SESSION IV
TRANSPORT MECHANISMS (2)

ROLE OF HYDROUS METAL OXIDES IN THE
TRANSPORT OF HEAVY METALS IN
THE ENVIRONMENT

G. FRED LEE

Institute for Environmental Sciences, University of Texas, Dallas, Texas

INTRODUCTION

The presence of heavy metals in natural water systems has in some instances caused significant ecosystem degradation. In the past the concern for heavy metals in the environment generally focused on milligram per liter concentrations of these contaminants. Studies conducted during the past few years have shown that while concentrations of many heavy metals approaching milligram per liter levels will cause acute lethal toxicity; microgram per liter levels will cause significant adverse sublethal toxicity such as impairment of an organism's reproductive capacity. Water quality standards for many heavy metals in many states are still at the milligram per liter concentrations. Within a few years, these will be reduced by a hundred- to a thousandfold in order to protect aquatic ecosystems and man from excessive exposure to heavy metals. As water quality standards are reduced to lower levels, a much better understanding of the aqueous environmental chemistry of heavy metal contaminants must be available in order to properly evaluate the significance of a heavy metal discharge to the environment.

One of the first steps in any systematic study of the aquatic chemistry of a potential contaminant is an elucidation of the principal modes of transport and reservoirs for the contaminant. For heavy metal contaminants such as copper, cadmium, lead, zinc, nickel, mercury, iron, and manganese, the hydrous oxides of aluminum, iron, and manganese may play dominant roles in determining reservoirs and modes of transport of these metals. This paper discusses the potential significance of hydrous metal oxides in the envir9nmental chemistry of heavy metal contaminants in natural water systems.

The ability of hydrous metal oxides to interact with heavy metals has been known for many years. One of the methods used to concentrate heavy metals from aquatic systems prior to analysis involves the co-precipitation of these metals with ferric or aluminum salts. Kolthoff and Sandell¹ in their classical text on Quantitative Inorganic Analysis discuss the co-precipitation of heavy metals with ferric, manganese and aluminum salts.

Sandell² in his book on colorimetric metal analysis presents an extensive discussion of the use of hydrous metal oxides for the recovery of heavy trace metals from dilute aqueous solutions. The advent of new, more sensitive analytical techniques, such as atomic absorption spectrophotometry and improved complexing agents, has virtually eliminated the use of hydrous metal oxides as coprecipitation agents as a means of improving sensitivity of analytical procedures. Renewed interest was developed in this technique in the 1940s and 1950s as a means of recovery of radioisotopes from natural water systems. Overman and Clark³ in their text on radioisotope techniques discuss the factors influencing the recovery of radioisotopes from environmental samples by precipitation techniques.

In order to understand the role that hydrous metal oxides may play in the environmental chemistry of heavy metal contaminants, it is essential to have some knowledge of the environmental chemistry of hydrous metal oxides. Stumm and Morgan⁴ in their text on Aquatic Chemistry provide an introduction to the general topic of the aqueous environmental chemistry of hydrous metal oxides. This work should be consulted for further information on this topic. They also briefly discuss the potential importance of hydrous metal oxides in the transport and aqueous environmental chemistry of various heavy metals such as cobalt, iron, nickel, and zinc in natural water systems.

Parks⁵ summarizes the factors controlling the sign and magnitude of surface charge of oxides and mineral oxides, especially hydrous metal oxides. He notes that the metal oxides exhibit ion exchange properties and the ion exchange capacity of the simple oxides arises from the existence of a pH dependent surface charge. According to Parks, in acid solutions, the surface charge is positive and therefore the hydrous metal oxides act as anion exchangers. In basic solutions, the surface charge is negative and they are cation exchange particles. In neutral solutions, the surface charge is mixed, both plus and minus and the particles show a limited capacity for exchange of both cations and anions. Parks also discusses the hydroxo complexes in solutions of a precipitate from hydrous metal oxides. Much of the discussion on the solution characteristics of hydroxo complexes is directly applicable to the complexes on the surfaces of hydrous metal oxide coatings on various types of particles as well as on discrete hydrous metal oxide particles.

Parks⁵ notes that the charge on hydrous metal oxides is instrumental in determining the state of dispersion, rheology, and the extent to which the solids act as ion exchangers for sorption sites. He further notes that it is possible that these materials could play important roles in the concentration of trace elements in natural water systems.

Morgan⁶ reported that the oxidation of manganese(II) by dissolved oxygen yielded a stoichiometric MnO_1.9 only under highly alkaline conditions. Oxidation under other conditions led to considerable adsorption of manganese(II) from solution.

Hem⁷ presented an extensive discussion of the environmental chemistry of manganese in natural waters. His work, as well as that of Stumm and Morgan⁴ and Delfino and Lee⁸ should be consulted for further information on the environmental behavior of manganese(II) and manganese hydrous oxide. In general, it has been found that under reducing conditions (absence of dissolved oxygen) the manganese dioxide is reduced to manganese(II). Under oxidizing conditions, however, both manganese(II) and manganese(IV) are present. Although thermodynamically unstable, manganese(II) can exist in the presence of dissolved oxygen because of the slow rates of reaction between the two species under acid-neutral or slightly alkaline pH conditions.

Stumm and Lee^9,10 studied and reviewed the aqueous environmental chemistry of iron. As contrasted to manganese, under oxidizing conditions such as in the presence of dissolved oxygen, ferric iron is the only species found in slightly acid to alkaline pH range. Ferrous iron is stable in the presence of dissolved oxygen under strongly acid conditions such as would occur in extreme cases of acid mine drainage. In addition to precipitating as a hydroxide, ferrous iron can precipitate as a carbonate under conditions of moderate to high carbonate alkalinity. Ferric iron precipitates as a hydroxide, generally in an amorphous form.

