Characteristics and Techniques of Fermentation Systems PDF

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/275953938 Characteristics and Techniques of Fermentation Systems Chapter · January 2010 CITATIONS...

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/275953938 Characteristics and Techniques of Fermentation Systems Chapter · January 2010 CITATIONS READS 4 33,890 2 authors: Marcel Gutiérrez-Correa Gretty Villena Universidad Nacional Agraria La Molina Universidad Nacional Agraria La Molina 48 PUBLICATIONS 1,291 CITATIONS 49 PUBLICATIONS 577 CITATIONS SEE PROFILE SEE PROFILE All content following this page was uploaded by Marcel Gutiérrez-Correa on 07 May 2015. The user has requested enhancement of the downloaded file.  Characteristics and Techniques of Fermentation Systems 183 7 Characteristics and Techniques of Fermentation Systems Marcel Gutiérrez-Correa and Gretty K. Villena 1. INTRODUCTION Biotechnology is a word that began to be used early in the past century but whose activities are as old as social human activities. Originally, the concept of biotechnology was circumscribed to the field of biochemical engineering, especially to the area of industrial microbiology and enzyme technology. However, the term has acquired a broader meaning and sometimes the actual conceptual borders are not clear enough to distinguish it from other classical technologies. The impressive advances attained in molecular biology, and more recently in the so-called “omics” fields, help biotechnology to include an expander range of activities in the chemical-biological sector. Although the definition given by Bu’Lock (1987) as the use of simple biological agents such as dead or living cells, enzymes and cell components represents a narrow view of biotechnology, combined with that given by the European Federation of Biotechnology (Knorr 1987) as the integrated use of biochemistry, microbiology and engineering for the technological applications of microbial, tissue culture and their parts, may be used for a more rational definition. Thus, biotechnology is the application to engineering of biological processes carried out by cells of any type, enzymes, cell parts or cell components to produce goods and services (Gutiérrez-Correa 2008a). Basic knowledge from molecular biology and other biological fields, chemistry, physics, informatics and engineering sciences are the support for the development of modern biotechnologies. Biotechnological tools developed from the basic sciences such as recombinant DNA technology, metabolic engineering and directed evolution as well as process development, bioprocess modeling, scale-up rationale and bioreactor design are of paramount importance in any field of application. The production of foods by the processing of raw materials from plant or animal origin dates from more than two million years ago and reached a major milestone with the control of fire. The first form of biotechnology that can be traced back to ancient times is the production of fermented foods such as wine, bread or cheese. The combined activities of heating of foods and In “Comprehensive Food Fermentation and Biotechnology” (A. Pandey, C.R. Soccol, C. Larroche, E. Gnansounou, C.-G. Dussap, eds.), Volume I, Chapter 7, p.183-227. Asiatech Publisher, Inc. New Delhi. 184 Food Fermentation Biotechnology the application of fermentation resulted in significant improvements in the safety and quality of foods (de Vos 1999). These properties, together with improving the food production to satisfy the growing and ageing global population, continue to be major areas in food biotechnology. Food biotechnology, apart from the direct use of plants and animals, is mainly concerned with the application of (live) food-grade microorganisms in industrial processes as well as the introduction of functionality in food components (Kuipers 1999; Vaughan 1999). In this century, bioprocesses are expected to replace the chemical industrial processes since they are environment-friendly based on natural activities carried out by living things, thus giving sustainability to the industrial sector (Gavrilescu & Chisti 2005). Also, a bio-based economy is steadily replacing the old oil-based economy (Gutiérrez-Correa 2008b, 2009). In this bioeconomy, new production systems are being developed so that the biomass and genetic resources could be the sources of energy and (bio)chemical products, including food ingredients (Gutiérrez-Correa 2008b). New production systems are integrated production systems (IPS) with the aim of producing foods, energy and industrial compounds with a maximal energy use and minimal environmental damage, thus sustainability is actually possible (Gutiérrez-Correa 2009). As shown in Fig. 7.1, IPS depends on two well-defined production systems: 1) the agro-refinery for the production of crops and biomass under sustainable agronomical practices but with a strong input of modern biotechnologies such as high-yielding genetic modified (GM) varieties, biocides and biofertilizers, thus contributing to diminish any environmental damage and, as required, to regenerate the soil fertility and health; 2) the biorefinery which is the central production model for the industrial activities in bioeconomy. It is an industrial concept based on petrochemical refineries that uses biomass as raw matter and biocatalysts (cells and enzymes) in combination with chemical and thermochemical processes for the production of a broad range of products that would eventually replace those produced by from oil (Kamm & Kamm 2004; Kamm et al 2000). In this sense, the biorefinery is analogous to the petroleum refinery in that in it the biomass is “cracked” into separated components and each component is converted to a separately marketed product. To be efficient, the biorefineries should operate with two fundamental biological systems: cell factories and enzyme systems (Gutiérrez-Correa 2009). Cell factories are genetically designed microorganisms (or other type of cells) by metabolic engineering to utilize and convert several substrates into (bio)chemical products (Das et al 2007; Gutiérrez-Correa 2008b; Hugenholtz & Kleerebezem 1999; Klein-Marcuschamer et al 2007; Raab et al 2005; Stafford & Stephanopoulus 2001; Yakandawala et al 2008). On the other hand, the enzyme systems are enzyme complexes modified by directed evolution to be adapted to the conditions required by certain bioprocesses for a very efficient activity (Bornscheuer 2005; Chatterjee & Yuan 2006; Eijsink et al 2005; Hibbert & Dalby 2005; Turner 2003). Food biotechnology will also be improved with this new technologies which will be applied both to the production of classical food products and to develop new ones. In any case, cell cultivation techniques have to be used either for the production of cell biomass and metabolites or for the in situ modification of food raw materials by microbial fermentation. In the context of industrial  Characteristics and Techniques of Fermentation Systems 185 Maximum energy use Modern Biotechnology Good Agronomic Practices Biofuels & Bioproducts (No-till farming, Biological N2 fixation) Biorefinery Plant Agro-refinery Biomass Foods Crops Enzyme Systems Recycled Cell Factories nutrients Soil Pesticides (minimize) Fertilizers Minimize nutrient losses (optimize) Minimun environmental damage Fig. 7.1. Integrated production systems in bioeconomy aims to produce foods, energy and industrial compounds with a maximal energy use and minimal environmental damage (modified from Gutiérrez-Correa 2009). biotechnology, the term “fermentation” refers to the growth of large quantities of cells under anaerobic or aerobic conditions within a vessel, called a fermenter or bioreactor. This chapter aims to describe the fermentation techniques that should be used to accomplish the production of microbial biomass, enzymes, metabolites and fermented food products. 2. THE FERMENTATION SYSTEM As stated above, many food biotech products are prepared with either microbial biomass – alive or unviable – or microbial enzymes and metabolites. The production of these microbial ingredients follows the general biotechnological process as depicted in Fig. 7.2. Microorganisms are the source of the biological processes that may derive in a market product (or service). However, native microorganisms are usually not suitable for direct use in the biotechnological process due to the parsimony which is a characteristic of all living things. Thus, industrial microorganisms are subjected to several types of genetic manipulations to increase the yields and productivities (Queener & Lively 1986; Stanbury & Whitaker 1984). Fermenters or bioreactors are the central part in the biotech process since they provide the link between the raw materials and final products by conducting the fermentation (Cooney 1983). In this chapter we use fermenter and bioreactor indistinctly to describe a vessel in which a fermentation process is carried out with the goal of minimizing the cost of producing a product or service; however, the term bioreactor is also used in a wider sense to comprise any ambient – like a lake or a pond for the cultivation of microalgae – or a transgenic plant or animal species for the production of a valuable therapeutic protein. 186 Food Fermentation Biotechnology Microorganism (Biological process) Raw matter Product Bioreactor (carbon, energy & (biomass, enzymes (fermenter) nutrient sources) or metabolites) Process control Fig. 7.2. Diagram of a biotechnological process. In this sense, a fermentation system is a complex and dynamic set of microorganisms or cells converting the materials and energy into more biomass and metabolic products within a controlled environment. Microbial elements of the fermentation system may be either a population, as in the majority of fermentation processes or a community as in mixed culturing and traditional fermented foods, where microbial cells are in different states of growth and development, unless the culture is synchronic. Thus, a fermentation system is inherently bound to a bioreactor that provides the controlled environment so that the specific biological and technological demands are met. As discussed later, fermentation systems can operate in several ways and within different types of fermenters, giving rise to several categories of them; however, all fermentation processes follow the same steps. 