UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Retention of chromated copper arsenate (CCA), a wood preservative, in soil Gerencher, Eva 1989

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1989_A6_7 G47.pdf [ 6.31MB ]
Metadata
JSON: 831-1.0097469.json
JSON-LD: 831-1.0097469-ld.json
RDF/XML (Pretty): 831-1.0097469-rdf.xml
RDF/JSON: 831-1.0097469-rdf.json
Turtle: 831-1.0097469-turtle.txt
N-Triples: 831-1.0097469-rdf-ntriples.txt
Original Record: 831-1.0097469-source.json
Full Text
831-1.0097469-fulltext.txt
Citation
831-1.0097469.ris

Full Text

RETENTION OF CHROMATED COPPER ARSENATE (CCA), A WOOD PRESERVATIVE, IN SOIL By EVA GERENCHER B.Sc, The University of British Columbia, 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Soil Science) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1989 © Eva Gerencher, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date OdoL^ 0(* , / t f ? DE-6 (2/88) ii ABSTRACT Chromated copper arsenate (CCA) is a biocidal chemical used by the wood preservation industry to extend the service life of wood products. CCA is currently the most commonly used wood preservative in British Columbia. Environmental concerns regarding the fate of CCA solutions accidentally released or chronically spilled to unpaved ground at wood preservation facilities has led to the present investigation on the retention and release of CCA constituents in soils. Batch and column tests were used to evaluate the adsorption and desorption of arsenic, chromium and copper in several B.C. soils. Batch tests were carried out using varying concentrations of CCA and individual arsenic, chromium and copper solutions. Migration of arsenic, chromium and copper in a soil column following application of a single slug dose of 2% CCA solution was investigated. Continuous leach column experiments were also conducted to evaluate adsorption of CCA constituents in soils. The column soils were sectioned and subjected to a sequential extraction procedure following influent CCA solution breakthrough to determine the partitioning, mobility and availability of As, Cr and Cu. Results from the adsorption experiments were used to identify important soil characteristics influencing the attenuation of CCA in the subsurface environment. Adsorption isotherm data was found to be well described by the Freundlich equation. The Fe and Al amorphous and crystalline hydrous oxide component of the soils studied was highly correlated with arsenic adsorption while the percent organic carbon was significantly correlated with chromium and copper retention in the soils. The clay content of the soils was poorly correlated with retention of CCA constituents in the soils studied. Desorption of arsenic, chromium and copper from soils previously equilibrated with CCA solutions showed that the extent of metal release is dependent on the amount of metal retained. At high initial CCA solution concentrations chromium was desorbed to a greater extent than arsenic or copper. This result suggests the following order for mobility of CCA constituents in the soils: chromium > copper > arsenic. iii Chromium, and to a lesser extent copper, adsorption was lower in CCA equilibrated batch tests than in single solute batch tests, particularly at high initial metal solution concentrations. Arsenic adsorption was not affected by the presence of codissolved solutes in CCA solution. The migration of arsenic, chromium and copper in a soil column following application of a single dose of 2% CCA revealed that chromium was the most mobile of the CCA constituents. Breakthrough curve plots showed that arsenic was the most strongly retained CCA constituent in both soils tested. Copper was also strongly retained in the soil columns however complete breakthrough was achieved indicating that the soils had reached a finite capacity for copper retention. Chromium was again the most weakly held CCA constituent in the continuous leach column test. Partitioning of the metals in the column soils showed that arsenic, chromium and copper were largely associated with the ammonium oxalate extractable fraction. Only copper was found to have a significant amount of retained metal in the exchangeable pool. The results suggest that environmental monitoring at CCA wood preservation facilities should focus on the potential for chromium migration in the subsurface. iv TABLE OF CONTENTS PAGE ABSTRACT ii TABLE OF CONTENTS iv LIST OF FIGURES vi LIST OF TABLES viii ACKNOWLEDGEMENTS ix 1.0 INTRODUCTION 1 2.0 LITERATURE REVIEW 4 2.1 Distribution of Arsenic, Chromium and Copper in the Environment 4 2.1.1 Arsenic - 4 2.1.2 Chromium 7 2.1.3 Copper 9 2.2 Soil and Solution Chemistry of Arsenic, Chromium and Copper 12 2.2.1 Introduction 12 2.2.2 Adsorption Isotherms 14 2.2.3 Soil/Solution Chemistry of Arsenic 18 2.2.4 Soil/Solution Chemistry of Chromium 25 2.2.5 Soil/Solution Chemistry of Copper 33 3.0 MATERIALS and METHODS 38 3.1 Soil Chemical and Physical Characterization 38 3.2 Adsorption Experiments - Batch Testing 41 V 3.2.1 CCA equilibration with soils as a function of time 41 3.2.2 Adsorption as a function of soil/solution ratio 42 3.2.3 CCA equilibration with soils 42 3.2.4 Arsenic, chromium and copper equilibration with soils 42 3.2.5 Desorption kinetics of CCA 44 3.3 Adsorption Experiments - Column Testing 45 3.3.1 Slug Dose Column Testing 45 3.3.2 Breakthrough Curve Column Testing 47 4.0 RESULTS AND DISCUSSION 48 4.1 Adsorption Experiments - Batch Testing 48 4.1.1 Adsorption of As, Cr and Cu as a function of time 48 4.1.2 Effect of soil/ solution ratio on As, Cr and Cu adsorption 48 4.1.3 CCA batch equilibration experiments 56 4.1.4 Desorption of As, Cr and Cu 79 4.1.5 Equilibration with individual metal solutions 82 4.2 Adsorption Experiments - Column Testing 96 4.2.1 Migration of a single slug dose of 2% CCA in a soil column 96 4.2.2 Breakthrough Curve 99 5.0 SUMMARY 111 6.0 CONCLUSIONS 115 7.0 REFERENCES 118 vi LIST OF FIGURES FIGURE PAGE 2.1 Eh-pH diagram for arsenic 20 2.2 Mole fraction distribution for arsenic species 20 2.3 Eh-pH diagram for arsenic showing fields of relative stability for the most important arsenic species in the presence of sulfur and barium 21 2.4 Potential Eh-pH diagram for 10'6M chromium in water 26 2.5 Chromium speciation as a function of pH and total Cr(VI) 28 2.6 The solubility of various chromium minerals at fixed redox potential 28 2.7 Stability field diagram for copper 34 3.1 Column construction 46 4.1 Adsorption of 0.02% CCA by LA Ap soil as a function of time 49 4.2 Adsorption of 2% CCA by LA Ap soil as a function of time 50 4.3 Adsorption of 0.02% CCA by WH Bhf1 soil as a function of time 51 4.4 Adsorption of 2% CCA by WH Bhf 1 soil as a function of time 52 4.5 Adsorption of 0.02% CCA by LA B soil as a function of time 53 4.6 Adsorption of 2% CCA by LA B soil as a function of time 54 4.7 Arsenic adsorption isotherm for CCA equilibrated soils 57 4.8 Chromium adsorption isotherm for CCA equilibrated soils 58 4.9 Copper adsorption isotherm for CCA equilibrated soils 59 4.10 Arsenic adsorption by surface soils equilibrated with CCA 61 4.11 Arsenic adsorption by subsurface soils equilibrated with CCA 62 4.12 Chromium adsorption by surface soils equilibrated with CCA 63 4.13 Chromium adsorption by subsurface soils equilibrated with CCA 64 4.14 Copper adsorption by surface soils equilibrated with CCA 65 4.15 Copper adsorption by subsurface soils equilibrated with CCA 66 4.16 Arsenic adsorption isotherm for single solute and CCA equilibrated Langley Ap soil 84 VII 4.17 Chromium adsorption isotherm for single solute and CCA equilibrated Langley Ap soil 85 4.18 Copper adsorption isotherm for single solute and CCA equilibrated Langley Ap soil 86 4.19 Arsenic adsorption isotherm for single solute and CCA equilibrated subsurface soils 88 4.20 Chromium adsorption isotherm for single solute and CCA equilibrated Whonnock Bhf1 soil 89 4.21 Chromium adsorption isotherm for single solute and CCA equilibrated Whonnock Bhf2 soil 90 4.22 Chromium adsorption isotherm for single solute and CCA equilibrated Milner B soil 91 4.23 Chromium adsorption isotherm for single solute and CCA equilibrated Langley B soil 92 4.24 Copper adsorption isotherm for single solute and CCA equilibrated Milner b soil 93 4.25 Copper adsorption isotherm for single solute and CCA equilibrated Langley B soil 94 4.26 Total metal concentration versus depth for the Langley Ap slug dose soil 98 4.27 Total metal concentration versus depth for the Whonnock Bhf1 slug dose soil column 98 4.28 Breakthrough curve for the Langley Ap soil column experiment 101 i 4.29 Breakthrough curve for the Whonnock Bhf1 soil column experiment 102 4.30 Partitioning of arsenic in the Langley Ap soil column following a sequential extraction scheme 105 4.31 Partitioning of chromium in the Langley Ap soil column following a sequential extraction scheme 105 4.32 Partitioning of copper in the Langley Ap soil column following a sequential extraction scheme 106 4.33 Partitioning of arsenic in the Whonnock Bhf1 soil column following a sequential extraction scheme 107 4.34 Partitioning of chromium in the Whonnock Bhf1 soil column following a sequential extraction scheme 107 4.35 Partitioning of copper in the Whonnock Bhf1 soil column following a sequential extraction scheme 108 viii LIST OF TABLES TABLE PAGE 3.1 Classification of soils used in the study 39 3.2 Selected properties of the soils studied 40 3.3 Metal concentrations corresponding to CCA solution strengths 43 4.1 Percent metal adsorbed vs soil/solution ratio 55 4.2 Percent metal adsorption by soils at initial CCA concentrations of 0.02% and 2% 67 4.3 Freundlich equation and Langmuir equation coefficient for the adsorption of arsenic, chromium and copper by 10 soils 69 4.4 Multiple regression equations relating arsenic, chromium and copper adsorption capacity parameters to properties of the soils studied 70 4.5 Simple correlation matrix - relationship between the Freundlich K parameter and selected soil properties 71 4.6 Desorption results for an initial concentration of 2% CCA equilibrated with the LA Ap and WH Bhf1 soil horizons 80 4.7 Comparison of percent metal adsorption by soils equilibrated with 0.02% and 2% CCA solution and the equivalent concentrations of individual metal solutions 83 4.8 The aqua regia extractable metal content of column soils after CCA application and subsequent leaching with deionized water 97 ix ACKNOWLEDGEMENTS I would like to thank my research supervisor, Dr. L.M. Lavkulich, for his support during the course of this study. I wish to express my appreciation to Dr. L.E. Lowe for his encouragement and helpful discussions. I wish to extend my very special thanks to Dr. D.E. Konasewich for his guidance, encouragement, support and friendship. I would also like to thank members of my committee, Dr. H. Schreier and Dr. K. Hall for their patience and guidance during the long period of this study. I am grateful to Ms. Eveline Wolterson for her expert technical assistance in using the ICP, without which this thesis could not have been completed. I would also like to thank Mr. Bernie von Spindler for his innovative equipment modifications and technical help. I wish to express my sincere appreciation to Dr. K. Morin for his invaluable advice and encouragement. My special thanks to Cindy Ott, Craig Siemens and my family for never letting me forget where my priorities should lie. I would also like to thank Rosalind Maracle for her help in the production of the theses. The support of the Science Council of British Columbia is gratefully acknowledged. 1 1.0 INTRODUCTION Contamination of groundwater resources is a major environmental concern, both from a human health and resource degradation point of view. Chemical releases from industrial facilities are one potential source of groundwater contamination. The wood preservation industry is an example of one such industry, in B.C. which has resulted in historical contamination of soils and groundwaters at several sites. As a result the most commonly used wood preservative, CCA (chromated copper arsenate) has been selected for study of subsurface attenuation mechanisms. The results of the study would aid in defining site selection criteria for wood preservation operations and aid in efforts to evaluate existing contaminated sites. Wood preservation involves the pressure or thermal impregnation of biocidal chemicals into wood to provide long term resistance to attack by fungi, insects, and marine borers. By extending the service life of available timber, wood preservation reduces the harvest of already stressed forestry resources and ensures safe working conditions where timbers are used as support structures. There are currently 18 wood preserving facilities in B.C. Thirteen of these facilities use the waterborne preservative, chromated copper arsenate, to treat a variety of wood products. A rapid upward trend in CCA usage between 1982 and 1984 was noted in wood preservation surveys commissioned by Environment Canada (Henning and Konasewich, 1984 a, b, c). Since the 1984 survey, three more CCA treatment facilities have started operation in B.C. The increased demand for chromated copper arsenate treated wood products has resulted from widespread acceptance of the products by homeowners for uses such a patio construction, playground equipment, landscaping, foundation plywood and outdoor furniture. Other uses of CCA such as treatment of marine pilings, fence posts and structural timbers have been shown to be very effective, and it is anticipated that the market for CCA treated products will broaden. Efforts to deregulate use of other wood preservative chemicals such as creosote and pentachlorophenol may also be contributing to the increased use of CCA in wood preservation. 2 The most common CCA formulation currently in use in Canada is known internationally as the Type C formulation. CCA type C is a mixture of copper oxide, chromic and arsenic acids. The concentration of active ingredients in a commercial 50% concentrate mixture of CCA is; 23.75% Cr0 3 , 9.25% CuO, and 17.0% As 2 0 5 (Koppers, 1985). CCA is typically purchased by wood preservation facilities as a 50% concentrate. It is stored in tankage and diluted to a working strength of 1.5 - 4.0%. The degree of dilution is dependent on the wood species and product type. CCA is unique among the wood preservative chemicals in that the metal components are chemically fixed within the wood cells via a complex reaction with the wood sugars to form highly insoluble precipitates (Dahlgren and Hartford, 1972). As a result wood products treated with CCA have a non-staining, clean, paintable surface. The CCA treating process is essentially a closed system with effort made to contain, collect and reuse the chemical to the greatest extent possible. However, depending on the type, extent, and effectiveness of containment surfaces employed to control drippage of chemical from freshly treated wood, CCA contamination of the ground surface and stormwater runoff may occur. Solid waste generation at CCA facilities is confined to debris and sludges which are periodically removed from collection sumps, retort cylinders and chemical tankage. In many instances such wastes are disposed in landfills, or in designated areas of the wood treating facility. Concern regarding the fate of CCA released to the environment as a result of wood preservative use prompted the Environment Protection Service - Pacific Region, in 1983, to let a contract to assess all CCA treatment facilities in B.C. The terms of the contract were to assess the facilities on the basis of current pollution control practices in place to minimize release of CCA to the environment and to insure adequate practices for worker protection. A wide variety of control practices were observed and the investigators recommended that a code of practice be developed for the industry. Several research needs were also identified. One of these was the need to clarify the fate of CCA solutions which are spilled and/or chronically dripped on unpaved ground in the yard site. While some operations take extensive measures to contain such release to the environment, other operations have virtually no containment measures. However, sampling in the vicinity of one site, which was considered to have a high degree of control measures, showed 3 large quantities of arsenic, chromium and copper in soils from a drainage ditch surrounding the site (Konasewich and Gerencher, 1986). Several of the yard sites are located near environmentally sensitive water-bodies and others are located near actively used aquifers. The industry in B.C. has little data on the mobility of CCA in soil-water systems and the threat to groundwater has not been defined. Studies to evaluate the migration and attenuation of contaminants on a site specific basis can be prohibitively expensive. Therefore, a more comprehensive approach, where quantitative data can be used to evaluate the migration/attenuation processes of contaminants in natural soil media is required. The complexity of natural soils, however, makes it infeasible to take into account all of the chemical and physical processes occurring during metal migration in soil. The major objective of this thesis is to examine the relative importance of various soil properties on the attenuation and release of arsenic, chromium and copper in soils which may result from industrial losses of CCA. To accomplish this objective, batch equilibration tests and column studies were conducted with CCA, arsenic, chromium and copper. 4 2.0 LITERATURE REVIEW 2.1 Distribution of As, Cr, Cu In the Environment 2.1.1 Arsenic Arsenic ranks twentieth in elemental abundance in the earth's crust, occurring at general levels of 2-5 ppm. Although not a major constituent of the lithosphere, arsenic is a major constituent of at least 245 different minerals. Arsenic is rarely encountered as a free element. In minerals, it exists in 3 possible oxidation states - metallic, trivalent and pentavalent (Boyle and Jonasson, 1973). Arsenic is found in trace amounts and minor quantities in nearly all the common sulfide minerals, in sedimentary iron and manganese ores, and in rock phosphate which can contain as much as 2000 ppm arsenic (Walsh and Keeney, 1975). The most common arsenic containing mineral is arsenopyrite (FeAs2). The atomic radii of As(lll) (0.59A) and As(V) (0.47A) are similar to those of Si(IV)(0.4A), AI(III)(0.51A) and Fe (lll)(0.64A). Hence, ionic substitution of arsenic is possible in various oxide and silicate minerals (Boyle and Jonasson, 1973). Natural Levels in Soil Arsenic is naturally present in soils at levels ranging from 0.1 to 100 ppm with an average of 5-6 ppm. The weathering of parent materials is the main source of arsenic in the soil environment. Soils overlying sulfide-ore deposits can contain several hundred ppm arsenic, present as the unweathered sulfide minerals or inorganic anion. Inorganic arsenate in soil is associated with iron and aluminum oxides and may be bound to organic matter (NAS, 1977). Anthropogenic Inputs In North America, most arsenic produced is a by-product of the smelting of non-ferrous metal ores. The demand for arsenic for use in agricultural biocide products has declined with the increasing use 5 of organic formulations, developed after World War II. Arsenic production by the non-ferrous metal industry has therefore become a troublesome contaminant whose removal costs during smelting and refining exceed its value (NAS, 1977). The only facility in North America currently recovering arsenic is the ASARCO smelter in Tacoma, Washington from which 90% of the condensed product is arsenic trioxide. In the mid-1940's, agricultural industry use accounted for approximately 80% of total consumption of arsenic. Currently arsenic compounds are still used in insecticides, desiccants, herbicides and fungicides and algicides, however to a much smaller extent. Fuel combustion is another source of arsenic discharge to the environment. Arsenic is present in coal and may be associated with metal sulfides, clay minerals or organic materials in the coal bed (Woolson, 1983). Petroleum oils contain very little arsenic, however, oil shales from Colorado were found to contain 82 ppm arsenic (Myers and Wunderlich, 1974 in: Woolson, 1983). Arsenic is also a major component in inorganic wood preservative compounds. Although exact figures were not reported, the USDA (1980) stated that the "quantity of arsenic used in the treatment of wood has tripled since 1970. This growth trend is expected to continue as the arsenical preservative compounds find wider application in lumber, recreational furniture, etc." Other minor uses of arsenic include; feed additives (organic arsenicals), drugs, riot-control agents, and as an additive in glass. The most significant contamination of soils with arsenic has resulted from the use of inorganic arsenicals such as As 2 0 3 , NaAs0 2, for one or more of the following uses: insecticides, herbicides, soil sterilants, silvicides, desiccants (Woolson et al., 1971). Large residues have been reported from the use of lead arsenate in orchard soils. Annual application rates of lead arsenate, beginning the 1900's, have ranged from 34-100 kg As/ha. Woolson et al. (1971) reported an As residue of 2,553 ppm for a Washington state orchard soil, with an overall average residue level of 165 ppm for 58 contaminated soil samples. The use of inorganic arsenicals as biocides has decreased in recent years. Sodium arsenite use was banned by the USDA in 1968, and organic arsenicals, carbonates and organic phosphates have largely replaced the agricultural uses of the inorganic arsenicals. 6 Accumulation of arsenic in soils may also occur as a result of burning coal and smelting or ore as emission sources. It is well documented that small-diameter particulate matter can escape most stack scrubbers and be lost to the atmosphere (USDA, 1980). Crecelius et al (1975) measured soil As levels in the vicinity of the ASARCO smelter in Tacoma, Washington. Soil as levels were found to be highest near the stack (380 ppm) with the levels decreasing with distance downwind from the stack. Plambeck and Smith (1976) in a study of arsenic contamination resulting from gold mine operations near Yellowknife, NWT reported that arsenic residues in the top 2 cm of soil exceeded 10,000 ppm within 0.8 km of the nearest gold smelter. Wood treated with inorganic metal salts such as chromated copper arsenate is another potential source of arsenic in the soil environment. There have been few assessments of arsenic releases from wood preservation facilities. During the decommissioning of one wood preservation operation in the Lower Mainland, soil arsenic concentrations of up to 22,000 ppm were found (Gough and Konasewich, 1985). Environmental Significance Contamination of groundwater sources is a major environmental concern, both from a human health and resource degradation point of view. Several CCA facilities in B.C. are located in remote areas where groundwater is the source of drinking water. Ingestion of contaminated groundwater is of particular concern with respect to human health because of the high efficiency of absorption of ingested metals (Scow et al., 1985). Arsenic is generally considered to be a human carcinogen and has recently been classified as a Group A carcinogen compound by the USEPA (1986) based on evidence for excess cancer risk for skin and lung cancers in humans. In general, the major characteristics of acute arsenic poisoning in humans are gastrointestinal disturbances, and cardiac abnormalities. Subacute exposures (eg. drinking moderately contaminated well water) results in nausea, vomiting, diarrhea, leg cramps, and disorders of the peripheral nervous, cardio-vascular, hepatic and integumentary systems (NAS, 7 1977). The drinking water limit for arsenic is 0.05 mg/L (CCREM, 1987). For discharges of arsenic to land, the sorptive/retentive capacity of the soil is the major factor influencing the leaching of arsenic to the groundwater table. This will be discussed in section 2.2. 