Biophysico-Chemical Processes of Anthropogenic Organic Compounds in Environmental Systems

By Baoshan Xing, Nicola Senesi and Pan Ming Huang

In contrast to the classical books which largely focus on separate, individual physicochemical and biological aspects, this book aims to integrate the frontiers of knowledge on the fundamentals and the impact of physicochemical and biological interactions and processes of AOCs in soil, sediment, water and air.  The specific objectives of this book are to address: (1) fundamental biophysico-chemical processes of AOCs in the environment, (2) occurrence and distribution of AOCs in air, water, and soil, and their global cycling, (3) the state-of-the-art analytical techniques of AOCs, and (4) restoration of natural environments contaminated by AOCs.  The book also identifies the gaps in knowledge on the subject matter and as such provides future directions to stimulate scientific research to advance the chemical science on biophysico-chemical interfacial reactions in natural habitats.

By virtue of complex nature of the interactions of AOCs with different environmental components and matrixes, no single available technique and instrument is satisfactory yet for determining their fate, transport, availability, and risk in the environment.  In order to fully understand the biophysico-chemical interactions and processes of AOCs in the environment, it is critical to know chemical, physical and biological properties of AOCs and their analytical techniques. The book is unique because of its multidisciplinary approach as it provides a comprehensive and integrated coverage of biophysico-chemical reactions and processes of AOCs in various environments, associated analytical techniques, and restoration of natural environments contaminated by AOCs. 

• Language: English
• Publisher: Wiley
• Release date: May 16, 2011
• ISBN: 9781118002117

Book preview

Series Preface

Scientific progress is based ultimately on unification rather than fragmentation of knowledge. Environmental science is the fusion of physical and life sciences. Physical, chemical, and biological processes in the environment are not independent but rather interactive processes. Therefore, it is essential to address physical, chemical, and biological interfacial interactions in order to understand the composition, complexity, and dynamics of ecosystems. Keeping separate these domains, no matter how fruitful, one cannot hope to deliver on the full promise of modern environmental science. The time is upon us to recognize that the new frontier in environmental science is the interface, wherever it remains unexplored.

The Division of Chemistry and the Environment of the International Union of Pure and Applied Chemistry (IUPAC) has approved the creation of an IUPAC-sponsored book series entitled Biophysico-Chemical Processes in Environmental Systems published by John Wiley & Sons, Hoboken, New Jersey. This series addresses the fundamentals of physical–chemical–biological interfacial interactions in the environment and the impacts on (1) the transformation, transport, and fate of nutrients and pollutants; (2) food chain contamination and food quality and safety; and (3) ecosystem health, including human health. In contrast to classical books that focus largely on separate physical, chemical, and biological processes, this book series is unique in integrating the frontiers of knowledge of both fundamentals and impacts of interfacial interactions of these processes in the global environment.

With the rapid developments in environmental physics, chemistry, and biology, it is becoming much harder, if not impossible, for scientists to follow new progress outside their immediate area of research by reading the primary research literature. This book series will capture pertinent research topics of significant current interest and will present to the environmental science community a distilled and integrated version of new developments in biological physical, and chemical processes in environmental systems.

This book, Biophysico-Chemical Processes of Anthropogenic Organic Compounds in Environmental Systems, is volume 3 of the Wiley-IUPAC series. This book comprises 22 chapters by renowned experts in the topic and is unique in integrating both fundamentals and impacts of interfacial interactions of physical, chemical, and biological processes pertaining to adsorption, transformation, bioavailability, toxicity, and transport processes of anthropogenic organic compounds in the air–water–soil environment and their global cycling. Further, the most modern techniques used for sampling, extraction, and instrumental analyses and various means for the restoration of natural environments contaminated by organic pollutants are treated.

This book can be used as an advanced reference source on biological, physical, and chemical processes and performance, analytical techniques, and restoration of anthropogenic organic compounds in the global environment for senior undergraduate and graduate students in environmental sciences and engineering. It is an essential reference for chemists and biologists studying environmental systems, as well as for geochemists, environmental engineers; and soil, water, and atmosphere scientists. It will serve as a useful resource book for professors, instructors, research scientists, professional consultants, and other individuals working on environmental and ecological systems.

P. Ming Huang

Nicola Senesi

Series Editors

Preface

Anthropogenic organic compounds (AOCs) are synthetically made organic chemicals. They range from gasoline components (e.g., benzene, toluene, xylene) to emerging contaminants such as endocrine-disrupting chemicals. Because of their wide use and disposal, AOCs are commonly found in our environment such as the water we drink, the air we breathe, and the soil from which we obtain our food. These compounds are often toxic and can severely deteriorate an ecosystem. They can also bioaccumulate through food chains and cause various diseases (and even death) to organisms, including humans. AOCs behave differently in various environmental media which differ in their different physical, chemical, and biological components and processes. Therefore, an in-depth and more complete understanding of the biological, physical, and chemical processes of AOCs in environmental systems is essential for the development of innovative management strategies for sustaining the environment and ecosystem integrity.

Physical, chemical, and biological interfacial interactions and processes govern the fate, transport, availability, exposure, and risk of AOCs. However, the fundamentals of many physicochemical and biological interfacial reactions of AOCs and their impacts on ecosystems remain largely unknown. As a result, predictive models for their fate, transport, and risk in different media are often off target. To advance the frontiers of knowledge on the subject matter would require a concerted and comprehensive effort of scientists in relevant physical and life sciences such as chemistry; mineralogy; geochemistry; microbiology; ecology; environmental engineering; and soil, atmospheric, and aquatic sciences. In addition, physical, chemical, and biological reactions and processes of AOCs in the environment are not independent but rather interactive and closely interrelated. Therefore, it is essential to systematically address these interactive processes and interactions through an interdisciplinary approach. Scientific progress in advancing the understanding of environmental fate and behavior of AOCs is based ultimately on integration rather than separation of knowledge across scientific disciplines.

To achieve the goal of knowledge integration, this book entitled, Biophysico-Chemical Processes of Anthropogenic Organic Compounds in Environmental Systems, brings together world-renowned scientists on the subject matter across scientific disciplines to integrate the current state-of-the-art knowledge, especially the latest discoveries, development, and future prospects on the research of AOCs in the environment. By virtue of the complex nature of the interactions of AOCs with different environmental components and matrixes, no single available technique, instrument, or model is satisfactory yet for determining their fate, transport, availability, and risk in the environment. In order to fully understand the biological, physical, and chemical interactions and processes of AOCs in the environment, it is critical to know the chemical, physical, and biological properties of AOCs and their analytical techniques. This book is unique because of its multidisciplinary approach, which provides a comprehensive and integrated coverage of biological, physical, and chemical reactions and processes of AOCs in various environments, associated sampling and analytical techniques, and restoration of natural environments contaminated by AOCs.

There are 22 chapters in this book, and these chapters are divided into four parts. Part I contains four chapters focusing on the fundamental biological, physical, and chemical processes of AOCs in the environment; Part II, with seven chapters, presents the occurrence and distribution of AOCs in air, water, and soil, and their global cycling; Part III, containing six chapters, discusses the state-of-the-art sampling methods and current analytical, biological, spectroscopic, and microscopic techniques for monitoring and studying AOCs; and Part IV consists of five chapters emphasizing the restoration of natural environments contaminated by organic pollutants.