2.1. The fermentation process The fermentation process itself typically takes place in a bioreactor or fermenter, but the whole production process can be composed of multiple process steps that refine the raw materials to the end-product. These process steps include raw material treatments such as blending and sterilization as well as end-product processing such as filtration, concentration, drying, and finally packaging (Fig. 7.3). In the heart of the fermentation process, i.e. in the bioreactor the control of acidity and temperature of the growth medium is important. The constant supply of nutrients as well as gases such as CO2 and O2 is also monitored. Oxygen is the most important gaseous  Characteristics and Techniques of Fermentation Systems 187 Downstream processing Product extraction Biomass & purification Packaging separation Biomass Inoculum development Production Fermented Commercial Stock Shake >1 fermenter Bioreactor(s) broth Product culture Flask Effluent treatment Sterilization & disposal Medium preparation Design & formulation Raw matter Fig. 7.3. Main activities of a fermentation process. substrate in aerobic fermentation, while carbon dioxide is the most important gaseous by-product of the process. 2.1.1. Fermentation process steps An established fermentation process is divided in the following steps: (a) medium preparation for the growth and production of the process cell at the stages of inoculum development and main fermentation; (b) medium sterilization as well as all the ancillary equipment to assure an aseptic environment; (c) inoculum or seed production which is a pure culture in an enough quantity to inoculate the production fermenter; (d) production stage which is conducted in the main fermenter for the product formation; (e) downstream processes to separate and purify the fermentation product; and, (f) treatment and disposal of effluents produced by the process. All these steps are interrelated and the success of the fermentation process depends on an adequate optimization that should be done during the development of the process. Medium preparation: Culture media used in industrial fermentations should contain all necessary elements for the synthesis of both, the cell biomass and metabolic products; moreover, they should satisfy the technical objectives of the process by offering a suitable environment for the growth and product formation, and at the same time they must be economically profitable. Although medium 188 Food Fermentation Biotechnology design for large-scale fermentations has historically been considered by the industrial microbiologists and biochemical engineers much more a matter of art than science (Miller & Churchill 1986), recent advances in this field are oriented to transform it into a systematic science (Dahod 1999). The fundamental knowledge available about the microorganisms, substrates and process are the basis for a rational science-based approach in medium formulation in industrial processes (Ertola et al 1995). One should consider the following steps in developing an industrial medium: design, formulation, optimization and preparation. Medium design is the selection of the necessary components to achieve growth and formation of expected products according to the specific microbial physiology, while formulation is the setting up of each component concentration according to the design and the selection of raw materials that contain those medium components. Due to the lack of complete information on the metabolism of each strain and the variability of raw materials available, it is always necessary to optimize the medium composition. Finally, medium preparation occurs each time a new batch is set and consists of the dosage and blending of raw materials to obtain the requested components’ concentrations as well as sterilization. There are at least three strategies for a rational design of fermentation media. First is to design it according to the elementary composition of a specific microorganism and of the main product that is produced. This information can be attained through either an elemental chemical analysis or through the searches in the published information on the same or related species (see for instance, Atkinson & Mavituna 1983). Second is to design according to the stoichiometry of growth and product formation. For this, a chemical equation should be established as follows (Bailey & Ollis 1986): CH m Ol + aNH 2 + bO2 → Ybc CH p On N q (biomass ) + Ypc CH r Os N t ( product ) + cH 2 O + dCO2 (1) where, Ybc is the yield coefficient of biomass based on carbon; Ypc is the yield coefficient of product based on carbon; a, is the stoichiometric coefficient for nitrogen requirement, and b for oxygen. From eq. 1 it is possible to estimate the true yield coefficient based on substrate for biomass (Yx/S) and for product (Yp/S) considering the “one-carbon formula” (OCF) of biomass and product: c (OCF weight )biomass Y x = Yb (2) S (OCF weight ) substrate c (OCF weight )product Y p = Yp (3) S (OCF weight ) substrate However, there are only few industrial processes in which there is a straightforward stoichiometric relationship between both sides of eq. 1. Moreover, cell physiology is very complex and metabolic pathways for some metabolites are too intricate, which make it impossible to represent the growth and product formation in a single chemical equation. Likewise, all the cells derive part of the  Characteristics and Techniques of Fermentation Systems 189 carbon from the substrates to obtain energy for maintenance which is a cost that will reduce the theoretical true yield values. The third strategy for medium design is based on the yield factors and maintenance energy. Let us consider that carbon substrate is consumed for at least three physiological activities (ΔS): growth energy, assimilation and maintenance energy, and this will end in a given amount of biomass (ΔX). Then we can derive a growth yield factor as follows: ΔX total produced ΔX produced Y x′ = = (4) S (ΔS )growth energy + (ΔS )assimilation + (ΔS )maintenance energy ΔStotal consumed The same reasoning can be applied to the case of product formation considering that this proceeds from a fraction of carbon substrate that ends in it, thus giving a product yield factor: Y p′ ΔP = (5) S ΔStotal consumed Yield factors are not constant and they vary with the species, conditions and even with the operating modes of the fermentation system. On the other hand, maintenance energy may represent either a small fraction of the substrate consumed that can be ignored or a significant fraction that can be over 50% of the consumed substrate as in secondary products (Stambury & Whitaker 1984). Medium formulation will be done according to the design employed for a particular situation and it will consist of determining the amounts of each component taking into consideration the yields, and, very frequently, the volumetric productivity of biomass (Γb) or product (Γp), for a given operating time (tl), a given biomass concentration (x) and under a specific production rate (qp): 1 tl Γp = t ∫0 q p xdt (6) l For media preparation, the source of the components is frequently complex materials that are sub- products of the agricultural and industrial activities, which also contain the majority of the mineral requirements (see: Dahod 1999; Miller & Churchill 1986). Organic complex materials are most popular for the medium preparation, and more related to a bioeconomy concept (Thomsen 2005). However, the use of chemically defined medium is gaining popularity in some commercial fermentations, particularly for the preparation of biological products (Zhang & Greasham 1999). Although these media are still not frequently developed for industrial processes, they do exhibit favorable characteristics at large-scale that are not observed with the traditional complex media. The production of some biological products requires the addition of inducers or precursors to the medium. Thus, for the production of vitamin B12 the addition of cobalt ions, glycine, threonine, ä-aminolevulinic acid or compatible solutes such as betaine (found in high contents in sugar beet molasses) and choline proves to be beneficial (Martens et al 2002). 190 Food Fermentation Biotechnology In spite of the information available about many microbial species and industrial processes, a rational science-based design and formulation of media is still difficult due to several factors. In this sense it is necessary to apply an optimization step by using the statistical techniques other than the classical – costly and time consuming-approach of changing one component while fixing the others at a certain level. For the development of a new medium, a two-steps strategy is recommended. In the first stage, important factors are considered to determine which components – and physical conditions–are the most promising for optimization. In this step, the Plackett-Burman design is most recommended. In the second stage, the most important ingredients are then optimized by response surface techniques. Readers can find more details by referring Greasham & Inamine (1986) and Strobel & Sullivan (1999). 