2.1 Batch Bioreactors
The batch bioreactor is the oldest and most used bioreactor in industry (Bellgardt, 2000b; Branyik et al., 2005). Its historical and most familiar use is in the production of alcoholic beverages (beer, wine, whiskey, etc.) and bread. Batch bioreactors combine all the necessary ingredients and then operate until the desired product concentration is reached at which point the product is extracted. In well-known processes where the final product is relatively cheap, product concentration can be correlated with time, leading to some process automation, lower capital needs, and lower operational costs (Bellgardt, 2000b). Batch bioreactor systems are also useful in modeling environmental issues (Fogler, 2005).
Biological application and experience have led to a differentiation based on substrate input or sterilization frequency. The simplest and least applicable variant is the batch cultivation system (Bellgardt, 2000b). Bioreactor sterilization is undertaken prior to the start of the process, followed by the medium being fed into the bioreactor creating a high substrate concentration (Bellgardt, 2000b; Williams, 2002). Inoculated microorganisms are introduced into the batch bioreactor at a low concentration to allow proper growth, which is practically uncontrollable until the process is finished. Ideally, the product is extracted once a satisfactory concentration is achieved, but the product in the batch cultivation system is also extracted if a necessary ingredient has been exhausted (Bellgardt, 2000b). Finally, the bioreactor is cleaned, and the process starts over again with bioreactor sterilization.
The need for more control over the biological process created the fed-batch (also known as the semibatch) cultivation system, which is the most widely used batch bioreactor. This deviation is a variable volume process that introduces additives at specific time intervals, gradually creating a more responsive and friendly growth environment (Bellgardt, 2000b). In other words, the bacteria receives the right amount and type of nutrients at the appropriate growth stage, creating a more efficient and controllable process. The final result is a product that can be adjusted or extracted when it achieves the desired properties.
The fed-batch and batch cultivation systems share the same cleaning and sterilization process in which the bioreactor operation is stopped and the bioreactor is emptied. This stoppage creates considerable costs and operational downtime. The repeated or cyclic system, which can be applied to both batch and fed-batch cultivation systems, may be installed in order to maximize the productivity. The cyclic cultivation system does not enter the cleaning and sterilization process, but rather empties a portion of the bioreactor while preserving part of the batch for the next cycle. Another method to increase productivity is cell retention techniques such as fluidized beds, membranes, or external separators. These options allow multiple cycles without cleaning and sterilization, which is initiated only if it is deemed that mutation risks exceed tolerable levels (Bellgardt, 2000b).
Variations of the batch bioreactor try to limit problems or expand batch bioreactor applications, but some systematic advantages and disadvantages exist. For the most part, batch bioreactors have lower fixed costs due to the simple concept, design, and process control (Bellgardt, 2000b; Donati and Paludetto, 1999; Williams, 2002); however, variable costs are generally higher for several reasons. First, cleaning and sterilization often add significant downtime and labor costs (Donati and Paludetto, 1999; Williams, 2002). These costs, however, can be limited in the cyclic cultivation system. Second, batch bioreactors have heat recovery difficulties leading to high environmental impact and energy consumption (Donati and Paludetto, 1999; Schumacher, 2000; Williams, 2002). Third, the additive nature of fed-batch and cyclic cultivation systems force the operator to prepare several subcultures for inoculation, which adds further variable cost pressures (Williams, 2002). Finally, batch bioreactors are not steady-state processes. The biological matter grows uncontrollably, leading to a changing environment that can bring about safety issues, runaway growth, or unexpected products when mutations occur (Westerterp and Molga, 2006).
Runaway reactions are unlikely in biological systems, but the variable environment can create conditions that change the competitive situation favoring a different bacterial species than the initially dominant one (Hoffmann et al., 2008). Batch bioreactors have limited, albeit relatively simple, process control that can lead to inconsistent or unwanted products, especially in a batch cultivation system. This problem can get even more pronounced in operations with a high potential contact amid pathogenic microorganisms or toxins, adding to variable costs if more stringent cleaning and sterilization procedures are needed (Williams, 2002).
