United States Environmental Protection Agency Office of Research and Development Office of Solid Waste and Emergency Response EPA/540/4-89/001 March 1989 &EPA Superfund Ground Water Issue Ground Water Sampling for Metals Analyses Robert W. Puls and Michael J. Barcelona The Regional Superfund Ground Water Forum is a group of ground-water scientists, representing EPA's Regional Superfund Offices, organized to exchange up-to-date information related to ground-water remediation at Superfund sites. Filtration of ground-water samples for metals analysis is an is- sue identified by the Forum as a concern of Superfund deci- sion-makers. Inconsistency in EPA Superfund cleanup prac- tices occurs where one EPA Region implements a remedial action based on unfiltered ground-water samples, while another Region may consider a similar site to be clean based on filtered ground-water samples. RSKERL-Ada and EMSL-Las Vegas have convened a technical committee of experts in the areas of ground-water geochemistry, inorganic chemistry, colloidal trans- port and ground-water sampling technology to examine this is- sue and provide technical guidance based on current scientific information. Members of the committee were Robert W. Puls, Bert E. Bledsoe and Don A. Clark of RSKERL; Michael J. Barcelona, Illinois State Water Survey; Phillip M. Gschwend, Massachusetts Institute of Technology; Terry F Rees, USGS-Denver; John W. Hess, Desert Research Institute (EMSL-LV); and NicholousT Loux, ERL-Ath- ens. This document was written by Robert W. Puls and Michael J. Barcelona and edited by all members of the committee. For further information contact Robert Puls, RSKERL-Ada, FTS 743-2262; Bert Bledsoe, RSKERL-Ada, FTS 743-2324; Jane Denne, EMSL-LV, FTS 545-2655. The findings and recommendations of the committee were that use of a 0.45 micron* filter was not useful, appropriate or repro- ducible in providing information on metals mobility in ground- water systems, nor was it appropriate for determination of truly "dissolved" constituents in ground water. A dual sampling ap- proach was recommended, with collection of both filtered and unfiltered samples. If the purpose of the sampling is to deter- mine possible mobile contaminant species, the unfiltered samples should be given priority. This means that added em- phasis is placed on appropriate well construction methods, materials and ground-water sampling procedures. For accu- rate estimations of truly "dissolved" species concentrations, fil- tration with a nominal pore size smaller than 0.45 microns was recommended. It was further concluded that filtration could not compensate for inadequate construction or sampling proce- dures. Background/Support Information Filtration of ground-water samples for metal analyses will not provide accurate information concerning the mobility of metal contaminants. This is because some mobile species are likely to be removed by filtration before chemical analysis. Metal con- taminants may move through fractured and porous media not only as dissolved species, but also as precipitated phases, poly- meric species or adsorbed to inorganic or organic particles of colloidal dimensions. Colloids are generally considered as par- ticles with diameters less than 10 microns (Stumm and Morgan, 1981). Numerous investigators have suggested the facilitated transport of contaminants in association with mobile colloidal particles. Kim et al. (1984) suggested that sorption to ground- water colloidal material caused the mobilization of some radio- nuclides in Gorleben ground waters. Saltelli et al. (1984) stud- ied americium percolation in glauconitic sand columns and at- tributed the unretained fractions to migrating colloidal species. * Micron = mm = 10'6 meter Superfund Technology Support Centers for Ground Water Robert S. Kerr Environmental Research Laboratory Ada, OK Environmental Monitoring Systems Laboratory Las Vegas, NV ------- These colloids were either homogeneous hydrous precipitates, or were formed from the adsorption of the radionuclide onto colloidal size mineral particles. Colloidal particles generated in batch experiments by Sheppard et al. (1979) were shown to adsorb significant quantities of radionuclides. Further work by Sheppard et al. (1980) concluded that the transport of radionu- clides by colloidal clay particles must be considered in any con- taminant transport model. Champlin and Eichholz (1968) showed that the movement of radioactive sodium and ruthe- nium in sand beds was associated with particulate matter of micron dimensions. Gschwend and Reynolds (1987) demon- strated that submicron ferrous phosphate colloids were sus- pended and presumably mobile in a sand and gravel aquifer. Studies by Yao et al. (1971) and O'Melia (1980) indicate that colloidal particles in the range 0.1 to 1.0 micron may be most mobile in a sandy, porous medium. Kovenya