Sewer Processes: Microbial and Chemical Process Engineering of Sewer Networks

Citation preview

SEWER PROCESSES Microbial and Chemical Process Engineering of Sewer Networks

Thorkild Hvitved-Jacobsen

SEWER PROCESSES Microbial and Chemical Process Engineering of Sewer Networks

CRC PR E S S Boca Raton London New York Washington, D.C.

Library of Congress Cataloging-in-Publication Data Hvitved-Jacobsen, Thorkild. Sewer processes : microbial and chemical process engineering of sewer networks / Thorkild Hvitved-Jacobsen. p. cm. Includes bibliographical references and index. ISBN 1-56676-926-4 (alk. paper) 1. Sewage—Microbiology. 2. Sewage—Analysis. 3. Sewerage—Design and construction. I. Title. TD736 H85 2001 628.3—dc21

2001052524

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage or retrieval system, without prior permission in writing from the publisher. The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC for such copying. Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe.

Visit the CRC Press Web site at www.crcpress.com © 2002 by CRC Press LLC No claim to original U.S. Government works International Standard Book Number 1-56676-926-4 Library of Congress Card Number 2001052524 Printed in the United States of America 1 2 3 4 5 6 7 8 9 0 Printed on acid-free paper

1.1 1.2 1.3 1.4 1.5

Introduction and Purpose Sewer Developments in a Historical Perspective Types and Performance of Sewer Networks The Sewer as a Reactor for Microbial Processes A New Approach

2.1 2.2 2.3

Redox Reactions Chemical Kinetics in a Microbiological System References

3.1 3.2 3.3

Quality of Wastewater Microbial Reactions and Quality of Substrate References

4.1 4.2

Air–Water Equilibrium Conditions Air–Water Transport Processes

Table of Contents

4.3 4.4 4.5

Odorous Compounds in Sewer Networks Reaeration in Sewer Networks References

5.1 5.2

Illustration of Aerobic Transformations in Sewers A Concept for the Aerobic Microbial Transformations of Wastewater in Sewers Process Descriptions Sewer Process Model Oxygen Mass Balance and Modeling in Sewers Anoxic Transformations in Sewers References

5.3 5.4 5.5 5.6 5.7

6.1 6.2 6.3 6.4 6.5

7.1 7.2 7.3 7.4

8.1 8.2 8.3

Hydrogen Sulfide in Sewers — A Historical Overview Hydrodgen Sulfide in Sewer Networks Anaerobic Microbial Transformations of Organic Matter in Sewers An Integrated Aerobic-Anaerobic Model Concept for Microbial Wastewater Transformations References

Methods for Field-, Pilot-, and Bench-Scale Studies Methods for Determination of Components and Parameters for Sewer Process Modeling Final Remarks References

Wastewater Design — An Integrated Approach for Wastewater Treatment Structural and Operational Impacts on Wastewater Quality Transformations in Sewers Tools for Prediction of Sewer Processes

Factors Affecting Mechanical and Barrier Properties

8.4 8.5 8.6

Model Simulations of Sewer and Treatment Plant Interactions Sewer Processes in an Integrated and Sustainable Perspective References

This book serves a dual purpose. First, it will be of use to students in environmental engineering by enabling them to understand sewer networks from a process engineering perspective. Second, this work is a practical reference intended to help planners, designers, and operators comprehend and control the effects of processes taking place in sewer networks. Practicing engineers will find the contents useful, in that it adds a process dimension to the design and operation of sewer networks. Traditionally, books dealing with sewer systems have been devoted to hydraulics and pollutant transport phenomena. In this context, urban drainage and wet-weather performance, as well as the sewer’s interaction with treatment plants and receiving waters, were main focal points. With its concentration on in-sewer chemical and microbiological processes and their environmental effects, this book is different. It adds a new dimension by incorporating sewer processes into management and engineering. A well-known example is the generation of sulfides under anaerobic conditions. This book investigates the conditions under which sulfides are produced and investigates how anaerobic conditions affect wastewater quality in terms of biodegradability by preserving substances that enhance denitrification and phosphorus removal. In an aerobic sewer, however, removal of these easily biodegradable organic substances and creation of less biodegradable particles in terms of biomass may occur. It can be shown that aerobic conditions may improve conditions for in-sewer treatment of wastewater and result in a positive interaction with subsequent mechanical and physicochemical treatment processes. These examples illustrate that a sewer is not just a collector and transport system but also a chemical and biological reactor that must be considered as an integral part of the urban wastewater system.

Preface

The residence time of the wastewater in a sewer may, under dry-weather conditions, be of the same order of magnitude as that found in treatment plants. In-sewer chemical and microbiological processes are, therefore, of specific interest under such conditions. This volume shows in detail that the microbiological transformations of wastewater in sewers relate directly to the treatment processes and, thereby, to the changes in effluent quality. This book deals broadly with wastewater as it occurs in households and industries, and not with activated sludge within a treatment plant. From a process perspective, it is crucial to understand this difference! Although the book focuses on wastewater in sewer systems, it is not about sewer processes only. Rather, its theory and data are relevant to any wastewater or other waste subject to microbial transformation. My intent is for the book to contribute to an understanding of the sewer as a chemical and microbiological reactor for the transformation of wastewater, and that the processes within it influence not just the sewer system but the entire urban wastewater system. Chemical and biological processes in wastewater “start at the sink” and not at the inlet to treatment plants — or in the receiving waters during combined sewer overflows. The text offers a fundamental understanding of the chemical and microbiological processes that take place within sewers and shows how the processes apply to engineering issues in wastewater and sewer networks. It is thus a work of process engineering relating to sewers, which can also be used as an engineering guide to troubleshooting sewer problems. The organization of the book follows from these two perspectives: one general and one specific. Chapter 1 offers an overview providing an understanding of the sewer as a process reactor. Chapters 2 and 3 stress chemical and microbiological fundamentals that are needed to understand processes in sewers. In Chapter 4, the transport phenomena between the wastewater phases and the sewer atmosphere are dealt with, particularly in terms of reaeration and odor emission. Chapters 5 and 6 investigate the aerobic and anaerobic processes in sewers. Besides sulfides, the major objective of Chapters 5 and 6 is to establish a conceptual understanding of aerobic and anaerobic processes and develop a corresponding model. The elements within sewer processes, such as components of wastewater and modeling parameters, require both bench- and field-scale methodologies. Such methodologies are the main subjects of Chapter 7. Chapter 8 includes case studies illustrating the chief points of the book. The theory and findings of this book are based on dozens of sewer projects carried out during the past 20 years. A precipitating cause of these investigations was the shift, begun in the early 1980s in Denmark, from small, local treatment plants to a centralized system requiring an extensive collection network. One of the challenges faced was to find ways of controlling sulfide formation in the many pressure mains constructed to carry wastewater over long distances.

Preface

As a result of these investigations, it gradually became evident that in-sewer processes under aerobic conditions had further implications in terms of removing easily biodegradable organic matters from wastewater intended for advanced treatment. It is on the basis of this work, some of it carried out with the assistance of MSc. and Ph.D. students, that the present work is based. I am especially grateful to my former Ph.D. students Per Halkjaer Nielsen, Niels Aagaard Jensen, Kamma Raunkjaer, Hanne Loekkegaard, Jes Vollertsen and Naoya Tanaka, each of whom devoted at least three years of work to improving the scientific understanding of sewer processes. I am also thankful to colleagues who reviewed the text, especially Dr. Jes Vollertsen of Aalborg University and Professor José Saldanha Matos of the University of Lisbon, Portugal. Both contributed important insights and suggestions to the manuscript. Aalborg University provided me with time for research and also offered other support necessary to produce this book. I especially acknowledge the careful and professional help of the department secretary, Kirsten Andersen, who typed this text and many of its precursors over the past 20 years. In addition, I would like to thank Lizzi Levin who created the drawings for the text. Writing a book affects — some might say infects — one’s family life, and I am extremely grateful to my wife Kirsten for her support and understanding during the two years this book was being written. Her patience was often tried and never found wanting.

Sewer Systems and Processes

The flows of wastewater originating from the water supply of a community and runoff from precipitation on urban surfaces are typically collected and conveyed for treatment and disposal. The system used for this purpose is called a sewer network or a collection system that consists of individual pipes (sewer lines) and a number of installations, such as inlet structures and pumps, to facilitate collection and transport. The efficient, safe and cost-effective collection and transport of wastewater and urban runoff have been identified as key criteria to be observed. In this context, the word “safe” means that public health, welfare and environmental protection have high priority. The demand for solutions toward more sustainable water management in the cities is a new challenge. A sewer network is subject to great variability. During dry-weather periods, the flow rates reflect the behavior of the community, typically with a variability of about a factor of 10 over day and night. During wet-weather periods — and in those sewer pipes receiving both municipal wastewater and stormwater runoff, i.e., the combined sewer network — the flow rates are often increased with a factor of 100–1000 for extreme rainfall events compared with the average dry-weather flow. It is clear that efforts in research and practice have been devoted to developing systems and procedures for design and operation of sewer networks under such conditions. During the last 20–30 years, emphasis has been on drainage phenomena in terms of flow conditions for the sewer network and integrated solutions comprising the treatment plant performance and receiving water impacts during wet-weather periods. Urban drainage has been a big issue in both research and practice.

Because of the basic requirements of collection and transport, sewer networks are normally dealt with from a physical point of view, i.e., the hydraulics and sewer solids transport processes have been focal points. From this point of view, new design and operational principles have been developed, to a great extent supported by numerical procedures and the ever-increasing capacity of computers. Under wet-weather conditions, the hydraulics and solids transport phenomena in a sewer play a major role, and the chemical and microbiological processes are typically of minor importance. Not surprisingly, interests devoted to urban drainage have focused on the physical behavior of the sewer. However, under dry-weather conditions that may occur around 95% of the time in many countries, chemical and biological sewer processes may exert pronounced effects on sewer performance and on the interaction between the sewer and subsequent treatment processes. Possibly, because researchers’ and operators’ interests have been devoted to wet-weather conditions, the biological and chemical performances of a sewer, i.e., the sewer as a “chemical and biological reactor,” have not been of great concern. It is, however, apparent that the sewer cannot be neglected as a chemical and biological process system. These processes may have impacts on the sewer itself, the treatment plant, the environment and the humans in direct or indirect contact with the sewer. Textbooks dealing with collection systems have normally been devoted to planning, design, operation and maintenance focusing on the physical processes. The present text will primarily be concerned with chemical and biological processes in sewers under dry-weather conditions, and it will emphasize the microbiological aspects. There are several examples that illustrate the importance of these sewer processes and call for their control and consideration in practice. The impact of sulfide produced under anaerobic conditions is the most well-known example. Sulfide is a serious health risk for humans and is a malodorous compound that may also create corrosion problems in the sewer network. Anaerobic conditions may, however, preserve and produce those easily biodegradable substrates that enhance advanced wastewater treatment in terms of improved conditions for denitrification and biological phosphorus removal. In an aerobic sewer, removal of these easily biodegradable organic substances and production of less biodegradable particles may occur. Aerobic conditions may, therefore, improve conditions for in-sewer treatment of the wastewater and result in a positive interaction with subsequent mechanical and physicochemical treatment processes. These examples show that a sewer is not just a collection and transport system but also a process system that must be considered an integral part of the urban wastewater system. In conventional design and management practice, treatment of wastewater is assumed to take place entirely within the treatment plant, while a sewer network serves the sole purpose of collecting and transporting wastewater from source to treatment. The concept of considering the sewer as a process reactor

Introduction and Purpose

also serves the purpose of breaching this rather rigid understanding of a sewerage system. When considering the processes “starting at the sink,” a number of aspects for improved engineering can be more correctly taken into account. Furthermore, it is important that more holistic approaches expressed in terms of sustainability, public health, environmental protection and enhancing the standard of living for the general population can be considered. Figure 1.1 illustrates that sewers, treatment plants and receiving water systems should not, from a process point of view, be viewed as stand-alone units. It is the purpose of this text to provide a fundamental basis for sewer processes and demonstrate how this knowledge can be applied when dealing with design, operation and maintenance of collection systems. The overall criteria of a sewer network to observe efficient, safe and cost-effective collection and transport of wastewater, integrated with the surroundings, are still valid, however, they are expanded with a process dimension. A fundamental basis in especially applied microbiology and chemistry will be needed to get the full benefit of the text. However, whenever needed, specific aspects of microbiology and chemistry will be dealt with when relevant for the understanding of in-sewer processes. As previously mentioned, the text will primarily focus on the microbiological processes in the sewer and, to some extent, also the chemical processes, especially the physicochemical processes. The corresponding environmental engineering relevance will be included as being the ultimate goal. Hydraulics and solid-transport phenomena, installation details, materials and the traditional

design and management will mainly be included when directly related to the biological and chemical processes.

Sewers were known as early as ancient Roman times. These systems were constructed to convey storm runoff from urban areas, protecting them against flooding. Such drainage systems were also in use in both Europe and America in the 16th and 17th centuries. Generally, it was prohibited to discharge wastes from households into these storm drains. The sewer network we know today is a relatively new infrastructure in the cities. Not until the middle of the 19th century did it become common to construct underground wastewater collection systems in European cities. London and Paris were among the first, but other cities followed rapidly. The first sewers developed from the storm drains, which now were allowed to receive waterborne wastes from flush toilets, converting the drains into combined sewers. A major reason for collecting the wastewater was the enormous problem of unpleasant smell from the open sewers, cesspools and privies and the requirements for space in the streets in the densely populated cities. In the middle of the 19th century, it was correctly concluded by some far-seeing persons that “substances” excreted by cholera-infected humans were transported into the drinking water systems. However, the concept of a deliberate separation of wastewater and drinking water was not generally accepted and probably was not an initiating factor in the establishment of underground sewers. Later, it became quite clear that a sewer was a “technical hygienic and sanitary installation” that efficiently reduced epidemic diseases. This characteristic is still valid and is a major reason why underground sewer networks are still expanding, even in developing countries under conditions of limited financial resources. The sewage collected in the cities was typically discharged without any kind of treatment. Problems like bacterial contamination, odors, dissolved oxygen depletion and fish kills were identified in downstream receiving waters. Even today, such problems are well known, and other problems like eutrophication and toxicity of heavy metals and organic micropollutants have been added. The “end of pipe” solution to these problems in terms of wastewater treatment plants was introduced. Although wastewater treatment plants — with different degrees of treatment — are now common worldwide, the process of further development of treatment is still progressing. We still suffer from the aftereffects of this development by a narrow distinction between the sewer as a transport system and the treatment plant for pollutant reduction. The older parts of the cities in Europe and the United States were typically served by combined sewer networks with outfalls from where the excess

Types and Performance of Sewer Networks

wet-weather flow was routed untreated into receiving waters. Although separate sanitary and storm sewers have dominated construction for the last 50–100 years, the old combined systems are still in operation, often being improved and often equipped with basins to detain wet-weather flows. The development that has taken place is the result of 100–150 years of enormous investments. All over the world, this has left us with a sewer and treatment plant infrastructure that will be in use for an unknown future. We will still see developments in terms of technical improvements and sustainable solutions. However, we will not, as a general trend, see the wastewater collection and treatment concept replaced by, e.g., centralized collection of “solid” human excreta or on-site solutions. This could have been a realistic option for further development 150 years ago — not now!

Design and operation of sewers affect sewer processes, and, what is considered important in this context, knowledge of sewer processes can be actively considered in the design and operation of a sewer network. The type of sewer determines, to a great extent, if aerobic or anaerobic processes proceed. Furthermore, ventilation of sewers may affect the buildup and dispersion of odorous and toxic substances produced by microbiological processes. There are three main types of sewer networks: sanitary sewers, storm sewers and combined sewers. Each of these types has specific properties related to sewer processes. (1) Sanitary sewer network: sanitary sewers — often identified as separate sewers — are developed to collect and transport wastewater from residential areas, commercial areas and industries. The wastewater transported in these sewers typically has a relatively high concentration of biodegradable organic matter and is therefore biologically active. Wastewater in these sewers is, from a process point of view, a mixture of biomass (especially heterotrophic bacteria) and substrate for this biomass. The flow in sanitary sewers may be controlled by gravity (gravity sewers) or pressure (pressure sewers). In a partially filled gravity sewer, transfer of oxygen across the air–water interface (reaeration) is possible, and aerobic heterotrophic processes may proceed. On the contrary, pressurized systems are full flowing and do not allow for reaeration. In these sewer types, anaerobic processes will, therefore, generally dominate. The residence time in the sewer network affects the degree of transformation that may occur. The residence time depends on the size of the catchment and specific sewer characteristics, such as slope and length. The residence time is often relatively high in a pressure sewer, principally during nighttime hours.

(2) Storm sewer networks: storm sewers or stormwater sewers are constructed for collection and transport of stormwater (runoff water) originating from impervious or semipervious surfaces like streets, highways, parking lots and roofs. Surface waters typically enter these networks through inlets located in street gutters. The storm sewers function during wet-weather periods and typically divert the runoff water into watercourses with no or limited treatment. In terms of in-sewer chemical and microbial processes, these systems are of minor interest and will only be given limited attention in the text. On the other hand, constructions like detention and retention ponds may, as parts of such sewer networks, act as chemical and biological treatment systems. (3) Combined sewer networks: municipal wastewater and urban runoff are collected, mixed and transported in combined sewers. In terms of the processes that proceed in the combined sewers, these systems generally operate like sanitary sewers during dry-weather periods. However, because of the ability to serve runoff purposes, combined sewer networks are designed differently compared with separate sewers and include elements like overflow structures often including detention basins. These constructions may influence a number of process details. Furthermore, the combined systems are subject to a higher degree of variability in the processes compared with the sanitary sewers because of the regular shift in the flow conditions. Combined sewer networks may be constructed as either gravity sewer lines or pressure pipes — or as a combination of both of these types. The characteristic features of these three types of sewer networks are depicted in Figure 1.2.

The Sewer as a Reactor for Microbial Processes

The three types of sewer networks described represent the main types. In practice, however, sanitary sewers may often appear in partially operated separate sewered catchments, i.e., they may to some extent receive runoff water. Other alternative sewer systems include, for example, the vacuum sewers that are typically small systems, operated locally. The sewer processes take place in a complex system. They proceed in one or more of the five phases: the suspended water phase, the biofilm, the sewer sediments, the sewer atmosphere and the sewer walls, and by exchange of relevant substances across the interphases. Processes that proceed in the sewer system affect other parts of the urban system, i.e., the urban atmosphere with malodorous substances. Furthermore, wastewater treatment plants and local receiving waters receive not just those substances discharged into the sewer but also products that are the result of the sewer processes (Figures 1.1 and 1.3). The sewer is dominated by heterotrophic microorganisms that degrade and transform wastewater components. These processes proceed under redox conditions determined by the availability of the electron acceptor. The importance of the processes for the sewer and the surroundings is not just caused by the removal and transformation of organic substrates — the electron donor — but is also a result of transformation of the electron acceptors exemplified by the formation of hydrogen sulfide from sulfate.

The design characteristics and operation mode of a sewer network determine, to a great extent, which redox conditions prevail. Table 1.1 gives an overview of sewer system characteristics associated with the process conditions. Primarily, aerobic and anaerobic conditions arise, whereas anoxic conditions only exist if nitrate — or oxidized inorganic nitrogen substances — is found in the wastewater. The magnitude of the reaeration, which is closely related to the design and operation of the sewer, is a fundamental process that determines if aerobic or anaerobic conditions exist. Under aerobic conditions, degradation of easily biodegradable organic matter is a dominating process. If dissolved oxygen or nitrates are not available, strictly anaerobic conditions occur, and sulfate is the potential external electron acceptor, resulting in the formation of hydrogen sulfide, a phenomenon well known among practitioners dealing with collection systems. Furthermore, fermentation under anaerobic conditions plays a major role in formation of malodors. In addition to the redox conditions associated with the characteristics of the sewer network as outlined in Table 1.1, a number of other sewer characteristics influence the process conditions. The following examples illustrate the close relations between design and operation characteristics and process conditions: • turbulence and flow of wastewater that affect reaeration and release of odorous substances into the sewer atmosphere • ventilation of the sewer system that affects release of odorous and toxic substances into the urban atmosphere • the water depth-to-diameter ratio of the sewer that affects process conditions in terms of reaeration and relative amount of biofilm in the sewer • the wastewater velocity and shear stress that affect the buildup of sewer biofilms and deposits The relation between the design and operation characteristics and the

Aerobic

+ Oxygen

Anoxic

Oxygen + Nitrate Oxygen Nitrate + Sulfate (+CO2)

Anaerobic

Partly filled gravity sewer Aerated pressure sewer Pressure sewer with addition of nitrate Pressure sewer Full-flowing gravity sewer Gravity sewer with low slope and deposits

A New Approach

process conditions is emphasized to draw attention to the fact that knowledge of sewer processes can actively be used to design sewers to observe specific needs. What is required is, of course, a quantification of this knowledge in terms that can be used in engineering the sewer network, i.e., in planning, design and management. Knowledge that can be expressed in terms of simulation models is of specific interest. In this way, it should, as an example, not be unrealistic to “design” wastewater for intended treatment processes. Wastewater characteristics play an important role in the nature of the sewer processes and to what extent they proceed. A number of parameters like temperature and pH and quality characteristics in terms of the biodegradability of the organic matter and the amount of active biomass available are crucial for the outcome of the transformations. Microbial transformations and generally not chemical transformations characterize the sewer environment in terms of quality transformations of the wastewater. On the other hand, the physicochemical characteristics, e.g., diffusion in the biofilm and exchange of substances across the water-air interface, play an important role and must be integrated with the microbial transformations. The hydraulics and the sewer solids transport processes have a pronounced impact on the sewer performance. These physical processes, however, are typically dealt with in hydraulics and are, therefore, only included in the text when directly and closely related to the chemical and biological processes.

A sewer network and any corresponding treatment have traditionally been separately designed and operated. Two different and separate functions have been dealt with: the sewer system must collect and convey the wastewater to the treatment plant, and the treatment plant must reduce pollution load into the receiving water according to the quality standards set. Consequently, sewers are often just considered input systems at the boundaries where they are connected with wastewater treatment plants and overflow structures that discharge untreated wastewater into watercourses during rainfall. This traditional approach to sewer performance needs considerable improvement. Against this background, a sustainable and integrated dimension of wastewater management in sewer networks is needed. The safe and efficient collection and conveyance of wastewater to treatment and disposal are still main concerns. The consideration of sewer processes as an element in the design and operation of sewers will give a new dimension to the overall objective of sewer management and contribute to improved sustainability. Therefore, the technical systems must be considered holistically:

• the sources of wastewater • the sewer as a physical, chemical and biological reactor for the wastewater components being transported • the interaction with the treatment plant • the consequences for the local receiving waters The fact that the sewer is a reactor for chemical and biological processes has only played a limited role when considering the sewer’s function. It is remarkable that the quality of wastewater is typically defined at the point of inlet to a treatment plant or at a point of discharge. The fact that wastewater is subject to transformations in a sewer network — often supported by a relatively long hydraulic retention time — has generally not been considered and included in the entire management of wastewater. A similar way of neglecting microbial and chemical processes in the other urban subsystems (the treatment plant and the local receiving waters) where wastewater appears, would be quite unthinkable. Heterotrophic bacterial processes dominate transformations in the sewer. There are similarities with the corresponding processes in biological treatment plants. It is, however, from the very beginning, important to emphasize that transformations of wastewater under sewer conditions and in activated sludge or biofilm systems proceed differently. The processes in sewers, treatment plants and receiving waters must be dealt with as an entity; however, they must also be considered with their own specific characteristics. The conveyance aspects of a sewer network in terms of capacity, pollution control, structural integrity and cost are beyond dispute. However, a new approach is needed to include the chemical and biological processes in a sewer more actively in the design and operation procedures. The first step is to give those sanitary and environmental engineers dealing with the sewers a comprehensive text that covers the relevant scientific background and engineering approaches of sewer processes. As far as the author knows, a text for this purpose has until now not been available. It is the hope that this book can serve this purpose. Because the chemical and biological process aspects, until now, have played a limited role in the engineering of collection systems, a corresponding limited knowledge and experience exist. The text is not able to give complete answers to all relevant questions, but it hopefully gives the reader new and valuable knowledge on the sewer as a chemical and biological reactor.

Chemical and Physicochemical Aspects of In-Sewer Processes

Chemical and physicochemical characteristics, conditions and processes are crucial for any biological system. A chemical basis also defines the conditions for the microbiologically initiated quality changes of wastewater under transport in sewers. Equilibrium and process-related aspects are important. The objective of this chapter is to highlight fundamental chemical and physicochemical aspects of general importance for sewer systems and in-sewer processes. The contents are selected with this in mind, and the focus is on chemical and physicochemical concepts adapted to this purpose. The chapter is written to serve as a solid background for understanding process-related aspects considering the sewer as a reactor.

Microbial transformations of wastewater organic matter in a sewer include what are basically considered biochemical processes, i.e., changes of chemical components initiated by living organisms. The presence of the heterotrophic biomass in wastewater, biofilms and sediments of a sewer is central for these biochemical processes. The biomass makes use of the organic matter in the wastewater for two fundamental purposes: the organic matter (the substrate) is the carbon source for microbial growth needed for the formation of new cells, and it is the source for energy needed for maintaining life (Figure 2.1). The anabolic processes provide the substances necessary for growth of new biomass.

The catabolic processes provide the energy needed for production of new cell biomass as well as for the maintenance of the fundamental functions in the existing biomass. The energy accumulated in the organic matter is made available for the microorganisms by catabolism, i.e., a “degradation” process performed by oxidation of the organic matter (substrate). The organic matter is, thereby, the electron donor. The corresponding reduction of an external electron acceptor is performed when either dissolved oxygen, DO (aerobic process), nitrate (anoxic process) or sulfate (anaerobic process) are present. These energy-producing reactions are termed respiration processes. They require the presence of an external compound that can serve as the terminal electron acceptor of the electron transport chain. However, under anaerobic conditions, fermentation processes that do not require the participation of an external electron acceptor can also proceed. In this case, the organic substrate undergoes a balanced series of oxidative and reductive reactions, i.e., organic matter reduced in one step of the process is oxidized in another. The fundamental understanding of the microbial processes in wastewater is based on the fact that substrate utilization for growth of biomass takes place parallel to its removal for energy purposes by an electron acceptor. Figure 2.2 shows the general concept and examples where an external electron acceptor is involved. These fundamental microbial transformations take place in the water phase, in the biofilms and in the sediments of the sewer.

Redox Reactions

The microbial catabolic processes, which proceed in wastewater, provide the biomass with energy. These processes include two process steps: oxidation of organic matter and reduction of an electron acceptor. The entire oxidation-reduction process, or redox process, consists basically of transfer of electrons from the electron donor (the organic matter) to the relevant electron acceptor, i.e., from the oxidation step to the reduction step. Figure 2.3 outlines the basic concept of a redox process involving the chemical components A, B, C and D. The oxidation of component A (electron donor) to component C produces an electron. This electron is transferred to the reduction step, thereby initiating transformation of component B (electron acceptor) into component D. The total redox process is shown in Figure 2.3. The energy relations associated with the redox processes in wastewater follow the general rules of thermodynamics (Castellan, 1975; Atkins, 1978). The Gibbs’ free energy, G, of the system is the major thermodynamic function defining the state — and the change in state — of the biochemical redox processes. At constant temperature and under constant pressure, G is equal to the maximum work, which can be produced by the redox process: ∆G = ∆H − T ∆S

(2.1)

where G H T S

Gibbs’ free energy (kJ mole 1) enthalpy (kJ mole 1) temperature (K) entropy (kJ mole 1K 1)

G is, as a thermodynamic work function, a measure of the driving force (the work-producing potential) of a redox process. For a naturally occurring process,

Redox Reactions

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)

CO2/HCOO (formate) H+/H2 CO2/CH3COO (acetate) S/H2S, HS CO2/CH4 SO24 /H2S, HS NO2 /NO NO3 /NO2 NO3 /N2 Fe3+/Fe2+ O2/H2O

0.43 0.41 0.29 0.27 0.24 0.22 +0.36 +0.43 +0.74 +0.77 +0.82

G is therefore negative. The majority of common redox processes are exothermic, i.e., heat producing with H < 0, in their natural direction. Often, they are so highly exothermic that the term T S has little influence on the magnitude of G. From a fundamental point of view, the Gibbs’ free energy released by the redox process is caused by the difference in the electron potentials of the oxidation and the reduction step of the redox process (Figure 2.3). The larger the difference in electron potential between these two half-reactions, i.e., the larger the potential for transfer of electrons from the oxidation step to the reduction step, and the more electrons transferred, the more energy is released according to the following equation: ∆G o′ = − nF ∆Eo′ = − nF ( Eo′ ,red − Eo′ ,ox )

(2.2)

where n number of electrons transferred ( ) F Faraday’s constant equal to 96.48 (kJ mole 1 V 1) ∆Eo′ potential of electron acceptor, ∆Eo′ ,red , minus potential of electron donor, ∆Eo′ ,ox (V) The potential of selected half-reactions listed in order of increasing potential are shown in Table 2.1. This table is often referred to as the “electron tower.” A low potential means that there is a high tendency to produce electrons (oxidation), and a high potential means a corresponding relative low tendency, i.e., preference for reduction. As an example, if the redox pair number (3) in Table

2.1 is combined with redox pair number (11), oxidation of carbon in acetate will take place to CO2 equivalent with the reduction of O2 to H2O. For this reaction, Eo in Equation (2.2) can be directly calculated: ∆Eo′ = Eo′ ,red − Eo′ ,ox = 0.82 − ( −0.29) = 1.11V

(2.3)

It is important to understand that Tables 2.1 and 1.1 basically express the same thing. Table 2.1 is, however, energy-related and, therefore, also expresses the fundamental of the processes: to proceed if Gibbs’ free energy is released. Table 1.1 more pragmatically expresses the same thing by relating the processes directly to the availability of the relevant electron acceptors. Referring to the information given in Table 2.1 and taking acetate as an example of an organic matter, a change from aerobic to anoxic and further to anaerobic process conditions will take place by moving from redox pair number (11) to (9) and further to (6). Aerobic processes are microbiologically preferred because they correspond to the highest energy release [cf. Equation (2.2)]. Table 2.1 states the redox relations at standard conditions. Extended information on the distribution of the redox pairs — still under equilibrium conditions but under varying redox potential and pH — is given in a Pourbaix diagram. Figure 2.4 is an example of such a diagram for the binary sulfur and oxygen system in water at 1 atm and 25 C with the sum of the concentrations of

Redox Reactions

all sulfur-containing ions equal to 0.1 mM (3.2 gS m 3). Because Figure 2.4 does not include sulfur compounds like sulfite and thiosulfite, it indirectly shows that these species are not thermodynamically stable under the conditions depicted. This fact is, of course, no obstacle for the occurrence of such components.

Redox reactions in wastewater of sewer systems concern the catabolic energy-producing processes. In this respect, organic matter is the electron donor that undergoes oxidation (Figure 2.3). Concerning the reduction step of the redox reaction, the heterotrophic microorganisms may use different electron acceptors. If oxygen is available, it is the terminal electron acceptor, and the process proceeds under aerobic conditions. In the absence of oxygen, and if nitrates are available, nitrate becomes the electron acceptor. The redox process then takes place under anoxic conditions. If neither oxygen nor nitrates are available, strictly anaerobic conditions occur, and sulfates or carbon dioxide (methane formation) are potential electron acceptors. Table 1.1 gives an overview of these process conditions related to sewer systems.

The stoichiometry of the redox processes relevant for wastewater is important for balancing these processes. Therefore, procedures are needed for this purpose. A basic definition important for the determination of process stoichiometry is the oxidation level, OX, defined as follows: an imaginary charge of an element for being stabilized in a molecule compared with a corresponding low stability of the single atom. The oxidation level is not a fundamental theorem; however, it is useful for practical purposes. The understanding of OX is related to the formation of the electronic configuration around the atoms of molecules compared with the configuration of the single atoms. It is thereby also related to the formation of covalent bonds between atoms in a molecule. The major elements in organic matter relevant for the transformations in wastewater of sewers are C, O, H, N and S. The classical understanding of the stability of these elements is associated with the configuration of the electrons

around the atom for the formation of stable molecules. Carbon is the central element that shows specific characteristics that will be dealt with later in this chapter. It is important to notice that the following refer to the elements C, O, H, N and S as constituents of organic matter (the electron donor) and not to these elements as part of an electron acceptor. The oxidation levels of the relevant electron acceptors will, however, also be dealt with. To understand what is behind the definition of the oxidation level and to apply this understanding correctly, it is crucial to realize that a redox process is fundamentally an exchange of electrons (cf. Figure 2.3). The understanding of the willingness for chemical substances to undergo a transformation has its origin in the electron structure of the elements and the corresponding degree of stability. From a pragmatic point of view, the electrons around the nucleus of an element are organized in shells. The shell number, n, starting with 1, refers to the order — organized around the nucleus — that the electrons occupy when successively increasing the atomic number. The maximum number of electrons, N, that can exist in a shell is given by the following formula: N

2n 2

(2.4)

Focusing on the first three shells, and thereby also including the elements that are the most interesting for organic matter, the maximum number of electrons in each of these shells is therefore: 1st shell: 2 2nd shell: 8 3rd shell: 18 Within each shell, the electrons are present in orbits that can be interpreted as a space of the shell where a maximum of two electrons with opposite spin may occupy a position. The shells and the orbits will be filled with electrons in order of lowest energy, i.e., according to a maximum of stability of each element. The configuration of the first 18 elements of the periodic system includes the atoms that are of specific interest when dealing with organic matter in wastewater and that are important for the microbial processes in sewers (Table 2.2). Referring to Table 2.2, all elements have a tendency to reach maximum of stability, i.e., to obtain the same structure as He, Ne and Ar. Taking hydrogen as an example, it will be stable with two electrons in the 1st shell, and the same orbit. By being surrounded by two electrons, H equals a structure corresponding to He. H can obtain an approach to this structure by supplying its electron to other electronegative atoms that will share one of their electrons with H in the

Redox Reactions

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

H, hydrogen He, helium Li, lithium Be, beryllium B, boron C, carbon N, nitrogen O, oxygen F, fluorine Ne, neon Na, sodium Mg, magnesium Al, aluminum Si, silicon P, phosphorus S, sulfur Cl, chlorine Ar, argon

1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

1 2 3 4 5 6 7 8 8 8 8 8 8 8 8 8

1 2 3 4 5 6 7 8

same orbit. By definition, the oxidation level, OXH, is thereby equal to +1, and the two electrons in the orbit that H now shares with the other atom form the bond that ties the two atoms together in the molecule that is formed. Both O and N have, as shown in Table 2.2, two electrons in the 1st shell that is thereby stabilized corresponding to the structure of He. A stabilization in the four orbits of the 2nd shell requires eight electrons. A stabilization of O and N corresponds to an electron configuration where each of these atoms obtains a structure comparable with Ne. The O and N atoms require an additional supply of two and three electrons, respectively, to establish the configuration of Ne, 2 and OXN 3. Sulfur is an atom with a configuration in the 3rd i.e., OXO shell equal to the 2nd shell for oxygen, i.e., with six electrons. By accepting two more electrons, S will approach Ar in terms of stability. The oxidation level of S 2. is, therefore, the same as for O, i.e., OXS As an example, Figure 2.5 shows both a simple (plane) and a more advanced (tetrahedral) picture of the electronic structure around the oxygen atom. In addition to the existing six electrons in the outer orbits, oxygen requires another two electrons to be stable, i.e., according to the definition of OX and the text: 2. OXO The following is an overview of oxidation levels for elements as major constituents of organic matter in wastewater:

OXH OXO OXN OXS

+1 2 3 2

For all practical purposes, the following must be observed: (1) The oxidation levels for each of the elements H, O, N and S in organic matter remain unchanged by the microbiological processes that proceed in sewer systems. (2) n OXE 0 for the elements included in a molecule, where n is the number of atoms for each of the elements E. Taking H2O as an example, this statement results in the following: nH OXH

nO OXO

2 ( 1) 1( 2)

0

When observing these two basic requirements, the following two examples show the two extremes of OX for carbon: Methane, CH4: Carbon dioxide, CO2:

OXC OXC

4 +4

The oxidation level of carbon, OXC, occurs because of a fixed OXO-value closely related, and linear to, the COD-value of the C-component in question. The following equation establishes this relationship: OXC

4 1.5

COD TOC

(2.5)

Redox Reactions

Example 2.1: Determination of OXC for acetate

Example 2.1 shows that OXC for acetate is 0. This is also the case for carbohydrates (CH2O)n. Other organic components typical for wastewater may have a different OXC-value, protein often from 0 to 0.4, whereas lipids typically show OXC-values between 1 and 2.

As depicted in Figure 2.3, electrons are transferred from the oxidation step to the reduction step of the redox reaction. The number of electrons exchanged is the fundamental basis for establishing the stoichiometry of the redox process. This fact is crucial when establishing a mass balance, as will be done by modeling sewer processes (cf. Chapters 5 and 6). The OX value is, by definition, a key element in determination of this number. In order to determine the transfer of electrons, it should be realized that if organic matter as the electron donor is oxidized, i.e., degraded, no changes in the OX-values for the elements H, O and N will take place in the oxidation step when the following waste products are produced: CO2, H2O, and NH3 / NH+4 . The number of electrons exchanged in a redox reaction with an organic matter as electron donor is, therefore, determined only by the change in the oxidation level for carbon. The unit for this exchange (electron equivalent, e eq) is: 1e

eq

NA (number of electrons exchanged per equivalent of component)

where NA (Avogadro’s number) is equal to 6.023 1023. The electron equivalent (e eq) number for a redox process where organic matter is the electron donor is, therefore, the following: e eq

nC

OXC

where nC is the number of C-atoms per mole (mole 1).

(2.6)

The balancing of redox reactions based on the transfer of electrons from the oxidation step to the reduction step follows a procedure based on the calculation of OXC and e eq. These steps for a process are as follows: Balance for the electrons (e ) Balance of the charge Balance for H Control of O The use of this procedure is exemplified (cf. Examples 2.2 to 2.5). The balance for an organic matter with a simple formula CH2O (a C:H:O ratio as for a carbohydrate) is shown in Example 2.2. (1) (2) (3) (4)

Example 2.2: Balancing of the oxidation step for an electron donor, CH2O

Examples 2.3 to 2.5 are concerned with the balancing of equations for electron acceptors relevant for sewer processes. Example 2.3 shows the procedure for an aerobic process step, Example 2.4 for an anoxic process and Example 2.5 for an anaerobic process.

Redox Reactions

Example 2.3: Reduction of electron acceptor, O2 (aerobic process)

Example 2.4: Reduction of electron acceptor, NOⴚ 3 (anoxic process)

Example 2.5: Reduction of electron acceptor, SOⴚ 4 (anaerobic process)

Example 2.2 shows how the oxidation of organic matter (the electron donor) is balanced by using the production of electrons as the central element. The cases in Examples 2.3 to 2.5 are balances for electron acceptor reductions under aerobic, anoxic and anaerobic conditions, respectively (all are relevant for processes in wastewater of sewer networks). The following exemplifies how the total balance of a redox reaction is completed taking an electron donor and an electron acceptor (cf. the outline of the total redox reaction in Figure 2.3). Example 2.2 is, in this respect, used as an example of an (microbial) oxidation of an electron donor (organic matter) under anoxic conditions, i.e., with reduction of NO3 as electron acceptor (cf. Example 2.4). Oxidation of CH2O, Example 2.2: 1 CH2 O OH 4

1 CO2 4

3 H2 O e 4

(2.7)

Reduction of NO3 , Example 2.4: 1 NO3 5

6 H 5

e

1 N2 10

6 H2 O 10

(2.8)

Chemical Kinetics in a Microbiological System

Equations (2.7) and (2.8) can be added directly because the number of electrons produced equals the number of electrons consumed. If this is not the case, equalization must be done as the preliminary step. After multiplication with H2O, the final equation showing the four and realizing that H+ + OH stoichiometry of the total redox reaction is as follows: CH 2O

4 H 5

4 NO3 5

CO2

7 2 H2O N2 5 5

(2.9)

This equation is rather simple and requires no procedure as described. However, those redox reactions that cannot be directly overseen will certainly require a well-defined stepwise procedure to establish stoichiometry and a corresponding mass balance.

