacteria produce succinic acid also over a total concentration of 50 g/l.

Looking at the CFU and the sugar consumption when working with Ca(OH)

, we can

say that either the Ca(OH)

is an inhibitor for the succinic acid production or the lack

of NH

OH is responsible for the lower productivity. Because these are the only two

parameters that were changed. At the beginning of our experiments we tried to run a

growth phase with Ca(OH)

instead of NH

OH as base and the bacteria did not grow to

the same cell densitiy achieved with NH

OH as base. So Ca(OH)

does have a negative

25

Figure 3.10: sugar consumption and CFU for fermentation F42 and fermentation F43

effect on the growth rate of AFP184 compared to NH

OH. From fermentation 34, 35 and 36 it can be seen that the system was nitrogen limited and large amounts of ammonium sulfate had to be added in order to reach the same cell density as a NH

OH neutralized fermentation. Therefore, it was decided to use Ca(OH)

only in the production phase and run the aerobic phase with NH

OH as base.

In fermentation 44 we tried to aerate the system with 10 l/h to achieve another growth phase to increase the CFU but it had no effect. The CFU did neither decrease nor in- crease. So either two hours were just not enough time to let the bacteria adapt to the aerobic conditions and let them begin to grow again or the concentration of succinic acid and base in the media were inhibiting them.

Another problem would occur if we would let the bacteria adapt to aerobe conditions

again. Due to their metabolism, the bacteria might start to metabolise the succinic acid,

because it is a intermediate in the citric acid cycle, thus decreasing the yield of succinic

acid at the end of the fermentation.

C HAPTER 4 Conclusion

Together with preliminary experiments the results have shown that a succinic acid con- centration in the media of approximatley 50 g/l or more inhibit further succinate pro- duction.

NH

OH should be used as base in the aerobic growth phase in order to obtain the desired growth rate and cell density.

It is possible to keep the productivity over a longer time by small additions of Ca(OH)

, this was proved by the higher sugar consumption in fermentation F43 compared with the sugar consumption in fermentation F42.

An inserted 2 hour aerobic phase during the anaerobic production phase had no effect on the cell viability or productivity.

26

C HAPTER 5 Future work

Further investigation should be done to determine if there is an optimum in the Ca(OH)

to NH

OH ratio when used in a fermentation. This also would show whether Ca(OH)

is inhibitory or toxic to the AFP184 at a certain concentration.

For fermentations that are run with Ca(OH)

there have to be done some changes in the downstreaming process, because calciumsuccinate does not dissolve in methanol that easy. So some kind of pre-treatment has to be done before the succinate salt can recov- ered and purified into free succinic acid crystals. If this is not possible then the whole downstreaming process has to be changed.

Studies with different bases and different base combinations could be done to improve downstream processing properties of the broth and investigate how AFP184 responds to other base chemicals. This could be of importance when incorporating succinic acid production in a biorefinery that has access to cheap base chemicals other than ammonium hydroxide.

27

A PPENDIX A Data from fermentations

Fermentation 34 to 36

Fermentation 34 Fermentation 35 Fermentation 36

Time [hours] OD550 OD550 OD550

0 1,61 1,91 1,64

2 3,47 3,44 2,77

4 8,76 9,44 9,26

6 18 16,64 18,88

8 19,92 28,64 29,52

(NH4)2SO4 amount 26,7g 40g 50g

Table A.1: Fermentation 34 to 36

28

29

Fermentation 37

OD

ρ reactor weight CFU c

c

[kg/l] [kg] [10

/ml] [g/l] [g/l]

T0 1,56 1,028 81,86 0,22 100,3

T2 2,23 1,029 81,85 2,23 103,3

T4 10,04 1,021 81,84 6,5 95,8

T6 19,36 1,028 81,87 33 86,4

T8 43,44 1,013 81,92 52 72,6

T24 35,44 0,999 84,23 33 44,4 61

m

[g] 563,47

m

[g] 614,88

Table A.2: Fermentation 37

30

Fermentation 39

OD

ρ reactor weight CFU c

[kg/l] [kg] [10

/ml] [g/l]

T0 1,38 1,04 82,01 0,14 96,9

T2 2,40 1,02 82,00 1,65 94,6

T4 8,24 1,035 82,00 8 92,7

T6 19,52 1,03 82,02 22 89,4

T8 38,88 1,00 82,03 54 71,8

T24 1,05 85,00 22 6

T25 1,06 86,58 36,7

T26 1,07 87,30 35,3

T28 1,07 87,20 34,6

T30 1,06 87,19 9 31

T48 1,06 87,10 23,3

m

[g] 1100,23

m

[g] 752,33

Table A.3: Fermentation 39

The OD

value after T24 was not measured due to precipitation of crystals.