et al. (1972) con- cluded that particles in the range 0.1 to 0.5 mm were most mo- bile in soil column studies. As much as 200 ppb copper, lead and cadmium was found associated with colloidal material in size range 0.015-0.450 mm by Tillekeratne et al. (1986). Rapid transport of plutonium (Pu) in core column studies by Champ et al. (1982) was attributed to colloidal transport, with 48% of the Pu associated with colloids in the size range 0.003-0.050 mm and 23% in the range 0.050-0.450 mm. Reynolds (1985) using carboxylated polystyrene beads ranging from 0.10 to 0.91 mm in size, recovered 45% of the 0.91 mm size beads, and greater than 70% of 0.10 and 0.28 mm size beads in laboratory sand column effluents. Lake and estuarine studies by Baker et al. (1986) and Means and Wijayaratne (1982) demonstrated the importance of natu- ral colloidal material in the transport of hydrophobic contami- nants. Carter and Suffet (1982) found that a significant fraction of "dissolved"DDT in surface waters was bound to colloidal hu- mic material. Takayanagi and Wong (1984) found over 70% of the total inorganic colloidal particles. Analytical methods used to determine "dissolved" metal con- centrations have historically used 0.45 micron filters to sepa- rate dissolved and particulate phases. If the purpose of such determinations is an evaluation of "mobile" species in solution, significant underestimations of mobility may result, due to col- loidal associations. On the other hand, if the purpose of such filtration is to determine truly dissolved aqueous species, the passage of colloidal material less than 0.45 microns in size may result in the overestimation of dissolved concentrations (Bergseth, 1983; Kim et al. 1984; Wagemann and Brunskill, 1975). Kennedy et al. (1974) found errors of an order of magni- tude or more in the determination of dissolved concentrations of aluminum, iron, manganese and titanium using 0.45 micron filtration. Sources of error were attributed to filter passage of fine-grained clay particles. Additionally, filtration of anoxic ground-water samples is very difficult without iron oxidation and colloid formation, causing a removal of previously dissolved species to be filtered. Filter loading and clogging of pores with fine particles may also occur, reducing the nominal size (Danielsson, 1981). Filtration should be viewed as only one approach for determining the "true" solution geochemistry of ground water, and others should be applied whenever possible. Purpose of Sampling It is important to identify the purpose of ground-water sampling before decisions regarding filtration, centrifugation or other phase separation techniques are made. Is it to determine the mobility of contaminants or to determine in situ aqueous geochemistry? The following definitions are also useful for consideration of this issue: (1) Total Contaminant Load Per Unit Volume of Aquifer = Mobile + Immobile Species. (2) Mobile Species = Dissolved + Suspended Species. (3) Dissolved = Free Ions + Inorganic Complexes + Low Molecular Weight Organic Complexes. (4) Suspended = Adsorbed + Precipitated + Polymeric + High Molecular Weight Organic Complexes. For an assessment of mobility, all mobile species must be con- sidered, including suspended particles acting as adsorbents for contaminants. While not all suspended species may necessar- ily be sufficiently mobile or toxic to pose a health risk, a conser- vative approach is proposed at this time until more definitive data are available. Contaminant transport models which ac- count for an additional aqueous mobile colloidal phase have been proposed by Avogadro and DeMarsily (1984) and Enfield and Bengsston (1988). A principle objective in a sampling effort for testing a geochemi- cal speciation model is to obtain estimates of the free ion activi- ties of the major and trace elements of interest. Since there are relatively few easily performed analytical procedures for mak- ing these experimental estimates, an alternative procedure is to test the analytically determined dissolved concentrations with model predictions including both free and complexed species. More and more remedial investigations are utilizing such mod- els to make predictions about contaminant behavior based on dissolved concentrations. It is not the purpose of this report to suggest how to perform these analytical determinations, but as noted above, the use of a 0.45 micron filter as the operational definition of "dissolved" may be inappropriate. Analytical tech- niques such as ion selective electrodes, ion exchange and po- larography may be more accurate. Research utilizing these and other techniques to correlate "dissolved" with filter size is recommended. If one adopts the conservative approach with no filtration for contaminant mobility estimations, increased importance is placed on proper well construction, and