2.1.2 Chromium Chromium is found in group VIB of the periodic table and is a member of the first transition series. Chromium is not found as a free element. Chromite, a spinel mineral with Cr(lll) on octahedral sites and iron (II) on tetrahedral sites is the most important naturally occurring chromium mineral (Langard, 1982). Chromium may also exist in the plus VI oxidation state but the reduced form is most commonly found in the environment. The average level of chromium in the earth's crust is 100 ppm. Ultramafic igneous and volcanic rocks are enriched in chromium compared to felsic rocks. Phosphorites and shales may also contain appreciable amounts of chromium (1800, 20 ppm respectively). The atomic radii of Cr(ll) and Cr(VI) are similar to Si, Al and Fe. Therefore, ionic substitution of chromium in various oxide and silicate minerals is possible. Natural Levels in Soil Soil chromium levels are influenced by the composition of the parent material from which it is formed. Hence, the background level of chromium in soils varies greatly. Bowen (1979), in a compilation of data from a number of sources, reported Cr concentrations in soils ranging from 5 to 1500 ppm. An average concentration of 43 ppm in Canadian soils was provided by Cary (1982). Soils derived from ultramafic igneous rock, such as basalt, are more likely to contain higher levels of chromium than soils derived from sandstone. 8 Anthropogenic Inputs Chromium is not currently mined in North America. Chromite, an ore composed of varying mixtures of iron and chromium oxides is the only commercial source of chromium metal. The industrial use ! of chromium is extremely diverse but can be divided into three major industrial sectors. 1. Metallurgy: Chromium is alloyed with iron, nickel, cobalt and other metals in the manufacture of steel. The form of chromium released from these operations would be largely trivalent or elemental. 2. Chemical Industry: Chromates are manufactured from chromite ore for raw material use in other industries, including; catalyst, paint, pigment, textile, fungicide, corrosion inhibitor, wood preservative products, dry cell batteries. Chromium effluents, emissions and solid waste from these industrial processes will enter various environmental compartments including soils. 3. Coal: Chromium emissions from the burning of coal will be introduced the soil environment mainly through landfilled fly ash. The amount of chromium in the fly ash has been correlated for particle size (Cary, 1982). Cr(VI) may be found in strongly alkaline fly ash (Phung et al., in Cary 1982). Soil residues in the form of slag and other solid waste from chromium steel production and chromate manufacture have caused reclamation problems and pollution of water resources (NRCC, 1976). Environmental Significance In general, hexavalent chromium is of greater concern as a toxic environmental pollutant than trivalent chromium. Trivalent chromium is rapidly immobilized in soils through a combination of solid 9 phase controls on solubility and adsorption to soil colloids. Therefore Cr(lll) is relatively unavailable to plants and has very low mobility in soil systems. Hexavalent chromium is both water soluble and a strong oxidant. In the absence of the reducing media such as organic matter or soluble Fe, Cr(VI) will be available for plant uptake and leaching processes. Numerous studies have reported Cr(VI) contamination of drinking water and groundwater from wastes disposed of through recharge basins and landfills (Griffin et al., 1977; Breeze, 1973; Robertson, 1975; Stollenwerk and Grove, 1985). The presence of Cr(VI) is generally indicative of industrial pollution, although some naturally occurring Cr(VI) exists. Hexavalent chromium contamination of groundwater, resulting from Cr(VI) discharge to a tailings pond was evaluated by Stollenwerk and Grove (1985). Aquifer material was composed of alluvium and found to have a finite capacity to adsorb Cr(VI) from solution. Cr(VI) was found to be relatively mobile compared to other heavy metals. As mentioned in the previous section, groundwater contamination is of particular concern with respect to human health (Scow et al., 1985). In humans small amounts of Cr(lll) are required for normal glucose, protein and lipid metabolism (NRCC, 1976). The toxicity of chromium has been attributed primarily to Cr(VI) which has been shown to produce liver and kidney damage, internal hemorrhage, dermatitis and respiratory problems. In addition, it has been classified as a Group A carcinogen by the USEPA (1986). The carcinogenicity of chromium has been linked to lung cancers developed by persons exposed in an occupational setting. The incidence of lung cancer appears to increase with the period of exposure to Cr(VI) (NIOSH, 1983). 2.1.3 Copper Copper is the first element of subgroup 1B of the periodic table and a member of the first transition series. Copper is widely distributed geologically and geographically, however, copper ore bodies represent only a fraction of known rock formations. The average abundance of copper in the earth's crust is estimated at 24 to 55 ppm. In nature, copper forms sulfides, sulfosalts, sulfates, 10 carbonates and other compounds and also occurs as the native metal (Parker, 1981). Chalcopyrite (CuFeS2) and bornite (Cu5FeS4) are common copper minerals. The copper contents of fine grained sedimentary rocks such as clays and shales and basic igneous rocks (e.g. basalt) are higher than in coarse grained sedimentary rocks and acid igneous rocks. Natural Levels in Soil Copper levels in soil reflect the composition of the parent material. Total amounts have been reported as ranging from 1 - 5 0 ppm (Swaine, 1969), but may be much higher if derived from mineralized parent materials. The solubility of some forms of copper contribute to copper accumulation in soil via groundwater solutions draining a mineralized area. An example would be a seepage area downslope from a bedrock copper deposit. Anthropogenic Inputs Anthropogenic sources of copper in soil include: a) historical and current mining and smelting of copper ores b) urban development, traffic, contaminated dust, dumped waste materials c) sewage sludge d) fertilizers, pesticides, wood preservatives. The long history of copper mining, particularly in Southwest England, has contaminated large areas of land to varying degrees by mining, processing and smelting of the ore minerals. Copper concentrations of up to 2000 ppm have been reported in soils from such areas (Thornton, 1979). Mining and smelting activity of copper and nickel ores in the Sudbury region of Canada has resulted in similar levels of metal accumulation in soils. Hutchinson and Whitby (1974) have shown that vegetative species abundance increases with increasing distance from the smelter in Sudbury. Industrial development and associated waste has elevated the copper content of soils in urban areas relative to rural soil copper levels. The sources are varied but would include soot and domestic 11 coal ash and municipal incinerators. Dumping of municipal (domestic) refuse is another source of copper to urban soils. Soil copper levels in excess of 100 ppm have been reported in urban soils (Thornton, 1979). Sewage sludge is frequently applied to agricultural and garden soils as a soil ameliorant and fertilizer. Depending on the source of the sewage sludge, it may contain elevated levels of heavy metals, including copper. Long term application rates may cause deleterious accumulation of copper in these soils. The coincident application of copper with phosphate rock fertilizer products has been reported by Swaine (1969). Superphosphate and triple superphosphate may contain copper up to (and occasionally greater than) 1000 ppm. Copper sulfates have been used as fungicides and fertilizers. The levels of copper introduced to soils from these inputs have seldom been reported to cause toxic effects. Copper deficiency is more often the problem. However, results of soil contamination studies at CCA operations have found levels to 27,000 ppm copper in soil (Gough and Konasewich 1985). Environmental Significance Copper is an essential and beneficial element in human metabolism and is generally considered to be non-toxic. The presence of copper in a water supply, although not constituting a hazard to health, may interfere with the intended domestic uses of the water. Copper in public water supplies enhances corrosion of aluminum and zinc. It also imparts an undesirable bitter taste to water. Staining of laundry and plumbing fixtures occurs at copper concentrations above 1.0 mg/L (Health and Welfare Canada, 1978). The ingestion of large quantities of copper may result in nausea, vomiting and death. Individuals with Wilson's disease (copper toxicosis), an inherited genetic disorder are susceptible to adverse reactions and even death from the ingestion of copper at normal levels of dietary intake. 12 2.2 Soil/Solution Chemistry of Arsenic, Chromium and Copper 2.2.1 Introduction The purpose of this section is to review the existing literature on chemical attenuation mechanisms that affect the aqueous concentrations of arsenic, chromium and copper and therefore govern their geochemical behavior in soils. Following a brief description of the major attenuation mechanisms expected to occur in the subsurface, the solution chemistry of each metal will be discussed. The final section in this review will discuss the soil components responsible for the attenuation of As, Cr and Cu in soils. Chemical attenuation refers to reactions that occur between solutes in solution and soil surfaces. These reactions change the distribution and concentration of the inorganic species in pore waters and hence their mobility in soil profile. The attenuation processes can be grouped under 2 headings: 1) precipitation/dissolution and 2) adsorption/desorption. Precipitation Precipitation of solid phases is an important attenuation mechanism. In cases where the solid phase of an element is present or can form in the subsurface and rates of precipitation/dissolution are rapid, the equilibrium solution concentration will be controlled by the solubility of the solid phase i even though adsorption/desorption reactions may be occurring. McBride (1989), however, states that in acid mineral and organic soils, the soils remain highly under-saturated with respect to precipitation of cationic metals as hydroxides or carbonates, even where high levels have been added to the soils. Coprecipitation with iron and aluminum may be a significant chemical attenuation mechanism for some cationic and anionic species during liming of soils (Jahiruddin et al., 1986; Artiole and Fuller, 1979). 13 Adsorption Adsorption/desorption is the second, and possibly more important chemical process controlling the concentration of elements in pore waters in soils. Adsorption/desorption is a complex process controlled by the geologic matrix and the hydrochemical environment. Hydrous oxides of Al, Fe and Mn, amorphous aluminosilicates (allophane) and organic material are important specific adsorbents for metals in solution (Mott, 1981; McBride, 1989; Huang et al., 1977; Barrow, 1985). Hydrochemical conditions influence adsorption by 1) controlling ion speciation via solution pH, Eh, ionic composition 2) providing ions that compete for adsorption sites, and 3) affecting the net surface charge on amphoteric adsorbents. Rai et al. (1984) summarized the most important solid matrix and hydrochemical variables for soil attenuation of arsenic, chromium and copper. Geologic Matrix Hydrochemical Environment Important Adsorbents Important Solution Variables Complexing Competing Element Fe oxide Mn oxide Clays Org. C pH Eh Ions Ions Arsenic X X X X X (moderate) Chromium X X X X X X X X Copper X X X X X X X X The table illustrates that useful generalizations and broad grouping of elements with similar geochemical behaviour can be made. However, Rai et al. (1984) pointed out that although considerable descriptive and qualitative information is available for some elements the capability still does not exist to predict quantitatively the adsorption behavior based solely on mineralogy and groundwater composition. Both cationic and anionic elements are removed from soil solution by specific adsorption reactions and non-specific or ion exchange reactions. Nonspecifically adsorbed ions are electrostatically and indifferently attracted to positive or negative charges on the soil surface. Examples of ions which fall into this category are Cl", N03", both are derived from very strong acids (Mott, 1981). 14 Specifically adsorbed anions may be derived from both strong and weak acids. All the anions in this class, with the exception of F are oxyanions. The specificity of the anion reaction with the soil surface is via ligand exchange. In other words, the oxygen ions on a hydrous oxide surface can be replaced by oxyanions such as arsenate or chromate. The oxyanion then enters into 6-fold coordination with Al3* or Fe3* ions and the surface of the adsorbate is rendered more negative (Mott, 1981; Bohn et al., 1979). Using arsenate as an example: M - OH 2 x* + H 2 As0 4 -> M - OAsH 2 ^ + H 20 Specific adsorption of most cationic species to oxide surfaces increases with increasing pH. The increasing adsorption with pH results from hydrolysis of the cations and a decrease in positive charge on amphoteric adsorbents. Ion exchange reactions will predominate at high solute concentrations when the specific adsorption capacity is exceeded (Rai et al., 1984). A mechanism of specific cation adsorption is shown below (Huang et al., 1977). Hydrolysis is followed by an adsorption reaction: IvT + mH 20 <-> M(OH)mx m + mH* M(OH)mx m + S-OH f» S -M(OH) m / m 2.2.2 Adsorption Isotherms Adsorption isotherms are a convenient method of describing the amount of solute adsorbed by a soil solid phase as a function of the equilibrium concentration of the solute (Bohn et al, 1979). The major reason for studying soil/solution phase interactions is to aid in predictions of the rate of quantity of solute leaching from a soil profile. Many equations have been developed to summarize the results from adsorption experiments. The two most common are the Freundlich equation and Langmuir equation. Barrow (1985) has argued that the purpose of these equations is merely to summarize data and to permit interpolations. Similarly, Mott (1981) has stated that the insights gained into the mechanisms of solute sorption through adsorption isotherms are mostly qualitative. It is important to recognize that a good fit of sorption data to a particular adsorption isotherm equation does not constitute proof of any specific sorptive mechanism (Travis and Etnier, 1981). The inability of sorption isotherm equations to distinguish adsorption from secondary precipitation 15 is an example of their limitations to prove sorption mechanisms (Veith and Sposito, 1977). According to Barrow (1985) development of a mechanistic description of sorption requires; 1) detailed knowledge of the reaction between ions and charged surfaces to be incorporated and 2) recognition that not only the effect of concentration but also that of time, pH, and temperature must be related to sorption. Despite these limitations, the use of simple adsorption equations such as the Freundlich and Langmuir equations can be considered a valid approach to summarize data and provide qualitative interpretations. The Freundlich isotherm is defined by the following relationship: x/m = kc 1 / n where: x/m = amount of adsorbate adsorbed per unit weight of adsorbent c = total adsorbate concentration in solution at equilibrium k and n = empirical constants The linear form of the equation is most often used so that the two adjustable parameters are easily estimated either by graphical means or by linear regression: log (x/m) = 1/n log C + log k A plot of log x/m vs log C yields a straight line with a slope of 1/n and an intercept of log k. One limitation of the Freundlich equation is the inability to predict a maximum quantity of adsorption. In addition, the basis of the Freundlich equation is empirical, without a theoretical foundation (Bohn et al, 1979). However, an underlying principle implied by the equation is that the energy of adsorption decreases as the adsorption density increases. Variation of the adsorption energy suggests surface site inhomogeneity, or possible surface interaction between adsorbed species. In other words, the initial sites occupied appear to be energetically more favorable and as these sites are occupied the specificity of the adsorption reaction decreases. 16 Several researchers have cautioned that although application of the Freundlich isotherm allows for easy curve fitting due to the flexibility of the two constants, accuracy is not guaranteed if the data are extrapolated beyond the experimental points (Bohn etal., 1979; Barrow, 1985; Travis and Etnier, 1981; Kinniburgh, 1986). The Freundlich equation was used to describe arsenic sorption by A and B horizons of five West Virgina soils (Elkhatib et al., 1984). Arsenite sorption conformed to the Freundlich isotherm over a concentration range of 5 to 1000 mg As(lll)/L. Amacher et al. (1988) reported that the Freundlich equation described the rapid initial retention of hexavalent chromium by soils. The Freundlich isotherm has been used more frequently to describe Cu adsorption by soil, especially where the adsorption surface is heterogenous and multiple sites exist (Sanders, 1980; Benjamin and Leckie, 1981). Jarvis (1981) reported that Cu sorption could be described by the Freundlich isotherm when initial Cu concentrations were 20 uM or higher. Presence of soluble organic matter may have influenced the amount of Cu remaining in solution at low initial concentrations. Kurdi and Doner (1983) also found that Cu sorption by different soil types conformed to a Freundlich but not a Langmuir equation. The Langmuir adsorption isotherm was developed by Langmuir (1918) to describe the adsorption of gases by solids. The equation has been used extensively in the soil science literature to describe the adsorption of ionic solutes by solids. The Langmuir equation is shown below. x/m = kCb 1 + kC where: k = a constant related to the binding strength b = maximum amount of adsorbate that can be adsorbed (or monolayer capacity) The equation can be rearranged to the linear form: C/(x/m) = 1/kb + C/b 17 If the adsorption data conform to the Langmuir equation, a plot of C divided by x/m vs C yields a straight line with a slope of 1/b and intercept 1/kb. This expression assumes a finite number of surface sites which distinguish it from the Freundlich equation. Other assumptions of the Langmuir equation are (Bohn et al., 1979; Rai et al., 1984; Travis and Etnier, 1981): 1. Adsorption occurs on specific sites with no electrostatic or chemical interactions between adsorbate ions. 2. For a given adsorbate, the binding energy for all surface sites is the same. 3. The binding energy is independent of adsorption density. 4. Maximum adsorption is limited to monolayer coverage of the surface sites. Multi-site Langmuir adsorption models have also been developed to resolve the problem of nonlinearity in plots of C/(x/m) vs C. Travis and Etnier (1981) in a survey of sorption relationships for reactive solutes in soil reviewed the literature describing the development of multi-site adsorption equations using the example of phosphate sorption by soil. These equations take into account the possibility that several energetically distinct sites are present on the sorbent surface. Alternatively the adsorption model suggests that 2 (or more) distinct mechanisms of adsorption occur on similar sites. The Langmuir isotherm is often used to describe sorption of reactive solutes by soil. Livesey and Huang (1981) used both the simple Langmuir equation and the two surface (i.e. multi-site) Langmuir equation to describe the adsorption of arsenic at low and higher concentrations ranges. Arsenic sorption by humic acid has been found to conform to the Langmuir equation (Thanabalasingam and Pickering, 1986). Griffin et al., 1977 described sorption of Cr(lll) and Cr(VI) by kaolinite and montmorillonite clay minerals. Interpretation of the data was aided by application of the Langmuir equation. Stollenwerk and Grove (1985) used the two-surface Langmuir equation to describe hexavalent chromium sorption by alluvinum. As with the study by Livesey and Huang (1981) the separation was based on the solute concentration ranges used. Zachara et al. (1989) also used 18 a two-part Langmuir equation to describe chromate anion sorption by subsurface soil horizons. The authors noted that the isotherm plots implied a high affinity of the solid surface for Cr0 4 2 ' at low concentration with affinity decreasing with surface saturation. Benjamin and Leckie (1981) also found that Cu sorption was described by the Langmuir isotherm at small adsorption densities, but that a higher adsorption density the data fit a Freundlich isotherm. In summary, in most adsorbate/adsorption studies investigators have reported sorption behavior to be consistent with the Langmuir model at low adsorption density but not at higher surface coverage (Benjamin and Leckie, 1981). In many studies, the Langmuir equation has been preferred to the Freundlich equation because its parameters have physico chemical significance representing the extensive (adsorption capacity) and intensive (affinity) properties of the adsorbent for the adsorbate. However, with the concurrent limitations of the Langmuir equation in appreciating adsorption site differences and its inability to effectively model adsorption over any but the most dilute concentration ranges it is not a suitable model for retention of ions in soils. Barrow (1985) has stated that the only reason to use a mathematical description of retention curves is to seek a way of summarizing behavior by a few numbers. For this reason he advocated using as simple an equation as possible provided the relationship holds over the concentration range studied. Use of the Freundlich equation, if applicable, would therefore be a suitable choice to present retention curve data. 2.2.3 Soil/Solution Chemistry of Arsenic Arsenic is a metalloid of trie Group VA elements. Its environmental chemistry is quite similar to another member of that periodic group, phosphorus, although the behavior of arsenic is more metallic in nature than phosphorus which is essentially covalent. Arsenic can exist in 4 valence states -3, 0, +3 and +5. in aqueous solutions +3 and +5 are the only important species. The chemical species of arsenic present in aqueous media will depend on the pH, pe (or Eh) and the dissolved oxygen content. An Eh-pH diagram for arsenic is shown in Figure 2.1. The outlined areas represent the Eh-pH conditions under which the indicated species are predominant. In general, the As(V) species are dominant under oxidizing conditions and occupy a relatively larger 19 stability field than the As(lli) species. The As(lll) species are more toxic, more soluble and more i mobile than the oxidized state (Penrose, 1974). The mole fraction distribution of arsenate species is shown in Figure 2.2. The pH at which the fraction of each species is equal to 0.5 corresponds to the pKa. H 3As0 4 <-> H2As04" + H* pk, = 2.24 H 2 AsO/ <-> H A s O / + FT pk, = 6.94 H A s O / <-> A s O / + H* pk3 = 11.5 The predominant arsenic (V) species under oxidizing conditions are HjAsGY and HAs04 2" with the undissociated acid becoming important at pH <4.5 and A s O / occupying a greater mole fraction at pH > 9.25. Under reducing conditions the most important species are H 3As0 3° and H 2As0 3 ' (Figure 2.1). Precipitation Precipitation/dissolution reactions of arsenic in soils have not been extensively studied. However, several investigators have suggested that different solid phase arsenic compounds may control aqueous As concentrations. Eh-pH stability diagrams showing the occurrence or absence of various dissolved arsenic species and solids have been constructed by Ferguson and