2.1.2. Sterilization In a fermentation process, the bioreactor and all additions (medium, air and all the ancillary equipment) must be completely sterile, at least from the engineering point of view. According to an operational or engineering definition of sterilization given by Bader (1986), it is the destruction or removal of all forms of life from a defined system through either a discontinuous or a continuous process. However, it is also important to consider that in some fermentation processes one is frequently not as much concerned with sterilization as with actual prevention of contamination. Contaminants may affect the fermentation process since they can also use the nutrients contained in the culture medium, thus, lowering the process productivity, degrading the fermentation product or producing metabolic products that make downstream processing more difficult, or they may even invade the final product. Contamination can be prevented by using the pure cultures as inoculum, sterilizing fermentation medium, vessels and ancillary equipment, and using aseptic techniques during the whole process. Sterilization can be carried out by chemical, thermal, radiation and physical process, being thermal and physical retention the most used. Under the thermal sterilizing conditions, death of microorganisms follows a first-order kinetics, considering N as the number of viable microorganisms present in the medium, under a death rate k, for a period of time t: dN = – kN (7) dt Upon integration, eq. 7 gives: N = N0e–kt (8) The reaction death rate k is a function of the environmental conditions, the organism and its history (Bader 1986). For practical reasons, it is useful to express eq. (8) in a logarithmic form: N ln = – kN (9) N0  Characteristics and Techniques of Fermentation Systems 191 where, N0 is the number of viable microorganisms at the beginning of the thermal treatment and N is the final number of viable microorganisms after a period of thermal treatment t. As stated elsewhere, N is taken to represent the probability that at least one viable microorganism is still present in the medium after sterilization (Trilli 1986). Inverting eq. 9, one can obtain an expression to be used as “sterilization criterion, Del factor or design criterion” (∇): N0 ∇ = ln = kt (10) N However, in batch sterilization, the temperature profile describes three phases-heating up, holding, and cooling down – each having a killing effect on the microorganisms. Thus, the final design criterion results from the combined activity of these phases: ∇Total = ∇heating + ∇holing + ∇cooling (11) Since many culture media may contain heat-labile components, it is important to consider a “quality criterion” (∧) (Trilli 1986), assuming the degradation rate of such components is first order: C0 ∧= = k2 t (12) C where, C0 and C are the initial and final concentrations of the heat-labile critical component, respectively. This subject is taken into account in high-temperature-short-time or continuous sterilization procedures. Air employed as carrier of oxygen in aerobic processes and any other gases which may be fed into reactor must be sterile. Although there are a number of ways of sterilizing the air, only three methods have found permanent application. These are heat, filtration through fibrous material and filtration through granular material. Heat is generally too costly for full-scale operation and air sterilization is generally carried out by filtration. Sterilization by filtration is employed to remove all the particles and microorganisms (bacteria, spores, yeast, mould, algae) from mass flows. According to the method of functioning the filters are categorized into surface or membrane filters (screen filters) and deep-bed filters. Sterile air filters are operated for around one year in industrial practice, and are required to withstand several sterilization runs. The quality of filter is measured on the basis of amount of material retained. 2.1.3. Inoculum development The preparation of a population of microorganisms from a dormant stock culture to a state useful for inoculating a final productive stage is called inoculum development. Preparation may range in scale and purpose from a small inoculum for a bioassay to 1m3 for the production of a vitamin or antibiotic in a 200 m3 fermenter. Inoculum aims to (i) minimize the loss of viable micro-organisms during the recovery from dormancy, (ii) obtain a genotypically identical copy of the population 192 Food Fermentation Biotechnology that was stored, (iii) increase biomass, and (iv) develop the culture to a physiological state suitable for the performance in the final production stage (Hunt & Stieber 1986). All initial conditions should be kept at fixed values if reproducible fermentations are to be achieved. Thus, the criteria given by Stambury & Whitaker (1984) are important when considering a protocol for inoculums development: (a) The inoculum must be in an active and healthy status to minimize the duration of the lag phase in the subsequent fermentation; (b) It must be available in sufficiently large volume to provide an inoculum of optimum size (0.1 – 10% of the medium volume); (c) The inoculum must be in a suitable morphological form; (d) It must be free of contamination; (e) The inoculated biomass must retain its product-forming capabilities. The inoculum development, which invariably involves both laboratory-based and plant-based steps, usually begins with the transfer of cells from an agar slant to a shaken flask by means of an inoculation wire fixed to a metal holder or a glass rod. Inoculum is transferred to the production plant, usually to a seed vessel. From this point of view, it is under plant control, which is more likely to be automated than the laboratory-based stages. It is important, therefore, that what is transferred is as consistent, in terms of size and quality, as possible, so that the control of the fermentation plant can be as automatic as possible. The preferred situation for the usual batch operation would be a set of process stages which are so reproducible that the sequence could be controlled in terms of time alone. However, we should consider that if the producing strain is already altered due to either contamination or the appearance of genetic variants, the inoculum will be inevitably contaminated or will contain genetic variants. Thus, the production batch will fail. In this sense, it extremely important that the “master culture” is very well conserved according to its taxonomic and genetic conditions (for a description of strain conservation techniques see Chang & Elander, 1986 and Monaghan et al 1999). The purity of the inoculum should be continuously checked throughout the whole seed culture process. Normally, this is done by using standard slow microbiological techniques but results need to be as fast as possible. In this sense, modern molecular fingerprinting techniques could speed up to know any contamination or even the presence of undesirable genetic variants during the seed process so that the production stage can still be saved. The classical method recommended by Lincoln (1960) about testing the colony type on agar plates seeded with the master culture and before initiating the inoculum development is still used (Fig. 7.4). There are several considerations that are recommended for developing the industrial inocula and these should be determined for each strain and each process. Simple activities such as the first transfer from a stock culture by using the inoculating wire loops cause wide variations in the number of cells that are seeded to the first shaken flask. This has evaluated by Webb & Kamat  Characteristics and Techniques of Fermentation Systems 193 Stock cultures 20ml each 19 vol. 1 vol. Master strain Culture until Culture until max. Log phase max. Log phase Store at – 20 C; Test 3% for purity and productivity Seed fermenters 5% Inoculum development Fig. 7.4. Classical steps in inoculum development. (1993) for Saccharomyces cerevisiae who recommended to inoculate the slants of stock culture with liquid cultures and use liquid inoculum instead of loop transfers to avoid variations of up to a 11-fold range. A common practice is to divide the first shake culture in aliquots to be used in preparing several rounds of inocula to supply them regularly over a long period of time. The conservation of these lab-seed cultures is a matter of consideration. McDaniel & Bailey (1968) found that the standard inoculum of Streptomyces viridoflavus and other microorganisms for the laboratory and pilot plant use, stored in the gas space of a liquid nitrogen refrigerator was the most satisfactory of several methods tested. The morphological and physiological condition of the inoculum is a consideration that is recommended to evaluate. Solvent production increased when the stock cultures of Clostridium butylicum were heat-shocked and three seed cultures were then performed (Gapes et al 1983). However, for Clostridium acetobutylicum neither the heat shocking nor the treatment with aqueous ethanol or butanol gave improvements in solvent production during the subsequent fermentation, but significantly improved production was obtained when culture transfers during the inoculum development procedure were made at the time of maximum cell motility (Gutierrez & Maddox 1987). If a good inoculum process is developed at the lab scale, this could be scaled-up to higher production levels, as has been shown for the production of 194 Food Fermentation Biotechnology 1,3-propanediol using Klebsiella pneumonia (Cheng et al 2007). In the case of filamentous fungi, the morphological and physiological status of the inoculum vary according to the species and the process that will be conducted (Papagianni 2004). Many fungal industrial processes require a pellet morphology that it has to come with the inoculum. For the production of penicillin V by Penicllium chrysogenum, the process performance was improved in the main culture by increasing the stirrer speed in the third preculture. The pellet size in an air-lift tower loop reactor was reduced to about half, cell growth was influenced only slightly, but the production phase was extended, and the final penicillin concentration was doubled (Möller et al 1992). For solid-state fermentation (SSF) and biofilm fermentation using filamentous fungi spore inoculum is preferred since this processes depend on cell adhesion for the development of a biofilm. Cell adhesion also triggers physiological responses that are different from those in the suspended growth (Gutiérrez-Correa & Villena 2003; Villena et al 2001). Carrier-attached starter cultures have been recommended to inoculate SSF reactors since even they contain mycelial propagules, these are already activated on cell adhesion and they continue growing in this attached form (Tengerdy 1992). A very important consideration to take into account for inoculum development is the amount of it to be seeded in the main fermenter. However, this is not only a matter of size but is related to both physiology and genetics. In most of the cases, fermenter precultures (seed fermenters) must be made in order to have enough inoculum for the main fermenter since if the main fermenter is started with too little inoculum, growth is delayed and the product formation rate can be unsatisfactory. The optimal inoculum concentration for the main fermenter determines the number of stages needed of the fermenter preculture. Frequently, the following inoculum concentrations have been recommended: bacteria, 0.1 – 3%, actinomycetes and fungi, 5 – 10%, and