Chemical kinetics is concerned with the rate aspects of chemical reactions, including the description of the rate expressions in homogeneous as well as in heterogeneous systems. Because microbial processes combine the activity of living organisms with transport of chemical reactants and products between a bulk water phase, across a cell wall and inside a cell, all microbiological processes are heterogeneous, by definition. For practical reasons, however, microbial processes in the wastewater phase can be considered homogeneous, whereas processes in sewer biofilms are heterogeneous. The metabolism of microorganisms is complex. However, the metabolic pathways followed by anabolic and catabolic processes need to be described in simple terms to be applied for design and operation of urban wastewater systems.

As already stressed, processes in the wastewater phase can, from a practical point of view, be considered homogeneous. The reactions may depend on the concentration of a relevant reactant and may often be described as either zero-order (0-order) or first-order (1-order) reactions.

A 0-order reaction is independent of the concentration of the reactants, i.e., the reaction rate is proportional to a constant multiplied with the concentration of a reactant raised to the power of zero:

dC dt

kC 0

k

(2.10)

where C concentration of reactant (g m 3) t time (h or d) k 0-order rate constant (g m 3 h 1 or g m 3 d 1) If the initial concentration of C for t t0 is C0, Equation (2.10) yields, by integration from t0 to t: C C0

k (t t0 )

(2.11)

In a microbial system a 0-order reaction may proceed under conditions where the biomass or substrate concentrations are high compared with their changes. Such conditions are not typical for the wastewater phase in a sewer. However, a 0-order reaction may proceed when factors such as surface area available for adsorption limit the reaction rate.

A 1-order reaction is dependent on the concentration of a reactant raised to the power of 1, i.e.: dC dt

kC

(2.12)

where C concentration of reactant (g m 3) t time (h or d) k 1-order rate constant (h 1 or d 1) If the initial concentration of C for t t0 is C0, Equation (2.12) yields, by integration from t0 to t: C

C0 e

k ( t t0 )

(2.13)

As shown in Equation (2.13), a 1-order reaction is equivalent to an exponential change of a component. A great number of microbial processes under sewer conditions are considered to follow this type of kinetics, and the 1-order

Chemical Kinetics in a Microbiological System

reaction is, therefore, the traditional way to describe microbial decomposition of organic matter in wastewater. An example of a process expected to follow 1-order kinetics is shown in Example 2.6. Example 2.6: First-order kinetics for BOD removal under aerobic conditions in wastewater Wastewater is transported under aerobic conditions in a half-full intercepting gravity sewer pipe for 4 hours. It is assumed that transformation of the organic matter only proceeds in the wastewater phase and follows a 1-order removal kinetics. The following wastewater and systems characteristics are given as follows: BOD5 is 200 g O2 m 3; temperature is 20 C; pipe diameter is 0.3 m; 1-order removal rate of BOD5 is 0.05 h 1 According to Equation (2.13): BOD5 (t

4 hr)

BOD5 (t

0)e

0.05 4

164 g O2 m

3

Under the conditions considered, a BOD-removal equal to 18% takes place during transport in the sewer.

Monod (1949) divides microbial growth into six well-defined phases (see Figure 2.6): (1) Alag phase (a-b) where the microorganisms may adapt to the environment (2) An acceleration phase (b-c) where the growth rate increases (3) An exponential growth phase (c-d). When the growth conditions are nonlimited, exponential growth may take place, and changes in biomass and substrate are maximal. According to Equations (2.12) and (2.13), exponential growth kinetics follows the expressions (2.14) and (2.15) corresponding to a maximal and constant specific growth rate and a minimal and constant generation time: dX dt

X

(2.14)

where X concentration of active microbial biomass (g m 3) t time (h or d) specific growth rate constant (h 1 or d 1) X

X0 e

( t t0 )

(2.15)

(4) A phase (d-e) for declining growth with a negative rate of acceleration for the growth (5) A stationary phase (e-f) where the microorganisms obtain their maximal population size. The substrate in this phase is sufficient for maintaining the population; however, it does not allow further increase. (6) An endogenous phase (f-g) corresponding to death and inactivation of the biomass, e.g., caused by reduced environmental conditions for survival Equations (2.14) and (2.15) are important under sewer conditions, not just because nongrowth limited conditions may exist. The equations are also the basis for description of the kinetics when the growth of the biomass takes place under substrate or other external environmental conditions limiting the growth rate. Substrate-limited growth in terms of reduced availability of both the electron donor and the electron acceptor is common in wastewater of sewer systems. Based on the concept of Michaelis-Menten’s kinetics for enzymatic processes, Monod (1949) formulated, in operational terms, the relationship between the actual and the maximal specific growth rate constants and the concentration of a limiting substrate [cf. Equation (2.14)]: max

S (S K S )

(2.16)

Chemical Kinetics in a Microbiological System

where maximum value of µ under nonlimited growth conditions (h 1 or d 1) S concentration of substrate (g m 3) KS saturation constant (g m 3) max

The relationship between and S as depicted in Figure 2.7 is relevant because it quantifies the importance of a substrate in terms of its concentration on 1/2 max for S KS. For this the growth rate. As seen from Equation (2.16), reason, KS is also named the “half saturation constant.” Equation (2.16) and the corresponding curves shown in Figure 2.7 are called the Monod expression and Monod curve, respectively.

As previously mentioned, all biologically initiated reactions are basically heterogeneous. However, for practical reasons, the processes in the suspended phase can be considered homogeneous. Processes in biofilms proceed by exchange of electron donors and electron acceptors with the surrounding bulk water phase. These processes are, therefore, heterogeneous. The simple kinetics for uptake of soluble substrate of the bacteria in a biofilm is traditionally described by a combination of mass transport across the water/biofilm interface, transport in the biofilm itself and the corresponding relevant biotransformations. Transport through the stagnant water layer at the biofilm surface is described by Fick’s first law of diffusion. Fick’s second law of diffusion and Michaelis-Menten (Monod) kinetics are used for describing the combined transport and transformations in the biofilm itself (Williamson

and McCarty, 1976a and 1976b; Harremoës, 1978). A detailed description of biofilm kinetics is found in Harremoës (1978) and Henze et al. (1995). In the following, an overview of biofilm kinetics relevant for sewer processes will be given. As depicted in Figure 2.8, mass transport of substrate from the bulk water phase takes place through a fluid boundary layer (liquid film) and into a biofilm followed by a combined diffusion and utilization of the substrate in the biofilm. The diffusion of the substrate from the bulk water phase across the stagnant liquid film is described by Fick’s first law (cf. Figure 2.8): J

ADw

S x

ADw

Sw

Sl Ll

(2.17)

where J A Dw S x S/ x

flux of substrate in the x direction (g s 1, g h 1 or g d 1) cross-sectional area through which flux is occurring (m2) molecular diffusion coefficient in water (m2 s 1, m2 h 1 or m2 d 1) substrate concentration (g m 3) distance (m) concentration gradient of substrate in the x direction (g m 3 m 1)

The biofilm is considered homogeneous, having a well-defined thickness, and the transport in the biofilm is determined by molecular diffusion. As this transport process is relatively slow, it is normally the limiting process for the

Chemical Kinetics in a Microbiological System

transformation in the biofilm. If Fick’s first law [Equation (2.17)] is applied for the transport in the biofilm, the following is valid: 2

J x

A Df

S

(2.18)

x2

where Df molecular diffusion coefficient in the biofilm (m2 s 1, m2 h 1 or m2 d 1). If no reaction proceeds in the biofilm, the flux of substrate, J, into the film is constant (equal to 0) and, consequently, J/ x 0: 2

Df

S

x2

1 J A x

0

(2.19)

Equation (2.19), which concerns a situation without processes in the biofilm, can be extended to include transformation of a substrate, an electron donor (organic matter) or an electron acceptor, e.g., dissolved oxygen. If the reaction rate is limited by just one substrate and under steady state conditions, i.e., a fixed concentration profile, the differential equation for the combined transport and substrate utilization following Monod kinetics is shown in Equation (2.20) and is illustrated in Figure 2.8. Equation (2.20) expresses that under steady state conditions, the molecular diffusion determined by Fick’s second law is equal to the bacterial uptake of the substrate. Df

d2S dx

2

k0 f

S KS

S

(2.20)

where KS k0f

saturation constant (g m 3) 0-order rate constant per unit volume in biofilm (g m 3 s 1, g m 3 h 1 or g m 3 d 1)

The 0-order rate constant, k0f , is a maximal volume-based characteristic value for a biofilm and a substrate: k0 f

qmax X f

where qmax maximal uptake rate of substrate in the biofilm (h 1) Xf bacterial density (concentration) in the biofilm (g m 3)

(2.21)

Equation (2.20) is nonlinear and has no analytical solution. However, analytical approximations exist (Harremoës, 1978; Henze et al., 1995). The biological processes in biofilms are either described by 1-order or 0-order kinetics. However, the 0-order reaction is of specific importance for sewer biofilms as is also the case for treatment processes of wastewater in biofilters. The saturation constant, KS, is normally insignificant compared with the substrate concentration, and the biofilm kinetics [cf. Equation (2.20)], is therefore 0-order. As shown in Figure 2.8, two different conditions exist: the biofilm is either fully penetrated or partly penetrated, corresponding to either a fully effective or a partly effective biofilm. The distinction between these two situations can be expressed by means of a dimensionless constant, , called the penetration ratio (Harremoës, 1978). For each of these two situations, the flux of substrate across the biofilm surface can neglect the stagnant liquid film being calculated [Equations (2.23) and (2.25)]: Fully penetrated biofilm: 2 D f Sw k0 f L2f

ra

1

(2.22) (2.23)

k0 f L f

Partly penetrated biofilm: 2 D f Sw k0 f L2f

ra

2 D f k0 f Sw0.5

1

k1/ 2 Sw0.5

(2.24)

(2.25)

where Sw substrate concentration in the bulk water phase (g m 3) ra biofilm surface flux (g m 2 s 1, g m 2 h 1 or g m 2 d 1) k1/2 1/2-order rate constant per unit biofilm surface area (g0.5 m 0.5 s 1, g0.5 m 0.5 h 1 or g0.5 m 0.5 d 1) Lf biofilm thickness (m) Equations (2.23) and (2.25), for ra, are interesting because they relate the 0-order reaction in the biofilm to the substrate conditions in the bulk water phase. As seen from these two equations, a fully penetrated biofilm is of 0-order with respect to the bulk water phase, and a partly penetrated biofilm is of

Chemical Kinetics in a Microbiological System

1/2-order. Sewer biofilms are typically relatively thick with a high rate. These biofilms are, therefore, normally partly penetrated by substrate and of 1/2-order with respect to the substrate conditions in the wastewater phase. Figure 2.9 shows the theoretical transition from 1/2-order to 0-order kinetics in a biofilm. The redox reactions taking place in a sewer biofilm require that diffusion of both electron donor and electron acceptor be considered. The steady state mass balance for these two components is [cf. Equation (2.20)]: D f ,ox

d 2 Sox dx 2

k0 f ,ox

D f ,red

d 2 Sred dx 2

k0 f ,red

Sox KS ,ox

Sred Sox KS ,red

Sred

Sox Sred K S ,ox Sox KS ,red Sred

(2.26)

(2.27)

These second-order nonlinear differential equations have no explicit solution but can be solved numerically. The limiting substrate for the biofilm transformations is the one that penetrates the shortest distance into the biofilm. Equations (2.26) and (2.27) are, thereby, reduced to an equation corresponding to Equation (2.20). If the limiting substrate cannot be identified, approximations based on Equation (2.25) can be developed.

Hydrolysis is an enzymatic process by which complex organic compounds that cannot directly be used as substrates by the bacterial biomass can be broken down into simple molecules (see Section 3.2.3). The complex hydrolyzable

substrate, XS, may thereby be converted into soluble compounds, SS. These molecules can be transported from the bulk water phase across the bacterial cell wall and into the cell where the substrate utilization processes proceed. A simple — and generally accepted — approach of a kinetics for the hydrolysis is a 1-order description (Henze et al., 1995): dXS dt

k H XS

(2.28)

where XS kH

hydrolyzable substrate (gCOD m 3) 1-order hydrolysis constant (d 1)

However, in some situations, it can be assumed that the particulate substrate is completely covered with bacteria that excrete the exoenzymes. As it is assumed that enzymes are excessively present, the hydrolysis rate is constant per unit area available for hydrolysis, i.e., a surface-based kinetics model: dMS dt

kA A

(2.29)

where MS kA A

mass of substrate (gCOD) surface-based hydrolysis constant (gCOD m 2 d 1) surface available for hydrolysis (m2)

A combination of the concepts behind Equations (2.28) and (2.29) is applied in the activated sludge model for the kinetics of the hydrolysis processes (Henze et al., 1987). This combined concept, originally proposed by Dold et al. (1980), includes a saturation type of expression and a heterotrophic biomass with a maximum capacity for hydrolysis: dXS dt

kn

KX

XS / X Bw X Bw XS / X Bw

where kn hydrolysis rate constant (d 1) XBw heterotrophic active biomass in the water phase (gCOD m 3) KX saturation constant for hydrolysis ( )

(2.30)

Chemical Kinetics in a Microbiological System

Equation (2.30) describes the two extremes where either availability of hydrolyzable substrate or biomass may limit the hydrolysis. If hydrolyzable substrate is available in excess, i.e., Xs > XBw, the biomass density, XBw, is considered proportional to the enzymatic activity, and a situation equivalent to Equation (2.29) exists, i.e., the rate of hydrolysis is 1-order with respect to XBw. Contrarily, if XS 1–0.5 m (Figure 3.5). The classification of wastewater in terms of size distribution is normally done from a practical point of view. Typically, a distinction is made between soluble, colloidal and suspended components (Figure 3.6). While this definition for determining what “solids” are is rational as far as physical transport processes in sewers are concerned, when dealing with the microbial processes for sewer conditions, an extension of the “solids” definition is required. Particles larger than about 10 4 m cannot be transported through the cell wall and are, therefore, from a microbial point of view, considered particles. In order to avoid confusion, how the size distribution is considered, e.g., from either a physical or a physiological point of view, must be clearly distinguished.

Wastewater is a mixture of very complex organic and inorganic compounds. The traditional way of characterizing wastewater components can be found in any textbook dealing with treatment of wastewater, e.g., Metcalf and Eddy, Inc. (1991) and Henze et al. (1995b). The organic fractions and some inorganic compounds, mainly oxygen, nitrate, ammonia, sulfate and sulfide, are of specific importance when microbial processes are dealt with in sewers. In addition, settleable solids, heavy metals and organic micropollutants may give specific impacts.

Microbial Reactions and Quality of Substrate

The traditional way of characterizing organic matter in domestic wastewater is in terms of bulk parameters like BOD, COD and TOC. Wastewater characterization by direct measurement of organic constituents has been performed in only a few studies (Nielsen et al., 1992). In these studies, the main components, relevant from a biochemical point of view, have been identified as substances originating from foodstuff: • carbohydrates • proteins • lipids (fats) In addition to these three main groups of organic components, volatile fatty acids (VFAs), amino acids, detergents, humic substances, organic fibers, etc., have been found. Raunkjaer et al. (1994) have modified methods for the determination of carbohydrates, proteins and lipids to be applied for wastewater. Henze et al. (1995b) estimate the average composition of these three organic fractions in wastewater and propose a corresponding conversion to COD units (Table 3.3). The table also includes stoichiometric formulas for carbohydrate and protein, proposed by Kalyuzhnyi et al. (2000). It should be noticed that the formulas shown in Table 3.3 give different compositions of the organic fractions. As an example, the nitrogen content of protein according to the formulas proposed by Henze et al. (1995b) and Kalyuzhnyi et al. (2000) is 8.8% and 16.7%, respectively. Based on the conversion factors given in Table 3.3, the composition of wastewater samples taken at the inlets of four wastewater treatment plants in Denmark is shown in Figure 3.7. The wastewater in the corresponding sewer catchments mainly originates from domestic sources, and the network mainly consists of gravity sewer sections. The results show that the three components — carbohydrates, proteins and lipids — make up a significant part of the

Microbial Reactions and Quality of Substrate

organic matter. Furthermore, the organic residue may include intermediate products from the degradation of these components. Average dissolved fractions of carbohydrates and proteins are shown in Table 3.4. Protein only exists in a dissolved form after transport in a sewer network, whereas lipids, per definition, are nondissolved. The biochemical processes are closely related to the specific nature of the different organic substrates. Although each of the fractions of organic matter, carbohydrates, proteins and lipids include a great number of specific molecules, they are chemically related and have common characteristics as substrate for heterotrophic organisms. The overview given in Figure 3.8 is absolutely simplified; however, it links together and shows the major pathways of the substrate. A different behavior of the three groups of organic matter is therefore expected under transport in a sewer network. The transformation of carbohydrates, proteins and lipids has been investigated during transport in an intercepting gravity sewer under aerobic conditions. It was seen that dissolved carbohydrates and, to some extent, proteins were removed, whereas the concentration of lipids was almost unchanged (Figure 3.9).

When wastewater components are applied as model parameters for simulation of the transformations taking place in a sewer, the fundamental requirements depend on model objectives. A basic requirement (following the concept depicted in Figure 2.2) is that the following two types of components be included when focusing on the heterotrophic microbial transformations: • active, heterotrophic biomass including cell biomass and extracellular polymer substances (EPS) • substrates that are readily biodegradable and hydrolyzable

Carbohydrates Proteins Lipids

C10H18O9 C14H12O7N2 C8H6O2

C6H12O6 C4H6ON

1.13 1.20 2.03

Microbial Reactions and Quality of Substrate

Average (n 13 16) Standard deviation

18

10

28

28

31

78

6

7

4

6

10

11

These components can be expressed in units of COD, i.e., gO2 consumed as mass and gO2 m 3 as concentration. This type of description is well known from modeling activated sludge processes in wastewater treatment plants (Henze et al., 1987, 1995a, 2000). Taking the Activated Sludge Model No. 2 (ASM2) as a basis, the following components are included (Henze et al., 1995a): SF SA SS XS XAUT XPHA XPAO XI SI XBw

readily (fermentable) biodegradable substrate volatile acids/fermentation products readily biodegradable substrate (SS SF + SA) slowly biodegradable substrate autotrophic, nitrifying biomass stored polyhydroxyalkanoate phosphorus-accumulating organisms inert, nonbiodegradable, particulate organics inert, nonbiodegradable, soluble organics heterotrophic biomass

For wastewater, as it occurs in the influent to wastewater treatment plants, the autotrophic biomass, and the phosphate-accumulating biomass, including the stored polyphosphate, is expected to be close to zero (Henze et al., 1995a). This corresponds to the fact that the conditions for growth of nitrifying biomass and phosphorus-accumulating organisms in sewer systems are far from being optimal. Therefore, the following is approximately correct when focusing on wastewater in sewer networks: XAUT + XPHA + XPAO = 0

The sum of the following components, considering SS mately equals the total COD:

(3.4) SF + SA, approxi-

Microbial Reactions and Quality of Substrate

SF + S A + XS + XI + SI + X Bw = CODtot

(3.5)

The components referred to in Equation (3.5) are determined according to the activated sludge concept relevant at the influent to wastewater treatment plants. Basically, they are also present in the wastewater of sewer systems. However, when considering sewer processes, a slightly different approach compared with the activated sludge concept is needed. Details, in this respect, will be given in Chapters 5 and 6. Briefly, the explanation is as follows: • The biodegradability within a period of time corresponding to the residence time in a sewer network (typically between a few hours and one day) must be reasonably detailed. Therefore, the fast biodegradable fractions considered as SS and fast hydrolyzable substrate must be included as separate fractions. On the contrary, what is not biodegradable within 1–2 days is of minor interest. As a consequence, there is no need to distinguish between a rather slowly biodegradable, particulate fraction of a substrate and a fraction that is inert, whatever it is — soluble or particulate. • It is fundamentally important that the different COD fractions in wastewater be quantified and determined by direct measurement methods. The number of fractions must be minimized, determined by the details desirable and required, for example, for modeling purposes. Based on these criteria, the following fractions are found to be relevant in the case of heterotrophic wastewater processes in sewer systems (HvitvedJacobsen et al., 1998, 1999): SF SA SS XSn XBw

fermentable, readily biodegradable substrate fermentation products readily biodegradable substrate (SS SF + SA) hydrolyzable substrate, fraction n; n typically equal to 2 or 3: n 1 (fast degradable), n 2 (slowly degradable); n 1 (fast degradable), n 2 (medium degradable), n 3 (slowly degradable) heterotrophic active biomass

The mass balance according to this concept is, therefore, as follows: SF + S A + XSn + X Bw = CODtot

(3.6)

The number of fractions of hydrolyzable substrate is determined by the quality of the wastewater, i.e., when fractions can be identified in terms of their

Microbial Reactions and Quality of Substrate

different rates of hydrolysis. Typically, two or three — often just two — fractions will be needed. Two fractions are generally sufficient when considering a low number of wastewater samples. However, when several inlets into a sewer network exist, the variability in the hydrolysis rate often requires three fractions of XSn. In addition to the COD-fractions given for the wastewater phase, corresponding parameters may be needed when including the biofilm and the sediments. For a rather simple modeling of heterotrophic processes in the biofilm, the heterotrophic active biomass, XBf , in units of gCOD m 2, may be required. Depending on the details in the description of the anaerobic processes in the biofilm and the sediments, a sulfate-reducing and methane-producing biomass may be introduced. Figure 3.10 shows a “typical” composition of wastewater as COD-fractions when applied in the ASM2 model and the corresponding values, exemplified with two fractions, for the characterization of wastewater in sewer networks. The word “typical” must, of course, be associated with a rather high variability, even for domestic wastewater. Although the values corresponding to the ASM2 concept reflect quality characteristics of a primary effluent from municipal wastewater treatment, the two sets of characteristics are considered rather identical types of wastewater and are, therefore, interesting to compare. It is logical that there are differences in the fractions, although the matrix, wastewater, is basically the same in the two situations. Different methods have been applied for its characterization, depending on the objective, either for simulation of the processes in activated sludge or for simulation of wastewater processes in sewers. It is, however, interesting to note that SA (VFAs) is estimated differently in the two cases, although it is determined by the same analytical method. In addition to what has been mentioned, it is interesting to compare the SS fraction defined by the ASM2 methodology with the SS and XS1 fractions determined for characterization of wastewater in sewers. Different oxygen uptake rate (OUR) methods for characterization of these fractions are applied (cf., Chapter 7). The example in Figure 3.11 that is relevant for characterization of wastewater in sewers shows a measured OUR versus time curve. The course of this curve reflects that the biomass activity depends on the quality characteristics of the wastewater sample. It briefly illustrates what is typical: after a few hours (5–6), the readily biodegradable fraction, SS, is depleted, and a fraction, XS1, considered as fast hydrolyzable, is used up after about 14 hours. As seen from Figure 3.10, the SS and XS fractions appear to equal the SS fraction determined by the method used in the ASM2 concept. In Chapter 7, it will be shown in detail how these COD-fractions in raw wastewater can be determined and how they, as substrate for the biomass, influence its activity and correspond with the rate of change of oxygen consumption.

The kinetics of biofilms is dealt with in Section 2.2.2. In the following, specific characteristics related to the processes will be considered. All types of sewer biofilms are produced at surfaces exposed to the water phase and also, to some extent, at the sewer air surfaces where aerosols are present and the humidity is high. Biofilms in sewers are often referred to as “slimes” and consist mainly of microorganisms, extracellular polymeric substances (EPS) produced by the microorganisms and adsorbed organic and inorganic compounds from the wastewater.

Microbial Reactions and Quality of Substrate

Aerobic biofilms in gravity sewers usually have a thickness of a few mm, depending mainly on the flow conditions. The anaerobic biofilm thickness in pressure mains is usually lower (maximum of a few hundred m) due to a higher water velocity and, therefore, a higher shear stress during the pumping periods and a lower bacterial growth rate. In each case, however, examples of very thick biofilms may be found where a significant increase in the friction factor has led to different hydraulic conditions for the biofilm (Characklis and Marshall, 1990; USEPA, 1985). The thin biofilms in pressure mains typically have a rather smooth surface, whereas the biofilms in gravity sewers are fluffy. The real biofilm surface in a typical gravity sewer is much larger than what may be calculated based on the inner surface of a sewer pipe. The microorganisms in sewer biofilms are embedded in a matrix of EPS that consists mostly of polysaccharides produced by the bacteria (Characklis and Marshall, 1990). The EPS fraction is the largest organic fraction in the biofilm, i.e., up to about 90% of the total organic content. Only limited studies of the total composition of sewer biofilms in terms of carbohydrates, proteins and humic substances have been undertaken (Figure 3.12). Corresponding information on the composition of anaerobic biofilms in pressure mains does not yet exist. Significant biomass production can take place in a gravity sewer biofilm. The biomass generated in the biofilm detaches and is, together with the biomass produced in the water phase, transported to the treatment plant or via overflow structures into receiving waters. A simple method to assess the amount of

biomass produced is to make an estimate from the consumption of the electron acceptor (O2, NO3 and SO24 ) and the yield constant of the process as shown in Example 3.1. Example 3.1: Biomass production in the biofilm of an intercepting sewer A 4 km intercepting sewer with diameter D 0.5 m is flowing half full. The DO consumption rate, rf , of the sewer biofilm is measured, and an average value of 0.6 gO2 m 2 h 1 was estimated. The biofilm yield constant of the heterotrophic biomass was not measured but was estimated as Yf 0.55 gCOD biomass produced per gCOD substrate consumed. Only aerobic heterotrophic transformations in the biofilm are expected to proceed. Calculate an estimate of a daily production of biomass in the biofilm of the sewer. The total amount of substrate that undergoes transformation is either used for growth of biomass or degraded by respiration (cf. Section 2.1.1). The rate of production of biomass is, therefore, as follows: rB

Yf 1 Yf

0.6 gCOD m 2 h

0.55 0.6 1 0.55

1

0.73 gCOD m 2 h

1

And, the daily production of biomass in the biofilm of the sewer is as follows:

Microbial Reactions and Quality of Substrate

MB

rB

2

0.73

D 4000 24 10

3

kgCOD

0.5 4000 24 10 2 55.3 kgCOD

3

kgCOD

A prerequisite condition for the validity of this result is that only aerobic microbial processes proceed. If anaerobic processes take place in the inner part of the biofilm, a reduction in the biomass production will take place, for example, see comments related to Figure 6.2. The biomass produced in the biofilm is detached to the water phase by erosion and by sloughing. Single bacteria and small parts of the biofilm erode continuously from the surface, while bigger parts may slough intermittently. Sloughing may take place when large changes in shear stress or substrate conditions occur, e.g., during wet-weather periods. However, the mechanisms are not well understood. In a steady state situation or over longer time periods, the total detached biomass will be equal to biomass production. Particles may be trapped on the biofilm surface or in voids of the biofilm where any organics may be hydrolyzed and further take part in the transformation processes. A number of factors influence adsorption and desorption of particles, such as particle size, surface charge, pH, etc., as well as biofilm surface properties and bulk water flow pattern. Studies of model biofilms have shown that water flows into the biofilm in small channels, making the prediction of transport of particles as well as soluble compounds complex (Norsker et al., 1995). An exchange of heavy metals between the bulk water phase and the biofilm takes place (Gutekunst, 1988). The concentrations of heavy metals in the biofilm may be considered indicative of the preceding wastewater transport of heavy metals. Heavy metals originating from short-term increases in concentration in the bulk water phase may be trapped in the biofilm and then be slowly released.

The occurrence of sewer sediments is primarily determined by the physical characteristics of wastewater solids and the hydraulic conditions. Basically, sewers should be designed and operated in a way that does not result in permanent deposits. This ideal performance of a sewer is not generally observed, and sediments may be more or less temporarily accumulated in sewers. In combined sewer networks, sediments may settle under dry-weather conditions when the wastewater velocity and shear stress at the bottom are low and be

eroded and transported in the suspended water phase under wet-weather conditions. The solids originating from the sewer sediments may, therefore, turn up in the combined sewer overflows (CSOs) into the receiving waters.

As briefly mentioned, the occurrence of sediments in sewers is closely related to the hydraulics and physical process characteristics and the properties of the solids. This text does not include a quantitative description of the physical processes in sewers, such as sedimentation, deposition and erosion, and the corresponding physical characteristics of sewer solids. A great number of publications and textbooks address these subjects. A comprehensive description with a broad range of literature references can be found in Ashley and Verbanck (1998). Other important publications in this respect are found in Ashley (1996), Hvitved-Jacobsen et al. (1995) and Hvitved-Jacobsen (1998). Perspectives and relations to the chemical and biological sewer processes are outlined in Ashley et al. (1999). The physical characteristics of sewer deposits can be described in terms of individual particle and bulk properties. The hydraulic and structural conditions in the sewer, together with the nature of the inputs, will control the type of material that deposits at a given location. Crabtree (1989) has proposed a sewer sediment taxonomy that is relevant mainly in terms of physical properties but also to chemical and biological processes (Table 3.5). The taxonomy is based on four primary classes with a fifth class B comprising agglutinated or cemented class A material. As seen from Table 3.5, organic matter constitutes an essential part of sewer sediments, however, generally with a low biodegradability. Class D (sewer biofilm) is included in the taxonomy (Section 3.2.7). Class A sewer sediment material is most commonly found in combined sewer networks.

Table 3.6 shows some selected chemical characteristics for class A sediments. Numerous studies have been collated to demonstrate the variability.

Only a few studies have been directly concerned with chemical and biological processes in sewer sediments. However, relatively high anaerobic activity in terms of H2S formation of sediment deposits compared with what is generally observed in sewer biofilms is observed (Section 6.2.5). This activity may indicate H2S formation in the deep parts of the sediment caused by the production

D E

A C

Coarse, granular bed material — widespread Mobile, fine-grained, found in slack zones, in isolation and overlying type A Organic pipe wall slimes Fine-grained mineral and organic material found in CSO storage tanks 1.21 1.46

1.72 1.17 17-32-52 1-22-80

1-6-30 29-45-73

1-62-83 1-69-85

3-61-87 5-55-71

1-6-20 4-9-80

3-33-90 0

61 22

7 50

Microbial Reactions and Quality of Substrate

TS, total solids (g kg 1) VS, volatile solids (%) COD (g kg 1)* BOD5 (g kg 1)* BOD, 4 hours (mg kg 1)* Organic N (mg kg 1) Ammonia, NH4-N (mg kg 1)

550–800 4.5–10 25–70 4–14 400 800 100

350–820 1–19 6–270 1–90 100–700 200–1500 10–300

*Dry weight.

of readily biodegradable organic matter by anaerobic hydrolysis and fermentation and the effects of such organic products being transported by diffusion into the upper layer of the sediment. It has also been shown that a biofilm may fast develop at the surface of sewer sediments and influence the cohesion of the sediment surface, thereby influencing the resistance to resuspension (Vollertsen and Hvitved-Jacobsen, 2000). In the same study, it was observed that methanogenesis (methane formation) in the sediments caused the formation of gas cavities that decreased the strength of the sediment surface. Crabtree (1986) postulated the existence of an aging process of sewer deposits caused by interactions between sediments and wastewater. Ristenpart (1995) investigated the occurrence of this aging in terms of the variability of specific components in sediments of different ages (Table 3.7). The differences in the magnitudes of the parameters shown in Table 3.7 correspond well with the course of (anaerobic) microbial processes in the sewer deposits. Newly deposited sediments not only have the highest pollutant potential but also show the lowest critical shear stress for erosion. Such sediment types may exert the highest impacts on receiving waters from combined sewer overflows.

Bulk density (kg m 3) TS, total solids (g kg 1) VS, volatile solids (% of dry matter) pH ( ) BOD5 (g kg 1 wet weight) COD (g kg 1 wet weight)

1200 355 27.0 5.68 31.6 95.6

1510 705 8.8 7.11 12.5 55.3

1840 812 2.4 7.66 2.7 19.0

References

Ashley, R.M. (ed.) (1996), Solids in sewers, Water Sci. Tech., 33(9), 298. Ashley, R.M. and M.A. Verbanck (1998), Physical processes in sewers, Proceedings from Congress on Water Management in Conurbations, Bottrop, Germany, June 19–20, 1997, pp. 26–47. Ashley, R.M., T. Hvitved-Jacobsen, and J.-L. Bertrand-Krajewski (1999), Quo vadis sewer process modelling? Water Sci. Tech., 39(9), 9–22. Benefield, L.D. and C.W. Randall (1980), Biological Process Design for Wastewater Treatment, Prentice Hall, Englewood Cliffs, NJ, p. 526. Bjerre, H.L., T. Hvitved-Jacobsen, S. Schlegel, and B. Teichgräber (1998), Biological activity of biofilm and sediment in the Emscher river, Germany, Water Sci. Tech., 37(1), 9–16. Boon, A.G. and A.R. Lister (1975), Formation of sulphide in rising main sewers and its prevention by injection of oxygen, Prog. Water Tech., 7 (2), 289–300. Characklis, W.G. and K.C. Marshall (eds.) (1990), Biofilms, John Wiley & Sons, New York. Christ, O., P.A. Wilderer, R. Angerhöfer, and M. Faulstich (2000), Mathematical modeling of the hydrolysis of anaerobic processes, Water Sci. Tech., 41(3), 61–65. Crabtree, R.W. (1986), The discharge of toxic sulphides from storm sewage overflows — a potential polluting process, WRC (Water Research Centre) report ER 203E. Crabtree, R.W. (1989), Sediments in sewers, J. IWEM (Institution of Water and Environmental Management), 3(6), 569–578. Gutekunst, B. (1988), Sielhautuntersuchungen zur Einkreisung schwermetalhaltiger Einleitungen, Institut für Siedlungswasserwirtschaft, Universität Karlsruhe, Band 49. Henze, M., W. Gujer, T. Mino, and M. v. Loosdrecht (2000), Activated sludge models ASM1, ASM2, ASM2d and ASM3, Scientific and Technical Report No. 9, IWA (International Water Association), p. 121. Henze, M., P. Harremoës, J. la Cour Jansen, and E. Arvin (1995b), Wastewater Treatment — Biological and Chemical Processes, Springer-Verlag, New York, p. 383. Henze, M., C.P.L. Grady Jr., W. Gujer, G. v. R. Marais, and T. Matsuo (1987), Activated sludge model no. 1, Scientific and Technical Report No.1, IAWPRC (International Association on Water Pollution Research and Control). Henze, M., W. Gujer, T. Mino, T. Matsuo, M.C. Wentzel, and G. v. R. Marais (1995a), Activated sludge model No. 2, Scientific and Technical Report No. 3, IAWQ (International Association on Water Quality), p. 32. Hvitved-Jacobsen, T. (ed.) (1998), The sewer as a physical, chemical and biological reactor II, Water Sci. Tech., 37(1), 357. Hvitved-Jacobsen, T. and J. Vollertsen (1998), An intercepting sewer from Dortmund to Dinslaken, Germany — prediction of wastewater transformations during transport, Report submitted to the Emschergenossenschaft, p. 35. Hvitved-Jacobsen, T., J. Vollertsen, and P.H. Nielsen (1998), A process and model concept for microbial wastewater transformations in gravity sewers, Water Sci. Tech., 37(1), 233–241. Hvitved-Jacobsen, T., J. Vollertsen, and N. Tanaka (1999), Wastewater quality changes during transport in sewers — An integrated aerobic and anaerobic model concept for carbon and sulfur microbial transformations, Water Sci. Tech., 39(2), 242–249. Hvitved-Jacobsen, T., P.H. Nielsen, T. Larsen and N. Aa. Jensen (eds.) (1995) , The sewer as a physical, chemical and biological reactor I, Water Sci. Tech., 31(7), 330.

Jahn, A. and P.H. Nielsen (1998), Cell biomass and exopolymer composition in sewer biofilms, Water Sci. Tech., 37(1), 17–24. Kalyuzhnyi, S., A. Veeken, and B. Hamelers (2000), Two-particle model of anaerobic solid state fermentation, Water Sci. Tech., 41(3), 43–50. Levine, A.D., G. Tchobanoglous, and T. Asano (1985), Characterization of size distribution of contaminants in wastewater; treatment and reuse implications, J. WPCF, 57, 805. Levine, A.D., G. Tchobanoglous, and T. Asano (1991), Size distributions of particulate contaminants in wastewater and their impact on treatability. Water Res., 25, 911. Logan, B.E. and Q. Jiang (1990), Molecular size distributions of dissolved organic matter, J. Env. Eng., 116, 1046. Matos, J.S. and E.R. de Sousa (1996), Prediction of dissolved oxygen concentration along sanitary sewers, Water Sci. Tech., 34(5–6), 525–532. Metcalf and Eddy, Inc. (1991), Wastewater Engineering — Treatment, Disposal and Reuse, 3rd edition revised by G. Tchobanoglous and F. L. Burton, McGraw-Hill, Inc., New York, p. 1334. Nielsen, P.H. and T. Hvitved-Jacobsen (1988), Effect of sulfate and organic matter on the hydrogen sulfide formation in biofilms of filled sanitary sewers, J. Water Pol. Contr. Fed., 60(5), 627–634. Nielsen, P.H., K. Raunkjaer, N.H. Norsker, N.Aa. Jensen, and T. Hvitved-Jacobsen (1992), Transformation of wastewater in sewer systems — A review, Water Sci. Tech., 25(6), 17–31. Norsker, N.-H., P.H. Nielsen, and T. Hvitved-Jacobsen (1995), Influence of oxygen on biofilm growth and potential sulfate reduction in gravity sewer biofilm, Water Sci. Tech., 31(7), 159–167. Raunkjaer, K., T. Hvitved-Jacobsen, and P.H. Nielsen (1994), Measurement of pools of protein, carbohydrate and lipid in domestic wastewater, Water Res., 28(2), 251–262. Raunkjaer, K., T. Hvitved-Jacobsen, and P.H. Nielsen (1995), Transformation of organic matter in a gravity sewer, Water Env. Res., 67(2), 181–188. Ristenpart, E. (1995), Sediment properties and their changes in a sewer, Water Sci. Tech., 31(7), 77–83. Standard Methods for the Examination of Water and Wastewater (1998), 20th Edition, APHA, AWWA and WEF. Stanier, R.Y., J.L. Ingraham, M.L. Wheels, and P.R. Painter (1986), The Microbial World, Prentice Hall, Englewood Cliffs, NJ. Tanaka, N. (1998), Aerobic/anaerobic process transition and interactions in sewers, Ph.D. dissertation, Environmental Engineering Laboratory, Aalborg University, Denmark. Tanaka, N. and T. Hvitved-Jacobsen (1999), Anaerobic transformations of wastewater organic matter under sewer conditions. In: I.B. Joliffe and J.E. Ball (eds.), Proceedings of the 8th International Conference on Urban Storm Drainage, Sydney, Australia, August 30–September 3, 1999, pp. 288–296. USEPA (1985), Odor and corrosion control in sanitary sewerage systems and treatment plants, U.S. Environmental Protection Agency, EPA 625/1-85/018, Washington D.C. Vollertsen, J. and T. Hvitved-Jacobsen (2000), Resuspension and oxygen uptake of sediments in combined sewers, Urban Water, 2(1), 21–27. WEF (1994), Wastewater biology: The life processes, a special publication prepared by Task Force on Wastewater Biology: The Life Processes, chaired by M. H. Gerardi, Water Environment Federation (WEF), p. 184.