Gavis (1972), Hem (1977) and others. However, only the oxides and sulfides of arsenic have been considered as solid phases in these systems. Wagemann (1978) examined 14 different metals to determine their potential as a controlling influence on total dissolved arsenic in freshwater. Four of the metals examined (Ba, Cr, Fe, Cu) were considered possible candidates. An Eh-pH diagram which included barium arsenate as a solid phase was constructed and is shown in Figure 2.3. The diagram illustrates that in the presence of barium (eg. barium sulfate) in oxidized waters above pH 4, i Ba3(As04)2 was a stable solid species. The stability calculations carried out by Wagemann used natural levels of free cation in solution. Therefore, if the solution concentration of chromium or copper was considerably above these levels (1 and 2 ug/L respectively) these cations could be Figure 2.1 Eh-pH diagram for arsenic. Solid species are enclosed in parentheses in cross hatched areas, (from Ferguson and Gavls, 1972) Figure 2.2 Mole fraction distribution of arsenic species, (from Sadiq et al, 1983) 21 10 II 12 13 F i g u r e 2.3 Eti-pH diagram for arsenic showing fields of relative stability for the most Important arsenic species in the presence of sulfur and barium, (from Wagemann et al, 1978) 22 expected to control aqueous arsenic concentrations in the neutral pH range by forming stable solid phases. Sadiq et al (1983) provided solubility relationships of various arsenate minerals. Ca 3(As0 4) 2 was the most stable arsenate mineral in alkaline conditions. Other solid phases which may precipitate in soil and sediments were listed in order of decreasing stability: Mn3(As04)2 >Cd3(As04)2 >Pb3(As04)2 >Cu3(As04)2 >AIAs04 >FeAs04. These types of solubility - controlling solid phases are expected to be highly dependent on redox and pH conditions. Livesey and Huang (1981) in a study measuring the ionic activity products for the Al, Fe*3, Ca, Mg, Mn and Pb solid phase arsenates found that soil solution levels of all arsenates were far below saturation limits. The authors concluded that adsorption, not precipitation, controlled solution arsenic concentration. In general, the literature reviewed suggests that the solubility of As species in natural systems is controlled to a greater extent by redox conditions, pH and adsorption reactions rather than by solubility equilibria. Adsorption of Arsenic in Soil The soil chemistry of arsenic is similar to that of phosphorus, especially in aerobic systems where the behavior of the arsenate anion closely resembles that of orthophosphate. The differences lie in arsenic's lability and reduced tendency to form organic complexes. Studies of arsenic retention in soils have shown the adsorption is largely controlled by the extractable amorphorus and crypto-crystalline hydrous oxides of iron and aluminum (Jacobs et al., 1970; Wauchope, 1975; Livesey and Huang, 1981; Fordham and Norrish, 1983). Woolson et al. (1971) in a survey of 58 arsenic containing surface soil samples reported that soils with a high reactive Fe content contained predominantly Fe-As compounds. Soils low in iron oxide content contained arsenic complexed with reactive aluminum or precipitated with exchangeable Ca. Arsenic adsorption studies with a number of pure phase minerals have shown that Fe- and Al- oxides have a high capacity, on the basis of surface area, for As retention relative to clay minerals (Anderson et al., 1976; Frost and Griffin, 23 1977; Huang, 1975; Leckie et al., 1980; Waugen et al., 1982). Arsenic adsorption on amorphous and crystalline hydrous oxides exhibits a marked pH dependence, with mobility of As(V) increasing at alkaline pH (pH 7-9) (Rai et al., 1984; Scow et al., 1985). Maximum As(V) adsorption occurs at acidic pH's while that of As(lll) occurs between pH 7 and 9. The pH of the adsorption edge and maxima are a complex function of the pka of the acid anion, the adsorbing species, the adsorption stoichiometry and surface charge characteristics (Hingston, 1981; Leckie et al., 1980). The relationship between adsorption and pH indicates that the primary species sorbed for As(V) is HjAsCV and for As(lll) is H 3As0 3°. In general, anion adsorption decreases as the solution pH increases and the isoelectric or point of zero charge (PZC) of the amphoteric adsorbent is approached. Several researchers have shown that with increasing adsorption density the adsorption of anionic species such as arsenate renders the surface of the oxide more negative and reduces the pH of the PZC (Hingston, 1968, 1972; Anderson ef al., 1976; Huang et al., 1977; Benjamin and Leckie, 1980). This observation is consistent with an anion adsorption mechanism of ligand exchange. Although most studies of anion sorption have used mineral models as substrates, the adsorption mechanisms postulated should be applicable to soils. Benjamin and Bloom (1981) proposed the following arsenic sorption reactions with hydrous oxides: SOH 2* + A s O / + H* <-> SOH 2As0 4H-SOH 2* + As0 4 3" + 2H* <-> SOH 2 As0 4 H 2 These reactions are consistent with experimental observations of pH increase via proton consumption and reduction in net surface positive charge (Rai et al., 1984). Few studies have been conducted on arsenic adsorption to clay minerals, largely because clays do not have an appreciable anion exchange capacity. However, arsenic retention has been correlated to clay content in a number of studies, although the results are generally difficult to interpret due to the presence of hydrous oxide coatings. Frost and Griffin (1977) reported that the adsorption 24 of arsenic anions from municipal landfill leachate solutions by kaolinite and montmorillonite was strongly pH dependent. Maximum sorption of As(V) occurred at pH 5, therefore, the principal pentavalent form being adsorbed was thought to be H2As04". The anion adsorption sites on clays are associated with the exposed octahedral cations on broken clay edges. The authors postulated that the tetrahedral configuration of the H2As04" ion and apparently high bonding energy associated with the metal surface bond are favorable for arsenic adsorption by clay minerals without surface activation by protons (i.e. independent of the PZC for clay edges). Interestingly, their results showed more than twice the arsenic sorption capacity for montmorillonite as compared with kaolinite. On the basis of broken clay edges as sorption sites for arsenic retention, the sorption capacity of the 2 clays should have been the same (or slightly higher for kaolinite). A possible explanation may be presence of reactive Fe- and Al- on the surface of the montmorillonite minerals. Fordham and Norrish (1983) also noted increased arsenic retention by kaolin and muscovite flakes carrying heavy deposits of fine grained iron oxides as compared with retention by clean flakes of the same material. Livesey and Huarlg (1981) correlated As adsorption maxima with sesquioxide components of soils and to a lesser extent clay content. However, they also cautioned that the sesquioxide components may be present as coatings on clay surfaces. Therefore, retention must be attributed to the oxide coatings not the clay minerals themselves. A paucity of data exists on arsenic retention by the organic matter component in soils. Only one paper was located. Thanabalasingam and Pickering (1986) investigated the sorption of As(lll) and As(V) species by two humic acids (HA I and HA li). They found the amount of arsenic sorbed by the humic acids to vary with arsenic concentration, pH, inorganic content of the substrate and valence state. As(V) was retained to a greater extent than As(lll). Maximum adsorption of As(V) occurred at pH 5-6, in the pH region where the monovalent ion, H-AsO/, was the dominant species. For As(lll) the principal adsorbing species were H2As303" and As 30 5". Adsorption of arsenic by HA decreased in alkaline media, likely because the humic acids dissolved. Uptake also declined at pH <5. The authors attributed this observation to increased protonation of both adsorbate and adsorbent. Phosphate, sulfate and carbonate anions were shown to compete successfully with As(V) for adsorption sites. The affinity sequence for sorption on the humic acid substrate was 25 H 2P0 4 ' > HjAsOV > S0 4 2 ' > C0 3 2". This suggested to the authors that the substrate was behaving as an anion exchanger. Therefore, because anion exchangers usually contain amino functional groups, the following interaction may be taking place: R-NH2 + HX <-> R-NH3*X R-NH3*X' + V <-> R-NH3*V + X The equation is consistent with the experimental observation that adsorption decreases with progressively alkaline conditions due to enhanced deprotonation of weak acids and competition for anion exchange sites. The adsorptive capacity for humic acids may, therefore, be related to the number of amino groups present in the organic matter. The presence of certain competing anions and complexing ions may affect arsenic adsorption by soil components. Woolson et al. (1973) in a study on the effect of phosphorus additions on arsenic toxicity to corn showed improved corn yields as the amount of P added was increased relative to As. Phosphate has been shown to strongly outcompete As(V) on hydrous oxides, soils and sediments by a number of researchers (Barrow, 1985; Goldberg, 1986; Hingston et al., 1971; Roy ef al., 1986). Other anions such as N032", Cl' and S0 4 2" even when present in excess apparently have little effect on As retention in soils (Livesey and Huang, 1981). 2.2.3 Soil/Solution Chemistry of Chromium Chromium is found in Group VI B of the periodic table and is a member of the first transition series. The trivalent and hexavalent oxidation states are the most important under ambient environmental conditions of Eh and pH. Chromium exists as the trivalent cation, Cr*3, and its hydrolysis products under reducing and moderately oxidizing conditions. Hexavalent chromium as the oxyanion Cr04 2" is the predominant form of chromium under strongly oxidizing conditions. Figure 2.4 shows an Eh - pH diagram of chromium species in water. The most stable ionic state of chromium is Cr(lll) which in an octahedral complex will have a coordination number of 6 and kinetically be relatively 26 Figure 2.4 Potential Eh-pH diagram for 10 8M chromium in water. The region enclosed with (+•) signs is where most Eh-pH measurements of matural waters will lie. (from Cary, 1982) 27 inert (Cary, 1982). It is not surprising then that the stability field for Cr(lll) extends over a wide range of Eh - pH conditions, whereas Cr(VI) is confined to strongly oxidizing conditions (Figure 2.4). Cr(lll) has a tendency to form very stable complexes with negatively charged organic and inorganic species. In the absence of anionic species, Cr(lll) can react with water in neutral solutions to form colloidal hydrous oxides (NRCC, 1976). Below pH 8 the hydroxy species Cr(lll) ions Cr(OH)2' and Cr(OH)2* are predominant Bartlett and Kimble (1976) observed precipitation of Cr(lll) at pH 5.5 presumably made up of macromolecules with Cr ions in six coordination with water and hydroxy groups. Therefore, between pH 6 and 11 aqueous concentrations of Cr(lll) are largely controlled by the low solubility of Cr(OH)3 and other solid phases. Under strongly oxidizing conditions the hexavalent species become important. The proportion of Cr(VI) present as HCr0 4', Cr042" and Cr 20 7 2" is both pH and concentration dependent. A predominance diagram using both pH and total Cr(VI) as variables is shown in Figure 2.5. The most important species at environmentally relevant concentrations are HCr04" and Cr0 4 2 '. The dichromate anion, Cr207 2", becomes important at solute concentrations >1000 mg/L (or 19.23 mM/L). Precipitation Figure 2.6 illustrates the solubility of Cr(lll) mineral at moderately oxidizing conditions (pe + pH = 10). FeCr 20 4, a relatively abundant Cr(lll) mineral and of very low solubility is likely to be the solubility - controlling solid phase under these conditions (Rai et al., 1984). Cr(OH)3 and Cr(lll) coprecipitated with Fe oxides have been suggested as solubility controlling solids by a number of researchers (Bartlett and Kimble, 1976a, 1976b; Griffin et al., 1977). Cr(lll) and Fe (III) have similar ionic radii (0.63A and 0.64A, respectively) and several investigators have noted that during weathering chromium is found to associate with ferric rich materials, such as podzolic B horizons, laterites and iron oxides (Grove and Ellis, 1980). Therefore Cr(lll) concentrations in pore waters may be controlled by the presence of Fe oxides. PH 2 o 1 o T 0 -1 c -2 o - 3 o> o -4 —t 1 1 — r C r 2 O r ? " H ? 0 r O 4 H C r O , "A" \6.0_ " ^ 2 0 . 0 mg I ' mg I - ' f - I I L CrO 4 -2 -1 0 1 2 3 4 5 6 7 8 9 10 PH Figure 2.5 Chromium speciation as a function of pH and total Cr(VI). (from Sengupta et al, 1986) Figure 2.6 The solubility of var ious chromium minerals at f ixed redox potential of (pe + pH = 10). (from Ra i et al, 1984) 29 Solubility controlling solids have not been identified for Cr(VI) largely because of anticipated reduction to the less soluble Cr(lll) form. Therefore the relative importance of solid phase control on Cr(VI) mobility cannot be evaluated. Oxidation and Reduction Reactions of Chromium in Soils Cr(lll) and Cr(VI) exhibit marked differences in their geochemical behavior (Rai et al., 1984). Cr(lll) is strongly adsorbed by clays, organic matter and hydrous oxides through specific adsorption and ion exchange reactions (Bartlett and Kimble, 1976; Griffin et al., 1977; James and Bartlett, 1983). Mobility of Cr(lll) in soil solution may also be restricted by precipitation as Cr(OH)3 and possibly divalent metal chromates at neutral to alkaline pH's (Bartlett and Kimble, 1976). Cr(VI), on the other hand, exists as an oxyanion, Cr042", and at high concentrations (>10'2M) as the dichromate anion Cr207 z". The chromate anion is moderately adsorbed by acidic subsoils that are high in iron hydrous oxides. However, Cr(VI) compounds are generally water soluble under oxidizing conditions, therefore, in the absence of hydrous oxides and at neutral to alkaline pH's they have a relatively high migratory potential in soil. The mobility of Cr(VI) compounds is of particular concern, because these compounds are generally more toxic than Cr(lll) compounds (Cary, 1982). It is apparent that the oxidation state of chromium will determine its behavior and mobility in the subsurface. Therefore, it is important to determine the factors controlling the oxidation state of solution phase chromium. Reduction of Cr(VI) -> Cr(lll) Several investigators (Bartlett and Kimble, 1976; James and Bartlett, 1983; Cary et al., 1977; Artiole and Fuller 1979; Bloomfield and Pruden, 1980; Grove and Ellis, 1980) have examined the role of particulate organic matter, dissolved Fe, dissolved sulfides, and pH on reduction of Cr(VI) to Cr(lll). Chromium (VI) was found to readily convert to Cr(lll) compounds under acidic conditions in the presence of an electron donor such as organic matter. Bloomfield and Pruden (1980) noted insignificant reduction in aerobic soils above pH 5 while under anaerobic conditions Cr(VI) was 30 extensively reduced at pH 6.65 in the presence of 0.5% dried plant matter. Dissolved Fe2* and S 2 ' will be important electron donors under anaerobic conditions while easily oxidized organic compounds are responsible for aerobic reduction of Cr(VI). James and Bartlett (1983) used gallic acid to reduce Cr(VI) in limed soils. The gallic acid rapidly reduced Cr(VI) held by soil colloids in exchangeable form except in limed samples of soils high in amorphous, organically complexed Al and Fe sesquioxides. Unfortunately, quantitative assessments of Cr(VI) reduction to Cr(lll) with varying concentrations (and types) of electron donors under variable Eh-pH conditions were not found in the literature. In summary, the rate and quantity of Cr(VI) reduction to Cr(lll) has important implications for Cr(VI) migration in the subsurface. Soils with a large pool of electron donors such as soluble and particulate organic material under mildly to strongly reducing, acidic conditions will rapidly reduce added Cr(VI) to the more stable, strongly adsorbed arid lower solubility Cr(lll) form. Therefore, disposal of Cr(VI) wastes would only pose a pollution hazard if the concentration of Cr(VI) exceeded both the adsorbing and reducing capacities of the soil. Oxidation of Cr(lll) to Cr(VI) The oxidation of Cr(lll) by dissolved oxygen under conditions approximating those in soil systems is too slow to be considered significant (Schroeder and Lee, 1975; Rai et al., 1986). Bartlett and James (1979) were the first investigators to qualitatively confirm more rapid Cr(lll) oxidation to Cr(VI) in soils with a high Mn-oxide content than in other soil types. Rai et al. (1986) studied the kinetics of Cr(lll) oxidation to Cr(VI) by reaction with manganese dioxide. They concluded that Cr(lll) species are oxidized to Cr(VI) by Mn-oxides over the entire pH range of most groundwaters. As a result, more Cr would be mobile in groundwaters contacting soils and sediments that contain Mn-oxides than would be expected from solubility constraints for Cr(lll) solids. However, the oxidative process is likely to be modified in soils and sediments with naturally occurring reductants such as organic matter. 31 Adsorption of Chromium in Soils Although Cr(VI) is the species of interest in environmental studies, the adsorption behavior of chromium compounds in soils is that of both the trivalent and hexavalent form. The adsorption of Cr(lll) in soils has not been extensively studied. Its attenuation in soils is largely attributed to solid phase formation, especially above pH 5. Adsorption may, however, be an important attenuation mechanism at low pH (<4.5). Korte et al. (1976) correlated Cr(lll) attenuation with soils high in free Fe oxides and Mn. Cary et al. (1977) in a study to evaluate techniques for maintaining plant available Cr in soil reported the reduction of added Cr(VI) to Cr(lll) in acidic soils. The Cr(lll) formed was found to be insoluble and unavailable for plant uptake with properties suggesting that it was a mixed Cr(lll) and Fe hydrous oxide. Grove and Ellis (1980) removed a large part of a soil Cr(lll) pool using oxalate and dithionite extractions. In a subsequent paper (Grove and Ellis, 1986), the authors reported that Cr(lll) added to a soil system greatly reduced the extractable levels of water soluble organo-Fe complexes. In agreement with the results of Cary et al. (1977) they interpreted their results by proposing the formation of a mixed hydrous oxide of Fe(lll) and Cr(lll). Therefore, although the hydrous oxide component of the soil is effective in attenuating Cr(lli), the reaction is more likely to be via precipitation then adsorption. Griffin et al. (1977) investigated the adsorption of Cr(lll) by kaolinite and montmorillonite. Precipitation, as hydrous chromic hydroxide, was found to be the major removal mechanism above pH 5. Below pH 4, Cr(lll) species were reported to be strongly adsorbed by both kaolinite and montmorillonite by cation exchange. In the pH range 1.5 to 4, 30 to 300 times more Cr(lll) than Cr(VI) was adsorbed by the clay minerals. Bartlett and Kimble (1976a) also correlated Cr(lll) adsorption with the clay content of soils. Organic complexation is another mechanism of Cr(lll) attenuation by soils. Bartlett and Kimble (1976a) found that sodium pyrophosphate extracted more Cr(lll) from a B horizon soil than from an Ap horizon with a higher organic matter content. Furthermore, the Cr(lll) organic complexes were 32 formed under acid conditions yet up to half the concentration remained stable after the pH was raised to neutrality. The authors suggested that Cr(lll) was largely complexed with fulvic rather humic acids. James and Bartlett (1983) reported the formation of soluble organic complexes of Cr(lll) with citric acid and fulvic acids. The organic acids effectively complexed Cr(lll) and prevented its precipitation up to approximately pH 7.5. In the absence of organic ligands Cr(lll) is precipitated between 4.5 and 5.5. In soil, organic complexation with humic and fulvic acids would modify the attenuation behavior of Cr(lll) by preventing its adsorption to clay minerals and hydrous oxides and enhancing its solubility at neutral pH. The adsorption behavior of Cr(VI) in soils is not well documented. Most studies have been conducted using pure mineral phases rather than more complex adsorbents such as soil. Research with model adsorbents suggests that amphoteric mineral phases with high points of zero charge (PZC) such a Fe- and Al- oxides and to lesser extent, clay minerals, may be effective adsorbents of chromate anions at pH levels of 2-7 (MacNaughton, 1977; Leckie et al., 1980; Rai et al., 1984; Zachara et al., 1988). Chromate adsorption, as with many protoloyzable oxyanions, is highly pH dependent and is affected by both the ionic strength of solution and surface protonation characteristics of the mineral phase (Ainsworth et al., 1989). Amorphous iron oxide (Fe2CyH20(am)), a common surface coating of subsoil particles, has a particularly high adsorption capacity for Cr(VI). Chromate adsorption increases with decreasing pH as a result of protonation of the surface hydroxyl sites and aqueous speciation of the chromate anion (Zachara ef al., 1969). Stollenwerk and Grove (1985) and Zachara et al. (1989) have correlated chromate retention with subsurface soils enriched with Fe-oxides, particularly at acidic pH's. The soils contained very little organic matter, inhibiting the reduction of Cr(VI) to Cr(lll). Griffin et al. (1977) and Zachara et al. (1988, 1989) have demonstrated chromate adsorption by clay minerals, in particular kaolinite. Hydrous oxides are thought to bind chromate via surface coordination, with indirect evidence suggesting outer-sphere complexation of adsorbing solute molecules (Rai et al., 1986; Hayes, 1987; cited in Ainsworth et al., 1989). The presence of competing ions may significantly reduce CrO/" adsorption in the subsurface environment. Surface sites that bind Cr0 4 2 ' do not appear to be 33 specific for the chromate anion. Elevated levels of dissolved C02(g), H 4Si0 4, S0 4 2 ' and phosphate anions in soils and groundwater have been shown to dramatically reduce Cr042" adsorption (Zachara et al, 1989; Benjamin and Leckie, 1981; Stollenwerk and Grove, 1985). 