spore suspension, 1 – 5 × 105/l culture medium (Crueger & Crueger 1984). A clear understanding of microbial growth kinetics is necessary not only for the management of a large-scale process but also for a rational inoculum development. If we consider a closed system for growing a microorganism from a finite concentration of nutrients and a small initial population, cell number and biomass will describe a kinetic behavior as shown in Fig. 7.5, where at least four clear growth phases are observed. During the lag phase, microorganisms adapt to the new environment. Duration of this phase has a profound influence on the process productivity so it is recommended to avoid it by inoculating the culture with an actively growing (and sometimes with a late log) inoculum on as similar as possible culture medium and with an appropriate concentration of cells or biomass. When cells are well adapted to the new environment they grow at a constant rate following an autocatalytic first-order kinetics where growth rate is proportional to cell or biomass concentration as shown in eq. 13 (µ = specific growth rate, h–1): dx = µx (13) dt this new phase is called exponential or “log”. Exponential growth continues until changes in the environment like depletion of nutrients and increase of waste metabolic products thus growth slows  Characteristics and Techniques of Fermentation Systems 195 down or even stops describing a third phase known as “stationary”. Eventually, the culture will enter a last phase where cell numbers or biomass constantly decreases due to depletion of cell energy reserves. Both log and stationary phases are of main interest from the industrial point of view since primary and secondary metabolites are produced, respectively. Stationary Deceleration Death log biomass Log dx α x dt Lag Time Fig. 5. Cell growth kinetics. During log phase growth rate is proportional to cell concentration Other reason related to the amount of inoculum used in the main fermenter is rooted on the genetics of the microorganism that is being used. The appearance of variants that may display an undesired phenotype is not uncommon among microorganisms. Although mutation rate is very low, the number of variants will increase with the number of generations that a given culture will have during exponential growth. Given that loss-of-function mutations occur in 1 in 105 or 106 newly divided cells, and even rarer gain-of-function mutations occur in 1 in 109 – 1010 bacteria, the possibility of beneficial mutations leading to organisms fitter than the inoculated strain is ever-present (Ferenci 2008). Also, the final proportion of variants in the population increases with fermentation scale (Trilli 1986). If eq. 13 is expressed in terms of a period of time t from a total biomass inoculum X0, the final biomass concentration x in a given fermentation volume V, then equation 14 is obtained: Vx = eµt (14) X0 196 Food Fermentation Biotechnology Considering that the number of generations G is related to the doubling time td and the specific growth rate µ by: μt G= (15) ln 2 then, substituting eq. 15 into eq. 14 yields: Vx = eGln2 (16) X0 or G = 1.44 (ln V + ln x + ln X0) (17) Thus, as expressed in eq. 17 there is a linear relation between the number of generations and the logarithm of fermenter volume. If the variants have growth rates higher than the parental, then their impact in the fermentation outcome will be dramatic. Consider that λ is the rate of appearance of variants (expressed as the number of variants per genome per generation), α is the ratio of the specific growth rate of the variant to that of the parent strain, and xp and xgv are the biomass concentration of parent and variant, respectively, the fraction of variants after G generations will be: xp α + λ −1 = (18) xgv + x p (α − 1) + λ 2G (α+λ−1) As it can be understood from eq. (17) and (18), the higher the amount of inoculum the lower the number of generations since the frequency of undesired variants is related with increasing fermentation scale. Equation (18) is also used to follow the replacement of recombinant cells (bearing a recombinant plasmid) by plasmid-free cells by changing l by p, the plasmid loss probability per generation (Ollis & Chang 1982). 2.1.4. Production stage The central part of a fermentation process is the production stage which is carried out in a bioreactor. As mentioned above, the goal of a bioreactor is to minimize the costs of the specific product while retaining the market quality. This includes achieving a high product yield, a high productivity, and–increasingly important – a high reproducibility (Lidén 2002). Thus, the bioreactor has to do several complex and interrelated tasks (Table 7.1). It should be clear, however, that minimizing the cost of the bioreactor also means minimizing the costs of the whole fermentation process since this principally depends on the cost-determining part or parts of the process. If, for instance, running the bioreactor is cost determining, then maximization of the volumetric productivity (ΓP) of the bioreactor is fundamental. Recalling eq. 6, considering the concept of product yield based on the substrate (YP/S), and considering that substrate is taken at rate rS, we can have:  Characteristics and Techniques of Fermentation Systems 197 Table 7.1. Some tasks of the bioreactor.* Functions Containment (maintenance of sterility; recombinant cells) Addition of gaseous nutrients (oxygen) Addition of nutrient solutions (carbon and nitrogen source; other nutrients) Removal of gaseous by-products (carbon dioxide; hydrogen) Control of the environment (physical: temperature, shear rate, flow motion, pressure; chemical: pH; physicochemical: foam) Suspension (free-living cells, particulate matter, flocks) Dispersion (two-phase systems) *Modified from Lidén (2002). 1 tl tl ∫0 S Γp = −Y p rS dt (19) But if, on the other hand, the downstream processing is cost determining, then maximization of the product concentration (P) in the bioreactor is the objective. This can be expressed for a batch operation as: V P= Y p rS (20) S F and for a continuous operation under a liquid flow F in a bioreactor of volume V, as: V P = −Y p rS (21) S F From equations (19) to (21) it is clear that the cell catalyst play a fundamental role in the process. Thus design of bioreactors should be carefully considered when developing a fermentation process but it is not further described here since it is beyond the aim of this chapter (readers are referred to Asenjo and Merchuk (1995), Chisti (1989), and van’t Riet and Tramper (1991)). It is interesting to mention that in the last few years several models of mini- and microbioreactors have been developed. Small-scale cultivation has the advantages of parallelization, automatization and cost reduction for medium constituents especially in studies employing isotopically labeled tracer substrates or substrates for mammalian cells (Prokop et al., 2004; Yun and Yoon 2005). These small bioreactors are very useful for the development of fermentation process at less cost speeding up many areas of bioprocessing (Lamping et al., 2003). Thus small-scale bioreactors aim to achieve this acceleration as a result of their inherent high-throughput capability, which results from their ability to perform many cell cultivations in parallel (Betts and Baganz, 2006). Of high importance in such small-scale studies is the transferability to the later large-scale process (Kumar et al. 2004). 198 Food Fermentation Biotechnology In the production stage cells are grown either to be used as commercial product (e.g., baker’s yeast, mushrooms, single cell protein, etc) or to produce a (bio)chemical product (aminoacids, enzymes, vitamins, pigments, etc) as a consequence of their metabolic activities. From the point of view of industry, microbial metabolites have been traditionally classified following Gaden’s rationale as (a) ‘growth associated’ products arising directly from the energy metabolism of carbohydrates supplied, (b) indirect products of carbohydrate metabolism and (c) products apparently unrelated to carbohydrate oxidation (Gaden 1959). Also, Bu’Lock and Powell (1965) considered that there would be two types of metabolites: “secondary” and “general or primary” accorded to its relevance to the growth where they are produced in different moments when the microorganism is cultured. Primary metabolites (ethanol, lactic acid, etc) are produced during the exponential growth (“trophophase”) and follow the same trend as the biomass while secondary metabolites are produced when the culture begins the stationary phase (“idiophase”) (Fig. 7.6). Industrially, the most important primary metabolites are amino-acids, nucleotides, vitamins, solvents and organic acids. These are made by a diverse range of bacteria and fungi and have numerous uses in the food, chemical and nutriceutical industries. Many of these metabolites are manufactured by microbial fermentation rather than chemical synthesis because the fermentations are economically competitive and produce biologically useful isomeric forms (Sanchez & Demain 2008). The traditional view is that secondary metabolites (i) do not contribute to the growth or survival of the producer (ii) are highly sensitive to the conditions stimulating their production (for example, medium composition) (iii) often have complex structures and (iv) have production rates that are decoupled from the doubling time of the cells. However, in recent years the idea that ‘secondary’ metabolites might have other functions, ranging from controlling gene expression to supporting growth, iron acquisition in microbial communities or pheromones, has become increasingly compelling (Kell et al 1995; Price-Whelan et al 2006). Secondary metabolism is brought on by exhaustion of a nutrient, biosynthesis or addition of an inducer, and/or by a growth rate decrease. These events generate signals which effect a cascade of regulatory events resulting in chemical differentiation (secondary metabolism) and morphological differentiation (morphogenesis) (Demain 1998). Thus, it is clear that production of these two types of metabolites is a complex one and that process control should be critical. The operation of the main fermenter should be carefully conducted to obtain higher productivities and yields. However, although there are several factors to control, temperature, agitation and