This book is an informative and important reference book for scientists, engineers, and professionals who are interested in the biological, physical, and chemical processes and interactions of AOCs in environmental systems. This book is also a critical addition to the existing literature on the subject matter. Further, this book can be used by undergraduate and graduate students, instructors and professors in the disciplines of environmental science and engineering; aquatic, soil, marine, and atmospheric sciences; geosciences; and ecological, biological, and chemical sciences. Again, the book chapter authors are leading authorities in their respective fields of research. Each chapter was rigorously reviewed externally as for refereed journal articles.

We sincerely thank all chapter authors and reviewers who graciously volunteered their time and effort, and contributed their knowledge and wisdom to improve the quality and clarity of this book. We are also highly grateful to the staff of IUPAC and John Wiley & Sons for their strong support and great cooperation in the publication of the book.

Baoshan Xing

Nicola Senesi

Pan Ming Huang

Contributors

Dr. Hassan Y. Aboul-Enein, Pharmaceutical and Medicinal Chemistry Department, Pharmaceutical and Drug Industries Research Division, National Research Center, Cairo, Egypt

Dr. Imran Ali, Department of Chemistry, Jamia Millia Islamia (Central University), New Delhi, India

Dr. Hans Peter H. Arp, Norwegian Geotechnical Institute, Department of Environmental Engineering, Oslo, Norway

Dr. Damià Barceló, Department of Environmental Chemistry, IDAEA-CSIC c/Jordi Girona, Barcelona, Spain

Dr. Mark J. Benotti, Applied Research and Development Center, Southern Nevada Water Authority, River Mountain Water Treatment Facility, Las Vegas, Nevada

Dr. Stephen A. Boyd, Department of Crop and Soil Sciences, Michigan State University, East Lansing, Michigan

Dr. Marisa Bueno-Montes, Institute of Natural Resources and Agrobiology of Seville—CSIC, Seville, Spain

Dr. Chuncheng Chen, Beijing National Laboratory for Molecular Sciences, Key Laboratory of Photochemistry, Institute of Chemistry, Chinese Academy of Sciences, Beijing, China

Dr. Robert L. Cook, Department of Chemistry, Louisiana State University, Baton Rouge, Louisiana

Dr. Songyan Du, Department of Environmental Sciences, Rutgers, the State University, New Brunswick, New Jersey

Dr. Marinella Farré, Department of Environmental Chemistry, IDAEA-CSIC c/Jordi Girona, Barcelona, Spain

Dr. Jay Gan, Department of Environmental Sciences, University of California, Riverside, California

Dr. Daniel W. Gerrity, Applied Research and Development Center, Southern Nevada Water Authority, River Mountain Water Treatment Facility, Las Vegas, Nevada

Dr. Kai-Uwe Goss, Department Analytical Environmental Chemistry, Helmholtz-Center for Environmental Research—UFZ, Leipzig, Germany

Dr. Peter Grathwohl, Center for Applied Geosciences, University of Tübingen, Tübingen, Germany

Dr. Adam P. Hitchcock, Brockhouse Institute for Materials Research, McMaster University, Hamilton, Ontario, Canada

Dr. Pan Ming Huang (deceased), Department of Soil Science, University of Saskatchewan, Saskatoon, Saskatchewan, Canada

Dr. Wesley H. Hunter, Department of Environmental Sciences, University of California, Riverside, California

Dr. Hongwei Ji, Beijing National Laboratory for Molecular Sciences, Key Laboratory of Photochemistry, Institute of Chemistry, Chinese Academy of Sciences, Beijing, China

Dr. Cliff T. Johnston, Department of Agronomy, Purdue University, West Lafayette, Indiana

Dr. Kevin C. Jones, Lancaster Environment Centre, Lancaster University, Lancaster, United Kingdom

Lina Kantiani, Department of Environmental Chemistry, IDAEA-CSIC c/Jordi Girona, Barcelona, Spain

Dr. Hans-Peter E. Kohler, Department of Environmental Microbiology, Eawag, Swiss Federal Institute of Aquatic Science and Technology, Dübendorf, ZH, Switzerland

Dr. Rai S. Kookana, CSIRO Land and Water, Glen Osmond, Australia

Dr. Klaus Kummerer, Department of Health Sciences, University Medical Center, Freiburg, Germany

Dr. David A. Laird, National Laboratory for Agriculture and the Environment, USDA-ARS, Ames, Iowa

Dr. John R. Lawrence, Environment Canada, National Hydrology Research Centre, Saskatoon, Saskatchewan, Canada

Dr. Bengang Li, College of Urban and Environmental Sciences, Peking University, Beijing, China

Dr. Hui Li, Department of Crop and Soil Sciences, Michigan State University, East Lansing, Michigan

Dr. Wanhong Ma, Beijing National Laboratory for Molecular Sciences, Key Laboratory of Photochemistry, Institute of Chemistry, Chinese Academy of Sciences, Beijing, China

Richard E. Meggo, Department of Civil and Environmental Engineering, University of Iowa, Iowa City, Iowa

Dr. Claudia Moeckel, Lancaster Environment Centre, Lancaster University, Lancaster, United Kingdom

Dr. Lee A. Newman, Biology Department, Brookhaven National Laboratory, Upton, New York

Dr. Jose-Luis Niqui-Arroyo, Institute of Natural Resources and Agrobiology of Seville—CSIC, Seville, Spain

Dr. José-JulioOrtega-Calvo, Institute of Natural Resources and Agrobiology of Seville—CSIC, Seville, Spain

Dr. Bo Pan, College of Environmental Science and Engineering, Kunming University of Science and Technology, Kunming, China

Dr. Janusz Pawliszyn, Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada

Dr. Sandra Pérez, Department of Environmental Chemistry, IDAEA-CSIC c/Jordi Girona, Barcelona, Spain

Dr. Joseph. J. Pignatello, Department of Soil and Water, Connecticut Agricultural Experiment Station, New Haven, Connecticut

Dr. David A. Reckhow, Department of Civil and Environmental Engineering, University of Massachusetts, Amherst, Massachusetts

Sanja Risticevic, Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada

Dr. Lisa A. Rodenburg, Department of Environmental Sciences, Rutgers, the State University of New Jersey, New Brunswick, New Jersey

Dr. Jerald L. Schnoor, Department of Civil and Environmental Engineering, University of Iowa, Iowa City, Iowa

Dr. Nicola Senesi, Department of Biology and Environmental and Agroforestal Chemistry, University of Bari, Bari, Italy

O. Samuel Sojinu, Department of Chemical Sciences, Redeemer's University, Mowe, Ogun State, Nigeria

Dr. Myrna J. Simpson, Department of Chemistry, University of Toronto, Toronto, Ontario, Canada

Dr. Andre J. Simpson, Department of Chemistry, University of Toronto, Toronto, Ontario, Canada

Dr. Shane A. Snyder, Applied Research and Development Center, Southern Nevada Water Authority, River Mountain Water Treatment Facility, Las Vegas, Nevada

Dr. Shu Tao, College of Urban and Environmental Sciences, Peking University, Beijing, China

Dr. Brian J. Teppen, DepartmentofCropand Soil Sciences, Michigan State University, East Lansing, Michigan

Dr. Dajana Vuckovic, Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada

Dr. Ji-Zhong Wang, School of Earth and Space Sciences, University of Science and Technology of China, Hefei, Anhui Province, China

Dr. Zhaohui Wang, Beijing National Laboratory for Molecular Sciences, Key Laboratory of Photochemistry, Institute of Chemistry, Chinese Academy of Sciences, Beijing, China

Dr. Jason C. White, Department of Analytical Chemistry, Connecticut Agricultural Experiment Station, New Haven, Connecticut

