Read an Excerpt
CHAPTER 1
Introduction
1.1 WHAT IS A BIOFILM?
The simple definition of a biofilm is "microorganisms attached to a surface." A more comprehensive definition is "a layer of prokaryotic and eukaryotic cells anchored to a substratum surface and embedded in an organic matrix of biological origin" (Wilderer and Characklis 1989). The importance of biofilms has steadily emerged since their first scientific description in 1936 (Zobell and Anderson 1936) and the first recognition of their ubiquity in the 1970s (Marshall 1976; Costerton et al. 1978). It is now estimated that planktonic microorganisms constitute less than 0.1% of the total aquatic microbial life (Costerton et al. 1995); thus, biofilms seem to constitute the preferred form of microbial life.
The ability of bacteria to attach to surfaces and to form biofilms can become an important competitive advantage over bacteria growing in suspension. Bacteria in suspension can be washed away with the water flow, but bacteria in biofilms are protected from washout and can grow in locations where their food supply remains abundant. The physical structure of the biofilm also allows for distinct biological niches that facilitate the growth and survival of microorganisms that could not compete successfully in a completely homogeneous system. Furthermore, microbial activity in biofilms can modify the internal environment (e.g., pH, O2, metabolic products, or disinfectant concentration) to make the biofilm more hospitable than the bulk liquid (Rittmann and McCarty 2001).
Within a biofilm, a variety of microbial groups can contribute to the conversion of different organic and inorganic substrates. For example, when a wastewater contains a mixture of conventional and xenobiotic organic pollutants, biodegradation of the xenobiotics requires a population of slow-growing organisms – those capable of degrading the xenobiotics. The slow growers could be washed out of a suspended-growth process, since all the biomass has the same growth rate, which normally is controlled for the benefit of the bacteria that degrade the conventional organic pollutants. However, in a biofilm, the slow-growing bacteria can establish themselves deeper inside the biofilm, protected from loss, while the conventional pollutants are removed near the biofilm-fluid interface (Rittmann et al. 2000). The same situation can occur for slow-growing microorganisms that are undesirable, such as enteric pathogens that do not grow well in environmental conditions.
1.2 GOOD AND BAD BIOFILMS
Some biofilms are good, providing valuable services to human society or the functioning of natural ecosystems. Other biofilms are bad, causing serious health and economic problems. Figure 1.1 illustrates a number of good and bad biofilms, which are discussed briefly in this section.
Biofilms have been used to treat wastewater since the end of the 19th century. For example, the first trickling filter (Figure 1.1a) was placed in operation in England in 1893. Wastewater flowed into a basin filled with broken stones (the substratum) from the top and trickled down over the stones. A biofilm grew attached to the rocks, which provided a specific surface area of about 40 m2/m3 (Tchobanoglous et al. 2003). Trickling filters remain in common use today using either rock media or plastic media. The latter came into play in the 1970s and increased the substratum's surface area to about 200 m2/m3.
Rotating biological contactors (RBCs) (Figure 1.1d), also developed in the 1970s, have plastic biofilm media attached to a rotating axle. The media are partially submerged in the wastewater and continuously rotate, providing intermittent contact of the biofilm with the wastewater and atmospheric oxygen.
Low maintenance and stable operation are advantages of trickling filters and rotating biological contactors. However, the volumetric conversion rates for these systems are relatively low. Biofilm reactors with larger specific surface areas were developed starting in the 1980s. Biological filters (Figure 1.1b,c) for the treatment of wastewater and water use gravel-sized granular medium with specific surface areas up to 1000 m2/m3, which allow higher volumetric conversion rates. To prevent clogging, these biological filters have to be backwashed. Fluidized bed reactors (Figure 1.1f) are operated with increased upflow water velocities that suspend small carrier particles in the water phase. In airlift reactors (Figure 1.1e), the carrier particles are suspended in the circulating water flow that is caused by the injection of air. Specific surface areas from 2000 to 4000 m2/m3 can be achieved with the small carrier particles used in fluidized bed or airlift reactors. Added advantages can include better control of the biofilm due to uniform shear on the biofilm particles and no problems with liquid or gas distribution in the bottom of the reactor. In the moving bed biofilm reactor (Figure 1.1g), the carrier material has a density similar to the density of water. As a result, even larger particles (> 5 mm) can be suspended using a mixer or by aerating the reactor to create airlift pumping. With suspended support media, the moving bed biofilm reactor does not have to be backwashed as biofilm detachment is caused by particle-particle collisions within the system, but using larger carrier material results only in moderate specific surface areas of 330 m2/m3 (Rusten et al. 2000).