The fed-batch cultivation system makes process control more challenging by creating a variable volume process. Any control mechanisms, therefore, require much more labor or capital (Bellgardt, 2000b; Donati and Paludetto, 1999; Simon et al., 2006; Williams, 2002). According to Simon et al. (2006), a fed-batch system can have thousands of control variables requiring a modern and powerful supervisory control and data acquisition system, programmable logic controllers, trained personnel, and an 8-year upgrade cycle, all of which eliminate or limit upgradability of older systems or construction of larger batch bioreactor systems (Heijnen and Lukszo, 2006; Simon et al., 2006). The complexity limits practical batch bioreactor application beyond a certain size, while other bioreactor modes enjoy economies of scale for much larger operations (Donati and Paludetto, 1999; Heijnen and Lukszo, 2006; Simon et al., 2006; Williams, 2002).
Some of the batch system costs can be offset by its flexibility. Batch bioreactors are able to produce the desired product consistently. They are also capable of producing several types of products with the same equipment or making the same type of product with different equipment. Significant product modifications can also be implemented online (Donati and Paludetto, 1999; Heijnen and Lukszo, 2006). These traits offer flexibility and competitive advantages to batch bioreactor operations; however, many problems and complications are encountered when these bioreactor schemes are used for multiple separation processes, which is often the case in industry (Barakat and Sorensen, 2008).
Most batch bioreactors operate in a changing external environment especially with respect to product and ecological demands (Heijnen and Lukszo, 2006). Researchers are able to take batch bioreactors and investigate reactions, both chemical and biological, for which data are unavailable or have never been documented, while limiting contamination and experimental or dangerous risks (Donati and Paludetto, 1999). These research bioreactors should be used for scaling purposes with care since most reactions and biological growth are affected by hydrodynamics, which are a function of bioreactor scale and type.
Ultimately, batch bioreactors contain biological matter that tends to mutate. Growth periods, therefore, need to be kept short and controlled to prevent these microbial mutations, which could produce inconsistent or undesirable products (Williams, 2002). Some fermentation processes, however, are characterized by biological matter that mutates very little allowing for long reaction times (Donati and Paludetto, 1999). Either way, a positive side effect of the controlled growth period is a higher conversion level (Williams, 2002).
A specific batch bioreactor application depends on multiple internal and external factors; however, general rules of thumb and process-specific improvements can be employed to make a smarter and more profitable selection. Batch bioreactor selectivity is based on the following factors: economic balance, production scale, reaction times, production flexibility, and the nature of the process and product (Donati and Paludetto, 1999). Typically, batch bioreactors are used for smaller operations, specialty products, long growth periods (bioreactor of choice by elimination), operations in which flexibility is vital, unsteady processes, and experimental development (Donati and Paludetto, 1999; Simon et al., 2006; Williams, 2002).
Batch bioreactor operation can be made more efficient by implementing several simple managerial procedures. First, a disturbance strategy should be developed by which personnel are trained to respond and actively scan for problems in the process leading to âlines of defenseâ that limit contamination and loss of product (Westerterp and Molga, 2006). These âlines of defenseâ should include an operating condition within which personnel and management are comfortable, an early warning system, and a reaction procedure to accidents and malfunctions including proper training and equipment (Westerterp and Molga, 2006). Second, a decision support framework (DSF) should be developed so that all personnel and management are familiar with operating costs, benefits, objectives, etc. The DSF will make production more efficient and profitable; it provides a clear outline of benefits and costs associated with general and specific options. General models, such as ANSI/ISA88 or ANSI/ISA95, are available and can be applied to all batch bioreactors (Heijnen and Lukszo, 2006). Finally, two improvement strategies can be implemented to make batch reactions more efficient. The âcook bookâ or ârecipeâ approach has been shown to improve yields in batch process operations. The user is able to adjust the biological reaction online as needed and is able to draw on extensive experience and/or knowledge to have better process control and product quality and consistency. The second strategy, production schedule optimization, has been proven effective in situations where products are made with different equipment, or equipment is used to make different products by optimizing capacity utilization (Schumacher, 2000).