Dr. Baoshan Xing, Department of Plant, Soil, and Insect Sciences, University of Massachusetts, Amherst, Massachusetts

Huishi Yuan, College of Urban and Environmental Sciences, Peking University, Beijing, China

Dr. Eddy Yongping Zeng, State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy Science, Guangzhou, China

Yanxu X. Zhang, College of Urban and Environmental Sciences, Peking University, Beijing, China

Dr. Jincai Zhao, Beijing National Laboratory for Molecular Sciences, Key Laboratory of Photochemistry, Institute of Chemistry, Chinese Academy of Sciences, Beijing, China

Part I

Fundamental Biophysico-Chemical Processes of Anthropogenic Organic Compounds in the Environment

Chapter 1

Interactions of Anthropogenic Organic Chemicals with Natural Organic Matter and Black Carbon in Environmental Particles

Joseph. J. Pignatello

1.1. Introduction

Sorption refers to the process of molecular exchange between molecules in a gas or liquid phase and a solid phase (the sorbent). Sorption to natural solids typically plays a major, fundamental role in a compound's transport, reactivity, and bioavailability in the environment. The topic of sorption to natural solids is huge and extends over many decades. This chapter emphasizes sorption to organic substances in soils and sediments, including natural organic matter (NOM) and black carbon (BC) materials. These organic substances usually dominate the sorption of nonionic compounds except when present in very low levels or when water is scarce. The chapter does not cover sorption to anthropogenic organic waste liquids or semiliquids, such as weathered oil residues (Jonker and Barendregt 2006) and coal tar (Bayard et al. 2000; Khalil et al. 2006), which can play a significant role as sorbent when present in soil or sediment at high levels. The chapter emphasizes sorption from the aqueous phase, since sorption at the relative humidities existing in most environments approaches that under water-saturated conditions.

The chapter is not meant to be an exhaustive review, but a personal perspective of the author that emphasizes the author's own work and the more recent literature. It considers sorption separately from the perspective of the sorbate and the sorbent. Although this distinction is artificial (obviously, they form a complex), it is useful here as an organizational aid.

1.1.1. Nature of the Sorbents

Natural organic matter (NOM) is defined as the organic substances remaining after advanced decomposition of biomass below temperatures where pyrolysis becomes important. It includes humic substances (humic and fulvic acids, humin) and geologically older organic matter present in kerogen and coal. Biomass sources of NOM derive mainly from photosynthetic processes of plants and algae and secondary processes of fungi and heterotrophic bacteria. Natural organic matter can exist in a variety of states: dissolved molecules or molecular aggregates, colloidal particles, surface patches or coatings on minerals, intimate complexes with clay-size minerals, and discrete particles. At the primary level NOM is a heterogeneous mixture of functional units within charged, polydisperse molecules that include nonpolar alkyl, carbohydrate-like, protein-like, lignin-like, heterocyclic, and polyaromatic moieties (Hayes and Clapp 2001; Schulten and Schnitzer 1997). This author hesitates to illustrate a hypothetical NOM structure for the reader, since consensus does not exist for one. Most of our knowledge about the structure and composition of NOM has been gained from studies on dissolved NOM (DNOM) from natural waters or soil extracts. The distinction between dissolved and colloidal is arbitrary but operationally it is defined by many researchers as the filtration membrane cutoff of 0.45 µm. Most DNOM molecules do not exist individually, but rather are self-associated by hydrogen-bonding and other weak forces in aggregates. Molecules of NOM complex strongly with trivalent metal cations (e.g., Fe and Al) and less strongly with divalent cations and other inorganic ions. The size of individual NOM molecules depends on the source, and may range from a few hundred to 10⁵ daltons (Da) or larger. Most researchers believe that solid NOM can be characterized as a three-dimensional phase, consisting primarily of macromolecules with a fraction of molecules below 10³ Da. The reader is referred to various treatises and reviews on humic substances (Aiken et al. 1985; Greenland and Hayes 1981; Hayes et al. 1989; Leenheer 2009; Schnitzer and Khan 1978; Stevenson 1994; Sutton and Sposito 2005). Valuable background on geologically aged NOM is given by Allen-King et al. (2002). Solid NOM may be characterized as a random-network macromolecular organic solid. The closest analogy is lignin, the random–network polymer in plants derived from phenylpropane units substituted randomly with methoxyl, phenolic, carbonyl, and quinoid groups and giving plants their woody character (Glasser and Kelley 1987).

Black carbon is the carbonaceous byproduct of the incomplete combustion of biomass or fuels (Goldberg 1985). It includes char, in which the original material is carbonized in the solid state, and soot, which is formed by the condensation of precursors from the gas phase. Black carbon is a small but important component of total organic carbon in undisturbed soils and sediments, but can be more prevalent in deep-sea and marine shelf sediments, in areas of frequent or recent fires, and in areas of high industrial activity (Masiello and Druffel 1998; Skjemstad et al. 1999). Soot is a component of atmospheric aerosols (Schmidt and Noack 2000). Black Carbon plays important roles in various geo- and bio-geochemical processes, the global carbon cycle, carbon sequestration, radiative heat balance of the planet, and pulmonary toxicity of aerosols (Lighty et al. 2000; Ramanathan and Carmichael 2008). The potential importance of BC as an adsorbent of environmental pollutants was first noted in the pioneering work of Gustafsson and Gschwend (Gustafsson and Gschwend 1997; Gustafsson et al. 1997).

Black carbon is not a single material, but a continuum of materials whose properties depend on source stock and formation conditions. The BC body is composed of single and short stacks of polyaromatic (graphene) platelets rimmed with ketone, ether, hydroxyl, quinoid, carboxyl, and other functional groups. The BC platelet—averaging 7–17 fused rings for biomass chars depending on formation conditions (Brewer et al. 2009; Kaneko et al. 1991)—is considered well represented by the terraced (0001) basal surface of graphite (Donnet et al. 1993). Thus, the BC surface may be considered micrographitic in character. The stacks are arranged in a highly disordered fashion creating a pore network (Boehm 1964; Goldberg 1985; Palotás et al. 1996). The platelets may contain five- and seven-membered rings that introduce curvature, contributing to the disorder (Harris and Tsang 1997; Shibuya et al. 1999). Raw BC is typically highly porous and thus highly surface-active. The pores lie mainly in the micropore (<2 nm) and mesopore (2–50 nm) size ranges (IUPAC definitions). The high surface area and presence of abundant micro- and mesoporosity render raw BC a stronger sorbent of organic compounds than NOM, especially at low concentration. The pore structure of BCs has sometimes (Boehm 1964; Xia and Ball 1999) been likened to that of activated carbons (ACs), which are produced under controlled conditions of temperature and gas composition. However, some of the porosity of BC may be inaccessible because of poor connectivity, plugging by uncombusted residues, or plugging by natural substances in the environment.

In most places the chapter considers the sorbent behavior of NOM and BC separately, although a certain degree of overlap in their natures is to be recognized.

1.1.2. The Sorption Process in General

Sorption may be categorized by the type of intermolecular forces that characterize the sorbate-immobile phase interaction: chemisorption, physisorption, and ion exchange. Chemisorption is typified by strong bonds (e.g., covalent and ligand–metal coordination bonds) in which orbital mixing predominates. It is less frequently encountered in environmental systems than are physisorption and ion exchange. Unless the covalent bond breaks reversibly—in most relevant reactions it typically does not—chemisorption will lead to loss of identity of the starting molecule. Physisorption and ion exchange are characterized by the weak intermolecular forces discussed later, and the identity of the starting molecule is preserved on desorption. This chapter does not deal with chemisorption.