Good biofilms also are ubiquitous in the environment. For example, bacteria in the subsurface normally grow as biofilms on the soil matrix (Figure 1.1h) and can help remove contaminants from the soil or ground water. Pumping nutrients, electron donors, or electron acceptors into the soil can enhance in situ biodegradation of contaminants. Natural attenuation of contaminants can occur if environmental conditions in the soil are favorable for biofilm development and metabolic reactions leading to contaminant destruction or immobilization (National Research Council 2000). In rivers, lakes, and coastal areas of the sea, a large fraction of bacterial activity is located in biofilms colonizing stones and sediments (Figure 1.1i). Biofilms also occur naturally in soils and on the roots of plants. All these naturally occurring biofilms are crucial for cycling nutrients in the Earth's biosphere.
Bad biofilms occur in many situations. For instance, biofilms are major problems in dental hygiene (Figure 1.1l), infectious diseases (e.g., cystic fibrosis), and infections related to medical implants (e.g., catheters, heart valves, contact lenses). Growth of biofilms in drinking-water distribution systems is another example of unwanted biofilms (Figure 1.1m). Using organic matter and ammonia present in treated water, bacteria form biofilms in distribution-system pipes. The biofilms and their metabolic reactions cause the water quality to deteriorate in terms of public health and aesthetics. Reduced heat or mass transfer can be the result of biofilms growing on heat exchangers, condenser, and membranes (Figure 1.1j,k).
Bad biofilms often cannot be prevented, as they develop even under adverse conditions (extreme pH values, temperatures up to 95°C, high shear conditions, or disinfectants). Removing biofilms is often difficult in technological systems without a direct access to the exposed surfaces. One prime example is a membrane system that is used for water purification and is prefabricated in spiral-wound modules. Once biofilms develop in these modules, they generally cannot be cleaned again and often need to be discarded. Unwanted biofilms growing on ship hulls (Figure 1.1n) increase the drag forces resulting in a significantly decreased fuel efficiency of these ships. Finally, biofilm growth on metal surfaces has been shown to be a major factor in promoting corrosion (Figure 1.1o).
Overall, biofilms play significant roles in many natural and engineered systems. Understanding the mechanisms of biofilm formation, growth, and removal is the key for promoting good biofilms and reducing bad biofilms. The two definitions at the beginning of this section underscore that a biofilm can be viewed simply or by taking into account complexities. The "better" definition depends on what we want to know about the biofilm and what it is doing. Mathematical modeling is one of the essential tools for gaining and applying this kind of mechanistic understanding of what the biofilm is and is doing.
1.3 WHAT IS A MODEL?
A mathematical model is a systematic attempt to translate the conceptual understanding of a real-world system into mathematical terms (National Research Council 1990). A mathematical model is only as good as the conceptual understanding of the processes occurring in the system. If the conceptual model is good, the model should reproduce the relevant phenomena. If it is not good, we know that we must improve the conceptual model. Thus, a model is a valuable tool for testing our understanding of how a system works.
Creating and using a mathematical model require six steps.
1. The important variables and processes acting in the system are identified. As a simple example for biofilms, substrate and active biomass are variables, and processes can include utilization and diffusion for substrate and synthesis and decay for active biomass.
2. The processes are represented by mathematical expressions. Continuing the biofilm example, substrate utilization can be represented by Monod kinetics and diffusion by Fick.s law.
3. The mathematical expressions are combined together appropriately in equations that express balances on mass, energy, or momentum. Again, for the same simple example, the mass balances are on substrate and active biomass at any position inside the biofilm.
4. The parameters involved in the mathematical expressions are given values appropriate for the system being modeled. For example, substrate utilization involves the maximum specific utilization rate and the Monod half-maximum-rate concentration.
5. The equations are solved by a technique that fits the complexity of the equations. Very simple systems can be solved with purely analytical solutions, but numerical solution techniques are needed for systems that are more complicated.
6. The model solution outputs properties of the system that are represented by the model's variables. For example, the model may output the concentrations of substrate at all positions in the biofilm and the flux of substrate into the biofilm's outer surface.
Modeling is a powerful tool for studying biofilm processes, as well as for understanding how to encourage good biofilms or discourage bad biofilms. The main reason is that biofilms naturally have complex interactions of microbiological, physical, and chemical processes. Even the simplest, most homogeneous biofilm develops concentration gradients from the interplay of diffusion with utilization. When a biofilm has complex physical and microbiological structures, many more processes interact. A mathematical model is the perfect means to connect the different processes to each other and to weigh their relative contributions.