Langmuir and Whittemore¹¹ discuss the characteristics of hydrous ferric oxide precipitates in natural water Systems. Ferric iron tends to form complexes with natural water organics. Subsequent sections of this paper deal with the interactions of organics and ferric iron with respect to their potential significance in iron transport.

The aqueous environmental chemistry of aluminum is somewhat simpler than that of manganese and iron due to the single oxidation state involved. There have been numerous studies on the various characteristics of aluminum hydroxide precipitates. Hem et al.¹² have reviewed this work and discussed their studies which demonstrate a reaction between aluminum hydroxide and dissolved silica to form an alumino silicate mineral with a clay-like structure. Schenk and Weber¹³ and Porter and Weber¹⁴ found that ferric iron interacts with silica to form soluble complexes. Elderfield and Hem¹⁵ have investigated the characteristics of aluminum hydroxide in natural waters and have found that there are Al(OH)₃ particles present in the solution under conditions where aluminum hydroxide should be soluble. These particles are formed on aging aluminum hydroxide. Elderfield and Hem propose that these materials represent an aged polymeric aluminum hydroxide complex which is in the process of forming gibbsite. Hahn and Stumm¹⁶ studied the adsorption of aluminum on a silica dispersion. They reported that the destabilization of silica dispersions results from specific adsorption of positively charged hydroxo aluminum complexes onto the negatively charged colloid surface causing a decrease and ultimately a reversal of the sign of the particle's surface potential.

There are a number of practical applications of the use of the sorption ability of hydrous metal oxides in water and waste-water treatment. No attempt will be made to review the literature in this area. However, such a review would show that the use of iron or aluminum in domestic and industrial water treatment for the purpose of coagulation is often accompanied by the removal of significant amounts of trace metals, organic contaminants and other chemical species from the water. It is reasonable to propose, based on the previous studies, that the normal water treatment practice of coagulation tends to reduce the concentrations of these materials that enter domestic and industrial process or drinking water. Additional evidence for the potential significance of hydrous metal oxides of iron and aluminum in removing trace contaminants from natural waters is provided by the studies that were done during the 1950s on the removal of radioisotopes in water and wastewater treatment processes.

Hydrous metal oxides can arise from a variety of sources including the weathering of various mineral species. They enter natural water systems from both surface and groundwater. Generally in a groundwater system they would occur in the reduced oxidation states such as manganese(II) and iron(II). Upon contact with the natural water which contains oxygen, they oxidize to the hydrous metal oxide. The relative rates of oxidation of iron and manganese have been studied in detail. It has been reported by Stumm and Lee¹⁰ and Morgan and Stumm¹⁷ that while iron is readily oxidized by dissolved oxygen to the ferric form in the alkaline-neutral to slightly acid pH range, manganese on the other hand requires a much higher pH for equivalent rates of oxidation. A considerable part of the manganese oxidation may take place at the surface of particles such as calcite where there is a microzone of higher pH. Also, the manganese oxidation may be mediated to a considerable extent by microorganisms.

In lakes with anoxic sediments which have reducing conditions, it is generally found that both iron and manganese would tend to migrate in the sediments through the interstitial waters until they come in contact with oxygen, where a precipitation of the hydrous metal oxide should occur. Generally, the precipitation of iron would occur first. In lakes with anoxic hypolimnia, considerable concentrations of iron and manganese in their reduced state do build up in the water column below the thermocline. As a result of thermocline erosion, generally due to high-intensity wind stress, there could be a continual production of hydrous metal oxides which would become part of the epilimnion. Stauffer and Lee¹⁸ have studied this mode of transport for phosphorus in Lake Mendota as well as other Wisconsin lakes. It is the most significant source of phosphorus during the summer. This source is one of the dominant controls of the blue-green algal blooms in the lake throughout the summer period.

Since the hypolimnion often contains higher concentrations of iron and manganese in their reduced forms, thermocline erosion and leakage of hypolimnetic waters at the thermocline sediment interface may be important sources of freshly precipitated hydrous metal oxides in the surface waters of lakes.

There have been numerous studies which point to the potential significance of hydrous metal oxides in influencing chemical contaminants in the environment. Jenne¹⁹ has proposed that the hydrous metal oxides of manganese and iron are the principal control mechanisms for cobalt, nickel, copper, and zinc in soils and freshwater sediments. He states that the common occurrence of these oxides as coatings allows them to exert a chemical activity far in excess of their total concentrations. He further indicates that the uptake or release of these metals from these oxides is a function of such factors as increased metal ion concentration, the concentrations of other heavy metals, pH, and the amount and type of organic and inorganic complex formers in solution. Jenne claims that the information available on the factors that control copper, zinc, nickel, and cobalt in natural waters suggests that the organic matter, clays, carbonates, and precipitation as discrete oxides or hydroxides cannot explain the aqueous environmental chemistry of these elements. According to Jenne, this explanation must include, as one of the dominant factors, the environmental behavior of the hydrous oxides of iron and manganese. The primary basis for Jenne's remarks is the literature on the behavior of these metals in soil systems. It is certainly reasonable to extend this behavior to the aquatic sediment systems since, in some respects, they are somewhat like some soils. There are significant differences, however, between sediments and soils that must be considered in any specific location and care must be exercised in extrapolating soil chemistry studies to the area of aquatic chemistry of sediments.

Morgan and Stumm¹⁷ have presented a general review of the role of hydrous metal oxides in limnological transformations. Other studies in this area include the work of Shimomura et al.,²⁰ who found that mercuric ions in the presence of chloride could be adsorbed onto ferric hydroxide. This adsorption was independent of pH but was dependent on the chloride, bromide and iodide content.