2.2.4 Soil/Solution Chemistry of Copper Copper is the first member of the Group 1B elements, the first transition series. In aqueous solutions copper exists in the +1 and +2 valence states. Under oxidizing conditions and pH values less than 7, Cu 2* is the dominant species, with hydrolysis species of Cu 2* becoming important at higher pH values. The pk of the first hydrolysis product of Cu 2*, CuOH*, is approximately 8. Copper adsorption strongly depends on pH, because CuOH* rather than Cu 2* is the preferred surface species for many adsorbents (McBride, 1981; Davis and Leckie, 1978). At pH 4 most of the copper will be present as the Cu 2* form, however, the proportion of CuOH* increases tenfold for each unit increase of pH (James and Barrow, 1981). Therefore, the adsorption of aqueous Cu occurs most rapidly between pH 4.5 and 5.5. Copper may also be removed from solution via complexation reactions with anions such as carbonates and phosphates as well as soluble organic compounds. Figure 2.7 illustrates a stability field diagram for copper. Precipitation Solution concentrations of copper may be solubility controlled at high pH, by precipitation of the oxide, the hydroxide or the hydroxy carbonate species of copper. Although solubility control may not be a dominant mechanism for Cu retention in soils at ambient Cu concentrations, it may become important at high Cu concentrations. Lindsay (1979) calculated trie relative stability of different Cu solids. He showed that under oxidizing conditions, cupric ferrite (CuFe204) and soil-Cu are very stable, while under reducing conditions cuprous ferrite (Cu2Fe204) is the stable phase. In non-calcareous soils, a large fraction soil-Cu has been reported to be "residual" or not extracted by weak acid, but solubilized by other extractants Figure 2.7 Stability-field diagram for copper, (from NRCC, 1979) 35 used to dissolve hydrous oxides or clays (McBride, 1981). The soil-Cu present in this pool can be considered occluded and coprecipitated copper. McBride (1978) suggested that Cu in soils may reside within oxide structures, since coprecipitation of Cu 2* in aluminum and iron hydroxides occurs readily. Several investigators have reported that formation of Cu oxide solid phases during adsorption of Cu by montmorillonite and kaolinite clays (Bingham et al., 1964; Farrah and Pickering, 1976; Abd-Elfattah and Wada 1981). Frost and Griffin (1977) in a study on Cu adsorption from landfill leachates by clay minerals concluded that precipitation is an important Cu removal mechanism above pH 6.5. McBride (1982) measured the adsorption of Cu 2* on aluminum hydroxide and oxyhydroxides as a function of pH. Precipitation of Cu(OH)2 was observed on the gibbsite adsorbent, while adsorption reactions dominated on the oxyhydroxides. Adsorption of Copper in Soil Copper is strongly adsorbed in soil by three important mechanisms: complexation by soil organic matter, specific adsorption and ion exchange (Rai et al., 1984). A fairly large body of literature exists for copper adsorption by soils, particularly for organic matter chelation and complexation and adsorption by clays (Pickering, 1979; Barrow, 1981; McBride, 1981). In subsoils low in organic matter content and at low aqueous Cu concentrations, specific adsorption of Cu by Fe, Al and Mn oxides is believed to be the most important retention mechanism (Benjamin and Leckie, 1981; Jarvis, 1981; McBride, 1982). With the exception of Pb2*, Cu 2* is the divalent transition element considered to be most strongly sorbed by hydrous oxides (McBride, 1981). Mn oxides have been reported to have a higher affinity for Cu 2* than Fe- and Al- oxides, likely due to the low PZC of Mn oxides of pH 2 (McKenzie, 1980; Golden et al., 1986). The adsorption of copper on amphoteric soil hydrous oxides is strongly pH -dependent (Jarvis, 1981; Benjamin and Leckie, 1981; Rai et al., 1984). The strong pH dependence of Cu adsorption is attributed not only to surface charge distribution on the adsorbent but also to adsorption of Cu as Cu(OH)* which is 36 more strongly bound to the surface of the hydrous oxide than the free hydrated ion, Cu 2* (Davis and Leckie, 1980). Specific adsorption of Cu by layer silicates also appears to be pH-dependent (McBride, 1981; Rai et al., 1984). According to McBride (1981) simple ion exchange is the mechanism of Cu 2* adsorption on clays at pH values below 5. Specific adsorption occurs at higher pH because layer silicate surfaces apparently promote hydrolysis of Cu 2* accompanied by the release of protons, whereby the adsorbed copper becomes less exchangeable (Farrah and Pickering, 1976; Tiller etal., 1984). However, McBride (1989) disputes evidence for specific adsorption based on ESR experiment with Cu 2* and Cr3* where it appeared that in the low pH range smectite clays promote hydrolysis beyond that observed in aqueous solution. Conversely, at high pH complete hydrolysis of the metal to form the neutral metal hydroxide appeared to be limited. Therefore, although clay exchange sites contribute to the retention of Cu in layer silicate soils, the limited solubility of Cu-oxide is possibly more important in attenuating the movement of Cu in these soils. The literature is in general agreement that in surface soils, the soil organic matter is the dominant component controlling adsorption of copper (Pickering, 1979; James and Barrow, 1981; Davies, 1980; McBride, 1981, 1989). Carboxylic, carbonyl and phenolic functional groups are important in Cu-organic matter bonding (McBride, 1978). Goodman and Cheshire detected small quantities of Cu 2* porphyrin complexes in peats. Copper is almost unique in its ability to form inner sphere complexes with soil organics at low pH (McBride, 1981). However, at higher Cu 2* loading, ESR studies indicate that Cu 2* is largely coordinated with oxygen-containing ligands of a type that do not bond very covalently (McBride, 1989). The higher molecular weight humic acids have been shown to be the most important fraction for binding Cu to organic matter in largely insoluble forms in the soil (Davies, 1980). McLaren et al. (1981) reported that humic and fulvic acids complexed approximately equal amounts of ionic copper in solution. Several types of organic molecules serve as ligands to form soluble complexes with Cu, including oxalic, citric, malic, tartaric and others (Stevenson and Ardakani, 1972). The fraction of Cu complexed by organic ligands is much higher at low pH (below pH 6) (McBride, 1989). 37 Soluble organic complexes affect the Cu adsorption by soils. McBride (1981) showed that adsorption of Cu 2* by montmorillonite was inhibited to various extents by soluble organic complexing agents, particularly EDTA and citrate Cu complexes. Fulvic acid was of intermediate effectiveness in preventing Cu 2* adsorption by the clay exchange sites. Similarly humic acids reduced Cu adsorption on kaolinite by interaction with the clay mineral surface and formation of aqueous Cu-humate complexes (Gupta and Harrison, 1982). Adsorption of Cu-organic complexes by layer silicates is unlikely because of their net negative charge (Bloomfield et al., 1976). Organic ligands have a variable effect on Cu adsorption by hydrous oxide components of soil. Both enhanced and inhibited adsorption of Cu 2* has been reported in the presence of soluble organic complexes (McBride, 1981). Fulvic and humic acids can be adsorbed by oxides via ligand exchange reactions whereby an oxide-organic-metal complex maybe formed (Davis and Leckie, 1980). In contrast, Laxen (1985) reported inhibited Cu adsorption on hydrous ferric oxide as a result of Cu complexation with soluble humic acids. Inorganic ligands have been shown to enhance adsorption of Cu 2* by forming stable surface-metal-ligand complexes (Davis and Leckie, 1978; Elliot and Huang, 1981). Clark and McBride (1985) reported enhanced adsorption of Cu 2* on allophane in the presence of phosphate. The authors suggested that a Cu-phosphate surface complex had formed and was bonded to the allophane as a Cu-phosphate ion pair. Benjamin and Bloom (1981) and Benjamin (1983), on the other hand, concluded that adsorption of anions is unaffected by the presence of cations unless the cations form a surface precipitate such as the Cu(OH)2 solid phase. McBride (1989) has postulated that the metal/phosphate ratio and the loading level of these ions on oxide surface sites was likely responsible for the observations of ion pair formation and surface precipitation of a metal-phosphate phase. 38 3.0 MATERIALS AND METHODS 3.1 Soil Physical and Chemical Characterization Ten soil samples were used in the study. The soils and their classifications are shown in Table 3.1. A variety of surface and subhorizon soils were chosen in order to 1) represent the major soil types found in the Lower Mainland, and 2) to investigate CCA adsorption to varying quantities of the three major soil components; clays, hydrous oxides and organic matter. The representation of Lower Mainland soils was felt to be important because of the concentration of wood preservation facilities in this area. The soil samples were air-dried and ground to pass through a 2 mm sieve. This soil was used for all subsequent determinations. Soil pH was measured in water in a 1:1 soil-water suspension and in 0.1 M CaCI2 in a 1:2 soil-water suspension (After Black et al, 1965). Exchangeable cations and total exchange capacity of the soils was determined by extraction with pH 7 ammonium acetate (Black et al, 1965). A colorimetric method was used to determine percent nitrate-nitrite nitrogen. Phosphorus was quantitated using the Bray P1 extractant (Black et al, 1965). Total soil organic carbon was determined by Leco Analyser. Extractions to determine iron aluminum oxyhydroxide content were carried out using: 1) acid ammonium oxalate (McKeague and Day, 1966); 2) sodium pyrophosphate Bascomb, 1968); and 3) sodium citrate-bicarbonate-dithionite (Mehra and Jackson, 1960). The extracted Fe and Al content was determined by atomic absorption spectrophotometry on a Perkin-Elmer 306 spectrophotometer. Total copper, chromium, and arsenic was determined by acid digestion (Rantala and Loring, 1973) followed by Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP) (Jarrell-Ash, Series 1100 Atom Comp) of the digest. Particle size analysis was performed using the hydrometer method (Day, 1950). Specific surface was measured by sorption of N 2 at liquid N 2 temperatures using the B.E.T. Adsorption Theory. A Quanta-Chrome Corporation Quantasorb Surface Area Analyzer, located in the Department of Mining and Metallurgy, was used for surface area measurement. Selected properties of the study soils are shown in Table 3.2. TABLE 3.1 Series Name Langley (LA) Langley (LA) Ladner (L) Crescent (CT) Crescent (CT) Grevell (G) Marble Hill (MH) Milner (MR) Whonnock (WH) Whonnock (WH) Classification of Soils Used in the Study Horizon Classification Ap Humic Luvic Gleysol; silty clay, poorly drained Btg As above Ap Humic Luvic Gleysol; silt loam, poorly drained Ap Orthic Gleysol; silty clay loam; poorly drained Bg As above Ap Orthic Regosol; sandy loam; moderately well drained Bf Orthic Humo-ferric Podzol; silt loam; well drained Bfcc Luvisolic Humo Ferric Podzol; silt loam; moderately well drained Bhf1 Duric Ferro-Humic Podzol; sandy loam; imperfectly drained Bhf2 As above 40 TABLE 3.2 Selected Properties of the Soils Studied pH SOIL Horizon water Langley Ap 4.8 Langley B 4.8 Ladner Ap 4.4 G r e v e l l Ap 5.4 Cloverdale Ap 5.1 Cloverdale B 4.5 M a r b l e h i l l B 5 . 5 Milner B 4.8 Whonnock Bhf 1 4.5 Whonnock Bhf2 4 . 8 CEC S p e c i f i c PH T o t a l C (meq/ Surface CaC12 (%) 100 g) (m2/kg) 4.3 10.10 54 5.56 4.5 0.61 30 54.92 4.1 2.43 21 5.85 5.4 1.32 13 3.99 4.8 1.55 17 7.42 4.4 0 . 64 13 7.98 4 . 9 1.13 15 10.16 4 . 4 1.51 22 13.15 3.8 8.05 72 2.90 4 .1 6.26 n /a n/a SOIL Horizon Fe (o) (%> Langley Ap 1 .13 Langley B 0 .26 Ladner Ap 0 .63 G r e v e l l Ap 0 .56 Cloverdale Ap 0 .84 Cloverdale B 0 .65 M a r b l e h i l l B 0 .57 Milner B 0 .63 Whonnock Bhf 1 0 .89 Whonnock Bhf2 1 .16 Al(o) Fe(p)2 - Al(p) ( %) (%) (%) 0 .98 0.78 0.93 0 .35 0.28 0.24 0 . 17 0.40 0 .19 0 .12 0.08 0.04 0 . 19 0.33 0.15 0 .17 0.26 0.13 0 .75 0.27 0.40 1 .38 0 .16 0.50 1 .90 0 .53 1.90 2 . 48 0.45 1.85 F e ( d )3 , Al(d) Clay (%) ( %) (%) 1.49 0 .71 31.20 1.83 0 .33 45.50 0. 67 0 .10 26.14 0.71 0 .25 14.00 1.25 0 . 11 26. 98 0.81 0 .07 23.10 1.14 0 .38 10 .20 1.12 0 .55 11.10 1.84 1 . 98 23.60 1. 90 2 .35 n/a 2' ammonium oxalate extractable i r o n and aluminum 3\ pyrophosphate extractable i r o n and aluminum d i t h i o n i t e extractable i r o n and aluminum 41 3.2 Adsorption Experiments - Batch Testing A 50 percent concentrate solution of chromated copper arsenate (Type C) was supplied by Pacific Wood Preservation Services Ltd. The concentrate consists of: 23.75% Chromium as Cr03; 9.25% copper as CuO; and 17% Arsenic as As 2 0 5 . The concentrate was used to prepare test solutions for all adsorption experiments, with the exception of those conducted with individual metal solutions. Test solutions of 2% and 0.02% CCA were selected because these concentrations represent the concentration used in the wood treatment process and typical concentrations in runoff waters, respectively. 3.2.1 Equilibration Time Three soils (LAAp, WHBhfl, and LAB) were equilibrated with a 0.02% and 2% solution of CCA to determine the time required to reach adsorption equilibrium. Ten grams of air-dried soil was suspended in 40 mLs- of the CCA solution and shaken on a reciprocal shaker for between 1 and 48 hours at room temperature. A 60 mL capacity polypropylene bottle was used as a reaction vessel. The pH was allowed to vary during the equilibration of the soil/solution. At the end of each reaction period, the suspensions were centrifuged in a Sorvall Model RC2B temperature-controlled centrifuge at 10,000 RPM for 10 minutes and then filtered through Whatman No. 4 filter paper to remove coarse fibrous organic matter. The supernatant was then analyzed for copper, chromium, i and arsenic by ICP. Samples of the initial CCA solution were analyzed prior to adding the solution to the soil samples to determine the initial solute concentrations. The amount of metal adsorbed on the soil was calculated from the difference between the initial and final As, Cr, and Cu concentrations in solution. Blanks were performed with both soils and polypropylene containers. Soil blank metal concentrations were subtracted from the experimental results. No detectable As, Cr, or Cu sorption onto the containers was noted. The final equilibration time chosen for ail subsequent batch adsorption experiments was 24 hours. Although further retention of arsenic and chromium could be expected in some soils after -this 42 period, the amount was not considered significant. Problems with surface abrasion of soil particles and potential temperature fluctuations during longer equilibration periods prompted the decision to restrict all subsequent batch adsorption experiments to 24 hours. 3.2.2 Sorption as a Function of Soil/Solution Ratio Batch tests were conducted on two soils, Ladner Ap and Milner B to investigate the effect of varying soil/solution ratios on adsorption of CCA components. Two concentrations of CCA were used, 0.02% and 2.0%, in four different soil solution ratios; 1:4, 1:8, 1:20, and 1:40. The CCA solutions were equilibrated with the two soils as described in Section 3.2.1. Following the 24 hour equilibration period the suspension was centrifuged, coarse filtered, and the solute solution was analyzed for As, Cr, and Cu concentrations by ICP analysis. The amount of metal adsorbed on the soil was calculated from the difference between the initial and final As, Cr, and Cu concentrations in solution. 3.2.3 CCA Equilibration With Soils Batch adsorption tests were conducted with 10 soils and varying concentrations of CCA. The investigation was conducted to quantify the amount of As, Cr, and Cu adsorbed by each soil. Ten grams of air dried soil was suspended in 40 mis of CCA solution with initial CCA solutions ranging from 0.001% to 3.0%. Metal concentrations corresponding to CCA solution strengths are given in Table 3.3. The pH of each solution was recorded before and after CCA equilibration with soils. Following the 24 hour equilibration period the suspension was centrifuged and the supernatant was filtered and analyzed as described in the previous sections. 3.2.4 Arsenic, Chromium and Copper Equilibration With Soils Batch adsorption tests were conducted with five soils (LA Ap, LA B, MR B, WH Bhf1, WH Bhf2) and individual metal solutions prepared from arsenic pentoxide, chromic acid and copper(ll) 43 TABLE 3.3 Metal concentrations corresponding to CCA solution strengths % Arsenic Chromium Copper CCA pH (umol/mL) (umol/mL) (umol/mL) 0.001 4.2 0.028 0.044 0.019 0.002 4.0 0.054 0.085 0.038 0.005 3.8 0.150 0.257 0.138 0.01 3.5 0.312 0.527 0.282 0.02 3.3 0.630 1.084 0.550 0.05 3.0 1.47 2.45 1.34 0.10 2.7 3.03 5.08 2.69 0.50 2.3 14.76 24.74 13.16 1.0 2.1 29.64 49.77 26.65 1.5 1.9 44.81 75.13 40.23 2.0 1.8 67.18 112.84 60.19 3.0 1.6 95.05 158.55 84.46 44 chloride. All three metal compounds are used in CCA formulations and are very soluble in water. The individual metal batch tests were carried out to determine the adsorption parameters for each metal in the soils studied and thereby allow comparison with adsorption determined in CCA formulation batch studies. Differences in adsorption parameters would suggest competitive effects between the anions and cations in solution. t" Metal concentrations were chosen to be consistent with the concentration ranges in CCA batch studies. The pH of each metal solution was adjusted to the pH of the corresponding CCA solution using 0.1 M HCL. Forty mLs of Cu, Cr or As solution of varying concentration was equilibrated with 10 g of soil at room temperature for 24 hours. The suspensions were centrifuged at 10,000 RPM for 10 minutes and coarse filtered (Whatman No.4) to remove fibrous organic material. Solution pH was measured and the supernatant analyzed for As, Cr, and Cu concentration by ICP. 3.2.5 Desorption Kinetics of CCA Desorption studies were carried out on two soils (LA Ap, WH Bhf1) previously equilibrated with CCA solution concentrations of 0.025%, 0.05%, 0.25%, 0.5%, 1.0%, and 2.0% CCA. The soil suspensions were centrifuged at 10,000 RPM for 10 minutes and a 20 ml aliquot was removed and replaced with 20 ml of distilled water. The sample was then resuspended and shaken for 8 hours. The soil suspension was centrifuged and 20 ml of the supernatant was removed for As, Cr, Cu analysis and replaced with 20 ml of distilled water. This procedure was repeated once more with 4 hours of shaking and 3 more times with 2 hours of shaking, resulting in 5 desorption steps for each sample. Desorbed As, Cr, and Cu was calculated as the difference between the equilibrium metal concentration at each desorption step and the equilibrium metal concentration prior to the desorption step. 3.3 Adsorption Experiments - Column Testing 45 Two sets of column experiments were carried out using CCA solution and the Langley Ap and Whonnock Bhf soils. The first column experiment was designed to investigate the vertical migration of Cr, Cu, and As after application of a single slug dose of 2.0% CCA followed by intermittent additions of water. This condition is meant to simulate a small spill of CCA followed by typical Lower Mainland precipitation levels. This experiment may also simulate chronic drippage of CCA solution from freshly treated lumber to unpaved ground. The second column experiment was conducted to investigate the Cr, Cu, and As solute breakthrough under aerobic, unsaturated (but "satiated") conditions. Although the two column experiments were conducted under different experimental conditions, the same column construction was used (Figure 3.1). The less than 2 mm fraction of air dried soil was packed into plexiglass columns 30 cm long with a 3 cm inside diameter. The soil was packed in 1 - 2 cm increments to minimize particle segregation and stratification. The columns were packed to a soil depth of 20 cm and the exact soil weight was recorded to allow calculation of bulk density. A 2 cm layer of acid washed Ottawa sand was placed above and below the soil plug to allow uniform dispersion of solution over the soil surface. A 3 mm diameter circle of 150-mesh nylon was placed above the column drain to trap fine particles. A rubber stopper fitted with a 1 cm diameter plexiglass rod was used as the column outlet. 3.3.1 Slug Dose Column Testing Three replicate soil columns were constructed for each soil. Two acid-washed sand columns were also set up to compare solute migration in a non-reactive medium. Soil blanks were also performed by percolating acidified water (at pH 3.3) through two soil columns constructed for that purpose. The columns were initially wetted with 2 pore volumes of 0.01 M CaCI2 followed by 2 pore volumes of water. The hydraulic conductivity was estimated for each column by recording the length of time Figure 3.1 Column construction 47 for a 3 cm head of free water to infiltrate the column. Following this initial wetting period, a 10 ml slug dose of 2.0% CCA solution was added to each column. The CCA solution was permitted to penetrate the soil over a period of 5 days before the first water wash. One pore volume of water was then added to the soil column in two doses of 35 mL each. The effluent collected was then analyzed for Cr, Cu, and As by ICP. This procedure was repeated 5 more times at 5 - 6 day intervals. The total amount of water added in the 6 column washes was representative of the volume of Lower Mainland rainfall over a winter season. Following the effluent collection period, the columns were sectioned into 8 equal parts and the total metal concentrations in each section determined by digestion with aqua regia digestion. A reverse aqua regia digestion (3 parts HN03:1 part HCl) was used because of the high organic matter content of the soils. 3.3.2 Breakthrough Curve Column Testing Three replicate soil columns were constructed for each soil. A multichannel Manostat peristaltic pump was used to deliver a constant flux of solution at a rate of 0.3 mL/min. This rate of flow resulted in delivery of between 6 and 7 pore volume displacements per day. The columns were initially wetted with 2 pore volumes of 0.01 M CaCI2 followed by 2 pore volumes of distilled water. The influent was then switched to 0.02% CCA. The effluent was collected in pore volume aliquots until one of two conditions were met: 1) breakthrough - where the ratio of effluent concentration to influent concentration (C/Co) equals 1 or, 2) steady state - where the effluent concentration was unchanged or changing very slowly at a value below that of the influent. Arsenic, chromium and copper concentrations in the influent and effluent were measured using ICP analysis. After breakthrough was established, the column soils were sectioned into 4 equal parts for further analysis. The column soils were subsequently subjected to a sequential extraction consisting of a distilled water leach, MgCI2 leach, ammonium oxalate extraction and (reverse) aqua regia digestion (Tessier et al., 1979). The purpose of these extractions was to determine the partitioning, mobility and availability of Cr, Cu and As in each column following influent CCA addition. 48 4.0 RESULTS AND DISCUSSION 4.1 Adsorption Experiments - Batch Testing 4.1.1 Adsorption of As, Cr and Cu as a Function of Time The results for the 0.02% CCA solution, shown in Figure 4.1 through 4.6 indicate that while adsorption equilibrium was established for arsenic and copper in less than 2 hours, chromium had still not reached equilibrium at 48 hours in the LA Ap and Wh Bhf soils. Adsorption equilibrium was established for all three metals in less than 2 hours in the LA B soil. Equilibration with 2.0% CCA solution, in the LA Ap and WH Bhf1 soils, resulted in a longer time period required to reach adsorption equilibrium. The equilibration time data suggest that retention of chromium in the soils is occurring by more than one mechanism. It is important to note that both soils in which this secondary mechanism of adsorption was noted contain an appreciable organic matter content. The WH Bhf1 soil also contains elevated levels of hydrous oxides.The length of time required to reach equilibrium in these soils, following the rapid initial adsorption, may indicate reduction of the hexavalent to the trivalent species. The length of time taken to reach equilibrium in the case of the more concentrated solution (2%) may also be attributed to a secondary adsorption mechanism occurring, possibly ion exchange. ^ 4.1.2 Effect of Soil/Solution Ratio on As, Cr and Cu Adsorption Two soils were equilibrated with CCA solutions at varying soil/solution ratios to evaluate changes in the amount of arsenic, chromium and copper adsorbed. The results indicate that the percent metal adsorbed increases appreciably as the soil/solution ratio increases (Table 4.1). The effect is not as pronounced for arsenic in the Milner B soil, perhaps because of this soil's apparent affinity for arsenic. The change in attenuation capacity of the soils is also greater at higher initial metal 49 C n Z3 O < cr h -z LU O O o < r — Ld 3 5 0 300.4 o ^ 0 " 2 5 0 H 2 0 0 H 150 1 00 5 0 H 0-4 -e e- -e e- -© AT I s e e w ARSFN IC A A A-A-A CHROMIUM • • • • • C O P P E R 1 1 1 I I | I I I I | I I I I | | | | | | | | | | | | | | | , | | | , p | M, , 0 5 . 1 0 . 