Air–Water Equilibrium and Mass Transfer — Odors and Reaeration in Sewers

Most sanitary and combined sewer networks consist of pipes designed to flow as open channels, i.e., with a free water surface. The wastewater flows downstream in such pipes by the force of gravity with a velocity of flow that depends principally on the pipe slope and frictional resistance. Typically, the design velocity is between 0.6 and 3 m s 1 to avoid blockage of the pipe by sewer solids accumulated at low flow conditions and to prevent damage of the sewer at a high flow. Under such conditions, exchange of substances between the sewer atmosphere and the wastewater phase may depend on and affect sewer processes. The following two phenomena are, to a great extent, affected by the equilibrium conditions and the mass transfer that takes place across the air–water interface in a sewer. • Odors in sewer networks: odor formation in sewer networks takes place under anaerobic conditions. Although such conditions typically exist in pressure mains and full-flowing gravity sewers, downstream-located, partly filled pipes and sewer structures like pumping stations and manholes with a free water surface may give rise to emission of odorous compounds. Hydrogen sulfide is a well-known example, but a great number of volatile organic compounds (VOCs) produced by fermentation processes also result in malodors. • Reaeration in sewer networks: the presence of dissolved oxygen in wastewater of sewer systems determines if, and to what extent, aerobic and anaerobic processes proceed. The air–water oxygen transfer (the reaeration) determines the potential of aerobic transformation and corresponding removal of wastewater components in many sewer

networks. The reaeration in a sewer is, therefore, a fundamental process affecting the microbial processes.

A first approach in this respect is described by the distribution coefficient KA for a more or less volatile compound, A, between the gas phase and the water phase: KA

yA xA

(4.1)

where KA yA xA

distribution coefficient or partition coefficient ( ) mole fraction of A in the gas phase [moles (total moles) 1] mole fraction of A in the water phase [moles (total moles) 1]

Equation (4.1) expresses that the ratio of the concentrations of A in the gas phase and the water phase, respectively, is a constant at equilibrium. This constant is temperature dependent but is independent of the quantity of A as long as dilute solutions are dealt with.

The concept of mole fraction of a component used in Equation (4.1) is a convenient measure of concentration when dealing with trace quantities and dilute solutions, often experienced in environmental systems. This is especially the case with transport phenomena and equilibrium between phases, where it results in simple quantitative expressions. The phenomena of interest when dealing with the exchange of odorous compounds and oxygen between wastewater and a sewer atmosphere are, in this respect, relevant examples. The fundamentals of mole fraction can be illustrated for a binary system consisting of two components, A and B. The mole fraction of A (considering a gas phase) is defined as follows: yA

NA ( N A NB )

(4.2)

Air–Water Equilibrium Conditions

where NA NB

moles of A moles of B

The mass balance (in the actual case for the gas phase) is, therefore, as follows: yA

yB

1

(4.3)

For trace quantities of A in air, yA can be expressed corresponding to a situation where 1 “mole” of air (mainly consisting of N2, O2, Ar and CO2) has a “molar volume” of approximately 22.4 l mole 1 at 0 C and 1 atm and a “molar weight” of 29 g mole 1: yA

c1A 1/ 22.4

c1A 0.0446

(4.4)

where c1A is the molar concentration of component A in air (moles L 1). Correspondingly, for dilute aqueous solutions, 1 mole of water equals 18 g, i.e., xA can be expressed as follows: xA

c2 A 1000 /18

c2 A 55.56

(4.5)

where c2A is the molar concentration of component A in water (moles L 1).

The relative volatility, , is a constant that under equilibrium conditions can be used to express the distribution of a volatile compound between a gas phase made of A and water vapor and a water phase containing A. This constant is for a component A defined as follows: A

y A / ywater x A / xwater

(4.6)

where A

ywater xwater

relative volatility ( ) mole fraction of water vapor in the gas phase [moles (total moles) 1] mole fraction of water in the water phase [moles (total moles) 1]

For a dilute solution of water, which is a reasonable approximation for normal wastewater, xwater is approximately equal to 1, and Equation (4.6) is therefore reduced to the following: A

y A / ywater xA

(4.7)

The most widely used and still simple theoretical approach for describing a gas-liquid equilibrium for a volatile compound A is expressed by Henry’s law: pA

yA P

(4.8)

HA xA

where pA partial pressure of a component A in the gas phase (atm) P total pressure (atm) HA Henry’s law constant for A [atm (mole fraction) 1] Under equilibrium conditions and at constant temperature, Henry’s law defines the relative amount of a volatile compound in the gas phase as a function of the relative concentration in the water phase., i.e., Henry’s law quantifies the degree of tendency of a volatile compound to escape from the liquid phase. The law applies to dilute solutions of A, i.e. for solutions where xA is close to 0 provided that A does not dissociate or react in the water phase. Table 4.1 gives some examples of Henry’s law constants and boiling points for selected odorous and nonodorous compounds. The list includes values for hydrocarbons that frequently appear in sewer networks from, for example, industrial sources and street runoff. Example 4.1: Solubility of oxygen in water The partial pressure of oxygen is measured with a gas detector in the atmosphere of a sewer network and found to be equal to 0.18 atm. This value is, probably because of oxygen consumption of the wastewater and limited ventilation, slightly lower than in the city atmosphere, where it was measured at 0.21 atm. Determine at 25 C the solubility of O2 (i.e., at equilibrium conditions) in the wastewater (considered as water) of the sewer. From Equation (4.8) and Table 4.1: xO2

pO2 HO2

0.18 43800

4.110

6

Air–Water Equilibrium Conditions

Volatile sulfur compounds (VSCs)

Nitrogenous compounds

Acids (VFAs)

Aldehydes and ketones

Inorganic gases Selected nonodorous compounds

Hydrocarbons

Methyl mercaptan Ethyl mercaptan Allyl mercaptan Benzyl mercaptan Dimethyl sulfide Dimethyl disulfide Methylamine Ethylamine Dimethylamine Pyridine Indole Scatole Formic Acetic Propionic Butyric Valeric Acetaldehyde Butyraldehyde Acetone Butanone Hydrogen sulfide, H2S Ammonia, NH3 Nitrogen, N2 Oxygen, O2 Carbon dioxide, CO2 Methane, CH4 Pentane, C5H12 Hexane, C6H14 Heptane, C7H16 Octane, C8H18

6 35 69 195 37 110 6.4 17 7 115 254 265 100 118 141 162 185 21 76 56 80 59.6 33.4 195.8 183 78.5 161.5 36 69 98 125

200 200 110 63 0.55 0.55 1.3 0.5

0.010 0.011 0.009 0.012 0.026 5.88 6.3 1.9 2.8 563 0.834 86,500 43,800 1,640 40,200 70,400 80,500 46,900 19,400

Henry’s constant, HA, is temperature dependent with a positive dHA/dT value. Each compound has its specific dependence. These values may be found in Sander (2000) for a reciprocal Henry’s law constant, kHo 56.29 / H A cf. Example 4.2.

Equation (4.5) converts mole fraction to moles L 1: cO2

55.56 xO2

55.56 4.110

6

2.2810

4

moles L 1

And further transformation into units of g m 3, knowing that the molar weight of O2 is 32 g mole 1: cO2

2.2810

4

32103

7.3 mg O2 L 1

7.3 g O2 m

3

As previously mentioned, the simple equilibrium approach requires that the relevant volatile compounds exist in a nondissociated molecular form in the water phase. For several substances, this is not the case. Hydrogen sulfide is, as an odorous compound, an important example, with sulfide chemical species related according to the following equilibrium: water-air transfer

pKa1 7.0

H2 S (g)

H2 S (aq)

(gas)

(aqueous)

pKa 2 14

HS (ion)

S2

(4.9)

(ion)

where the equilibrium constants, Ka1 and Ka2, determine the ratio between the concentrations, C, at equilibrium: K a1

K a2

CH CHS CH2S (aq) CH CS2 CHS

(4.10)

(4.11)

The release to the atmosphere is strongly dependent on the pH because only the molecular form and not the dissociated forms can be emitted, e.g., at a pH about 7, an equal amount of H2S and HS exists in the water phase. Increase of the pH will, therefore, at equilibrium conditions and at a constant total sulfide concentration, reduce the hydrogen sulfide concentration in the overlying sewer atmosphere (Figure 4.1). Therefore, when applying Henry’s law [Equation (4.8)], only the nondissociated molecular form, H2S, should be taken into account. The nondissociated H2S (aq) form can be determined based on Equations (4.9) to (4.11). The sulfide ion, S2 , only exists in measurable amounts above a

Air-Water Equilibrium Conditions

pH of about 12. In the case of wastewater, only the equilibrium between H2S (aq) and HS in Equation (4.9) is normally relevant, and Equation (4.12) determines the ratio between CHS and CH2S (aq) at the actual pH: pH

pK s1 log

CHS CH2S (aq)

(4.12)

where pKs1 is 7.0 at 25 C. The curves shown in Figure 4.2 are the combined result of Henry’s law [Equation (4.8)] and the pH-dependent dissociation of H2S [Equation (4.12)] as briefly illustrated in Example 4.2. Although the curves are describing equilibrium conditions, they are essential for the evaluation of the potential risk for odor problems. Turbulence in the wastewater, e.g., caused by drops, and the degree of ventilation in the sewer are important for establishing the equilibrium conditions shown in Figure 4.2. Release of H2S into the atmosphere at pumping stations and hydraulic jumps may be rather high. Such conditions are described by Matos and de Sousa (1992), who have developed a model for prediction of hydrogen sulfide buildup in the atmosphere of a gravity sewer pipe (cf. Section 4.3.2). Example 4.2: Equilibrium distribution of sulfide between water and a gas phase Calculate at pH 7.0 and 15 C the concentration of H2S in ppm in a sewer atmosphere in equilibrium with wastewater with a total (H2S + HS )-concentration

of 1.0 gS m 3. It is considered to be approximately correct to estimate pKs1 7.0 at 15 C, whereas the temperature dependency of Henry’s constant must be considered. i.e., approximately at pH 7.0 according to Equation (4.12): CH2S (aq) 1.0 0.5

0.5 gS m

3

From Table 4.1, HH2S 563 atm (mole fraction) 1 at 25 C. The temperature dependency of kHo 56.29 / HH2S is found in Sander (2000) and HH2S at 15 C is thereby estimated to be 441 atm (mole fraction) 1, i.e., about 22% lower than at 25 C.† The mole fraction of H2S in the water phase is determined according to Equation (4.5), and a molecular weight of S equal to 32 g (mole) 1: xH2S

0.510 3 32 55.56

2.8110

7

†In addition to the equation given by Sander (2000) that was used for determination of the temperature dependency of Henry’s constant for H2S, HH2 S , the following equation proposed by Clarke and Glew (1971) may be used:

log HH2 S 104.069 4423.11T

1

where HH2 S is Henry’s constant [atm (mole fraction)

36.6296 log T 1]

0.01387T

and T is the temperature (K).

Air–Water Transport Processes

The partial pressure of H2S according to Equation (4.8) is as follows:

pH

2S

441 2.8110 7

123.910 6 atm 124 ppm

The small deviation between this value and what is shown in Figure 4.2 is due to the use of different constants. In addition to hydrogen sulfide used as an example, a number of other compounds in wastewater, for example, NH3 and VFAs, exist in both a molecular form and an ionic form. Therefore, conditions similar to those described for H2S exist for these substances.

A number of approaches to the air–water mass transport exist. In relation to transport processes in sewer networks, the main developments have been directed toward the air–water oxygen transfer. The following are the main theoretical descriptions that are relevant in this respect: • Two-film theory (Lewis and Whitman, 1924): the theory is based on molecular diffusion through two stagnant films, a liquid and a gas film, at the air–water interface. • Penetration theories (Higbie, 1935; Danckwerts, 1951; Dobbins, 1956): according to the penetration theory, diffusion of gases takes place into elements of water transported by turbulence to the surface. • Surface renewal theory (King, 1966): the theory describes the replacement of a surface liquid film by the action of eddies that move between the bulk water phase and the surface film. The surface renewal rate thereby determines the exchange between the surface and the bulk water. Further details on air–water mass transfer phenomena may be found in Thibodeaux (1996) and Stumm and Morgan (1981).

The two-film theory considering molecular diffusion through stagnant liquid and gas films is the traditional way of understanding mass transfer across the air–water boundary. As briefly described, other theories exist. However, the two-film theory gives an understanding of fundamental phenomena that may lead to simple empirical expressions for use in practice.

The transport process, according to the two-film theory, of a volatile component across the air–water interface is depicted in Figure 4.3. The figure illustrates a concept that concentration gradients in both phases exist and that the total resistance for mass transfer is the sum of the resistance in each phase. Although mass transfer across the water-air interface is difficult in terms of its application in a sewer system, it is important to understand the concept theoretically. The resistance to the transport of mass is mainly expected to reside in the thin water and gas layers located at the interface, i.e., the two films where the gradients are indicated (Figure 4.3). The resistance to the mass transfer in the interface itself is assumed to be negligible. From a theoretical point of view, equilibrium conditions exist at the interface. Because of this conceptual understanding of the transport across the air–water boundary, the theory for the mass transport is often referred to as “the two-film theory” (Lewis and Whitman, 1924). According to the two-film theory, it is appropriate to consider the transport of volatile components between the water phase and the air phase in two steps: from the bulk water phase to the interface and from the interface to the air, or vice versa. The driving force for the transfer of mass per unit surface area from the water phase to the interface and from the interface to the air phase is determined from the difference between the actual molar fractions, xA and yA, and the corresponding equilibrium values, x*A and y*A : k2 A ( x*A

JA

=

−

xA )

(4.13) (4.14)

Air–Water Transport Processes

where JA k1A k2A

flux rate of component A [moles (total moles) 1 s 1 m 2] gas phase mass transfer coefficient (s 1 m 2) water phase mass transfer coefficient (s 1 m 2)

Which of the Equations [(4.13) or (4.14)] is determined to be the most important depends on which part of the boundary has the major resistance to the mass transport. If, as an example, the major resistance exists in the water film of the boundary, i.e., k2A < k1A, Equation (4.13) is the relevant description of the flux rate. The two equilibrium values of the molar fraction, x A* and y A* are fictitious, however, each is determined from Henry’s law, i.e.: y*A

HA xA P

(4.15)

yA

HA * xA P

(4.16)

Equations (4.15) and (4.16) can be substituted in either Equation (4.13) or Equation (4.14). In the case of a dominating resistance in the water film, Equation (4.13) is reformulated by substitution of Equation (4.16): JA

k2 A x A

yA HA P

(4.17)

Each of the mass transfer coefficients k1A and k2A can be interpreted as a molecular diffusion coefficient, D, divided by a film thickness, z, for the gas phase and the water phase, respectively, i.e., k D/z. However, this interpretation has no meaning in practice because of the lack of knowledge on the thickness of the two films. General expressions for the air–water mass transfer can be derived by solving Equations (4.13) and (4.14), for x*A and y A* respectively, and substituting the results in both Equations (4.15) and (4.16). The following two expressions are obtained: JA

k1A k2 A k2 A P k1A HA

yA P HA

xA

KL

yA P HA

xA

(4.18)

JA

k1A k2 A HA x k1A HA A P k2A P

yA

KG x A

HA P

yA

(4.19)

where KG KL

overall mass transfer coefficient referring to the gas phase (s 1 m 2) overall mass transfer coefficient referring to the water phase (s 1 m 2)

From Equations (4.18) and (4.19), the following two expressions are derived: 1 KL

1 KG

1 k2A

1 k1A

P H A k1A

(4.20)

HA P k2A

(4.21)

The two pairs of equations, Equations (4.18), (4.19) and (4.20), (4.21) are equally valid, but typically, the two corresponding equations, Equations (4.18) and (4.20), are applied. Equation (4.20) expresses that the total resistance to mass transfer across the air–water boundary is equal to the sum of the resistances across the liquid film and the gas film. The importance of the magnitude of Henry’s constant is, in this respect, evident. For high values of HA, e.g., exemplified by O2, the resistance mainly exists in the water film, and turbulence in a sewer will, therefore, enhance the water-air transfer process. The importance of turbulence in the water phase is reduced for odorous components with a relatively low HA value, and turbulence in the air phase will correspondingly increase the release rate (Table 4.1). As seen from Equations (4.20) and (4.21), these facts also depend on the k1A/k2A ratio that varies according to system characteristics. Liss and Slater (1974) have, based on the value of HA, evaluated which type of mass transfer resistance exists. They propose the following criteria, valid for most systems (cf. Table 4.1): • Flow through the liquid film controls the mass transfer if HA > 250 atm (mole fraction) 1. • The resistance in both the water and the air film may be of importance if HA is between 1 and 250 atm (mole fraction) 1. • The flow conditions are controlled by the air film if HA < 1 atm (mole fraction) 1. This situation corresponds not only to compounds with a

Odorous Compounds in Sewer Networks

relatively low volatility but also to compounds that are reactive in the water phase, e.g., like NH3. As can be seen from Table 4.1, all three situations are relevant for odorous compounds. Concerning the air–water oxygen transfer (reaeration), resistance primarily exists in the liquid film. A major problem in the quantification of air–water transport phenomena in terms of the rate expression [Equation (4.18)] is to find appropriate values for KL. As far as sewer systems are concerned, the most well-established knowledge concerning air–water mass transfer is on reaeration (Section 4.4). Specific details related to odor formation and reaeration will be dealt with in Sections 4.3 and 4.4, respectively.

Anaerobic conditions in wastewater of sewer systems may give rise to formation of volatile substances that are typically identified by a number of problems, like malodors, health risks and corrosion. The conditions and processes related to such problems are dealt with in Section 3.2.2. Further information on measurement, modeling and control of odors can be found in Stuetz and Frechen (2001). The anaerobic microbial processes responsible for the formation of odor-causing substances produce inorganic gases and VOCs. The malodorous inorganic gases are primarily ammonia (NH3) and hydrogen sulfide (H2S). As depicted in Figure 3.3, a great number of VOCs can be produced. The most common organic substances associated with odors and produced from wastewater organic matter are shown in Table 4.2. A great number of organic compounds known as potential odor-causing substances have been identified in domestic wastewater (Raunkjaer et al., 1994; Hvitved-Jacobsen et al., 1995; Hwang et al., 1995). Generally, VFAs are known as anaerobic decomposition products of carbohydrates, e.g., starch. Mercaptans are primarily produced from proteins. Several of the compounds shown in Table 4.2 arise from the anaerobic decomposition of organic matter containing sulfur and nitrogen. Only a few studies have been concerned with measurements of specific odorous compounds in sewer systems. Hwang et al. (1995) have analyzed the influent wastewater in a study of malodorous substances in wastewater at different steps of sewage treatment (Table 4.3). Although the results in Table 4.3 only represent examples, they are interesting for several reasons. First of all, the table shows that several of the odorous compounds may appear in wastewater from a sewer system in relatively high

Volatile sulfur compounds (VSCs)

Nitrogenous compounds

Methyl mercaptan Ethyl mercaptan Allyl mercaptan Benzyl mercaptan Dimethyl sulfide Dimethyl disulfide Thiocresol Methylamine Ethylamine Dimethylamine

CH3SH C2H5SH CH2苷CHCH2SH C6H5CH2SH (CH3)2S CH3SSCH3 CH3C6H4SH CH3NH2 C2H5NH2 (CH3)2NH

Pyridine

Acids (VFAs)

Aldehydes and ketones

1 0.2 0.05 0.2 1 0.3–10 0.1 1–50 2400 20–80 4

Indole

1.5

Scatole

0.002–1

Acetic Butyric Valeric Formaldehyde Acetaldehyde Butyraldehyde Acetone Butanone

CH3COOH C2H5COOH C3H7COOH HCHO CH3CHO C2H5CHO CH3COCH3 C2H5COCH3

15 0.1–20 2–2600 370 0.005–2 5 4600 270

Odorous Compounds in Sewer Networks

Hydrogen sulfide Carbon disulfide Methyl mercaptan Dimethyl sulfide Dimethyl disulfide Dimethylamine Trimethylamine n-Propylamine Indole Skatole

23.9 0.8 148 10.6 52.9 210 78 33 570 700

15–38 0.2–1.7 11–322 3–27 30–79 — — — — —

concentrations, especially when compared with H2S. It should also be noticed that the concentrations are those observed in wastewater. What appears in the air phase may depend, for example, on Henry’s constant (cf. Sections 4.1 and 4.2). Thistlethwayte and Goleb (1972) reported investigations of the composition of sewer air. The main part of the samples was taken in a sewer transporting mixed municipal wastewater with a maximum residence time around 4 hours. BOD5 of the wastewater was ranging from 300–350 g m 3, and the temperature was typically about 24 C. The authors divided the components in the sewer air into four groups. These groupings were determined not only by the chemical nature of the components but also according to their respective concentrations. Ammonia was not included in this scheme: (1) (2) (3) (4)

Carbon dioxide: CO2 Hydrocarbons and chlorinated hydrocarbons Hydrogen sulfide: H2S Odorous gases and vapors such as mercaptans, amines, aldehydes and VFAs

The typical composition of sewer air reported by Thistlethwayte and Goleb (1972) is shown in Table 4.4. This investigation did not distinguish between components from inlets, e.g., industrial sources and components produced in the sewer. Group 1 (CO2) indicates that microbial degradation of wastewater organic matter takes place in the sewer. In terms of odor, the other groups (2–4) are relevant. In spite of the fact that the investigation did not include the sources of the components found in the sewer atmosphere, group 2 probably is a result of

inputs to the system. The components included in groups 3 and 4, however, are interpreted as resulting from anaerobic processes. The results reported by Thistlethwayte and Goleb (1972) indicate that the concentrations of the constituents of groups 3 and 4 tend to be related. That is to say, the constituents of group 4 (a, b and c) tend to vary according to the levels of the H2S concentrations, roughly in the ratio of 1:50 to 1:100. They conclude that this observation suggests that although the H2S concentration alone may not be a sufficient measure of potential sewer air odor levels, H2S concentration measurements are probably sufficient for most studies of sewer gases.

Release of odorous compounds including hydrogen sulfide from the wastewater into the overlaying atmosphere is a fundamental process for evaluating odor problems. As long as an odorous compound remains in the water phase, odor problems do not exist. It is important that odorous substances undergo degradation in the wastewater phase under aerobic conditions. Organic sulfur compounds seem to be fast degradable, whereas this is not the case for nitrogen compounds (Hwang et al., 1995).

(1) Carbon dioxide, CO2 (2) Hydrocarbons and chlorinated hydrocarbons a. Hydrocarbons, mainly aliphatics C6–C14 and mostly C8–C12 (petrol) b. Chlorinated hydrocarbons, mostly trichlorethylene with ethylene dichloride and some carbon tetrachloride (3) Hydrogen sulfide, H2S (4) Odorous gases and vapors a. Sulfides (mostly methyl mercaptan and dimethyl sulfide; some ethyl mercaptan) b. Amines (mostly trimethylamine and dimethylamine; some diethylamine) c. Aldehydes (mostly butyraldehyde)

0.2–1.2% up to 500 ppm 10–100 ppm 0.2–10 ppm 10–50 ppb 10–50 ppb 10–100 ppb

Odorous Compounds in Sewer Networks

For sewer systems, three major transport phenomena for odorous compounds may be important: • the air–water transfer process for odorous compounds describing the emission from the wastewater into the sewer atmosphere • the ventilation of the sewer network responsible for the emission of odorous compounds from the sewer atmosphere into locations where malodors should not be accepted • adsorption of gases on the moist sewer walls followed by degradation processes The fundamentals of the first aspect are dealt with in Sections 4.1 and 4.2, concerning the equilibrium relations and the transport processes, respectively. Furthermore, equilibrium aspects of the emission of hydrogen sulfide from the water phase into the sewer atmosphere are included in Section 4.1 as relevant and illustrative. Emission in terms of mass transfer of an odorous compound requires information as shown in Section 4.2. The determination of the overall mass transfer coefficient, KL or KG [cf. Equations (4.18) and (4.19)] is crucial in this respect. As previously mentioned, KL and Equation (4.18) are typically applied. Approaches have been suggested for the determination of KL values for odorous compounds based on knowledge of the molecular diffusion coefficient and the experience gained from air–water oxygen transfer in terms of K LO2 values (cf. Section 4.4). In Section 4.2.2, it was mentioned that the mass transfer coefficient, k, according to the two-film theory, is equal to D/z for each of the two films. Contrary to this theory, the surface renewal theory implies that k D0.5/z. The value of n in the following expression (4.22) is not well defined from a theoretical point of view. However, it appears that n is about 1 in a slow-flowing sewer and that it approaches 0.5 in turbulent conditions. n

KL K LO2

DL DLO2

(4.22)

The resistance to oxygen transfer across the air–water interface mainly exists in the water film. Therefore, Equation (4.22) should only be applied to compounds that are comparable to oxygen, i.e., according to Liss and Slater (1974), those that have an HA value greater than ~250 atm (mole fraction) 1. The occurrence of hydrogen sulfide in the sewer atmosphere is an important example for illustrating odor problems and other negative effects associated with sulfide that will be further dealt with in Section 6.2.6. According to Table 563 atm (mole fraction) 1, and H2 S, therefore, observes the 4.1, HH2 S

requirement for applying Equation (4.22) to determine the mass transfer coefficient, K LH2S . A number of approaches have been suggested for the determination of the molecular diffusion coefficient, D, of a component in water (Othmer and Thakar, 1953; Scheibel, 1954; Wilke and Chang, 1955; Hayduk and Laudie, 1974; Thibodeaux, 1996). Based on these five references, the diffusion coefficient ratio DLH2S / DLO2 was found to vary within the interval 0.78–0.86 with an arithmetic mean value equal to 0.84. This value can be inserted in Equation (4.22) as a first estimate to determine K LH2S . Equation (4.22) and the empirical expressions for K LO2 outlined in Table 4.7 are the basis for the determination of the mass transfer coefficient for H2S, K LH2S , and thereby, the emission of H2S from the wastewater into the sewer atmosphere. Further details relevant in this respect are dealt with in Section 4.4. A great number of processes and sinks related to the sulfur cycle in a sewer affect to what extent hydrogen sulfide is an odor problem. Figure 4.4 outlines the major pathways that also will be major subjects for detailed descriptions in Chapter 6. Although not all aspects depicted in Figure 4.4 can be easily quantified, they should be included in an evaluation of odor problems associated with sewage transport. The transformation and transport rates that are involved in the sulfur cycle shown in Figure 4.4 determine to what extent the relevant components will exist in the different phases of the sewer system. As already shown — and further focused on when dealing with the concrete corrosion in Section 6.2.6 — the

Odorous Compounds in Sewer Networks

impacts of hydrogen sulfide are primarily associated with its occurrence in the gas phase. In this respect, it is interesting that although there is a rather fast release of sulfide from the water phase to the sewer atmosphere, the concentrations measured in the atmosphere typically only correspond to 2–20% of the equilibrium concentration (Pomeroy and Bowlus, 1946; Matos and de Sousa, 1992). Adsorption of H2S on the sewer walls followed by oxidation to sulfuric acid is considered the main reason for this observation (cf. Section 6.2.6). Equation (4.23) is a simple, however, in many cases, probably a realistic mass balance for sulfide in the sewer atmosphere. Compared with the absorption rate of H2S on the sewer walls, the amount that is released by ventilation from a sewer into the urban atmosphere is often considered insignificant, and the following mass balance is, therefore, approximately valid (cf. Figure 4.4): Qatm

Qe Qads

(4.23)

where Qatm quantity of H2S buildup in the sewer atmosphere (g) Qe quantity of H2S emitted from the wastewater (g) Qads quantity of H2S adsorbed and then oxidized on the sewer walls (g) Matos and de Sousa (1992) and Matos and Aires (1995) have, based on empirical expressions for the emission rate and adsorption rate on the sewer walls of H2S, used Equation (4.23) as a basis for a model for forecasting the buildup of H2S in the sewer atmosphere. The expressions included in this model are detailed in Matos and de Sousa (1992). In addition to the VOCs produced in sewers under anaerobic conditions, such components may originate from external sources, e.g., industries (Corsi et al., 1995; Olson et al., 1998). Typically, these VOCs are hydrocarbons and similar products. The emission is influenced by a number of sewer system parameters, hydraulic conditions, wastewater characteristics and physicochemical properties for the VOC component. The emission of such incoming VOCs thereby follows the general approach described in this text for odorous substances.

Two different types of odor measurements can be performed, either analytical measurements or sensory measurements (cf. Section 7.1.4). Sensory measurements are either performed by the human nose or electronic detectors and, therefore, relate to the effects of the odor (Sneath and Clarkson, 2000; Stuetz et

al., 2000). Sensory measurements are extremely useful; however, in terms of prediction and modeling, the analytically based measurements are crucial. On the other hand, the large number of potential odorous substances is a major obstacle for determination without having a general reliable indicator. Thistlethwayte and Goleb (1972) made an important, however, somewhat dubious statement when they concluded that although the H2S concentration alone may not be a sufficient measure of potential sewer air odor levels, H2S concentration measurements are probably sufficient for most studies of sewer gases. Their statement was based on rather limited measurements in sewers; however, it corresponds to the theoretical considerations concerning odor production from anaerobic microbial processes in sewers dealt with in Chapter 3 and what has been mentioned in the previous sections of this chapter. Other authors, primarily dealing with odors related to wastewater treatment plants, have also observed that H2S can be considered a relevant indicator for an odor level (Gostelow and Parsons, 2000). From an engineering point of view, hydrogen sulfide is especially relevant. In addition to odor, several human health-related problems are potentially associated with the occurrence of hydrogen sulfide at concentration levels relevant for sewer networks. In this respect, it is interesting to compare values from Table 4.5 with Figure 4.2. The levels indicated in Table 4.5 depend on human sensitivity and time of exposure. It is important to notice that H2S loses its characteristic smell at about 50 ppm, and, therefore, no direct possibility for its detection exists. The density of H2S is slightly higher than that of the atmosphere (relative density 34/29). Therefore, H2S has a tendency to accumulate in, for example, pumping stations and manholes. As it is typically not detected by its smell at those concentrations, where it is life threatening, instruments or alarm systems for its monitoring must generally be used when working in sewer systems. Although the hydrogen sulfide concentration in the sewer atmosphere or the

Threshold odor limit Unpleasant and strong smell Headache, nausea and eye, nose and throat irritation Eye and respiratory injury Life threatening Immediate death

0.0001–0.002 0.5–30 10–50 50–300 300–500 >700

Reaeration in Sewer Networks

2

Minor Medium Considerable

surroundings is the more correct indicator of odor problems, a pragmatic approach is, as a first estimate, to let hydrogen sulfide in the wastewater phase of a sewer network be the indicator of the potential risk. Problems related to the occurrence of hydrogen sulfide have been intensively reported in the literature (cf. this chapter and Section 6.2.6). Although a great number of factors affect the relation between the occurrence and the problems identified, Table 4.6 can be considered a relevant approach. The table gives a simplified estimate, and indication of “medium” problems must not be considered equivalent to “no need for control.” The partial pressure of H2S on a volumetric basis in the atmosphere in equilibrium with a water phase of sulfide (H2S + HS ) is at a pH of 7, approximately equal to 100 ppm (gS m 3) 1 (Figure 4.2). It is clear that under equilibrium conditions, much lower concentrations than those corresponding to the values shown in Table 4.6 may result in odor and human health problems. This is also seen from the fact that Henry’s constant for H2S is rather high, HH2S 563 atm (mole fraction) 1 at 25 C (Table 4.1). However, under real conditions in sewer networks, conditions close to equilibrium rarely exist because of, for example, ventilation and adsorption followed by oxidation on the sewer walls. Typically, the gas concentration found in the sewer atmosphere ranges from 2–20% and is normally found to be less than 10% of the theoretical equilibrium value (Melbourne and Metropolitan Board of Works, 1989). Another reason why a total concentration of 0.5 gS m 3 typically does not give rise to problems is that wastewater contains small amounts of heavy metals that will combine with free sulfide to form insoluble metal sulfides. Iron is typically present in wastewater in relatively high concentrations.

The DO mass balance in wastewater of sewer systems is fundamental for the microbial processes. The low solubility of oxygen in water, relatively high resistance to mass transfer across the air–water interface and potentially high removal rate of DO are major reasons for the fact that DO is often a limiting factor

for the aerobic microbial processes in wastewater. A quantification of the mass transfer across the air–water interface in a sewer, taking into account both process and system relevant aspects, is highly important and needed for a quantification of the aerobic microbial processes.

The first basic aspect is the solubility of oxygen in water and wastewater in equilibrium with an overlaying atmosphere [cf. Henry’s law, Equation (4.8), and Example 4.1]. The following formula is generally applied for the determination of the solubility of oxygen in (clean) water in equilibrium with the atmosphere: SOS

P pS (14.652 0.41022T 0.00799T 2 760 pS

0.0000773T 3 ) (4.24)

where SOS dissolved oxygen saturation concentration in water (in equilibrium with the atmosphere) (gO2 m 3) P the actual air pressure (mmHg) pS saturated vapor pressure at temperature T (mmHg) T temperature ( C) The temperature dependency of SOS is generally more important than the dependency of the pressure. As seen from Equation (4.24), SOS is close to 14.65 g m 3 when the temperature is approaching 0 C, and at 15 C, it is 10.04 g m 3. However, a reduced partial pressure of O2 in a sewer atmosphere should be considered (cf. Example 4.1). Equation (4.24) refers to clean water. Inorganic and organic soluble components in wastewater have an impact on the solubility: SOS,ww

SOS

(4.25)

where SOS,ww

dissolved oxygen saturation concentration in wastewater (gO2 m 3) ratio of solubility of O2 in wastewater to that of clean water; correction factor ( )

The value of may vary depending on the type of wastewater, typically in the interval of 0.8–0.95.

Reaeration in Sewer Networks

Based on theoretical considerations, typically supported by a great number of experimental data, empirical equations have been developed for determination of the reaeration in pipes. Equations relevant for sewer pipes will be dealt with. When considering sewer pipes, reaeration is traditionally dealt with using an approach similar to Equations (4.18) and (4.19), however, formulated in different units: F

K L a( SOS SO )

K LO2 ( SOS SO )

(4.26)

where F KL a

K LO2 SO

rate of oxygen transfer (g m 3 s 1, h 1 or d 1) overall oxygen transfer coefficient (s 1, h 1 or d 1) oxygen concentration in bulk water phase (g m 3)

The overall oxygen transfer coefficient is defined as follows: KL a

KL a

KL A / V

K L dm1

(4.27)

where oxygen transfer velocity ( m s 1, m h 1 or m d 1) water-air surface area, A, to volume of water, V (m 1) hydraulic mean depth of the water phase, i.e., the cross-sectional area of the water volume divided by the water surface width (m)

KL a dm

Equation (4.26) may be reformulated to include temperature dependency and corrections caused by the wastewater matrix: F

K LO2 (20) ( SOS SO )

(T 20) r

(4.28)

where

r

correction factor for surface-active agents ( ) temperature coefficient for reaeration ( )

The value of depends on the amount of surface-active agents (detergents) in the wastewater. Typically, the value is relatively close to 1, about 0.95. It is

generally accepted that the temperature coefficient, r, is 1.024 (Elmore and West, 1961). The value of the overall oxygen transfer coefficient, K L a K LO2 , is central for the determination of the rate of oxygen transfer, F. Table 4.7 summarizes a number of empirical expressions proposed for the determination of K L a in gravity sewers (Jensen, 1994). Only the formulas for K L a by Parkhurst and Pomeroy (1972), Taghizadeh-Nasser (1986) and Jensen (1994) have been developed for sewer pipes. Taghizadeh-Nasser (1986) performed the investigation in a pilot sewer, whereas the formulas developed by Parkhurst and Pomeroy (1972) and Jensen (1994) were based on measurements in real sewers. Parkhurst and Pomeroy (1972) made investigations based on an oxygen mass balance in sewers that were cleaned for sediments and biofilm. Jensen (1994) based his formula on the one developed by Pomeroy and Parkhurst (1972) and measurements of the reaeration by a direct methodology using krypton-85 as radiotracer (cf. Chapter 7). The expressions in Table 4.7 show that sewer systems and flow characteristics determine the magnitude of K L a. Figure 4.5 illustrates how K L a varies with the flow in a gravity sewer with a diameter D 0.7 m and a slope s 0.003 at a temperature of 15 C. The figure also depicts the corresponding water depth-to-diameter ratio (y/D) and a full-flowing pipe at about 530 m3 h 1 (147 l s 1). Referring to Figure 4.5, all three expressions for KLa show a rather steep de0.2. The model by cline above a water depth-to-diameter ratio, y/D Taghizadeh-Nasser (1986) shows what is considered an unrealistically high potential for reaeration at conditions of an almost full-flowing pipe. The two other models are identical, as indicated by the mathematical expressions.

(1) Krenkel and Orlob (1962) (2) Owens et al. (1964) (3) Parkhurst and Pomeroy (1972)

0.121(u s)0.408 dm0.66 0.00925u 0.67dm1.85 0.96(1 0.17 Fr 2 )(s u)3/ 8 dm1

(4) Tsivoglou and Neal (1976)

B u s

(5) Taghizadeh-Nasser (1986)

0.4u(dm / R)0.613 dm1

(6) Jensen (1994)

0.86(1 0.2 Fr 2 )(su)3/ 8 dm1

*Where Fr is the ug 0.5 dm0.5 , Froude number ( ); u is the mean velocity of flow (m s 1); g is the gravitational acceleration (m s 2); s is the slope (m m 1); R is the hydraulic radius, i.e., the cross-sectional area of the water volume divided by the wetted perimeter (m); and B is the coefficient given as a function of water quality and intensity of mixing ( ) (here about 2360).

Reaeration in Sewer Networks

Special sewer structures like junctions, manholes, bends, weirs and drops may give rise to a turbulence that is increased compared with the hydraulic conditions that exist under normal sewer pipe flow. The turbulence introduced by these structures increases the air–water oxygen transfer, and the formulas in Table 4.7 are no longer valid. These special sewer structures typically have their own site-specific characteristics, and a simple empirical description of the reaeration at sewer drops and falls that includes only the most important parameters is needed. The following two equations, Equations (4.29) and (4.30), express a pragmatic solution to this approach:

so =

SOS − SO,u SOS − SO,d

where so SO,u SO,d SOS

DO deficit ratio for reaeration at sewer fall ( ) DO concentration upstream sewer structure (gO2 m 3) DO concentration downstream sewer structure (gO2 m 3) DO at equilibrium (saturation concentration) (gO2 m 3)

(4.29)

(1) Thistlethwayte (1972) (2) Pomeroy and Lofy (1977) (3) Matos (1992)

1 + 0.20H e0.41H 2

e0.45 H − 0.125 H

*Where H is the fall height (m), i.e., the difference in elevation of the hydraulic energy line upstream and downstream from the fall.

And the following expression: = 1 − so−1

(4.30)

where is the DO efficiency coefficient for reaeration at sewer fall ( ). Both so and depend on the same basic characteristics as the KLa value, i.e., sewer structure characteristics, wastewater quality and temperature. Simple empirical equations expressed in terms of a major important parameter, the height, H, have been developed (Table 4.8). The expression by Thistlethwayte (Table 4.8), is derived from experiments in polluted watercourses. The other two expressions are based on data obtained in sewer systems. The expression proposed by Matos (1992) can be applied in small sewers, and is appropriate to fall heights lower than about 1.75 m.

References

The expressions presented in Table 4.8 are compared in Figure 4.6. The constants included in the expressions are a result of the investigations performed by the authors mentioned in the table. These constants may vary according to a number of site-specific conditions including temperature and wastewater quality characteristics.