Lockwood and Chen²¹ reported that mercury(II) adsorption by manganese oxides was rapid, taking place in a few minutes, when added to aged suspensions of MnO₂ at low ionic strength. The addition of sodium chloride at 0.6 M repressed adsorption below pH 9 but not above pH 10. The addition of sodium perchlorate at 0.6 M decreased the rates of adsorption by an order of magnitude. They propose that the uncharged Hg(OH)₂ is the adsorbed species. They concluded that MnO₂ may be an important mercury scavenger in fresh and brackish waters.

Krauskopf²² discussed the factors controlling the concentrations of Zn, Cu, Pb, Hi, Cd, Ni, Co, Hg, Ag, Cr, Mo, W, and V in seawater. He noted that one of the principal mechanisms for controlling the concentrations of these various elements in seawater was the adsorption on hydrated ferric oxide or manganese dioxide.

Schindler²³ discusses the heterogeneous equilibria for hydrous metal oxides in seawater and specifically notes the potential importance of these oxides in the seawater systems. Slowey et al.²⁶ noted that the distribution of copper, manganese, and zinc was related to the distribution of ferric hydroxide in these waters.

Gibbs²⁵ has demonstrated the potential importance of metallic coatings of hydrous metal oxides on river-borne sediments as one of the major phases responsible for the transport of transition metals in natural water systems. He notes that these metallic coatings are mainly ferric hydroxide.

Posselt et al²⁶ studied the sorption of metal ions and tensioactive organic substances found on hydrous manganese dioxide. The metal ions investigated included Ag⁺, Ba²⁺, Mg²⁺, Mn²⁺, Nd³⁺, Sr²⁺, anionic, nonionic and cationic surface active agents.

They found that the rates of sorption of all the metals studied were rapid, with equilibrium attained within a matter of several minutes. The cationic organic solute was also rapid. However, neither the anionic or nonionic organic solute sorbed to any significant extent. An exchange-type mechanism appears to be the principal mode of metal ion sorption. The equilibrium distributions between the sorbate and sorbent fit a Langmuir sorption equation with equilibrium capacities in the range of 0.1-0.3 mole/mole of manganese dioxide.

Murray et al.²⁷ reported that group one and two cations are strongly sorbed on manganese dioxide. They found that the sorption was independent of small pH changes at high concentrations of the cations and was highly pH dependent at low concentrations of the cations. They propose that in dilute solutions the adsorption occurs as counter ions in the diffuse double air, while in the high concentration the sorption occurs within the manganite lattice. They also reported that Ni²⁺, Cu²⁺ and especially Co²⁺ exhibited marked specific adsorption on the MnO₂.

Hingston et al.²⁸ studied the adsorption of selenite on goethite and found that the specific adsorption increased the pH of the suspension and the negative charge of the oxide surface. They proposed that the mechanism of adsorption involved the release of water molecules from the surface when selenite ion is adsorbed.

The above discussion is not meant to represent a comprehensive review of the previous studies on potential significance of hydrous metal oxides on the environmental chemistry of various contaminants. Instead, it is designed to be illustrative of the types of studies that have been done in this area. It is clear from the literature that hydrous metal oxides could play a very significant role in a wide variety of chemical contaminants in natural water systems. Certainly any discussion on the transport and cycling of heavy metals in natural waters would be incomplete if it did not consider the role of hydrous metal oxides. It is somewhat surprising to find that in the fall of 1972, a conference was held that was sponsored by the US Environmental Protection Agency, National Science Foundation and Battelle-Columbus Laboratories, concerned with the cycling and control of metals in the environments. Examination of the proceedings of this conference²⁹ points to the fact that none of the authors chose to discuss the potential significance of hydrous metal oxides as a mode of transport of heavy metals in natural water systems. It is felt that this is a significant deficiency of the 1972 conference since it is certainly improper to discuss heavy metal contaminant cycling without considering the potential role of hydrous metal oxides.

Some of the factors that could be significant in influencing the role that hydrous metal oxides play in metal ion transport are discussed below.

The review by Jenne¹⁹ should be consulted for further information on the potential role of various factors in controlling the transport of heavy metals in environmental systems, especially for references on factors controlling heavy metals in sediments and soils.

The age of a hydrous metal oxide precipitate could play a significant role in the ability of these precipitates to interact with heavy metals and other chemical contaminants. Morgan and Stumm¹⁷ summarized the work of several investigators (also see Morgan and Stumm³⁰) on the sorption characteristics of manganese dioxide for various metallic species. They found that freshly precipitated manganese dioxide has a very significant sorption capacity for heavy metals. They also found that this sorption capacity had a marked pH dependence in the neutral to slightly alkaline pH range with increasing sorption with increasing pH. However, as pointed out by Lee (see Morgan and Stumm¹⁷) this high sorption capacity and marked pH dependence may only be applicable to freshly precipitated hydrous metal oxides. The aging of the precipitate would likely reduce sorption capacity as a result of molecular rearrangements which improve the crystallinity of the precipitate. In addition, the sorption capacity of hydrous metal oxides may be changed significantly with age due to the sorption of other materials in solution. Of particular concern is the role of natural water organics on the sorption process. This will be discussed further in a subsequent section.

Of potential significance in the role of heavy metal oxides on metal ion transport is the fact that the degree of interaction between the heavy metals and hydrous metal oxides is likely to be dependent on whether the heavy metal was present at the time of formation of the hydrous metal oxide precipitate or coating. It has been known for some time that much greater incorporation of chemicals into ferric hydroxide precipitates occurs when the precipitation takes place in the presence of the contaminant. For example, Malhotra et al.³¹ found that almost twice as much phosphate was incorporated into a ferric hydroxide floc if the formation of the floc took place in the presence of the phosphate as compared to addition of the preformed floc to a phosphate solution.

