Radioactivity in the Marine Environment (1971)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Marine Chemistry 143 TABLE 5 Ratios of Elemental Concentrations in Deep Waters Cd and in Surface Waters C^ of the Pacific Ocean Element CdeeplCsurf Li 1.0 Na 1.0 K 1.0 Kb 1.0 Cs 1.0 Mg 1.0 1.02 Ca Sr <1.1 Ba 4 Ra 4 CO2 1.2 S as SO;2 P as phosphate ions SiasSi(OH)§ N as NOJ Mn 1.0 >10.0 >10 >10.0 .- 1 Mo 1 U 1 Br ~1 posed organic matter and deposited with sediments in the coastal ocean. Consequently, open ocean sediments are rela- tively less important as a reservoir for reactive elements than are the coastal bottom deposits. Downward transport and subsequent decomposition of the particulate remains of marine organisms are the major cause of chemical inhomogeneities within the ocean. Vari- ous elements fixed by marine organisms in the surface wa- ters of the ocean are transported to, and released in, depths where solution or oxidation of the debris takes place. The net result is a depletion of these elements in surface waters and an enrichment in deep waters. The ratio of the concentration of an element (normalized to a standard salinity) in the deep waters to that in the sur- face waters of the Pacific Ocean provides a convenient index of the effectiveness of this process. Values are given in Table 5 for several elements. Nitrogen, phosphorus, and silicon are removed very effi- ciently, the last element primarily as the opaline tests of diatoms and radiolaria, and the first two as basic constit- uents of the organic material itself. The fact that the nitrogen/phosphorus ratio (16/1) is nearly constant in all the major open ocean water masses and in all major types of organisms attests to the dominance of organic activity in the generation of inhomogeneities in the concentrations of these two elements. Carbon is depleted not only by its in- corporation in organic molecules but also by its fixation and removal as CaCO3 (in foraminifera, coccoliths, pteropods). Calcium owes its depletion to CaCO3 and possibly to cal- cium phosphate precipitation. The absence of significant concentration variations for relatively unreactive elements such as cesium strongly suggests that transport by organisms is unimportant compared to transport by physical mixing in the water column. Nonsystematic variations in the strontium/chlorinity and cesium/chlorinity ratios have been reported, but the results are difficult to interpret. Possibly, the strontium concentra- tion of seawater is altered by the dissolution of large num- bers of celestite (strontium sulfate) radiolaria. For cesium no mechanism is evident. The depletion factors given in Table 1 can be easily con- verted to residence times for the element in surface water, Ts, relative to transfer to the deep sea. Assuming that such elements as Na, K, Mg, Cl, and S are transported only by physical mixing, we can use their residence times as refer- ences. From simple material-balance considerations, it can be shown that Ts element = "deep (Tf chlorine) where C^ and Cdeep are the concentrations of the element in surface and in deep water, respectively. Radium, for ex- ample, because of its fourfold enrichment in deep water, apparently resides in surface water only one quarter the time that chlorine resides there. The silicon residence time is on the order of one tenth the chlorine residence time. The residence time of a relatively nonreactive element is about 20 years in the mixed layer for the world ocean-this estimate is based upon physical mixing processes solely. Thus, barium or radium would have passage times of about 5 years in the mixed layer, while silicon would be removed in about 2 years. The variations with depth of many trace elements in sea- water are poorly known, as indicated in Table 2. For such elements, another approach is needed to delimit their reac- tivities in biological cycles. One possibility is to use the chemical composition of marine plankton. The degree to which a given element is enriched by marine organisms should be related to the importance of its transport by par- ticles. Three types of solid phases must be considered: organic detritus, calcareous exoskeletal materials, and opa- line exoskeletal materials. The total ocean-wide productions can be estimated from the deep-water excesses (concentra- tion in deep water minus concentration in surface water) of silicate, of total dissolved inorganic carbon or carbonate alkalinity, or of nitrate and phosphate. The deep Pacific excesses are as follows: P 3 X10-6 moles/liter N 5 X10-6 moles/liter C 4X10-* moles/liter Si 2 X10-4 moles/liter Carbonate alkalinity 2X10~4 equivalence/liter