aeration may be the most important. The transfer of energy, nutrients, substrate and metabolite within the bioreactor must be brought about by a suitable mixing device. The efficiency of any one nutrient may be crucial to the efficiency of the whole fermentation. Most industrial fermentations are aerobic processes meaning that the production microbe requires oxygen to grow. The oxygen demand is met by sparging air through the fermentation vessel and the use of an agitator increases the amount of dissolved oxygen. One of the most critical factors in the operation of a fermenter is the provision of adequate gas exchange. Oxygen is the most important gaseous substrate for microbial metabolism, and carbon dioxide is the most important gaseous metabolic product. When oxygen is required as a microbial substrate, it is frequently a limiting factor in fermentation. Because of its low  Characteristics and Techniques of Fermentation Systems 199 Trophophase Idiophase Biomass concentration Product concentration Biomass Secondary product Primary product Time Fig. 7.6. Time course profiles of cell growth, primary and secondary metabolites production. solubility, only 0.3 mM O2, equivalent to 9mg/l, dissolves in one liter of water at 20°C in an air/water mixture. This amount of oxygen will be depleted in a few seconds by an active and concentrated microbial population unless oxygen is supplied continuously. In contrast, during the same period the amount of other nutrients used is negligible compared to the bulk of concentrations. Therefore most aerobic microbial processes are oxygen limited. The mass transfer of oxygen into liquid can be characterized by the oxygen transfer rate (OTR) or by the volumetric oxygen transfer coefficient (kLa), which are critical parameters for bioreactor function. The oxygen transfer rate and the volumetric oxygen transfer coefficient are dependent on several factors as shown in Table 7.2. Oxygen transfer rate can be defined as: Table 7.2. Factors that affect oxygen transfer rate Vessel geometry: diameter, capacity Mixing properties: power, impeller configuration and size, baffles Aeration system: sparger rate, geometry, location Agitation: mechanics, pneumatics Nutrient solution: composition, density, viscosity Microorganism: morphology, concentration, suspended, immobilized or biofilm Antifoam agent used: vegetal oils, silicone antifoams Temperature: as temperature rises oxygen solubility decreases Pressure: increases oxygen solubility 200 Food Fermentation Biotechnology OTR = kLa (C* – CL) (22) where, CL = concentration of dissolved oxygen in the fermentation broth, C* = is the saturated dissolved oxygen concentration, a = is the gas/liquid interface area per liquid volume, and kL = is the mass transfer coefficient. The critical oxygen concentration (Ccrit) is the term used to indicate the value of the OTR or oxygen absorption rate which permits respiration without hindrance. When the system is in steady state there should not be oxygen accumulation in any point of the bioreactor, thus: kLa (C* – CL) = q0x (23) where, q0 = specific oxygen consumption rate and x is the biomass concentration. When CL = 0 (maximum OTR) the bioreactor can withstand a maximum biomass concentration: k L aC * xmax = (24) q0 q0 x or (kLa)crit = (25) (C * − Ccrit ) Generally the critical oxygen concentrations are 5 – 25% of the oxygen saturation value in cultures. At oxygen absorption rates which are lower than the critical concentrations, respiration rate is correlated with the O2 concentration in the solution. Above this value, no dependence between respiration rate and dissolved oxygen has been observed. In Newtonian fluids, such as those occurring in yeast and bacterial fermentations, the critical oxygen concentration is constant and is not affected by fermentation conditions. In non-Newtonian solutions, such as those occurring with filamentous microorganisms, the critical oxygen concentration has been shown to be dependent on fermentation conditions. It is desirable to guide the production process along the path to guarantee that the process will produce the product in such a way that it meets predefined quality specifications. To reach this requirement it is necessary the availability of on-line measurement techniques as reliable as possible along with sensitive and precise instruments. Although the number of monitored state variables is usually greater in small-scale bioreactors, in large-scale production bioreactors only a minimal set of the most relevant variables is monitored. This is mainly aiming to minimizing the contamination risk and avoiding wrong controller actions caused by potentially malfunctioning sensors. Some control factors are depicted in Fig. 7.7. Thus, in industrial production open-loop control schemes are preferred (Dusseljee & Feijen 1990; Kristiansen 1987; Lübbert & Simutis 2001; Sonnleitner 1999; Wang 1986). 3. DOWNSTREAM PROCESSES AND EFFLUENT TREATMENT For any commercial fermentation process the recovery and purification of the fermentation product  Characteristics and Techniques of Fermentation Systems 201 Biomass Agitation Concentration/factor level Product pH Sugar consumption Temperature Time (h) Fig. 7.7. Some fermentation parameters considered in control systems. is one of the most critical steps. High recovery yields are necessary for a product to compete successfully in the market since the cost of product recovery is generally greater than 50% of total production costs. In most fermentation processes product concentration is low and recovery in commercial quantities require extensive purification procedures (Table 7.3). In the fermentation process all activities carried out to have a microbial product ready for the market are known as Downstream Processes. The complexity of the downstream processes in a fermentation process is determined by the required purity of a given product which in turns depends on its final use. Thus, therapeutic recombinant proteins should be more than 99% pure while some enzymes can be only 30% pure or even they are sold as a dry fermented product (as some enzymes produced by solid state fermentation used for animal feed). In general, recovery process that has to be used in a given fermentation process is influenced by (a) the type of fermentation, (b) the physicochemical properties of product and sub-products or the presence of undesirable contaminants, (c) product concentration in the fermented culture, (d) product localization (intra- or extra cellular), (e) product stability, (f) scale of production, (g) required specifications, and (h) effluent treatment. In any case, the first step is biomass separation either to be processed (as in the case of intra cellular products or it is the final product by it selves) or to be discarded (in extra cellular products). Fig. 7.8 gives an outline of the main downstream processes and further information may be found several papers (Asenjo 1990; Calton et al 1986; Hatti-Kaul and Mattiasson 2001; Schügerl 2005). 202 Food Fermentation Biotechnology Table 7.3. Final concentration of some industrial fermentation products. Metabolite Concentration (g/l) Amino-acids 30 – 170 Antibiotics 0.2 – 30 Amylases 0.1 – 3 Baker’s yeast 50 – 60 b-glucosidase 0.001 Citric acid 110 Glucose oxidase 0.2 Insulin 0.08 Interferon 0.001 – 0.01 Monoclonal antibodies 0.001 – 0.005 Riboflavin 0.1 – 7.0 Vitamin B12 0.06 Bioreactor Biomass separation Centrifugation, filtration Extracellular metabolite Intracellular metabolite Cell disruption Physical, chemical enzymatic Centrifugation/sedimentation Solid-liquid separation extraction, filtration Evaporation, distillation, adsorption, Concentration precipitation, ultrafiltration Purification Chromatographic techniques Formulation Crystallization, lyophilization, atomization Product Fig. 7.8. Downstream process activities and techniques.  Characteristics and Techniques of Fermentation Systems 203 According to our current view about environmental pollution effluents from fermentation must be subjected to some type of treatment before disposal. There are three other main reasons for effluent treatment: (a) high dissolved or solid organic matter content, (b) some microorganisms could have produced toxins or are potentially harmful or are recombinant, (c) water cost and availability. In various fermentation processes effluents can be considered as by-products and are sold as animal feed supplements or soil conditioners. Some solid fermentation wastes can be also used to generate steam and electricity. A wise approach to face up to fermentation effluents is under the concept of IPS as was described in section 1 of this chapter. 4. TYPES OF BIOREACTORS There are several types of bioreactors according to different criteria of classification but innovations are constant, particularly in either oxygen transfer, shear induced by stirring, control of water activity in organic phase systems or waste biotreatment (Desshuses et al 997). The most common bioreactors will be shortly described based on the source of agitation that is used. Since bioreactions should occur within a homogeneous environment so that transfer phenomena cannot be a limited factor, mixing is need. This is accomplished by agitation of the system which can be carried out mechanically by means of stirrers or pneumatically by a gas (mostly air). Table 7.4 presents a comparison between those bioreactors. 