1 5 2 0 2 5 3 0 3 5 4 0 4 5 50 TIME (hrs) Figure 4.1 Adsorption of 0.02% CCA by LA A p soil as a function of time. 50 3 0 0 0 0 0 7 1 1 1 1 I 1 1 ' ' I I I I I I I ' I I | I I I I | I I I I | I I I I | I I I I | I I I I | i | | | I • 0 5 10 15 20 25 30 35 40 45 50 TIME (hrs) Figure 4.2 Adsorption of 2% C C A by LA A soil as a function of time 51 3 5 0 CP CD z> o h -< . c r 3 0 0 H 2 5 0 2 0 0 »e—e e e- -e- -o L U 1 5 0 O Z . O O 1 0 0 _ _ l < L J ' 5 0 • • ~4~- #-o o o o o ARSENIC CHROMIUM . • • • • C O P P E R 0 i i i i i i i i i i i i i i i i i i i I i i i i i i i i i i i i i i i i i i i i i i i i i i i i i 0 5 , 10 15 2 0 2 5 3 0 3 5 4 0 ' 4 5 5 0 TIME ( h r s ) Figure 4.3 Adsorption of 0.02% CCA by WH B h l 1 soil as a function of time 52 40000 TIME (h r s ) Figure 4.4 Adsorption of 2% C C A by WH B h ( 1 soil as a function of time 53 C P zs O h— < cr LU o z o o < h— LU 350 3 0 0 -2 5 0 -200 -1 5 0 -100 50--e-. A _ •A A -A o o o o o ARSENIC AAA-A^ CHROMIUM • • • • • COPPER 0 — i i i | i i — i — i — | — i — n — i | i i i i | i—rn—r~|—i—i—i—i—r~r 0 10 15 20 25 30 i I i i i i i i i i i i i i i i 35 40 45 50 TIME (hrs ) Figure 4.5 Adsorption of 0.02% CCA by LA B soil as a function of time 54 .1 0000 < 2 0 0 0 -o ~t i i i i | i i i i | i i i i | i i i i | M i i | i i i i | i i i i | i i i i | i i i i | i i 0 5 10 15 . 20 2 5 3 0 35 40 45 50 TIME (h r s ) Figure 4.6 Adsorption of 2% C C A by LA B soil as a function of time 55 TABLE 4.1 P E R C E N T METAL ADSORBED vs SOIL/SOLUTION RATIO MILNER B - 0.02% Adsorbed Metal Percent Adsorbed Metal grams mL As Cr ' Cu As Cr Cu s o i l s o l u t i o n Ratio (ug/g) (ug/g) (ug/g) % % % 10 40 1:4 171.1 136.5 128 100 64. .7 99.7 5 40 1:8 366 . 6 179.2 250 100 43. .3 99 2 40 1:20 837 . 4 225 538 99.5 21; .7 85.2 1 40 1:40 1615 246.4 691.2 96 11, .9 54.7 MILNER B - 0.10% Adsorbed Metal Percent Adsorbed Metal grams mL As Cr Cu As Cr Cu s o i l s o l u t i o n Ratio (ug/g) (ug/g) (ug/g) % % % 10 40 1: 4 814 522 587 99 . 8 51 . 7 95 . 6 5 40 1:8 1671 774 945 99 .5 37 . 1 75 2 40 1:20 3941 1124 1182 93 .8 21. .5 37 .5 1 40 1: 40 6067 1516 1302 72 .2 14 .  5 20 . 7 LADNER Ap - 0.02% Adsorbed Metal Percent Adsorbed Metal grams mL As Cr Cu As Cr Cu s o i l s o l u t i o n Ratio (ug/g) (ug/g) (ug/g) % % % 10 40 1:4 164 .4 94.5 125.1 96.1 44, .8 88. ,7 5 40 1:8 299.9 115.6 244.9 72.4 27 . , 9 95. , 1 2 40 1:20 555.4 180.6 600.8 66 17 , . 4 96, .8 1 40 1:40 755.6 198.4 1120.4 44.9 9, .6 97 , . 4 LADNER Ap - 0.10% Adsorbed Metal Percent Adsorbed Metal grams s o i l mL s o l u t i o n Ratio As (ug/g) Cr (ug/g) Cu (ug/g) As % Cr % Cu % 10 40 1:4 737 285 597 90 .3 28 .2 97 , 2 5 40 1:8 1223 403 1163 72 .8 19 .3 92 , .3 2 40 1:20 2055 1028 2170 48 .9 19 .7 68 , .9 1 40 1:40 2314 1236 2313 27 .5 11 .8 36. .7 56 solution concentration, particularly for arsenic and copper. Low soil/solution ratios are often used in batch equilibration experiments for the purpose of establishing a maximum adsorption capacity. In the present study, with three metals as co-solutes, a low soil/solution ratio (eg., 1:40) may result in significant solute-solute rather than solute-adsorbent interactions. In addition, low soil/solution ratios are not realistic in environmental terms, where under natural ground water conditions the soil/solution ratio rarely exceeds 1 (Fritz and Hall, 1988). On the basis of the data shown, it was decided to use the 1:4 soil/solution ratio in all subsequent batch equilibration experiments. This soil/solution ratio would serve to maximize solute-sorbent reactions as well as conform to environmentally relevant conditions. 4.1.3 CCA Batch Equilibration Experiments Ten soil samples were equilibrated with varying concentrations of chromated copper arsenate to evaluate adsorption of arsenic, chromium and copper. The results are shown in Figures 4.7 through 4.15. Figures 4.7 to 4.9 show the adsorption of arsenic, chromium and copper by the three soils which differ the most in organic matter content, hydrous oxide content and clay content. These three soils were chosen for study first because on the basis of their varying characteristics they would be expected to show the greatest variation in adsorption of the three metals. Figure 4.7 clearly shows differences in attenuation of arsenic by the three soils. Both the Langley Ap and Whonnock Bhf, soils appeared to have a very high affinity for arsenic while the Langley B soil adsorbed significantly less arsenic from solution. The Langley B soil appeared to reach a maximum adsorption capacity at the 2% CCA concentration, after which arsenic was slightly desorbed. Arsenic also appears to be approaching a maximum sorption concentration in the Langley Ap soil. Chromium adsorption by the three soils is shown in Figure 4.8. As with arsenic adsorption, the Langley Ap and Whonnock Bhf, soils adsorbed more chromium than the Langley B soil. However, chromium desorption is evident at high solute concentration. The amount of chromium adsorbed by the soils was less than half the amount of arsenic adsorbed. Figure 4.9 illustrates the adsorption of copper by the three soils. Copper adsorption follows the same trend as arsenic and chromium 57 3 5 0 EQUILIBRIUM CONCENTRAT ION ( u m o l / m L ) Figure 4.7 Arsenic adsorption isotherm for CCA equilibrated soils 58 2 0 0 1 7 5 1 5 0 o E 3 1 2 5 g 1 0 0 cr o GO 7 5 Q < o 5 0 2 5 -0 - i P f - r C K K ^ K ) LANGLEY Ap • QD-Q-Q WHONNOCK Bhf 1 A A ^ A - A LANGLEY B 0. i 1 1 1 i i " I i i i i | i i i — i — r ~ i — i — i — i — | — i — i — i — r — i — i — i — i — i — f A - , — r 2 0 4 0 6 0 8 0 1 00 120 1 40 EQUILIBRIUM CONCENTRAT ION ( u m o l / m L ) Figure 4.8 Chromium adsorption isotherm for C C A equilibrated soils 59 1 2 0 H 0 5 10 15 2 0 2 5 3 0 3 5 4 0 4 5 5 0 5 5 6 0 EQUILIBRIUM CONCENTRATION ( u r n o l / m L ) Figure 4.9 Copper adsorption isotherm for C C A equilibrated soils 60 with respect to soil affinity. Pronounced desorption of copper is noted at the highest concentration while both the Langley Ap and the Whonnock Bhf, soils show slight desorption. The results indicate that the initial choice of the Langley Ap, Langley B, and Whonnock Bhf, soils which showed the greatest variability in organic matter, hydrous oxide and clay content was appropriate for the purpose of illustrating extremes in attenuation of metals. The approach was then extended to include seven additional soils. Results for arsenic, chromium and copper adsorption by all the surface and subsurface soils studied are shown in Figures 4.10 to 4.15. Table 4.2 lists the percent metal adsorbed by the soils at initial CCA concentrations of 0.02% and 2%. Adsorption of arsenic by all surface and subsurface soils is shown in Figure 4.10 and Figure 4.11. Arsenic attenuation is highest in the Langley Ap and Whonnock B soils and lowest in the Langley B soil. All these soils, with the exception of the Whonnock Bhf horizons, appear to be reaching a maximum capacity for arsenic adsorption. This result is confirmed in Table 1 where it can be seen that the percent arsenic adsorbed from dilute CCA concentrations is higher than from more concentrated CCA solution for all soils except the Whonnock Bhf horizons. Chromium adsorption by the surface and subsurface soils shows a marked trend toward desorption at high concentrations for all soils except surface soils low in organic matter such as Cloverdale Ap, Grevell Ap and Ladner Ap (Figures 4.12-4.13). These soils showed the most consistent uptake of chromium over the concentration range studied. All of the soils showed a higher affinity for chromium at dilute CCA concentrations (Table 4.2). The Langley Ap and Whonnock Bhf1 and Bhf2 soils removed the largest amount of copper from solution (Figures 4.14 and 4.15). Desorption of copper at high initial concentrations occurred in the Langley Ap, Langley B, Marble Hill B, Cloverdale B and Whonnock Bhf1 soils. With the exception of the Whonnock Bhf1 and Bhf2 horizons, which contain a high amount of organic matter, the surface soils adsorbed copper to a greater extent than the subsurface soils. 61 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.10 Arsenic adsorption by surface soils equilibrated with CCA 4 5 0 62 o o o o o Whonnoc k Bhf 1 •***-*-* Whonnoc k Bh f 2 • o-ee-e Mnrb leh i l l b B - B D B H I ] Mi lner B A-AA-A-A L ang l ey B <H><H>0 C l o ve rda l e B 0-f i i i i i i i i i i i i i i i i i i i | i i i i i i i i i | i i i i i i i i i i i i i i i i i i i | i i i i i i i i i 0 20 40 6 0 8 0 100 120 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.11 Arsenic adsorption by subsurface soils equilibrated with CCA 63 150 cn O E Q LxJ m o Q < o 100H 50 H 0 y I I I I I I I I I | I I I I I I I I I | I I I I I I I I I | I I I I I I I I I | I I I I I I I I I | I I I I I I I I I | I I I I I I I I I 0 20 40 60 80 100 120 EQUILIBRIUM C O N C E N T R A T I O N ( u m o l / m L ) 140 Figure 4.12 Chromium adsorption by surface soils equilibrated with CCA 64 225 / / / o o o o o Whonnock Bhf 1 *•**-*-* Whonnock Bhf2 Marblehil l B Mcra-a Milner B AAA-A-A Langley B Cloverdale B 0 ~yi i i i i i i i i | i i i i i i i i i | i i i i i i i i i | i i i i i i ) i i | i i i i i i i i i | i i i i i i i i i 0 50 100 150 200 250 • 300 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.13 Chromium adsorption by subsurface soils equilibrated with C C A 65 CP O E Q LU m cr o LO Q < o 120H 100 o o o o o LANGLEY Ap • B o o f l LADNER Ap A A * * * CLOVERDALE Ap • • • • • GREVELL Ap "i ' i i I i i i i I i i i — i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — i — i — i — i — i — r — r 10 20 30 40 50 60 70 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.14 Copper adorption by surface soils equilibrated with C C A 66 2 0 0 o o o o o Whonnock Bhf 1 Whonnock B h f 2 »-»a«»-o Marb leh i l l B o a o & Q Mi lner B A A A - A - A Lang l ey B 00<K^ C l ove rda l e B S—• — g 11 i i i i i i i i | i i i i i i i i i i i i i i i i i i i | i i i i i i i i i [ i i i i i i i i i | i i i i i i i i i i i i 0 20 4 0 6 0 8 0 100 120 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.15 Copper adsorption by subsurface soils equilibrated with CCA 67 TABLE 4.2 Percent metal adsorption by soils at i n i t i a l CCA concentrations of 0.02% and 2% Arsenic Soil Name 0.02% 2% LA Ap 99 84 LA 13 99 42 G Ap 97 57 CT Ap 96 55 L Ap 95 49 CT B 98 48 MH B 99 69 MR B 99 86 WH Bhfl 99 97 WH Bhf2 100 99 Chromium Copper 0.02% 2% 0.02% 2% 71 33 99 66 33 9 99 29 56 22 99 38 48 21 99 33 45 20 99 33 31 17 98 30 44 21 99 28 72 22 99 29 71 37 99 62 91 42 99 57 68 The adsorption isotherm data were fitted to the Freundlich and Langmuir equations to determine which model best described the data. The Freundlich parameters 1/n and k, as well as the Langmuir parameters b and k are tabulated in Table 4.3. The Freundlich parameter 1/n is equal to the slope of the regression line. The Freundlich k parameter was later used to determine correlation coefficients of soil properties with adsorption of the metals by the soils. The coefficients of determination (r2), for the relationship Log x/m and Log C for the Freundlich equation and C/x/m and C for the Langmuir equation, are also shown (Table 4.3). The Freundlich equation described the experimental results well over the entire CCA concentration range used. Table 4.3 suggests that the data also conforms to the Langmuir equation. However, plots of the linear form of the Langmuir equation appeared to resolve not into one straight line but rather into two or more straight lines at different concentration ranges. This indicates that the simple Langmuir equation is inadequate to describe the adsorption results over the entire concentration range studied. Similar results have been reported by several investigators (Posner and Bowden, 1980; Holford, 1982; Travis and Etnier, 1981) and have been explained on the basis of a nonconstant energy of adsorption and the presence of different surface sites for adsorption (Sposito, 1982). Holford (1982) suggested that the Freundlich constant k functions as an index of sorption capacity of soil surfaces for solute species. Comparison of sorption capacity using this parameter agrees quite well with the results obtained for the adsorption isotherms and is therefore a useful index of adsorption capacity of a soil for a given solute. A multiple regression analysis was conducted to predict the best combined effect of soil properties on arsenic, chromium and copper sorption parameters. The best four variable equations for Freundlich k values for arsenic, copper and chromium are shown in Table 4.4. A simple correlation matrix is shown in Table 4.5. The matrix reports correlation coefficients for simple regression of one variable to another. The value represented in the matrix (r) when squared indicates the variation of the value of one variable as predicted by the other. For example, the correlation TABLE 4.3 Freundlich equation and Langmuir equation coefficient for the adsorption of arsenic, chromium and copper by 10 soils. ARSENIC S o i l Name Freundlich Parameters Langmuir Parameters 1/n k r k b r LA Ap 0 .738 72 . 4 4 0 .99 0.99 255 .3 0 .99 B 0 . 455 21 .00 0 . 9 3 0 . 2 1 3 77 .5 0 .98 L Ap 0 . 5 5 5 24 . 5 5 0 . 9 9 0.256 188 . 7 0 . 9 4 G Ap 0 .563 18 .20 0 .99 0.182 172 .4 0 .92 CT Ap 0 .555 16 .59 0 .99 0.192 151 .5 0 .92 CT B 0 .476 20 .89 0 .99 0.33 130 .9 0 .95 MHB 0 . 4 6 1 46 .78 0 .98 1 . 0 4 196 . 1 0 .99 MRB 0 . 4 5 6 87 . 1 0 0 . 9 7 1.8 2 7 7 . 1 0 . 9 9 W H B h f 1 0 . 5 5 5 200 . 44 0 .98 3 .28 3 1 7 . 7 0 .98 W H B h f 2 0 .513 3 1 3 . 3 3 0 .99 8 .9 3 7 0 . 5 0 .95 CHROMIUM S o i l Name Freundlich Parameters Langmuir Pacnmotnc.i 1/n k r k b r LA Ap 0.515 9.91 0.97 0.182 98 .3 0.92 B 0 .707 1. 64 0.97 -0.012 -99 .7 0.11 L Ap 0.573 5.20 0.97 0.029 124 .1 0.84 G Ap 0 .670 3.87 0 .98 0 .027 135 .6 0.88 CT Ap 0 .795 2.26 0 . 99 0.009 205 .1 0.43 CT B 0.705 1.85 0.97 0.07 38 .8 0.81 MHB 0.714 3.08 0.99 0.031 105 .7 0.90 MRB 0.608 6.37 0.99 0.093 97 .5 0.97 WHBhf1 0.550 9.64 0.98 0.129 118 .4 0.93 WHBhf2 0 .582 16.60 0.99 0.11 203 .3 0.97 COPPER S o i l Name Freundlich Parameters Langmuir Parameters 1/n k r k b r LA Ap 0.329 50 .12 0. 95 19.02 109. 9 0. 99 B 0 . 454 9 .77 0. 63 -0.352 20. 3 0 . 88 L Ap 0.508 18 .62 0. 92 0.79 86. 6 0. 99 G Ap 0.514 17 .38 0. 92 . 0.9 80. 6 0. 99 CT Ap 0.521 16 . 60 0. 89 0.46 89. 8 0. 95 CT B 0 .504 12 .27 0 . 88 0.97 58 . 7 0 . 98 MHB 0.414 16 .98 0. 88 1.19 60. 4 0. 99 MRB 0.331 18 .07 0. 97 0.66 63. 1 0. 98 WHBhf1 0 .409 30 .20 0 . 99 1.21 109. 8 0 . 98 WHBhf2 0 .375 36 .31 0 . 99 0.408 153. 6 0 . 95 70 TABLE 4.4 M u l t i p l e r e g r e s s i o n equations r e l a t i n g a r s e n i c , chromium and copper a d s o r p t i o n c a p a c i t y parameters to p r o p e r t i e s of the s o i l s s t u d i e d . M u l t i p l e Regression Equation A r s e n i c F r e u n d l i c h k = 30.61 - 2.64(%0C) - 22.78(%Fe(d)) + 136.5(%A1(d)) - 0.154(%c1 ay) 0.95 Chromi urn F r e u n d l i ch k - 3 . 1 7 - 0 . 0 2 2 U 0 C ) + 7 . 2 7 ( « F e ( o ) ) + 3 . 7 2 U A 1 (o) ) + 0 . 0 3 4 ( % c l a y ) 0.82 Copper F r e u n d l i ch k 5.87 + 2.69(%0C) + 7.14(%Fe(d)) + 0 . 8 3 U A 1 (d)) + 0.077(%clay) 0.79 %0C = % o r g a n i c carbon. %Fe(o) = % ammonium o x a l a t e e x t r a c t a b l e i r o n %Fe(d) = % d i t h i o n i t e e x t r a c t a b l e i r o n %A1(o) = % ammonium o x a l a t e e x t r a c t a b l e aluminum % A l ( d ) = % d i t h i o n i t e e x t r a c t a b l e aluminum TABLE 4.5 Simple c o r r e l a t i o n matrix - r e l a t i o n s h i p between the F r e u n d l i c k k parameter and s e l e c t e d s o i l p r o p e r t i e s F r e u n d l i c h K S o i l P r o p e r t y A r s e n i c Chromi urn Copper % Organic carbon 0.670 0.864 0. 909 Fe ( o )1' 0. 463 0.727 0.749 Fe (Pr - 0.4 14 0. 585 0.799 Fe ( d )3' 0. 547 0.489 0.515 Al (o) 0. 941 0.776 0.518 Al (p) 0.953 0.814 0. 628 Al (d) 0.977 0.775 0.533 % c l a y -0.169 -0.094 0 . 187 s u r f a c e area -0.293 -0.360 -0.199 ammonium o x a l a t e e x t r a c t a b l e Fe and Al pyrophosphate e x t r a c t a b l e Fe and Al d i t h i o n i t e e x t r a c t a b l e Fe and Al 72 coefficient r for the relationship between the Freundlich k value for arsenic and the percent ammonium oxalate extractable aluminum is 0.941. This indicates that 88.5% (r2) of the variation in the arsenic Freundlich k parameter for the soils studied can be predicted by the percent ammonium oxalate extractable aluminum. Arsenic Results of the regression analysis showed that approximately 95% of the variation in the Freundlich k values for arsenic could be predicted by a multiple regression equation which includes organic carbon, dithionite extractable Fe, dithionite extractable Al, and clay as independent variables. Table 4.5 shows that dithionite extractable aluminum is the major soil component controlling the adsorption of arsenic in the soils studied. Ammonium oxalate extractable and pyrophosphate extractable aluminum also correlated well with the adsorption capacity of soils for arsenic. The clay content and surface area of the soils were the least correlated soil properties with arsenic retention. The Freundlich k values for arsenic ranged from 16.6 for the Cloverdale Ap horizon to a high of 313.33 for the Whonnock Bhf2 horizon (Table 4.3). On the basis of the Freundlich k values, arsenic retention by the soils studied can be divided into two groups: 1) low adsorption capacity: L Ap > LA B > CT B > G Ap > CT Ap 2) medium to high adsorption capacity: MH B < LA Ap < MR B « WH Bhf1 < WH Bhf2. The variable charge components of the soil, Fe and Al-oxides , and to a lesser extent organic matter were the soil properties most closely correlated to arsenic retention in the soils studied. Arsenic (V) in soil solution exists as the oxyanion with the predominant species at pH 2 to 6 being the monovalent anion, H 2As0 4 Many researchers have reported significant correlation between arsenic adsorption and oxalate-extractable amorphous Al and Fe compounds (Jacobs et al, 1970; Wauchope, 1975; Anderson et al, 1976; Livesey and Huang, 1981; Fordham and Norrish, 1983). 73 The adsorption of arsenic has also been investigated on Al and Fe oxides such as goethite and gibbsite (Hingston et al, 1971; Anderson and Malotky, 1979). The affinity of soils with elevated Fe and Al hydrous oxides, in particular the Whonnock Bhf, and Bhf2 horizons, is extremely high for arsenic (Figure 4.11) The high adsorption capacity and strong affinity of these soils for arsenic suggest a specific adsorption mechanism is responsible for retention. The mechanism of specific arsenate adsorption on Al and Fe oxyhydroxides is considered to be ligand exchange with surface hydroxyls and/or surface aqua groups (Hingston et al, 1971; Anderson et al, 1976; Bohn, 1976). In general, anion adsorption increases as the solution pH decreases and the amphoteric substrates become more positively charged. Hendershot (1978) measured the PZC's (point of zero charge) of a number of variable charge soil horizons in B.C. and reported a range of 3.3 to 6.1. One of the soils used in this study, Marblehill B, was measured by Hendershot (1978) to have a PZC of 4.83. PZC values for model adsorbents such as Fe 2 0 3 and Al203 have much higher PZC's of 8.5 and 6.9, respectively. It is not surprising then that those soils without an appreciable variable charge component did not adsorb arsenic to a great extent. Organic matter also correlated positively with arsenic adsorption. As stated in the literature review, however, very few studies have investigated adsorption of arsenic by organic matter. Thanbalasingam and Pickering (1986) reported maximum adsorption of As(V) by two humic acids to occur in the pH 5 to 6 range with adsorption decreasing in alkaline media, possibly because of dissolution of the humic acids. The authors suggested that the major anion retention sites are amino groups. Although organic matter appears to play a role in arsenic adsorption in this study its importance relative to the hydrous oxide component of the soil is unclear. Examination of the Freundlich k value in Table 4.3 for the soil with the highest organic matter content, Langley Ap, shows that this soil ranks only fourth in sorption capacity for arsenic. The Langley Ap soil horizon is not enriched in Fe- and Al- hydrous oxides to the same extent as the podzolic soils used in the study. In addition, the high organic matter content of the LA Ap soil sample together with its fine texture may cause masking or blocking of hydrous oxide binding sites in this soil. 74 The low adsorption capacity soils are characterized by a high clay content and low organic matter and hydrous oxide content, with the exception of the Grevell Ap soil which is low in all soil components. The predominant clay minerals in the soil samples are vermiculite and montmorillonite (Luttmerding 1981) which do not have an appreciable anion exchange capacity, even at acidic pH conditions. Therefore, the lack of correlation between arsenic adsorption and clay content is not surprising. Many investigators have reported a positive correlation between clay content in soils and arsenic retention (Frost and Griffin, 1977; Huang, 1980; Livesey and Huang, 1981; Fordham and Norrish, 1983). The presence of sesquioxide coatings, as discrete colloidal precipitates on clay surfaces is a possible explanation for these results. Therefore, it is concluded that the hydrous oxide component of the soil is the most important in the removal of arsenic from soil solution. Chromium The Freundlich k values for chromium (Table 4.3) ranged from 1.64 for the Langley B horizon to 16.60 for the Whonnock Bhf2 horizon. The range of Freundlich k values is not as great for chromium adsorption by the soils as compared to arsenic adsorption. Figures 4.12 and 4.13 show only moderate affinity of the soils for chromium retention. Chromium retention by soils, can be divided into 3 groups: 1) low adsorption capacity: LA B < CT B < CT Ap 2) moderate adsorption capacity: MH B < G Ap < L Ap < MR B 3) relatively high adsorption capacity: WH Bhfl < LA Ap < WH Bhf2 The relationship between soil properties and adsorption of chromium is not as clear as that for arsenic. Desorption of chromium at high concentration lowers the Freundlich k sorption capacity value, rendering this method of comparison for adsorption characteristics, less effective than with arsenic. To better delineate the soil properties controlling the removal of chromium from solution, a multiple regression analysis was performed using the Freundlich k values as the dependent variable and various soil properties as the independent variables. The best four variable equation for Freundlich k values is shown in Table 4.4. A correlation matrix is provided in Table 4.5. 