Equilibrium Sorption Phenomena

We have already discussed several equilibrium processes, and these are summarized in Table 2.8 along with some new ones for this section. Most chemists, especially those dealing with pollutant fate and transport phenomena, prefer to work with a system at equilibrium. This makes the mathematical expressions much simpler, and it allows us to ignore many poorly understood kinetic processes. In this section, we will look at pollutant sorption (attraction) phenomena between the aqueous phase and other phases present in natural water. We will use several terms to describe these processes, but in general they all mean the same thing. For example, when metals associate with particles in water, it is usually through an ion exchange mechanism on the surface of the particle and it is technically correct to use the term "adsorption" to describe the process. However, when hydrophobic pollutants associate with particles, it is more of a solvation process, since the interaction is not site- or charge-

TABLE 2.8. A Summary of Important Equilibrium Constants Used in Environmental Chemistry

K Name Mathematical Expression Description

TABLE 2.8. A Summary of Important Equilibrium Constants Used in Environmental Chemistry

K Name Mathematical Expression Description


Typical equilibrium constant

, |A-||H+| " |HA|

A common expression used to represent the

equilibrium concentration in a chemical reaction


Henry's law constant

Partial pressure of x Water con. of x (M)

Used to describe the equilibrium partitioning of a pollutant between the gas and liquid phase (units of atm-L/mol)


Organic matter-water partition coefficient

Cone, of X in DOM Cone, of X in water

Used to describe the partition or binding of pollutants between dissolved organic matter and water (L/kg)


Octanol-water coefficient

Cone, of X in octanol Cone, of X in water

Used to model the partitioning of pollutants in biota (bioconcentration) (unitless)


Sediment-water distribution coefficient

Cone, of X in sediment (mg/kg) Cone, of X in water (mg/L)

Used to show how pollutants adsorb to a solid phase (units of L/kg)


Sediment-water partition coefficient

Cone, of X in sediment (mg/kg) Cone, of X in water (mg/L)

Used to show how pollutants partition onto or into another phase (i.e.. organic matter on a sediment particle) (units of L/kg)


Organic carbon-water partition coefficient corrected for the presence of organic matter

Kd or Kp Fraction of organic matter

Used to correct the Kd or Kp for the presence of organic matter coatings, (units are the same as Kd or K„)

specific, and we use the term "partitioning." Some researchers use the term "sorption" to include both processes, since some organic compounds are slightly polar in nature and the associated process can be a mixture of partitioning and adsorption. You should try to keep these terms straight, since some researchers become very agitated when the terms are used incorrectly.

2.7.1 Sorption Surfaces

Clays. There are a variety of surfaces in natural waters, including inorganic and organic colloids (mineral phases and humin—one form of natural organic matter). Colloidal particles are defined as very small particles that do not settle out of the solution during the time scale of interest, which can range from hours to decades in length. Colloidal particles can consist of very small inorganic minerals, natural organic matter (generally classified as dissolved), or a combination of both. In addition, the mineral phases can be coated with precipitants such as iron or manganese oxides and hydroxides. Due to small particles' high ratio of surface area to volume, they can account for a large amount of adsorbed and/or partitioned mass of the pollutant in solution. Natural organic matter was mentioned at the end of Section 2.4.3, where representative structures were given. NOM comes in a variety of sizes and forms, and there seem to be an endless number of ways to characterize and describe it (discussed later). For the moment, we will classify NOM as either (a) dissolved or (b) sorbed to particles (colloidal and large particles) or humin (insoluble chunks of NOM). There is no officially defined size of colloids, but environmental chemists usually filter natural water samples through a 0.20- to 0.45-|mm filter and call everything that passes through the filter "dissolved," including colloidal. The reader should be aware that there are other definitions of dissolved, but we will use this most common definition in our discussions.

If you have had a course in geology, you know that many minerals exist in nature. Soil scientists interested in sorption phenomena study many forms of par-ticulate matter (those that do not pass a 0.45-|mm filter), especially aluminum oxides, iron oxides and hydroxides, manganese oxides and hydroxides, and clay minerals. Clays and NOM coatings on inorganic particles are important in the adsorption phenomena of all pollutants, including metals and organics. Iron and manganese minerals and precipitants, meanwhile, are important primarily for sorption of metals and polar or ionic organics.