The concentrations of succinic acid were not measured because HPLC analyzes dissolved

components. The precipitated crystals could neither be dissolved in water or alcohol nor

in sodium hydroxide or hydrochloric acid solutions.

31

Fermentation 41

OD

ρ reactor weight CFU c

c

[kg/l] [kg] [10

/ml] [g/l] [g/l]

T0 1,48 1,02 81,92 1,25 93,8 0,5

T2 2,97 1,04 81,92 3,91 97,2

T4 8,28 1,02 81,92 8,6 90,4

T6 26,8 1,02 82,01 21 85,5

T8 38 0,92 82,02 64 67,5 1,14

T10 1,015 83,08 50 49,3 2,4

T12 1,03 83,35 24 39,3 12,1

T14 1,02 83,59 38 27,4

T16 1,02 83,88 25 18,8 24,2

T18 1,05 86,12 20 48,5 25,3

T20 1,06 86,62 12 42,9

T22 1,06 87,00 15 37,2

T24 1,07 87,35 11 34,6 33,9

T26 1,07 87,69 5 31,3

T28 1,06 87,99 7 29,3

T30 1,07 88,35 3 27,9

T32 1,07 88,67 3 26,2 27,4

m

[g] 844,46

m

[g] 497,44

Table A.4: Fermentation 41

32

Fermentation 42

OD

ρ reactor weight CFU c

[kg/l] [kg] [10

/ml] [g/l]

T0 1,34 1,02 82,02 0,35 94,6

T8 36,72 1,01 82,06 39 66,6

T10 38,88 1 82,39 56 44,2

T12 44,32 1,01 82,59 61 34,3

T14 39,2 1,02 82,76 50 25,6

T16 40,08 1,01 82,94 52 15,8

T18 42 1,02 83,03 51 11,7

T20 38,4 1,02 83,06 54 8,5

T22 38,8 1 83,07 44 7,3

T24 45,68 1,01 83,07 29 6,7

m

[g] 741,41

m

[g] 402,94

Table A.5: Fermentation 42

33

Fermentation 43

OD

ρ reactor weight CFU c

c

[kg/l] [kg] [10

/ml] [g/l] [g/l]

T0 1,77 1,03 81,84 0,61 104,1

T8 37,2 1,01 81,90 42 76,2 0,87

T10 36,8 1,00 82,16 70 57,3 12,9

T12 1,02 82,93 48 44,4 12,8

T14 1,03 83,08 68 31,3 12,0

T16 1,05 84,48 68 60,6 11,6

T18 1,03 84,73 42 51,1 11,6

T20 1,01 84,94 33 40,7 11,0

T22 1,00 85,10 51 35,9 10,8

T24 1,03 85,26 36 32,1 10,9

T38 1,03 85,69 34 15,6 11,9

m

[g] 1025,22

m

[g] 707,69

Table A.6: Fermentation 43

34

Fermentation 44

OD

ρ reactor weight CFU c

c

[kg/l] [kg] [10

/ml] [g/l] [g/l]

T0 1,76 1,03 82,30 0,43 93,9 0,88

T8 38,48 1,02 82,29 76 67 1,32

T10 1,02 82,60 64 52,2 11,8

T12 1,02 82,83 60 40 10,9

T14 1,03 83,00 17 28,2 11,2

T16 1,03 83,13 14 19,8 11,2

T18 1,04 85,60 38 77,1 11,1

T20 1,03 85,75 37 72,1 11,3

T22 1,04 85,86 47 65 10,9

T24 1,05 86,00 47 59,1 14,2

T34 1,05 87,18 14 52,3 13,7

T40 1,06 87,69 14 40,6 12,9

m

[g] 681,39

m

[g] 366,15

Table A.7: Fermentation 44

Altogether 1,6 l of 250 g/l Ca(OH)

-solution was added.