Physisorption may be further categorized by how the sorbate mixes with the solid, whether by adsorption or absorption. Adsorption is the association of a molecule at the interface between a fluid and the surface of a solid whose atomic/molecular lattice cannot be penetrated by sorbate molecules. Absorption (sometimes called partitioning) is the intermingling (dissolution) of a molecule within the atomic/molecular lattice of a solid. The only solids allowing lattice penetration in this way commonly found in the environment are composed of organic matter. While the terms adsorption and absorption have utility, it is more at the phenomenological level than the mechanistic. Adsorption to a lattice-impenetrable solid may encompass several qualitatively different processes: (1) the resting of molecules on discrete surface sites, (2) the partitioning of molecules into an ordered microscopic hydration phase near the surface, (3) the condensation of the molecules into a liquid-like state in small pores, and (4) the layering of molecules on the surfaces of water films that coat particles. Processes (2) and (3) can be regarded mechanistically as more akin to dissolution than to association at a discrete surface. Likewise, absorption in natural organic matter may involve residence of some molecules within closed pores and/or at specific molecule-scale sites within the phase; it is not clear whether a distinction between adsorption and absorption is meaningful in such cases, at least at the molecular scale.

Sorption of a chemical species x from the aqueous phase (xw) can be written as a series of steps in a thermodynamic cycle (Fig. 1.1):

(1.1)

equation

Step 1 is the transfer of x from water to the gas phase, encompassing desolvation of x and collapse of the preexisting solvation shell to the bulk water structure. Step 2 is the reorganization of the sorbent matrix needed to accommodate x, such as the formation of a cavity to fit x, or displacement of existing water molecules or other organic molecules or ions at the site. Step 3 establishes interactions of x with the now-prepared sorbent. Sorption from the vapor phase is the same except without step 1. The Gibbs free energy of sorption ΔGsorp is the net change in free energy summed over all the steps. The ΔGsorp is related to the dimensionless thermodynamic sorption coefficient Ksorp by

(1.2) equation

where R is the universal gas constant, T is temperature [in Kelvins], a is the activity in the solid or fluid state, and is the corresponding reference state activity. The value of will depend on the choice of the reference state for the sorbed phase. The relationship between Ksorp and a conventional sorption coefficient defined by one of the isotherm models (see Section 1.3.1) will likewise depend on the choice of the sorbed phase reference state.

Figure 1.1. Sorption as a series of steps in a thermodynamic cycle: (1) transfer from water to gas phase; (2) preparation of the (water-wet) sorbent; (3) transfer from the gas phase to the prepared site.

Note that if the sites are heterogeneous in energy, and possibly will be dependent on concentration; hence, Ksorp will be concentration-dependent. However, sorption always tends toward linearity (i.e., a constant ratio of sorbed to fluid-phase concentrations) as the concentration in the fluid phase tends toward zero. In the environment many organic contaminants can be present at concentrations in the fluid phase ranging from vanishingly small to the limit of solubility or vapor pressure.

1.1.3. Overview of Weak Intermolecular Forces in Physisorption

Noncovalent intermolecular attractive forces are listed in Table 1.1. For physisorbing compounds these so-called weak forces act simultaneously and additively in combinations appropriate to the structures of the interacting species.

Table 1.1 Weak Intermolecular Attractive Forces.

1.1.3.1. London–van der Waals and Coulombic Forces

The forces between a nonionic molecule and an uncharged site may include: dipole–dipole, the interaction between permanent dipoles (Type 1 in Table 1.1); dipole–induced dipole, the attraction of a permanent dipole with the dipole that it induces in its neighbor; (Type 2 in Table 1.1); and induced dipole-induced dipole (Type 3 in Table 1.1), the mutual attraction of momentary dipoles produced by the synchronization of electronic motion in interacting neighbors. These are commonly called dipolar, induction, and dispersion forces, respectively. They are often referred to collectively as London–van der Waals, or simply van der Waals (vdW) forces. The strength of individual vdW forces depends on the separation distance to the inverse sixth power, how the molecules are oriented, and applicable molecular properties of the interacting species such as dipole moment, ionization potential, and polarizability (Table 1.1). Except for very small molecules the most important of the vdW interactions is thought to be dispersion (Hunter 2004).

Placing a charge on either the sorbate or the site leads to charge–dipole (Type 4 in Table 1.1) and charge–induced dipole (Type 5 in Table 1.1) interactions, both proportional to the charge and the sixth power of the separation distance. When the sorbate and site are fully and oppositely charged, a Coulombic force (Type 6 in Table 1.1) exists between the charges that is proportional to the first power of the separation distance. While the Coulombic force usually dominates the energy of interaction between ions, it is important to realize that since ions are not ideal point charges, vdW forces contribute significantly. This is true even for simple inorganic ions, as witnessed in chromatographic separations (Fritz 2005). Moreover, the uncharged regions of the organic ion will undergo the other weak interactions permitted by structure and orientation.

1.1.3.2. Hydrogen (H-) Bonding

Hydrogen bonding (type 7 in Table 1.1) is the interaction between an acidic proton of a donor and the lone pair electrons of an acceptor (AH :B). It is included among the noncovalent forces; however, whereas weak H bonds are due mainly to dipolar interactions, the covalent nature of the H bond increases with its strength, such that very strong H bonds are essentially three-center, four-electron covalent bonds (Gilli and Gilli 2000). The most important type of H bond in environmental systems occurs when the atoms A and B terminate in O, N, or S. These ONS H bonds are highly oriented (A–H–B angle, 180° ± 15°, except in intramolecular H bonds), and in most relevant environmental systems range in enthalpy from 6 to 70 kJ/mol.

Gilli et al. (2009) categorize ONS H bonds into ordinary (Ohb), charge-assisted (+/−CAhb, +CAhb, and −CAhb), and resonance-assisted (RAhb) leitmotifs:

(1.3) equation

(1.4) equation

(1.5) equation

(1.6) equation

(1.7)

equation

The Ohb and +/−CAhb leitmotifs differ only in the degree of proton transfer between A and B. Formation of an RAhb takes place when H bonding is accompanied by π–conjugated bond delocalization, such as in the carboxylic acid dimer:

The leitmotifs +CAhb, −CAhb, or RAhb are the strongest H bonds. The strength of the H-bond increases as the difference in pKa values of AH and BH+ (ΔpKa = pKa,AH – pKa,BH+) approaches zero (Gilli et al. 2009). Thus, water (ΔpKa = 14) forms only weak H-bonds with itself [8 kJ/mol (Silverstein et al. 2000)] and with most ONS functional groups. Water forms stronger H bonds with carboxylic acids, phenols having strongly electron-withdrawing groups, and protonated nitrogen groups. Table 1.2 groups H bonds in terms of strength between organic functional groups relevant to the structures of NOM and many common contaminants. Particularly strong H bonds exist in carboxylic acid dimers, in amide dimers, in conjugate pairs (RCO2H/RCO2−, ArOH/ArO−), between RCO2H or RCO2− and certain nitrogen or protonated nitrogen compounds, and between phenols and certain nitrogen compounds (Table 1.2).

Table 1.2 Estimated Strengths of Intermolecular H-Bonded Pairs for Some Relevant Functional Group Pairs.

Source: Gilli et al. (2009).

Hydrogen bonds are also possible between ONS donors and the π bond in alkenes or aromatic rings (the so-called π–H bond); between ONS acceptors and acidic C–H groups, and between acidic ONS protons and aliphatic halogen atoms. As these H bonds are, with some exceptions, much weaker than ONS-type H bonds, they may not be able to compete with water when water is abundant.