purging and sampling proce- dures to eliminate or minimize sources of sampling artifacts. Sources of Sampling Artifacts vs. "real" Ground-water Environment The disturbance of the subsurface environment as a result of well construction and sampling procedures presents serious ------- obstacles to the interpretation of ground-water quality results. Some degree of disturbance of natural conditions is inevitable. However, the impact of improper well construction and sam- pling techniques can permanently bias the usefulness and in- tegrity of wells as sampling points. Several aspects of well con- struction and sampling procedures must be carefully consid- ered to avoid errors associated with the introduction of foreign particles or the alteration of ambient subsurface conditions which may affect natural dissolved or suspended materials. Well Construction The design, drilling, and construction of monitoring wells have been identified as particularly important steps in the collection of representative water chemistry and hydrologic data. Several references have emphasized the minimization of both the dis- turbance and the introduction of foreign materials (USEPA, OSWER-9950.1, 1986; Barcelona, et al., 1983; Barcelona et al., 1985) because of the potential impact on water chemistry. The RCRA Technical Enforcement Guidance Document (USEPA, OSWER-9950.1, 1986) suggests that the well must allow for sufficient ground-water flow for sampling, minimize passage of formation materials into the well, and exhibit suffi- cient structural integrity to prevent collapse of the intake struc- ture. It should be recognized, however, that the well must first provide a representative hydraulic connection to the geologic formation of interest. Without the assurance of this hydraulic integrity, the water chemistry information cannot be interpreted in relation to the dynamics of the flow system or the transport of chemical constituents. More specific guidance is therefore necessary to maintain or restore the natural hydraulic conductivity of the formation in the vicinity of the screened portion of monitoring wells through the drilling, construction and development procedures. The litera- ture on water well technology can be most helpful in this regard since minimal disturbances of the subsurface is a common goal in maximizing both the yield of water supply wells and the rep- resentativeness of water samples and hydraulic information from monitoring wells (Driscoll, 1986). To insure the long-term integrity of monitoring wells, particularly with respect to excluding foreign particles and permitting the passage of mobile (i.e., dissolved and suspended) contaminants, specific items which should be observed are: 1) If no alternative to the use of drilling muds or fluids exists, these materials must be removed from the well bore and adja- cent formations by careful well development (Driscoll, 1986). This guidance also applies to the removal of the low permeabil- ity "skin" which is caused by abrasion, oxidation and invasive muds which may seal the well bore from the screened interval and bias in situ determininations of hydraulic conductivity (Faust and Mercer, 1984; Moench and Hsieh, 1985; Faust and Mercer, 1985). Pumping rates during development should be docu- mented and care should be taken not to exceed these rates during purging or sampling since further development and well damage may aggravate suspended particulate and turbidity problems even in properly designed wells. 2) The emplacement of grouts and seals to isolate the screened interval must be carefully done. The use of tremie pipes and frequent checking of the depth of emplacement of clay or cement grouts during well construction are strongly en- couraged. It is also important to take care to follow manufacturer's guide- lines on the hydration of cement or expanding cement as grouts or seals. Excess water addition and grading of cement compo- nents or materials due to free fall through standing water can permanently damage the well's integrity (Evans and Ellingson, 1988). 3) Casing and screen materials must be selected to retain their integrity in the subsurface environment (i.e., avoid iron, steel), minimize bias to water samples and insure that screen openings are not reduced by the buildup of corrosion products or by compression (USEPA, OSWER-9950.1, 1986). These effects can be checked by repeat determinations of in situ hy- draulic conductivity over the useful life of the well. Redevelop- ment and replacement of the well should be considered if dete- rioration or significant changes in hydraulic conductivity are observed. Erratic water level readings and sudden changes in turbidity or purging behavior of monitoring wells prior to sam- pling are warning signs of possible loss of material integrity. 