The term clay has two meanings: a clay particle is any particle smaller than 2 |mm in size, regardless of composition, whereas a clay mineral is distinguished by its chemical composition and crystallographic structure. These two definitions tend to overlap, since most particles in the <2-|mm fraction of most soils and sediments are some form of clay mineral.

We will discuss two clay mineral phases, kaolinite and montmorillonite, shown in Figures 2.20 and 2.21, respectively. We will begin with a summary, and then we elaborate on the terms used. Clay minerals are phyllosilicates, hydrous aluminum-silicate sheet structures. Clays minerals are composed of alternating sheets of (a) silicon in tetrahedral coordination with oxygen and (b) aluminum in octahedral coordination with oxygen. Kaolinite is composed of one-to-one (1: 1) layers, each com-

Octahedral Aluminium StructureAl2o3 Octohedral

posed of one of each kind of sheet, with a chemical structure of Al2O3-2SiO2-2H2O. Montmorillonite is a 2:1 layered clay (each octahedral sheet is bounded by two tetrahedral sheets, to form a layer), with a chemical structure of Na2O-7Al2O3-22SiO2-nH2O or CaO-7ALO3-22SiO2-nH2O.

In order to understand how and why adsorption of metals to clays occurs, we must further expand on clays' chemical structure and three-dimensional shape. Clays have a characteristic structure of layers composed of two alternating types of sheets. One sheet consists of Al3+, O2-, and OH- ions, where the negative ions form an octahedral structure around the Al3+. The relative numbers of Al3+, O2-, and OH- must satisfy the valences of the entire continuous structure in two dimensions. This sheet is commonly referred to as the gibbsite sheet or octahedral sheet, since it has the same general chemical formula [Al2(OH)6] as the mineral gibbsite. The second type of sheet is composed of Si4+, O2-, and OH- ions. The Si4+ ion forms the center of a tetrahedron of oxygen atoms, while the bases of the tetrahedrons form hexagonal rings. This sheet is referred to as the silica sheet or tetrahedral sheet of the clay structure.

Clay structures consist of layers composed of various combinations of octahedral and tetrahedral sheets. The simplest combination, one of each sheet, forms a kaolinite clay. Each octahedral sheet is linked to one tetrahedral sheet through the sharing of oxygens by the Si and Al atoms. This results in the structures shown in Figure 2.20 for kaolinite. Clays with this structure are referred to as 1:1-type clays, since for every octahedral sheet there is one tetrahedral sheet. Note how combinations of O2- and OH- are used to satisfy the valence charge and result in a neutral structure. The resulting clay crystal is built up with a succession of the 1 : 1 (gibbsite-silicate) layers, one on top of another. Successive layers of kaolinite are relatively difficult to separate because of the hydrogen bonding between gibbsite and silicate sheets in adjacent 1: 1 layers. The hydrogen in the OH- from the gibbsite sheet is bound to the O2- in the silicate sheet of the adjacent layer. This rigid structure will become important in the next section, when we discuss isomorphic substitution and surface charge.

Another common form of clay is the 2:1 structure, where an octahedral sheet is sandwiched between two tetrahedral sheets. An idealized clay of this type is mont-morillonite, illustrated in Figure 2.21. These 2:1 clays have very interesting properties with respect to absorption phenomena. For example, each 2: 1 layer is rigidly held together, but adjacent 2: 1 layers may be loosely held together depending on the chemicals or ions that are present in the interstitial area (the space between adjacent 2:1 layers). In the absence of interstitial ions, dry montmorillonite layers are held together by a combination of electrostatic forces (resulting from isomorphic substitution discussed later) and van der Waals dispersion forces (between adjacent O2- groups from each layer) (Sposito, 1984). The interstitial spaces between adjacent layers in 2 : 1 clays can be expanded in the presence of water (or other solvents), as water molecules commonly occupy these interstitial areas. Water acts to hold the layers loosely together through hydrogen bonding or by the presence of hydrated ions that may be present in the interstitial space. In contrast, adjacent kaolinite layers are held so tightly together that essentially no ions can migrate between the layers. The importance of the expandable nature of 2:1 clays is that it provides more surface area for diffusion and sorption of metal ions or organic pollutants.