ASCE (1989), Sulfide in wastewater collection and treatment systems, ASCE (American Society of Civil Engineers) Manuals and Reports on Engineering Practice 69, 324. Clarke, E.C.W. and D.N. Glew (1971), Aqueous nonelectrolyte solutions, Part VIII. Deuterium and hydrogen sulfides solubilities in deuterium oxide and water, Can. J. Chem., 49, 691–698. Corsi, R.L., S. Birkett, H. Melcer, and J. Bell (1995), Control of VOC emissions from sewers: A multi-parameter assessment, Water Sci. Tech., 31(7), 147–157. Danckwerts, P.V. (1951), Significance of liquid-film coefficients in gas absorbtion. Indust. and Eng. Chem., 43(6), 1460–1467. Dagúe, R.R. (1972), Fundamentals of odor control, J. Water Poll. Control Fed., 44, 583–595. Dobbins, W.E. (1956), The nature of the oxygen transfer coefficient in aeration systems. In: B. J. McCabe and W. W. Eckenfelder Jr. (eds.), Section 2.1 of Biological Treatment of Sewage and Industrial Wastes, Reinhold Publishing Corp., New York, pp. 141–148. Elmore, H.L. and W.F. West (1961), Effects of water temperature on stream reaeration, J. Sanit. Eng. Div., 87, 59. Gostelow, P. and S.A. Parsons (2000), Sewage treatment works odour measurement, Water Sci. Tech., 41(6), 33–40. Hayduk, W. and H. Laudie (1974), Prediction of diffusion coefficients for non-electrolysis in dilute aqueous solutions, J. A.I.Ch.E., 20, 611–615. Higbie, R. (1935), The rate of absorbtion of a pure gas into a still liquid during short periods of exposure, Am. Inst. Chem. Eng. Trans., 31, 365–390. Hvitved-Jacobsen, T., K. Raunkjaer, and P.H. Nielsen (1995), Volatile fatty acids and sulfide in pressure mains, Water Sci. Tech., 31(7), 169–179. Hwang, Y., T. Matsuo, K. Hanaki, and N. Suzuki (1995), Identification and quantification of sulfur and nitrogen containing odorous compounds in wastewater, Water Res., 29(2), 711–718. Jensen, N.Aa. (1994), Air–water oxygen transfer in gravity sewers, Ph.D. dissertation, Environmental Engineering Laboratory, Aalborg University, Denmark. King, C. J. (1966), Turbulent liquid phase mass transfer at a free gas-liquid interface, Indust. and Eng. Chem., 5, 7. Krenkel, P.A. and G.T. Orlob (1962), Turbulent diffusion and the reaeration coefficient, J. Sanit. Eng. Div., 88(SA2), 53. Lewis, W.K. and W.G. Whitman (1924), Principles of gas absorption, Indust. and Eng. Chem., 16(12), 1215. Liss, P.S. and P.G. Slater (1974), Flux of gases across the air-sea interface. Nature, 247, 181–184. Matos, J.S. (1992), Aerobiose e septicidade em sistemas de drenagem de águas residuais, Ph.D. thesis, IST, Lisbon, Portugal. Matos, J.S. and C.M. Aires (1995), Mathematical modelling of sulphides and hydrogen sulphide build-up in the Costa do Estoril sewerage system, Water Sci. Tech., 31(7), 255–261.

Matos, J.S. and E.R. de Sousa (1992), The forecasting of hydrogen sulphide gas build-up in sewerage collection systems, Water Sci. Tech., 26(3–4), 915–922. Melbourne and Metropolitan Board of Works (1989), Hydrogen sulphide control manual — Septicity, corrosion and odour control in sewerage systems, Technological Standing Committee on Hydrogen Sulphide Corrosion in Sewerage Works, vols. 1 and 2. Olson, D.A., S. Varma and R. L. Corsi (1998), A new approach for estimating volatile organic compound emissions from sewers: Methodology and associated errors, Water Env. Res., 70(3), 276–282. Othmer, D.F. and M.S. Thakar (1953), Correlating diffusion coefficients in liquids, Industrial and Engineering Chemistry, 45(3), 589–593. Owens, M., R.W. Edwards, and J.W. Gibbs (1964), Some reaeration studies in streams, Ing. J. Air Pollut., 8, 469. Parkhurst, J.D. and R.D. Pomeroy (1972), Oxygen absorption in streams, J. Sanit. Eng. Div., ASCE, 98(SA1), 121–124. Pomeroy, R.D. and F.D. Bowlus (1946), Progress report on sulfide control research, J. Sewage Works, 18, 597–640. Pomeroy, R.D. and R.J. Lofy (1977), Feasibility study on in-sewer treatment methods, NTIS No. PB-271445, USEPA, Cincinnati, OH. Raunkjaer, K., T. Hvitved-Jacobsen, and P.H. Nielsen (1994), Measurement of pools of protein, carbohydrate and lipid in domestic wastewater, Water Res., 28(2), 251–262. Sander, R. (2000), Henry’s law constants. In: W.G. Mallard and P.J. Lindstrom (eds.), Chemistry WebBook, NIST Standard Reference Database Number 69, National Institute of Standards and Technology, USA, http://webbook.nist.gov/chemistry. Scheibel, E.G. (1954), Liquid diffusivities, Ind. Eng. Chem., 46, 2007–2008. Sneath, R.W. and C. Clarkson (2000), Odour measurement: A code of practice, Water Sci. Tech., 41(6), 23–31. Stuetz, R. and F.-B. Frechen (eds.) (2001), Odours in Wastewater Treatment — Measurement, Modelling and Control, IWA Publishing, p. 437. Stuetz, R.M., R.A. Fenner, S. J. Hall, I. Stratful, and D. Loke (2000), Monitoring of wastewater odours using an electronic nose, Water Sci. Tech., 41(6), 41–47. Stumm, W. and J.J. Morgan (1981), Aquatic Chemistry: An Introduction Emphasizing Chemical Equilibria in Natural Waters, John Wiley & Sons, New York, p. 780. Taghizadeh-Nasser, M. (1986), Gas-liquid mass transfer in sewers (in Swedish); Materieöverföring gas-vätska i avloppsledningar, Chalmers Tekniska Högskola, Göteborg, Publikation, 3:86 (Licentiatuppsats). Thibodeaux, L. J. (1996), Environmental Chemodynamics, John Wiley & Sons, New York, p. 593. Thistlethwayte, D.K.B. (ed.) (1972), The Control of Sulfides in Sewerage Systems, Butterworth, Sidney, Australia. Thistlethwayte, D.K.B. and E. E. Goleb (1972), The composition of sewer air, Proceedings from the 6th International Conference on Water Pollution Research, Israel, June 1972, pp. 281–289. Tsivoglou, E.C. and L.A. Neal (1976), Tracer measurement of reaeration, III: Predicting the reaeration capacity of inland streams, J. Water Pollut. Contr. Fed., 48(12), 2669. U.S. National Research Council, Division of Medical Sciences (1979) Hydrogen sulfide, report by Committee on Medical and Biological Effects of Environmental Pollutants, Subcommittee on H2S.

References Vincent, A. and J. Hobson (1998), Odour Control, CIWEM (Chartered Institution of Water and Environmental Management) monograph of Best Practice No. 2, Terence Dalton Publishing, London, p. 32 Wilke, C. R. and P. Chang (1955), Correlation of diffusion coefficients in dilute solutions, J. A.I.Ch.E., 1(2), 264–270.

Aerobic and Anoxic Processes — Process Concept and Model

The aerobic microbial processes in wastewater are efficient, and the corresponding rates of transformation may be relatively high. In sewer networks with aerobic conditions and a relatively long residence time, significant changes in the wastewater quality in terms of a reduced amount of biodegradable substrate and production of biomass may take place. Such quality changes affect the subsequent treatment processes. In the case of nutrient removal, the aerobic transformations of the wastewater in a sewer may cause a reduced capacity for denitrification and biological phosphorus removal. On the other hand, when primary treatment is required, a positive interaction may exist in terms of reduced amounts of soluble and biodegradable organic matter and an increased amount of biomass, i.e., particulate organic matter. The sewer is a reactor for processes that interact in either a positive way or a negative way with the downstream processes at what is traditionally called the “treatment plant.” Fundamentally, the extent of the aerobic transformations of organic matter in a sewer depends on the presence of an active heterotrophic biomass, electron donors and the electron acceptor. The continuous supply of the electron acceptor, oxygen, is, in this respect, crucial. The reaeration process often limits the transformations and is a key process. This chapter deals with the microbial transformations of wastewater under aerobic conditions in a sewer network. It emphasizes the transformations of the organic matter and includes processes in both the water phase and the biofilm. Furthermore, transformations of particles in suspension originating from sewer sediments are included. A concept and a corresponding model for the integration of the major microbial processes, i.e., growth of the heterotrophic biomass, the respiration and the hydrolysis, are also dealt with. The basic chemical and biological aspects of sewer processes are focused on in Chapters 2 and 3. The reaeration process is dealt with in Chapter 4.

The basic theoretical aspects of aerobic and anaerobic processes relevant for wastewater in sewer networks are focused on in Chapters 2 and 3. Figure 5.1 briefly illustrates an important difference between an aerobic and an anaerobic process exemplified with the transformations of protein in a wastewater sample originating from a sewer system. Under aerobic conditions, suspended protein components were significantly increased, and the soluble part was correspondingly reduced. This change is interpreted as the result of a growth process of the bacterial biomass. Under anaerobic conditions, no significant transformations of soluble and particulate protein took place. The results shown in Figure 5.1 are in agreement with an increase in the oxygen uptake rate as illustrated in Figure 3.11. These examples demonstrate that organic matter components, size fractions and turnover rates of wastewater are affected by the presence of oxygen. They are also in agreement with the basic concept of growth and substrate utilization shown in Figure 2.2. Almeida (1999) made transformation studies of wastewater components in a gravity sewer. The sewer has a length of 7.2 km and a typical retention time of 1.5 hours. An average slope equal to 0.007 and several drops resulted in a sewer dominated by aerobic processes. In addition to the organic components (CODtot, CODsol and BOD), other relevant parameters (ammonia, nitrate, TSS

Illustration of Aerobic Transformations in Sewers

CODtot CODsol BOD5 NH3

6 19 7 6

80 80 20 79

0.94 0.95 0.95 0.98

and VSS) were included. The estimated average removal percentages are shown in Table 5.1. Although experimental studies under sewer conditions are subject to high variability, it can be concluded that the removal of the components shown in Table 5.1 is mainly attributed to the activity of the heterotrophic biomass. Theoretical considerations and a number of studies like the one performed by Almeida (1999) clearly demonstrate that the heterotrophic biomass is central for understanding aerobic transformations. Studies by Stoyer (1970), Stoyer and Scherfig (1972), Koch and Zandi (1973), Pomeroy and Parkhurst (1973) and Green et al. (1985) have also focused on removal of organic matter in sewers, primarily in terms of BOD and COD. The heterotrophic biomass in wastewater is typically not a limiting factor for the aerobic transformations in a sewer. The limitation is typically caused by the supply of the electron acceptor (oxygen), i.e., the reaeration. Contrary to what has been proposed by several authors, it would typically serve no useful purpose to inject biomass (sludge) in a sewer line to enhance the treatment processes unless a significant amount of oxygen is continuously supplied to the wastewater (cf. Example 5.1).

Example 5.1: Reaeration and aerobic transformations in a sewer A gravity sewer pipe with a diameter D 0.5 m and a slope s 0.003 m m 1 is flowing half full under stationary conditions, i.e., the DO concentration is constant and equal to about 0.3 gO2 m 3. The pipe is made of concrete, and the roughness is 1.0 mm. The sewer is an interceptor and serves a separate sewered catchment. The wastewater originates from domestic sources and has a temperature of T 15 C. The characteristics of the wastewater are approximately as depicted in Figure 3.10, i.e., the potential process rates for the aerobic transformations are relatively high. Only aerobic processes in the water phase are considered in the example.

Calculate under these conditions the removal of COD in units of gCOD m h 1 and in units of gCOD m 3 km 1.

3

Solution: The aerobic transformations are considered limited by the reaeration, and Equation (6) in Table 4.7 is used to calculate the rate of oxygen supply. Based on the formula of Colebrook and White for a full-flowing circular pipe, the flow and velocity of the wastewater are determined as follows for the half-filled pipe: u = 0.97 m s−1 Q = 95 L s−1

The hydraulic mean depth of the water phase is:

8

D2 D

=

8

D=

8

⋅ 0.5 = 0.196 m

The rate of oxygen supply at the air–water interface is at 15 C: F = 0.86(1 + 0.2Fr 2 )(su)3 / 8 dm−11.024T −20 ( SOS − SO )   0.972 3/8 −1 −5 = 0.86  1 + 0.2  (0.003 ⋅ 0.97) (0.196) 1.024 (10 − 0.3) × 9.81 0.196   = 4.64 gO2 m −3 h −1

That is, 4.64 gCOD m 3 h 1 is removed. Per unit length of the sewer, the transformation is as follows: 4.64 −3

0.97 ⋅ 10 ⋅ 3600

= 1.33 gCOD m −3 km −1

It is readily seen that the “natural” reaeration in the half-filled sewer results in a rather limited rate of transformation of the organic matter. Under DO nonlimiting conditions, i.e., above 2–4 gO2 m 3, the quality of the wastewater would probably result in a rate of transformation that was 2- to 5-fold as high (cf. Figure 3.11). In a combined sewered area, the pipe would typically be designed with a relatively lower dry-weather flow rate. Under such conditions, the oxygen supply

A Concept for the Aerobic Microbial Transformations of Wastewater

per unit volume of wastewater would be higher, resulting in an enhanced removal rate of the COD. The example shows that in addition to wastewater quality characteristics, the sewer system characteristics are crucial in determining the extent of the transformations.

When transformations of wastewater organic matter in sewers have been dealt with, bulk parameters like BOD and COD have typically been used as measures of organic matter and organic matter removal, e.g., as indicators for treatment. Simple, experimentally determined removal percentages have been used for prediction purposes (Stoyer, 1970; Koch and Zandi, 1973; Pomeroy and Parkhurst, 1973; Green et al., 1985; Almeida, 1999). Investigations of organic matter transformations under aerobic conditions in a gravity sewer were carried out by Raunkjaer et al. (1995) and included soluble and particulate fractions of proteins, carbohydrates, lipids and VFAs. The methods for analyses of these fractions were directly developed for their applicability to a wastewater matrix (Raunkjaer et al., 1994). Although the investigations of both Raunkjaer et al. (1995) and Almeida (1999) showed that removal of COD — measured as a dissolved fraction — took place in aerobic sewers, a total COD removal was more difficult to identify. From a process point of view, it is clear that total COD is a parameter with fundamental limitations, because it does not reflect the transformation of dissolved organic fractions of substrates into particulate biomass. The dissolved organic fractions (i.e., VFAs and part of the carbohydrates and proteins) are, from an analytical point of view and under aerobic conditions, considered to be useful indicators of microbial activity and substrate removal in a sewer. The kinetics of the removal or transformations of these components can, however, not clearly be expressed. Removal of dissolved carbohydrates can be empirically described in terms of 1-order kinetics, but a conceptual formulation of a theory of the microbial activity in a sewer in this way is not possible. The conclusion is that theoretical limitations and methodological problems are major obstacles for characterization of microbial processes in sewers based on bulk parameters like COD, even when these parameters are determined as specific chemical or physical fractions. It is considered important to identify a rather simple, however, generally well-accepted concept for the microbial transformations that can be extended when further knowledge on sewer processes exists. Although details may be missing in a simple description, advantages in terms of possibilities for a sound

use of the concept as a modeling basis increases because of less components, processes and parameters to be experimentally determined. Different approaches may arise when defining the narrow gap between what is a simple and informative description of a system and what is considered more correct; however, because of the details, useless for interpretation of experimental results and for practical application. Basically, a concept for microbial transformations in sewer networks should cover soluble and particulate components and relevant processes in the water phase, in the biofilm and in the sewer sediments. In addition, mass transfer between these phases and an air–water transfer of oxygen should be taken into account (Figures 1.3 and 5.2). Although only the aerobic microbial activity will be focused on in the concept presented in this chapter, anoxic and anaerobic processes should be considered possible extensions (cf. Chapter 6). The concept for description of aerobic, microbial transformations of wastewater organic matter in a sewer should include processes that are relevant for description of wastewater quality aspects. The heterotrophic biomass, the organic substrates and the relevant electron acceptor (dissolved oxygen, DO) are considered fundamental in this respect. The fundamental understanding of the concept selected is based on the generally accepted fact that substrate utilization for growth of biomass takes place parallel with its removal for energy purposes by the electron acceptor (Figures 2.2 and 5.3). These transformations take place in the water phase, the biofilm and the sediment.

A Concept for the Aerobic Microbial Transformations of Wastewater

Compared to the traditional BOD and COD removal concept, which considers organic matter as “degradable” in a fictitious “removal process,” the concept described has moved to highlight biomass as being the real active component, depending on the nature and availability of organic substrates and electron acceptor. The heterotrophic biomass is, therefore, in terms of its activity, the central component of such a concept.

Microbial processes in terms of sludge (biomass) production, BOD and nutrient removal have been focused on when dealing with biological wastewater treatment. Therefore, and because the sewer is the system for input to the wastewater treatment plant, there is a basis as well as a perspective to establish a process concept for the sewer that can be integrated with the activated sludge processes. In activated sludge wastewater treatment processes, organic matter is subdivided into a number of fractions. The major components for the aerobic, heterotrophic processes of activated sludge have been identified as being heterotrophic biomass and different fractions of biodegradable and inert organic matter (cf. Section 3.2.6). Several researchers have contributed to the fundamental steps toward this understanding, e.g., Kountz and Forney (1959), McKinney and Ooten (1969), Gaudy and Gaudy (1971), Marais and Ekama (1976), Gujer (1980) and Dold et al. (1980). These early findings were the basis for a comprehensive description of activated sludge processes formulated in model terms by Grady et al. (1986) and as the “Activated Sludge Model No. 1” by Henze et al. (1987). Further developments on organic carbon transformation kinetics and associated methodologies in activated sludge have taken place, e.g., Sollfrank and Gujer (1991) and Henze et al. (1995, 2000). Differences between the microbial processes in a sewer and in a wastewater treatment plant should be considered for two major reasons. First, take into

account which process-related aspects are relevant for each of the two systems. Second, to realize that different conditions for the microbial processes exist in activated sludge compared with wastewater and biofilm in a sewer. Concerning the first aspect, advanced wastewater treatment is, as far as microbial processes are concerned, evaluated by the ability and efficiency of organic carbon, nitrogen and phosphorus removal from the water phase. In the sewer, however, only the heterotrophic organic carbon transformations are relevant to focus on in terms of wastewater quality changes and “treatment processes.” On the contrary, if advanced treatment of the wastewater is required, it is important to preserve the readily biodegradable organic matter during transport in a sewer because of the subsequent denitrification and biological phosphorus removal processes in the treatment plant. Furthermore, the sulfur cycle in sewers is of importance when conditions for hydrogen sulfide formation exist during wastewater transport. Concerning the second aspect mentioned, i.e., the different process conditions in sewers and treatment plants, the activated sludge processes proceed in a system characterized by a high concentration of sludge flocs including biologically active biomass and nonactive particulate organic matter. From this point of view, the activated sludge consists of the living heterotrophic biomass, primarily, slowly hydrolyzable organic substrate and non- or almost nonbiodegradable substances. In an activated sludge system, the active biomass typically exists under substrate limited growth conditions. Compared with this dense floc system, the wastewater in a sewer is characterized by a low concentration of active biomass in an exponential growth phase. Corresponding process conditions in terms of substrate limited or unlimited growth and hydrolysis of entrapped particulate substrate may be affected by the biomass concentration and are, therefore, expected to result in different process conditions for wastewater in the sewer and in the activated sludge system. Furthermore, biofilm and sediment processes in the sewer will add to the differences between sewer and treatment plant processes. In addition, the sewer processes are exposed to a composition of the organic substrate as it appears more or less directly discharged from households and industry and different from what may exist at the inlet to and in a wastewater treatment plant. The sewer processes are especially relevant for the most biodegradable organic fractions of the wastewater and for sewer systems with a relatively high residence time. These aspects were the starting points for a sewer process model. The first attempt to establish a conceptual understanding and corresponding model for microbial carbon transformations of wastewater under sewer conditions was proposed and applied by Bjerre et al. (1995, 1998a, 1998b). The model developed was based on the activated sludge concept and included as major processes, the biomass growth, the hydrolysis of particulates and the decay of biomass (Figure 5.4). Experimental results confirmed that the concept could be

A Concept for the Aerobic Microbial Transformations of Wastewater

applied for description of processes in the water phase when extended to include two to three fractions of hydrolyzable substrate characterized by their different rates of hydrolysis (Bjerre et al., 1998b). Methods for characterization of wastewater components in terms of different COD fractions and parameters for stoichiometric and kinetic description of the microbial processes were primarily developed based on procedures for activated sludge characterization. However, as a consequence of the different growth conditions for the heterotrophic biomass in activated sludge and in wastewater of a sewer, a modified procedure and a different interpretation of the oxygen uptake rate (OUR) versus time was used (Bjerre et al., 1995, 1998b). In addition, biofilm batch experiments and field investigations were performed to include biofilm and sediment processes as simple surface flux descriptions (Bjerre et al., 1998a). Based on this modified activated sludge concept, it was possible to produce acceptable model simulation results for the water-phase processes of the heterotrophic carbon transformations in sewers. However, problems were identified for the description of the heterotrophic biomass decay. A major problem was the magnitude of the 1-order decay rate constant with respect to the biomass concentration. Henze et al. (1987) and Kappeler and Gujer (1992)

assumed this constant to be 7–10% and 5% of the maximum specific growth rate ( H ) for activated sludge, respectively. Bjerre et al. (1995) found the decay rate constant for wastewater from a sewer system to be 15% of H, and Vollertsen and Hvitved-Jacobsen (1998) estimated values on the order of 40–60% of H for suspended sewer sediments. Such decay rates for bacteria are unrealistic [cf. Kurland and Mikkola (1993)]. Vollertsen and Hvitved-Jacobsen (1998), therefore, concluded that the concept until now adopted should be reconsidered. Taking this into consideration and based on new experiments as well as results from previous investigations [e.g., Tempest and Neijssel (1984) and Russel and Cook (1995)], Vollertsen and Hvitved-Jacobsen (1998) concluded that a nongrowth-related substrate removal process should be considered. This process could be interpreted as a maintenance energy requirement of the heterotrophic biomass. Compared with this process, the decay of biomass and endogenous respiration were considered less important for wastewater and were omitted as part of the heterotrophic processes in a sewer. This fact was confirmed by Tanaka and Hvitved-Jacobsen (1998) based on their observations of heterotrophic biomass activity under varying aerobic and anaerobic conditions. Only if the readily biodegradable substrate produced by hydrolysis was insufficient to support the maintenance energy requirement of the biomass, was biomass considered to undergo endogenous respiration. Results from addition of readily biodegradable substrate to sewer solids subject to aerobic conditions for several days confirm that this process is likely to take place (Vollertsen and Hvitved-Jacobsen, 1999). The possibility of endogenous respiration is required, otherwise the mass balance of the COD cannot be observed, because the rate constant for the maintenance energy requirement is constant and not considered depending on, for example, the availability of substrate. Although the maintenance energy requirement of the biomass defined as a nongrowth-related substrate utilization process is considered real, the overall concept of biomass-related processes still reflects a pragmatic approach. It is, however, important to state that in addition to the general requirement of using a simple concept including the major microbial processes considered important for the objective, the COD mass balance must be observed. It is also important to accept that any change in a concept will affect the value of the parameters in the corresponding conceptual model. This means that because the biomass decay is replaced with the biomass maintenance energy requirement, in addition to the increased number of hydrolyzable fractions, this will result in parameters that are different compared with what is normally recommended and accepted for the activated sludge model. As an example, a biomass yield constant based on the sewer process concept is defined differently compared to what is the case for the activated sludge model concept. This fact calls for theoretically based and experimentally reliable methodologies to determine process parameters.

A Concept for the Aerobic Microbial Transformations of Wastewater

The concept and methodologies to determine components and parameters must be in agreement. If this is not the case, the concept is nothing but theoretical. Based on the findings and considerations mentioned, the concept proposed for the heterotrophic biomass processes in wastewater under aerobic sewer conditions is depicted in Figure 5.5. The difference between this concept and the concept behind the activated sludge model is clear. It is not a point for discussion which concept is “correct,” but it is fundamentally important to consider the validity of a concept formulated in relatively simple terms under given constraints. The difference in the consideration of sewer and treatment plant processes is, however, no obstacle for the corresponding integration of these two systems, including the receiving waters. What is important is the fact that COD-fractions in terms of their role in biotransformation processes are compatible and well defined at the interfaces between the sewer and the two subsystems: the treatment plant and the local receiving waters. The concept shown in Figure 5.5 is in accordance with the basic criterion that has been highlighted and discussed and finally depicted in Figure 5.3. The

biomass is the central component that, by its activity, is the driving force for the changes of the organic components of the wastewater. It seems correct to deal with microbial transformations based on this fundamental statement. However, it also creates basic problems. It is somewhat problematic that the biomass in terms of its activity is determined based on an oxygen uptake rate measurement, at the same time as it is materialized as a COD-fraction. This fundamental “conflict,” mixing biomass activity and biomass as something being materialized, is the strength and the weakness of the concept. Explicitly expressed methodologies for determination of the biomass, e.g., applying genetic engineering methods, are at present not realistic alternatives, although a future potential may exist (Vollertsen et al., 2000). Only a careful validation of the concept performed under different external conditions for the biomass can make this dual situation meaningful. The concept including methods for determination of related characteristics is — when this basic requirement is observed — considered theoretically sound and, at the same time, operational from a practical point of view.

The microbial transformations of the wastewater described in the concept shown in Figure 5.5 deal with the COD components defined in Section 3.2.6. The figure also depicts the major processes that include the transformations of the organic matter (the electron donors) in the two subsystems of the sewer: the suspended wastewater phase and the sewer biofilm. The air–water oxygen transfer (the reaeration) provides the aerobic microbial processes with the electron acceptor (cf. Section 4.4). Sediment processes are omitted in the concept but are indirectly taken into account in terms of a biofilm at the sediment surface. Water phase/biofilm exchange of electron donors and dissolved oxygen is included in the description. All subsystems influence the integrated in-sewer processes, although the water phase processes are typically most important. Reaeration is a process that, when there are low DO concentrations in the wastewater, limits the rate of the aerobic processes (cf. Example 5.1). The relative importance of the processes in the suspended water phase and in the biofilm may vary, e.g., determined by the area/volume ratio of the sewer pipe. When considering the details of the sewer process concept, the processes of the subsystems are theoretically described at different levels. The most detailed description is done for the water phase by including the biomass/substrate relationship. Fewer details in terms of a description at an empirical level are included for the reaeration and the processes in the biofilm.

Process Descriptions

Theoretical knowledge is available for a detailed description of the biofilm processes (Characklis, 1990; Gujer and Wanner, 1990). However, a fundamental requirement to establish applicable experimental procedures for determination of components and process parameters delimits the use of details. A simple description of the biofilm processes in terms of a surface flux model according to the description in Section 3.2.2 is selected. The dilemma related to the details of a process description is fundamental. Generally, one should reach a point where a process description and the experimental potential for its quantification are optimal. The criteria for this optimum is a quantification of those processes relevant for the sewer performance focused on, expressed with sufficient accuracy, applicable under real conditions and at the same time in agreement with methods for a sound determination of model parameters. The reaeration process is described in Section 4.4. In the following, focus will be on expressions for the microbial processes. Components applied in these expressions are shown on p. 54. A list of nomenclature is given in Appendix A, pp. 229–232.

Growth of suspended biomass under limiting conditions of organic substrate and DO follows the classical Monod formulation (Monod, 1949) (cf. Section 2.2.1). This was confirmed in studies with wastewater based on laboratory and field experiments performed under gravity sewer conditions by Bjerre et al. (1995) and Bjerre et al. (1998a). The formulation of the heterotrophic growth rate of suspended biomass in water phase, rgrw, in units of gCOD m 3 d 1, therefore, follows the concept of heterotrophic growth in the “Activated Sludge Model No. 1” (Henze et al., 1987): rgrw =

H

SO SF + S A X Bw K Sw + (SF + S A ) KO + SO

(T − 20) w

(5.1)

where 1 H maximum specific growth rate (d ) KSw saturation constant for readily biodegradable substrate (gCOD m 3) KO saturation constant for DO (gO2 m 3) w temperature coefficient for the water phase process ( ) T temperature ( C)

Suspended biomass growth results in the removal of readily biodegradable substrate. A yield constant, YHw, typically about 0.55 g COD biomass produced

per g COD substrate consumed, has been observed by Bjerre et al. (1998a). The consecutive energy-producing process manifests itself in a corresponding DO consumption ratio of 1 YHw.

Considerations behind introduction of a nongrowth-related consumption of substrate and corresponding uptake of DO are discussed in Section 5.2.2. The concept outlined in Figure 5.5 accounts for a maintenance energy requirement of the biomass, which in addition to the growth (yield)-related energy requirement, removes readily biodegradable substrate and, if this is not available, the biomass itself by lysis. The latter is needed to observe the fundamental COD mass balance. The maintenance energy requirement concept is considered reasonable in a sewer network where the biomass often exists under organic carbon conditions of unlimited growth and where biomass decay in the suspended water phase is considered less important. The maintenance energy requirement rate, rmaint, of the suspended biomass is as follows: rmaint = qm

SO X Bw KO + SO

(T − 20) w

(5.2)

where qm is the maintenance energy requirement rate constant (d 1).

Details concerning biofilm growth and activity in sewers are at present not available to the same extent as is the case for the suspended biomass. Therefore, a simple expression for biofilm growth and respiration compared to other well-known deterministic biofilm models, e.g., as described by Gujer and Wanner (1990), is selected. Laboratory and mixed field/laboratory studies have confirmed that half-order kinetics for DO surface removal rates may be a reasonable approximation for sewer biofilm (Raunkjaer et al., 1997; Bjerre et al., 1998b). These results also showed the influence of readily biodegradable substrate. Furthermore, temperature dependency limited by diffusion is included (Nielsen et al., 1998). The following equation for the aerobic growth rate was therefore used: rgrf = k1/ 2 SO0.5

YHf 1 − YHf

SF + S A A K sf + (SF + S A ) V

(T − 20) f

(5.3)

Process Descriptions

where k1/2 half-order rate constant (gO20.5 m −0.5d −1 ) YHf biofilm yield constant [gCOD, biomass (gCOD, substrate) 1] A/V wetted sewer pipe surface area divided by the water volume, i.e., R 1, where R is the hydraulic radius KSf saturation constant for readily biodegradable substrate in biofilm (gCOD m 3) This growth expression requires a minimum of kinetics and stoichiometric coefficients to be determined, and no hydraulic details are included. The dynamics of sewer biofilm detachment are not quantitatively known, and a steady state biofilm with a biomass release to the bulk water phase, equal to the biomass growth within the biofilm, is therefore an estimate. It should be noticed that biomass growth and respiration for bulk water phase include details that are not taken into account in the simple half-order biofilm description. As an example and a consequence, the two yield constants, YHw and YHf, are differently interpreted in terms of the substrate requirement of the biomass (Figure 5.5). Example 5.2: Aerobic processes in sewer biofilms Equation (5.3) includes the influence of both the electron donor (organic substrate) and the electron acceptor (DO) on the growth of biofilm biomass. The expressions for these dependencies are of the Monod type and 1/2-order kinetics, respectively (cf. Section 2.2). Investigations have been performed to exemplify the DO surface removal rates from biofilms grown on different types of wastewater (Bjerre et al., 1998b). Such investigations may indicate if Equation (5.3) can be considered an appropriate description of the aerobic activity. The wastewater for these studies originates from an open sewer system, the Emscher river, Germany. The results of the experiments are outlined in Table 5.2, and further details are shown in Figures 5.6 and 5.7. The experiments outlined in Figure 5.6 are performed under organic substrate nonlimiting conditions. Figure 5.7 shows corresponding experiments, however, performed under stepwise aerated conditions without the addition of organic substrate. These experiments show the importance of the organic matter on the aerobic biofilm activity in terms of the biodegradability of the wastewater. The results of this study outlined in Table 5.2 and Figures 5.6 and 5.7 follow what is expressed in Equation (5.3). This equation observes theoretically sound concepts and is, therefore, found to be an appropriate expression for the estimation of biofilm activity.

r r r

70

130

280

280

36

23

10

r

52

70

52 r

0.93 086

0.08 SO0.45 0.10 SO0.35

0.2–8 0.2–6.5

130–250

130–230

90–180

80 0.2–5.5

1.5–6

— 0.83

0.085 0.11 SO0.53

80 0–1.5

0.98

0.07 SO0.46

Process Descriptions

Hydrolysis of particulate substrates produces readily biodegradable substrate for the biomass (cf. Figure 5.4 and Section 3.2.3). The kinetics of the hydrolysis, following the concept of the activated sludge model one, is described in Section 2.2.2. The following interpretation of hydrolysis of wastewater in a sewer is considered: particulate substrate is available in the bulk water phase, and biomass in the bulk water and biofilm — assuming a reduced activity in the biofilm — is taken into account. Under these conditions, the rate of hydrolysis, rhydr, for each of the hydrolyzable fractions, n, is as follows: rhydr = khn

XSn / X Bw SO  A X Bw + X Bf   K Xn + XSn / X Bw KO + SO  V

(T − 20) w

(5.4)

where khn hydrolysis rate constant, fraction n (d 1) KXn saturation constant for hydrolysis, fraction n (gCOD gCOD 1) efficiency constant for the biofilm biomass ( ) The existence of different fractions of particulate substrates in terms of their specific hydrolysis rates is an important finding that originates from investigations of wastewater and resuspended sediments (Bjerre et al., 1995; Bjerre et al., 1998a; Vollertsen and Hvitved-Jacobsen, 1998; Tanaka and HvitvedJacobsen, 1998a). Typically, two to three fractions must be considered to interpret the hydrolysis when it occurs in wastewater of sewer systems (cf. Section 3.2.6).

A number of fundamental requirements and constraints must be observed when establishing a concept of understanding and a corresponding model within environmental engineering. This is, of course, also the case when dealing with sewer processes. As the starting point, the processes must be theoretically understood and interpreted in an integrated way before they can be formulated conceptually. This work must be continuously repeated, and corrections must be made by iterative procedures influenced by results from relevant experiments and observations. Not until a point has been reached where an acceptable agreement between a concept and its experimental verification has been obtained, can a concept be defined and, temporarily, successfully formulated. The observation is that mass balances are, in this respect, a basic engineering requirement. A relatively simple concept including central but also a

Sewer Process Model

limited number of components and parameters to be experimentally determined should be a major goal. In addition to the kinetics of the sewer processes described in Section 5.3, the stoichiometry of the transformations of the components is crucial for the mass balance. The stoichiometry of the biomass/substrate relationships is, according to the activated sludge model concept, determined by the heterotrophic biomass yield constant, YH, in units of gCOD gCOD 1. As depicted in Figure 5.5, the yield constant is an important factor related to the consumption of both SS and SO for the production of XBw. Wastewater in sewers includes different and varying species of heterotrophic microorganisms. A simple relationship between biomass growth and substrate utilization is needed. Several studies performed with different types of wastewater and sewer solids have shown that a simple description is possible and acceptable (Bjerre et al., 1995; Vollertsen and Hvitved-Jacobsen, 1999). The fundamental understanding of the biomass yield constant follows the activated sludge model concept, however, there is a major difference (cf. Figures 5.3 and 5.5). When introducing the maintenance energy requirement, a part of the readily biodegradable substrate is used for nongrowth-related purposes and should not be taken into account when determining the yield constant. Therefore, YHw in the sewer process concept is defined as follows: YHw = −

dX Bw dSS ,growth

(5.5)

According to the relatively simple empirical description of the biofilm processes, the biomass concentration of the biofilm, XBf , expressed in units of gCOD m 2 of the biofilm surface, is considered constant. The biomass growth rate within the biofilm is, therefore, assumed to result in a corresponding rate of biomass detachment and release to the bulk water phase, i.e., the biofilm is considered in a steady state. The remaining components, XBw, SS, XSn and SO, dealt with in the concept are subject to changes during transport of the wastewater in the sewer. As a result, the following mass balances for these components are expressed in terms of a set of coupled differential equations: ∂X Bw = rgrw + rgrf − rmaint ∂t

(5.6)

Note: rmaint is only relevant if SS is not sufficient for maintaining the biomass. ∂SS 1 1 =− rgrw − rgrf − rmaint + ∑ rhydr,n ∂t YHw YHf

(5.7)

Suspended biomass growth Maintenance energy requirement Biofilm biomass growth Hydrolysis, fraction 1 Hydrolysis, fraction 2 1/YHw 1 1/YHf 1 1 1 1

*If SS is not sufficiently available to support the biomass maintenance energy requirement.

(2) (3) (4) (5) (6)

(1) Reaeration 1 1* 1

YHw)/YHw 1 (1 YHf)/YHf (1

1

rrea [4.28, Equation (6) from Table 4.7)] rgrw (5.1) rmaint (5.2) rgrf (5.3) rhydr, n 1 (5.4) rhydr, n 2 (5.4)

Oxygen Mass Balance and Modeling in Sewers

∂XS ,n ∂t

= −rhydr,n

(5.8)

Note: for each of the hydrolyzable fractions, n: 1 − YHf −∂SO 1 − YHw = −rrea + rgrw + rgrf + rmaint ∂t YHw YHf

(5.9)

where rrea is the rearation rate. This set of coupled differential equations can — as also expressed for the activated sludge model concept — be formulated in terms of a matrix. This matrix includes the relationships between the relevant components, processes, expressions, process rates and coefficients (Table 5.3). The mass balances shown in Equations (5.6) to (5.9) can be identified as columns in the matrix. Methods for determination of kinetic and stoichiometric parameters and components relevant for the conceptual model shown in Table 5.3 will be dealt with in Chapter 7. These symbols are given in Appendix A.

The overall DO mass balance in a sewer network is shown in Figure 5.8 (Matos and de Sousa, 1996). The figure outlines the central role of the DO

concentration in the bulk water phase. For several reasons, this is natural: the flow of oxygen occurs into the water phase with the reaeration as the driving force, and the DO consuming processes have their source of oxygen in this phase. Furthermore, measurement of the DO concentration typically takes place in the water phase. The DO mass balance in a sewer can be established at different levels. Basically, it is either expressed by the conceptual model, i.e., Equation (5.9) that is the same as column number 5 in the matrix, Table 5.3, or simply stated as follows (Parkhurst and Pomeroy, 1972; Jensen and Hvitved-Jacobsen, 1991; Matos and de Sousa, 1991, 1996):

dSO = input − output + K L a( SOS − SO ) − (rw − rf ) dt

(5.10)

where rw rate of DO consuming processes in the flowing wastewater (gO2 m h 1) rf rate of DO consuming processes in the biofilm (gO2 m 3 h 1)

3

For a sewer network with self-cleansing conditions, without drops and without considerable amounts of reduced substances produced by anaerobic processes in the deeper parts of the biofilm, the simplified mass balance in Equation (5.10) follows what is outlined in Figure 5.8. Equation (5.10) can be solved by numerical methods or analytically corresponding to different conditions of DO consumption (Matos and de Sousa, 1996). Different levels of description of the rate expressions, rw and rf, can be selected when using a simple mass balance as expressed in Equation (5.10). Matos and de Sousa (1996) propose a temperature-dependent, however, DO nondependent value for rw: rw = rw (20)

(T − 20)

(5.11)

A Monod type of DO dependency as used in Equations (5.1) and (5.2) can be added to this equation. The value of rw(20) depends on the quality of the wastewater and is subject to high variability. It can be determined from simple laboratory experiments. The values shown in Table 5.4 originate from different investigations of wastewater from sewers and clearly show that local knowledge is required to indicate a reliable value for rw(20). A simple description of the DO consumption rate, rf, in the biofilm, in units of gO2 m 3 h 1, is proposed by Parkhurst and Pomeroy (1972):

Oxygen Mass Balance and Modeling in Sewers

Boon and Lister (1975) USEPA (1985) Matos and de Sousa (1991) Huisman et al. (1999)

Wastewater subject to anaerobic conditions Young wastewater Small sewers Main sewers, 15 C

rf = 5.3SO (su)−0.5 R −1

11–16 2–3 0.1–0.3 (average values) 0.5 (night)–3 (day)

(5.12)

where s slope of sewer line (m m 1) u mean velocity of flow (m s 1) R hydraulic radius (m) Equation (5.12) shows a linear dependency in the DO concentration that is not in agreement with the results shown in Figure 5.6. Matos (1992) also found a discrepancy between Equation (5.12) and experimental results and substituted the expression 5.3 SO in Equation (5.12) with a constant equal to 10.9. This constant depends on biofilm and wastewater characteristics and should be determined from local measurements. In addition to the information given by Bjerre et al. (1998b) in Example 5.2, values of respiration rate measurements for sewer biofilms are shown in Table 5.5. It is important to notice that the empirical expressions (5.11) and (5.12) are simple, and, contrary to the DO mass balance in Equation (5.9) — as a part of the conceptual description of sewer processes — do not include a dynamic description of the wastewater quality changes taking place in a sewer. However, the simple DO mass balance expressed in Equation (5.10) may give useful information, as shown in Example 5.3. Generally, it is important that both simple and more complicated models exist. A model most appropriate to use will always depend on the actual objective and the information available.