144 Radioactivity in the Marine Environment TABLE 6 Elemental Concentration Factors for Marine Organisms (Expressed as the Logarithm)" TABLE 7 Concentration Factors between Seawater and Plankton Ash for Some Trace Elements" Element Plants Animals Al 4 4.5 As 3 2 Ba — 1 Be 3 3 B 0.1 0.5 Cd 2 5 Ca 0.5 1 Ce 2.5 2 Cs 0.5 1 a — -1 Cr 4.5 5 Co 3 2 Cu 2 3 F 0.1 - Ga 1.5 1.5 Au 3.5 1.5 I 3.5 - Fe 4.5 3.5 Pb 4.0 4.0 Li - -0.3 Mg -0.2 -0.2 Mn 3.5 3.5 Mo 1.5 1.5 Ni 2.0 3.0 Nb 3.0 5.5 P 3.7 4 Pu 3 3 K -0.1 0.5 Ra — - Ru 2.0 1.0 Sc 4.5 3.5 Si 2.3 2.0 Ag 3.0 3.0 Na -1.5 -0.8 Sr 1.0 0.5 S -0.5 -0.5 Sn 2.5 2.0 Ti 3.5 3.5 W 3.0 2.0 U - - V 2.5 2.0 Zn 3.0 4.0 Zr 3.0 3.2 "Concentration factors for organisms from the open and coastal areas are included. Data is insufficient to ascertain differences that might result from systemic variations in elemental concentrations in these two marine domains. For the total dissolved inorganic carbon, 3 X10"4 moles/ liter result from the combustion of organic carbon, and 1 X10-" moles/liter from the fallout of CaCO3. The carbo- nate alkalinity, nitrogen, phosphorus, and carbon data are internally consistent, and show that for each mole of CaCO3 dissolving in the deep sea, organic material containing 3 moles of organic carbon must be oxidized. The silicon data Concentration Factors (liters seawater per Plankton Ash (Mg/g) grams plankton ash) Ele- Seawater Plants Plants ment (Mg/liter) (Sargassum) Animals (Sargassum) Animals P 88 20,000 20,000 230 230 Ag 0.3 0.3 0.3 1 1 Al 1 65 300 65 300 B 4,450 1,200 140 0.27 0.031 Ba 20 120 52 6.0 2.5 Cd 0.11 8 13 72 120 Co 0.4 3 3 7.5 7.5 Cr 0.5 9 7 18 14 Cu 2 270 270 135 135 Li 170 6 40 0.04 0.2 Ni 6.6 27 12 4 2 Sr 8,100 8,500 930 1 0.1 Ti 1 26 120 26 120 "Based on unpublished data for marine plankton species from G. Thompson and V. T. Bowen (Woods Hole), H. Curl (Oregon State), G. Nicholls (Manchester), and K. K. Turekian (Yale). suggest that for each mole of CaCO3, there must be 2 moles of opaline silica precipitated. The corresponding weight ratios would be—dry organic material: CaCO3 : SiO2 = 0.7: 1.0: 1.3. The elemental concentration factors for marine organ- isms, with respect to seawater, are given in Table 6. Although there is an uncertainty of at least an order of magnitude in these values for many of the elements, they provide an entry to the problem of the dissemination of elements through biological activity. Where reliable concen- trations are known, we can estimate the deep-water ex- cesses of trace elements. If we assume, for example, that any element incorporated into plant tissue is released in a manner similar to that of phosphorus without fractionation, then the enrichment values in Table 6 assume an impor- tance. Those elements that are reported to be enriched to the same level as phosphorus, or higher (Al, Sc, Pb, Fe, and Cr in Table 6), would be expected to have significant deep- water enrichment. Such elements are removed from surface water by sinking particles. Since phosphorus can attain en- richments of an order of magnitude or so in the deep ocean relative to the surface ocean, elements that are enriched in marine plants by even one order of magnitude less than phosphorus would be expected to have detectable non- homogeneous vertical distributions, whereas, if the enrich- ment of an element relative to phosphorus is down by more than one order of magnitude, no significant vertical- concentration changes are to be expected. Concentration factors based upon direct measurements of elemental contents in ashed organisms are given in