There are many other possible configurations of gibbsite and silicate sheets, including the brittle platy minerals called micas. Muscovite is basically a montmo-rillonite layering structure with potassium between the two 2:1 layers. Potassium acts to collapse the 2:1 layers and holds the layers tightly together. Chlorite, another common mineral, is composed of an Mg-Al gibbsite-type layer sandwiched between two 2:1 layers.

Isomorphic Substitution. The chemical structure of clays lends itself to imperfections, and these imperfections are referred to as isomorphic substitutions. Essentially any cation with a coordination number of 4 or 6 can be substituted for Si4+ and Al3+ in individual sheets. Montmorillonite rarely, if ever, is found in the pure form. However, this substitution does not occur in kaolinite, which always has a chemical structure of Al4Si4Oi0(OH)8. For montmorillonite, the most common substitution is in the tetrahedral sheets, where Al3+ replaces Si4+. Greater varieties of substitution occur in the octahedral sheet. Most commonly, the substitution of Al3+ for Si4+ and Mg2+ for Al3+ leaves a deficiency of positive charges in the montmoril-lonite layers. These substitutions occur in the crystal lattice when the clay forms, resulting in a permanent charge. The charge deficiency may be compensated for in a variety of ways: (1) by replacement of O2- by OH-, (2) by introduction of excess cations into the octahedral sheet, which may have some of its cation sites unfilled, and (3) by adsorption of cations onto the surface of individual layers. Although all of these may happen, the last will be our main focus, since it occurs after the clay is formed and can account for removal of metals from solution. Overall, isomorphic substitution results in permanent, nonspecific, diffuse charges that are spread across the clay surface.

The Inorganic Hydroxy! Croup. In our previous discussions, we have presented the clay sheet as a continuous two-dimensional surface. However, clay particles are <2 mm in diameter; therefore, there will also be considerable clay edges present in a soil or sediment sample. The most common and reactive functional group in clays is the hydroxyl group that is exposed on the outer periphery of the clay on the truncated ends of the sheets. Two types of edges occur, with silanol groups originating from the silicon tetrahedral edge and aluminol groups originating from the aluminum octahedral edge. These are similar in reactivity and only differ in the fact that silanol groups do not form inner-sphere complexes (discussed earlier). The charge of silanol and aluminol groups is highly pH-dependent, so that these groups will in general be protonated at low pH values and deprotonated (anionic) at high pH values. This is important later when we discuss the adsorption of metal ions.

Measurement of Surface Charge. The actual charge of a soil/sediment suspension can be determined by a variety of experiments. The most common measure of charge is the point of zero charge (PZC), which is the pH value of a soil suspension at the point when the total net particle charge vanishes (Sposito, 1984). This can be determined by titrating a sample and measuring the mobility of the particles under an applied voltage. Another measure of charge is the point of zero salt effect

(PZSE). The PZSE is determined by locating the common point of intersection for several graphs of surface charge (oH in Figure 2.22) versus pH, each determined at a fixed ionic strength of the background electrolyte.

Factors Affecting Metal Sorption. Clearly both of these variables (pH and ionic strength) will affect adsorption of metal pollutants. For example, as the surface charge of a particle changes, by a change in either ionic strength or pH, the affinity of the surface for a metal pollutant will change. As the pH is increased, the particle surface becomes more anionic on average and absorbs more and more of the metal from solution. As we will see in Section 2.7.4, as more metal is adsorbed onto the solid, the observed Kd (the ratio of pollutant concentration on the clay to the water phase) will increase.

Other factors that affect the adsorption of metals include the oxidation state of the metal, composition of salts contributing to ionic strength, other surface-complexed cations, and the concentration of suspended solids. Summaries of these

Figure 2.22. Hypothetical PZSE determination by a simulated titration.