75 Results of the regression analysis indicated that organic carbon content was the most important soil component influencing the removal of chromium from soil solution. The attenuation of chromium in soils is complicated by the reduction of Cr(VI), the chromium species in CCA solution, to Cr(lll) under mildly oxidizing, acidic conditions in the presence of electron donors such as dissolved iron or organic matter. The environmental and substrate requirements for Cr(VI) reduction were all present in the soil/solution reactions studied. Therefore, with the exception of those soils with very low organic matter and sesquioxide content(LA B, CT B) reduction of Cr(VI) to Cr(lll) was likely to have occurred. Cr(lll) is present as the trivalent cation Cr3*, and its hydrolysis products in soil solution, while the oxyanions HCr04" and Cr042" are the dominant Cr(VI) species in soil solution. The geochemical behaviour of Cr(lll) and Cr(VI) in the subsurface is markedly different (Rai et al., 1984). Cr(lll), as a cationic species is strongly sorbed by clays, organic matter and hydrous oxides through specific adsorption and ion exchange reactions (Bartlett and Kimble, 1976a, 1976b; Hem, 1977; Griffin et i al., 1977). In addition, solution concentrations of Cr(lll) in soil-water systems are largely controlled by the low solubilities of Cr(OH)3 and possibly other divalent metal chromium compounds, particularly above pH 5 (Rai et al., 1986). Stollenwerk and Grove (1985) reported that attempts to desorb chromium(VI) adsorbed by alluvium were less successful the longer the contact time before desorption was initiated. This result suggests that either the Cr(VI) had slowly become part of the Fe hydroxide coatings on the alluvium or that it had reduced to Cr(lll) and co-precipitated with Fe-hydroxide. Therefore, in soils with a high amount of organic matter increased attenuation of chromium as the reduced species is expected to occur. The results suggest that the fairly high adsorption capacity for chromium in soils with an appreciable organic matter content, at dilute concentrations of CCA solution (Table 4.2), is most probably due to the reduction of Cr(VI) to the trivalent cationic species. At higher solution chromium concentrations, the sorption capacity of the soils was observed to decrease, to the point of desorption in some cases. The concentration of Cr(VI) at high CCA capacity of the soils. Therefore, direct correlation of chromium(VI) adsorption with soil properties is not easily made. Rather, soil factors influencing chromium(VI) reduction to 76 the more stable, less soluble, and more strongly adsorbed trivalent species should be evaluated in attempts to explain the retention of chromium by soils. Although soil contamination in the vicinity of CCA wood preservation facilities is not expected to approach the higher chromium concentrations used in this study, such contamination may occur as a result of an accidental or spill release. Under such conditions the Cr(VI) would remain the predominant species in soil solution and retention of Cr would have to be predicted on this basis. Hydrous oxides, in particular amorphous iron oxide, have been reported as effective adsorbents of chromate anions at pH levels of 2 to 7 (Rai et al, 1984; Zachara et al, 1988). As with many protolyzable oxyanions, chromate adsorption increases with a decrease in pH as a result of protonation of surface binding sites. Hydrous oxides are thought to bind the chromate anion, via surface coordination, however these surface sites do not appear to be specific for chromate. The presence of competing anions such as common groundwater constituents (H4Si04, S 0 4 2 C 0 2 (g)) and phosphate anions in soils have been shown to significantly reduce Cr0 4 2 adsorption (Stollenwerk and Grove, 1985; Zachara eft si., 1989). It would be reasonable to suspect that the other anionic constituent of CCA, the arsenate anion, would also effectively out-compete the chromium anion for adsorption sites. The desorption of chromium observed at high concentration may be evidence for this competitive effect. The lack of desorption noted for the low hydrous oxide and low organic matter soils (Figure 4.12) may be explained by the lack of arsenate competition for the soil surface binding sites. The batch experiments using individual metals will shed more light on this subject. Unfortunately, it is difficult to evaluate the contribution of organic matter to adsorption of Cr(VI) in soils, apart from its role in reduction reactions. The LA Ap soil appeared to adsorb Cr as effectively as the soils with both a high hydrous oxide and organic matter content. Therefore, the Cr species adsorbed in the LA Ap soil was likely the reduced Cr(lll) form while the WH Bhfl soil, with its high hydrous oxide content, had the potential to adosrb both the trivalent and hexavalent Cr species. Evidence for adsorption of Cr(VI) by the soil organic matter component was not found in the literature. In fact, organic matter was deliberately excluded from many studies on chromate 77 adsorption to avoid complications inherent in attempts to interpret the adsorption of both a cationic and anionic chromium species. Copper The Freundlich k values for copper (Table 4.3) ranged from 9.77 for the Langley B soil horizon to 50.12 for the Langley Ap soil horizon. Referring to Table 4.3, it is interesting to note that the correlation coefficients for copper adsorption were better for the Langmuir than the Freundlich equation, although both equations adequately described the adsorption data. For this reason, the Langmuir parameter b was used to compare the sorption capacity of the soils for copper. Although both the Freundlich k parameter and the Langmuir b parameter are considered to represent extensive properties (adsorption capacity) of the adsorbent for the adsorbate (Holford, 1982), the values of the Freundlich k tend to be less than half those of the Langmuir b. Nevertheless, the Freundlich k values have been shown to be highly correlated with sorption capacity, especially over wide concentration ranges. Therefore, provided the two sorption parameters are not used for direct comparison with each other, it is appropriate to use the parameter which best fits the experimental data. The Langmuir b values for copper ranged from 20.3 for the Langley B soil horizon to 153.6 for Whonnock Bhf1 soil horizon. All the soils showed very high affinity for copper at dilute CCA concentrations (Table 4.2). Copper adsorption appeared to reach a maximum capacity in most soils at high concentrations of CCA solution (Figures 4.14 and 4.15). Three of the soils, the Langley Ap, and Whonnock Bhf 1, were observed to desorb copper at the highest concentration of CCA used (3%). Both the Langley Ap and the WH Bhf1 soils exhibited high affinity for copper as shown in Figures 4.14 and 4.15. Therefore the sudden desorption noted cannot be explained on the basis of adsorption/desorption miechanism. However, the problems with desorption of copper do not appear to be reflected in the Langmuir b parameter. 78 On the basis of the Langmuir b values, adsorption of copper by the soils can be divided into 2 groups: 1) low-moderate adsorption capacity: LA B « CT B < MH B < MR B < G Ap < LAp < CTAp 2) high adsorption capacity: WH Bhf1 < LA Ap < WH Bhf2. To maintain consistency with the previous statistical analyses a multiple regression analysis using the Freundlich k values (Table 4.3) was conducted to predict the best combined effect of the soil properties on copper sorption parameters. The best four variable equation for the Freundlich k values in shown in Table 4.4 and a simple correlation matrix is provided in Table 4.5. The regression analysis indicated that organic carbon was the most important soil component for copper i adsorption from CCA solution. In aqueous solutions of pH values less than 7 the dominant copper species is Cu 2*. However, the preferred copper species for adsorption by many absorbents is the first hydrolysis product of Cu 2*, Cu(OH)* (McBride, 1981; Davis and Leckie, 1978). The pK for this reaction is approximately 8. Adsorption of copper, therefore, is strongly dependent on pH, occurring mostly rapidly between pH 4.5 and 5.5 (James and Barrow, 1981). This may explain the very high adsorption capacity (99%) of all the soils studied for copper at dilute CCA concentration (Table 4.2), where the pH was approximately 5.3. The pH of the equilibrium solution at the 2% CCA concentration, on the other hand, varied between 2.5 and 2.9 and the adsorption capacity of the soils was observed to decrease significantly for all soils except those with a high organic matter component (Langley Ap, Whonnock Bhf 1 and 2) (Table 4.2). The importance of soil organic matter on the adsorption of copper is well documented in the literature (Pickering, 1979; James and Barrow, 1981; McBride, 1981, 1989). The carboxylic, carbonyl and phenolic functional groups are considered to be the most important bonding sites for copper on particulate organic matter (McBride, 1978). At pH values below 6, Cu complexation by organic ligands to form soluble organic complexes becomes important (McBride, 1989). These soluble complexes may affect the adsorption of copper on clay and hydrous oxide binding sites. 79 McBride (1981) showed that adsorption of Cu 2* by montmorillonite was inhibited to various extents by EDTA and citrate Cu complexes. The author also discussed the variable effect of organic ligands on Cu adsorption by hydrous oxide components of the soil. It is possible that formation of soluble Cu-organic complexes may have enhanced or inhibited adsorption of Cr(lll) on soil adsorption sites. This will be discussed in section 4.1.5. The relative unimportance of layer silicates in adsorption of copper, especially at high concentrations, is evident in the low adsorption capacity of soils enriched in clay, LA B, CT B (Figure 4.14, 4.15; Table 4.2, 4.3.). According to McBride (1981) and Rai et al., (1984) specific adsorption of Cu by layer silicates is pH dependent and below pH 5 the dominant adsorption mechanism for Cu 2* on clays is simple ion exchange. The results of this study indicate that the decrease in attenuation capacity of soils at high CCA concentrations and low pH may be explained by a non-specific cation absorption mechanism such as ion exchange. The pH dependence of copper adsorption is particularly important for the hydrous oxide component of the soils. Cu 2* is strongly sorbed by hydrous oxides at pH values approaching the PZC of the substrates. Mn-oxides have been shown to have a high affinity for Cu 2*, possibly because of their relatively low PZC (~pH 2) as compared with Fe- and Al-hydrous oxides (McKenzie, 1980; Golden et al., 1986). Consequently, unless the hydrous oxide component of the soil is enriched in Mn-oxides or the solution pH is above pH 5, copper sorption by hydrous oxides will proceed by a non-specific ion exchange mechanism. 4.1.4 Desorption of As, Cr and Cu The rate of desorption of arsenic, chromium and copper from soils previously equilibrated with varying concentrations of CCA solutions was studied. Five desorption steps were carried out over a period of 18 hours. The results for the highest concentration of CCA used in the experiment (2%) are shown in Table 4.6. 80 TABLE 4.6 Desorption r e s u l t s f o r an i n i t i a l c o n c e n t r a t i o n of 2% CCA e q u i l i b r a t e d with the LA Ap and WH B h f l s o i l h o r i z o n s Langley Ap Time % Metal Desorbed (hrs) As Cr Cu 0 0.0 0.0 0.0 CO 5.4 31.4 11.4 10 8.7 44.7 16. 9 14 11.4 50. 7 20. 0 16 12.4 52.7 20.6 18 14. 3 54.9 22.0 Whonnock B h f l Time % Metal Desorbed (hrs) As Cr Cu 0 0.0 0.0 0. 0 8 1.2 38.8 17 . 6 10 2.1 54 .4 25. 1 14 2.8 62.3 29 . 4 16 . 3.6 66. 2 31. 9 18 4.3 68. 3 33. 5 81 Only a small amount of the metals was released by the soils at more dilute initial CCA concentrations, indicating that the extent of release is dependent on the amount of retained metal. This result also suggests that at lower concentrations the metal ions are retained by sites where they are more strongly held while at high concentrations the binding energy decreases (Amacher et al 1988). Their results support the results obtained for the adsorption experiments, showing that arsenic is the most strongly and specifically adsorbed of the metal constituents in CCA. The lack of desorption observed for arsenic, especially in the WH Bhfl soil, suggests that it may be irreversibly sorbed in the soil. Elkhatib et al (1984) reported a similar result for arsenite desorption in the A and B horizons of five West Virginian soils. The authors attempted to relate soil properties to desorption rate coefficients using multiple regression equations. The percent Fe 2 0 3 and pH were the soil properties most closely related to the arsenite desorption coefficients. The relationship may also explain the results of this experiment. The Whonnock Bhfl soil horizon contains an appreciable amount of iron and aluminum hydrous oxides. Chromium desorption was virtually the same in the two soils. The high organic matter Langley Ap horizon desorbed a little less chromium over the five desorption steps. This was probably because of its increased capacity to reduce the Cr(VI) to the less soluble, strongly adsorbed Cr(lll) species. However, the difference in the organic matter content of the LA Ap and WH Bhfl soil was very small, therefore the significance of this result is not clear. At CCA concentrations below 1%, Cr was very difficult to desorb, indicating reduction to the Cr(lll) species. At higher initial CCA concentrations a larger proportion of the chromium would be present as Cr(VI), therefore it is this species which is most readily desorbed. \ Table 4.6 shows that copper was also held strongly by the Langley Ap soil, confirming the importance of organic matter in adsorption of copper from soil solution. Based on this experiment the mobility of CCA components in soils with similar characteristics would follow the order: chromium > copper > arsenic. 82 4.1.5 Equilibration with Individual Metal Solutions Batch equilibration tests were carried out with six soils using individual metal solutions formulated from As 2 0 5 , Cr0 3 and CuCI2. The purpose of these tests was to determine if the adsorption parameters differed for the metals in single solute batch reactions versus multiple solute (CCA) batch reactions. Adsorption isotherms for the various soils with arsenic, chromium and copper solutions are shown in Figures 4.16 to 4.25. Table 4.7 shows a comparison of metal adsorption by soils equilibrated with 0.02% and 2% CCA solution and equilibrated with the equivalent concentration of individual metal solutions. Figure 4.16 illustrates the adsorption of arsenic by the Langley Ap soil from both individual metal and CCA solution. Arsenic appears to be adsorbed to a greater extent in the CCA equilibrated batch test. Two possibilities may explain this result. The first is the occurrence of solute-solute interactions in the form of co-adsorption of arsenic with chromium or copper in ,the CCA equilibrations. If this is the case a similar result to that obtained for arsenic would be expected for chromium and/or copper. Figures 4.17 and 4.18 show that this is not the case; both Cr and Cu are removed from solution to a greater extent from their respective individual metal solutions. A second possible explanation involves the pH of the equilibrating solutions. Arsenic(V) exists in soil solution as an oxyanion, therefore its adsorption by variable charge soil components is pH-dependent and increases with decreasing pH. In addition, the preferred species for adsorption is the monovalent anion H2Asb4" (pK2 = 6.94) which is present at higher concentration between pH 2 and 6. The pH of the 0.02% solution in the individual metal test was 6.6 which for the CCA equilibration it was 5.3. The pH difference was less pronounced between the 2% individual metal and CCA solutions, 3.3 and 2.9 respectively. In fact, adsorption of arsenic by the Langley Ap soil appears to be more similar for the two experiments at the 2% CCA and 2% individual metal concentrations. This result suggests that under less acidic conditions than those used on this study arsenic would be expected to be less strongly attenuated by similar soils and consequently be more mobile in the subsurface. 83 TABLE 4.7 Comparison of percent metal a d s o r p t i o n by s o i l s e q u i l i b r a t e d with 0.02% and 2% CCA s o l u t i o n and the e q u i v a l e n t c o n c e n t r a t i o n s of i n d i v i d u a l metal s o l u t i o n s S o i l Name LA Ap LA B MR B WH B h f l WH Bhf2 CCA 0. 02% 99 99 99 99 100 A r s e n i c As 0.02% 99 99 100 100 100 CCA 2% 87 42 86 97 99 As 2% 75 45 84 96 95 S o i l Name Chromium CCA Cr CCA Cr 0.02% 0.02% 2% 2% LA Ap LA B MR B WH B h f l WH Bhf2 71 3 3 72 71 91 77 42 68 97 89 33 9 22 37 42 52 24 31 49 56 Copper CCA Cu CCA Cu S o i l Name 0.02% 0.02% 2% 2% LA Ap 99 100 66 52 LA B 99 100 29 32 MR B 99 99 62 24 WH B h f l 99 99 n/a n/a WH Bhf2 99 99 n/a n/a 84 Figure 4.16 Arsenic adsorption isotherm for single solute and C C A equilibrated Langley Ap soil 400 85 20 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.17 Chromium adsorption isotherm for single solute and C C A equilibrated Langley Ap soil 86 2 0 0 cr> 150H "o Q 1 0 0 -LU 5 0 -1 o ° o o o S ing le M e t a l - L A Ap • CCA E q u i l i b r a t e d - L A Ap 0 - j 1 1 1 I I 1 1 1 1 1 I I 1 1 1 1 1 1 1 1 1 1 I I 1 1 1 1 1 i I I 1111111111111 I I 1111111 I I 1 11 1 11111 1111 1 11 i 111 I I 11 40 0 5 10 15 20 2 5 30 3 5 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.18 Copper adsorption isotherm for single solute and C C A equilibrated Langley Ap soil 'i 87 Adsorption of arsenic in the subsurface soils is shown in Figure 4.19. The amount of metal adsorbed appears to be very similar in both the CCA equilibrated and individual metal equilibrated soils. This result indicated that the sites responsible for arsenic sorption in these soils are specific for the arsenate anion and that neither copper or chromium effectively compete for these binding sites. One other result to note in Figure 4.19 is the increased arsenic retention in the LA B soil equilibrated with the single solute solution at high arsenic solution concentration. Adsorption of the three metal components in CCA solution would be largely via ion exchange at this high concentration with a nonspecific sorbent such as clay. Therefore competition for binding sites may be occurring in this soil at the high metal concentrations. The pH factor which may have contributed to the decreased adsorption capacity of the Langley Ap soil for arsenic from the single solution was not evident in the batch test with subsurface soils. Firstly, the pH differences at dilute concentrations of arsenic in solution were not as pronounced and secondly the subsurface soils are enriched with Fe- and Al-hydrous oxides. The strong affinity of arsenic adsorption for hydrous oxide surfaces may have overridden the pH effect. The amount of chromium removed from the single solute metal solution was higher than the amount removed from the CCA solutions for all the soils studied (Figures 4.17, 4.20, 4.22, 4.23; Table 4.7). The difference in removal of chromium from the two solutions is especially evident at high chromium i solution concentrations. The uptake of chromium by soils from the single metal solution and CCA solution is remarkably similar for all the soils with a high organic matter content (LA Ap, WH Bhf 1, i WH Bhf2). This indicates that organic matter is the principle component influencing the retention i of chromium by the soils. In addition, the results clearly show that the adsorption of chromium is significantly affected by the presence of competing ions. At high solution concentration, where chromium(VI) is likely to be the dominant species, the major competing anion is expected to be arsenic. Several investigators have reported that the adsorption of Cr(VI) generally decreases as the ionic strength of solution increases (Mayer and Schick, 1981; James and Bartlett, 1983; Stollenwerk and Grove, 1985; Zachara et al., 1987). Zachara et al. (1987) found that Cr(VI) 88 400 0 10 20 30 40 50 60 70 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.19 Arsenic adsorption isotherms for single solute and C C A equilibrated subsurface soils 89 4 0 0 20 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.20 Chromium adsorption isotherm for single solute and C C A equilibrated Whonnock Bhfl soil 90 Figure 4.21 Chromium adsorption isotherm for single solute and CCA equilibrated Whonnock Bhf2 soil 91 225 0 50 100 150 200 250 EQUILIBRIUM CONCENTRATION (umol/mL) Figure 4.22 Chromium adsorption isotherm for single solute and C C A equilibrated Milner B soil 92 Figure 4.23 Chromium adsorption isotherm for single solute and C C A equilibrated Langley B soil 93 150 0 j I I I I I M 1 1 | I I I I I I I I I | I I I I I I I I I | I I I I I I I I I | I M I I I I I I | I I l I M M I | M I I l I I l i 0 2 0 4 0 6 0 8 0 100 1 20 140 EQUILIBRIUM CONCENTRATION ( u m o l / m L ) Figure 4.24 Copper adsorption isotherm for single solute and C C A equilibrated Milner B soil 94 1 0 0 It -t i 0 i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i | i i i i i i i i i i i i i i i i i i i i i i i i i i i i i 0 1 0 2 0 3 0 4 0 5 0 6 0 EQUIL IBRIUM C O N C E N T R A T I O N ( u m o l / m L ) Figure 4.25 Copper adsorption isotherm for single solute and C C A equilibrated Langley B soil 95 adsorption was suppressed 50 to 80% in ion mixtures with solute concentrations typical of the subsurface-groundwater environment. The authors further stated that cationic solute constituents had little influence on Cr0 4 2 ' adsorption. Stollenwerk and Grove (1985) reported a decrease in chromate anion adsorption by alluvium in the presence of competing anions such as HP04 2' in particular and S0 4 2 ' to a lesser extent. Significant competition by the arsenate anion for the adsorption sites, therefore, will likely enhance the mobility of chromium in the subsurface. In a different vein, Benjamin (1983) investigated the effects of various cations, including Cu on sorption of Cr0 4 2 ' and Si0 4 onto Fe 2 0 3 H20. The possible interactions of adsorbate ions with each other and with the sorbent evaluated included: 1) competition for surface sites; 2) alteration of the surface electrical potential resulting in a change in the electrostatic interactions between the ions and the surface; or 3) precipitation of a new surface phase. Benjamin (1983) concluded that sorption of anions is unaffected by the presence of cations unless the cations form a surface precipitate, perhaps as the Cu(OH)2 solid species. When such a precipitate forms, sorption of the anions is increased. In the absence of a surface precipitate, the effect of cations on anion sorption was negligible, even at very high apparent adsorption densities of the metals. While such a precipitate was unlikely to have formed during this experiment due to the acidic conditions, it is possible that such a mechanism could operate under more neutral pH conditions. Frost and Griffin (1977) and McBride (1982) have shown that precipitation of Cu(OH)2 is an important mechanism for removal of copper from soil solution at pH values above 6.5. The amount of copper removed from solution by individual metal and CCA solutions could, unfortunately, only be evaluated for the Langley Ap, B and Milner B soil horizons. The initial copper concentration for the Whonnock soil equilibrations was too hjgh, and the solutions were found to contain precipitated copper. The results indicate that copper retention was not significantly enhanced using the single solute solutions for the Langley Ap and Milner B soils except at high initial concentrations. Copper adsorption by the Langley B soil from single solute and CCA solutions, as with the other two soils, was not significantly different at dilute initial copper concentrations (Table 4.7). However, significantly increased removal of copper from the single solute solutions occurred at an earlier point on the adsorption isotherm. 96 The adsorption of copper from single solute and CCA solutions is strongly influenced by the amount of organic matter the adsorbing soil contains, and to a lesser extent by the hydrous oxide component. The clay fraction of the soil did not correlate well with copper adsorption in the soils studied (Section 4.1.3). The results obtained for removal of copper from single solute versus CCA solutions is consistent with the relative affinities of the soils studied for adsorption of copper. In other words, soils with a high affinity and sorption capacity for copper will sorb copper to the same I extent regardless of the presence of competing ions (eg., Cr(lll)) or the ionic strength of the solution. However, at high adsorption densities, the available binding sites will be less specific for copper and competition with other ions in solution may occur. The adsorption isotherm data for the Langley Ap and Milner B soil illustrate this effect (Figures 4.18 and 4.24). The Langley B soil, which has a high clay content, has a correspondingly lower affinity and adsorption capacity for copper (Figure 4.25). Consequently ion-exchange becomes the dominant adsorption mechanism for copper retention by the Langley B soil at lower copper adsorption densities than observed for the Langley Ap and Milner B soils. The competitive effects of co-solutes appear to become more important when ion exchange is the dominant adsorption mechanism. Therefore, adsorption of copper will be relatively unaffected by co-dissolved solutes in soils with a high affinity for copper, except at very high copper concentrations (775 umol/mL in these experiments). Competitive effects from co-dissolved solutes will reduce the adsorption capacity of soils with a low affinity for copper at lower adsorption densities. 