If it is found that freshly precipitated hydrous metal oxides tend to have higher sorption capacities, then the regions where hydrous metal oxide formation is occurring could be areas where they would have their greatest influence on heavy metal transport in natural water systems. Particular attention should be given to boundary areas between anoxic and oxic systems such as the thermocline region in eutrophic lakes which have anoxic hypolimnia; selected trenches and fjords in the oceans and coastal zone; the boundary of anoxic sediments and the oxidized sediment, soil, or overlying water; and the point where anoxic groundwaters enter surface waters or are discharged to the surface through springs. The other boundary condition which could be important in influencing the role of hydrous metal oxides on heavy metal contaminants is the situation where there is neutralization of strongly acidic waters, such as the neutralization of acid mine drainage. At the point of neutralization, two phenomena could be occurring. One of these is the precipitation of the oxidized forms of iron and aluminum as the pH of the solution is raised. The other is the enhanced rates of oxidation of reduced forms of iron and manganese which take place at higher pH values. Morgan and Stumm¹⁷ have summarized the work in this area and have noted that the rate of oxidation of both iron and manganese increases by a factor of 100 for each pH unit increase in pH. A practical example of this phenomenon is found in the work of Theobold et al.³² in their studies on the junction of Deer Creek with Snake River in Colorado. The junction of these two rivers results in the precipitation of large amounts of iron and aluminum hydrous oxides arising from the neutralization of acidic waters. These studies have shown that the precipitates contain large amounts of various metallic species. Not only do the hydrous metal oxides exert a significant influence on the heavy metals, but also the heavy metals such as copper may have an influence on the hydrous metal oxides. For example, Stumm and Lee¹⁰ have found that the presence of copper greatly catalyzes the oxidation of ferrous sulfate by dissolved oxygen. The copper in turn would tend to interact with the ferric hydroxide formed under oxidations which take place in neutral to slightly alkaline conditions. A similar type of catalysis has been noted for the oxidation of manganese during water treatment as reported by Jenne.¹⁹

In order to better understand the role of hydrous metal oxides in the chemistry of contaminants in natural water systems, there is a need for research on the factors influencing the sorption characteristics of hydrous metal oxides as a function of age of the precipitate, sorption on the performed precipitates, and the role of organics and other chemical constituents on sorption.

Jenne¹⁹ has presented a review of the factors that could influence metal ion transport in soils and to a lesser extent in aquatic systems. In considering the aqueous environmental chemistry of metals in natural waters, the potential role of organic matter must be considered. Organics could play a dominant role in the transport of metals in natural water systems from several points of view. One of these would be the physical transport of particulate organics in which the metal and the organics would become associated either in the form of insoluble complexes or peptized colloidal species. Shapiro³³ found that extracts of natural water organics tended to influence the size of ferric hydrous oxide precipitates where a dominant size fraction occurred in the 0.1-0.45 mm size range in the presence of natural organic matter. In its absence, the size fraction was larger than this and the bulk of the precipitated ferric species could be removed on 0.45 mm pore size membrane filters. In this case, it appears that the organics played a dominant role in peptization of the iron species. This in turn could be of significance in interaction between the hydrous iron oxide and other organic and inorganic species in solution.

The work of Hall and Lee³⁴ has shown that natural water organic matter obtained under conditions which probably most closely simulate the material in the natural water systems is in true solution, provided the iron content of the solution is small. As ferric iron is added to the system, the organics and the iron become associated to form larger particles and tend to become colloidal in character. This observation probably explains the wide discrepancy among various investigators on the nature of natural water coloring matter; where some have claimed that it is colloidal, while others claim it is in true solution. Based on the Hall and Lee work, it appears that whether it is true solution or not depends directly on the amounts of iron present in the system.

In the South Atlantic and Gulf Coast areas of the United States, many of the rivers entering the estuaries contain large amounts of coloring matter which are derived from the leaching of forests and marshes. In general, the studies in -these areas have shown that the coloring matter tends to be in a precipitated form. Such a situation would likely be of significance to filter-feeding organisms such as oysters and certain crustaceans as a result of the fact that as part of their filter feeding process, they may pick up iron organic-color precipitates. Therefore, not only must there be an understanding of the occurrence and mode of transport of the hydrous metal oxides as it may affect the transport of heavy metals such as zinc, copper, nickel, cadmium, etc.; information must also be available on the role of natural organics in influencing the behavior of hydrous metal oxides. Natural organics could influence how various heavy metals interact with organics and with hydrous metal oxides. This in turn would influence how various types of organisms could obtain excessive exposure to potentially hazardous chemicals.

Recent studies by Hall³⁵ on the toxicity of zinc to algae demonstrate the potential significance of the interactions between Fe3+, complexing agents and trace heavy metals in affecting algal growth. Using standard AAP algal media, Hall found that the toxicity of zinc to Micro cystis in batch culture was a function of the order of addition of iron, EDTA and zinc to the culture medium. A different toxicity of zinc was found if the EDTA was added to the culture medium after the iron rather than before it. This pattern is probably the result of the fact that under the pH conditions that exist in the culture medium, hydrous ferric oxide would tend to be formed. The EDTA added to the culture medium is not a sufficiently strong complexing agent to prevent Fe(OH)₃precipitation and incorporation of zinc into the precipitate. Also, the zinc would tend to form complexes with the EDTA which may tend to cause it to be nontoxic to algae if the zinc-EDTA complex behaves like the copper-EDTA complex. The interactions between iron, complexing agents and heavy metals could be very important in determining the transport and toxicity of these elements to aquatic organisms.