4) Well design fundamentals with regard to the selection of a filter pack and screen size are among the most important is- sues in obtaining representative hydraulic and water quality in- formation. The exclusion of fines, clays, and silts can be achieved by selecting the grain-size distribution for the filter pack by multiplying the 50-percent retained size of the finest formation sample by a factor of two (Driscoll, 1986). The filter pack material should be cleaned and washed free of fines to insure that extraneous contaminants or particles are removed. The well screen slot openings should be chosen to retain 90% of the filter pack material after development. In natural packed wells it may be advisable to select a screen slot size which will retain at least 50% of the finest material in the screened inter- val. Minimizing slot screen width however, often leads to greater time and energy spent in well development. The need to docu- ment well development procedures cannot be overemphasized. Maintenance of the hydraulic performance of monitoring wells and the connection of wells to the zones of greatest hydraulic conductivity, where contaminant transport is most probable, should take equal importance to the collection of representative water quality data. Purging and Sampling Water that remains in the well casing between sampling peri- ods is unrepresentative of water in the formation opposite the screened interval. It must be removed by purging or isolated from the collected sample by a packer arrangement prior to the collection of representative water samples. Water level read- ings must be made carefully to avoid the disturbance of fines or precipitates which may enter or form in the well due to chemical reactions or microbial processes and accumulate on the inte- rior walls of the well casing screen or at the bottom of the well. ------- Similarly, it is important to purge the stagnant water at flow rates below those used in development to avoid further development, well damage or the disturbance of accumulated corrosion or reaction products in the well. The use of certain sampling de- vices, particularly bailers and air-lift arrangements, should be discouraged in order to avoid the entrainment of suspended materials which are not representative of mobile chemical con- stituents in the formation of interest. A note of caution should be voiced to encourage repetitive sam- pling of monitoring wells prior to judging the representativeness of determinations of hydraulic conductivity , water level read- ings and water quality data. The effects of the inevitable "trauma" due to drilling, sealing and development of monitoring wells can bias observations of water chemistry until the subsurface is al- lowed to equilibrate sufficiently (Walker, 1983). Estimates of the time to achieve equilibration vary substantially, particularly when drilling fluids are used in highly permeable formations (Brobst, 1984; Driscoll, 1986); however, periods of weeks to several months may be necessary before even major ionic con- stituents of ground water equilibrate to previous levels (Barcelona, et al., 1988). Recommendations for Sampling In general, the zone of interest must be isolated, the sample pumped slowly to minimize turbidity and sample collected in such manner as to eliminate O2 and CO2 exchange with the at- mosphere. No filtration for mobile metals determination is rec- ommended. If the unfiltered values exceed maximum contami- nant level concentrations for ground-water quality, additional analyses and re-evaluation of sampling artifacts are required. It should be emphasized that extreme differences between un- filtered and 0.45 mm filtered samples does not preclude the use of unfiltered data for risk assessment decisions. Significant particulate mobility may be occurring at such a site, and addi- tional analyses with other larger filters (e.g. >0.45 mm) may be most appropriate given the current size estimates for upper lim- its for mobile particles. Isolation of Sampling Zone Isolation ofthe samplingzone is necessary to minimizethe purge volume as well as to minimize air contact. This is especially important since Eh/pH conditions of the formation waters are notoriously sensitive to dissolved gases content. Inflatable pack- ers can be used to achieve isolation ofthe sampling zone. Pumping for Sample Collection It is recommended that a positive displacement pump can be used. Othertypes of sample collection (e.g., bailing) may cause displacement of non-mobile particles or significantly alter ground water chemistry leading to colloid formation (e.g., vacuum pumps). Surging must be avoided, and a flow rate as close to the actual ground-water flow rate should be employed. Acknowl- edging that this may be impossible or impractical in some in- stances, a pumping flow rate based on the linear ground-water flow rate and open screen area is proposed, where pumping flow rate -linear GW flow