Figure 2.22. Hypothetical PZSE determination by a simulated titration.

effects can be found in Ames and Rai (1978), Bell and Bates (1988), Looney et al. (1987), and Tichnor (1993).

Clay Particles in Nature. Clays in natural environments are rarely free of coatings; they are not "clean." They are normally coated with inorganic precipitates or organic molecules resulting from the degradation of plant and animal material (illustrated in Figure 2.21). Both of these types of coatings will affect surface charge, sometimes imparting a charge of their own. This brings us to the next topic, a discussion of coating on mineral surfaces.

2.7.2 Organic Matter

One of the most important factors influencing sorption phenomena is the presence of organic matter in a sample. Virtually all samples have some organic matter present, but the type and concentration can vary dramatically. In principle, the sources of organic matter are obvious: Any plant, animal, or excrement of these can be incorporated into a water or soil sample. As you can imagine, the chemical variability of the resulting compounds is unlimited. However, upon introduction into a natural system, they undergo complex microbial and abiotic transformations that produce a set of compounds generally referred to as fulvic, humic, and humin materials. For simplicity and consistency, the term "natural organic matter" will be used in this textbook to refer to any organic compound present in the sample. Compounds entering a natural system include proteins (polypeptides and nucleotides), lipids (fats, waxes, oils, and hydrocarbons), carbohydrates (cellulose, starch, hemicellu-lose, lignin), and porphyrins and plant pigments (chlorophyll, hemin, carotenes, and xantophylls) (Stumm and Morgan, 1996). The products of microbial digestion and degradation of these compounds make up NOM. A typical NOM sample contains a mixture of "fresh" organic matter additions as well as "aged" organic matter. Thus, a single NOM sample will contain thousands to tens of thousands of chemically different structures. Generally, 20-30% of the compounds in a NOM sample can be identified by conventional means as protein-like materials, polysaccharides, fatty acids, and alkanes (Schnitzer, 1986). The remaining 70-80% of the NOM consists of complex, altered residues of plants and animals. Molecular weights of NOM found "dissolved" in water range from 500 to 5000 atomic mass units (Thurman, 1985).

There have been intensive efforts to characterize the structure of NOM. These efforts are summarized in Thurman (1985), Hayes et al. (1989), and Suffet and MacCarthy (1989). However, no one expects to establish a single structural formula to describe NOM. We have simply attempted to identify important functional groups, molecular sizes, and chemical properties for these compounds. Two of these structures were shown earlier in Figure 2.7. The first is an early structure of fulvic acid. The second is a more elaborate conceptualization of NOM structure from Schultten and Schnitzer (1993). The key point of the latter figure is that there are numerous ionic sorption sites and hydrophobic centers in the large NOM molecules. These will be important when we discuss sorption of pollutants by NOM.

Other characterization attempts have concentrated on the chemical functionality of NOM. Researchers have devised chromatographic techniques to separate or fractionate NOM based on chemical properties, such as hydrophilicity and hydrophobicity (Leenheer, 1980; Leenheer and Huffman, 1976). These provide the environmental chemist with a means of characterizing different NOM molecules based on their reactivity, but you must realize that, as with most classifications, these may be rather arbitrary. A simple version of this classification identifies the functional groups observed in different NOM molecules. These groups are summarized in Table 2.9 (partially based on Killops and Killops, 1993).

Another important way of characterizing NOM is by changing the pH of a water sample and observing the behavior of the various NOM components (Hayes et al., 1989). Fulvic acids are the fraction of NOM that is soluble under all pH conditions. Humic acids are defined as the organic matter that is precipitated from an aqueous solution when the pH is decreased below 2. Given that the pH of most natural waters is between 5.5 and 9, both humic and fulvic acids will be present in most natural water samples. In contrast, humin is the fraction of NOM that is not soluble in water at any pH value. Thus, humin will be present in or associated with soil or sediment.