Filter-feeding organisms tend to be highly selective in the particle size of the food that they take in. It is possible that heavy metals could interact with hydrous metal oxides whose particle size would be influenced by natural water organics in such a manner as to make the potential contaminants more or less available than would occur in the absence of natural organics.

It is generally believed that the primary role of natural organics in influencing the transport of heavy metals is complexation. Often complexation is invoked as a means of explaining a lack of behavior according to simple mass action relationships. Frequently, what could be explained as the formation of soluble complexes could also be explained as reactions of hydrous metal oxides in many situations. Sometimes investigators utilize an increase in free metal ion concentration with decreasing pH as an indication of complexation. Actually, the same kind of behavior would be expected for metal ions associated with hydrous metal oxides of the colloidal size range. Any systematic study of the metal ion complexes of natural water Systems must include efforts to determine whether the metal ion transport is due to colloidal hydrous metal oxides. It is clear that additional studies must be done in this area before the role of soluble organic complexes in influencing metal ion transport versus that of hydrous oxide transport in natural water systems can be understood. For example, Sanchez and Lee³⁶ studied the factors controlling copper in Lake Monona, Madison, Wisconsin. This lake has received 1.5 million lb of copper sulfate for algae control over the past 50 years. The copper precipitates in the system and has become incorporated in the sediments. The purpose of the Sanchez and Lee study was to ascertain whether or not there is any evidence for soluble organic complexes of copper influencing the amount of copper in solution. Based on these studies, they concluded that soluble organic complexes play a very minor role in the chemistry of copper in this lake. The concentrations found could be readily explained by either the basic carbonates m an aerated system or the sulfides in an anoxic system. For further discussion of the possible role of organics in influencing the transport of metals in natural water systems, consult the review by Saxby³⁷ on the role of metal organic complexes in geochemical cycles.

Barber³⁸ has recently published the results of some studies on the role of organic complexing agents on controlling algae growth in natural waters. He notes that in some natural waters there are considerable amounts of toxicity to algae which can be eliminated by the addition of complexing agent. Similar types of results may occur whenever there are large amounts of hydrous metal oxides in the area since the metal ions which are responsible for the toxicity may be removed from solution by sorption onto the hydrous metal oxide. Kharkar et al.³⁹ have shown that many of the transition metals are not readily desorbable from iron hydroxide precipitates. Once the co-precipitation occurs, the possibility of removing these metals under natural water conditions is quite poor. Therefore, the interaction between heavy metals and hydrous metal oxides under aerobic conditions could represent a more or less permanent sink for potentially hazardous heavy metals. However, iron and manganese hydrous oxides dissolve under anoxic conditions which could potentially lead to a significant amount of metal contaminant release to the water. Normally, however, a buildup of heavy metals under anoxic conditions does not occur due to the fact that under these conditions there is concomitant production of hydrogen sulfide which forms highly insoluble compounds with many of the heavy metals. This was the situation that Sanchez and Lee³⁶ found for copper in Lake Monona. About the only aquatic environment where it would be expected that the reduction of hydrous metal oxides could result in significant metal contaminant release, excluding iron and manganese, is an environment where sulfide production is severely limited by the amount of sulfate present in the water. It is estimated by the author that such an environment would have a sulfate concentration of less than 1-2 mg/l.

There is substantial literature on the interactions between iron and aluminum hydrous oxides and phosphate in natural water systems. No attempt will be made in this paper to review this literature except to point out that iron and aluminum hydroxide can effectively remove phosphate from natural water systems. The interaction between these species is used as a basis for advanced waste treatment methods for phosphate removal from domestic waste-waters. More recently aluminum has been added to lakes to remove phosphate in a lake-wide treatment program. Dramatic results have been attained both in Sweden and in the United States using this kind of treatment. While the alleged benefits of adding alum to lakes are for phosphate removal, it cannot be ascertained without special studies whether the reduced algal growth is due to the aluminum hydroxide removal of trace metals which are necessary for algal growth. The significance of this kind of situation has been demonstrated in some unpublished work by the author in which an investigation was made on the potential benefits that might be derived by reducing the phosphorus content of various lakes by alum addition. Samples of the lake water were taken to the laboratory and various amounts of aluminum sulfate were added to the water. Algal growth that occurred with or without the addition of the alum was noted in the standard laboratory bioassay technique. In order to check whether or not it was the removal of phosphate that was the key to influencing algal growth, the samples that have been treated with alum received phosphate in an amount equal to that originally present and an algal assay was again conducted. In some samples, it was found that very poor growth occurred when the phosphorus was added back to the samples. It was reasoned that this poor growth was due to the alum floc removing trace metals to an extent that they then became limiting for algal growth in the media. One of the methods that could be used to correct this situation is to add a trace metal supplement to the sample which had received the alum treatment. When this was done, then reasonably consistent growth occurred which was more or less proportional to the amount of growth that was achieved without the addition of alum, indicating that phosphorus was one of the key limiting elements in the samples for the algal population present.

HYDROUS OXIDE COATINGS

Clay minerals and some other mineral species have significant cation exchange capacity. It is sometimes stated that they could play a dominant role in the transport of heavy metals. However, it is doubtful that the cation exchange capacity of layer silicates, such as clay minerals, plays a significant role in heavy metal transport for several reasons. First, the cation exchange capacity represents a small part of the sorption capacity of the natural water particulate matter for cations. Studies by Fruh and Lee⁴⁰ have shown, for the uptake of cesium on Lake Mendota sediments and on vermiculite, that cation exchange capacity represents a small part of the total sorption capacity for this species. It was found that a large part of the sorption could be removed by repeated washing with low-metal-content water. Another factor to consider is that competing for the cation exchange sites with the heavy metals of interest are the bulk metal species such as calcium, magnesium, and sodium which occur at concentrations many-fold over the heavy metals. Since, in general, cation exchange reactions have distribution coefficients of approximately the same order of magnitude for the various metallic species, it would be expected that calcium and magnesium would be the dominant ions occupying the cation exchange sites with very few of them being covered by metal ions of the heavy metal type. Jenne¹⁹ has noted that there is little relationship between the cation exchange capacity of the soils and the fixation of the heavy metals in the soils.