Boon and Lister (1975) Matos and de Sousa (1991) Norsker et al. (1995) Huisman et al. (1999)

15 C Small sewers, 20 C 20 C Main sewer, 15 C

0.7 0.15–1.5 (average values) 1.2–1.8 0.17–0.25

Example 5.3: DO concentration profiles in a sewer pipe under varying conditions of water depth-to-diameter ratio (Matos and de Sousa, 1996) DO concentration profiles are calculated and exemplified in Figure 5.9 according to Equation (5.10), assuming the following: DO concentration at the inflow to the pipe section: SO rw 5 g m 3 h 1 T 20 C s 0.005 m m 1 sewer diameter D 0.3 m

4.3 gO2 m

3

The hydraulic characteristics of the flow were calculated according to the Manning equation. The curves in Figure 5.9 illustrate the importance of the y/D ratio on the DO concentration. Although Equation (5.10) and thereby Figure 5.9, include DO consumption processes, it is interesting to compare Figure 5.9 with Figure 4.5. This figure depicts a corresponding influence of the y/D ratio on the magnitude of the reaeration, i.e., the supply with oxygen. As also illustrated in Example 5.1, these different angles of the DO mass balance show that reaeration often plays a central role in establishing aerobic conditions. Figure 5.9 also shows that DO equilibrium conditions may only be established at relatively low values of y/D.

Oxygen Mass Balance and Modeling in Sewers

The processes in the DO mass balance, and thereby the DO concentration, are subject to great variability. Especially in sewers with low DO concentrations, this variability may ultimately result in varying aerobic/anaerobic conditions, a case that will be further dealt with in Chapter 6. At a specific location, the flow, the wastewater quality and the temperature vary over time, daily and annually, and result in substantial changes in the DO concentration. Example 5.3 illustrated such relations with emphasis on the flow conditions. Example 5.4 will further illustrate this and especially focus on the impact of the temperature and the wastewater quality on the DO concentration. Example 5.4: DO concentration profiles in a sewer pipe subject to daily variations in flow, wastewater quality and temperature (Gudjonsson et al., 2001) This example supplements and extends what was illustrated in Example 5.3 related to the variability of DO in sewer networks. Example 5.3 was based on the simple DO mass balance expressed in Equation (5.10). This example will make use of the sewer process model that integrates the reaeration and the DO consuming processes (Table 5.3). Figure 5.10 shows measured values of the DO concentration in a sewer. The example originates from an investigation in an intercepting gravity sewer (diameter D 0.5 m; slope s 0.0023 m m 1). The DO variability over day and night is affected by the flow and the quality of wastewater (Gudjonsson et al., 2001). Around noon, the wastewater produced in the morning hours has an increased depleting effect on the DO concentration. Furthermore, a comparison between the curves measured over three different days from April through June shows the impact of the temperature on the rate of DO consumption. Figure 5.11 includes all measurements during this period and indicates the effect of a temperature increase from 9–14 C on the daily duration of anaerobic conditions defined as a DO concentration 1 gO2 m 3. However, experience and literature seem to indicate that if the DO concentration exceeds 0.2–0.5 gO2 m–3, sulfide problems are not typical. When sulfide problems are identified, sulfide concentration in the bulk water is a simple indicator for the corresponding problems (in networks with a free water surface). Sulfide concentrations of 0.5, 3 and 10 gS m 3 may be considered as low, moderate and high, respectively, in terms of problems that are typically reported. In countries with long pressure mains or high temperature of the wastewater in gravity sewers, significantly higher concentrations than 10 gS m 3 are reported (e.g., Thistlethwayte, 1972; Pomeroy and Parkhurst, 1977). The sulfide problem in sewers relates to a complex balance between a large number of wastewater quality-related factors and aspects that depend on the design of the sewer system. The level of the DO concentration in the bulk water, the magnitude of the relevant microbial process rates and exchange rates for

Hydrogen Sulfide in Sewer Networks

oxygen and sulfide outline the general nature of the important factors. It is crucial to “organize” such factors in a framework that can be applied when predicting the sulfide problem. In the following, seven main quality and system factors that influence the formation of sulfide will be briefly focused on.

Sulfate is typically found in all types of wastewater in concentrations greater than 5–15 gS m 3, i.e., in concentrations that are not limiting for sulfide formation in relatively thin biofilms (Nielsen and Hvitved-Jacobsen, 1988). In sewer sediments, however, where sulfate may penetrate the deeper sediment layers, the potential for sulfate reduction may increase with increasing sulfate concentration in the bulk water phase. Under specific conditions, e.g., in the case of industrial wastewater, it is important that oxidized sulfur components (e.g., thiosulfate and sulfite) other than sulfate may act as sulfur sources for sulfate-reducing bacteria (Nielsen, 1991).

Biodegradable organic matter is available in wastewater as a substrate for sulfate reduction. However, in wastewater from, for example, food industries with a relatively high concentration of readily biodegradable organics preferred by the sulfate-reducing bacteria, the sulfate reduction rate may be higher than in wastewater from households. However, also in domestic wastewater, readily biodegradable and fast hydrolyzable COD may be high in certain areas, owing to a shortage or reuse of water, leading to a higher potential for sulfide formation. Several specific organics, e.g., formate, lactate and ethanol, have been identified as particularly suitable substrates for sulfate-reducing bacteria (Nielsen and Hvitved-Jacobsen, 1988).

The temperature dependency of the sulfate reduction rate for single sulfate-reducing bacteria is high, corresponding to a temperature coefficient, , of about 1.13, i.e., a change in the rate with a factor Q10 3.4 per 10 C of temperature increase. Because diffusion of substrate into biofilms or sediments is typically limiting sulfide formation, the temperature coefficient is reduced to about 1.03 (Nielsen et al., 1998). The development of sulfate-reducing species that are adapted to low and to high temperatures, respectively, may result in an apparent reduced temperature dependency between winter and summer. This long-term temperature

dependency may have affected the sulfide production rates observed by Nielsen et al. (1998) in pressure mains at wastewater temperatures between 5 and 12 C.

Sulfate-reducing bacteria mainly exist between pH 5.5 and 9. A significant inhibition of the sulfate-reducing bacteria will, however, not take place below a pH of about 10. The relative distribution between the two sulfide components, with an oxidation level of 2, H2S and HS , depends on the pH. This aspect and the importance hereof is described in Sections 4.1.2 and 4.3.

Sulfide is primarily produced in the biofilm — and the sewer sediment if it occurs. The corresponding water phase concentration of the sulfide is calculated by knowing the relevant area/volume (A/V) ratio. In a pressure main or full-flowing gravity sewer, these parameters correspond to the area and the volume of the pipe, respectively. In a partly flowing gravity sewer pipe, A and V correspond to the wetted pipe surface area and the bulk water volume, respectively. As an example and considering other conditions unchanged, small diameter pressure pipes having A/V larger than big diameter pipes result in higher sulfide concentrations in the water phase (cf. Example 6.2).

The potential production of sulfide depends on the biofilm thickness. If the flow velocity in a pressure main is over 0.8–1 ms 1, the corresponding biofilm is rather thin, typically 100–300 m. However, high velocities also reduce the thickness of the diffusional boundary layer and the resistance against transport of substrates and products across the biofilm/water interphase. Totally, a high flow velocity will normally reduce the potential for sulfide formation. Furthermore, the flow conditions affect the air–water exchange processes, e.g., the emission of hydrogen sulfide (cf. Chapter 4).

The anaerobic residence time of the wastewater during transport is a factor that affects the level of sulfide concentration in the wastewater. The residence time in a pressure pipe is determined by the magnitude of wastewater inflow compared with the volume (length and diameter) of the pipe. The level of sulfide formation in a given pipe is subject to the diurnal variation of the inflowing

Hydrogen Sulfide in Sewer Networks

wastewater and to the precipitation pattern in combined sewered catchments (Figure 6.3).

Typically anaerobic conditions in pressure mains exist after a short initial aerobic phase, depending on the DO concentration of the incoming wastewater. The factors of importance for the sulfide production in a pressure sewer are dealt with in Section 6.2.3 and include wastewater and system characteristics.

The equation for determination of the sulfide produced in terms of the resulting concentration in the bulk water phase based on the areal sulfide production rate is: CS 2

CS 2

,

CS 2

d

,

u

ra

A tr V

(6.5)

where CS 2 CS 2 , d C S 2 ,u ra A/V tr

sulfide production transformed to the water phase (gS m 3) sulfide concentration at the end of the sewer section (gS m 3) sulfide concentration at the start of the sewer section (gS m 3) areal sulfide production rate (gS m 2 h 1) area-to-volume ratio, i.e. pipe surface area/pipe volume (m 1) anaerobic residence time in the pipe (h)

A number of simple equations have been developed for the prediction of the areal sulfide production rate, ra, in pressure pipes (Table 6.1 and Figure 6.4). All equations take into consideration that the sulfide generation is affected by

(1) ra

0.5 10

3

0.4 u BOD50.8SSO 1.139T 4 3

1.07T

(2) ra

0.228 10 COD

(3) ra

1 10 3 BOD5 1.07T

(4) ra

k* (CODS

(5) ra

a** (CODS

50)0.5

50)0.5 1.03T

Thistlethwayte (1972) Boon and Lister (1975)

20

Pomeroy and Parkhurst (1977)

20

1.07T

20

20 20

Hvitved-Jacobsen et al. (1988) Nielsen et al. (1998)

*k = 0.0015 typical Danish domestic wastewater without industrial sewage

k = 0.003 wastewater from mixed domestic and industrial (foodstuff) sources k = 0.006 wastewater mainly from foodstuff industries **a = 0.001–0.002 typical Danish domestic wastewater without industrial sewage a = 0.003–0.006 wastewater from mixed domestic and industrial (foodstuff) sources a = 0.007–0.010 wastewater with biodegradable organic matter from mainly foodstuff industries where BOD5 biochemical oxygen demand (gO2 m 3) COD chemical oxygen demand (gO2 m 3) CODS soluble COD (gO2 m 3) SSO4 sulfate concentration (gS m 3) k and a rate constants ( ) T temperature ( C) s = slope (m m 1) u = flow velocity (m s 1)

Hydrogen Sulfide in Sewer Networks

the concentration of the organic matter (BOD, total COD or soluble COD) and the temperature. Equations (2) to (5) imply that the sulfate concentration in the wastewater is high and nonlimiting for sulfide formation. This is typically the case for sulfate concentrations exceeding 5–15 gS m 3 (Nielsen and Hvitved-Jacobsen, 1988). Equations (1) and (2) correspond to the expected maximum formation rates under “optimal” conditions, i.e., they are conservative estimates. To some extent, Equations (4) and (5) take into consideration the quality of the wastewater in terms of its biodegradability by including a biodegradable fraction of the soluble COD (CODS 50) and different rate constants depending on the source of the wastewater. The expressions shown in Table 6.1 all include constants that have been found good approximate values based on experiments. These values may of course be adjusted to account for specific cases. As an example, different flow conditions in continuously and intermittently pumped mains may affect the transfer of substances and products across the biofilm-water interface, and, thereby, the production of sulfide (Melbourne and Metropolitan Board of Works, 1989).

The balance between the DO consuming processes in wastewater and biofilm and the reaeration determines whether anaerobic conditions and thereby sulfide problems in gravity sewers exist or not. Typically, hydrogen sulfide does not occur in wastewaters from gravity sewers if the DO concentration is higher than 0.2–0.5 g m 3 (USEPA, 1974). If the DO concentration becomes lower, either because of a high DO consumption rate or because of reduced reaeration, Equation (3) in Table 6.1 has been proposed for the prediction of sulfide (USEPA, 1974). However, even where

sulfide buildup in a gravity sewer is observed, ra is usually less than a third of what would be predicted for a full-flowing sewer where oxygen is completely excluded. Also, the other equations in Table 6.1 may result in a corresponding prediction of sulfide formation in gravity sewers at such low DO concentrations if this information is taken into account. The release of H2S into the sewer atmosphere and a possible oxidation of sulfide in the water phase must be taken into account as outlined in Figure 4.4 (Pomeroy and Parkhurst, 1977; Tchobanoglous, 1981; Wilmot et al., 1988). A rather simple evaluation of the potential for sulfide problems in gravity sewers with a diameter smaller than about 0.6 m can be done using the so-called Z-formula (Pomeroy and Parkhurst, 1977; ASCE, 1989; ASCE and WPCF, 1982): Z

BOD5 (s 0.5Q 0.33 ) 1 (P / b)1.07T

20

(6.6)

where T s Q P b

temperature ( C) slope ( ) flow (ft3 s 1, “cubic feet per second”), 1 m3 wetted pipe-wall perimeter (m) pipe width at the water surface (m)

35.314 ft3

The following variant of Equation (6.6) has been proposed by Boon (1995): Z

3BOD5 (s 0.5Q 0.33 ) 1 (P / b)1.07T

20

(6.7)

where T s Q P b

temperature ( C) slope (m per 100 m) flow (L s 1) wetted pipe-wall perimeter (m) pipe width at the water surface (m)

The size of the calculated Z-value determines the estimated magnitude of the sulfide problem (Table 6.2). As an example, a half-filled 0.6 m diameter gravity sewer, wastewater at 20 C with BOD5 250 g m 3, and sewer pipe slopes of s 0.03, 0.1 and 0.3%, respectively, results in Z-values as shown in Table 6.3. Z-values from Table 6.3 indicate that significant sulfide problems may easily occur in trunk and intercepting gravity sewers with moderate slope.

Hydrogen Sulfide in Sewer Networks

Z < 5000 5000 < Z < 10000 Z > 10000

Problems occur rather infrequently Risk of sulfide problems Risk of sulfide problems frequent

Because of the release of H2S into the atmosphere and sulfide oxidation in the water phase, it is more complex to predict the generation of sulfide that may occur in a gravity sewer at very low DO concentrations, compared with a pressure main where the DO concentration is zero. The Z-formula is, in this respect, an equation for screening purposes. However, Pomeroy and Parkhurst (1977) developed an empirical equation for the quantification of sulfide buildup in partly filled (gravity) sewers. This equation includes two terms: ra

M BOD51.07T

20

N (s u)3/ 8 dm1 RCS 2

M BOD51.07T

20

N (s u)3/ 8

P b

(6.8)

1

CS 2

where areal sulfide production rate (gS m 2 h 1) coefficient (m h 1) temperature ( C) coefficient slope (m m 1) flow velocity (m s 1) hydraulic mean depth of the water phase, i.e., the cross-sectional area of the water volume divided by the water surface width (m) R hydraulic radius, i.e., the cross-sectional area of the water volume divided by the wetted perimeter (m)

ra M T N s u dm

0.03 0.1 0.3

0.32 0.58 1.00

45 82 141

19,000 35,000 61,000

19,370 8740 4220

CS 2 sulfide concentration (gS m 3) P wetted pipe-wall perimeter (m) b pipe width at the water surface (m)

The first term in Equation (6.8) is basically equal to Equation (3) in Table 6.1, and thereby corresponds to the formation rate of sulfide. However, the areal sulfide formation rate is normally found to be lower in a gravity sewer than in a pressure main, probably because of the effects of the daily changing water level and flow conditions. The value of M is therefore typically lower than shown in Equation (3). Pomeroy and Parkhurst propose M 0.32 10 3 m h 1. The second term in Equation (6.8) corresponds to the sinks for sulfide in the water phase that, according to Figure 4.4, are primarily caused by oxidation in the water phase and emission into the sewer atmosphere. Pomeroy and Parkhurst (1977) propose values for N at two levels, N 0.96 and N 0.64. The first value corresponds to a median buildup of sulfide, whereas the last value is a conservative estimate for prediction of sulfide buildup in a sewer. The second term of Equation (6.8) shows that the removal of sulfide from the water phase is considered a 1-order reaction in the sulfide concentration. The term also includes elements related to the reaeration and, thereby, the emission of hydrogen sulfide [cf. Equations (3) and (6) in Table 4.7 and Section 4.3.2]. Although based on intensive investigations, Equation (6.8) must — as an empirical model — be applied with caution. It has already been mentioned that sulfide in gravity sewers probably will not appear if the DO concentration is higher than 0.2–0.5 gO2 m 3. Equation (6.8) which is valid for partly filled gravity sewers, has been reformulated by Matos (1992) to determine the downstream concentration of sulfide: CS 2

,d

C1 (CS 2

,u

C1 )eC2

(6.9)

where C1 C2 L

M BOD5 (s u) N

(3/ 8)

P b

LNs3/ 8 3600 dm u 0.625 sewer length (m)

Depending on sewer design and hydraulic conditions, sewer solids may temporarily or more permanently accumulate as sediments in gravity sewer networks (cf. Section 3.2.8). In sanitary sewers, this may depend on the daily

Hydrogen Sulfide in Sewer Networks

fluctuations in the flow and in combined sewer networks especially affected by the difference between dry- and wet-weather flows. At the sewer sediment surface, a corresponding biofilm will develop that performs as a “thick biofilm.” Anaerobic conditions in the sediment normally prevail, and corresponding processes proceed. Focusing on sulfide formation, the sediment is often simply taken into account by considering it covered with a biofilm. The potential for sulfide production in terms of the surface flux will typically exceed what is observed for sewer biofilms, e.g., being 50–100% higher (Schmitt and Seyfried, 1992; Bjerre et al., 1998).

A number of impacts and effects are closely related to the presence of anaerobic conditions in wastewater of sewers. The major aspects are as follows: • • • • •

health-related effects odor problems concrete and metal corrosion inflow of anaerobic wastewater into treatment plants combined sewer overflows into receiving waters The first three aspects are related to the release of volatile substances into the gas phase of the sewer and from there into the urban atmosphere. These volatile compounds are H2S and organic odorous compounds produced under anaerobic conditions in the wastewater or associated biofilm and sediment. The first two mentioned effects, human health and odorous-related effects, are dealt with in Chapter 4 in relation to the exchange processes between the water and the gas phase in a sewer and will not be further described. As referred to in Table 4.6, sulfide concentration levels of 2 gS m 3 can generally be associated with minor, medium and considerable problems, respectively.

Concrete corrosion is associated with the formation of hydrogen sulfide. Hydrogen sulfide-induced concrete corrosion has been a well-known major consequence of anaerobic conditions in sewers for many years (Parker, 1945a, 1945b, 1951; Fjerdingstad, 1969; Thistlethwayte, 1972; USEPA, 1974). Concrete corrosion is still a worldwide phenomenon that has great economic impact (Vincke et al., 2000). As long as sulfide remains in the water phase, no harmful effect will occur. The concrete corrosion problem is caused by hydrogen sulfide that from the gas phase is absorbed in the liquid film that exists on moist concrete surfaces in the

sewer system. The concrete surface that is especially corroded is typically close to the anaerobic water phase and turbulent areas from where the release of hydrogen sulfide may be high. Furthermore, the sewer crown is also exposed to corrosion. At such moist surfaces, oxygen will typically be available from the sewer atmosphere, and H2S is oxidized to sulfuric acid by microbial reactions: H2 S 2O2

(6.10)

H2 SO4

The principle of corrosion in a concrete sewer pipe is shown in Figure 6.5. The aerobic bacteria responsible for this oxidation of hydrogen sulfide to sulfuric acid belong to the aerobic and autotrophic Thiobacillus family (Sand, 1987; Milde et al., 1983). These bacteria may be active at rather low pH values. Thiobacillus concretivorus is active at pH values between about 0.5 and 5 and may produce solutions of sulfuric acid up to about 7%. To be active, it requires that other species of the Thiobacillus family bring down the pH value. Two of the species, Thiobacillus concretivorus and Thiobacillus neapolitanus, are, in addition to sulfide, also able to use thiosulfate and elemental sulfur as energy sources. The sulfuric acid that is generated on the moist sewer surfaces may react with the alkaline cement in the concrete. A simple stoichiometry of this reaction is given by the following expression: H2 SO4

CaCO3 (Cement)

H2 O CO2

CaSO4

(6.11)

(Gypsum)

If the formation rate of sulfuric acid is low, a major part of the sulfuric acid

Hydrogen Sulfide in Sewer Networks

will react with the cement and leave a material of loose bound inert components, e.g., sand and gravel. On the other hand, if the formation rate of sulfuric acid is relatively high, a part of the sulfuric acid will be washed away before reaction and turn up in the wastewater and here react with alkaline components under the formation of sulfate ions. The sulfate ions that are produced in the liquid film on the sewer walls may, as an associated effect, cause a chemical sulfate attack on the concrete material. Equations (6.10) and (6.11) imply that 32 g H2S-S through the formation of sulfuric acid has a potential for the reaction with 100 g of CaCO3 of the cement in the concrete. The corrosion rate of the concrete can be expressed as follows (USEPA, 1974): rcorr

100 f 32 A

(6.12)

where rcorr corrosion rate of the concrete surface (g m 2 h 1) f rate of hydrogen sulfide absorption on the concrete surface (g m h 1) A alkalinity of the concrete material in units of CaCO3*

2

Equation (6.12) can be reformulated. If the right side of the equation is divided by the density of the concrete that is about 2.4 106 g m 3, the area-based corrosion rate can be transferred to a corrosion rate given in units of depth per time: f (6.13) c 11.4 A where c is the corrosion rate (mm y 1). Equation (6.13) is an expression assuming that all sulfuric acid formed will react with the cement. For practical reasons, the following expression can be formulated: c

k11.4

f A

(6.14)

where k is the correction factor ( ). *A standard test procedure for determination of CaCO equivalents may be found in the Encyclope3 dia of Industrial Chemical Analysis, Interscience Publishers Division, John Wiley & Sons, New York, vol. 15, p. 230.

For systems where the formation rate of sulfuric acid is low, k is approaching 1. Where this formation rate is high, k may be about 0.3–0.4. In cases of severe concrete corrosion, a corrosion rate of 4–5 mm y 1 may be observed (Mori et al., 1991). A reliable concrete corrosion rate is difficult to predict. As already mentioned and also shown in Figure 4.4, it requires that several process and exchange rates in terms of primarily sulfide formation, emission to the sewer atmosphere, sulfide absorption and sulfide oxidation on the sewer walls can be determined. There are several examples from the literature that show that corrosion has seriously (and often quickly) deteriorated sewer networks (EWPCA, 1982; Aldred and Eagles, 1982; ASCE, 1989). Although corrosion is difficult to predict, the number of examples and extent of the problems observed have given a comprehensive knowledge of where and when concrete corrosion may exist. This knowledge can briefly be summarized as follows. A concrete corrosion rate is typically relatively low at circumstances with both moderate temperatures (< 20ºC) and relatively low sulfide concentrations in the wastewater (< 0.5 gS m 3). The following may increase the risk for corrosion: • The concrete sewer network or pumping station that is considered is located downstream of systems with risk for a high sulfide formation. Such systems may primarily include pressure mains and also gravity sewers with permanent deposits of sewer solids. • Systems that are exposed to excessive turbulence of anaerobic wastewater and a potential increased release of hydrogen sulfide. Systems with a risk for increased turbulence are inlet structures, drops, cascades, sharp bends and inverted siphons. As an example, changes in the flow regime from a pressure pipe into a gravity sewer may give rise to the release of hydrogen sulfide. Corrosion of the sewer pipe wall is often pronounced near the daily water surface level and at the crown, because such surface areas are especially exposed to moisture. Estimating an average corrosion rate, the sewer life expectancy can be calculated assuming the thickness of an allowable concrete loss.

Hydrogen sulfide is a weak acid that reacts with most heavy metals and produces a corresponding metal sulfide with a low solubility in water. With a divalent metal, the total reaction is as follows:

Hydrogen Sulfide in Sewer Networks

H2 S Me

MeS H2

(6.15)

Metal corrosion has often been observed in pumping stations and sewer structures with electronic equipment.

Inlet structures at treatment plants often increase the turbulence of the wastewater. A corresponding release of hydrogen sulfide and other odorous substances into the air from inflow of anaerobic wastewater may be a nuisance that will require treatment of the gas released. On the other hand, aeration of the anaerobic wastewater may quickly oxidize the sulfide and some of the organic odorous substances. However, the N-containing odorous substances are slowly oxidized and may be a problem whenever they occur in the inflow to the treatment plant (Hwang et al., 1995). Inflow of wastewater containing sulfide to activated sludge treatment plants may lead to a change in floc structure due to a reduction of Fe(III) in the flocs to FeS (Nielsen and Keiding, 1998). The change has been observed as a weakening of the floc strength leading to potential disintegration of the flocs. Release of up to 10% of the total organic matter of the flocs has been observed. Extracellular polymeric substances (EPS), colloids and loosely adhered bacteria have been identified as being released.

The need to control sulfide problems in sewer networks has caused the development of a great number of strategies and procedures for that purpose. The major effort to avoid sulfide-related problems should be dealt with in the planning and design phase of the sewer network. It is important to stress that, if possible, operational procedures should only be implemented for systems that have developed sulfide problems. However, pressure mains may require that such methods be applied and included in a design process. Control methodologies can be subdivided as follows: • Design procedures for active control of sulfide problems — procedures that are implemented in the design phase of the sewer network to reduce the conditions for sulfide formation. • Design procedures for passive control of sulfide problems — procedures used in the design phase that aim at reducing the effect of sulfide but not necessarily the formation. • Operational procedures for control of sulfide problems — procedures that

aim at reducing sulfide formation or related effects by the use of treatment processes and maintenance strategies. The procedures are typically implemented after sulfide problems have been identified. The three main types of sulfide control procedures give an overall subdivision, although some overlap between the three groups may exist. Some of these procedures may be relevant not just for the control of sulfide but also for the reduction of other odorous substances. An overview of the methods emphasizing the process aspects will be given. The descriptions of the methods will not aim at giving information on detailed design principles.

These “active” design procedures for sewer systems include the following: • • • •

increase of the reaeration of the wastewater reduction of the turbulence of anaerobic wastewater reduction of conditions for sewer solids accumulation reduction of the biofilm thickness

When designing sewer networks, particularly gravity sewers, reaeration is the major process that should be focused on to reduce sulfide formation and the formation of organic odorous substances (cf. Section 4.4). A number of hydraulic and systems characteristics can be managed to increase the reaeration rate and avoid or reduce sulfide-related problems. The hydraulic mean depth, the hydraulic radius, the wastewater flow velocity and the slope of the sewer pipe are, in this respect, important factors that are dealt with in Section 4.4. It should be stressed that it is not necessarily the objective to avoid sulfide formation (in the sewer biofilm), but the sulfide that occurs in the bulk water phase should be at a low concentration level. Therefore, the DO concentration in the bulk water phase should not be lower than about 0.2–0.5 gO2 m 3, sufficiently high to oxidize sulfide before a considerable amount is emitted to the sewer atmosphere. As mentioned in Section 6.2.6, the turbulence of anaerobic wastewater should be avoided because of the risk of sulfide problems. Otherwise, when dealing with aerobic systems, turbulence may enhance the reaeration. The thickness of the sewer biofilm affects sulfide formation. Reduction of the biofilm thickness by increasing the wastewater velocity may lead to reduced sulfide problems. At very low velocities in an arerobic gravity sewer, a biofilm thickness may be more than 50 mm; however, it may be substantially reduced to typically 1–5 mm when the velocity is increased. The thickness of

Hydrogen Sulfide in Sewer Networks

the biofilm is a function of the shear stress on the pipe wall caused by the wastewater passing by. The average shear stress around the walled perimeter is as follows: g s R

(6.16)

where average shear stress on the wetted surface (N m 2) density of wastewater (kg m 3) g gravitational acceleration (m s 2) s slope of the pipe (m m 1) R hydraulic radius, i.e., the cross-sectional area of the water volume divided by the welled perimeter (m) A critical shear stress, crit between 3 and 4 N m 2, has been recommended to prevent the formation of a thick biofilm. Melbourne and Metropolitan Boards 3.4 N m 2. In a small diameter pipe ( 1 m, the velocity should exceed 1.2–1.4 m s 1. It is important to design the sewer system to avoid permanent solids accumulation, as the development of deposits may cause an enhanced sulfide formation rate compared with a sewer biofilm. Solids deposition and resuspension in sewers will not be dealt with in this text. An overview of these physical properties is found in Ashley and Verbanck (1998). Examples of corrosion-resistant designs of sewer systems are found in Kienow and Pomeroy (1978) and ASCE (1989), for example.

In addition to the “active” design principles for reducing the risk of sulfide problems, a number of more “passive” principles exist. The following design considerations are especially relevant for the reduction of corrosion: selection of corrosion-resistant materials and design of a well-ventilated sewer system. The concrete material used for sewer construction is typically based on the use of Portland cement. Different types of Portland cement have not shown significant differences in the corrosion rate of concrete. However, an increase of the relative amount of cement used in the concrete reduces the corrosion rate — in units of mm y 1 — according to the increase in the alkalinity per unit volume of the concrete material (Grennan et al., 1980). The use of high-alkaline materials

as aggregates in concrete, e.g., limestone and dolomite, will also decrease the corrosion rate. Corrosion-resistant types of concrete exist, but they are typically relatively expensive. Such types may be based on supersulfated or aluminous cement or cement with a high content of sulfur. The concrete surface may also be protected by coating or painting, e.g., epoxy resin and coal-tar. The impermeability of such coatings is important to resist sulfide attack. Furthermore, alternatives for sewer construction exist in terms of PVC, ABS and PE plastics materials, for example. Also, pipes of glazed earthenware have shown good resistance to sulfide corrosion. Ventilation of sewers may not only reduce the hydrogen sulfide concentration in the sewer atmosphere but also the moisture that is a fundamental requirement for establishment of microbial activity on the sewer walls. It is important that the ventilation be well controlled; otherwise, odorous problems in the vicinity of the sewer network may occur. In some cases, operational procedures like treatment of the vented air, e.g., by wet oxidation, by chemical scrubbing or by passing a biofilter, may need to be considered.

Operational procedures for the control of sulfide problems have played an important role for existing sewer systems over the last 40–50 years. The reason is that sulfide problems have not always been considered and predicted in the design phase, or it has been acceptable to deal with sulfide problems in the daily operation of the sewer network. However, in pressure mains, sulfide formation may typically take place. In such systems, hydrogen sulfide control may be needed, and procedures that are operated by the municipality should be implemented. Table 6.4 outlines methods that may be used for such control. Some of these methods will be further considered. The details are described in the literature. Further information of relevance in this respect is found in Melbourne and Metropolitan Board of Works (1989), ASCE and WPCF (1982), ASCE (1989), USEPA (1974, 1985), Pomeroy et al. (1985) and Vincke et al. (2000). The methods outlined in Table 6.4 will often require the addition of chemicals. As a general concern, whether or not such methods give negative effects, e.g., on the successive wastewater treatment processes or in the receiving waters, must be considered. Methods that have an oxidizing effect may also typically reduce malodors. Chemical precipitation, however, is only effective for the control of sulfide problems. In the following, further interpretation of the methods shown in Table 6.4 will be given.

Hydrogen Sulfide in Sewer Networks

Prevention of sulfate-reducing conditions Prevention of adverse effects Methods aiming at specific effects on the biological system Mechanical methods Treatment methods

Addition to the wastewater of: air pure oxygen nitrate Chemical precipitation of sulfides by: iron(II) sulfate iron(III) chloride Alkaline substances increasing pH Chlorine Hydrogen peroxide Ozone Flushing Ball for detachment of biofilm Chemical Biological

These methods have typically been applied in systems with raising (pressure) mains, although they may also be used in gravity sewer systems. Injection of air: the oxygen in the injected air will prevent sulfate-reducing conditions in the sewer. The DO concentration in the wastewater establishes an aerobic upper layer in the biofilm, and sulfide produced in the deeper part of the biofilm or the deposits that may diffuse into the water phase will be oxidized (cf. Figure 6.2). The oxidation of sulfide will mainly proceed as a chemical process, although microbial oxidation may also take place (Chen and Morris, 1972). Factors that affect the oxidation rate of sulfide include pH, temperature and presence of catalysts, e.g., heavy metals. It is important to note that the amount of oxygen needed to avoid sulfate-reducing conditions is determined by the aerobic respiration rate of the wastewater and the biofilm and not the potential amount of total sulfide production in the sewer. The relatively low solubility of oxygen (9–11 gO2 m 3) in wastewater compared with the DO consumption rate typically requires that oxygen must be injected at several points of a sewer pipe to ensure aerobic conditions. This is, of course, expensive and requires manpower in terms of operation and maintenance. Furthermore, the readily biodegradable and fast hydrolyzable fractions of the organic matter may be depleted (Tanaka et al., 2000b). In the case of requirement for mechanical treatment, this is positive;

however, the opposite is the case if the subsequent treatment requires denitrification and biological phosphorus removal. A great number of technical systems for air injection can be implemented. Boon (1995) proposed a method that reduces the requirement for oxygen in a rising main. In this system, air is injected continuously at the discharge end of the pipe, where the wastewater is recirculated in a relatively short length of the sewer line close to its end. When injecting air in pressure mains, it must be taken into account that the volume of oxygen only amounts to about 20%. The inert N2 gas — to some extent mixed with the incoming air — will be collected at the top part of a pipe. An automatic vent system to remove this gas may be installed. The pumping performance of an air-injected pressure main can be evaluated using a model for the energy loss. Injection of pure oxygen: the injection of pure oxygen overcomes some of the problems compared with air. The solubility of pure oxygen in water is compared with oxygen in air increased with a factor of about 5, i.e., it is about 45–50 gO2 m 3. Furthermore, the amount of inert gas to be managed is reduced to about nil. The disadvantage is that pure oxygen must be available at the injection point, transported and kept in containers or produced at the site. Addition of nitrate: the addition of nitrate will establish anoxic conditions when DO is depleted and will, thereby, suppress the sulfate reduction. The theoretical details of the anoxic processes in the bulk water phase and in the biofilm on the suppression of sulfate-reducing conditions and the presence of sulfide in the water phase are not well understood (Abdul-Talib et al., 2001). However, nitrate should be sufficiently active to oxidize sulfide (Einarsen et al., 2000). The microbial activity of wastewater under anoxic conditions is lower compared with aerobic conditions (Abdul-Talib et al., 2001). This is important to consider, because a low nitrate uptake rate (NUR) compared with the oxygen uptake rate (OUR) in units of electron equivalents means a reduced transformation rate of the most biodegradable fractions of the organic matter. As mentioned under the point on injection of air, this may have implications in terms of treatment. Furthermore, a relatively low NUR value also has operational advantages because of a reduced demand for nitrate to suppress sulfide formation. Nitrate can be added as different salts, e.g., Ca(NO3)2, well known as a fertilizer product. Inflow of nitrate into a treatment plant and into receiving waters should typically be avoided. The rate of nitrate addition to the wastewater must be controlled (Bentzen et al., 1995; Einarsen et al., 2000). The effectiveness of nitrate addition to control sulfide has been investigated in a sewer network in the Lake Balaton catchment, Hungary (Jobbágy et al., 1994). Figure 6.6 shows the effect of nitrate addition on the H2S concentration in the headspace of a manhole located downstream of the station where nitrate was added. The importance of the anoxic removal of organic matter at the inlet

Hydrogen Sulfide in Sewer Networks

to the wastewater treatment plant is demonstrated in Figure 6.7. This figure also shows that the nitrate addition can be controlled in terms of low concentrations at the influent to the treatment plant.

Methods for chemical precipitation of sulfides: the negative effects of sulfide can be avoided by adding metal salts. The most common salts are iron(II) and iron(III) as sulfate or chloride (Hvitved-Jacobsen et al., 1988; Jameel, 1989). In anaerobic wastewater, Fe(III) will be reduced to Fe(II) and the following precipitation of the highly insoluble iron sulfide, FeS, will proceed as an almost immediate process: Fe

HS

FeS H

(6.17)

It is important to note that addition of iron salts keeps the sulfide as a precipitate in the wastewater and hinders the emission to the sewer atmosphere and the following negative effects. Precipitation of sulfide normally has no effect on the formation of sulfide. Sulfate is needed as an electron acceptor for the sulfate reduction process, but normally, it is available in unlimited concentrations in wastewater. If this is not the case, iron sulfate may support the process. The anaerobic fermentation processes still proceed, and the odorous organic substances produced are generally not affected by the addition of iron salts. The iron salts are normally added to the wastewater in dissolved form and under aerobic conditions. The iron(III) ion is, therefore, hydrated and reacts as an acid: Fe(H2 O)+++ x

Fe(H2 O) x 1 (OH)

H

(6.18)

Noncorrosive materials should be used for the dosing equipment. The addition of iron salts also reduces the alkalinity of the wastewater, i.e., it may decrease the pH value. An effect of a low alkalinity of wastewater is a potential reduction of the nitrification rate in the subsequent treatment process. Iron salts may be added before as well as after the sulfide is produced, i.e., either upstream or downstream in the sewer system. The very low solubility of FeS implies that the process, for all practical reasons, proceeds at 100% efficiency. The stoichiometry expressed by Equation (6.17) shows that addition of an iron salt requires 1.75 g Fe per g of (H2S + HS )-S. The precipitated FeS occurs in the wastewater as small suspended particles that make the color black. Iron salts normally occur in wastewater in small, however, not necessarily insignificant amounts. Therefore, a small amount of the sulfide produced in a sewer will typically be removed by precipitation of the iron that naturally occurs. Although a number of precautions have been mentioned for the control of sulfide by precipitation as FeS, this methodology is often considered not only acceptable but also cheap and efficient. Added iron salts may also be reused in the treatment process. Under aerobic conditions in the wastewater treatment plant, the amorphous FeS is fast oxidized, and the iron can be used for chemical removal of phosphate.