1.4 THE RESEARCH CONTEXT FOR BIOFILM MODELING
Mechanistically based modeling of biofilms began in the 1970s (e.g., Williamson and McCarty 1976; Harremoës 1976; Rittmann and McCarty 1980). The early efforts focused mainly on substrate flux from the bulk liquid into the biofilm. The mathematical model represented the biofilm as a simple "slab" in which substrate gradients are in one dimension, perpendicular to the substratum (i.e., the surface onto which the biofilm is attached). Experimental measurements were of the overall substrate-removal rate and the total biofilm accumulation.
Today, experimental techniques available for a detailed evaluation of structure and activity of biofilms have advanced significantly. Microsensors can be used to measure concentrations of many soluble compounds directly within the biofilm (e.g., oxygen, ammonia, nitrate, sulfide, pH). Thus, the availability of substrates and electron acceptors in different regions of the biofilm can be evaluated (Zhang and Bishop 1994a). Rapid advances in molecular biology and in situ hybridization techniques have resulted in the development of gene probe and microscopy techniques that permit the detailed analysis of microbial communities in complex biofilms (Lawrence et al. 1994; De Beer et al. 1997; Silyn-Roberts and Lewis 1997). For example, strain-specific and group-specific ribosomal RNA (rRNA)-targeted probes and confocal laser scanning microscopy (CLSM) are used to investigate the ecology of several diverse types of biofilms, the rumen of animals, activated sludge, and sulfate-reducing fixed-bed reactors (Stahl et al. 1988; Amann et al. 1992). Similarly, fluorescently labeled antibodies are used to examine natural microbial communities in complex environments such as soils or natural waters (Bohlool and Schmidt 1980). Techniques to study the spatial structure of biofilms are taking advantage of histological tools, such as micro-slicers (Zhang and Bishop 1994b).
With the application of these new techniques and tools, new experimental models to grow biofilms in the laboratory have also been developed. Examples are flow cells that can be directly placed on the stage of a microscope and used to observe biofilm development in real time. However, the use of flow cells is in most cases restricted to the initial stages of biofilm development (experiments generally shorter than 2 weeks and usually less than a few days) and to thin biofilms (conventional confocal laser scanning microscopy does not allow to image biofilms thicker than 100 µm). Flow cells are good examples of laboratory model systems to study certain features of biofilms in great detail but they neglect other features and operating conditions.
Motivated by the new experimental discoveries and enabled by increasingly powerful computers and numerical methods, mathematical models have evolved in parallel (Noguera et al. 1999a). The visualization of heterogeneous structures in biofilms (e.g., using images from confocal laser scanning microscopy) has triggered the development of a new generation of mathematical models in which the three-dimensional structure of the biofilm is simulated. The ability to perform in situ visualization of individual micro-colonies within a biofilm has fueled the creation of biofilm models that reproduce multi-species interactions. Because of the flexibility offered by modeling and because of the potential to integrate a multitude of processes into a single computational unit, mathematical modeling is becoming a more important tool in biofilm research.
1.5 A BRIEF OVERVIEW OF BIOFILM MODELS
Mathematical models come in many forms that can range from very simple empirical correlations to sophisticated and computationally intensive algorithms that describe threedimensional biofilm morphology and activity. The best choice depends on the type of biofilm system studied, the objectives of the model user, and the modeling capability of the user.
An example of such a very simple empirical approach is shown in equation (1.1), which describes the BOD5-removal efficiency of trickling filters in wastewater treatment:
[MATHEMATICAL EXPRESSION OMITTED] (1.1)
where BV,BOD is the BOD5 load per filter volume in kg/m3d, and F is the ratio of the flow rate approaching the trickling filter and the wastewater flow (National Research Council 1946). Like most empirical models, this one is based on finding patterns from a large quantity of data obtained under relevant operating conditions. An empirical model can be used to estimate the performance of similar systems as long as the operating conditions are within the range of the evaluated data. Most empirical models provide little insight into biofilm mechanisms, and they should not be used to predict performance outside the tested range of conditions.
Starting in the 1970s, several mathematical models were developed to link substrate flux into the biofilm to the fundamental mechanisms of substrate utilization and mass transport (Harris and Hansford 1976; Harremoës 1976; LaMotta 1976; Williamson and McCarty 1976; Rittmann and McCarty 1980; Rittmann and McCarty 1981). The major goal of these firstgeneration mechanistic models was to describe mass flux into the biofilm and concentration profiles within the biofilm of one rate-limiting substrate. The models assumed the simplest possible geometry (a homogeneous "slab") and biomass distribution (uniform), but they captured the important phenomenon that the substrate concentration can decline significantly inside the biofilm.
(Continues…)
Excerpted from "Mathematical Modeling of Biofilms"
by .
Copyright © 2006 IWA Publishing.
Excerpted by permission of IWA Publishing.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