The most important characterization of NOM in water defines whether it is present in the dissolved form or sorbed to a solid. For this characterization, we use another operational definition. Dissolved organic matter (DOM) is defined as the organic matter that will pass though a 0.45-|mm filter (Gelman type A/E glass fiber filters are usually used for this distinction). The organic matter retained by the filter is considered to be in the particulate form, usually sorbed to inorganic particles. NOM can be attached to inorganic particles through a variety binding mechanisms, including hydrogen bonding, van der Waals dispersion forces, cationic bridging, and hydrophobic effects.

2.7.3 Organic Sorbates

First, we will consider the adsorption of ionic pollutants, specifically metals. Metals, being cations in aqueous solutions, will be adsorbed to anionic sites (negative charges on a clay or depronated functional groups on the NOM). Solution conditions that favor the formation of negative sites will favor increased adsorption of metals. A good exercise at this point would be to return to Table 2.9 and Figure 2.7 and identify functional groups that may be important in attracting cationic pollutants.

Now consider the binding of organic pollutants, specifically nonpolar, non-ionizable pollutants such as polychlorinated biphenyls (PCBs). These are commonly referred to as hydrophobic pollutants. These types of pollutants are not attracted to ionized functional groups—in fact, they are repelled by these groups. Hydrophobic pollutants are attracted to hydrophobic centers in the NOM molecule and hydropho-bic mineral surfaces. The intermolecular forces responsible for these attractions are van der Waals dispersion forces, which simply follow the old saying "like dissolves like" (e.g., hydrophobic liquid coatings dissolve hydrophobic pollutants). In this regard, the attraction is not really adsorption but is more like a solution or dissolv-

2.7 EQUILIBRIUM SORPTION PHENOMENA 83 TABLE 2.9. Functional Groups Observed in Different NOM Molecules



Resulting Compound

Hydroxyl Carbonyl


Alcohol Phenol

Aldehyde Ketone Quinone Carboxylic acid







R = CH2 Indenyl R = O Furanyl R = NH Pyrryl R = S Thiophenyl


Indene Furan Pyrrole Thiophene

Benzene Pyridine






Iron (II) Porphyrin

Iron (III) Porphyrin

ing phenomenon. As noted earlier, environmental chemists regard this type of attraction as partitioning or sorption, where the pollutant dissolves or partitions into the hydrophobic center of the NOM. The difference between adsorption and sorption is the basis for environmental chemists using (a) distribution coefficients (Kd) to describe the adsorption of metals and (b) partition coefficients (Kp) to describe the partitioning of hydrophobic pollutants to environmental particles. As with the adsorption of metals, favored by conditions that create negative sites, sorption of hydrophobic pollutants is favored by conditions that promote the formation of hydrophobic centers in the NOM or coiling of the NOM.

2.7.4 Partition Coefficients, Kd and Kp

One of the most important parameters that can determine the fate of a pollutant in an aqueous system, especially in rivers, lakes, and groundwater, is its distribution coefficient (Kd) or partition coefficient (Kp) between different media. These coefficients are a measure of how a pollutant distributes itself between the water phase and the particulate (or solid) phase. Pollutants on the solid phase are considerably less bioavailable and therefore less toxic. These sorbed pollutants can also settle out of solution in lakes or become immobile in groundwater and be effectively removed, at least temporally, from the system. An example of the buildup of pollutants in sediments is given below. We will present ways of calculating these coefficients in Chapter 3.

Distribution coefficients are concerned with adsorption, defined as the net accumulation of matter (pollutants) at the interface between a solid and a liquid. The matter (pollutant) that accumulates at the surface is referred to as the adsorbate. The solid surface on which the pollutant accumulates is the adsorbent. Partition coefficients, as we discussed earlier, are concerned with the partitioning and induced-dipole interactions between two nonpolar compounds (i.e., PCB and hydrophobic regions of NOM). Even though adsorption is not occurring in this process, the effects on the system are similar, and the terms adsorbate and adsorbent are still used.