A number of chemicals have been used to control sulfide. The following are the most commonly used with different effects on the biological processes. Alkaline substances increasing pH: shock dosing with alkaline chemicals at

Hydrogen Sulfide in Sewer Networks

pH values > about 12 will degrade the biofilm. An essential part of it will be detached. The effect on the sulfide formation is only temporary until a new biofilm has been reestablished, typically after a few days. In practice, sodium hydroxide (NaOH), i.e., caustic soda, and calcium hydroxide [Ca(OH)2] have been used. Any negative effects on the subsequent wastewater treatment processes should be carefully considered. Chlorine: the use of chlorine (Cl2) as a sulfide-controlling agent is based on its poisoning effect on the biological system and its oxidizing effect on sulfide. Chlorine is nonspecific, and a great number of side reactions with wastewater components may occur. Even concentrations of about 50 g m 3 have not efficiently been able to reduce sulfide concentrations of about 5–10 gS m 3. Chlorine is an environmentally problematic chemical that generally cannot be recommended for use in sewer systems. Hydrogen peroxide and ozone: contrary to chlorine, hydrogen peroxide (H2O2) is rather specific to oxidize sulfide to S or SO24 with S as the main product at a pH < about 8. Furthermore, hydrogen peroxide may degrade the sulfate-reducing bacteria in the biofilm. Although H2O2 is relatively expensive, it may be economical in its use. A concentration of 5–10 g m 3 may be sufficient to suppress sulfide formation. Also, ozone has been used for sulfide control.

Removal of sewer biofilm and deposits by flushing and use of a “cleaning ball” for detachment of biofilm and resuspension of sewer sediments are examples of mechanical methods for reducing sulfide occurrence.

At inlet structures to treatment plants and pumping stations, sewer gases may be emitted to the atmosphere. To avoid odor nuisance, this gas may need treatment. Such methods have been widely used in the mining and coal gas industries to remove hydrogen sulfide from gases (Herrygers et al., 2000). Chemical and biological methods exist. Chemical scrubbing by absorption in alkaline or oxidative solutions, e.g., ozone, is possible. Biological methods in terms of biofilters are also widely used by passing the polluted gas through a filter medium. Efficient and simple biofilters may use packing materials like iron ore [hydrated haematite or iron(III) oxide], peat or heather. At the surface of such materials and under moist and aerobic conditions, a biofilm will develop. In this biofilm, hydrogen sulfide, odorous substances produced in the sewer and VOCs discharged with the wastewater may be degraded. Often, a rather short residence time, e.g. 30–60 seconds, is typically needed to treat the polluted sewer gas.

A different approach for sewer gas cleaning is to use bioscrubbers. In a bioscrubber, the polluted gas is absorbed into a liquid and then degraded in an activated sludge process.

The difference between aerobic and anaerobic transformations of wastewater organic matter is crucial. From a basic point of view, however, still related to the sewer systems, aerobic and anaerobic microbial processes have been dealt with in Chapter 3. The aerobic transformations and a corresponding conceptual model were the main subjects of Chapter 5. According to what was dealt with in Chapter 3 and in Section 6.2, the major processes that may proceed in wastewater under anaerobic conditions are briefly as follows: • Anaerobic hydrolysis: this process transforms the hydrolyzable substrate, XSn, into fermentable, readily biodegradable substrate, SF. • Fermentation: this process transforms the fermentable, readily biodegradable substrate, SF, into fermentation products, SA (VFAs). • Methanogenesis: a process that transforms the fermentation products (VFAs) into methane. • Sulfate reduction: a process for the formation of hydrogen sulfide These anaerobic processes have been investigated under sewer conditions (Tanaka and Hvitved-Jacobsen, 1998, 1999, 2000; Hvitved-Jacobsen et al., 1999; Tanaka et al., 2000a, 2000b). The results from experiments in the laboratory, in a pilot sewer and under full-scale sewer conditions combined with theoretical considerations have provided knowledge on the anaerobic transformations of carbon that can be expressed conceptually (Figure 6.8). The concept expressed in Figure 6.8 is described in relatively simple terms. The most important parts are shown with full-drawn lines, whereas the dotted lines are generally less important for the formulation of a sewer process model. The processes can be described in further details, however, the major concern has been to establish a concept for which components and parameters can be experimentally determined without unrealistic resources for laboratory and field studies. Methods for this determination will be dealt with in Chapter 7. One of the details omitted in the concept depicted in Figure 6.8 is the growth of the sulfate-reducing biomass. Characklis et al. (1990) proposed Equation (6.19) for determination of the stoichiometry for the sulfate reduction process using lactate as the carbon source for energy requirement and growth. This equation can be used to evaluate the importance of the simplification.

Anaerobic Microbial Transformations of Organic Matter in Sewers

CH3CHOHCOOH 0.43 H2 SO4

0.067 NH3

lactic acid

0.33 CH1.4 N 0.2 O0.4 0.96 CH3COOH 0.43 H2 S 0.7 CO2

0.94 H2 O

biomass

(6.19) The yield constant, Y, for the sulfate-reducing biomass is, according to Equation (6.19), in mass units: Y

(g biomass)/(g lactic acid)

0.083 g g

1

(6.20)

The relatively low yield constant makes it acceptable to omit the growth process for the sulfate-reducing biomass in the concept. As shown in Figure 6.8, the most important part of the anaerobic sulfur cycle in terms of the sulfate respiration process can be integrated with the anaerobic carbon cycle. A fractionation of the readily biodegradable substrate (SS) into SF and SA fits well to the anticipation that mainly SF is used by the sulfate-reducing biomass in sewer biofilms. By integrating the sulfide formation in this way, a simple conceptual approach is obtained instead of the traditional empirical descriptions as depicted in Table 6.1.

The process rates for the transformations of wastewater organic matter under anaerobic conditions are quite different compared with the aerobic transformation rates (cf. Chapter 5). In this respect, Figure 6.9 outlines, as an example, the mass balance and flows of organic matter for a number of experiments performed under laboratory conditions (Tanaka and Hvitved-Jacobsen, 1999). The components and processes are included according to the concept shown in Figure 6.8. The results shown in Figure 6.9 give an error in the overall mass balance caused by the fact that average values of the experiments are shown. No methanogenesis was observed (cf. Figure 3.2). The results in Figure 6.9 show that the production rate of readily biodegradable substrate (SS) by the anaerobic hydrolysis is larger than the removal rate. It is a major result that readily biodegradable substrate is not just preserved but produced. This fact has positive implications on the subsequent treatment processes in terms of denitrification and biological phosphorus removal. On the other hand, it is negative if wastewater treatment is only required in terms of BOD removal.

The two concepts for aerobic and anaerobic transformations of organic matter shown in Figure 5.5 and Figure 6.8, respectively, can be integrated. Although the active biomass is different for aerobic and anaerobic processes, the same fractions of organic substrates — readily biodegradable and hydrolyzable — are relevant in both cases. The fractions of substrate and their formation and utilization form a natural link between the aerobic and anaerobic heterotrophic processes. Integration between the aerobic and anaerobic systems for wastewater transformations in sewers makes it possible to describe the performance of changing aerobic-anaerobic conditions. Several examples described in Chapter 5, e.g.,

An Integrated Aerobic-Anaerobic Model Concept

Figures 5.10 and 5.11, show that such conditions exist in gravity sewers that are subject to a variability that proceeds in time and space. Changing aerobic and anaerobic conditions also appear in pressure mains when injecting air or pure oxygen. The integrated aerobic-anaerobic concept based on Figures 5.5 and 6.8 is shown in Figure 6.10. In the water phase, the concept includes growth of the heterotrophic biomass, growth and nongrowth-related oxygen consumption, hydrolysis and fermentation. Concerning the biofilm processes, e.g., sulfate respiration, relatively simple descriptions in terms of surface-based transformations are used. Sewer sediment processes are not included in the model but can often be substituted by those taking place in a biofilm (Section 6.2.5). The improvement for prediction of sulfide formation compared to the empirical equations shown in Table 6.1 is that the wastewater quality in terms of its biodegradability is more directly expressed and included. Furthermore, sulfide formation can be predicted under changing aerobic/anaerobic conditions in gravity sewers and pressure mains. The conceptual model in terms of a matrix formulation is outlined in Table 6.6, and examples of components and parameters are listed in Tables 6.5, 6.7 and 6.8. Methods for the determination of these components and parameters will be dealt with in Chapter 7.

XBw XBf XS1 XS2 SF SA SS SO COD SH2S

Heterotrophic active biomass in the water phase Heterotrophic active biomass in the biofilm Hydrolyzable substrate, fast biodegradable Hydrolyzable substrate, slowly biodegradable* Fermentable substrate Fermentation products (i.e. VFAs) Readily biodegradable substrate (SF + SA) Dissolved oxygen Total COD Total sulfide

20–100

gCOD m

3

~10

gCOD m

2

50–100

gCOD m

3

300–450

gCOD m

3

0–40 0–20 0–40 0–4 about 600 0–5

gCOD m gCOD m gCOD m gO2 m 3 gCOD m gS m 3

3 3 3 3

*Includes very slowly biodegradable and inert organic matter.

Equation g in Table 6.6 that is selected for the simulation of the sulfide production rate originates from Nielsen et al. (1998). The dissolved COD used in the empirical model [Table 6.1, Equation (5)] as a measure of the wastewater quality does not exist among the COD components of the sewer process model shown in Table 6.5. A substitute for the term (CODS 50) in the original model (Table 6.1), interpreted as biologically active COD components for sulfate production, must therefore be found. It is a well-known fact that sulfate-reducing bacteria use readily biodegradable organic matter like alcohols, lactate, pyruvate and some aromatic substrates but generally, not directly, higher carbohydrates. Based on studies in pilot pressure sewer and field investigations, Tanaka and Hvitved-Jacobsen (1998) and Tanaka et al. (2000a) found that, among different options, the term (SS + XS1) was an acceptable substitute for (CODS 50) (Figure 6.11). Theoretically, experimentally and from a modeling point of view, there are good reasons for this substitution. The aerobic-anaerobic model concept presented for prediction of sulfide formation in sewers has further applications. The model is developed to simulate organic matter transformations and the change in biodegradability during transport of wastewater in sewer networks. The model is also a tool for predicting quality changes of wastewater in sewers that interact with the successive treatment processes and the water quality processes taking place in water courses during combined sewer overflows (Hvitved-Jacobsen and Vollertsen, 1998).

b.

SO ) XBw

1

h. K L a 24 ( SOS

SA

0.86 (1

KO ) A / V (T 20) r

(T 20) w

(T 20) w

XBf A / V )

0.20 Fr 2 ) (su)3 / 8 dm1

SO

XBf A / V )

(T 20) KO /( S

KO )( XBw

XS1 )0.5

SO ) where K L a

g. kH2 S 24 10 3 ( SF

(T 20) w

KO ) ( X Bw

XBf A / V )

(T 20) f

(T 20) w

S A ))

SO )( XBw

( SF

SO ) XBw

XSn / X Bw ) KO /( SO

XSn / XBw ) SO /( KO

SF ) KO /( SO

/ XBw ) /( K Xn

f . q fe SF /( K fe

fe khn ( X Sn

d. khn ( XSn / XBw ) /( K Xn

(T 20) w

S A ) /( KSf

S A )) SO /( KO

YHf /((1 YHf ) A / V ( SF

S A ) /( KSw ( SF

c. qm SO /( KO

k1/ 2 SO0.5

e.

1 1 1 1 1

1/YHw 1/YHf 1 1

1

S A is not sufficiently available to support the biomass maintenance energy requirement.

H ( SF

SF

a.

*If

Aerobic growth in bulk water Aerobic growth in biofilm Maintenance energy requirement Aerobic hydrolysis, fast Aerobic hydrolysis, slow Anaerobic hydrolysis, fast Anaerobic hydrolysis, slow Fermentation Hydrogen sulfide production Reaeration 1

1

1 1 ( 1*)

1 1

(1 YHw)/YHw (1 YHf)/YHf 1

Equation a Equation b Equation c Equation d, n Equation d, n Equation e, n Equation e, n Equation f Equation g Equation h

1 2 1 2

An Integrated Aerobic-Anaerobic Model Concept

Maximum specific growth rate for heterotrophic biomass Suspended biomass yield constant for heterotrophics Saturation constant for readily biodegradable substrate Saturation constant for dissolved oxygen (DO) Temperature coefficient in the water phase Maintenance energy requirement rate constant 1/2-order rate constant Biofilm yield constant for heterotrophic biomass Saturation constant for readily biodegradable substrate Efficiency constant for the biofilm biomass Temperature coefficient in the biofilm Hydrolysis rate constant, fraction 1 (fast) Hydrolysis rate constant, fraction 2 (slow) Saturation constant for hydrolysis, fraction 1

H

YHw Ksw KO w

qm k1/2 YHf KSf

f

kh1 kh2 KX1 KX2

Saturation constant for hydrolysis, fraction 2 Anaerobic hydrolysis reduction factor Maximum rate for fermentation Saturation constant for fermentation Hydrogen sulfide production rate constant Temperature coefficient for hydrogen sulfide production

fe

qfe Kfe kH S S

2

KLa T SOS Fr u g s dm A/V r

1

6.7

d

0.55 1.0

gCOD gCOD 1 gCOD m

0.05

gO2 m

1.07 1.0 4 0.55

— d1 0.5 gO0.5 d 2 m GCODg COD

5

gCOD m

0.15 1.05 5 0.5 1.5

3

1 1

3

— — d1 d1 gCOD gCOD

0.5 0.14 3 20 2(3) 1.030

Oxygen transfer coefficient Temperature Dissolved oxygen saturation concentration at T°C Froude number u(g dm) 0.5 Mean flow velocity Gravity acceleration Slope Hydraulic mean depth Ratio of biofilm area to bulk water volume Temperature coefficient for reaeration

3

1

gCOD gCOD

1

— d

1

gCOD gCOD

gS2 m 2 h —

9.81

1.024

1

1

h1 C gO2 m 3 — ms 1 ms 2 mm 1 m m1 —

An Integrated Aerobic-Anaerobic Model Concept

䊊

䊉

The use of the integrated aerobic-anaerobic sewer process model depicted in Table 6.6 will be exemplified in Chapter 8. Compared with the different empirical models that have been developed for prediction of transformations in sewers, it integrates and expands the conditions that can be taken into account: • simulation under aerobic, anaerobic or changing aerobic-anaerobic conditions • simulation of transformations in gravity sewers and pressure mains The model is, in several aspects, a simplification of the processes that occur. As an example, it is important — and also possible when further data are available from experiments — to expand the model by including the oxidation of sulfide in a gravity sewer at low DO concentrations and the emission of hydrogen sulfide into the sewer atmosphere. From a general point of view, however, it is considered important only to deal with the most important aspects to keep the number of processes low and not include too many process parameters that are site specific.

The sewer model is designed from a conceptual point of view and has potential for further applications. In Section 4.3.3, it was concluded that the occurrence of sulfide can be used as a pragmatic measure of malodors. Therefore, the sewer process model also has potential for the prediction of odor problems. Furthermore, as dealt with in Section 8.5.2, the model also predicts the aerobic transformations of suspended sediment particles in sewers (Vollertsen and Hvitved-Jacobsen, 1998, 1999; Vollertsen et al., 1998, 1999). The model is also a potential tool for simulation of the impacts from combined sewer overflows.

Abdul-Talib, S., T. Hvitved-Jacobsen, J. Vollertsen, and Z. Ujang (2001), Anoxic transformations of wastewater organic matter in sewers — process kinetics, model concept and wastewater treatment potential, INTERURBA II. Aldred, M.I and B. G. Eagles (1982), Hydrogen sulfide corrosion of the Baghdad trunk sewerage system, Water Pollut. Contr., 81(1), 80–96. ASCE (American Society of Civil Engineers) (1989), Sulfide in wastewater collection and treatment systems, ASCE manuals and reports on engineering practice no. 69, p. 324. ASCE (American Society of Civil Engineers) and WPCF (Water Pollution Control Federation) (1982), Gravity sanitary sewer design and construction, ASCE manuals and reports on engineering practice no. 60 or WPCF manual of practice no. FD-5, p. 275. Ashley, R.M. and M.A. Verbanck (1998), Physical processes in sewers, Congress on Water Management in Conurbations, Bottrop, Germany, June 19–20, 1997. In Emschergenossenschaft: Materialen zum umbau des Emschersystems, Heft 9, 26–47. Bentzen, G., A.T. Smith, D. Bennett, N. J. Webster, F. Reinholt, E. Sletholt, and J. Hobson (1995), Controlled dosing of nitrate for prevention of H2S in a sewer network and the effects on the subsequent treatment processes, Water Sci. Tech., 31(7), 293–302. Bjerre, H.L., T. Hvitved-Jacobsen, S. Schlegel, and B. Teichgräber (1998), Biological activity of biofilm and sediment in the Emscher river, Germany, Water Sci. Tech., 37(1), 9–16. Boon, A.G. (1995), Septicity in sewers: Causes, consequences and containment, Water Sci. Tech., 31(7), 237–253. Boon, A.G. and A.R. Lister (1975), Formation of sulfide in rising main sewers and its prevention by injection of oxygen, Progress in Water Technology, 7, 289–300. Characklis, W.G., W.C. Lee, and S. Okabe (1990), Kinetics and stoichiometry of planktonic and biofilm (sessile) sulfate-reducing bacteria, report of Inst. Biological and Chemical Process Analysis, Montana State University, Bozeman, MT. Chen, K.Y. and J.C. Morris (1972), Kinetics of oxidation of aqueous sulphide by oxygen, Envir. Sci. Technol., 6, 529–537. Einarsen, A.M., A. Aesoey, A.-I. Rasmussen, S. Bungum, and M. Sveberg (2000), Biological prevention and removal of hydrogen sulphide in sludge at Lillehammer Wastewater Treatment Plant, Water Sci. Tech., 41(6), 175–182. EWPCA (European Water Pollution Control Association) (1982), Proceedings of EWPCA International State-of-the-art Seminar on Corrosion in Sewage Plants, Hamburg, January 28–29, 1982.

References Fjerdingstad, E. (1969), Bacterial corrosion of concrete in water, Water Res., 3, 21–30. Grennan, J.M., J. Simpson, and C.D. Parker (1980), Influence of cement composition on the resistance of asbestos cement sewer pipes to H2S corrosion, Corrosion Australasia, 5(1), 4–5. Herrygers, V., H. van Langenhove, and E. Smet (2000), Biological treatment of gases polluted by volatile sulfur compounds. In: P.N. L. Lens and L. H. Pol (eds.), Environmental Technologies to Treat Sulfur Pollution — Principles and Engineering, IWA Publishing, pp. 281–304. Hvitved-Jacobsen, T. and J. Vollertsen (1998), An intercepting sewer from Dortmund to Dinslaken, Germany, report submitted to the Emschergenossenschaft, Essen, Germany, p. 35. Hvitved-Jacobsen, T., K. Raunkjær, and P. H. Nielsen (1995) Volatile fatty acids and sulfide in pressure mains, Water Sci. Tech., 31(7), 169–179. Hvitved-Jacobsen, T., J. Vollertsen, and N. Tanaka (1999), Wastewater quality changes during transport in sewers — an integrated aerobic and anaerobic concept for carbon and sulfur microbial transformations, Water Sci. Tech., 39(2), 242–249. Hvitved-Jacobsen, T., B. Jütte, P. Halkjær Nielsen, and N.Aa. Jensen (1988), Hydrogen sulphide control in municipal sewers. In: H. H. Hahn and R. Klute (eds.), Pretreatment in Chemical Water and Wastewater Treatment, proceedings of the 3rd International Gothenburg Symposium, Gothenburg, Sweden, June 1–3, 1988, Springer-Verlag, New York/Berlin, pp. 239–247. Hwang, Y., T. Matsuo, K. Hanaki, and N. Suzuki (1995), Identification and quantification of sulfur and nitrogen containing compounds in wastewater, Water Sci. Tech., 29(2), 711–718. Jameel, P. (1989), The use of ferrous chloride to control dissolved sulfides in interceptor sewers, J. Water Pollut. Contr. Fed., 61(2), 230–236. Jobbágy, A., I. Szántó, G. I. Varga, and J. Simon (1994), Sewer system odour control in the Lake Balaton area, Water Sci. Tech., 30(1), 195–204. Kienow, K.K. and R.D. Pomeroy (1978), Corrosion resistant design of sanitary sewer pipe, ASCE (American Society of Civil Engineers) Convention and Exposition, Chicago, IL, October 16–20, 1978, p. 25. Matos, J.S. (1992), Aerobiose e septicidade im sistemas de drenagem de águas residuais, Ph.D. thesis, IST, Lisbon Portugal. Melbourne and Metropolitan Board of Works (1989), Hydrogen sulphide control manual — septicity, corrosion and odour control in sewerage systems, Technological Standing Committee on Hydrogen Sulphide Corrosion in Sewerage Works, vols. 1 and 2. Meyer, W.J. and G.H. Hall (1979), Prediction of sulfide generation and corrosion in concrete gravity sewers: A case study, J. B. Gilbert & Associates, A Division of Brown and Caldwell, Sacramento, CA. Milde, K., W. Sand, W. Wolf, and E. Bock (1983), Thiobacilli of the corroded concrete walls of the Hamburg sewer system, J. General Microbiology, 129, 1327–1333. Miljøstyrelsen (1988), Hydrogen sulfide formation and control in pressure mains, The Danish Environmental Protection Agency, project report no. 96, p. 109 (in Danish). Mori, T., M. Koga, Y. Hikosaka, T. Nonaka, F. Mishina, Y. Sakai, and J. Koizumi (1991), Microbial corrosion of concrete pipes, H2S production from sediments and determination of corrosion rate, Water Sci. Tech., 23(7–9), 1275–1282. Nielsen P.H. (1991), Sulfur sources for hydrogen sulfide production in biofilm from sewer systems. Water Sci. Tech. 23, 1265-1274. Nielsen, P.H. and K. Keiding (1998) Disintegration of activated sludge flocs in the presence of sulfide, Water Res., 32(2), 313–320.

Nielsen, P.H. and T. Hvitved-Jacobsen (1988), Effect of sulfate and organic matter on the hydrogen sulfide formation in biofilms of filled sanitary sewers, J. WPCF, 60, 627–634. Nielsen, P.H., K. Raunkjaer, and T. Hvitved-Jacobsen (1998), Sulfide production and wastewater quality in pressure mains, Water Sci. Tech., 37 (1), 97–104. Parker, C.D. (1945a), The corrosion of concrete 1. The isolation of a species of bacterium associated with the corrosion of concrete exposed to atmospheres containing hydrogen sulphides, Aust. J. Expt. Biol. Med. Sci., 23, 81–90. Parker, C.D. (1945b), The corrosion of concrete 2. The function of Thiobacillus concretivorus (nov. spec.) in the corrosion of concrete exposed to atmospheres containing hydrogen sulphide, Aust. J. Expt. Biol. Med. Sci., 23, 91–98. Parker, C.D. (1951), Mechanics of corrosion of concrete sewers by hydrogen sulfide, Sewage Ind. Wastes, 23, 1477–1485. Pomeroy, R.D. and F.D. Bowlus (1946), Progress report on sulphide control research, Sewage Works J., 18 (4). Pomeroy, R.D. and J.D. Parkhurst (1977), The forecasting of sulfide buildup rates in sewers, Prog. Water Techn., 9 (3), 621–628. Pomeroy, R.D., J.D. Parkhurst, J. Livingston, and H.H. Bailey (1985), Sulfide occurrence and control in sewage collection systems, Technical Report, US Environmental Protection Agency, USEPA 600/X-85-052, Cincinnati, OH. Sand, W. (1987), Importance of hydrogen sulfide, thiosulfate and methylmercaptan for growth of thiobacilli during simulation of concrete corrosion, Applied and Environmental Microbiology, 53(7), 1645–1648. Schmitt, F. and C.F. Seyfried (1992), Sulfate reduction in sewer sediments, Water Sci. Tech., 25(8), 83–90. Tanaka, N. and T. Hvitved-Jacobsen (1998), Transformations of wastewater organic matter in sewers under changing aerobic/anaerobic conditions, Water Sci. Tech., 37 (1), 105–113. Tanaka, N. and T. Hvitved-Jacobsen (1999), Anaerobic transformations of wastewater organic matter under sewer conditions. In: I. B. Joliffe and J. E. Ball (eds.), Proceedings of the 8th International Conference on Urban Storm Drainage, Sydney, Australia, August 30–September 3, 1999, pp. 288–296. Tanaka, N. and T. Hvitved-Jacobsen (2001), Sulfide production and wastewater quality — investigations in a pilot plant pressure sewer, Water Science and Technology, 43(5), 129–136. Tanaka, N., T. Hvitved-Jacobsen, and T. Horie (2000a), Transformations of carbon and sulfur wastewater components under aerobic-anaerobic transient conditions in sewer systems, Water Env. Res., 72(6), 651–664. Tanaka, N., T. Hvitved-Jacobsen, T. Ochi, and N. Sato (2000b), Aerobic-anaerobic microbial wastewater transformations and reaeration in an air-injected pressure sewer, Water Env. Res., 72(6), 665–674. Tchobanoglous, G. (ed.) (1981), Occurrence, effect and control of the biological transformations in sewers. Chapter 7 in: Metcalf and Eddy, Inc., Wastewater Engineering: Collection and Pumping of Wastewater, McGraw-Hill, New York, pp. 232–268. Thistlethwayte, D.K.B. (ed.) (1972), The Control of Sulfides in Sewerage Systems, Butterworth, Sidney, Australia, p. 173. USEPA (1974), Process design manual for sulfide control in sanitary sewerage systems, USEPA 625/1-74-005, Technology Transfer, Washington, DC. USEPA (1985), Odor and corrosion control in sanitary sewerage systems and treatment plants, USEPA 625/1-85/018, Washington, DC.

References Vincke, E., J. Monteny, A. Beeldens, N.D. Belie, L. Taerwe, D. van Gemert, and W.H. Verstraete (2000), Recent developments in research on biogenic sulfuric acid attack of concrete. In: P.N. L. Lens and L. H. Pol (eds.), Environmental Technologies to treat Sulfur Pollution — Principles and Engineering, IWA Publishing, pp. 515–541. Vollertsen, J. and T. Hvitved-Jacobsen (1998), Aerobic microbial transformations of resuspended sediments in combined sewers — a conceptual model, Water Sci. Tech., 37(1), 69-76. Vollertsen, J. and T. Hvitved-Jacobsen (1999), Stoichiometric and kinetic model parameters for microbial transformations of suspended solids in combined sewer systems, Water Res., 33(14), 3127–3141. Vollertsen, J., M. do C. Almeida, and T. Hvitved-Jacobsen (1999), Effects of temperature and dissolved oxygen on hydrolysis of sewer solids, Water Res., 33(14), 3119–3126. Vollertsen, J., T. Hvitved-Jacobsen, I. McGregor, and R. Ashley (1998), Aerobic microbial transformations of pipe and silt trap sediments from combined sewers, Water Sci. Tech., 38(10), 249–256 (read text pp. 257–264). Errata: Water Sci. Tech., 39(2), 234–241. Wilmot, P.O., K. Cadce, J.J. Katinic, and B. V. Kavanagh (1989), Kinetics of sulfide oxidation by dissolved oxygen, J. Water Poll. Control Fed., 60, 1264–1270.

Methods for Sewer Process Studies and Model Calibration

This chapter focuses on two main subjects. It will first deal with knowledge and methodologies of good practice in the study of chemical and microbial processes in wastewater collection systems. The information on such processes is provided by investigations, measurements and analyses performed at bench, pilot and field scale. Second, it is the objective to establish the theoretical basis for determination of parameters to be used for calibration and validation of sewer process models. These main objectives of the chapter are integrated: sampling, pilot-scale and field measurements and laboratory studies and analyses are needed to determine wastewater characteristics, including those kinetic and stoichiometric parameters that are used in models for simulation of the site-specific sewer processes. The main procedures for sewer process studies will be dealt with, however, primarily those that are directly related to the determination of process-relevant characteristics. Procedures and measurements of, e.g., sewer hydraulic and solids transport characteristics will not be included in the text. Although information from such measurements is relevant for sewer process model simulation and evaluation, literature is generally available for that purpose. The following are publications dealing with the hydraulic measurements in sewers: ASCE (1983) and Bertrand-Krajewski et al. (2000). An overview of the physical processes in sewers is found in Ashley and Verbanck (1998). This chapter will not include all relevant methods. What are considered key representative examples will be focused on.

A sound theoretical basis, reliable and relevant methodologies for investigations and a conceptual description of the integrated phenomena and processes

constitute a fundamental triangle for process studies with the objective of establishing a process model. Such a framework forms the basis for engineering tools that make it possible to design and manage environmental systems with a process dimension. A specific objective for the methodologies applied for sewer process investigations is often to procure model parameters. These methodologies have their origin in bench-scale, pilot-plant and field studies. General aspects including examples of such types of methodologies will be addressed. Experiments must be carefully considered according to the objectives set. It is a general and basic requirement for an experimental study that the mass balance for the components that are focused on is observed.

Analytical methods may, to some extent, be available for the determination of specific components. Such methods are generally found in Standard Methods for the Evaluation of Water and Wastewater (1998). Process studies are often performed as either batch or flow-reactor experiments. A general and important aspect of laboratory studies is that they can be performed under controlled process conditions. Reactor experiments must be elaborated to fulfill the specific objective of the study. In addition to the experimental setup, a carefully considered program for sampling, handling and analysis is required. A laboratory reactor experiment needs a lot of planning — and also typically experience — to be successful. The example of a biofilm reactor setup shown in Figure 7.1 demonstrates how an experiment can be performed under controlled conditions (Raunkjaer et al., 1997). The objective of the study is to determine substrate (acetate) and DO surface removal rates of biofilms that were grown on wastewater. Careful control is needed to do so during conditions where both the substrate and the DO should be studied as limiting factors for the removal rates. A great number of specific details that will not be dealt with here were considered for this experiment.

Pilot and laboratory studies are generally performed to control the conditions of the different factors that have an impact on the processes. The advantage of a pilot study compared with a laboratory experiment is the scale that is closer to the real system. Such “scale factors” may, e.g., be a more realistic volume-to-area ratio and a more correct flow regime. Drawbacks include the high demands for resources and high manpower requirement for manufacturing and

Methods for Field-, Pilot- and Bench-Scale Studies

operation. Furthermore, pilot studies may often be difficult to run and may require specific practical skills. Pilot sewer studies are often carried out in systems operating with recirculation. Specific care must be taken in systems where water-gas exchange processes form a part of the mass balance. Critical points are pumps and bends that may change the flow regime, air–water exchange processes, biofilm and particle structure. Figure 7.2 is a sketch of a pilot sewer used for sewer process studies (Tanaka and Hvitved-Jacobsen, 2000).

It is in the real sewer systems that the benefits of a proper engineering of the processes must be proved. However, the sewer is not an optimal system for detailed studies of sewer processes because of a poor ability for establishment of controlled conditions. Field investigations are needed to determine reasonable

values for those parameters that are normally not feasible to measure explicitly by laboratory or pilot-plant studies. Field experiments are generally performed by sampling and measurement in upstream and downstream stations of a sewer network. A volume of water in the sewer can be monitored by following the course of a tracer that is added in an upstream station. Substances like rhodamine, radiotracers and salts may typically be selected for that purpose. Sampling after the passage of such tracers is a convenient way to ensure that “corresponding samples” are taken and to avoid too much noise because of the variability in wastewater quality. It is generally preferred to carry out field experiments in long sewer lines that make it possible to measure relatively large differences in the wastewater quality compared with the level and the uncertainty for determination of the parameters in question. Furthermore, a sewer line without tributary sewers, infiltration and exfiltration must be preferred because of a less complicated sampling program with which to identify all inputs and outputs to the system. Although field investigations are difficult to manage and typically not “ideal” from a scientific point of view, they are needed. First, because the field investigations show the real performance and second, because results from such process studies are generally required for model calibration and validation.

In all types of experimental work, sampling, handling and analysis must be considered carefully. This means that these three procedures must observe the overall objective of the study.

Methods for Field-, Pilot- and Bench-Scale Studies

Samples that are representative for a system — or a part of a system — are typically required in sewer process studies. Locations for sampling are typically manholes and pumping stations. Samples from a sewer — whether they represent wastewater, biofilm or sediment — must be taken by observing systematic diurnal and seasonal changes and a variability in time and space. Sampling must generally be repeated to investigate the magnitude and the importance of these variations. Sampling is often performed automatically and with a corresponding measurement of the flow to allow for the determination of a mass transport. Handling of a sample in terms of investigations and studies is exemplified in Section 7.1.1. Handling by transport from the sampling site to a laboratory must ideally be done without changing the sample, i.e., without changes in the parameters that will be analyzed. The potential process rates and the time needed for transport must be considered. Process rates are typically lower under anaerobic conditions than under aerobic conditions. If possible, a wastewater sample is, therefore, often transported under anaerobic conditions.

A bulk water oxygen uptake rate (OUR) measurement is performed as a laboratory study. It will be separately dealt with because of its fundamental importance for sewer process studies. The OUR is an activity-related quantitative measure of the aerobic biomass influence on the relationship between the electron donor (organic substrate) and the electron acceptor (dissolved oxygen, DO). It is a measure of the “flow of electrons” through the entire process system under aerobic conditions (Figure 2.2). The OUR versus time relationship of wastewater samples from sewers becomes a backbone for analysis of the microbial system. This relationship is crucial for characterization of the suspended wastewater phase in terms of COD components and corresponding kinetic and stoichiometric parameters of in-sewer processes. A methodology for OUR measurements in wastewater systems was originally developed for the characterization of activated sludge in terms of COD components and process parameters (Ekama and Marais, 1978; Dold et al., 1980). It included OUR measurements of the activated sludge under substrate-limited and substrate-nonlimited growth conditions, typically performed by discontinuous addition of wastewater. Development of respirometry principles and techniques has taken place, e.g., motivated for the control of the activated sludge processes (Spanjers et al., 1998). The OUR measurement principle developed for nonseeded wastewater differs from what is typically used for activated sludge characterization. OUR versus time measurements performed typically during 1–2 days and in a batch

reactor will lead to a quantification of the processes and the COD fractions of the wastewater. An OUR experiment is carried out at a constant temperature, e.g., 20 C, and under DO nonlimiting-growth conditions for the biomass in the wastewater sample. Typically, the DO concentration may vary from 8–6 gO2 m 3 for the determination of one OUR value. Figure 7.3 is a schematic outline of the relation between such DO concentration versus time measurements and the calculation of the OUR value at time tn. A simplification of the aerobic heterotrophic in-sewer microbial processes is depicted in Table 5.3. By omitting the reaeration and the growth of the biofilm biomass in this description, the remaining processes proceed interactively in the water phase under the conditions established in the OUR experiments. Furthermore, the processes take place at a constant temperature and at DO nonlimiting growth conditions affecting the formulation of the relevant rate expressions (see Table 7.1). A characteristic example of an OUR experiment of a wastewater sample with readily biodegradable substrates and hydrolyzable substrates is shown in Figure 7.4. The figure shows that the original amount of readily biodegradable substrates within the first 4 hours of the experiment is depleted, meanwhile increasing the biomass concentration and, consequently, the respiration rate (the OUR value) from about 9–17 gO2 m 3 h 1. After this period, readily biodegradable substrate is only available as a result of production from hydrolysis that is immediately consumed because of a growth-related respiration rate and a maintenance requirement rate of the biomass. At about 15 hours after the start of the experiment, the fast hydrolyzable fraction (n 1) is depleted. After this time, only hydrolysis of the slowly hydrolyzable fraction (n 2) remains a source for production of readily biodegradable substrate. This interpretation of

Methods for Field-, Pilot- and Bench-Scale Studies

Biomass growth in 1/YHw bulk water 1 Maintenance energy requirement Hydrolysis, fraction 1 1 (fast) Hydrolysis, fraction 1 2 (slow)

1 1

1

(1 YHw)/YHw

Equation a

1*

1

Equation b Equation c, n

1

Equation c, n

2

*If SS is not present in sufficient concentration, XBw is used for endogenous respiration. a. µHSS/(KSw + SS)XBw b. qmXBw c. khn(XSn/XBw)/(KXn + XSn/XBw)XBw

the OUR versus time variability for a wastewater sample is in terms of a quantitative description formulated in the model concept (Table 7.1, according to Table 5.3). The example in Figure 7.4 shows what is often observed, that a good agreement between measured and simulated OUR values can be obtained by the rather simple description of the microbial processes in Table 7.1. Generally, it can be concluded that an OUR experiment reflects the different phases of activity that the heterotrophic biomass is exposed to, depending on the availability and quality of the substrate (Section 3.2.6). The principle outlined is crucial as an experimental basis for the transfer of the concept for microbial transformations of wastewater into a tool that can be applied for design and management of sewer systems.

Different types of equipment, depending on the resources available and the number of measurements required, can be used for determination of an OUR versus time curve. A rather simple and manually operated type was used by Bjerre et al. (1995). A still relatively inexpensive apparatus, simple to operate automatically, was designed by Tanaka and Hvitved-Jacobsen (1998). However, this type may introduce a minor error at especially low OUR values because of a potential release of oxygen into a headspace of nitrogen gas in the reactor. An advanced, however, also relatively expensive, type of equipment is designed and used by Vollertsen and Hvitved-Jacobsen (1999) (see Figure 7.5). This type is automatically operated and flexible under varying environmental conditions. The equipment shown in Figure 7.5 is produced in stainless steel with a 2.2 L reactor volume. It operates at a DO concentration between about 6–8 gO2 m 3 and with the temperature of the wastewater kept constant (20 C) by circulating water through a cap around the reactor. When the DO concentration is below about 6 gO2 m 3, an aeration cycle starts by blowing compressed air into the reactor and by opening the piston to the expansion chamber. After the end of the aeration period, the piston is closed with a preset time delay to ensure that the air bubbles are removed from the reactor. Data on DO measurements are automatically sampled, and OUR values are calculated for each period corresponding to a reduction of the DO concentration from about 8–6 gO2 m 3.

Methods for Field-, Pilot- and Bench-Scale Studies

In addition to sampling in a sewer followed by analysis of specific components or use of a sample for further laboratory or pilot-scale experiments, a number of direct or indirect measurements must typically be performed in the sewer itself. Important measurements related to sewer process studies are DO, reaeration, biofilm characteristics and odor.

Sensor measurements of DO in sewers require special attention because of the risk for clogging the DO-probe with gross solids. Figure 7.6 shows an arrangement to avoid this problem (Gudjonsson et al., 2001). The DO-meter is placed in a small float of polyurethane that keeps the probe submerged and lets the gross solids pass without being retained on the sensor. A reliable measurement of DO in a wastewater system requires that the surface of the sensor be regularly cleaned to avoid the development of a biofilm that otherwise will consume oxygen and disturb the measurement.

Determination of reaeration relies on the measurement of the air–water oxygen transfer coefficient (Section 4.4.2). Measurement of this coefficient — the reaeration coefficient — in gravity sewer lines follows basically the methods that have been developed for and applied in rivers. Methods for determination

of the oxygen transfer coefficient can be divided into two groups: direct and indirect methods. Indirect methods are generally based on the principle of a mass balance for dissolved oxygen. Direct methods of the oxygen transfer coefficient make use of relatively inert substances that have a constant air–water mass transfer coefficient compared with that of oxygen. An overview of direct and indirect methods for determination of the reaeration is found in Jensen and Hvitved-Jacobsen (1991). An indirect method to determine reaeration in gravity sewers was developed by Parkhurst and Pomeroy (1972). They made the measurements in gravity sewers where the biofilm was removed mechanically followed by a shock load with caustic soda. During the measurement period, the biological activity was suppressed in the water phase by a chemical substance. Measurement of upstream and downstream DO concentrations in the sewer determined the reaeration by using a simple DO mass balance. Jensen and Hvitved-Jacobsen (1991) developed a direct method for the determination of the air–water oxygen transfer coefficient in gravity sewers. This method is based on the use of krypton-85 for the air–water mass transfer and tritium for dispersion followed by a dual counting technique with a liquid scintillation counter (Tsivoglou et al., 1965, 1968; Tsivoglou and Neal, 1976). A constant ratio between the air–water mass transfer coefficients for dissolved oxygen and krypton-85 makes it possible to determine reaeration by a direct method. Sulfur hexafluoride, SF6, is another example of an inert substance that has been used as a tracer for reaeration measurements in sewers (Huisman et al., 1999).