35

Fermentation 61

OD

ρ reactor weight CFU c

c

[kg/l] [kg] [10

/ml] [g/l] [g/l]

T0 1,46 1,03 82,08 0,37 100,1

T8 36,48 1,01 82,11 45 69,6

T10 1,02 82,39 68 57,1

T12 1,02 82,64 27 48,1

T14 1,04 82,74 23 40,7

T16 1,03 82,94 28 35

T18 1,02 83,15 12 29,4

T20 1,05 83,33 24 24,5

T22 1,04 83,54 25 20,7

T24 1,03 84,85 15 53,4 16,9

T34 1,03 85,04 7,8 45,2

T40 1,04 85,79 6,8 38,7

T45 1,04 86,41 3,6 34 18,7

m

[g] 798,91

m

[g] 476,92

Table A.8: Fermentation 61

A PPENDIX B Declaration

Hereby I, Mario Winkler, certify, that this thesis is the result of my own work and in- vestigation, except where stated otherwise. Other sources are acknowledged by endnotes giving explicit references.

36

References

[1] http://en.wikipedia.org/wiki/Succinic_acid; 17.01.2006

[2] Andersson C.: Preliminary Investigation of E. coli Fermentation Using Five and Six Carbon Sugar Mixtures. 2004, Lule˚ a university of technology:

Lule˚ a

[3] Dale B.E.: ’Greening’ the chemical industry: research and development priorities for biobased industrial products; Journal of Chemical Technol- ogy and Biotechnology 78 (2003): p. 1093-1103

[4] Donnelly M.I., Sanville-Millard C.Y., Nghiem N.P.: Method to produce succinic acid from raw hydrolysates; Patent No.: US 20,030,017,559 Pub- lication Date: 23.01.2003

[5] Kamm B., Kamm M.: Principles of biorefineries; Appl. Microbiological Biotechnology (2004) 64: p.137-145

[6] Keseler I.M., et.al.: EcoCyc: a comprehensive database resource for Es- chericia coli. Nucleic Acids Research. 2005. 33: Database issue D334-D337 [7] Leib M., Carmo J. Pereira, John Villadsen: Bioreactors: a chemical engi- neering perspective; Chemical engineering science 56 (2001) p.5485-5497 [8] Lennartsson A.: Production of Succinic Acid by E.coli from Mixtures of

Glucose and Fructose. Thesis work 2005. Lule˚ a University of Technology, Div. Biochemical and Chemical Processengineering

[9] Mills G.L.: Bioreactor Succinic Acid Production. 2002

[10] Pyung Cheon Lee, Woo Gi Lee, Sunhoon Kwon, Sang Yup Lee and Ho Nam Chang: Succinic acid production by Anaerobiospirillum succiniciproducens: effects of the H2/CO2 supply and glucose concen- tration; Enzyme and Microbial Technology, 1999. 24: p.549-554

[11] Tchieu J.H., Norris V., Edward J.S., Saier M.H.: The complete phosphotransferase system in Escheria coli ; Journal of Molecular Micro- biology and Biotechnology. 2001. 3(3): p. 329-346

37

38

[12] U.S. Department of Energy: Energy, environmental and economics hand- book; 1997

[13] Vemuri G.N., Eitmann M.A., Altmann E.: Effects of growth mode and pyruvate carboxylase on succinic acid production by metabolically engi- neered strains of Escheria coli ; Appl. and Environmental Microbiology.

2002. 68(4): p. 1715-1727

[14] T.Werpy, G. Petersen, et al.: Top value added chemicals from biomass, Volume I: Results of screening for potential candidates from sugars and synthesis gas, Publication, available at: http://www.osti.gov/bridge; Au- gust 2004

[15] Sanjay Yedur, Kris A. Berglund, Dilum D. Dunuwila: Succinic acid pro- duction and purification; Patent No.: US 6,265,190 Publication Date:

24.07.2001

[16] J.G. Zeikus, M.K. Jain, P. Elankovan: Biotechnology of succinic acid pro-

duction and markets for derived industrial products; Applied Microbiology

and Biotechnology, 1999. 51: p.545-552