4.1. Stirred tank bioreactor The most popular bioreactor with mechanical internal agitation, lacking a loop is the Stirred Tank Reactor (Fig. 7.9A, F). This is a cylindrical vessel usually with a height to diameter (H/D) ratio equal or greater than 2; it is also equipped with either radial or axial-flow stirrers and baffles to increase fluid turbulence producing different flow patterns inside the vessel (in tall fermenters, installation of multiple impellers improves mixing). A Variomixing bioreactor was developed which contains four baffles that rotate intermittently at a rotational speed slower or similar to the speed of a centrally placed axial flow impeller. Rotational speeds of the baffles and impeller results in the highly turbulent flow regime characteristic of conventional bioreactors with high mixing Table 7.4. Comparison between bioreactors according to the source of agitation Characteristic Mechanically agitated Pneumatically agitated bioreactors bioreactors Mechanical complexity Complex Simple Shear stress High Low Turbulence distribution Irregular Uniform Cleaning Difficult Easy Versatility High Low Cost High Low 204 Food Fermentation Biotechnology and mass transfer capacities. Stagnant zones around crevices and crannies in which wall growth may commence are avoided since the baffles are never completely at rest (Larsen et al 2004). When aeration is required, the H/D ratio of the stirred bioreactor is usually increased. This provides for longer contact times between the rising bubbles and liquid and produces a greater hydrostatic pressure at the bottom of the vessel. Typically, only 70 – 80% of the volume of stirred reactors is filled with liquid; this allows adequate headspace for disengagement of droplets from exhaust gas and to accommodate any foam which may develop. Stirred tank bioreactors require a relatively high input of energy per unit volume and give higher mass transfer rates but shear forces are also higher being detrimental for fragile cells. However, the preferred technique for the large- scale production of biopharmaceuticals from animal cells is as homogeneous cultures in stirred- tank reactors. The reasons for this are the superior process control that is possible with homogeneous systems and the relative ease of scale-up from 1- to 1000-litre reactors. Despite the fact that animal cells are relatively large in comparison with bacteria and yeast and also lack a cell wall, the shear sensitivity of these cells has not proved to be a serious concern when scaled up to an efficient commercial production process, using culture media that contain both proteins and protective components (Chisti 2000; van der Pol & Tramper 1998). 4.2. Bubble column bioreactor Bubble column (Fig. 7.9 B) is the simplest of all bioreactors. It has an H/D ratio greater than 3 without any internal mechanical devices. In bubble column reactors, aeration and mixing are achieved by gas sparging; this requires less energy than mechanical stirring. Bubble columns are applied industrially for production of bakers’ yeast, beer and vinegar, and for treatment of wastewater. Agitation is due to the upward movement of gas (usually air) bubbles which they are produced by the action on a sparger mounted at the bottom. The sparger nozzles have to be evenly distributed over the cross bottom section to prevent heterogeneous flow patterns (at higher gas velocity) (Joshi et al 2002). Thus, reduced shear forces are produced in this bioreactor. Modified columns are possible, which are fitted with one or more sieve plates or other devices at intermediate positions in the column. These provide redispersion of the gas phase. The advantages are low capital cost, lack of moving parts, and satisfactory heat and mass transfer performance but foaming can be problem. 4.3. Airlift bioreactor Airlift Bioreactors are often chosen for culture of plant, animal cells, immobilized catalyst and non-Newtonian cultures as in polysaccharide production because shear level is low (Kang et al 2001). They have an H/D ratio greater than 10 without any internal mechanical devices. Gas is sparged into only part of the vessel cross section called the riser. Gas hold-up and decreased liquid fluid density cause liquid in the riser to move upwards. Gas disengages at the top of the vessel leaving heavier bubble-free liquid to recirculate through the downcomer. Airlift bioreactors  Characteristics and Techniques of Fermentation Systems 205 A B C D E Air Air Air F G H Fig. 7.9. Principal bioreactor configurations. Stirred tank (A, F), Bubble column (B), air-lift (C, D, G), and packed bed (E, H). configurations are internal-loop vessels (Fig. 7.9 C, G) and external-loop vessels (Fig. 7.9 D). In the internal-loop vessels, the riser and downcomer are separated by an internal baffle or draft tube. Air may be sparged into either the draft tube or the annulus. In the external-loop vessels, separated vertical tubes are connected by short horizontal section at the top and bottom. Because the riser and downcomer are further apart in external-loop vessels, gas disengagement is more effective than in internal-loop devices. Fewer bubbles are carried into the downcomer, the density difference between fluids in the riser and downcomer is greater, and circulation of liquid in the vessel is faster. Accordingly, mixing is usually better in external-loop than internal-loop reactors. In Propeller Loop Bioreactors, loop circulation is promoted by a propeller, which acts as a pump to force fluid either up or down through a central draft tube. In Jet Loop Bioreactors, an external loop is employed, with a mechanical pump to remove the liquid. Gas and recirculated liquid are injected into the tower through a nozzle. 4.4. Packed bed bioreactors Packed Bed Bioreactors are used with biofilms, immobilized or particulate biocatalysts, for example during the production of aspartate and fumarate, conversion of penicillin to 6-aminopenicillanic 206 Food Fermentation Biotechnology acid, and resolution of amino-acid isomers as well as waste water treatment (Fig. 7.9 E, H) (Gu & Syu 2004). Damaged due to particle attrition is minimal in packed beds compared with stirred reactors. Mass transfer between the liquid medium and solid catalyst is facilitated at high liquid flow rate through the bed. To achieve this, packed beds are often operated with liquid recycle. The catalyst is prevented from leaving the columns by screens at the liquid exit. Aeration is generally accomplished in a separated vessel because if air is sparged directly into the bed, bubble coalescence produces gas pockets and flow channeling or misdistribution. Packed beds are unsuitable for processes which produce large quantities of carbon dioxide or other gases which can become trapped in the packing. To overcome the disadvantages of packed bed, Fluidized Bed Bioreactors may be preferred. Because particles are in constant motion, channeling and clogging of the bed are avoided and air can be introduced directly into the column. Fluidized bed reactors are used in waste water treatment with sand or similar material supporting mixed microbial populations, and with flocculating organisms in brewing and production of vinegar. Another variation of the packed bed is the Trickle Bed Bioreactor. Liquid is sprayed onto top of the packing and trickles down through the bed in small rivulets. Air may be introduced at the base; because the liquid phase is not continuous throughout the column, air and other gases move with relative ease around the packing. Trickle-bed reactors are used widely for aerobic wastewater treatment. 4.5. Membrane bioreactors Membrane Bioreactors combine selective mass transport with chemical reactions, and the selective removal of products from the reaction site increases the conversion of product-inhibited or thermodynamically unfavorable reactions (Giorno & Droli 2000). Their potential applications have lead to a series of developments in several technology sectors: (1) the induction of microorganisms to produce specific enzymes; (2) the development of techniques to purify enzymes; (3) the development of bioengineering techniques for enzyme immobilization; and (4) the design of efficient productive processes (Tischer & Kasche 1999). The membrane can have a flat-sheet shape, assembled in a plate-and-frame module or a spiralwound module, or tubular-like, assembled in tube-and- shell modules; it can also have a symmetric or an asymmetric structure. The biocatalyst can be flushed along a membrane module, segregated within a membrane module, or immobilized in or on the membrane by entrapment, gelification, physical adsorption, ionic binding, covalent binding or cross-linking. 4.6. Photobioreactors Photobioreactors are technical systems for the production of phototrophic microorganisms, particularly microalgae. These systems have to be evaluated in their various configuration concepts regarding their potential productivity and economic feasibility. The most important and most obvious differences in microalgal production systems are the exposure of the microalgal culture to the environment (Lorenz & Cysewsky 2000). Open systems can be divided into natural waters (lakes,  Characteristics and Techniques of Fermentation Systems 207 lagoons, ponds) and artificial ponds or containers, erected in very different ways. Regarding the technical complexity, open systems such as the widespread raceway ponds may vary considerably, but they are still much simpler than more recent closed systems for the cultivation of microalgae. A covered open system has been proposed for the cultivation of transgenic seaweeds (Qin et al 2005). Most of these closed systems consist of tubular photobioreactors with tubes of various shapes, sizes and length as well as the transparent materials used (Pulz 2001). Closed photobioreactors are characterized by the regulation and control of nearly all the biotechnologically important parameters which produces a reduced contamination risk, no CO2 losses, reproducible cultivation conditions, controllable hydrodynamics and temperature, and flexible technical design (Ugwu et al 2002). In all cases the biotechnological solutions for optimum growth, with the main factors being light and turbulence, are key issues for success. Closed photobioreactors are currently tested for microalgal mass cultures in several configurations like tubular systems (glass, plastic, bags), flattened, plate-type systems, and ultrathin immobilized configurations. Vertical arrangements of horizontal running tubes or plates seem to be preferred for reasons of light distribution and appropriate flow. Also, airlift bioreactors are especially suitable for photosynthesis since it provides ordered mixing (light/dark cycles). Generally, in an internal loop airlift bioreactor, the region of the riser can be regarded as a dark region, and the region of downcomer as an illuminated region where illuminance may vary spatially (Wu & Merchuk 2003). 