Measurement of biofilm activity can be performed based on laboratory reactor experiments or with a technique combining biofilm growth taking place in a sewer followed by measurements in laboratory scale (Raunkjaer et al., 1997; Bjerre et al., 1998). Huisman et al. (1999) developed a sewer in situ biofilm respiration chamber. It includes a DO sensor and a chamber that can be pressed onto the sewer wall. It is designed to achieve an even and unidirectional flow distribution over the entire measurement area. Pure oxygen is injected for oxygenation.

A great number of methods have traditionally been used for sampling and measurement of odors. The need for a standard procedure has, in Europe, led to the development of a (draft) standard (CEN, 1999; Sneath and Clarkson, 2000). Odors can be detected at very low concentrations. Sampling, handling and

Methods for Determination of Components and Parameters

analysis of odors must be carefully carried out to be reliable. Procedures for odor measurements specifically related to sewer networks do not exist. The following outlines the general fundamentals of sampling and measurement of odors. Samples of odorous air for olfactometric analysis are usually collected into a sample bag and transported to a laboratory for analysis. A sample bag may be placed in a container, and the bag is gradually filled with odorous air when air is removed from the gap between the container and the bag using a vacuum pump. A laboratory where the measurement takes place must be free from odor and is typically air-conditioned with air filtration. The odor sample is placed in an olfactometer that basically is a device for dilution of the sample. Typically, the meter has two outlet ports: diluted odorous air flows from one, and clean odor-free air flows from the other. In dynamic olfactometry, panel members assess the two ports of the olfactometer. The assessors indicate from which of the ports the diluted sample is flowing. The measurement starts with a dilution that is large enough to make the odor concentration beyond the panelists’ threshold. This concentration is normally increased by a factor of two in each successive presentation. Only when the correct port is chosen and the assessor is certain that the choice is correct and not just a guess, is the response considered a true value. In the CEN procedure, the European odor unit per cubic meter, ouE m 3 is used. The odor concentration at the detection threshold is defined as equal to 1 ouE m 3. A sensor array named the electronic nose is a rapid and relatively simple technique that can be used for monitoring wastewater odors (Stuetz et al., 2000). The electronic nose uses sensors of varying affinities to characterize an odor without reference to its chemical composition.

A combination of laboratory and field experiments is required for determination of components and parameters for a sewer process model for simulation of the microbial transformations of organic matter (cf. specifically Sections 5.2–5.4, 6.3 and 6.4). Furthermore, additional information is needed to include the sulfide formation. Explicit determination of model components and parameters are preferred to indirect and implicit methods. However, to some extent, model calibration is typically needed to establish an acceptable balance between process details of a model and possibilities for direct experimental determination of model parameters. Determination of OUR versus time curves is without question the most

important experimental procedure needed for determination of parameters for a sewer process model (cf. Section 7.1.3). In addition, relevant hydraulic information (flow characteristics), sewer system characteristics (e.g., length of sewer line, diameter and slope) and specific in situ determined quality parameters (specifically, the temperature and the DO concentration) are required. Determination of OUR curves for the wastewater in upstream and downstream points of a sewer is needed to assess the impact of the sewer biofilm. The detailed methodologies for determination of model components and parameters can be categorized in four groups: (1) Determination of central model parameters (cf. Section 7.2.1). OUR versus time measurements of incoming wastewater to a sewer system modified by the addition of readily biodegradable substrate, e.g., acetate. (2) Determination of the biodegradability of wastewater organic matter (cf. Section 7.2.2). OUR versus time measurements of incoming wastewater to a sewer system. (3) Determination of model parameters by iterative simulation (cf. Section 7.2.3). An iterative simulation of the OUR curve for the incoming wastewater. (4) Calibration and validation of the sewer process model (cf. Section 7.2.4). OUR measurements of corresponding upstream and downstream wastewater samples followed by a simulation (calibration) with the sewer process model. These four procedures are all recommended to be performed in the order shown to achieve optimal parameter estimation followed by a final validation of the gravity sewer process model (Figure 7.7). In the case of design of a new sewer system, procedure number 4 is, of course, not relevant and kinetic parameters for the sewer biofilm must be evaluated and selected based on information from comparative systems. It is important to note that wastewater is subject to great variability in terms of its components and processes. Procedures 1 to 4, therefore, correspond to a typical analytical method for the determination of the characteristic components and the stoichiometric and kinetic parameters. Cases where the procedure described in Sections 7.2.1–7.2.4 is either difficult or not feasible to follow may exist. A detailed knowledge on wastewater characteristics and experience from laboratory and modeling studies may be crucial in such situations for finding alternative variants of the procedures 1 to 4. The procedures described in Sections 7.2.1 to 7.2.4 refer to the aerobic formulated sewer process model (cf. Table 5.3) whereas Section 7.2.5 deals with methods applied for determination of model parameters to include transformations of wastewater components under anaerobic conditions (cf. Table 6.6).

Methods for Determination of Components and Parameters

An OUR experiment is performed on a wastewater sample from the influent to the sewer system in question. The sample is modified by addition of readily biodegradable substrate, typically acetate or glucose. The following four process parameters are explicitly determined based on this experiment: • µH, maximum specific growth rate • KS, saturation constant for readily biodegradable substrate • YHw, yield constant • qm, maintenance energy requirement rate constant The experiment, or preferably a number of parallel experiments, is carried out until the originally available readily biodegradable substrate and the fast hydrolyzable substrate are depleted. For typical domestic wastewater, this is

the case after 1–2 days. At this point in the experiment, it is assumed that the lack of substrate suppresses the growth process. Readily biodegradable substrate (typically acetate or glucose) is then added to the wastewater, and the experiment continues until this substrate is also depleted and the OUR value slightly declines. The impact of the substrate added to the microbial system, where no growth of the biomass takes place, can thereby be evaluated. The idea behind the experiment is to let the biomass growth rate change from zero to its maximum value. By adding a known amount of substrate under controlled conditions, the interpretation of the OUR response can be described by the model concept (cf. Table 7.1) and the four central process parameters can be determined. The following two conditions are important for a successful outcome of the experiment: • Before adding the readily biodegradable substrate, the maintenance energy requirement of the active biomass should ideally correspond to the amount of readily biodegradable substrate produced by hydrolysis of the slowly biodegradable COD fraction, i.e., an equilibrium corresponding to an almost constant OUR must be seen. • The active biomass that is available must react directly on (i.e., show an immediate exponential growth response on) the readily biodegradable substrate that is added. These two conditions are crucial for a correct outcome of the experiment. If the experiment is not successful by using acetate or glucose, a more broad-spectral, however, well-defined type of substrate can be selected, e.g., a yeast extract. If the experiment is still not successful, a calibration procedure, i.e., an extended procedure number 4, is still an alternative. Figure 7.8 shows an example with six parallel OUR experiments. For illustration, the acetate is added in a varying amount when the readily biodegradable and the fast hydrolyzable substrate have been depleted and a rather constant, however, slowly decreasing OUR value is reached. This stage of the system corresponds to a situation where the biomass maintenance energy requirement is supported by the hydrolysis of the slowly hydrolyzable substrate. After the acetate is added, an immediate increase in the OUR value is observed, corresponding to the start of nonlimited growth of the active biomass. Further exponentially increased growth of the biomass takes place until the added amount of acetate is almost depleted and the growth drops to a level determined by the available slowly hydrolyzable substrate. It is essential that the determination of the four parameters by the experiment performed is based on expressions that can be derived from the model concept (Table 7.1). Furthermore, it must be noticed that a successful agreement between a measured and a simulated OUR curve, including the effect of the

Methods for Determination of Components and Parameters

substrate added, is an essential criterion for the validity of the concept for microbial wastewater transformations. Experiments show that this agreement is normally observed when applying the methodology described. This fact is a strong indication of the validity of the concept and is also crucial for the determination of the active biomass as the central component for the concept. This central position of the biomass, XBw, is reflected by the fact that it is proportional to the OUR value [cf. Equation (7.5)]. The success of the concept and the corresponding procedure for estimation of the parameters are crucial for the use of the concept in terms of a sewer process model that can be used for design and management purposes. Details concerning derivation of expressions for the four parameters mentioned, based on the model shown in Table 7.1, are found in Vollertsen and Hvitved-Jacobsen (1999). The final expressions are as follows: SS ,add

YHw

SO,growth SS ,add

ln H

OUR(t ) OUR(t0 ) t t0

(7.1)

(7.2)

KSw is determined from the slope of the decline for the OUR curve when the added SS is being depleted. H

qm

1 YHw SO,maint YHw SO,growth

(7.3)

Except for SS,add, SO,growth and SO,maint that are interpreted in the following, the symbols used in the Equations (7.1) to (7.3) are explained in Appendix A. The interpretation of the oxygen uptake, SO, from an experiment with addition of acetate in a given amount, SS,add, is shown in Figure 7.9 as an example originating from one of the experiments shown in Figure 7.8. As seen from Figure 7.9, SO is divided into two parts, a growth-related oxygen uptake ( SO,growth) and an oxygen uptake ( SO,maint) that corresponds to the maintenance energy requirement of the biomass. Example 7.1: Derivation of the expression [Equation (7.2)] for the maximum specific growth rate, ␮H, for the heterotrophic biomass As an example of the derivation of the Equations (7.1) to (7.3) based on the

Methods for Determination of Components and Parameters

conceptual model shown in Table 7.1, Equation (7.2) for determination of µH will be derived. Referring to Table 7.1, the OUR value at time t is as follows: OUR(t )

dSO,growth

dSO,maint

dt

dt

1 YHw YHw

SS H

K Sw

SS

qm X Bw

Because the growth process proceeds under substrate nonlimiting conditions, the following is approximately the case: SS K Sw SS

1

Therefore: OUR(t )

1 YHw YHw

H

qm X Bw (t )

OUR(t0 )

1 YHw YHw

H

qm X Bw (t0 )

From Table 7.1 it also follows that dXBw/dt µH XBw. In an integrated form, this equation is equivalent to the following:

X Bw (t ) H (t

X Bw (t0 )e t0 )

ln

H ( t t0 )

X Bw (t ) X Bw (t0 )

By substituting the expressions derived for OUR into this equation, it follows that: H (t

t0 )

ln

OUR(t ) OUR(t0 )

This equation is equal to Equation (7.2). The process parameters that are calculated based on the Equations (7.1) to (7.3) and the procedure for KSw are shown in Table 7.2 for each of the six experiments depicted in Figure 7.8. Apart from some inconsistency for the determination of qm when adding a small amount of readily biodegradable substrate compared with the values when adding higher amounts, Table 7.2 shows a reasonable agreement between the parameters calculated. It is considered important that the concept, Table 7.1 (used for the derivation of the expressions for determination of the parameters), corresponds with the experimental evidence. This is also observed from the simulation results depicted in Figure 7.8. For the verification of the concept, it is crucial that the biomass as a central component is correctly interpreted. This fact is demonstrated in Figure 7.8 by a rather good agreement between the measured and the simulated OUR value immediately before and after the addition of acetate. This relies on the fact that the OUR under nonlimiting-growth conditions is proportional with the biomass [cf. Equation (7.5)]. Figure 7.8 also shows a weakness of the concept by a less correct simulation of the biomass activity when the added substrate has been depleted.

YHw ( ) 1 H (d ) KSw (gO2 m 3) qm (d 1)

0.65 5.4 0.7 2.13

0.66 6.5 0.9 1.84

0.71 4.7 0.9 1.02

0.67 5.7 1.0 1.15

0.67 5.3 1.1 0.87

0.65 5.3 0.8 1.02

Methods for Determination of Components and Parameters

The OUR experimental procedure is used for determination of COD fractions for wastewater in terms of biomass and different fractions of substrate. As shown in Figure 7.4, an OUR experiment of a wastewater sample reflects the different phases of activity that the heterotrophic biomass is exposed to depending on the availability of the substrate. If an OUR experiment is carried out under both substrate-nonlimited and substrate-limited conditions, the different COD components can be determined. Such conditions exist during an OUR experiment where readily biodegradable substrate is available at the beginning of the experiment. The following COD fractions can be determined: (1) OUR experiment under substrate-nonlimited conditions: • SS, readily biodegradable substrate • XBw, heterotrophic biomass (2) OUR experiment under substrate-limited conditions: • XSn, hydrolyzable substrates Furthermore, the maximum specific growth rate, µH, can be determined where substrate-nonlimited conditions exist, and the saturation constant, KSw, can be found where such conditions are just being terminated. Determination of these two process parameter follows the principle that is described in Section 7.2.1. The determination of the COD components depends on the fact that the substrate uptake can be experimentally related to the OUR curve. The heterotrophic yield constant, YHw, that is experimentally determined from procedure number 1, Section 7.2.1, relates the oxygen uptake to the readily biodegradable substrate that is consumed irrespective of its origin, being either directly available or continuously produced from hydrolyzable COD fractions. Details concerning derivation of the expressions for the COD fractions are found in Vollertsen and Hvitved-Jacobsen (2001). The following expressions [Equations (7.4) to (7.6)] for determination of the components in the wastewater at t0 0 are derived in the case of two hydrolyzable fractions (cf. Figure 7.10 for interpretation and determination of the two fractions of oxygen uptake, SO1 and SO2): SS

X Bw

SO1 1 YHw

OUR 1 YHw H YHw

(7.4)

(7.5) qm

XS ,fast

SO2 1 YHw

(7.6)

These COD fractions can be determined by an OUR experiment typically performed during 0.5–2 days. The slowly hydrolyzable fraction of the COD, XS,slow, cannot be determined from the oxygen uptake because the degradation of this fraction takes considerable time and also interferes with the degradation of already produced biomass. A determination based on a COD mass balance is, however, possible: XS ,slow

CODtot

( X Bw

SS

XS ,fast )

(7.7)

Basically, the procedure for determination of XBw(t0) requires availability of readily biodegradable substrate at t 0, ideally not less than 10–15 gCOD m 3,

Methods for Determination of Components and Parameters

to ensure nonlimited growth. If this is not the case and the OUR curve is approximately constant, XBw(t0) can be estimated by Equation (7.5), assuming that µH 0. The biomass, XBw, may also be determined by adding readily biodegradable substrate, e.g., acetate, to the wastewater. The COD fractions can also be determined by iterative simulation methodologies based on a model corresponding to the matrix formulation in Table 7.1 and with parameters determined from procedure number 1 (Section 7.2.1). Successful use of this methodology requires, however, not only theoretical insight into sewer processes but also experience in calibration techniques.

Procedures number 1 and 2 described in Sections 7.2.1 and 7.2.2, respectively, show that the COD components and central parameters for the sewer process model can be explicitly determined by means of rather simple OUR experiments. However, some of the process parameters for the hydrolysis still need to be determined. These are as follows: • khn, hydrolysis rate constants • KXn, saturation constant for hydrolysis Well-known iterative methods for model calibration of the OUR curve can be used to determine these two constants when results from procedures number 1 and 2 are available. The model used for the determination of these parameters is based on the matrix formulated in Table 7.1. The values shown for khn and KXn in Table 6.7 may be used as a starting point for this iteration.

Procedures 1 to 3 described in the previous three subsections have typically been performed on wastewater samples at an upstream point of a sewer. The objective of these procedures has been to characterize the incoming wastewater to the sewer system in terms of COD fractions and process-relevant parameters. Contrary to this, the present procedure number 4 is performed with the overall objective of determining sewer process-related characteristics including the biofilm and reaeration. The characteristics of the water phase considered in procedures 1 to 3 are hereby extended to include all major processes relevant for the microbial transformations in gravity sewers, especially when dealing with aerobic processes. Further detailed characterization that is needed when including the anaerobic transformations will be dealt with in Section 7.2.5. The COD components of a downstream wastewater sample from a sewer can be compared with the corresponding values of an upstream sample that has

previously been taken. Considering approximate plug flow in the sewer, the difference between corresponding values of the COD fractions of these two wastewater samples reflects the result of the microbial processes during transport. These processes proceed in the wastewater phase as well as in the biofilm — and in the sewer deposits — all influenced by the actual reaeration. The OUR versus time curves for the two wastewater samples constitute the analytical basis to determine the components and the relevant process parameters of the wastewater (cf. procedures number 1 to 3). The mathematical description of the processes in the water phase, Table 7.1, is particularly integrated in procedure 3. The description of the reaeration requires that actual sewer and flow characteristics be available (cf. Section 4.4). Simulation procedure 4 is basically a calibration of the sewer process model for aerobic microbial transformations as described in the matrix formulation (Table 5.3). Both the biofilm processes and the reaeration are included. Initial values for the components and process parameters for this simulation originate from the sample taken at the upstream sewer station. When simulated values of the downstream COD components are acceptable, i.e., approaching the corresponding measured values, the calibration procedure is successfully completed. The major model parameters to be included in the calibration process are those relevant for the biofilm, especially k1/2 and KSf . After calibration, the model is ready for a successive validation process and later use in practice. A rather simple 1/2-order flux model for the biofilm processes was selected, although more detailed formulated models are well known, e.g., Gujer and Wanner (1990). The reason is that a simple and sound parameter estimation and calibration procedure that can be easily performed has been emphasized. For aerobic gravity sewers, procedure 4 is the ultimate calibration of the sewer process model. This is based on procedures 1 to 3 using information from upstream and downstream wastewater samples and by including local sewer systems and flow characteristics, temperature and DO concentration values of the wastewater in the sewer. Example 7.2 outlines the results of calibration and validation performed on a 5 km intercepting sewer line. Example 7.2: Calibration and validation of the sewer process model Procedures 1 to 4 described in Sections 7.2.1 through 7.2.4 are applied in this example for determination of wastewater COD fractions, model parameters and a corresponding calibration/validation of the sewer process model under aerobic and dry-weather conditions. The number of repeated tests — a total of 29 during different seasons — demonstrates not just the validity of the sewer process model depicted in Table 5.3 but also the validity of the concept behind the model formulated in Section 5.2.

Methods for Determination of Components and Parameters

The investigations were performed on a 5.2 km gravity sewer line in Denmark. A series of 29 corresponding upstream and downstream samples were taken according to procedure 4 and exposed to OUR measurements and analysis following procedures 1 to 3. Six of the samples (three originating from a summer period and three from a winter period) were used for determination of parameters that were considered universal for all 29 samples. The rest of the samples (11 from a summer period and 12 from a winter period) were used for validation. The field site used for calibration and validation of the sewer process model was an intercepting gravity sewer located between the city of Dronninglund and the wastewater treatment plant in Asaa in the northern part of Jutland, Denmark (Figure 7.11). The total load of wastewater to the sewer system in terms of Person Equivalents is 4350 PE with 3525 PE originating from Dronninglund. Almost no discharges from industries take place. Dronninglund is primarily served by a sanitary sewer network, and the three small villages are predominantly combined sewered. The wastewater from Dronninglund is transported 1.2 km in a pressure main followed by a 5.2 km intercepting gravity sewer made of concrete to the treatment plant in Asaa. No wastewater is discharged into the sewer at this stretch. The average wastewater flow during the 29 events varied between 0.013 and

0.016 m3 s 1, with a residence time between 2.96 and 3.11 hours. The average summer and winter temperature of the wastewater was 15.2 C and 8.2 C in station 1 and 12.7 C and 7.5 C in station 4, respectively (all are average values with a standard deviation between 0.5 and 1.0). The DO concentration varied typically between 1.0 and 3.0 gO2 m 3 in all stations. The average value and standard deviation for CODtot were 670 and 145 gCOD m 3 during summer and 450 and 70 gCOD m 3 during winter, respectively. Parameter values considered universal for all 29 events were determined for the six events that were selected for calibration. These universal values are shown in Table 7.3. The remaining parameters — dependent on the actual event — were determined separately. The result of the model simulation is validated by the ability of the model to predict quality changes of the wastewater organic matter during transport in the sewer from stations 1 to 4. These quality changes are defined in terms of the COD fractions XBw, SS and XS,fast. Figure 7.12 shows measured and simulated values for these three main COD fractions as absolute values at station 4 and as changes during transport. The agreement between the absolute values is not considered important for the validation but is shown for general information. However, it is crucial that the model be able to predict an average value of the changes of the COD fractions during dry weather. The results from this study support what has been dealt with throughout the text and especially focused on in Chapter 5: the concept’s ability to predict wastewater quality changes in a sewer. The example shows that the corresponding process model can be used to simulate the average dry-weather performance of the sewer and serve as a tool for process design and management. The sewer, with its relation to the subsequent treatment plant, is generally

KO KSw KSf K1/2 qm XBf YHf YHw r f w

gO2 m 3 gCOD m 3 gCOD m 3 gO20.5 m 0.5 d 1 d1 gCOD m 2 gCOD biomass (gCOD substrate) gCOD biomass (gCOD substrate) — — — —

1 1

0.5 1.0 5 6 1.0 10.0 0.55 0.55 1.024 1.05 1.07 0.15

Methods for Determination of Components and Parameters

considered, during dry-weather conditions, a “stationary” designed and operated system with regard to the microbial processes that proceed. Both sanitary sewers and combined sewer networks — taking into account a wet-weather functioning of the combined system — must be designed to operate under “average” dry-weather conditions without being subject to a detailed “short-time” control as far as wastewater quality changes are concerned. Because this is the case, the sewer process model can be used as a tool for prediction of the microbial transformations (i.e., quality changes of the wastewater) in the sewer. This information is important when dealing with integrated process management of the sewer and the wastewater treatment plant system.

Anaerobic processes in wastewater of sewer systems in terms of both the organic matter transformations and the sulfur cycle have been dealt with in Chapter 6. Particularly, Section 6.4 has focused on the integrated aerobicanaerobic sewer process model. From a conceptual point of view, the anaerobic

transformations have been less detailed compared with the aerobic processes in the water phase. Corresponding less detailed methods for description of the processes exist — or will be needed. As far as organic matter transformations are concerned, the process rates are significantly slower compared with aerobic transformations. Basically, readily biodegradable organic matter is preserved and even, to some extent, produced opposite to the situation when aerobic processes proceed. The sulfur cycle, until now included in the sewer process model, is relatively simply described following empirical expressions for sulfide formation. Other important processes in this respect, e.g., hydrogen sulfide emission and sulfide oxidation, still need to be included, however, and, most of all, investigated from a conceptual point of view. The methods for determination of the parameters concerning the anaerobic processes in the aerobic-anaerobic process model are less structured compared with the procedures that have been described in Sections 7.2.1 through 7.2.4. The following points constitute a rather simple approach in this respect: • determination of the VFAs to complete the description of the fermentation process in terms of fermentable, readily biodegradable substrate, SF, and fermentation products, SA, considered identical to the VFAs • determination of sulfide and a sulfide formation rate • determination of the formation rate for readily biodegradable substrate (anaerobic hydrolysis) • parameter estimation by calibration The first three methods can be considered more or less explicit methods for the determination of process components and parameters. The four groups of methods for investigation of anaerobic transformations will be outlined in the following.

The VFAs, primarily formate, acetate, propionate, n-butyrate and isobutyrate, can be determined analytically on an ion chromatograph (Standard Methods for the Examination of Water and Wastewater, 1998). Determination of fermentable, readily biodegradable substrate, SF, and fermentation products, SA, in units of COD requires that the VFA components be converted to this unit. The following example using formate demonstrates this: HCOO

1 O2 2

CO2

OH

(7.8)

The stoichiometry of Equation (7.8) shows that the COD-to-formate ratio is as follows:

Methods for Determination of Components and Parameters

16 45

0.36 gCOD (g formate)

1

Table 7.4 outlines corresponding stoichiometric relations for other VFA components. The relationship between readily biodegradable substrate, SS, fermentable, readily biodegradable substrate, SF, and fermentation products, SA, is defined as: SS

SF

(7.9)

SA

Total sulfide can be measured photometrically by the methylene blue method (Cline, 1969; Standard Methods for the Examination of Water and Wastewater, 1998). Reliable sensors exist for the measurement of hydrogen sulfide in air. Because sulfide is a very reactive component that is easily oxidized or reacts with heavy metals to produce precipitates, and because hydrogen sulfide is also easily emitted from the water phase, precautions must be taken under sampling, and the interpretation of analytical results must be careful. Laboratory experiments for the determination of sulfide production rates from anaerobic biofilms grown on wastewater can be performed in simple and more complicated operating biofilm reactors without or with a rotating drum (Nielsen, 1987; Norsker et al., 1995). Mixed laboratory and field experiments to determine the sulfide production rate with biofilm growth under field conditions followed by reactor experiments in the laboratory is described in Bjerre et al. (1998). Sulfide production rates can also be determined in real systems, taking corresponding upstream and downstream samples for analysis and knowing flow and system characteristics (Nielsen et al., 1998). Sulfide production rates have been measured by circulating wastewater in a pilot plant, full-flowing sewer system (Tanaka and Hvitved-Jacobsen, 2000). Field and pilot-plant studies for the determination of sulfide production rates may be

Formate Acetate Propionate Lactate Butyrate

0.36 1.08 1.53 1.08 1.84

feasible in pressure mains, however, under gravity sewer conditions, such methods are less useful because of partial emission and oxidation of sulfide.

Preservation and formation of readily biodegradable organic substrate, SS, is a major characteristic that is observed under anaerobic conditions in wastewater. An experimental procedure that can be applied for this purpose is crucial for the prediction of the anaerobic transformation of the organic matter and particularly as a basis for estimation of the anaerobic hydrolysis rate (cf. Figure 6.10). An experimental procedure carried out under changing aerobic and anaerobic conditions in an experimental setup according to an OUR experiment has been developed (Tanaka and Hvitved-Jacobsen, 1998). The principle for the determination of the formation rate of SS under anaerobic conditions is based on a comparison between two OUR experiments that are performed on the same wastewater sample. One is a normal OUR experiment (cf. Section 7.1.3). The other is carried out with one or two anaerobic periods of the duration of a few hours during the OUR experiment. The result of such experiments that are performed in parallel is shown in Figure 7.13. As can be interpreted by comparing the SS consumed in the two experiments (cf. Section 7.2.2) the difference in

Methods for Determination of Components and Parameters

oxygen uptake (OU) is equal to the generation of SS during the anaerobic periods, taking into account the aerobic yield constant, YHw. The production of SS depends on the duration of the anaerobic period, whereby an average production rate can be estimated. The SS produced is basically a net production; however, the consumption of substrate for the production of biomass is considered relatively low under anaerobic conditions in this experiment. It has been confirmed that under the conditions performed, no biofilm is generated and no sulfide production takes place (cf. Section 6.3 and Figure 6.9). By comparing the level of the OUR values before and after the anaerobic period, it is readily seen from Figure 7.13 that the activity of the heterotrophic biomass, XBw, is maintained during anaerobic conditions. This observation has been confirmed for at least up to about 24 hours (Tanaka and HvitvedJacobsen, 1999). The (aerobic) biodegradability of the organic matter in wastewater affects the anaerobic formation rate of SS. The results shown in Figure 7.14 correspond to the findings that a soluble COD value, CODS, below 50 gCOD m 3 has no real potential for sulfide formation (cf. Section 6.2.4), as this value typically corresponds to non- and slowly biodegradable, soluble organic matter. In the case where the anaerobic processes take place under conditions where consumption of SS by the sulfate-reducing biomass and the fermenting biomass must be considered, Equation (7.10) expresses the total anaerobic hydrolysis rate. This equation is based on the assumption that methane formation in sewers without sediment normally can be neglected (Section 3.2.2). rS ,tot

rS ,net

2rS ,s

rS ,ferm

(7.10)

where rS,tot rS,net rS,s rS,ferm

total anaerobic hydrolysis rate (gCOD m 3 h 1) net measured anaerobic formation rate of SS (gCOD m 3 h 1) production rate of sulfide (gS m 3 h 1) substrate-consuming rate by the fermenting biomass (gCOD m h 1)

3

Equation (7.10) is based on the assumption that Equation (6.4) is an acceptable description of the relation between organic matter consumption and sulfide production, i.e., 2 moles of CH2O with a COD value of 32 gCOD mole 1 is consumed by the production of 1 mole of H2S-S with a molar weight of 32 gS mole 1. Tanaka and Hvitved-Jacobsen (1999) have experimentally confirmed that a relatively high linear correlation coefficient between the production of SS and

CO2 exists under anaerobic conditions in wastewater. They also found that typically 50% of the CO2 was produced by the sulfate-reducing bacteria, the other half by the fermenting biomass. However, the net production rate of SS was typically about 70% of the total produced SS by anaerobic hydrolysis [Equation (7.10)]. Hence, this equation may, even in a reduced form, be valuable for the estimation of the production of readily biodegradable substrate under anaerobic conditions.

The three described groups of methodologies are experimental ways leading to the estimation of model parameters for the description of the anaerobic processes according to the aerobic-anaerobic conceptual model (Table 6.6). The determination of the remaining kinetic and stoichiometric parameters in this model, however, requires a calibration procedure, where the results of the above three described methodologies are used. Table 6.7 shows typical values of such parameters determined by the three methodologies followed by a model calibration. Focusing on the carbon flow in wastewater under anaerobic conditions, the corresponding process parameters shown in Figure 6.9 are based on utilization of results from the three mentioned procedures and a calibration of the aerobic-anaerobic sewer process model shown in Table 6.6.

This section focused on field, pilot-plant and bench-scale methodologies that can be applied to determine parameters and components for sewer process

References

models. Without sound methods for sewer process studies, any simulation model and process design and operationally related procedure become irrelevant as far as practical implementation is concerned. Sewer process modeling is a new discipline, and rather little information is available for the practitioner in terms of experience. The methodologies and investigations referred to in this section have formed an important basis for not only the sewer process concept but also for the determination of the corresponding model parameters. However, the variability of the wastewater characteristics is considerable, and the types of systems in which the microbial processes proceed are legion. Care must be taken when applying already available information on the sewer processes. In this respect, the methodologies for sewer process studies become crucial in aquiring new knowledge and information. Investigations in sewers take place in a nonattractive environment that is also often difficult to get into. This fact may have had substantial influence on the rather limited knowledge on the sewer performance, not just the sewer processes. From a microbiological point of view, however, the sewer is a highly interesting environment with high variability and diversity. Hopefully, this fact, highly supported by the evidence of the practical applications, will lead to improved methods for sewer process studies in the future.

ASCE (1983), Existing sewer evaluation and rehabilitation, ASCE (American Society of Civil Engineers) Manual and Report on Engineering Practice 62; WPCF (Water Pollution Control Federation) Manual of Practice FD-6, p. 106. Ashley, R.M. and M.A. Verbanck (1998), Physical processes in sewers, Congress on Water Management in Conurbations, Bottrop, Germany, June 19–20, 1997. In Emschergenossenschaft: Materialien zum Umbau des Emscher-Systems, Heft, 9, 26–47. Bertrand-Krajewski, J.-L., D. Laplace, C. Joannis, and G. Chebbo (2000), Mesures en hydrologie urbaine et réseau d’assainissement, Tec et Doc, Paris, p. 808. Bjerre, H.L., T. Hvitved-Jacobsen, B. Teichgräber, and D. te Heesen (1995), Experimental procedures characterizing transformations of wastewater organic matter in the Emscher river, Germany, Water Sci. Tech., 31(7), 201–212. Bjerre, H.L., T. Hvitved-Jacobsen, S. Schlegel, and B. Teichgräber (1998), Biological activity of biofilm and sediment in the Emscher river, Germany, Water Sci. Tech., 37(1), 9–16. CEN (1999), Air quality — determination of odour concentration measurement by dynamic olfactometry, European Committee for Standardisation, draft prEN 13725. Cline, D.J. (1969), Spectrophotometric determinations of hydrogen sulfide in natural waters, Limnology and Oceanography, 14, 454–458. Dold, P.L., G.A. Ekama, and G. v. R. Marais (1980), A general model for the activated sludge process, Prog. Water Tech., 12, 47–77. Ekama, G.A. and G. v. R. Marais (1978), The dynamic behaviour of the activated sludge process, Research report No. W 27, Department of Civil Engineering, University of Cape Town.

Gudjonsson, G., J. Vollertsen, and T. Hvitved-Jacobsen (2001), Dissolved oxygen in gravity sewers — measurement and simulation, Proceedings of the 2nd International Conference on Interactions between Sewers, Treatment Plants and Receiving Waters in Urban Areas (INTERURBA II), Lisbon, Portugal, February 19–22, 2001, pp. 35–43. Gujer, W. and O. Wanner (1990), Modeling mixed population biofilms. In: W. G. Characklis and K. C. Marshall (eds.), Biofilms, John Wiley & Sons, Inc., New York, pp. 397–443. Huisman, J.L., C. Gienal, M. Kühni, P. Krebs, and W. Gujer (1999), Oxygen mass transfer and biofilm respiration rate measurement in a long sewer, evaluated with a redundant oxygen balance. In I.B. Joliffe and J.E. Ball (eds.), Proceedings from the 8th International Urban Storm Drainage Conference, Sydney, Australia, August 30–September 3, 1999, vol. 1, pp. 306–314. Jensen, N.Aa. and T. Hvitved-Jacobsen (1991), Method for measurement of reaeration in gravity sewers using radiotracers, Research J. WPCF, 63(5), 758–767. Nielsen, P.H. (1987), Biofilm dynamics and kinetics during high-rate sulfate reduction under anaerobic conditions, Appl. Environ. Microbiol., 53(1), 27–32. Nielsen, P.H., K. Raunkjaer, and T. Hvitved-Jacobsen (1998), Sulfide production and wastewater quality in pressure mains, Water Sci. Tech., 37(1), 97–104. Norsker, N.H., P.H. Nielsen, and T. Hvitved-Jacobsen (1995), Influence of oxygen on biofilm growth and potential sulfate reduction in gravity sewer biofilm, Water Sci. Tech., 31(7), 159–167. Parkhurst, J.D. and R. D. Pomeroy (1972), Oxygen absorption in streams, ASCE, J. Sanit. Eng. Div., 98(SAI), 101. Raunkjaer, K., P.H. Nielsen, and T. Hvitved-Jacobsen (1997), Acetate removal in sewer biofilms under aerobic conditions, Water Res., 31(11), 2727–2736. Sneath, R.W. and C. Clarkson (2000), Odour measurement: A code of practice, Water Sci. Tech., 41(6), 25–31. Spanjers, H., P.A. Vanrolleghem, G. Olsson, and P.L. Dold (1998), Respirometry in control of the activated sludge process: Principles, IAWQ Scientific and Technical Report no. 7, p. 48. Standard Methods for the Examination of Water and Wastewater, 20th Edition (1998), American Public Health Association, American Water Works Association and Water Environment Federation, Washington DC. Stuetz, R.M., R.A. Fenner, S. J. Hall, I. Stratful, and D. Loke (2000), Monitoring of wastewater odours using an electronic nose, Water Sci. Tech., 14(6), 41–47. Tanaka, N. and T. Hvitved-Jacobsen (1998), Transformations of wastewater organic matter in sewers under changing aerobic/anaerobic conditions, Water Sci. Tech., 37(1), 105–113. Tanaka, N. and T. Hvitved-Jacobsen (1999), Anaerobic transformations of wastewater organic matter under sewer conditions. In: I. B. Joliffe and J. E. Ball (eds), Proceedings of the 8th International Conference on Urban Storm Drainage, Sydney, Australia, August 30–September 3, 1999, pp. 288–296. Tanaka, N. and T. Hvitved-Jacobsen (2000), Sulfide production and wastewater quality — investigations in a pilot plant pressure sewer, Proceedings from the 1st World Water Congress of the International Water Association (IWA), vol. 5, pp. 192–199. Tsivoglou, E.C. and L.A. Neal (1976), Tracer measurement of reaeration: III, Predicting the reaeration capacity of inland streams, J. Water Pollut. Control Fed., 48(12). Tsivoglou, E.C., J.B. Cohen, S.D. Shearer, and P.J. Godsil (1968), Tracer measurements of stream reaeration: II, Field studies, J. Water Pollut. Control Fed., 40(2).

References Tsivoglou, E.C., R.L. O’Connell, M.C. Walter, P.J. Godsil, and G.S. Logsdon (1965), Tracer measurements of atmospheric reaeration: I, Laboratory studies, J. Water Pollut. Control Fed., 37(10), 1343–1363. Vollertsen, J. and T. Hvitved-Jacobsen (1999), Stoichiometric and kinetic model parameters for microbial transformations of suspended solids in combined sewer systems, Water Res., 33(14), 3127–3141. Vollertsen, J. and T. Hvitved-Jacobsen (2001), Biodegradability of wastewater — a method for COD-fractionation, Proceedings of the 2nd International Conference on Interactions between Sewers, Treatment Plants and Receiving Waters in Urban Areas (INTERURBA II), Lisbon, Portugal, February 19–22, 2001, pp. 25–33.

Applications — Integrated Process Design and Operation of Sewers

It is not possible and not a purpose of this text to deal with a great number of specific details of the chemical and microbial processes that are relevant for the different types of collection systems that exist. The number of types and corresponding variability are legion, determined by tradition and site-specific demands. The entire text is, therefore, developed to focus on the general understanding of in-sewer chemical and microbial processes, the conditions for them to proceed and the process engineering of the urban wastewater system. The process interactions between the sewer network and the other parts of the urban wastewater system, in particular, the treatment plant, are important in this respect. With knowledge and tools for prediction of the in-sewer processes as the basis, the practitioner is encouraged to integrate these and their relevant corresponding impacts in the management of different types of sewer networks. Case studies selected for this chapter should provide guidance in this respect. Tools in terms of empirical and conceptual models that have been described in the previous chapters are considered useful. In addition to engineering students, planners, engineers and regulatory officials who are dealing with design, operation and management of sewer systems may find such applications of in-sewer processes beneficial.

In-sewer processes may have adverse effects on the sewer itself and the surrounding environment. Well-known examples are the health, corrosion and odor problems that are related to sulfide formation. Other examples are

associated with the interactions between the sewer and the treatment plant in terms of the quality changes of the wastewater that occur in the sewer network. The traditional approach for wastewater treatment plant design is to relate the treatment requirements to the flow and quality of the wastewater that exists at the inlet to the plant. This approach basically ignores the fact that “untreated” wastewater as it appears at the inlet to the treatment plant can be managed during transport in a sewer network to comply with the treatment processes. Contrary to this approach, the integrated process design concept does not just consider wastewater as being subject to quality transformations under transport in the collecting system. It is the fundamental difference compared with the traditional approach that these transformations are manageable by the designer and operator. For the treatment plant design and operation, it means that the “untreated” wastewater as it occurs at the inlet to the plant can be designed to comply with the treatment processes, i.e., the effluent standards. Figure 8.1 gives an overview of the two different approaches for wastewater management and thereby, outlines the concept of “wastewater design.”

The process conditions in terms of availability of the electron acceptor are crucial for the type of process — aerobic, anoxic or anaerobic — that proceeds

Structural and Operational Impacts on Wastewater Quality

A1: Flow regime and degree of turbulence, e.g., slope of sewer line, pipe diameter and degree of pipe uniformity A2: Ventilation, e.g., use of airtight manhole lids or forced ventilation

A3: Sewer capacity, i.e., residence time of the wastewater A4: Relative capacity, i.e., water depth-to-pipe diameter ratio A5: Flow velocity, i.e., shear stress

B1: Source control [households (e.g., water consumption, urine separation and use of kitchen grinders), discharges from industry and reduced infiltration] B2: Injection of oxygen or nitrate

Exchange of volatile compounds across the air–water interface, e.g., oxygen (reaeration that affects aerobic or anaerobic conditions) and release of odorous substances Release of odorous substances to the urban atmosphere and change of reaeration due to a lower atmospheric oxygen concentration Extent of the processes Reaeration and relative importance of microbial transformations in the water phase and in the biofilm Development of sewer biofilm and sediments; thereby an effect on the corresponding processes Sewer upstream quality that affects the in-sewer processes and the downstream quality of the wastewater Impact on process conditions (aerobic, anoxic or anaerobic)

(cf. Table 1.1). Table 1.1 also shows that the type of sewer to a great extent has an influence on the type of process that is favored. This fact is highly important because it can be used in the opposite direction. The sewer network designer and operator thereby have a guideline for which structural and operational conditions should exist to enhance the microbial processes that are preferred. Table 8.1 gives an overview of structural and operational measures that affect the type and course of the in-sewer dry-weather processes. The table exemplifies how process design and operation of sewer networks can be implemented. In other words, it is an overview of methods for “wastewater design.” The impact of the different structural or operational measures on the sewer processes and wastewater quality characteristics can be assessed by model simulation. Examples 8.1 and 8.2 illustrate how structural measures, primarily related to A4 in Table 8.1, affect the in-sewer processes.