4.2. Adsorption Experiments - Column Testing 4.2.1 Migration of a Single Slug Dose of 2% CCA in a Soil Column A soil column experiment was conducted to investigate the vertical migration of arsenic, chromium and copper after application of a single slug dose of 2% CCA (the usual working strength concentration at wood preservation facilities) followed by intermittent additions of water. The experiment was designed to simulate the fate of chronic CCA contamination at wood preservation facilities. For example, storage 97 areas for freshly treated wood are often unpaved and not roofed, resulting in chronic drippage of CCA solution to the ground. Migration of the contaminant in the soil will depend on the amount of rainfall, the infiltration properties of the soil, and the adsorption capacity of the soil for CCA components. Contaminant sorbed or precipitated at the soil surface will also be subject to wetting and drying conditions. The results for Langley Ap and Whonnock Bhfl soil columns are shown in Figures 4.26 and 4.27. The amount of arsenic, chromium and copper extracted using an aqua regia digest for each depth is provided in Table 4.8. TABLE 4.8 The aqua regia extractable metal content of column soils after CCA application and subsequent leaching with deionized water. Depth (cm) 2.5 5 7.5 10 12.5 15 17.5 20 Langley Ap Metal Concentration As Cr Cu umol/g 26. 98 0 0 0 0 0 0 0 27.42 12.33 8.2 0 0 0 0 0 31.63 0 0 0 0 0 0 0 Whonnock Bhfl Metal Concentration As Cr Cu umol/g 31.49 0 0 0 0 0 0 0 37.29 8.69 3.82 3.28 47 48 19 05 29.43 0 0 0 0 0 0 0 The results show that both arsenic and copper were very strongly retained in the upper 2.5 cm of the soil column. Chromium, on the other hand, migrated to the 7.5 cm depth in the Langley Ap soil column and to the lowest (20 cm) depth of the Whonnock Bhfl column. The movement of the three metals relative to each other is consistent with the findings in the batch equilibration tests. Enhanced retention of the chromium in the Langley soil may be attributed to the greater potential for reduction possible in this soil due to its high organic matter content relative to the WH Bhfl soil. The migration of chromium in soil column tests has also been related to the effects of flux (Alesii et al, 1980). It is therefore possible that the increased movement of chromium in the WH Bhfl was influenced by the rate at which the water infiltrated the soil column. The WH Bhfl has sandy loam texture while the LA Ap soil has a silty loam texture. The rate at which water was added to the 35 TOTAL METAL CONCENTRATION (umol/g) 30 25 20 15 10 H 5 o i l 2.5 7.5 10 12.5 15 D E P T H (cm) ArBonlo Chromium Copper 17.5 — i — 20 Figure 4.26 Total metal concentration versus depth for the Langley Ap slug dose soil column. 40 TOTAL METAL CONCENTRATION (umol/g) 30 -20 10 -I 2.5 n 7.5 10 12.5 D E P T H (cm) Arsenic t*8%l Chromium H I Copper HI 15 17.5 — i — 20 Figure 4.27 Total metal concentration versus depth for the Whonnock Bhfl slug dose soil column. 99 columns during the course of the experiment was not controlled. Rather, two aliquots of 35 mL each were added to the top of the soil column and the water moved through at a rate determined by the hydraulic conductivity of the column soil. Therefore, the increased mobility of the chromium in the WH Bhfl column is probably related to both soil affinity and texture of the soil. The significance of these results for chronic, low-level contamination of wood preservation facility soils relates to the relative mobility of the the three metal components of CCA. The most mobile constituent of CCA will likely be chromium while arsenic and copper will be retained in the near-surface soil layers. However, these results cannot be carelessly extended to actual field conditions. Both the LA Ap and WH Bhfl soils used in this study had a high organic matter content. The WH Bhfl soil was also enriched with hydrous oxides. The importance of the organic matter component of the soils with respect to promoting reduction of the hexavalent chromium to the less soluble, strongly sorbed trivalent species has been discussed previously. Copper adsorption is also very highly correlated with organic matter in soils. The importance of organic matter for attenuation of arsenic has also been noted previously. Unfortunately, wood preservation plants, and in fact most industrial facilities, are not likely to have soils enriched in organic matter. As a result, the strong attenuation of arsenic and copper, and to a lesser extent chromium, observed in these soils cannot be relied upon to mitigate the potential for leaching for soils low in organic matter. Consequently, results from laboratory studies such as this should be interpreted in a conservative manner. 4.2.2 Breakthrough Curve A second column experiment was conducted using the Langley Ap and Whonnock Bhfl soils for the purpose of producing breakthrough curves for arsenic, chromium and copper. Breakthrough is established when the effluent metal concentration is equal to influent metal concentration. A ratio of effluent concentration (C) to influent concentration (C) is commonly used to express breakthrough. 100 The column soils were subjected to a sequential extraction scheme, following breakthrough, to determine the partitioning of the arsenic, chromium and copper within various soil pools. Breakthrough curves for the Langley Ap and Whonnock Bhfl soils are shown in Figures 4.28 and 4.29. The curves show that chromium is the most weakly retained metal, indicated by its sharp breakthrough curve at relatively low pore volume displacement. Arsenic is held very strongly by both soils, particularly the WH Bhf 1 soil. Copper appears to be held more strongly by the Langley soil than the Whonnock soil based on the number of pore volume displacements required before breakthrough. Analysis of the results on a individual metal basis reveals that arsenic retention in the soil columns is very highly correlated with the hydrous oxide component of the soils. Arsenic is so strongly retained by the Whonnock Bhfl soil that its first appearance in the effluent required 280 pore volume displacements compared with approximately 60 in the Langley Ap soil. The Whonnock Bhfl contains a significant amount of amorphous and crypto-crystalline iron and particularly aluminum hydrous oxides. The strong retention of arsenic in the two soils is in agreement with the results obtained for arsenic adsorption in batch equilibration tests using CCA. An attempt was made to relate the results from the column study to the batch tests using equilibrium and adsorbed metal concentrations. The cumulative equilibrium arsenic concentration versus the total adsorbed arsenic concentration for the column study was plotted on an adsorption isotherm generated during batch equilibration tests. The resulting point lay on the y-axis which indicated that the amount of arsenic sorbed in the column experiment appeared to be 100% relative to batch test results. A possible explanation for this is the much smaller soil/solution ratio in the column experiment. The relationship between the number of specific binding sites for arsenic and soil mass is obviously not a simple one. The shape of the breakthrough curve for chromium reflects a mixture of both Cr(VI) and Cr(lll). Chromium breaks through very rapidly at low pore volume displacements illustrating a low affinity for the soil adsorption sites. At this stage; the dominant chromium species in solution is most likely HCr0 4'. The lack of retention of chromium as the chromate oxyanion is probably due to competition PORE VOLUME o Figure 4.28 Breakthrough curve for the Langley Ap column experiment 0 50 100 150 200 250 300 3 5 0 400 PORE VOLUME o ro Figure 4.29 Breakthrough curve for the Whonnock Bhfl column experiment 103 with the arsenate anion for binding sites. Chromium reached a relative metal concentration (C/C) of 0.5 in approximately 10 pore volume displacements in the WH Bhfl soil and approximately 25 pore volume displacements in the La Ap soil. As discussed in the previous section, chromium attenuation in soils appears to be significantly influenced by flux. Therefore, chromium would be expected to more mobile in a soil with a coarse texture and thus higher hydraulic conductivity. In addition, the high organic matter content of the Langley soil provides greater potential for chromium reduction and may also influence the rate of the reduction reaction. A secondary chromium retention mechanism appears to be occurring following the steep breakthrough curve. The attenuation of chromium by the column soils increases significantly over successive pore volumes suggesting that a progressively greater portion of the influent chromium concentration is being converted to the trivalent chromium species. In fact, chromium never fully achieves breakthrough. Retention of Cr(lll) species may involve precipitation and coprecipitation reactions with hydrous oxides as well as adsorption to the soil substrates. The secondary mechanism is evident in both the LA Ap and WH Bhfl soil columns. Recognition of the two part behaviour of Cr(VI) in the soils is important for pollution prediction. In the event of a major accidental release of CCA, chromium is likely to migrate very rapidly in the soil profile before slower reactions, such as reduction to the Cr(lll) species, can take place. In the absence of organic matter or other electron donor substances, chromium will continue to pose a threat to groundwater resources. The breakthrough curve plots show that both soils strongly retain copper to a certain point (pore volume displacement) after which copper fairly rapidly reaches complete breakthrough. This indicates that the soils appear to have a finite capacity to adsorb copper which is not evident for the other two constituents. Copper was not detected in the effluent from the LA Ap soil column for 175 pore volume displacements compared with approximately 85 pore volumes for the WH Bhfl soil column. The Langley Ap soil has a greater capacity to adsorb copper than the Whonnock Bhf 1 soil, probably because of its higher organic matter content. The rapid breakthrough following initial detection of copper in the effluent indicates that specific binding sites for copper are exhausted. 104 A competitive effect with Cr(lll) cations may also influence the rapid breakthrough of Cu. The results for copper retention in the column soils are consistent with batch equilibration studies. The nohspecificity of copper adsorption at high solution concentrations was also evident in the batch experiments and was considered to reflect a simple ion exchange adsorption mechanism. Results from the sequential extraction procedure conducted on the column soils after breakthrough are shown in Figures 4.30 to 4.35. It is immediately apparent that the largest pool for all 3 metals is the ammonium oxalate extractable fraction. Partitioning of arsenic between the various pools showed that the easily exchangable water soluble and MgCI2 extractable pools recovered less than 10% of the arsenic adsorbed. The aqua regia extractable pool, which represents the residual fraction, accounted for approximately 8% of the arsenic adsorbed. The residual pool may account for the metals associated with the organic fraction in a largley insoluble form. The ammonium oxalate extractable fraction comprised over 80% of the arsenic adsorbed in the LA Ap soil and over 90% in the WH Bhfl soil. The results are generally very similar for both the LA Ap and the WH Bhf 1 soil. The association of arsenic with the ammonium oxalate extractable pool is consistent with the high adsorptive capacity of soils rich in hydrous oxides for arsenic. Similar results were observed in the batch and column tests. The small amount of arsenic extracted by water and MgCI2 suggests that the arsenic is strongly held by the soil by specific adsorption reactions and will not be readily available for uptake by plants or leaching in the soil profile. The distribution of chromium between the various soil pools shows that most of the chromium was extracted from the ammonium oxalate and aqua regia extractable fractions. Less than 3% of the chromium was found in the easily extractable pool in the LA Ap soil and virtually no chromium was found in the water and MgCI2 extracts for the WH Bhfl soil. For the LA Ap soil, the proportion of ammonium oxalate and aqua regia extractable chromium was almost equivalent. Chromium i distribution in the WH Bhfl soil was greater for the ammonium oxalate extractable pool relative to 105 As EXTRACTED (ug Aa/g) (Thousands) 3 -Water MgCI2 Ammonium Oxalate Aqua Regla *§zzzzzzza R S S S J — 10 15 D E P T H (cm) 20 Figure 4.30 Partitioning of arsenic in the Langley Ap soil column following a sequential extraction scheme. Cr EXTRACTED (ug Cr/g) (Thousands) 0.5 10 15 D E P T H (cm) Figure 4.31 Partitioning of chromium in the Langley Ap soil column following a sequential extraction scheme. 106 Cu EXTRACTED (ug Cu/g) (Thousands) ' ' H Water 111 M g C I 2 Wffih Ammonium Oxalate MM Aqua Regia D E P T H (cm) Figure 4.32 Partitioning of copper in the Langley Ap soil column following a sequential extraction scheme. 107 As EXTRACTED (ug As/g) (Thousands) 1 0 -2 -Figure 4.33 Water OHl MgCI2 HH Ammonium Oxalate Aqua Regia 10 15 D E P T H (cm) 20 Partitioning of arsenic in the Whonnock Bhfl soil column following a sequential extraction scheme. Cr EXTRACTED (ug Cr/g) (Thousands) 2.5 1.5 1 0.5 mm wmk Water III MgCI2 HHi Ammonium Oxalate Aqua Regia 10 15 D E P T H (cm) mm 20 Figure 4.34 Partitioning of chromium in the Whonnock Bhfl soil column following a sequential extraction scheme. 108 Cu EXTRACTED (ug Cu/g) (Thousands) Water DEPTH (cm) Figure 4.35 Partitioning ol copper in the Whonnock Bhfl soil column following a sequential extraction scheme. 109 the residual pool. The LA Ap soil has a higher organic matter content than the WH Bhfl and hence has the potential to reduce a large proportion of Cr(VI). The Cr(lll) cation is likely to have a greater affinity for organic matter binding sites than Cr(VI). The chromium associated with the aqua regia extractable pool may represent the organically bound or complexed trivalent species. i Ammonium oxalate may extract the acid soluble, lower molecular weight, fulvic acid fraction of organic matter while the higher molecular weight fraction humic acid and particulate organic matter would be left for extraction by aqua regia. In the WH Bhfl soil, a larger proportion of the chromium would be present as Cr(VI). In addition, podzolic B horizons have a proportionally higher fulvic acid than humic acid content. Consequently, a smaller portion of the chromium would be expected in the residual or aqua regia extractable pool. The larger pool of ammonium oxalate extractable Cr suggests that either a smaller proportion of Cr(VI) was reduced to Cr(lll) or that the Cr(lll) had coprecipitated with Fe- or Al-hydroxides. The acidic pH of the CCA solution (pH 3.3) may cause dissolution of hydrous oxides allowing them to participate in coprecipitation reactions. Stollenwerk and Grove (1985) reported difficulty in removing Crfrom alluvium that had aged for 1.5 years. They postulated that Cr(VI) may have slowly become part of the Fe-hydroxide structure or that it had reduced to Cr(lll) and coprecipitated with Fe-hydroxide. Copper distribution in the breakthrough curve column soils shows a larger proportion of the retained metal in the water and the MgCI2 extractable fraction. Approximately 30% of the adsorbed copper was in the exchangeable pool. The ammonium oxalate extractable pool, however, remained the largest fraction for adsorbed copper for both soils. The residual pool accounted for only about 5% of the adsorbed copper. The results indicated that simple ion exchange is a dominant mechanism of adsorption for copper in both the LA Ap and WH Bhfl soils. Therefore, a relatively large amount of the adsorbed copper may be taken up by plants or displaced by more specifically binding solutes. The reason for the large the proportion of copper associated with the ammonium oxalate extractable pool, rather than the residual pool, may be explained by the formation of soluble organic complexes with acid soluble fulvic acids. Although copper adsorption has been most closely correlated with organic matter (McBride, 1981 and 1989), specific adsorption of Cu by Fe-, Al- and Mn-oxides has been shown in subsoils low in organic matter and at low aqueous concentrations (Benjamin and 110 Leckie, 1981; Jarvis, 1981; McBride, 1982). Both of the soils used have a high organic matter content, but the WH Bhfl soil has a high hydrous oxide content. However, copper partitioning in the two soils was virtually identical. A possible explanation may be precipitation of copper as Cu(OH)2(s). In summary, the breakthrough curve results indicate that chromium will be the most mobile CCA constituent in soils. Arsenic is very strongly adsorbed by soils enriched with hydrous oxides, and copper is strongly adsorbed by the organic matter component of soils. Partitioning of adsorbed arsenic, chromium and copper in the soils revealed that only copper is available in an exchangeable phase and all three metals are strongly associated with the ammonium oxalate extractable pool. 111 5.0 SUMMARY Chromated copper arsenate is a water soluble, biocidal chemical used by the wood preservation industry to extend the service life of wood products. CCA is currently the most commonly used wood preservative chemical in B.C. Concern regarding the fate of CCA released to the environment prompted the Environmental Protection Service - Pacific Region to conduct an assessment of all CCA facilities in B.C. (Konasewich and Henning 1988). The assessment identified several research needs, one of which was the need to clarify the fate of CCA solutions which are spilled and/or chronically dripped on unpaved ground at wood preservation facilities. Although the attenuation of the individual constituents of CCA (arsenic, chromium and copper) have been studied by other researchers, no studies on the fate of arsenic, chromium and copper as a chemical mixture have been conducted, to my knowledge. Therefore, this study was carried out to examine the relative importance of various soil properties on the attenuation and release of arsenic, chromium and copper, in soils. The soil samples were collected from locations in the Lower Mainland. The soils were chosen to vary in the amount of organic matter, Fe- and Al- hydrous oxides and clays they contained. Batch equilibration studies were carried out using both CCA solutions and individual metal solutions. Data from the equilibration studies was found to be well described by the Freundlich equation, over the entire concentration studied. Simple and multiple regression analysis was conducted to relate Freundlich k values to soil parameters. The results showed that, in general, the Fe- and Al- hydrous oxide component of the soils was highly correlated with arsenic adsorption while the percent organic carbon was significantly correlated with chromium and copper retention in the soils studied. Soils enriched in Fe- and Al-hydrous oxides adsorbed almost 30% more arsenic than soils low in this soil component. The mechanism of arsenic adsorption on Fe- and Al- oxyhydroxides is considered to be ligand exchange with surface hydroxyls and/or surface aqua groups. 112 The organic carbon content of the soils was shown to influence the retention of chromium in soils. Organic matter would function as an electron donor in the reduction reaction of Cr(VI) to Cr(lll). Retention of chromium in the soils is largely as the cationic Cr(lll) species which is less soluble and more strongly adsorbed than the anionic Cr(VI) species in the subsurface. At high initial concentrations of CCA however, a proportion of chromium would continue to be present as the Cr(VI) anion. Soils enriched in Fe- and Al- hydrous oxides appeared to adsorb the greatest amount of chromium at high initial CCA concentrations, under acidic conditions. Organic matter was shown to be the most important soil property influencing the retention of copper in the soils studied. Hydrous oxide enriched soils also showed an affinity for copper while the clay component of the soil was found to have a low adsorption capacity for copper. Desorption of arsenic, chromium and copper from soils previously equilibrated with varying concentrations of CCA solutions showed that the extent of release is dependent on the amount of retained metal. Only a very small amount of the metals was released by the soils at dilute CCA concentrations suggesting that at lower adsorption densities the metal ions may be specifically adsorbed while at high concentrations the binding energy decreases and the metals are less strongly held. Chromium was desorbed to a greater extent than copper or arsenic in both soils indicating that the mobility of CCA constituents in the soils would follow the order: chromium > copper > arsenic. Results from the individual metal batch equilibration tests showed that the amount of arsenic adsorbed by the subsurface soils was similar to arsenic adsorption in CCA equilibrated soils. This indicates that the sites responsible for arsenic adsorption in these soils are specific for arsenic and that neither copper or chromium effectively compete for these binding sites. Adsorption of arsenic in the surface soil, LA Ap, showed enhanced adsorption in the CCA equilibrated soil. This result was attributed to the lower pH of the CCA solutions and suggests that under less acidic conditions than those used in this study, arsenic may be more mobile in the soil profile. Chromium adsorption by the soils in the single solute and CCA equilibration tests showed a marked difference. Approximately 20% more chromium was retained by soils equilibrated with chromium 113 solutions compared with CCA equilibrated soils. At high initial CCA concentrations, where a significant proportion of chromium is present as the chromate anion, arsenic is likely to be the major competing anion. The increased attenuation of chromium in the individual metal solutions was greater at higher initial chromium concentrations. At lower initial metal concentrations the chromium appeared to be adsorbed to the same extent in both CCA and individual metal solutions suggesting that specific binding sites were reponsible for chromium retention at low adsorption densities. This result may also indicate retention of the chromium as the reduced trivalent species which is less subject to competitive effects form codissblved solutes. Adsorption of copper by soils with a high organic matter content and consequently a high adsorption capacity for copper was found to be relatively unaffected by the presence of codissolved solutes. However, at high initial concentrations of copper in solution competitive effects became more important. Competitive effects from codissolved solutes are evident at lower initial copper concentrations in soils with a low adsorption capacity for copper. The migration of arsenic, chromium and copper in a soil column after application of a single dose of 2% CCA solution was investigated. The results showed that after the addition of 420 mL of deionized water both the arsenic and copper were retained in the upper 2.5 cm of the soil column while the chromium had migrated to the 7.5 cm depth in the LA Ap column and the lowest depth in the WH Bhfl soil column. Although the results for the migration of the three metals relative to each other were consistent with batch test findings, the increased mobility of chromium in the WH Bhfl soil column suggested the influence of solution flux on the retention of chromium. It appears that the retention of chromium is enhanced if the soil solution moves more slowly through the soil column. The breakthrough curve column experiments support the results shown in the slug dose column tests. Arsenic is very strongly retained by both the LA Ap and the WH Bhfl soils. Chromium, on the other hand, broke through at low pore volume displacements indicating a low affinity for adsorption sites, in both soils. Over successive pore volume displacements, however, chromium appeared to be attenuated to a greater extent, suggesting retention as the reduced, trivalent form. The breakthrough curve results show that copper was very strongly retained by both soils, but broke through rapidly after a finite 114 capacity for copper adsorption appeared to be reached. It is uncertain whether the rapid breakthrough observed was due to a finite adsorption capacity or competition for adsorption sites by Cr(lll). A combination of these two possibilities is most likely the answer. Partitioning of the metals in the column soils showed that the arsenic, chromium and copper retained by soils was largely associated with the ammonium oxalate extractable fraction. This result indicates that the CCA constituents are preferentially bound to the iron and aluminum hydrous oxide component and the acid soluble organic matter component of the soils studied.Only copper was found to have a significant amount of retained metal in the exchangeable pool (water and MgCI2 extractable) 115 6.0 CONCLUSIONS Use of CCA treatment chemicals at a plant site introduces the potential for contamination of soils, groundwater and surface water. The extent of potential contamination is, in turn, dependent on the physical and chemical properties of the treatment chemical, plant design, and operating practices as well as site specific characteristics, including: soil type, geology, hydrology, climate, topography and drainage. This study was undertaken to identify important soil characteristics influencing the attenuation of CCA in the subsurface environment. The most important soil properties with respect to the attenuation of arsenic, chromium and copper i under moderately acidic, oxidizing conditions were found to be the Fe- and Al- amorphous and crystalline hydrous oxides and the organic matter component of soils. The hydrous oxides are of particular importance in the adsorption of the oxyanions, l-LAsO/ and HCr04" . Adsorption of arsenic was most highly correlated with the hydrous oxide component of soils. Arsenic was not subject to competitive effects from codissolved solutes in CCA solution. Therefore, it is likely that arsenic is adsorbed via ligand exchange reactions with surface hydroxyl and/or aqua groups in soils. Chromium(VI) retention in the soils studied appears to be influenced by reduction reactions to the more stable, less soluble trivalent chromium species. The reduction reaction has been shown to proceed fairly rapidly under acidic conditions, in the presence of an electron donor such as organic matter (James and Bartlett 1983). Cr(lll) attenuation in soils is largely attributed to solid phase formation, especially above pH 6. Adsorption may be an important mechanism at more acidic pH where Cr(lll) may be adsorbed by layer silicates or complexed with organic matter. Cr(lll) retention in soils enriched with Fe and hydrous oxides is likely by coprecipitation reactions with free iron to form a Cr(lll) and Fe hydrous oxide. 