5. OPERATING FERMENTATION SYSTEMS During the development of a fermentation process one of the most important decisions to be taken is that concerning to the bioreactor configuration and its operating conditions because more than 50% of the capital investment is referred to the bioreactor (Shuler & Kargi 1992). Several considerations have to be observed for this decision which includes the type of product and its recommended concentration as well as productivity, yields, substrate conversion and impurities or by-products. Also, the type of producing cells to be used is another consideration since this is related to their sensitivity to shear stress, rheological modifications to the fluid system, their oxygen demands and their genetic instability. Finally, bioreactor level–which is also related to the market size–determines the difficulty and complexity of the downstream processes that are required for the given product and its market specifications. Thus, the choice of bioreactor and its operating mode has to be done in the context of a fermentation system and this in the context of an integrated production system. We have already discussed how any fermentation process works and the main bioreactor configurations that are available. Now, we will briefly discuss the main operating systems that are available for a wise choice, and these may be divided as closed (batch), open (continuous) or semi-open (fed-batch) systems. A general abstract representation of these operating systems is given in Fig. 7.10, where the “control volume or region” is defined as “a region in the space throughout which all the variables of interest are uniform” (Sinclair 1987; Sinclair & Cantero 208 Food Fermentation Biotechnology 1990; Sinclair and Kristiansen 1987). In an operating fermentation system the following balance equation can be given by: 1 d (Vy ) V dt = ∑r gen − ∑ rcons + Dyi − γDy0 (26) where, y is any state variable, Srgen means the sum of all generation rates, Srcons is the sum of all consumption rates, V is the bioreactor operating volume, F is the flow rate, D is the dilution rate (D = Fi/V), γ = F0 / Fi, and sub-indices i and 0 denote the input and output of any state variable to the system, respectively. Fi F = flow rate Xvi Xv = viable cell concentration Xd = non-viable cell concentration Xdi S =substrate concentration Fo Si P = product concentration (δ) Xvo δ = separation factor Pi V = operating volume (δ) Xdo i, o = inlet and outlet conditions So Po Batch Continuous Fed-batch Cell Recycle Fi = Fo = 0 Fi = Fo = 0 Fi > 0; Fo = 0 Fi = Fo = 0 δ= 1 δ= 1 δ= 1 δ= 1 Fig. 7.10. General diagram of operating fermentation systems. 5.1. Batch system Batch systems are the second most commonly used fermentation systems and they are the simplest in operation. It operates as a closed systems where Fi = F0 = 0 so that the operating volume is constant. Maximum levels of C and N are limited by inhibition of cell growth so that biomass production is limited by the C/N ratio load and by the production of toxic waste products. Cells are harvested when biomass levels or product levels start to decline. The best advantage of batch processing is the optimum levels of product recovery. The disadvantages are the wastage of unused nutrients, the peaked input of labour and the time lost between batches. During batch processing, heat output (temperature), acid or alkali production (pH) and oxygen consumption (DO) will range from very low rates at the start to very high rates during the late exponential phase. Kinetically, in batch process m and qS (specific consumption rate) are at the maximum value and not controlled.  Characteristics and Techniques of Fermentation Systems 209 From eq. (26) and considering that dV/dt = 0, state variables can be balanced as followed: dy dt = ∑r gen − ∑ rcons (27) Considering that in each batch there is a significant time (tcsf) expended for cleaning, sterilizing filling, lag phase and harvesting, the total batch (tbatch) time will be given by: tbatch = tgrowth + tcsf (28) where tgrowth is the time required for growth. The batch time can be expressed as follows (Shuler and Kargi 1992): 1 xmax tbatch = μ ln x + tcsf (29) max 0 where, x0 is initial cell concentration at the beginning of the fermentation stage. Thus, taking into account eq. (4), batch production rate (Prbatch) will be: Y x′ S0 (Prbatch) = (30) S 1 xmax ln + tcsf μ max x0 Batch operations are well suited for secondary metabolite production. However, due to flexibility batch operations are preferred to continuous operations. 5.2. Continuous system Open fermentation systems are best known as continuous due to input-output interchange of materials. The culture medium may be designed such that growth is limited by the availability of one or two components of the medium. When the initial quantity of this component is exhausted, growth ceases and a steady state is reached, but growth is renewed by the addition of the limiting component at a continuous flow. Since input flow is equal to the output flow, bioreactor volume will be kept constant. Fresh media is continuously added into the bioreactor and at the same time bioreactor fluid is continuously removed. The cells continuously propagate on the fresh medium entering the reactor and at same time, products, metabolic waste products, and cells are removed in the effluent. Continuous culture reactors need to be shut down less frequently than batch systems. Although continuous reactors are rarely used for large scale production of specific products, they are widely used in waste treatment processes. The steady state of continuous processing is far easier to control. In the continuous processing, the rates of consumption of nutrients and those of the output chemicals are maintainable at optimal levels. Besides, the labour demand is also more uniform. Continuous culturing is highly selective and favours the propagation of the best-adapted organism in culture. There is a marked reduction 210 Food Fermentation Biotechnology in processing time, with the same holding capacity of equipment or, alternatively, a marked reduction in equipment size for the same rate of production. Furthermore, as a consequence of steady-state operation, the product can be expected to have greater uniformity. A continuous fermentation is more adaptable to instrumental control, and it is better integrated into such other parts of the over-all process as the preparation of medium and the recovery of product, which may be operated more economically and efficiently in a continuous manner (Maxon 1955). However, continuous processing may suffer from contamination, both from within (appearance of genetic variants) and outside. The fermenter design, along with strict operational control, should actually take care of this problem. Considering the general balance equation (27), γ = F0/Fi = 1, and since Fi = F0, dV/dt = 0, then the balance equation for continuous culture will be: dy0 dt = ∑r gen − ∑ rcons + Dyi − Dy0 (31) In continuous systems it is not possible that cell population can grow at µmax as in batch systems because the culture will be “washed out” (Sinclair and Kristiansen 1987). However, when either biomass or a primary metabolite is of interest continuous systems should be preferred. According to Shuler & Kargi (1992), most commercial fermentations operate with xmax/x0 ≈ 10 to 20 and continuous cultures will always have a higher production rate (Prcont) than batch systems: Prcont ≈ μ max Y x′ S0 (32) S then, the ratio of biomass production rates is: Prcont xmax = ln x + μ max tcsf (33) Prbatch 0 This ratio favors continuous systems but due to their disadvantages batch or fed-batch systems are used. 5.3. Fed-batch systems A third operating mode is considering a semi-open system where there is an input flow of sterile culture medium until the maximum volume accepted by the bioreactor is reached. This means that this system operates with a variable volume and it is call fed-batch. In the fed-batch system, a fresh aliquot of the medium is continuously or periodically added, without the removal of the culture fluid (Brown 1990). However, substrate is consumed as soon as in enters the control volume so that dS/dt ≈ 0, and dx/dt ≈ 0; thus, as long as the feeding continues the system is always at a quasi-steady state. The fermenter is designed to accommodate the increasing volumes. Balance equation for fed-batch systems is similar to eq. (31) and since V is not constant it can be derived as follows:  Characteristics and Techniques of Fermentation Systems 211 d (Vy ) Vdy ydV = + (34) dt dt dt Since: dV = Fi – F0 (35) dt Combining eq. (35) into eq. (34) yields: d (Vy ) Vdy = + y (Fi − F0 ) (36) dt dt Expressing eq. (36) per volume unit: 1 d (Vy ) dy = + Dy (1 − γ ) (37) V dt dt Combining equations (26) and (37): dy dt + Dy (1 − γ ) = ∑r gen − ∑ rcons + D ( yi − γ y0 ) (38) giving dy dt = ∑r gen − ∑ rcons + D ( yi − y0 ) (39) This equation is then similar to eq. (31) (Sinclair 1987; Sinclair and Kristiansen 1987). Conventionally there are two types of nutrient-feeding mode: without feedback control and with feedback control. In the former case the feed rate of nutrient is kept constant or changes are controlled in a predetermined manner, while the latter case is feedback-controlled, utilizing a measurable parameter as control index, such as pH, DO concentration, RQ, and nutrient concentration (Park 2004). Fed-batch is used to take advantage of the fact that substrate concentration in the control volume is maintained at a very low level. This situation may be advantageous in avoiding catabolite repression by the rapid use of carbon sources, maintaining the aeration capacity of the bioreactor, and avoiding some toxic effects of medium or excreted metabolic components. Fed-batch culture has proven to be the most robust technique to maintain growth rates just below the acetate threshold, particularly through exponential growth, in recombinant Escherichia coli for heterologous protein production (Johnston et al 2002). For large-scale applications, fed-batch, high-cell-density E. coli K-12 derivative cultivation strategies have proven suitable for considerably increasing the volumetric productivity of these processes (Lemuth et al 2008). Irrespective of more sophisticated closed- loop strategies, fed-batch cultivations are usually carried out with open-loop control via exponential 212 Food Fermentation Biotechnology or constant feeding. Exponential feeding maintains the specific growth rate at a constant level. The maximal biomass concentration that can be achieved with this strategy depends on sufficient oxygen supply and heat transfer capacities. At a constant feed rate, the specific growth rate gradually decreases due to declining carbon and energy source levels. 