Example 8.1: Relative importance of the biofilm to bulk water ratio on the DO consumption in a gravity sewer The total in-sewer respiration in terms of a DO removal rate originates in the case of no contribution from sewer deposits from the biofilm and the bulk water microbial processes. The total respiration rate is, therefore, expressed as follows: rtot =

A rf + rw V

where rtot A V rf rw

total rate of DO consuming processes in a sewer (gO2 m 3 h 1) area of biofilm (m2) volume of bulk water (m3) rate of DO consuming processes in the biofilm (gO2 m 2 h 1) rate of DO consuming processes in the bulk water phase (gO2 m 3 h 1)

Figure 8.2 illustrates the DO consumption rate of the biofilm, rf, in percentage of the total DO consumption rate, rtot, versus the A/V ratio. The calculations are based on DO and substrate nonlimiting conditions. The value of rf is considered constant and equal to 1.0 gO2 m 2 h 1, and rw varies from 2–20 gO2 m 3 h 1. In a pipe with a diameter of 500 mm, a daily variability of the water depth

Structural and Operational Impacts on Wastewater Quality

between 100 and 250 mm results in A/V ratios of 17 and 8 m 1, respectively. From Figure 8.2, it is seen that the corresponding DO consumption rate of the biofilm in percentage of the total rate is about 60 and 40%, respectively, if the bulk water respiration rate is 10 gO2 m 3 h 1. The relative importance of the biofilm compared with the bulk water transformations of the wastewater is not just a question of flow conditions in terms of a daily variation. It is possible in the design phase of the sewer network to control the contribution of the biofilm by selecting a different — smaller — size pipe.

Example 8.2: Reaeration and DO variability of a gravity sewer pipe The example illustrates how the flow conditions of a sewer pipe affect the reaeration and the resulting DO concentration in the wastewater. As an example, a gravity sewer with a pipe diameter of 500 mm and a slope s 0.003 m m 1 is selected. The sewer is without deposits but with a biofilm on the wetted perimeter. The DO consumption rate of the bulk water phase, rw, is at 10 C assumed to have a maximum value equal to 5 gO2 m 3 h 1, however, is limited by the magnitude of the reaeration. The DO consumption rate of the biofilm, rf , is considered a 1-order process in the DO concentration by following Equation (5.12). In Figure 8.3, the oxygen transfer coefficient, KLa, the flow velocity, u, the bulk water DO concentration and the DO consumption rate of the biofilm, rf , are all plotted versus the flow, Q, under steady state conditions in a gravity sewer pipe under the conditions given. Figure 8.3 shows that the sewer is full flowing at 775 m3 h 1 (215 L s 1). It also shows that the reaeration and the DO concentration vary considerably with the flow conditions. At rather low flow rates, the DO concentration is about 2–4 gO2 m 3, a level that is significantly reduced even at flow rates that are below those corresponding to a half-full flowing pipe. Example 8.2 shows that it is possible to control the magnitude of the aerobic transformation of the wastewater by the selected level of the flow compared with the capacity of the sewer. Referring to Table 8.1, it is important to stress that a specific structural change may have an impact on more than just one process, and that it can also affect the total resulting impact of a change in opposite directions. As an example, turbulence in the water phase may increase the reaeration rate of the water phase; increasing the emission rate of hydrogen sulfide and odorous substances thereby intensifies the negative effects of anaerobic conditions. Other aspects concern the complex temperature impacts of the in-sewer processes. Sewer processes, microbial as well as physicochemical like reaeration, are temperature dependent. The different processes interact, and the overall temperature dependency of a specific phenomenon may therefore not

necessarily exert the effects that are expected. As an example, the aerobic biodegradation rate increases with temperature, however, with a significantly higher rate than the reaeration rate (cf. Tables 6.7 and 6.8). An increase in the temperature, therefore, increases the aerobic microbial activity and, consequently, the DO consumption rate without being compensated by the reaeration. In a gravity sewer, this fact may result in a low DO concentration and, consequently, a DO limited microbial activity followed by a buildup of easily biodegradable organic matter, SS, that is primarily produced by hydrolysis of fast hydrolyzable substrate, XS1. The VFAs produced by fermentation may not to the same extent appear during winter when the DO concentration is typically high. Therefore, it is likely that wastewater becomes more biodegradable at a high temperature than at a low temperature — in fact, a quite different result would then be expected. On the other hand, XS1 may be available in a relatively higher concentration during a winter season because of a lower hydrolysis rate. These highly dynamic and complex in-sewer processes, their overall effect on the quality of the wastewater and the impact on the processes in a

Tools for Prediction of Sewer Processes

downstream located treatment plant can only be quantitatively evaluated by taking such interactions into account. This brief example of temperature dependency calls for a simulation tool in terms of a model that can be used for assessment of the quality transformations of wastewater in a sewer and the corresponding impacts. The empirical and conceptual models for prediction of the in-sewer processes that are dealt with in Chapters 4 through 7 are crucial in this respect.

Models are required to apply sewer processes engineering whenever it concerns the sewer itself or its interaction with the treatment plant processes. The text has not only provided such tools but also emphasized the fundamental understanding of the in-sewer processes from an application point of view, together with knowledge on sewer process studies as a prerequisite to generate model parameters. The focal point of the text concerns process design of sewer networks. However, without a fundamental knowledge of the nature of the in-sewer processes, tools for their prediction, interactions and impacts are not just useless but are also risky. The text deals with the microbial and chemical process engineering of sewer networks. It emphasizes dry-weather processes and not the wet-weather impacts that are primarily controlled by physical processes. Under such conditions, the physical in-sewer processes in terms of, for example, hydraulics, sediment and biofilm erosion and solids transport are important. A quite different approach must be applied when wet-weather conditions in sewers dominate. However, wet-weather performance of sewers also requires that sediment deposition be dealt with during dry-weather periods. It is the actual objectives and the availability of data that basically determine whether simple empirical equations or complex conceptual process models should be preferred. A fundamental requirement is, of course, that the conceptual model exists. Empirical equations should be used with precaution because too much specific information may be included. The risk of using empirical equations outside their “area of definition” must be carefully observed. The conceptual process models, however, tend to describe a phenomenon from its fundamental theoretical basis and typically provide the user with an integrated approach. A conceptual understanding behind a model gives a general applicable approach that is often not possible when using an empirical model. The corresponding requirement is, of course, an integrated understanding of the relevant system and its behavior without which the model is without sense. The risk

of using a complex conceptual model is that its complexity is ignored, i.e., model parameters are selected without a sufficient amount of data. The concept depicted in Figure 6.10 and in a model version shown in Table 6.6 integrates aerobic and anaerobic processes relevant for sewer networks. As a prototype model version, this concept exists under the name “Wastewater Aerobic/Anaerobic Transformations in Sewers,” WATS model (HvitvedJacobsen et al., 1999). It is important to stress that the process understanding behind the WATS model is under continuous development. The WATS model is, therefore, dynamic in the sense that progress in the fundamental understanding of the in-sewer processes will have a corresponding influence on the details and the extent of the processes included. What is important is the basic understanding of the integrated microbial processes and the physicochemical exchange and transport processes that are included. In particular, the activity and central position of the living biomass is crucial to observe. These fundamental characteristics are considered the strength of the WATS model concept. The specific processes and components included and the detailed description of the processes are under continuous consideration and development. At present, the WATS model includes the following main aspects and simulation options of the in-sewer carbon and the sulfur cycle under dry-weather conditions (cf. especially Chapters 5 and 6): • quality transformations for the wastewater organic matter (COD) in terms of its biodegradability • DO mass balance of the water phase • sulfide formation • description of transformations under varying aerobic and anaerobic conditions Briefly expressed, the WATS model simulates changes in the electron donor and the electron acceptor under both aerobic and anaerobic conditions. The WATS model is formulated in deterministic terms. However, an extension to include simple Monte-Carlo stochastic simulation is possible, taking into consideration a measured variability of the process parameters. Extension of the WATS model to integrate further dry-weather processes is considered important. Examples of such extensions are the description of the wastewater quality and nitrite/nitrate transformations under anoxic conditions and the emission of hydrogen sulfide into the sewer atmosphere followed by its transformation (oxidation) at the sewer walls. An extension of the WATS model to include wet-weather conditions requires a conceptual change by strengthening the physical processes in terms of solids deposition, erosion and transport. Quality aspects still play a role, however, in a different way, because the transformations that proceed during

Tools for Prediction of Sewer Processes

transport are considered to be of minor interest in the aerobic, diluted wastewater. The biodegradability of the solids that are eroded from the biofilm and the deposits may, in the case of combined sewer overflows into receiving waters, be of interest (cf. Section 8.5).

The hydraulic performance of sewer pipes can be described at different levels. In the case of nonstationary, nonuniform flow, the Saint Venant Equations should be applied. However, under dry-weather conditions, the Manning Equation is an adequate description of the wastewater flow in a gravity sewer pipe when considering the prediction of wastewater quality changes under transport. There are no grounds for using advanced hydraulic models because of the uncertainties in the prediction of the microbial transformations of the wastewater. The Manning Equation and the Equation of Continuity are the basic hydraulic tools for description of the wastewater flow in a pipe at stationary, uniform flow: Manning Equation: U = MR2 / 3s 0.5

(8.1)

where U flow velocity of wastewater (m s 1) M Manning number (m1/3 s 1) R hydraulic radius, i.e., the cross-sectional area of the water volume divided by the wetted perimeter (m) s pipe slope (m m 1) Equation of Continuity: Q = UA

(8.2)

where Q wastewater flow (m3 s 1) A cross-sectional area of the water phase (m2) Equations (8.1) and (8.2) are normally sufficient for describing wastewater flow in a sewer pipe related to the prediction of dry-weather in-sewer processes.

From a process point of view, it is typical to distinguish between wastewater collection in pressurized systems and gravity sewer networks. The reason is that anaerobic conditions typically prevail in pressure sewers, whereas in gravity sewers, conditions are established for aerobic processes. Different empirical models for pressure mains and gravity sewers have, therefore, been developed. The integrated aerobic-anaerobic WATS model has changed this situation. As an example, it is possible to use the model in a gravity sewer with changing aerobic and anaerobic conditions. As previously stressed, a number of in-sewer processes still need to be dealt with. Examples are the anoxic transformations and the processes related to the extended sulfur cycle, particularly, the oxidation of sulfide and the emission of hydrogen sulfide into the sewer atmosphere, including its further oxidation at the sewer walls. Combined use of empirical and conceptual models is still needed. Generally considered, removal and preservation of the biodegradable organic matter in sewers takes place under aerobic and anaerobic conditions, respectively. Removal of biodegradable organic matter in a sewer network is preferred when criteria relevant for physicochemical or mechanical treatment must be observed. Preservation of the biodegradable part of the COD is, however, important in the case of advanced biological treatment, i.e., related to denitrification and biological phosphorus removal. The microbial transformations of the wastewater that proceed in the sewer are integrated with the treatment processes. In other words, the “treatment process” of the wastewater has already started before it reaches the treatment plant. The following two examples describe two different perspectives of sewer and treatment plant interactions. In the Costa do Estoril example from Portugal, removal of organic soluble and colloidal substrates in the sewer will enhance the effectiveness of the physicochemical treatment before discharge of the treated wastewater to the coastal zone takes place. Contrary to this, preservation of readily biodegradable and fast hydrolyzable substrates is important during wastewater transport in the Emscher intercepting sewer in Germany. Such substrates will improve the conditions for denitrification and biological phosphorus removal in the treatment processes before the treated wastewater is discharged into the river Rhine. The simulation of the in-sewer processes for the two examples is performed by the WATS model (cf. Table 6.6). The quality of the wastewater that enters a sewer in terms of COD fractions from the Costa do Estoril catchment is

Model Simulations of Sewer and Treatment Plant Interactions

considered equal to the corresponding quality from the Emscher catchment. Thereby, it is possible to compare the results of the processes that proceed in the two different sewer systems.

The Costa do Estoril sewerage system serves about 720,000 inhabitants located in the western part of Lisbon, Portugal. The system includes a 26 km gravity interceptor sewer, some important gravity tributary sewers, nine pumping installations, an underground treatment plant and a long sea outfall with submerged diffusers (Figure 8.4). The treatment plant is in the process of being upgraded to advanced physicochemical treatment including filtration and disinfection of the wastewater. The interceptor with an internal diameter varying from 1.5–2.5 m has a low average slope of 0.0008 m m 1 along the sewer line. Except for the first reach, anaerobic conditions normally exist in the interceptor. Provided that aerobic conditions exist, the long intercepting sewer is well suited for microbial transformations that support the physicochemical treatment processes. The aerobic removal of readily biodegradable and fast hydrolyzable substrates and the production of slowly biodegradable organic matter in terms of biomass will enhance the efficiency of the entire treatment of

the wastewater. Aerobic conditions in the interceptor may be established by introducing a number of aeration stations in the sewer. Using the WATS model, Figure 8.5 shows the DO variability along the sewer line as a result of a scenario with aeration up to 70% of saturation when the DO concentration has been reduced to 0.5 gO2 m 3. As seen from the figure, a total of 14 aeration stations in the interceptor will be needed. The number of stations may, however, be reduced if pure oxygen or hydrogen peroxide is added. The result of the aerobic scenario in terms of quality changes of the wastewater in the intercepting sewer is shown in Table 8.2. The in-sewer treatment is assessed by the changes that take place in the easily biodegradable substrates (readily biodegradable and fast hydrolyzable substrates) and in the

Easily biodegradable substrates Slowly biodegradable substrates Total

160

70

90

56

440 600

480 550

40 50

9 8

Model Simulations of Sewer and Treatment Plant Interactions

relatively slow biodegradable substrates making up the rest of the COD. The WATS model simulations have shown that under the existing anaerobic conditions in the sewer, insignificant changes in these COD fractions proceed. Table 8.2 shows a moderate reduction in the total COD concentration under transport of the wastewater in the sewer. Also important is an increase in the slowly biodegradable or nonbiodegradable COD fractions — primarily corresponding to production of particulate organic matter (heterotrophic biomass). This COD fraction can undergo removal by the physicochemical treatment process. However, the most important result of the aerobic biotransformations in the sewer is the considerable reduction in the easily biodegradable, i.e., soluble and colloidal, COD fractions that otherwise would be discharged into the coastal zone. The predicted treatment in the interceptor is supported by corresponding aerobic transformations of the wastewater in the tributaries. In general, the slopes in the tributaries to the interceptor are relatively steep, and a number of sewer drops exist. The wastewater in these tributaries typically flows under aerobic conditions, and heterotrophic processes proceed. Field studies that have been undertaken by Almeida (1999) have shown that these transformations are in agreement with the results shown in Table 8.2. The example illustrates that a highly positive interaction in terms of treatment potential between aerobic microbial transformations in a sewer and a subsequent physicochemical treatment take place. The total outcome of this interaction in terms of pollutant reduction will depend on the effectiveness of the final treatment step. This result must be assessed by the effects of the resulting discharges of the relevant COD fractions into the receiving water, not as a percentage of a COD reduction at the outlet from the final treatment processes. A fundamental requirement for using a sewer as a treatment system followed by subsequent physicochemical or mechanical treatment is often the installment of aerators. Contrary to what has been proposed by a number of authors, the limitation in using the sewer as a treatment system is normally not the biomass. Addition (circulation) of activated sludge is, therefore — except for cases with excessive aeration — in general, of no interest. Substitution of oxygen with nitrate as electron acceptor is possible, but a reduced rate of transformation is expected.

The river Emscher in the Ruhr district, Germany, is a tributary to the river Rhine (Figure 8.6). Due to heavy industrialization in the area that took place about 100 years ago, the river Emscher and its tributaries were systematically developed as open sewers. When, in the 1950s, it became evident that biological treatment of the wastewater was urgently necessary, it was decided to

construct a treatment plant at the mouth of the river Emscher for treatment of the entire flow of the Emscher and also its tributaries. However, at the end of the 1980s, the open sewer concept was no longer able to fulfill new demands in terms of reduced odor nuisance, improved aesthetics and needs for recreational areas in the densely populated area. It was decided to gradually stop the transport of wastewater in open sewers, to decentralize wastewater treatment and to return the Emscher river and its tributaries to as natural a state as possible. A period of 25 years and a total investment of about 4.4 billion USD were planned for the realization of the restructuring program (Stemplewski et al., 1999). A drastic reduction in the production of wastewater in the Emscher catchment from 4.6 to 3.7 million person equivalents, primarily caused by industry, changed during the period from 1995–97 the need for wastewater treatment plant capacity. Additional treatment capacity was, contrary to previous expectations, not needed (Figure 8.6). However, to distribute the wastewater over the three existing wastewater treatment plants, an approximately 50 km intercepting sewer will have to be installed parallel to the river Emscher. The aim of this sewer is to collect wastewater from populated areas within the river catchment and divert it to two large plants, Bottrop and Dinslaken, each having a non-used treatment capacity (Figure 8.6). The following primary criteria for sewer performance were set: • to avoid permanent deposits of sewer solids • to ensure that no serious problems caused by sulfide formation occur • to preserve high biological degradability of the wastewater during transport

Model Simulations of Sewer and Treatment Plant Interactions

to enhance denitrification and biological phosphorus removal at the treatment plants The second and the third criteria were assessed from simulation using the WATS model. Parameters for this simulation originated from field experiments and OUR experiments of the future wastewater inflows to the intercepting sewer. These results, corresponding to the parameters shown in Tables 6.5, 6.7 and 6.8, were used for calibration of the model. Model simulations of selected sewer scenarios including both gravity sewers and pressure mains have shown that it was possible to avoid major sulfide problems and, at the same time, preserve the easily biodegradable COD fractions if a gravity sewer is appropriately designed (Figure 8.7). In this respect, it is crucial to establish flow conditions that prevent permanent deposits of sewer solids and reduce reaeration. Results from simulations with the WATS model show that aerobic and anaerobic processes take place simultaneously in the gravity sewer. It can be readily seen from the simulations that the DO concentration is reduced to a low value (Figure 8.8). The DO level is determined by the balance between the reaeration and the microbial oxygen uptake rate of the aerobic sewer processes. In the case of the Emscher sewer, reaeration takes place by means of the slope of the sewer line and the flow conditions designed to limit the aerobic microbial processes. The low reaeration rate causes a corresponding low aerobic activity of the microorganisms in the wastewater and the biofilm compared with the potential magnitude at a nonlimiting DO concentration. Therefore, a relatively low amount of the easily biodegradable fractions of COD will be removed. As far as the subsequent biological removal of nitrogen and phosphorus at the treatment plants in Bottrop and Dinslaken is concerned, this fact is very important to observe when transporting wastewater over a long distance with a corresponding long residence time. Compared with the present situation, and depending on the temperature, the daily average amount of easily biodegradable substrate will be increased about 25% at the two treatment plants. The simulations depicted in Figure 8.8 also show that a rather low hydrogen sulfide concentration is predicted in the gravity sewer. Only minor problems related to hydrogen sulfide production may therefore arise. Until now, the WATS model did not include sulfide release to the sewer atmosphere, sulfide oxidation or sulfide precipitation that may further reduce the concentrations shown. The predicted sulfide concentrations are, therefore, maximum levels. In case a natural capacity of iron salts in the wastewater to precipitate sulfide is inadequate, the sulfide concentrations are considered at a level that can be relatively easily controlled. Different scenarios of interceptor systems between Dortmund and Dinslaken were compared. As an example, sulfide formation in a gravity sewer

Model Simulations of Sewer and Treatment Plant Interactions

and a pressure main was simulated. Figure 8.9 shows higher sulfide concentrations in the wastewater of a pressure main compared with a gravity sewer. Different residence time and area/volume ratios of the two systems are major reasons for this. It is for both sewer systems seen that the highest sulfide concentrations exist in the upstream parts of the system because of a relatively high surface area/volume ratio. Sulfide reduction along the sewer reach mainly occurs because of dilution by the inflowing wastewater without sulfide. When comparing sulfide formation for the two sewers, it must, as already mentioned, be realized that the simulated results for the gravity sewer are considered to be overestimated. The risk of a low DO concentration, particularly in a long sewer, is the formation of hydrogen sulfide. The estimated sulfide problems in the case of a gravity sewer are considered marginal and controllable. To walk on the razor’s edge and avoid sulfide problems and aerobic removal of easily biodegradable organic matter therefore seems possible. The river Emscher is to be restructured to become an attractive urban river, a natural part of its surroundings, and with its own unique character. The 50 km

Sewer Processes in an Integrated and Sustainable Perspective

intercepting sewer, considered a reactor for microbial processes, is an important means for improved interaction between wastewater collection and wastewater treatment in existing plants. In terms of cost benefit, the sewer is a cheap alternative to three new wastewater treatment plants (Figure 8.6). Indirectly, the sewer is, as a bioreactor, a very important prerequisite for an ecological and recreational upgrading of the Emscher catchment area.

Although the text concerns in-sewer processes, it has further perspectives, because wastewater occurs in several systems. The text might, therefore, have been written not as a “sewer processes” book but as a text on “wastewater processes.” The author could have done so at the general level but could not have included a corresponding specific experience. At the end of the book, it is, however important to mention that the fundamental knowledge of the book is not limited to sewer networks. Wastewater occurs not just in pipes and open channels but, just as an example, in many countries, also in different types of wastewater treatment ponds. Taking wastewater processes into account, there is a perspective for further improvement of such treatment systems. In several cases, it has been stressed in the text that wastewater and activated sludge, as far as the microbial processes are concerned, perform differently. When concerned with microbial and physicochemical processes, this fact is always important to remember, i.e., to consider wastewater as wastewater and activated sludge as activated sludge.

Wet-weather processes have, in general, been excluded in the text, because they are based on a different concept and perform differently. Microbial and physicochemical processes are contrary to the physical processes dominating in sewers during dry-weather transport of the wastewater. When dealing with combined sewer networks in terms of pollutant loads during overflow events, dry-weather solids deposition and erosion and solids transport during high-flow events are, in addition to the rainfall/runoff hydraulic and sewer solids characteristics, the central physical in-sewer processes. Quite different process approaches are, therefore, required to describe dry-weather and wet-weather sewer performance. The most important quality aspect of wet-weather sewer performance is related to the combined sewer overflows (CSOs) and their impacts on the

adjacent receiving waters. In addition to the stormwater runoff (SWR) that transports pollutants from the rainwater and the urban surfaces, the CSOs furthermore include pollutants that originate from the diluted wastewater and the eroded sewer solids. The dominance of the physical processes in a sewer related to wet-weather discharges does not mean that microbial and physicochemical processes in sewers play no role. However, the nature of these processes is different compared with what is observed during dry weather. In this respect, transformation of the components during transport in the sewer is typically of less importance. The following microbial process aspects should, however, typically be considered: • the biofilm growth characteristics during the antecedent dry-weather period • the biodegradation characteristics of the sewer solids that are eroded from the sediments and the biofilms The in-sewer heterotrophic microbial processes concept that is dealt with under dry-weather conditions can be extended to include these two wet-weather quality aspects. Although the microbial processes play a minor in-sewer role, they are important in terms of the receiving water impacts from CSO discharges. This aspect is relevant for those impacts of the CSO discharges that are related to biodegradation of the organic matter, e.g., in terms of oxygen depletion or growth of heterotrophic biofilms in the receiving waters. The traditional description of receiving water impacts from CSO discharges related to the soluble and particulate fractions of COD is in such situations an approach with limitations. The soluble fraction includes substances with different biodegradation characteristics, i.e., inert, readily biodegradable and fast hydrolyzable COD, and the particulate fraction consists of different types of solids in terms of adsorption characteristics, settling rates and biodegradability (Vollertsen et al., 1999). Fundamental microbial characteristics, i.e., biomass growth and corresponding substrate utilization, are not taken into account in the traditional wet-weather approach. Furthermore, at present, the description is based on loads at the sewer/receiving water interface and normally has no process links to solids erosion and washout phenomena in the sewer. A new concept for improved CSO impact assessment must include physical and microbial characteristics and processes. As far as the microbial heterotrophic transformations are concerned, intensive investigations have shown that suspended particles originating from sewer sediments follow the concept for wastewater depicted in Figure 5.5 (Vollertsen and Hvitved-Jacobsen, 1998; Vollertsen and Hvitved-Jacobsen, 1999; Vollertsen et al., 1999). This finding is important, because it shows that the concept and corresponding model developed for transformations of wastewater in sewers

Sewer Processes in an Integrated and Sustainable Perspective

have potential for being expanded to include transformations of the CSO discharges into receiving waters. The concept seems to be valid across the interface between the sewer and the receiving water system. Figure 8.10 defines the corresponding integrated framework for transport and transformation. A successful implementation of the concept depicted in Figure 8.10 depends on a description of the pathways for sewer solids. A number of steps should be known in this respect, i.e., erosion of the sewer sediment and biofilm, solids transport through the sewer and the CSO structure, final degradation, and physical-chemical adsorption or sedimentation of specific solids fractions in the receiving waters. Only comprehensive and well-designed field investigations carried out in sewers and receiving waters will provide the experimental evidence for the complicated processes that the sewer solids will undergo. Wet-weather processes are subject to high variability. A simple deterministic model result in terms of the impacts on the water quality is out of scope. From a modeling point of view, a stochastic description is a realistic solution for producing relevant results. Furthermore, an approach based on a historical rainfall series as model input is needed to establish extreme event statistics for a critical CSO impact that can be compared to a water quality criterion. In terms of CSO design including water quality, this approach is a key point.

The urban water system that we are dealing with today, in principle, dates back to the 1800s. The growth of the cities during the industrial development of Europe and the United States required a technical solution for urban water management that could cope with the high quantities of water needed and the associated amount of wastewater produced. A large-scale system for supply of water and collection of wastewater was technologically developed as a pipe system in and out of the city. Sustainability is a new challenge for the urban water management of today. During the last decade, the large-scale pipe system out of the city, i.e., the sewer network, and the associated treatment concept for the wastewater has been proclaimed a nonsustainable solution for management of the wastewater. The existing centralized method of urban wastewater management is difficult to defend if sustainability is considered just met by small-scale and local solutions. Such solutions are, however, only more sustainable compared with a centralized solution, if they in total are winning. The author is not sure that this comparison is always made honestly when small-scale and local solutions are proclaimed sustainable. It must be realized that urban wastewater management is difficult to cope with in a sustainable way, because the entire idea is to use resources and pollute them! In spite of this, the sustainable development of the urban infrastructure must be seriously considered, particularly because the centralized idea behind the urban water cycle, without doubt, will exist for an unknown future. Except for continued improvement and upgrading of the urban wastewater system that is considered a general trend of a technological development, only a few technical solutions have contributed to an improved sustainability. The attempt to reduce water consumption and make the use of water more efficient in households and industry is correct in terms of sustainability as long as a high health standard of the population is maintained. In addition to a changed attitude of people, a number of water-conserving technologies have contributed to this development. However, reduced water consumption and renovation of the sewer network to reduce the infiltration may also create adverse effects for the existing sewer network. Such problems will, as an example, develop because of an increased amount of solids deposits in gravity sewers and a prolonged anaerobic residence time in pressure mains. Planners, designers and operators have a tremendous task to find both nonstructural and technical sustainable solutions to the future development of sewer networks. It is the author’s opinion that an extended integrated performance of the urban wastewater infrastructure — internally as well as with the surroundings — is fundamental for improved sustainability. The concept of the

References

sewer as a bioreactor has an overall objective to deal with integrated microbial interactions across the sewer/treatment plant interface, and it approaches in-sewer process interactions with the surroundings. In this respect, the sewer process concept is a sustainable approach. Contrary to the “hard” engineering solutions typically applied when implementing solutions for the collection of wastewater, the process concept tends to give the sewer a “value” as a process reactor integrated with the surroundings. This is a new approach for those who manage the sewer network today. Probably, the reader of the text has already experienced this. Briefly expressed, the sustainable approach of the sewer process concept can be interpreted by changing wastewater management from an “end-of-pipe treatment” to a “pipe and plant treatment.” This is not the only way of approaching a sustainable solution for an urban wastewater system, however, it is a contribution. But, it is certainly true that the sewer process concept tends to put much more focus on the dry-weather performance of the sewer than is typically done.

Almeida, M. do C. (1999), Pollutant transformation processes in sewers under aerobic dry weather flow conditions, Ph.D. dissertation, Imperial College of Science, UK, p. 422. Hvitved-Jacobsen, T., J. Vollertsen and N. Tanaka (1999), Wastewater quality changes during transport in sewers — an integrated aerobic and anaerobic model concept for carbon and sulfur microbial transformations, Water Sci. Tech., 39(2), 242–249. Stemplewski, J., S. Schlegel, A. Stein, W. Geisler, K.-G. Schmelz, T. Hvitved-Jacobsen and J. Vollertsen (1999), Restructuring the Emscher system, Proceedings from the 11th EWPCA (European Water Pollution Control Association) Symposium on Sewerage Systems — Cost and Sustainable Solutions, May 4–6, 1999, Munich, Germany, p. 14. Vollertsen, J. and T. Hvitved-Jacobsen (1998), Aerobic microbial transformations of resuspended sediments in combined sewers — a conceptual model, Water Sci. Tech., 37(1), 69–76. Vollertsen, J. and T. Hvitved-Jacobsen (1999), Stoichiometric and kinetic model parameters for microbial transformations of suspended solids in combined sewer systems, Water Res., 33(14), 3127–3141. Vollertsen, J., T. Hvitved-Jacobsen, I. McGregor and R. Ashley (1999), Aerobic microbial transformations of pipe and silt trap sediments from combined sewers, Water Sci. Tech., 39(2), 234–241.

Nomenclature

A list of components, constants and system characteristics related to a conceptual description of microbial processes of wastewater in sewers is provided. The list includes symbols for components and parameters relevant for processes under both aerobic and anaerobic conditions (see Chapters 5 and 6). COMPONENTS Dissolved Components SA SALK SF SO SOS SS SH2S

fermentation products (gCOD m 3) bicarbonate alkalinity (molHCO3 m 3) fermentable, readily biodegradable substrate (gCOD m 3) dissolved oxygen (gO2 m 3) dissolved oxygen saturation concentration (gO2 m 3) readily biodegradable substrate (gCOD m 3) total sulfide (gS m 3)

Particulate components XB heterotrophic active biomass (gCOD m 3 or gCOD m 2) • XBf heterotrophic biomass in the biofilm (gCOD m 2) • XBw heterotrophic active biomass in the water phase (gCOD m 3) XM methane-producing biomass (gCOD m 3) XSn hydrolyzable substrate, fraction n (gCOD m 3) • n 1: fast, n 2: slow • n 1: fast, n 2: medium, n 3: slow

Gas Component SCH4

methane (gCOD m 3)

STOICHIOMETRIC AND KINETIC CONSTANTS khn hydrolysis rate constant for fraction n (d 1) • n 1: fast, n 2: slow • n 1: fast, n 2: medium, n 3: slow kH2S hydrogen sulfide formation rate constant (h 1) k1/2 1/2-order rate constant (gO20.5 m 0.5 d 1 ) Kfe saturation constant for fermentation (gCOD m 3) • Kfef in the biofilm • Kfew in the water phase KA saturation constant for SA (gCOD m 3) • KAf in the biofilm • KAw in the water phase KALK saturation constant for alkalinity (molHCO3 m 3 ) KF saturation constant for SF (gCOD m 3) • KFf in the biofilm • KFw in the water phase KO saturation constant for dissolved oxygen (gO2 m 3) KS saturation constant for readily biodegradable substrate (gCOD m 3) • KSf in the biofilm • KSw in the water phase KXn saturation constant for hydrolysis, fraction n (gCOD gCOD 1) • n 1: fast, n 2: slow • n 1: fast, n 2: medium, n 3: slow qfe fermentation rate constant (d 1) • qfef in the biofilm • qfew in the water phase qm maintenance energy requirement rate constant (d 1) YH yield constant for heterotrophic biomass [gCOD, biomass (gCOD, substrate) 1] • YHf in the biofilm • YHw in the water phase YM yield constant for methane-producing biomass [gCOD, biomass (gCOD,substrate) 1] • YMf in the biofilm • YMw in the water phase maximum specific growth rate for heterotrophic biomass (d 1) H maximum specific growth rate for methane-producing biomass (d 1) M

Nomenclature

fe

efficiency constant for the biofilm biomass ( ) • A aerobic conditions • An anaerobic conditions efficiency constant for anaerobic hydrolysis ( )

FLOW AND SEWER SYSTEM CONDITIONS ratio of biofilm area to bulk water volume, i.e., R 1 (m 1) u (g dm) 0.5, Froude number ( ) gravitational acceleration (m s 2) hydraulic mean depth (m) air–water oxygen transfer coefficient, reaeration constant (s 1, h 1 or d 1) ** R hydraulic radius (m) s slope (m m 1) u mean flow velocity (m s 1)

A/V Fr g dm* KLa

GENERAL ENVIRONMENTAL PARAMETERS T temperature ( C) t time [s (second), h (hour) or d (day)] temperature coefficient ( ) • f biofilm • r reaeration • s sulfide production • w water phase SYSTEMATICS If a multiplication symbol is omitted, a space is put between units, e.g., gCOD/m3 or gCOD m 3 Symbols S X K k *The **The

soluble, dissolved component particulate component saturation constant rate constant

cross-sectional area of the water volume divided by the water surface width. cross-sectional area of the water volume divided by the wetted perimeter.

Selected Indices f H h M n

biofilm heterotrophic hydrolysis methane producing fraction number for hydrolyzable substrate • n 1: fast, n 2: slow • n 1: fast, n 2: medium, n 3: slow r reaeration s sulfide production w water phase

Activated sludge, 101–103 Aerobic transformations, 8, 13, 40–41, 106–112 Air–water equilibrium, 66–73 Air–water mass transfer, 73–77 Air–water oxygen transfer, see reaeration Anabolic process, 11–12 Anaerobic transformations, 8, 13, 41–43, 96–112, 158–160 Anoxic transformations, 8, 13, 40–41, 121–125, 154–155 Biochemical process, 11–13 Biodegradable substrate, see readily biodegradable substrate and hydrolyzable substrate Biodegradation, see aerobic, anaerobic and anoxic transformations Biofilm kinetics, 29–33 Biofilms, 56–59, 109–111 Biomass growth, 11–13, 58–59, 100–101, 104–107 Calibration of sewer process model, 191–195 Catabolic process, 12, 14 Chlorine, 153, 157 Combined sewers, 6

Components, see sewer process model and wastewater components Concept for transformation of wastewater, 99–106 Concrete corrosion, see corrosion Corrosion, 145–149 Degradation, 12 Denitrification, see anoxic transformations Design of sewer networks, 205–227 Dissolved oxygen, 86 Electron acceptor, 8, 12–14 Electron donor, 12–14 Electron equivalent for a redox reaction, 21 Electron structure for elements, 17–20 Fermentation, 12, 41–43 Field investigations, 173–174, 179–181, 192–195 First-order reaction, 26–27 Flow calculation, 213 Force mains, see pressure mains Gas-liquid equilibrium, see air–water equilibrium

Index Gas-liquid mass transfer, see air–water mass transfer Gibb’s free energy, 14–16 Gravity sewers, 5–6 Growth limitation, 27–29 Growth, see heterotrophic growth Henry’s law, see also air–water equilibrium, 68–69, 70–73, 75–77 Heterotrophic growth, 11–13, 107–108, 113 Hydraulic performance of sewers, 213 Hydrogen peroxide, 153, 157 Hydrogen sulfide, see sulfide Hydrolysis, 33–35, 43–46, 112 Hydrolyzable substrate, 45–46 Kinetics for heterogeneous reactions, 29–35 Kinetics for homogeneous reactions, 25–29 Kinetics of hydrolysis, 33–35 Laboratory investigations, 172 Maintenance energy requirement, 104, 108 Manning equation, 213 Methanogenesis, 41–43, 62 Modeling, see sewer process model Model simulations, 214–223 Mole fraction, 66–67 Monitoring in sewers, see field measurements Monod kinetics, 27–29 Nitrate addition, 154–155 Odors, 65, 77-85 Odor formation, 41-43, 77 Odor measurement, 180-181 Olfactometry, 181 Operation of sewer networks, 205-227 Oxidation level, 17-21 Oxygen mass balance 115-121 Oxygen transfer, see reaeration

Oxygen uptake rate (OUR), 55, 175–178 Ozone 153, 157 Parameters for sewer process model, 181–200 Particulate substrate, 46, 51, 54–56 Pilot plant studies, 172–173 Pourbaix diagram, 16 Pressure sewers, 5–6 Process model, see sewer process model Process overview, 7–9 Quality of wastewater, 37–40 Readily biodegradable substrate, 45–46, 51–56 Reaeration, 65, 85–91 Redox reactions, 11–25 Respiration, 12, 108–109 Respirometry, see oxygen uptake rate Sampling in sewers, 174–175 Sanitary sewers, 5 Scrubbing, 157–158 Sediments 42-43, 59-62 Separate sewers, 5–6 Septicity, see anaerobic transformations and sulfide Sewer atmosphere, 80 Sewer biofilms, see biofilms Sewer drop, 89–91 Sewer process concept, 99–106 Sewer process model, 112–115, 160–166, 211–213 Sewer process overview, 7–9 Sewer sediments, see sediments Sewer solids, see sediments Slimes, see biofilms Slope of sewer, 88, 117 Solids, see sewer solids Solubility of oxygen, 86 Soluble substrate, 46, 51, 54–56 Stoichiometry of redox reactions, 17–25, 134–135 Storm sewers, 6 Substrate, 11–13, 40–56

Index Sulfate reduction, see sulfide formation Sulfide, 70–71, 197–198 Sulfide control, 149–158 Sulfide effects, 83–85, 145–149 Sulfide emission, 70–73, 81–82 Sulfide formation, 42–43, 60–62, 135–139, 197–198 Sulfide gas, 80–83 Sulfide prediction in gravity sewers, 141–145 Sulfide prediction in pressure mains, 139–141 Sulfur cycles, 131–133 Sustainable wastewater management, 226–227 Temperature dependency of processes, 35–36 Treatment of sewer gases, 157–158 Treatment of sulfide, see sulfide control Two-film theory, 73–77

Validation of sewer process model, 191–195 Ventilation, 82–83 Volatile organic compounds (VOC’s), see odors Volatile fatty acids (VFA’s), 55, 69, 77–79, 196–197 Wastewater components, 46–56, 189–191, 196–197 Wastewater processes, see aerobic, anaerobic and anoxic transformations Wastewater quality, see quality of wastewater Wet weather processes 223–225 Yield, 58 Zero-order reaction, 25–26

Thorkild Hvitved-Jacobsen is a Professor of Environmental Engineering at the Department of Environmental Engineering, Aalborg University, Denmark. His research and teaching focus on environmental process engineering related to the urban wastewater collection and treatment systems, including urban drainage. His research has resulted in over 160 publications in primarily international journals and proceedings. He is the chairman of the IWA/IAHR Sewer Systems & Processes Worvking Group.