rate x 2 x screen ht. x well radius x 10 While an initial approximation, flow rates around 100 ml/min have been used to successfully sample ground-waters in a qui- escent mode. Additional research is needed in this area, particularly with re- spect to the appropriateness of this generic equation. An inex- pensive flow-through type cell set-up utilizing this approach was described by Garske and Schock (1986). Assessment of Water Constituents While Sampling Monitoring of the pumped ground water for dissolved oxygen, temperature, conductivity and pH aids in the interpretation or establishment of ground-water background quality. Gschwend and co-workers (personal communication) have observed that turbidity diminished dramatically after prolonged pumping, changing similarly, although possibly more slowly, than other water quality parameters (e.g., O2, conductivity). An initial esti- mate proposed for time of pumping necessary to collect water from a formation is around two times the time required to get plateau values for the above parameters. No Filtration for Mobile Fraction Determination Those samples intended to indicate the mobile substance load should not be filtered. Steps to preserve their integrity, such as acidification, should be performed as soon as possible. Filtration for Specific Geochemical Information Any filtration for estimates of dissolved subsurface species loads should be performed in the field with no air contact and immedi- ate preservation and storage. In-line pressure filtration is best with as small a filter pore size as practically possible (e.g., 0.05, 0.10 micron). Using a smaller pore size filter will require longer sample collection time, increasing the need for air exclusion from the sample (Laxen and Chandler, 1982; Holmetal., 1988). Polycarbonate membrane-type filters with uniform and sharp size cutoffs are recommended to minimize particle loading on the filter. Although membrane filters are more prone to clogging than fiber-type filters, the uniform pore size, ease of cleaning, and minimization of adsorptive losses from the sample tend to improve the precision and accuracy in the analytical data. The filter holder should be of material compatible with the metals of interest. Holders made of steel are subject to corrosion and may introduce non-formation metals to samples. Large diam- eter filter holders (e.g., > 47 mm) are recommended to reduce clogging and pore size reduction and for ease of filter pad re- placement. The use of disposable in-line filters are suggested for convenience if of sufficient quality. Prewashing of filters should be routinely performed. Work by Jay (1985) shows that virtually all filters require prewashing to avoid sample contami- nation. Quality assurance and quality control becomes increasingly important when adopting the above recommendations. The use ------- of field blanks and standards for field sampling is essential. Field blanks and standards enable quantitative correction for bias due to collection, storage and transport. Analysis of the filters them- selves and their particulate load is suggested as a check on mass balance and filtration effects on solid/solution separation efficiency. References Avogadro, A. and G. De Marsily 1984. The Role of Colloids in Nuclear Waste Disposal. In Scientific Basis for Nuclear Waste Management, Gary L. McVay, Ed., pp. 495-505. Backhus, D., P. R. Gschwend, M.D. Reynolds. 1986. Sampling Colloids in Groundwater, Abstract. EOS 67:954. Baker, J.E., PD. Capel, and S.J. Eisenreich. 1986. Influence of Colloids on Sediment-Water Partition Coefficients of Polychlorobiphenyl Congeners in Natural Waters. Environ. Sci. Technol. 20(11):1136-1143. Barcelona, M.J., G.K. George, and M.R. Schock. 1988. Com- parison of Water Samples from PTFE, PVC and SS Monitoring Wells. Illinois State Water Survey Internal Report Prepared for USEPA-EMSL, Las Vegas, NV, Aquatic and Subsurface Moni- toring Branch (CR #812165-02), 37 pp. Barcelona, M.J., J.P Gibb, and R.A. Miller. 1983. A Guide to the Selection of Materials for Monitoring Well Construction and Groundwater Sampling. Illinois State Water Survey Contract Report #327 prepared for USEPA-RSKERL, Ada, OK, and USEPA-EMSL, Las Vegas, NV. EPA-600/2-84-024, 78 pp. Barcelona, M.J., J.P. Gibb, J.A. Helfrich, and E.E. Garske. 1985. Practical Guide for Groundwater Sampling. Illinois State Water Survey Contract Report #374 prepared for USEPA-RSKERL, Ada, OK and USEPA-EMSL, Las Vegas, NV. EPA-600/2-85- 104, 94 pp. Bergseth, H. 1983. Effect of Filter Type and Filtrate Treatment on the Measured Content of Al and Other Ions in Groundwater. Acta Agriculturae Scandinavica 33 (1983) 353-359. Brobst, R.B. 1984. Effects of Two Selected Drilling Fluids on Ground Water Sample Chemistry; Monitoring Wells, Their Place in the Water Well Industry. Educational Session, NWWA Na- tional Meeting and Exposition, Las Vegas, NC. Carter, C.W and I.H. Suffet. 