While it is generally found that the distribution coefficients for the uptake of various cations are approximately the same for the sorption by clay minerals, Morgan and Stumm³⁰ found that the distribution coefficient for heavy metals or freshly precipitated manganese dioxide was greater than for alkaline or alkaline earth metals. Therefore, there could be a preferential sorption of heavy metals on hydrous metal oxides even in the presence of large amounts of other cations.

It should be noted that when considering the sorption capacity of mineral fragments for heavy metal species, consideration must be given to the possibility of hydrous metal oxide coatings on the surface of these particles which would in turn play a dominant role in the chemistry of heavy metals. For example, studies by Plumb and Lee⁴¹ have shown that taconite tailings derived from iron ore mining in the Mesabi Range in northern Minnesota tend to show significant sorption capacity for various metal ions such as copper, zinc, cadmium and phosphates. This capacity is manifested even though the tailings, which are composed primarily of quartz and of an iron-magnesium silicate (cummingtonite), were found to be a fraction of 1% soluble under Lake Superior conditions. Even though there was release of magnesium and silicates to a solution. there was no release of the trace amounts of copper, zinc, and nickel present in the silicate lattices. Actually, there was uptake of these species by the mineral fragments. This behavior can be explained by the fact that the iron which was released from the taconite particles would precipitate on the surface as a ferric hydroxide and would tend to remove phosphate and the heavy metals. A significant part of this removal was likely to be associated with surface coatings of the hydrous metal oxides on the surface of the mineral fragments.

A similar type of situation could develop for several types of natural water precipitates. Of particular importance would be that of the calcium carbonate species such as calcite, argonite and dolomite. These species in natural waters would tend to have a microzone of higher pH surrounding the particle than the bulk solution. This microzone of higher pH could readily promote the oxidation and precipitation on the surface of the calcium carbonate species. This hydrous metal oxide on the surface would tend to sorb various metal species from solution as has been found for the pure metal oxides investigated by Morgan and Stumm.¹⁷

Plumb and Lee⁴¹ found that drying the taconite tailings at 105^oC for 1 h markedly changed the initial release of copper. Under these conditions, it was found that the copper present in the tailings was initially partially released. After a period of time, however, this copper was adsorbed to a significant extent back onto the tailings. It is reasoned that this drying step markedly changed the surface character of the hydrous iron oxide which inhibited its initial copper sorption. The resorption after extended periods of time on the dried tailings is related to the diffusion of fresh iron from the cummingtonite particle lattice to the surface where copper sorption could occur once again. This change in the surface structure of the taconite tailings was readily demonstrated by the amount of ammonium acetate leachable copper. Plumb and Lee found that the amount of the ammonium acetate leachable copper from taconite tailings was quite small under conditions where tailings had never been dried. However, drying caused a significant increase in the ammonium acetate-leachable copper. This further points to a change in the surface chemistry of the tailings most probably related to the hydrous metal oxide on the surface of the particle such as cummingtonite.

It is important to emphasize that the control of heavy metals by mineral fragments with hydrous oxide coatings may actually be a tertiary or possibly a quaternary system where organic matter in the form of colloidal compounds or dissolved species or combinations of both may actually be involved. Few studies have been done on tertiary systems of this type involving heavy metals. Wang et a1.⁴² have conducted some studies on tertiary systems involving clay minerals, organics and pesticides. It was found that the sorption of pesticides on clays was both enhanced and retarded with the presence or absence of certain types of organic compounds. In one case, a certain type of organic would enhance the sorption of parathion on montmorillonite, while another organic would inhibit parathion sorption on montmorillonite.

A somewhat analogous situation developed when nitrilotriacetic acid was added to a solution which had been equilibrated with Lake Monona sediments. As noted above, these sediments contained large amounts of copper as a result of the fact that copper has been added to the lake for algae control. The copper was found in an aerated system to be controlled by the basic carbonate. Upon addition of NTA to the solution in the presence of sediments, it was found that the copper decreased in concentration rather than increased. Since the concentrations of NTA used were in the order of a milligram per liter, there should have been significant complex formed between the copper in NTA. Therefore, it was postulated that this soluble complex was strongly sorbed by the hydrous ferric oxide from the oxidation of the iron sulfides present in the sediments. In other words, the hydrous metal oxides tended to cause NTA to remove copper from solution rather than making it available for interactions with aquatic organisms.

Hem and Skougstad⁴³ conducted a study on the incorporation of copper into ferric hydroxide floc. They found that significant copper incorporation occurred in the acid to neutral pH range. Under alkaline conditions, it was not possible to distinguish between incorporation in the floc and the direct precipitation of copper hydroxide.

One of the most pronounced examples of the sorption capacity of hydrous metal oxides for trace metals is found in the manganese nodules from the oceans (see Goldberg⁴⁴). Numerous studies have shown that these nodules contain large amounts of trace metals. The concentration of some trace metals in the nodules is sufficient to cause serious consideration of nodule mining for recovery of heavy metals. While the exact mechanism of incorporation is not known, it is likely to involve a sorption of the metal ions on the hydrous metal oxides.