Marine Chemistry 145 Table 7 in the arbitrary units of liters of seawater per gram of ash. On this scale, the concentration factor for phos- phorus is about 200 (or more, if we use the lower phospho- rus concentration of surface waters). These more recently determined values provide an additional frame of reference for considerations of vertical distributions of elements in the marine environment. CONCLUSION The chemical factors governing the dispersion of species in- troduced into the marine environment are in need of better definition, especially with regard to coastal environments. The involvement of many elements in primary plant produc- tivity is not known. The comparative chemistries of oxic and anoxic waters warrant much additional work. The na- tures and roles of suspended inorganic and organic phases are poorly understood. An elaboration of such problems would provide firmer bases from which to systematize the general dispersion problem. In particular, we see need for chemical investigations along the following lines: 1. Elemental concentrations should be sought for the marine plants that are most important in the fixation of carbon in both coastal and open ocean areas. The relative importance in the coastal oceans of attached plants to planktonic species with regard to the uptake of specific ele- ments warrants attention. 2. For the coastal ocean, the relative effects of interac- tions of introduced species with nonliving particulate or- ganic and inorganic phases and with viable organic phases should be ascertained. 3. The distribution of heavy metals in the open ocean and their distribution among dissolved, colloidal, and particulate states should be determined. SUMMARY Although initially incorporated in the biosphere, or dis- solved in seawater or associated with particles, materials in- troduced into the ocean are eventually removed from seawater and deposited on the ocean bottom. The available data are systematized for certain important radionuclides in seawater and sediments. Marine environments may be divided into coastal ocean (above the continental shelf) and open ocean, where the water is deeper than 1,000 m. Physical and biological con- ditions are different in these ocean areas, and these in turn influence the behavior of the elements in the ocean. Ele- ments are classified as (A) conservative elements, (B) those exhibiting well-developed and regular variability in concen- tration with depth or ocean basin or both, and (C) those whose concentration is independent of depth or ocean basin. Elements in class A are generally unreactive in sea- water. Elements in class B are usually involved in biological cycles. Elements in class C are not well understood. Ele- ments have also been classified according to their residence time and reactivity in seawater. The coastal ocean is characterized by rapid mixing of substances, partial retention close to coast of solids and certain reactive elements, relatively intense biological activ- ity, and abundance of particles. Many reactive elements be- come associated with particles and are deposited near the continents. Open ocean waters are characterized by relative defi- ciency of nutrients in surface waters and relative scarcity of organisms and particles. Elements with long residence times tend to accumulate in open ocean waters. Sediments depositing in deep ocean areas are less important as a reser- voir for reactive elements than are sediment deposits in coastal areas. Downward transport and subsequent decom- position of particles is the dominant cause of depletion of the element in surface waters and enrichment in subsurface waters. SELECTED REFERENCES These references were selected to provide background infor- mation about marine chemistry and an introduction to the literature. Barnes, C. A., and M. G. Gross. 1966. Distribution at sea of Colum- bia River water and its load of radionuclides, p. 291-301. In Disposal of radioactive wastes into seas, oceans, and surface waters. IAEA, Vienna. Burton, J. D. 1965. Radioactive nuclides in sea water, marine sedi- ments, and marine organisms, p. 425-476. In J. P. Riley and G. Skirrow [ed.] Chemical oceanography. Vol. 2. Academic Press, New York. Chester, R. 1965. Elemental geochemistry of marine sediments, p. 23-80. In J. P. Riley and G. Skirrow [ed.] Chemical ocean- ography. Vol. 2. Academic Press, New York. Culkin, F. 1965. The major constituents of sea water, p. 121-162. In J. P. Riley and G. Skirrow [ed.] Chemical oceanography. Vol. 1. Academic Press, New York. Dietrich, G. 1963. General oceanography-An introduction. Wiley, New York. 588 p. Duursma, E. K. 1966. Molecular diffusion of radioisotopes in inter- stitial water of sediments, p. 355-369. In Disposal of radioactive wastes into seas, oceans, and surface waters. IAEA, Vienna. Goldberg, E. D. 1965. Minor constituents in sea water, p. 163-196. In J. P. Riley and G. Skirrow [ed.] Chemical oceanography. VoL 1. Academic Press, New York. Harvey, H. W. 1960. The chemistry and fertility of sea waters. Cambridge Univ. Press, London.