1982. Binding of DDT to Dis- solved Humic Material. Environ. Sci. Technol. 16(11 ):735-740. Champ, D.R., WF Merritt, and J.L. Young. 1982. Potential for Rapid Transport of Pu in Groundwater as Demonstrated by Core Column Studies. In Scientific Basis for Radioactive Waste Man- agement. Vol. 5, Elsevier Sci. Publ., NY. Champlin, J.B.F., and G.G. Eichholz. 1968. The Movement of Radioactive Sodium and Ruthenium through a Simulated Aqui- fer. Water Resour. Res. 4(1):147-158. Danielsson, L.G. 1982. On the Use of Filters for Distinguishing Between Dissolved and Particulate Fractions in Natural Waters. Water Res. 16, 179-182. Driscoll, FG. 1986. Ground Water and Wells. 2nd Ed., Johnson Division, St. Paul, MN, 1108 pp., pp. 497, 438-9, 722, 725-26. Enfield, C.G. and G. Bengtsson. 1988. MacromolecularTrans- port of Hydrophobic Contaminants in Aqueous Environments. Ground Water 26(1 ):64-70. Evans, L.G. and S.B. Ellingson. 1988. The Formation of Ce- ment Bleed-Water and Minimizing Its Effect on Water Quality Samples. In Proceedings of the Ground Water Geochemistry Conference, pp. 377-389. Hyatt Regency, Denver, CO, Asso- ciation of Ground Water Scientists and Engineers. NWWA-Water Well Journal Publishing Company, Dublin, OH. Faust, C.R. and J.W Mercer. 1984. Evaluation of Slug Tests in Wells Containing a Finite Thickness Skin. Water Resour. Res. 20(4):504-506. Faust, C.R. and J.W. Mercer. 1985. Reply Water Resources 21(9):1462. Garske, E.E. and M.R. Schock. 1986. An Inexpensive Flow- Through Cell and Measurement System for Monitoring Selected Chemical Parameters in Ground Water. Ground Water Monitor- ing Review 6(3): 79-84. Gschwend, P.M. and M.D. Reynolds. 1987. Monodisperse Ferrous Phosphate Colloids in an Anoxic Groundwater Plume. J. of Contaminant Hydrol. 1(1987):309-327. Holm, T.R., G.K. George and M.J. Barcelona. 1988. Oxygen Transfer Through Flexible Tubing and its Effects on Ground Water Sampling Results. Ground Water Monitoring Review. 8(3):83-89. Jay, PC. 1985. Anion Contamination of Environmental Water Samples Introduced by Filter Media. Analytical Chemistry 57(3):780-782. Kennedy, V.C. and G.W Zellweger. 1974. Filter Pore-Size Ef- fects on theAnalysis of Al, Fe, Mn, and Ti in Water. Water Resour. Res. 10(4):785-790. Kim, J.I., G. Buckau, F Baumgartner, H.C. Moon and D. Lux. 1984. Colloid Generation and theActinide Migration in Gorbelen Groundwaters. In Scientific Basis for Nuclear Waste Manage- ment, V 7, Gary L. McVay, Ed., Elsevier, NY, pp. 31-40. Kovenya, S.V., M.K. Mel/EnikovaandA.S. Frid. 1972. Studyof the Role of Mechanical Forces and Geometric Conditions in the Movement of Highly Dispersed Particles in Soil Columns. Pochvovedeniye. 10, 133-140. Laxen, D.P.H. and I.M. Chandler. 1982. Comparison of Filtra- tion Techniques for Size Distribution in Freshwaters. Analytical Chemistry 54(8): 1350-1355. ------- Means, J.C. and R. Wijayaratne. 1982. Role of Natural Col- loids in the Transport of Hydrophobic Pollutants. Science 215(19):968-970. Moench, A.F. and PA. Hsieh. 1985. Comment on "Evaluation of Slug Tests in Wells Containing a Finite Skin" by C.R. Faust and J.W. Mercer. Water Resour. Res. 21 (9): 1459-1461. O'Melia, C.R. 1980. Aquasols: The Behavior of Small Par- ticles in Aquatic Systems. Environ. Sci. Technol. 14(9): 1052- 1060. Reynolds, M.D. Colloids in Groundwater. 1985. Masters The- sis. Mass. Inst. of Tech. Boston, MA. Saltelli, A., A. Avogadro and G. Bidoglio. 1984. Americium Fil- tration in Glauconitic Sand Columns. NuclearTechnol. 67, 245- 254. Sheppard, J.C., M.J. Campbell and J.A. Kittrick. 1979. Reten- tion of Neptunium, Americium and Curium by Diffusible Soil Par- ticles. Environ. Sci. Technol. 13(6):680-684. Sheppard, J.C., M.J. Campbell, T Cheng and J.A. Kittrick. 1980. Retention of Radionuclides by Mobile Humic Compounds. Environ. Sci. Technol. 14(11):1349-1353. Stumm, W and J.J. Morgan. 1981. Aquatic Chemistry. John Wiley and Sons, Inc., NY. Takayanagi, K. and G.T.F. Wong. 1984. Organic and Colloidal Selenium in South Chesapeake Bay and Adjacent Waters. Marine Chem. 14:141-148. Tillekeratne, S., T. Miwa and A. Mizuike. 1986. Determination of Traces of Heavy Metals in Positively Charged Inorganic Col- loids in Fresh Water. Mikrochimica Acta 6:289-296. Wagemann, R. and G.J. Brunskill. 1975. The Effect of Filter Pore-Size onAnalytical Concentrations of Some Trace Elements in Filtrates of Natural Water. Intern. J. Environ. Anal. Chem. 4:75-84. Walker, S.E. 1983. Background Water Quality Monitoring: Well Installation Trauma. In Proceedings of the Third National Sym- posium on Aquifer Restoration and Ground Water Monitoring, NWWA, Fawcett Center, Columbus, OH. pp. 235-246. Yao, K., M.T Habibian and C.R. O'Melia. 1971. Water and Waste Water Filtration: Concepts and Applications. Environ. Sci. Technol. 5(11):1105-1112. ------- |