Peterson⁴⁵ found, in a study of iron and manganese encrustations on rocks taken from northern Wisconsin lakes, that these encrustations were deficient in heavy metals compared to the surrounding sediments. The reason that the marine nodules tend to concentrate heavy metals and the fresh water nodules tend to be deficient in heavy metals is not known at this time.

SUMMARY AND CONCLUSIONS

Jenne has proposed that the hydrous metal oxides of manganese and iron are nearly ubiquitous in soils and sediments, both as partial coatings on other minerals and as discrete oxide particles. He proposes that these oxides act as a sink for heavy metals. The available evidence discussed in this paper strongly supports Jenne's proposal. It has been found that the uptake and release of heavy metals is influenced by the pH of the solution and by the presence of organic and inorganic complexes. Further, it is noted that one should not judge the potential role of heavy metals in influencing the aqueous environmental chemistry of copper, zinc, nickel, cadmium, etc. based on the concentration of iron and manganese in the solution. Actually, as noted by Jenne, often hydrous metal oxide sorption activity is far in excess of what would be predicted, based on their concentrations as a result of their occurring on the surfaces of various types of mineral and detrital particles. This situation is further complicated by the fact that the reactions between the hydrous metal oxides and the heavy metals are often nonstoichiometric. Also, many of these reactions are not reversible. This makes simple solution equilibria and the simple chemical kinetic approach essentially nonapplicable to the systems. The system is further complicated by the fact that the hydrous metal oxides would not be expected to show constant characteristics as a function of the age of the oxide. A freshly precipitated oxide, such as ferric hydroxide or manganese dioxide, would likely have markedly different sorption characteristics than aged oxides or hydroxides. Further, in the presence of natural waters, it is possible that the natural water organic matter could play a significant role in uptake and release of heavy metals on hydrous oxide coatings or discrete particles.

It is evident from the discussion that while there is no doubt that hydrous metal oxides are important sinks and modes of transport for heavy metals in the environment, the quantitative magnitude of this role is not known for a variety of natural water conditions. It is clear that, as greater emphasis is placed on the control of heavy metals in the environment by water pollution control regulatory agencies, a much better understanding of the interactions between heavy metals and hydrous metal oxides must be available in order to affect technically sound and economically feasible control programs.

Acknowledgment - This paper was supported by the Institute for Environmental Sciences, University of Texas at Dallas.

REFERENCES

1. KOLTROFF, I. M. and SANDELL, E. B., Textbook of Quantitative Inorganic Analysis, 3rd edn., Macmillan, New York, 1953.

2. SANDELL, E. B., Calorimetric Determination of Traces of Metals, 3rd edn., Interscience, New York, 1959.

3. OVERMAN, R. T. and CLARK, H. M., Radioisotopes Techniques, McGraw-Hill, New York, 1960.

4. STUMM, W. and MORGAN, J. J., Aquatic Chemistry, Wiley, New York, 1970.

5. PARKS, G. A., Aqueous surface chemistry of oxides and complex oxides minerals, in Equilibrium Concepts in Natural Water Systems, Advances in Chemistry Series, Vol. 67, pp. 121-160, American Chemical Society, Washington DC, 1967.

6. MORGAN, J. J., Chemical thermodynamics, in Equilibrium Concepts in Natural Water Systems, Advances in Chemistry Series, Vol. 67, pp. 1-29, American Chemical Society, Washington DC, 1967.

7. HEM, J. D., Chemical Equilibria and Rates of Manganese Oxidation, US Geol. Survey Water-Supply Paper 1667-A, A1-A64, 1963.

8. DELFINO, J. J. and LEE, G. F., Chemistry of manganese in Lake Mendota, Wisconsin, Environ. Sci. Technol 2, 1094-1100 (1968).

9. STUMM, W. and LEE, G. F., The chemistry of aqueous iron, Schweizerische Zeitschruft fur Hydrology XXI, 295-319(1960).

10. STUMM, W. and LEE, G. F., Oxygenation of ferrous iron, Ind. Eng. Chem. 53, 143-146 (1961).

11. LANGMUIR, D. and WHITTEMORE, D. O., Variations in stability of precipitated ferric oxyhydroxides, in Nonequilibrium Systems in Natural Water Chemistry, Advances in Chemistry Series, Vol.106, pp.209-234 (1971).

12. HEM, J. D., ROBERSON, C. E., LIND, C. J., and POLZER,W. L., Interactions of Aluminum with Aqueous Silica at 25^oC, US Geol. Survey Water-Supply Paper 1827-E, 1973.

13. SCHENK, J. and WEBER, W. J., Chemical interactions of dissolved silica with iron II and iron III, J. Am. Wat Wks. Ass. 60, 199-212 (1968).

14. PORTER, R. A. and WEBER, W. J., Interaction of silicic acid with iron III and uranyl ions in dilute aqueous solutions, J. Inorganic and Nucl. Chem. 33, 2443-2449 (1971).

15. ELDERFIELD, H. and HEM, J. D., The development of crystalline structure in aluminum hydroxide polymorphs on aging, Min. Magazine 39, 89-96(1973).

16. HAHN, H. H. and STUMM, W., Coagulation by AI(III)-the role of adsorption of hydrolyzed aluminum in the kinetics of coagulation, in Adsorption from Aqueous Solution, Advances in Chemistry Series, Vol.79, pp. 91-111, American Chemical Society, Washington DC, 1968.

17. MORGAN, J. and STUMM, W. with discussion by LEE, G.F. Role of multivalent hydrous metal oxides on limnological transformations, Proceedings 2nd International Conference of Water Pollution Research, Pergamon Press, 1965.

18. STAUFFER, R. and LEE, G. F., The role of the thermocline in controlling blue-green algal blooms, to appear in the Proceedings of Eutrophication Process Modeling Workshop, Utah State University (September 1973).