116 Adsorption of the chromate oxyanion, HCr04", is highly correlated with the hydrous oxide component of the soils studied. However, the soil binding sites do not appear to be specific for the chromate anion. The presence of the arsenate anion in CCA solution significantly reduced chromium adsorption by the soils. Adsorption of copper was most highly correlated with the organic matter component of the soils studied. Precipitation may also be an important copper removal mechanism above pH 6. Copper appears to be nonspecifically adsorbed by the layer silicate and hydrous oxide component of soils. Reduced adsorption of the copper cation was observed at high initial metal concentrations in CCA solution as compared with single solute copper solution. All three metal constituents of CCA were found to be strongly adsorbed by the soils studied at dilute i initial concentrations of CCA. This result indicates that CCA is specifically adsorbed by the soil components at low adsorption densities. Both the batch and column studies show that of the three metal constituents in CCA, chromium has the most potential to migrate iri the soil profile. In the absence of organic matter or other electron donor substances such as Fe2*, a major accidental release of CCA could result in a significant threat of chromium contamination to groundwater resources. The slug dose column experiment also showed that low level CCA contamination, such as chemical dripped from freshly treated lumber, may result in significant subsurface migration of chromium, particularly in areas of high rainfall. Arsenic and copper in this experiment were both retained in the upper 2.5 cm of the soil column. Copper retention in soils is very highly correlated with the presence of organic matter. Unfortunately, organic matter is seldom a significant component in soils at industrial facilities. Therefore, under acidic conditions (pH < 5) copper may also be subject to leaching in the soil profile. Arsenic, on the other hand, is more likely to become mobile under moderately alkaline conditions. The presence of calcium, however, will control solution phase concentrations of arsenic by precipitation reactions. Recommendations for future work on the retention of CCA in soils include: 117 1) chromium speciation to confirm the reduction of Cr(VI) to Cr(lll), particularly in soils with a high organic matter component. 2) Both batch and column adsorption/desorption studies using soils with low organic matter contents. In summary, high levels of iron and aluminum hydrous oxides and high levels of organic matter are the most siginificant soil components for the attentuation of CCA in soils. The clay content, although not significantly correlated with adsorption of CCA components in soils, may influence retention of CCA by providing a physical barrier to the leaching of metals in the soil profile. Characterization of the subsurface environment for both soil types and presence and depth to groundwater should be undertaken as part of every environmental monitoring effort at CCA wood preservation facilities. Information from subsurface characterization together with the findings in this study will aid in evaluation and remediation of CCA contamination at CCA facilities. 118 7.0 REFERENCES Abd-Elfattah, A. and K. Wada. 1981. Adsorption of lead, copper, zinc, cobalt, and cadmium by soils that differ in cation exchange materials. J. Soil Sci. 32:271-283. Ainsworth, C . C , D.C. Girvin, J.M. Zachara and S.C. Smith. 1989. Chromate adsorption on goethite: effect of aluminum substitution. Soil Sci. Soc. Am. J . 53:411-418. Alesii, B.A., W.H. Fuller and M.V. Boyle. 1980. Effect of leachate flow rate on metal migration through soil. J. Environ. Qual. 9:119-126. Amacher, M.L., H.M. Selim and I.K. Iskandar. 1988. Kinetics of chromium (VI) and cadmium retention in soils, a nonlinear multireaction model. Soil Sci. Soc. Am. J. 52:398-408. Anderson, M.A. J.F. Ferguson and J. Gavis. 1976. Arsenate adsorption on amorphous aluminum hydroxide. J . Colloid Interface Sci. 54:391-399. Anderson, M.A. and D.T. Malotky. 1979. The adsorption of protolyzable anions on hydrous oxides at the isoelectric pH. J. Colloid Interface Sci. 72:413-427. Artiole, J. and W.H. Fuller. 1979. Effect of crushed limestone barriers on chromium attenuation in soils. J. Environ. Qual. 8:503-510. Barrow, N.J. 1985. Reaction of anions and cations with variable-charge soils, pp. 183-230. ]n: Advances in Agronomy, Volume 38. Academic Press. New York. Barrow, N.J. 1987. Reactions wityh Variable-Charge Soils. Martinus Nijhoff Publishers. Dodrecht, Netherlands. Bartlett, R.J. and J.H. Kimble. 1976a. Behavior of chromium in soils: Trivalent forms. J.Environ. Qual. 5:379-383. Bartlett, R.J. and J.H. Kimble. 1976b. Behavior of chromium in soils: Hexavalent forms. J. Environ. Qual. 5:383-386. Bascomb, D.L. 1968. Distribution of pyrophosphate extractable iron and organic carbon in soils of various groups. J. Soil Science 19:251-268. Benjamin, M.M. 1983. Adsorption and surface precipitation of metals on amorphous iron oxyhydroxide. Environ. Sci. Technol. 17:686-692. Benjamin, M.M. and J.p. Leckie. 1981. Multiple-site adsorption of Cd, Cu, Zn, and Pb on amorphous iron oxyhydroxide. J. Colloid Interface Sci. 79:209-221. Bingham, FT . , A.L. Page and J.R. Sims. 1964. Retention of Cu and Zn by H-montmorillonite. Soil Sci. Soc. Am. Proc. 28:351-354. Black, C A . 1965. Methods of Soil Analysis, Part 1 and 2. Agronomy 9:1-1572. Bloomfield, C. and G. Pruden. 1980. The behavior of Cr(VI) in soil under aerobic and anaerobic conditions. Environ. Pollut. Series A. 23:103-114. Bohn, H.L. 1976. Arsenic Eh-pH diagram and comparisons to the soil chemistry of phosphorus. Soil Science. 121:125-127. 119 Bohn, H.L., B.L. McNeal and G.A. O'Connor. 1979. Soil Chemistry. John Wiley & Sons, New York. , Bowen, H.J.M. 1979. Environmental Chemistry of the Elements. Academic Press, London. Breeze, V.G. 1973. Land reclamation and river pollution problems in the Croal Valley caused by waste from chromate manufacture. J. Appl. Ecol. 10:513-525. Boyle, R.W. and I.R. Jonasson. 1973. The geochemistry of arsenic and its use as an indicator element in geochemical prospecting. J . Chem. Explor. 2:251-296. Cary, E.E. 1982. Chromium in air, soil and natural waters, pp. 49-64. |n: S. Langard (ed), Biological and Environmental Aspects of Chromium. Elsevier Biomedical Press. Amsterdam. Canadian Council of Resource and Environment Ministers. 1987. Canadian Water Quality Guidelines. Environment Canada. Crecelius, E.A., M.H. Bothner and R. Carpenter. 1975. Geochemistries of arsenic, antimony, mercury and related elements in sediments of Puget Sound. Environ. Sci. Technol. 9:325-333. Dahlgren, S.E. and W.H. Hartford. 1972. Kinetics and mechanism of fixation of Cu-Cr-As wood preservatives. Hozforschung 26:62-69. Davies, B.E. 1980. Applied Soil Trace Elements, pp.259-284. John Wiley & Sons Ltd, New York. Davis, J.A. and J.O. Leckie. 1978. Effect of adsorbed complexing ligands on trace metal uptake by hydrous oxides. Environ. Sci. Technol. 12:1309-1315. Davis, J.A. and J.O. Leckie! 1980. Surface ionization and complexation at the oxide/water interface. 3. Adsorption of anions. J. Colloid Interface Sci. 74:32-43. Day, P.R. 1950. Physical basis of particle size analysis by the hydrometer method. Soil. Sci. 70:363-374. Elkhatib, E.A., O.L. Bennett and R.J. Wright. 1984. Arsenite sorption and desorption in soils. Soil Sci. Soc. Am.J. 48:1025-1030. Elliot, H.A. and C P . Huang. 1981. Adsorption characteristics of some Cu(ll) complexes on aluminosilicates. Water Res. 15:849-855. Farrah, H. and W.F. Pickering. 1976. The sorption of copper species by clays. I. Kaolinite, II. Illite and montmorillonite. Aust. J. Chem. 29:1167-1184. Fritz, S.J. and S.D. Hall. 1988. Efficacy of various sorbic media in attenuation of selenium. J. Environ. Qual. 17:480-484. Fordham, A.W. and K. Norrish. 1983. The nature of soil particles, particularly those reacting with arsenate, in a series of chemically treated samples. Aust. J. Soil Res. 21.:455-477. Frost.D.V. and R.A. Griffin. 1977. Effect of pH on adsorption of arsenic and selenium from a landfill leachate by clay minerals. Soil Sci. Soc. Am. J . 4V.53-57. Griffin, R.A., A.K. Au and R.R. Frost. 1977. Effect of pH on adsorption of chromium from landfill-leachate by clay minerals. J. Environ. Sci. Health A12:431-449. 120 Goldberg, S. 1986. Chemical modeling of arsenate adsorption on aluminum and iron oxide minerals. Soil Sci. Soc. Am. J . 50:1154-1157. Golden, D.C., J.B. Dixon and C.C. Chen. 1986. Ion exchange, thermal transformations, and oxidizing properties of birnessite. Clays Clay Miner. 34:511-520. Gough, G. and D.E. Konasewich. 1985. Spill incidents and consequences at Kopper's facility. Paper given at an EPS Chemical Management Workshop on Control of Chemical Releases from Wood Treating Facilities, 27 and 28 March, 1985. Grove, J.H. and B.G. Ellis. 1980a. Extractable chromium as related to soil pH and applied chromium. Soil Sci. Soc. Am. J. 44:238-242. Grove, J.H. and B.G. Ellis. 1980b. Extractable iron and manganese as related to soil pH and applied chromium. Soil Sci. Soc. Am. J . 44:243-246. Health and Welfare Canada. 1978. Guidelines for Canadian Drinking Water. Minister of National Health and Welfare. Ottawa, Ontario. Hem, J.D. 1977. Reactions of metal ions at surfaces of hydrous iron oxide. Geochim. Cosmochim. Acta 41:527-538. Henning, F.A and D.E. Konasewich. 1984a. Characterization and the Assessment of Wood Preservation Facilities in British Columbia. Environmental Protection Service, Pacific and Yukon Region, West Vancouver, B.C. Henning, F.A. and D.E. Konasewich. 1984b. Description and Assessment of Four Eastern Canadian Wood Preservation Facilities. Environmental Protection Service, Ottawa, Ontario. Henning, F.A. and D.E. Konasewich. 1984c. Overview and Assessment of Selected Canadian Wood Preservation Facilities. Environmental Protection Service, Ottawa, Ontario. Hendershot, W.H. 1978. Surface charge properties of selected soils. Ph.D. Diss., Univ. of B.C., Vancouver, B.C. Hingston, F.J. 1981. A review of anion sorption, pp. 51-90, ]n: M.A. Anderson and A.J. Rubin (eds), Adsorption of Inorganics at Solid-Liquid Interfaces. Ann Arbor Science Publ., Ann Arbor, Michigan. Hingston, F.J., R.J. Atkinson, A.M. Posner and J.P. Quirk. 1968. Specific adsorption of anions on goethite. 9th Int. Congress Soil Sci. Trans. 1:669-678. Hingston, F.J., A.M. Posner and J.P. Quirk. 1971. Competitive adsorption of negatively charged ligands on oxide surfaces. Discuss. Faraday Soc. 52:334-342. Hingston, F.J., A.M. Posner and J.P. Quirk. 1972. J . Soil Sci. 23:177-192. Holford, I.C.R. 1982. The comparative significance and utility of the Freundlich and Langmuir parameters for characterizing sorption and plant availability of phosphate in soils. Aust. J. Soil Res. 20:233-242. Huang, C P . , H.A. Elliot and R.M. Ashmead. 1977. Interfacial reactions and the fate of heavy metals in soil-water systems. J.W.P.C.F. 4:745-756. Jacobs, L.W., J.K. Syers and D.R. Keeney. 1970. Arsenic sorption by soils. Soil Sci. Soc. Am. Proc. 34:750-754. 121 Jahiruddin, M., B.J. Chambers, NT . Livesey and M.S. Cresser. 1986. Effect of liming on extractable Zn, Cu, Fe and Mn in selected Scottish soils. J. Soil Sci. 37:603-615. James, B.R. and R.J. Bartlett. 1983. Behavior of chromium in soils. VI. Interactions between oxidation-reduction and organic complexation. J. Environ. Qual. 12:173-176. James, R.O. and N.J. Barrow. 1981. Copper reactions with inorganic components of soils including uptake by oxide and silicate minerals, pp. 47-68. ]n: J.F. Loneragan, A.D. Robson and R.D. Graham (eds), Copper in Soils and Plants. Academic Press, New York. Jarvis, S.C. 1981. Copper sorption by soils at low concentrations and relation to uptake by plants. Soil Sci. 32:257-269. Kinniburgh, D.G. 1986. General purpose adsorption isotherms. Environ. Sci. Technol. 20:895-904. Konasewich, D.E. and E. Gerencher. 1986. Assessment of arsenic(lll) presence in CCA facility yard soils and drainage waters. Prepared for the Environmental Protection Service - Pacific Region. Koppers Co. Inc. 1985. Material Safety Data Sheets - Wolmanac Concentrate 50%. Koppers Co. Inc., Pittsburg, Pennsylvania. Korte, N.E., J. Skopp, W.H. Fuller, E.E. Niebla and B.A. Alesii. 1976. Trace element movement in soils: influence of soil physical and chemical properties. Soil Science 122:350-359. Kurdi, F. and H.E. Doner. 1983. Zinc and copper sorption and interaction in soils. Soil Sci. Am. J. 47:873-876. Langard, S. 1982. Biological and Environmental Aspects of Chromium. Elsevier Biomedical Press. Amsterdam. Laxen, D.P.H. 1985. Trace metal adsorption/coprecipitation on hydrous ferric oxide under realistic conditions. The role of humic substances. Water Res. 19:1229-1236. Leckie, J.O., M.M. Benjamin, K.Hayes, G. Kaufman and S. Altman. 1980. Adsorption/coprecipitation of trace elements from water with iron oxyhydroxide. EPRI-RPP-910. Electric Power Research Institute, Palo Alto, CA. Livesey, NT. and P.M. Huang. 1981. Adsorption of arsenate by soils and its relation to selected chemical properties and anions. Soil Science. 131:88-94. Luttmerding, H.A. 1981. Soils of the Langley-Vancouver map area. RAB Bulletin 15, British Columbia Soil Survey. Vol. 6. B.C. Ministry of the Environment. McBride, M.B. 1978. Retention of Cu 2*, Ca 2*, Mg2*,and Mn2* by amorphous alumina. Soil Sci. Am. J . 42:27-31. McBride, M.B. 1981. Forms and distribution of copper in solid and solution phases of soil. ]n: J.F. Loneragan, A.D. Robson, and R.D. Graham (eds), Copper in Soils and Plants. Academic Press. New York. McBride, M.B. 1989. Reactions controlling heavy metal solubility in soils. ]n: B.A. Stewart (ed), Advances in Soil Science. Volume 10. Springer-Verlag. New York. McKeague, J.A. and J.H. Day. 1966. Dithionite and oxalate extractable iron and aluminum as aids in differentiating various classes of soils. Can. J . Soil Sci. 46:13-22. 122 McKenzie, R.M. 1980. The adsorption of lead and other heavy metals on oxides of manganese and iron. Aust. J. Soil Res. 18:61-73. McLaren, R.G., R.S. Swift and J.G. Williams. 1981. The adsorption of copper by soil materials at low equilibrium concentrations. J. Soil Sci. 32:247-256. MacNaughton, M.G. 1977. Adsorption of chromium(vT) at the oxide-water interface. In: H.E. Drucker and R.E. Wildung (eds), Biological Implications of Metals in the Environment. National Technical Information Service, CONF-750929. Springfield VA. Mehra, O.P. and M.L. Jackson. 1960. Iron oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate. Clays and Clay Minerals 5:317-327. Mott, C.J.B. 1981. Anion and ligand exchange, In: D.J. Greenland and M.H.B. Hayes (eds), The Chemistry of Soil Processes. John Wiley and Sons, Ltd. New York. National Research Council. 1977. Arsenic. National Academy of Sciences. Washington, D.C. National Research Council of Canada. 1978. Effects of arsenic in the Canadian environment. NRCC. Ottawa, Ontario. National Research Council of Canada. 1976. Effects of chromium in the Canadian environment. NRCC. Ottawa, Ontario. National Research Council of Canada. 1979. Effects of copper in the Canadian environment. NRCC. Ottawa, Ontario. NIOSH. 1983. National Institute for Occupational Safety and Health. Registry of Toxic Effects of Chemical Sunstahces (RTECS). Vol.2. Parker, A.J. 1981. Introduction: the chemistry of copper, ]n: J.F. Loneragan, A.D. Robson and R.D. Graham (eds), Copper in Soils and Plants. Academic Press, New York. Penrose, W.R. 1974. ^rsenic in the marine and aquatic environments. Analysis, occurrence, and significance. CRC Critical Reviews in Environmental Control. 4:465-482. Pickering, W.F. 1979. Copper retention by soil/sediment components, pp.217-253. ]n: J.O. Nriagu (ed), Copper in the Environment. John Wiley & Sons, New York. Plambeck, J.A. and R.A. Smith. 1976. A survey and analysis of the nature and extent of heavy metal contamination of soils and vegetation in the area of Yellowknife, NWT. Environment Canada. Report EPS DSS 50-04870. Posner, A.M. and J.W. Bowden. 1980. Adsorption isotherms: should they be split? J . Soil Sci. 31:1-10. Rai, D., J.M. Zachara, A.P. Schwab, R.L. Resch, D.C. Girvin, R.L. Schmidt and J.E. Rogers. 1984. Chemical attenuation rates, coefficients, and constants in leachate migration. Vol. 1. A critical review. EPRI-EA-3356-Vol.1. Electric Power Research Institute. Palo Alto, CA. Rai, D., J.M. Zachara, L.E. Eary, D.C. Girvin, D.A. Moore, C T . Resch, B.M. Sass, and R.L. Schmidt. 1986. Geochemical behavior of chromium species. EA-4544. Electric Power Research Institute. Palo Alto, CA. Rantala, R.T. and D.H. Loring. 1973. Technical notes: new low-cost teflon decomposition vessel. Atomic Adsorption Newsletter 12:97-99. 123 Robertson, F.N. 1975. Hexavalent chromium in the groundwater in Paradise Valley, Arizona. Groundwater. 13:516-527. Roy, W.R, J.J. Hassett and R.A. Griffin. 1986. Competitive coefficients for the adsorption of arsenate, molybdate and phosphate mixtures by soils. Soil Sci. Soc. Am. J . 50:1176-1182. Sadiq, M., T.H. Zaida and A.A. Mian. 1983. Environmental behavior of arsenic in soils; theoretical. Water, Air and Soil Pollut. 20:369-377. Sanders, J.R. and C. Bloomfield. 1980. The influence of pH, ionic strength and reactant concentrations on copper complexing by humified organic matter. Journal of Soil Sci. 31:53-63. Schroeder, H.A. and G.F. Lee. 1975. Potential transformations of chromium in natural waters. Water, Air and Soil Pollut. 4:355-365. Scow, K., M. Byrne, M. Goyer, L.Nelken, J. Perwak, M.Wood and P.Cruse. 1985. An exposure and risk assessment for arsenic. EPA-440/4-85-005. Sengupta, A.K., D. Clifford and S. Subramonian. 1986. Chromate-ion exchange process at alkaline pH. Chem. Rev. 20:1174-1184. Sposito.G. 1984. The Surface Chemistry of Soils. Oxford Univ. Press, New York. Stevenson, F.J. and M.S. Ardakani. 1972. Organic matter reactions involving micronutrients in soils, ]n: J.J. Mortvedt et al (eds), Micronutrients in Agriculture. Soil Sci. Soc. Am., Madison WI. Stollenwerk, K.G. and D.B. Grove. 1985. Adsorption and desorption of hexavalent chromium in an alluvial aquifer near Telluride, Colorado. J. Environ. Qual. 14:150-155. Swaine, J . 1969. Trace element content of soils. Technical Communication No. 48 of the Commonwealth Bureau of Soil Science, Rothamstead Experimental Station, Harpenden. Tessier, A., P.G.C. Campbell and M. Bisson. 1979. Sequential extraction procedure for the speciation of particulate trace metals. Analytical Chemistry. 5J.:844-851. Thanabalasingam, P. and W.F. Pickering. 1986. Arsenic sorption by humic acids. Environ. Pollut. B12:233-246. Thornton, I. 1979. Copper in soils and sediments, In: J.O. Nriagu (ed), Copper in the Environment. John Wiley and Sons, Inc. New York. Travis, C C . and E.L. Etnier. 1981. A survey of sorption relationships for reactive solutes in soil. J . Environ. Qual. 10:8-17. United States Department of Agriculture. 1980. The Biologic and Economic Assessment of Pentachlorophenol, Inorganic Arsenicals, Creosote. USDA Technical Bulletin 1658-1. United States Environmental Protection Agency. 1986. Carcinogenic potential of arsenic compounds in drinking water. EPA/600/1-86/003. Veith, J.A. and G. Sposito. 1977. On the use of the Langmuir equation in the interpretation of "adsorption" phenomena. Soil Sci. Soc. Am. J . 41:697-702. 124 Walsh, L.M. and D.R. Keeney. 1975. Behavior and phytotoxicity of inorganic arsenicals in soils, pp. 35-53. In: E.A. Woolsen (ed), Arsenical Pesticides. ACS Symposium Series No. 7. Wagemann, R. 1978. Some theoretical aspects of stability and solubility of inorganic arsenic in the freshwater environment. Water Res. 12:139-145. Wauchope, R.D. 1975. Fixation of arsenical herbicides, phosphate and arsenate in alluvial soils. J. Environ. Qual. 4:355-358. r Woolsen, E.A., J.H. Axley and P.C. Kearney. 1971. The chemistry and phytotoxicity of arsenic in soils. I. Contaminated field soils. Soil Sci. Soc. Am. Proc. 35:938-943. Woolsen, E.A. 1983. Emissions, cycling and effects of arsenic in soil ecosystem, pp. 51-92. ]n: B.A. Fowler (ed), Biological and Environmental Effects of Arsenic. Elsevier Science Publishers. Zachara , J.M., D.C. Girvin, R.L. Schmidt and C T . Resch. 1987. Chromate adsorption on amorphous iron oxyhydroxide in the presence of major groundwater ions. Environ. Sci. Technol. 21:589-594. Zachara, J.M., C.E. Cowan, R.L. Schmidt and C.C. Ainsworth. 1988. Chromate adsorption on kaolinite. Clays Clay Miner. 36:317-326. Zachara, J.M., C.C. Ainsworth, C.E. Cowan and C T . Resch. 1989. Adsorption of chromate by subsurface soil horizons. Soil Sci. Soc. Am. J . 53:418-428. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097469/manifest

Comment

Related Items