1.1.3.3. Interactions between π-Conjugated Systems

It has been realized for a long time that special weak interactions may occur between arene units. Arene is defined as aromatic and related cyclic π conjugated systems. Arene–arene interactions are believed to play a role in many chemical and biological phenomena (Hunter et al. 2001; Hunter and Sanders 1990; Janiak 2000; Meyer et al. 2003; Schmidt-Mende et al. 2001). It now appears likely that they play a role in environmental partitioning as well.

The nature of arene–arene interactions is not fully understood. The strongest interactions occur when one unit is π-electron rich (donor, ) and the other is π-electron poor (acceptor, ). These interactions are known as π-stacking or π–π electron donor–acceptor (π–π EDA) interactions. The units undergoing π–π EDA interaction typically associate in a sandwich, but more often in a parallel displaced fashion (Table 1.1). According to current understanding (Cockroft et al. 2007; Gung and Amicangelo 2006; Hunter et al. 2001; Hunter and Sanders 1990; Sinnokrot and Sherrill 2006), the π–π EDA complex is a hybrid structure that can be written

(1.8)

equation

representing quadrupolar, vdW, and charge transfer interactions between ring systems.

In solution, is believed to dominate the energy. In general, π systems have a quadrupole vector perpendicular to the plane of the nuclei, where the π cloud on either side of the plane is polarized oppositely to the σ-bond framework (type 8 in Table 1.1). Complexation may result if interactions between quadrapoles is net attractive. A classic example is the 1 : 1 complex between benzene and hexafluorobenzene (Williams 1993). The monomers have quadrapole moments of opposite sign and magnitude (−33.3 and 31.7 C/m², respectively (Vrbancich and Ritchie 1980). Benzene is electron rich in the π cloud and electron poor in the σ bond system, while hexafluorobenzene is polarized in the opposite sense. Thus, attractive forces occur between the π clouds and to a lesser extent between the σ systems. The 1 : 1 C6H6: C6F6 complex forms a solid that crystallizes in a parallel displaced arrangement and melts 20 K higher than either monomer (Williams 1993). While neither monomer has a permanent dipole moment, the 1 : 1 gas–phase complex has a dipole moment of 0.44 Debye, about one-eight of a charge (Steed et al. 1979). All VdW forces between rings and their substituents are believed to be minor because they largely cancel out with those previously existing between momomers and displaced solvent molecules (Cockroft et al. 2007; Hunter 2004). (However, they obviously would play an important role in gas-solid adsorption.) The CT structure is the dative bond formed by one electron transfer from the highest occupied molecular orbital of to the lowest unoccupied molecular orbital of . It contributes primarily to the excited state, often yielding an absorbance band in the near UV–visible region, but may contribute appreciably to the ground state energy in strong complexes (Gung and Amicangelo 2006).

The π–π EDA bond enthalpy approaches that of a strong H bond (Foster 1969; Perry et al. 2007). It scales with the respective donor and acceptor ability of the opposing units determined by substituents that donate or withdraw electron density, especially through the π system. The flow of electron density is illustrated below:

Multiple alkyl and/or –NR2 groups confer strong donor character. Donor strength also trends with π cloud polarizability; thus, polyaromatic hydrocarbons (PAHs) are π donors whose strength appears to increase with fused ring number, at least up to 4. Especially strong electron withdrawing groups include –NO2, –NR3+, –CF3, –C(=O)R, –SO2R, and –CN. Halogen has opposing σ-withdrawing and π-donating effects; yet a ring may become net π-accepting as the number of halogens exceeds a critical number, as occurs with F. Especially favorable complexation occurs when the acceptor is a charged heteroaromatic ring. In this case stability is aided by a charge–quadrupole (π−cation) interaction (Type 9, Table 1.1) (Qu et al. 2008). An example is the complex in water between phenanthrene and o-phenanthroline mono- or dication (Wijnja et al. 2004), leading to a dramatic increase in the apparent solubility of phenanthrene.

The π–π EDA complexes form readily in water and in both polar and apolar organic solvents (Breault et al. 1998; Ferguson and Diederich 1986; Foster 1969), although the relation with solvent polarity is complex (Breault et al. 1998).

Like-polarized arenes ( ) associate much more weakly than do oppositely polarized arenes, and their complexes tend to orient T-shaped in order to maximize π–σ and minimize π–π interactions. Thus, benzene and small PAHs orient roughly T–shaped in their crystals (Newcomb 1994), and benzene orients T-shaped in its most stable gas–phase complex (Iimori et al. 2002; Morimoto et al. 2007).

1.2. Sorption from the Perspective of the Sorbate

1.2.1. Thermodynamic Driving Forces in Sorption

The contribution of each force listed in Table 1.1 to the net driving force for sorption from aqueous solution depends on the difference in free energy of the interaction with water compared to that with the sorbent. Teasing apart the contributions of individual noncovalent forces in sorption has been a challenge. The following discusses the state of our knowledge about the contributions of individual noncovalent forces to the sorption free energy. Such a discussion must consider both positive and negative driving forces; examples of the latter are solvation and steric hindrance to sorption.

1.2.1.1. Desolvation and the Hydrophobic Effect

A major driving force for sorption from the aqueous phase of nonionic compounds, whether to minerals or organic matter, is the exclusion of the organic solute from water known as the hydrophobic effect. This is supported by a vast and relentless stream of data. The term hydrophobic effect refers to the forces that limit the solubility of apolar molecules—or parts thereof—in water. It is also terminology that refers to the clustering of hydrophobic units such as surfactants (surface-active agents) into micelles, the folding of biological molecules, and the behavior of solutes and water near apolar surfaces. Because of the central role of the hydrophobic effect in chemistry and biology, many studies have been devoted to a resolution of its underlying nature. For reviews, see Chandler (2005), Lazaridis (2001); and Southall et al. (2002).

While concordance has not yet been reached on all its aspects, researchers are in general agreement that the hydrophobic effect arises from disruption of the cohesive energy of water, not from any special attraction between hydrophobic molecules nor of any special repulsion between apolar entities and water molecules. The disruption originates in the greater ordering, and fewer—if not stronger—water–water H bonds within the hydration shell surrounding the hydrophobic entity than in bulk water itself, a situation that raises an entropy and/or enthalpy penalty for dissolution in water. It should be noted that hydrophobicity may have a different connotation when speaking of the distribution of an organic molecule among qualitatively different microdomains within a solid (i.e., NOM phases, narrow pores, and near surfaces) because water exists in a different organizational state in those environments than it does in bulk water.

The following are the salient features of the hydrophobic effect according to contemporary understanding.

There is no evidence for long-range forces between hydrophobic entities. The solvation free energy for alkanes in water is pairwise-additive (Wu and Prausnitz 2008); that is, the Henry law constant, which is proportional to the free energy of solvation, is linearly related to the number of C–C bonds. This means that the hydrophobic effect is primarily a local phenomenon limited to only a few angstroms (Chandler 2005; Wu and Prausnitz 2008).

Aggregates of nonpolar molecules less than 1 nm in radius are unstable in water (Chandler 2005; Southall et al. 2002). This means that nonpolar entities such as alkanes (Wu and Prausnitz 2008) and PAHs (Wijnja et al. 2004) have no tendency to self-associate in water much below their solubility limit. This and the previous point are consistent with a solvent-centered, rather than solute-centered origin, of the hydrophobic effect. Since there appears to be no special driving force for association of hydrophobic entities, terms such as hydrophobic bonding and hydrophobic interactions are misleading.