19. JENNE, E. A., Controls on Mn, Fe, Co, Ni, Cu, and Zn concentrations in soils and water: the significant role of hydrous Mn and Fe oxides, Advances in Chemistry Series, Vol. 73, Trace Inorganics in Water, pp. 337-387, 1968.

20. SHIMOMURA, S., NISHIHARA, Y., FUKUMOTO, Y., and TANASE, Y., Adsorption of mercuric ion on ferric hydroxide, J. of Hygienic Chem. 15, 84-89 (1969).

21. LOCKWOOD, R. A. and CHEN, KY, Adsorption of Hg(II) by hydrous manganese oxides, Environ. Sci. Technol. 11, 10281034 (1973).

22. KRAUSKOPF, K. B., Factors controlling the concentrations of thirteen rare metals in seawater, Geochim. cosmochim. Acta 9, 1-32 (1956).

23. SCHINDLER, P. W., Heterogeneous equilibria involving oxides, hydroxides, carbonates and hydroxide carbonates, in Equilibrium Concepts in Natural Water Systems, Advances in Chemistry Series, Vol.67, pp. 196-221, American Chemical Society, Washington DC, 1967.

24. SLOWEY, J. F., HEDGES, D. H., ISBELL, F. R., and HOOD, D. W., Distribution studies of Cu, Mn, and Zn in the Gulf of Mexico, Geophys. Union (A) 46, 548(1965).

25. GIBBS, R. J. Mechanisms of trace metal transport in rivers, Science 180, 71(1973).

26. POSSELT, H. S., ANDERSON, F. I., and WEBER, W. J., Cation sorption on colloidal hydrous manganese dioxide, Environ. Sci. Technol. 2,1087-1093 (1968).

27. MURRAY, D. J., HEALY, T. W.,and FUERSTENAU, D. W., The adsorption of aqueous metal On colloidal hydrous manganese oxide, in Adsorption from Aqueous Solution, Advances in Chemistry Series, Vol.79, pp.74-81, American Chemical Society, Washington DC, 1968.

28. HINGSTON, F. J., POSNER, A. N., and QUIRK, J. P., Adsorption of selenite by goethite, in Adsorption from Aqueous Solution, Advances in Chemistry Series, Vol. 79, pp.82-90, American Chemical Society, Washington DC, 1968.

29. ANONYMOUS, Cycling and control of metals, Proceedings of Environmental Research Center, US Environmental Protection Agency, Cincinnati, Ohio, 1973.

30. MORGAN, J. and STUMM, W., Colloid-chemical properties of manganese dioxide, J. Coil. Sci. 19, 347-369 (1964).

31. MALMOTRA, S. K., LEE, G. F., and ROMLICH, G. A., Nutrient removal from secondary effluent by alum flocculation and lime precipitation, J. Air Wat Pollut. 8, 487-500 (1964).

32. THEOBOLD, P. K., JR., LAKIN, H. W., and HAWKINS, D. B., The precipitation of aluminum, iron, and manganese at the junction of Deer Creek with the Snake River in Summit County, Cob., Geochim. cosmochim. Acta 27, 121-132 (1963).

33. SHAPIRO, J., The effect of yellow organic acids on iron and other metals in water, J. Am. Wat. Wks. Ass. 56, 1062-1082 (1964).

34. HALL, K. and LEE, G. F., Molecular size and spectral characterization of organic matter in a meromictic lake, Wat. Res. 8, 239-251(1974).

35. HALL, R. H., personal communication, Procter and Gamble Corporation, Cincinnati, Ohio, 1973.

36. SANCHEZ, I. and LEE, G. F., Sorption of copper on Lake Monona sediment-effect of NTA on copper release from sediments, Wat. Res. 7, 587-593 (1973).

37. SAXBY, J. D.,Metal-organic chemistry of the geochemical cycle, Revs. Pure Appl Chem. 19, 131-150(1969).

38. BARBER, R. T., Organic ligands and phytoplanklon growth in nutrient-rich seawater, in Trace Metals and Metal-Organic Interactions in Natural Waters, pp.321-338, Ann Arbor Science Publications, 1973.

39. KHARKAR, D. P., TUREKIAN, K. K., and BERTINE, K.K., Stream supply of dissolved silver, molybdenum, antimony, selenium, chromium, cobalt, rubidium, and cesium to the oceans, Geochim. cosmochim. Acta 32 (3) 285 (1968).

40. FRUR, E.G. and LEE, G. F., unpublished results, University of Wisconsin.

41. PLUMB, R. H. and LEE, G. F., Effect of Taconite Tailings on Water Quality in Lake Superior, report to Reserve Mining Company, University of Wisconsin, Madison, 1973.

42. WANG, W. C., LEE, G. F., and SPYRIDAKIS, D., Adsorption of parathion in a multi-component solution, Wat. Res. 6, 1219-1228 (1972).

43. HEM, J. D. and SKOUGSTAO, M. W., Co-precipitation effects in solution containing ferrous, ferric and cupric ions, US Geol. Survey Water-Supply Paper 1459-E, 95-110(1960).

44. GOLDBERG, E. D., The oceans as a chemical system, in The Sea, vol.2, Composition of Seawater, Comparative and Descriptive Oceanography (Hill, ed.), 1963.

45. PETERSON, J., Aqueous environmental chemistry of lead, PhD thesis, University of Wisconsin, Madison, 1973.

References as:"Lee, G.F., 'Session IV Transport Mechanisms: Role of Hydrous Metal Oxides in the Transport of Heavy Metals in the Environment,' Proc. of Symposium of Transport of Heavy Metals in the Environment, IN: Progress in Water Technology 17: 137-147 (1975)."