5.4. High-cell-density cultures A primary goal of fermentation research is the cost effective production of desired products using high productivity techniques. Increasing cell mass or product concentration is one of the most effective methods for improving productivity in the fermentation process. Advances in protein engineering, the completion of numerous bacterial and fungal genome sequencing projects, and the isolation of new genes from extremophiles have led to an increased number of useful proteins. However, to enable recombinant proteins to play a role in applications where larger quantities are required production technologies that are more efficient and robust are desirable. Overall protein productivity can be improved by increasing the product of two variables: (i) the amount of recombinant protein per cell (specific productivity) and (ii) the amount of cell mass per unit of volume and time (cell productivity) (Srinivasan et al 2002). Submerged high-cell-density cultivations (HCDC), with dry biomass concentrations higher than 100 g/l of various bacteria and yeasts have been obtained. HCDC were first established for yeasts to produce single-cell protein, ethanol and biomass. Later, dense cultures of other mesophiles producing various types of product were developed (Riesenberg & Guthke 1999). High-cell-density fermentations offer many advantages over traditional fermentations in that final product concentrations are higher and downtime and water usage are reduced, yet overall productivity is improved, resulting in lower setup and operating costs. Two main strategies for HCDC have been probed. First, fed-batch culturing in several bioreactor configurations as was discussed in section 3.3. Second, cell recycling systems in which the maintenance of high cell density in the bioreactor is obtained by using a cell separator, centrifugal separation of cells, immobilized cells or recycling membrane-filtered cells. Cells are recycled by the fermenter and a small amount of effluent was exchanged for fresh medium to avoid nutrient depletion thus operating continuously. The balance equation for cell recycling systems can be established by including a separator factor (δ) into equation (31): dy dt = ∑r gen − ∑ rcons + δDx (40) Increasing the efficiency of the separator, thus lowering the value of δ, cell density will also increase. However, this technique has several drawbacks, including: substrate inhibition, limited oxygen transfer capacity, the formation of growth inhibitory by-products, and limited heat dissipation (Lee 1996). Control of the process is a difficult task for both strategies but results are in most cases impressive using multivariate statistical control, artificial neural networks, fuzzy control and  Characteristics and Techniques of Fermentation Systems 213 knowledge-based supervision (expert systems). 6. CLASSIFICATION OF FERMENTATION SYSTEMS In this chapter we have so far discussed the main characteristics of fermentation systems in a general way so that they can be applied, in principle, to any fermentation system. Most texts of industrial microbiology, biochemical engineering and biotechnology do not discuss in detail their own unspecific classification of fermentation systems. The most common way of classification is to separate fermentation systems in submerged and solid-state, considering biocatalyst immobilization a form of the former. Thus, it seems that this common classification is based on engineering criteria where the amount of free-water of the system is the most important of them and microorganisms are biocatalysts. Eventually, some subtle reference is given to the fact that in solid-state fermentations microorganisms grow adhered to a solid support. However, important biological criteria are seldom considered in the classification of fermentation systems. A very important biological characteristic of cells is referred to signaling processes. Microbial signaling is produced as a response to a stimulus or to communicate between cells within a population (Kolenbrander 2002; Waters and Bassler 2005). Cell adhesion is a characteristic found in almost all cell types across biological kingdoms. Adhesion to surfaces is not a random process but it is genetically guided through several surface molecular markers as well as specialized cell structures. Once cells adhere to a surface several signaling steps are fired so that several cell molecular, physiological and morphological changes take place giving rise to a biofilm. A biofilm is an assemblage of microbial cells that is irreversibly associated (not removed by gentle rinsing) with a surface and enclosed in a matrix of primarily polysaccharide material (Dolan 2002; O’Toole et al 2000). Thus, the concept of a biofilm presumes either a population or a community of microorganisms living attached to a surface. Biofilms can be developed on either biotic or abiotic surfaces from a single species or as a community derived from several species (David and O’Toole 2000; Fenchel 2002). This way of growth is the prevailing lifestyle of microorganisms including bacteria, yeast, filamentous fungi and even micro-animals (Armstrong et al 2001; Burzio et al 1997; Gilbert & Lappin-Scott, 2000; Watnick & Kolter, 2000). Fermentation systems are classified according to biological properties of cells. Thus, we consider three categories of fermentation systems: (i) Submerged Cell Fermentation (SCF) commonly referred as “submerged fermentation”, (ii) Surface adhesion fermentation (SAF) and (iii) Immobilized Cell Fermentation (ICF). These three fermentation categories are depicted in Fig. 7.11. 6.1. Submerged cell fermentation This is the most common fermentation system that is used in fermentation industry. For most of the history of microbiology, microorganisms have primarily been characterized as planktonic, freely  Submerged Cells Surface Adhesion Immobilized Cells Biofilm Solid state 214 Food Fermentation Biotechnology Fungal pellets Fungal biofilm Koji tray Alginate beads (Air-lift) (Stirred tank) (Packed bed) Fig. 7.11. Main categories of fermentation systems. Surface adhesion fermentation comprises both biofilm and solid-state fermentations.  Characteristics and Techniques of Fermentation Systems 215 suspended cells and described on the basis of their growth characteristics in nutritionally rich culture media (Dolan 2002). Also, the classical perception of microorganisms as unicellular life forms is almost entirely based on the pure-culture mode of growth; since microbial suspensions can be diluted to a single cell and studied in liquid culture, this mode of growth has traditionally predominated in the study of microbial physiology, pathogenesis and industrial applications in the research laboratory. However, this type of fermentation gave rise after industrial penicillin production in the 1940’s to biological engineering and to the huge developments in all fields of submerged fermentation technology. Since most of the topics considered in this chapter are directly applied to this category no further discussion is given here. 6.2. Surface adhesion fermentation This category was first proposed by Gutiérrez-Correa & Villena (2003) based on the natural tendency of cells to adhere to surfaces. As discussed above, cell adhesion is a biological mechanism responsible of this tendency and depends on cell-guided activities at the molecular level. We consider that this category may be divided in two: (i) Solid-state fermentation, and (ii) Biofilm fermentation (Fig. 7.11). 6.2.1. Solid-state fermentation (SSF) SSF is a process used for the production of fermented food, animal feed, fuel, enzymes and pharmaceuticals, which involves the growth of microorganisms (mainly fungi) on moist solid substrates in the absence of free-flowing water. SSF processes exhibit several advantages over SF, including improved product characteristics, higher product yields and productivities, easier product recovery and reduced energy requirements. Also, mixed culture SSF for enzyme production give higher yields than single culture SSF (Castillo et al 1994; Dueñas et al 1995; Gutierrez- Correa et al 1998; Gutierrez-Correa & Tengerdy 1997, 1998, 1999). Since SSF processes have been used for centuries there is a great number of references and many excellent reviews have also been published (Doelle et al 1992; Hölker et al 2004; Krishna 2005; Pandey 2003; Soccol et al 2003; Suryanarayan 2003;). Several SSF bioreactor models have been designed following two general categories: laboratory-scale and pilot and industrial-scale (Durand 2003; Fasidi et al 1996). Many designs have been published that belong to the first category including static and agitated models but only few models are used in commercial production. In general, however, many types of SSF bioreactors can be run at the bench scale level with small quantities of substrate but their scale-up is difficult due to heat generation and heterogeneity with natural supports. 6.2.2. Biofilm fermentation For many bacteria, commitment to surface attachment and subsequent growth as a biofilm is a 216 Food Fermentation Biotechnology highly regulated event (Newell et al 2009). It should be noted that adhesion and subsequent differential gene expression to generate phenotypes distinct from those of free-living organisms are two unifying processes of the biofilm concept (Ghigo, 2003; O’Toole et al 2000). Although bacterial biofilms have been extensively studied at different levels (Kierek-Pearson & Karatan 2005), fungal biofilms have received little attention. Filamentous fungi are naturally adapted to growth on surfaces and in these conditions they show a particular physiological behavior which it is different to that in submerged culture; thus, they can be considered as biofilm forming organisms according to our former concept. Differential physiological behavior of most attached fungi corresponds principally to a higher production and secretion of enzymes and also to a morphological differentiation which is absent in submerged cultures as it has been found in solid-state cultures (Akao et al 2002; Biesebeke et al 2002). We have found that filamentous fungi–Aspergillus niger-actually form biofilms and in this way are more productive than in the classical submerged fermentation (Villena & Gutiérrez-Correa 2006, 2007a); this has been corroborated by other authors. Likewise, A. niger biofilms can withstand low w