The hydrophobic effect is sometimes described as being entropy-driven. In fact, both entropy and enthalpy play a role, depending on the solute and temperature (Southall et al. 2002). The dissolution of hydrocarbons is unfavorable in cold water because of entropy, and in hot water because of enthalpy. Smaller solutes tend to force ordering of neighboring water molecules, resulting in an entropy penalty for dissolution, whereas larger solutes tend to be more effective at breaking water–water H bonds, resulting in an enthalpy penalty.

For water at the interface of an organic liquid or an extended hydrophobic surface, each water molecule participates in about one fewer H bonds than in bulk water (Chandler 2005). However, the H bonds near a hydrophobic surface appear to be stronger; the enthalpy of the H bond of water in the first hydration shell of argon in water is 2.1 kJ/mol greater than in bulk water (Silverstein et al. 2000).

Some OH groups of water at the water–liquid interface of hydrocarbons or chlorinated hydrocarbons orient toward the organic phase in the fashion, –O–H organic. That this interaction is attractive is indicated by the shift to lower energy in the stretching frequency of such OH groups relative to that of the O–H protruding into air at the water–water vapor interface (Moore and Richmond 2008). This is consistent with a generally attractive interaction between water and hydrophobic surfaces.

Due to vdW interactions of water with the hydrophobic entity, the interfaces between water and organic liquids or between water and hydrophobic surfaces are sharp, with no evidence for a vacuum-like gap (Chandler 2005; Moore and Richmond 2008).

While the hydration shell of apolar entities may be more ordered than bulk water, both the iceberg structure originally proposed by (Frank and Evans 1945) and the crystalline water cage structure that forms around the clathrate hydrates of noble gases and some small hydrocarbons at suitable temperature and pressure (Bontha and Kaplan 1999; Jeffrey 1984) are probably exaggerations of its true nature.

Any given functional unit of a molecule introduces opposing forces for sorption. On one hand, its size and polarizability tend to disrupt the structure of bulk water and drive the molecule to the solid; on the other hand, its permanent polarity tends to drive the molecule to the aqueous phase. This is best exemplified by the halogens.

Halogen (except F) is large and polarizable, contributing to molecular hydrophobicity. By contrast, halogen polarizes the C–X bond because of its electronegativity. We know that bond polarization plays a role because hydrophobicity (as octanol–water partition coefficient, Kow) increases with halogen substitution more so for sp²- than sp³-substituted carbon (alkane < alkene < aromatic ring). The same is true for most functional groups bonded to carbon through O, N, or S. This is so for two main reasons. (1), sp² carbon is more electronegative and competes better for the σ- electrons than does sp³ carbon; and (2), in sp² systems, delocalization of a halogen electron pair into the π system counteracts the bond dipole moment.

The question arises about the importance of dispersion as a driving force for sorption. Many will point to the general relationship between molecular size of a series of apolar compounds (or apolar functional groups) and sorption intensity as support for the importance of dispersion interactions with the sorbent. However, it is believed that dispersion forces between a hydrocarbon entity and its water solvation shell are similar in magnitude to those between the entity and its hydrocarbonaceous neighbors when dissolved in a hydrocarbon solvent (Israelachvili 1992). It follows that, in the process of exchange between the aqueous phase and the sorbent, dispersion interactions roughly cancel out, and dispersion per se is not a major driving force in sorption, but rather the effect of molecular size is manifested in the hydrophobic effect. By contrast with solution–solid sorption, gas–solid sorption is highly driven by dispersion, since intermolecular forces in the gas phase are negligible.

1.2.1.2. Hydrogen Bonding and Dipolar Interactions

Hydrogen bonding favors sorption only to the extent that the net free energy of H bonding with the sorbent is greater than that with water. Hydrogen bonding opportunities are greater in water because of the sheer abundance of H-bond donor and acceptor groups—111 and 55.5 mol/L, respectively. On the other hand, NOM and BC both are rich in carboxylate and phenolate groups, that tend to form stronger H bonds than water, especially with solutes that have phenolic, aromatic amine, amide, and carboxylate groups (Table 1.2).

Given the abundance of H-bonding groups in NOM, H bonding of contaminants with H-bonding substituents almost certainly takes place in this phase. Nevertheless, direct evidence for this has been elusive. Dixon et al. (1999) observed a progressive downfield shift in the NMR signal of 4-fluoroacetophenone with increasing fulvic acid concentration in methanol–water, but whether this shift was due to H bonding is questionable since it was not reversed by addition of a six fold excess of acetophenone, which arguably should have outcompeted the fluoro derivative for H-bond acceptor sites interacting with the ketone O. Welhouse and Bleam (1993a, 1993b) showed by NMR that atrazine dimerizes in CCl4 and also forms H bonds (as donor or acceptor) of moderate strength with mono functional molecules. With bifunctional molecules such as carboxylic acids and amides, atrazine forms strongly H-bonded cyclic complexes by simultaneously accepting and donating a H bond, as in the structure below:

Working with the trifluoroethyl derivative of atrazine in 10% humic acid at pH 11.8, Chien and Bleam (1997) inferred H bonding of atrazine to DNOM by the apparent absence of NMR resonances characteristic of atrazine dimers. The contribution of H bonding to atrazine sorption, however, remains unclear.

In order to gain insight into the driving forces for sorption to NOM, Borisover and Graber (2003) transformed regular soil–water isotherms to soil–hexadecane isotherms (Fig. 1.2) by converting aqueous concentration to the thermodynamically equivalent n-hexadecane concentration via Henry's constant ratios or solubility ratios. This amounts to switching the reference state from the pure liquid/subcooled liquid to the dilute solution in an inert solvent (n-hexadecane) capable only of dispersion. The resulting rebuilt isotherms (Fig. 1.2) therefore represent the difference in free energy of interactions in the water-wet solid and the inert solvent, and thus highlight interactions with NOM of a more polar nature. Compounds showing exceptionally strong sorption to a peat soil (45% OC) were those capable of forming strong H bonds with carboxyl and phenolic groups on NOM: 3-nitrophenol, phenol, pyridine, benzoic acid, benzyl alcohol, atrazine, and 2,4-dichlorophenol. Compounds of intermediate sorption strength included trichloromethane and those capable of serving as H-bond acceptors to form weaker H bonds: acetophenone, anisole, 2-chloronitrobenzene, and nitrobenzene. Compounds sorbing more weakly were apolar aliphatic and aromatic compounds. Interestingly, chlorinated aliphatic compounds sorbed more strongly than did chlorinated aromatic compounds. This suggests that dipolar interactions between the bond dipoles of chlorinated aliphatics, which are greater than those of chlorinated aromatics, engage in dipolar interactions with NOM to favor their transfer from the hydrocarbon phase.

Figure 1.2. Pahokee peat-n-hexadecane sorption isotherms (Borisover and Graber 2003). The concentration in hexadecane is calculated by the concentration in water times the ratio of the solubility in hexadecane to that in water.