As soil and sediments wash into lakes and streams, the particles aggregate and form larger particles that will settle in calm (quiescent) waters. When these particles have accumulated pollutants, the settling of these particles to the sediments can act as a removal mechanism (referred to as a sink) for pollutants. Over time, and when clean water and sediments return to the water body, the pollution will be buried by clean material and removed from interaction with the ecosystem. Such an example is shown in Figure 2.23 for PCBs in Lake Hartwell in South Carolina (United States). Note that as you move down into the sediment from the watersediment interface, you are moving back in time. Figure 2.23a shows an area of the lake that is subject to considerable mixing and input of pollutants. This is indicated by the variable but high presence of PCBs in each section of the sediment column. In Figure 2.23b, the PCB-contaminated sediments start to be buried by cleaner, more recent deposited sediment. Figure 2.23c shows an even further burial of contaminated sediments. Thus, the accumulation of pollutants on soil-sediment particles is an important factor in fate and transport processes. We will return to distribution

Sediment Core G26

20 30

PCB Concentration (ppm)

20 30

PCB Concentration (ppm)

Figure 2.23. PCB concentration as a function of sediment depth in Lake Hartwell Sediments. [Data from Germann (1988).]

coefficients and partition coefficients in the lake, stream, and groundwater modeling chapters.

2.7.5 Ion Exchange Phenomena for Ionic Pollutants

Another way of representing the adsorption process is as an ion exchange reaction. Here we visualize the negative surface or edge sites of a clay or environmental particle as being saturated, or nearly saturated, by native cations such as H+, Na+, and

Figure 2.23. continued

K+. Cationic pollutants, such as heavy metals, generally have a higher charge (or charge density) and therefore a higher affinity for these sites and displace the native, readily exchangeable ions. Thus, we can model the process as ion exchange. There will be a finite number of sites for cations to adsorb to the surface, and the relative abundance of these sites on the particles of a soil or other material is expressed as the ion exchange capacity. A more formal definition is the number of moles of adsorbed ions that can be desorbed from a unit mass of solid under a given set of conditions (i.e., temperature, pressure, solution composition, solid-solution ratio, etc.). Soil scientists refer to this measure as the cation exchange capacity (CEC), which is defined as the concentration of sorbed cations that can be readily exchanged for other cations. The CEC is usually reported in units of meq/100g soil, as an exchangeable charge per mass, and is an indication of a soil's ability to store nutrients or absorb metals.

As a general rule, the affinity of a soil or sediment for a metal cation will increase with the tendency of the cation to form inner-sphere complexes. For a series of uniform valence metals, this tendency is directly related to ionic radius, R, for two reasons:

1. The ionic potential (z/R; charge/radius) decreases with increasing R.

2. A larger radius implies a greater tendency for the metal to polarize (distort) in response to an electric field (the surface charge of the soil particle).

Using these two guidelines, selectivity sequences for metal ions can be established based on ionic radii:

Cs+ > Rb + > K + > Na + > Li+ Ba2+ > Sr2 + > Ca2+ > Mg2+ Hg2+ > Cd2+ > Zn2+

where adsorption of the ion increases from left to right. Unfortunately, ionic radius alone is not sufficient to predict selectivity for transition metals. Extensive experimentation has established the following order of selectivity (the Irving-Williams order) (Stumm and Morgan, 1981):

Mn2+ < Fe2+ < Co2+ < Ni2+ < Cu2+

Outer-sphere complexes, meanwhile, are responsible for the effect of pH on metal adsorption. As pH increases above the pHZPC (the pH where there is no surface charge), the net surface charge of the particle increases, thus increasing the electrostatic attraction of a mineral surface for the metal. Usually NOM is present to some degree and competes with the surface for the metal. If the NOM is present in the dissolved phase, the adsorption to the surface will decrease if the NOM has a greater affinity for the metal than does the surface. If the affinity of the NOM for the metal is less than that of the surface, little or no change in the surface adsorption will occur due to the presence of NOM. This scenario will be complicated further if the NOM subsequently adsorbs to the mineral surface, a common phenomenon in nature.

Continue reading here: Transformationdegradation Reactions

Was this article helpful?

0 0