1.2.1.3. π–π EDA Interactions

When permitted by structure, π–π EDA interactions can only add to the driving force of sorption, since water is incapable of π–π EDA interactions. Many pesticides, explosives, antibiotics and other classes of environmental contaminants contain arene groups with strong π–donor or π-acceptor character. In addition, certain natural compounds in soil can act as strong acceptors—there are many examples of quinones among plant signaling chemicals and allelochemicals, such as juglone (5-hydroxy-1,4-naphthoquinone). Heteroaromatic amines have been detected in surface and subsurface waters contaminated by shale oil or coal liquification wastes (Sims and O'Loughlin 1989). Humic substances are rich in π-acceptor units, including quinones, charged heterocyclic amines, and aromatic rings having multiple electron-withdrawing carbonyl groups. It has been postulated that the CT bands from internal quinone–hydroquinone complexation partially account for absorbance of humic substances in the visible region (Del Vecchio and Blough 2004). Humics may also contain strong π-donor units such as alkyl-substituted rings, polyaromatic rings, and pyrrole-type heterocyclic rings. The graphene platelets of black carbon, with their polyaromatic surfaces, may contain both electron-rich and electron-poor regions, depending on their size and the distribution of functional groups along their rims, which may attract π-acceptor and π-donor molecules, respectively (Zhu and Pignatello 2005a) (see Fig. 1.3).

Figure 1.3. (a) Free-energy relationship between excess adsorption on nonporous graphite and complexation with π donors, naphthalene, phenanthrene, or pyrene in chloroform measured by NMR for mononitrotoluene, 2,4-dinitrotoluene, and trinitrotoluene. The excess adsorption free energy is postulated to originate from π–π EDA interactions on the graphene surface. (bottom) Strong π acceptors (e.g., trinitrotoluene) interact with electron-rich regions of the surface, while strong π donors (e.g., phenanthrene) interact with electron-poor regions. Surface charge separation may be due to step/edge defects (graphite) or rim functionality [black carbon (BC)]. A hypothetical BC platelet with potential donor and acceptor regions is indicated.

The term π-π interactions has been used rather loosely in the more recent environmental literature. Speculation is the norm, and the nature of the force is frequently misunderstood. Nevertheless, evidence is emerging for a contribution of π–π EDA interaction to sorption of some compounds by NOM, as well as by BC.

Mixing of substituted pyridines and triazine herbicides with DNOM is reported to generate CT absorbance at 460 nm, suggestive of π–π EDA (Müller-Wegener 1987). Sorption of the π-donor compounds, pentamethylbenzene, naphthalene, and phenanthrene, to several soils increased with decreasing pH from 7 to 2.5 (Zhu et al. 2004), whereas no similar pH effect occurred for non-π-donor hydrophobic compounds. Other possible systematic affects varying with pH were ruled out. These π-donor solutes may interact with π-acceptor sites in NOM, such as aromatic rings with multiple carboxyl groups or charged aromatic and heteroaromatic amines, whose acceptor ability increases with degree of protonation (i.e., –CO2H and –NH+= are more electronegative than –CO2− and –N=, respectively). Pairwise complexation in methanol–water between donors, pentamethylbenzene, naphthalene or phenanthrene, and each of the following model NOM acceptors was identified spectroscopically: 1,3,5-benzenetricarboxylic acid, 1,4,5,8-naphthalenetetracarboxylic acid, and pyridine. Donor complexes with these model acceptors was pH-dependent and gave a CT band in the UV/visible spectrum and upfield NMR chemical shifts indicative of face-to-face association. On the basis of the free-energy relationship to be discussed in Section 1.2.2, π-π EDA interactions made a small contribution ( 5%–8%) to the free energy of phenanthrene sorption in a soil (Zhu and Pignatello 2005b).

Other, more indirect evidence for π–π EDA interactions of PAHs with NOM or DNOM exists. Polycyclic aromatic hydrocarbons (PAHs) consistently sorb more strongly to soils than do PCBs of comparable hydrophobicity (Kow) (Allen-King et al. 2002; Chiou et al. 1998; Cornelissen et al. 2004; van Noort 2003); PAHs are strong π-donors, while PCBs are not strong donors or acceptors. Isotherms of PAHs to DNOM are frequently nonlinear (Laor and Rebhun 2002; Polubesova et al. 2007), suggesting that specific interactions take place. In terms of the n-hexadecane reference state, aromatic hydrocarbons sorption to Pahokee peat (Borisover and Graber 2003) follows the order, phenanthrene ≥ naphthalene > benzene, the same order as their π-donor capablility.

A number of papers have postulated π–π charge transfer processes between pesticides (triazine, urea, and bipyridylium herbicides) and humic substances based on elevated free radical concentrations measured by ESR spectroscopy (e.g., Senesi et al. 1995; Sposito et al. 1996). Whatever the source of free radicals, their presence cannot be related to the reversible π–π EDA complex referred to here, as no electron transfer to produce free radicals takes place. Moreover, free radicals would likely react with O2 or couple with NOM, resulting in loss of solute identity.

Evidence also exists for π–π EDA interactions with elemental carbonaceous materials. Zhu and Pignatello (2005a) found that adsorption of nitroaromatics on nonporous microcrystalline graphite and on BC (wood charcoal) is far greater than predicted by the hydrophobic effect, based on a calibration set of compounds, and in accord with their π-acceptor strength (mono- < di- < trinitrotoluene). Hydrogen bonding of the nitro groups was ruled out. Adsorption of PAHs on the same solids was likewise greater than predicted by the hydrophobic effect and followed the π-donor strength (naphthalene < phenanthrene). Complexation between the PAH donors and the nitroaromatic acceptors was observed in chloroform. These complexes displayed upfield shifts of NMR spectral frequencies due to ring current effects, which is indicative of face-to-face association. They also gave CT bands in the visible region often seen with π–π EDA complexes. The association constant followed the order in expected strength of interaction; namely, mono- < di- < tri-nitrotoluene with a given PAH, and naphthalene < phenanthrene < pyrene with a given nitroaromatic. Figure 1.3 shows the strong relationship between the free energy of molecular complexation in chloroform solution and the excess free energy of nitroaromatic adsorption on graphite based on the hexadecane reference state. Chen et al. (2007b) confirmed π–π EDA interactions between polynitroaromatic compounds and the graphene-like surface of carbon nanotubes using a similar approach as Zhu and Pignatello (2005a).

Taken together, the results indicate that the graphene surface may be amphoteric with respect to π-interactive adsorbates; referring to Figure 1.3, electron rich regions of the surface attract strong π acceptors while electron-poor regions attract strong π donors. Polarization of the graphite surface near defects and edges is visible by scanning tunneling microscopy (McDermott and McCreery 1994). Figure 1.3 further presents a hypothetical structure of BC showing a potential donor region overlying the polarizable polyaromatic center of the sheet, and acceptor regions in the vicinity of π-acceptor moieties along the rims.

1.2.1.4. Steric Effects

Steric hindrance is possible both in partitioning and adsorption, but very little attention has been paid to it until relatively recently. Steric effects in partitioning to NOM have not been systematically investigated. Partitioning into a solid phase involves the opening of a cavity for the incoming molecule if one does not already exist. The free energy of partitioning from water must include a term for the difference in free energy of cavitation in the solid and in the aqueous phase.

The excess free energy required for opening a spherical cavity of radius r in a liquid is the sum of a volumetric term (4/3 πr³E), where E is the volumetric energy density, and a surface term (4πr²γ), where γ is surface tension (Hummer et al. 1998; Southall et al. 2002). The volumetric term dominates for small molecules, whereas the surface term dominates for very large molecules or extended surfaces. Cavitation of a solid phase strongly depends on its viscoelastic properties and free volume distribution. For a flexible solid in its equilibrium state, the free-energy cost for forming a cavity (cavity penalty) increases systematically with penetrant size for a homologous series of penetrants; atoms or very small penetrants may fit into existing free volume, thus requiring little or no cavitation, whereas complete exclusion from the internal phase would be reached at some very large size. Suppressed sorption on the basis of molecular size for large molecules has been reported to occur in phospholipid bilayers (Dulfer and Govers 1995; Gobas et al. 1988; Kwon et al. 2006; Yamamoto and Liljestrand 2004) ordinarily regarded as being a partition medium.

Steric effects are known to play a role in the adsorption of organic compounds to matrix-impenetrable solids (Kärger and Ruthven 1992), both porous and nonporous, and have been claimed for natural solids. There are at least two sources of steric effects, illustrated in Figure 1.4: deviation from molecular planarity size exclusion. (1) Since interactions depend on close molecular approach, deviation from molecular planarity should reduce contact area with a flat surface, other molecular properties being equal (2) Steric effects in microporous solids can be manifested by size exclusion at pore throats.

Figure 1.4. Illustration of the contact area and size exclusion hypotheses for steric effects in adsorption.

The contact area hypothesis for adsorption to a smooth surface has been tested for alkanes. As they are able to readily adopt a planar conformation, normal alkanes are expected to achieve closer contact with a surface than cycloalkanes, which are restricted by ring puckering. Closer contact favors vdW forces, whose magnitudes are inversely related to separation distance to the sixth power (Table 1.1). The difference in sorption coefficient between n- and c-alkanes, however, is small: the ratio Kn/Kc for gas-phase transfer to the surfaces of liquids and nonmicroporous inorganic solids is close to unity (0.83–1.62) and tends to decrease with the number of carbons (Endo et al. 2008b). However, the difference in interaction enthalpy may be greater than it would appear by the ratio Kn/Kc, since the entropy penalty for transfer to a surface or phase is greater for the n-alkane because it has more degrees of freedom in the gas phase than does the cycloalkane.

The contact area hypothesis has also been invoked to explain sorption trends for BC materials and sediments with high levels of BC. For example, in sediment systems polychlorinated biphenyl (PCB) congeners in which coplanarity of the rings is impaired due to ortho chlorine substitution have a lower sorption coefficient (Barring et al. 2002; Bucheli and Gustafsson 2001; Jonker and Smedes 2000; Jonker et al. 2004; van Noort et al. 2002), greater fast-desorbing fraction (van Noort et al. 2002), and higher bioavailability (Jonker et al. 2004) than do coplanar congeners with the same number of chlorines. However, noncoplanar congeners have an inherently weaker tendency to interact with themselves, as witnessed by their higher subcooled vapor pressures (Schwarzenbach et al. 2002). They also partition less favorably into octanol from water (Schwarzenbach et al. 2002) for reasons that are not clear, since close approach applies to both solvents. The coplanarity effect for PCB adsorption to soot and soot-like materials from water held after normalizing the sorption coefficient by Kow (Jonker and Koelmans 2002b).

Since BC materials are highly porous, an alternative explanation for the coplanarity effect is size exclusion (molecular sieving) in pores; a noncoplanar congener has a larger critical diameter and therefore would be more restricted in the passages that it could enter. It is well known that diffusion in zeolites becomes severely hindered as the minimum critical molecular diameter approaches the pore diameter (Kärger and Ruthven 1992). Anthracene and phenanthrene are both planar and have nearly identical KOW values (Schwarzenbach et al. 2002), yet anthracene consistently sorbs more strongly than does phenanthrene to BC materials (Jonker and Koelmans 2002b) and BC present in sediments (Cornelissen et al. 2004). Size exclusion may also explain the decline in Langmuir adsorption capacity of PAHs on activated carbon with increasing size (Walters and Luthy 1984), although no mechanism was offered in that study. Adsorption of n-hexane in some activated carbons is much greater than is cyclohexane, said to reflect size exclusion (Endo et al. 2008b).

Systematic evidence has now been presented for size exclusion in charcoal BC for a series of planar aromatic compounds, both polar and apolar (Pignatello et al. 2006a; Zhu and Pignatello 2005a). For this sample of BC about 80% of the porosity exists in pores up to 2 nm wide. To normalize for hydrophobic effects, nonporous graphite was used as the reference state; BC–graphite isotherms were constructed from the experimental BC–water and graphite–water isotherms:

(1.9) equation

In this reaction, K is the distribution concentration ratio between the sorbed and the solution phases, and q is the sorbed concentration normalized by surface area. The BC–graphite isotherms are shown in Figure 1.5. Benzene and monosubstituted benzenes sorb somewhat more strongly on BC than on graphite, which could be true or merely an artifact of the technique for measuring surface area (Braida et al. 2003). Nevertheless, in Figure 1.5a it can be seen, that the char-graphite distribution ratio decreases—that is sorption becomes weaker relative to graphite—increasing number of ring substituents, regardless of the compound's polarity. The effect is considerable; for example, adsorption of tetramethylbenzene to BC is about an order of magnitude weaker than benzene at constant sorbed concentration on graphite. Likewise, the char-graphite distribution ratio decreases with increasing fused ring size (Fig. 1.5b), revealing a one-order-of-magnitude difference in benzene and phenanthrene affinities for BC at constant concentration on graphite. Since these compounds are all planar (although some of the nitro groups may orient in a nonplanar relationship with the benzene ring), these results conclusively show that a size exclusion effect is operative that restricts the internal pore network surface area available for adsorption as molecular size increases. Support for steric effects is also evident in the work of Nguyen et al. (2007), who found that in two chars the maximum sorption capacity increased in the order of decreasing molecular diameter among planar compounds: phenanthrene < naphthalene < 1,2-dichlorobenzene/1,2,4-trichlorobenzene < 1,4-DCB (Nguyen et al., 2007).

Figure 1.5. Char-graphite distribution coefficients normalized by surface area of aromatic compounds as a function of molecular size. Each point represents the mean and standard deviation of 11-18 data measured over a range of concentrations. [Data from (Zhu and Pignatello, 2005).] TOL = toluene; XYL = 1,4-dimethylbenzene; 124 TMB = 1,2,4-trimethylbenzene; 124 TCB = 1,2,4-trichlorobenzene; 1235 TeMB = 1,2,3,5-tetramethylbenzene; 1245 TeMB = 1,2,4,5-tetramethylbenzene; 12 DCB = 1,2-dichlorobenzene; BNTL = benzonitrile; MNT = 4-nitrotoluene; DNT = 2,4-dinitrotoluene; TNT = 2,4,6-trinitrotoluene; BEN, benzene; NAPH = naphthalene; PHEN = phenanthrene.

1.2.1.5. Sorption of Ionic and Ionizable Compounds

Because of the deprotonation of carboxyl and phenoxyl groups NOM contains an abundance of charged sites where ion exchange of organocations with native cations (M+) may occur. It is often difficult to quantify the contribution of NOM to cation exhange in whole soils because cation exchange sites are abundant on mineral surfaces as well. Organic anions may also show a tendency to sorb, particularly if the nonionic parts of the molecule are large; the classic examples are tetra- and pentachlorophenoxide (Schellenberg et al. 1984), whose sorption may occur as an ion pair with an inorganic cation and is weaker than sorption of the neutral molecule. Perfluoroalkanoic acids and perfluoroakane sulfonic acids (Ahrens et al. 2009; Higgins and Luthy 2006) also undoubtedly sorb as the ion pair, since their pKa values are well below 1. For further information on ion sorption, refer to Schwarzenbach et al. (2002).

Sorption, of ionic or ionizable molecules containing multiple functional groups presents a much more complex situation. This will be illustrated with two antibiotic compounds whose sorption to NOM has been studied in some detail—sulfamethazine and tetracycline and their analogs.

Sulfamethazine (Fig. 1.6) may exist in water