146 Radioactivity in the Marine Environment Home, R. A. 1969. Marine chemistry. Wiley (Interscience), New York. 568 p. International Atomic Energy Agency. 1966. Disposal of radioactive wastes into seas, oceans, and surface waters. Vienna. 898 p. Kautsky, H. 1966. Possible accumulation of discrete radioactive elements in river mouths, p. 163-173. In Disposal of radioactive wastes into seas, oceans, and surface waters. IAEA, Vienna. Lowton, R. J., J. A. Martin, and J. W. Talbot. 1966. Dilution, dis- persion and sedimentation in some British estuaries, p. 189-204. In Disposal of radioactive wastes into seas, oceans, and surface waters. IAEA, Vienna. Osterberg, C. L., N. Cutshall, V. Johnson, J. Cronin, D. Jennings, and L. Frederick. 1966. Some non-biological aspects of Colum- bia River radioactivity, p. 321-333. In Disposal of radioactive wastes into seas, oceans, and surface waters. IAEA, Vienna. Riley, J. P., and G. Skirrow [ed.]. 1965. Chemical oceanography. Academic Press, New York. 2 vol. Templeton, W. L., and A. Preston. 1966. Transport and distribution of radioactive effluents in coastal and estuarine waters of the United Kingdom, p. 267-288. In Disposal of radioactive wastes into seas, oceans, and surface waters. IAEA, Vienna. Turekian, K. K. 1965. Some aspects of the geochemistry of marine sediments, p. 81-126. In J. P. Riley and G. Skirrow [ed.] Chemical oceanography. Vol. 2. Academic Press, New York. Turekian, K. K. 1969. The ocean, streams, and atmosphere, p. 297- 323. In K. H. Wedepohl [executive ed.] Handbook of geo- chemistry. Vol. 1. Springer-Verlag, Berlin. Wedepohl, K. H. [executive ed.]. 1969. Handbook of geochemistry. Springer-Verlag, Berlin. 2 vol.

Chapter Six MARINE SEDIMENTS AND RADIOACTIVITY E. K. Duursma, M. G. Gross Sedimentary particles have a substantial capacity to remove radionuclides from the ocean, either depositing them on the bottom out of contact with man or concentrating them on particles that may be picked up by filter-feeding organ- isms, some of which are used by man for food (Templeton and Preston, 1966; Chapters 7 and 8, this report). The role of marine sediments in these processes has been studied less than most other aspects of marine radioactivity, as was al- ready pointed out by Koczy and Rosholt (1962). This is perhaps surprising but understandable. Marine sediments deposited at great depths and far from waste-disposal locali- ties contain little fallout and waste activity. The explana- tion is that a soluble radionuclide introduced into the sea will have a fate similar to that of the natural elements (see Chapter 5) and that the yearly input of radionuclides into marine sediments is related to the residence time of stable elements in seawater. For the deep ocean this input to the sediments has remained quite small since the mid-1950's, when man-made radionuclides first appeared in large amounts in the oceans. Because of the diverse composition of marine sediments, it is still difficult to predict a complete quantitative picture for areas with highly contaminated seawater. It is possible, however, to approximate satisfactorily the general distribu- tion and deposition behavior to be expected and the relative magnitude of the effects, based on laboratory experiments and field studies. To understand the role of sediments in the uptake of radionuclides, it is necessary to know the capacity of sedi- ments and sedimentary particles, in relation to their physical and chemical properties, to sorb (or desorb) radionuclides. It is also necessary to know in which areas of the ocean sediments are being deposited or in which large quantities of suspended sediment might sorb and remove radionuclides from seawater. These problems will be discussed in relation to the known mechanisms of sediment-radionuclide inter- actions. In addition, a preliminary evaluation of the possible role of sedimentary materials in removing radionuclides from contaminated seawater will be discussed. SORPTION OF RADIONUCLIDES BY MARINE SEDIMENTS Sediment Composition Sediments can sorb radionuclides of a variety of elements. Our limited knowledge prevents accurate predictions, but important relationships and general principles can be dem- onstrated. For instance, several investigators have reported that radionuclides are more concentrated in fine-grained than in coarse-grained sediments (Hamaguchi, 1962; Noshkin and Bowen, 1965;Kautsky, 1966;Templeton and Preston, 1966). Experiments by Duursma and Eisma (un- published) with various radionuclides and mostly fine- 147

148 Radioactivity in the Marine Environment IU ' V .x * Ce(X) * g* y V Pm(v) io5- S S^ o ? Co(A) + l?j. + Ru(o) fit ® ® Fe(+) ®^i Zr(®) f— ifiio4- o 0 9^ Q Zn(0) x *fe u. u o 0,03- z o T| x Cs(X) F— Z> . . * cc • *• . Sr(») c/5 0 10-3 1- ia' 1 io io2 MEDIAN DIAMETER (MICR0NS) 103 FIGURE 1 Distribution coefficients (K) for ten radio- nuclides as a function of the median grain size (D$Q), in microns, in various marine sediments. Determined with the thin-layer technique (Duursma and Bosch, 1969); filters with 10-mg sediment are exposed to 150 ml seawater enriched with radionuclides. The marine sediments used were sampled by French, German, U.S., Soviet, Italian, and Finnish research vessels at the following locations (Duursma and Eisma, unpublished): Area Depth (m) Location Mediterranean 2,710 43°N,7'/4°E Black Sea (used oxygenated) ? 44° N, 35tt°E Caribbean I 15°N,69°W Romanche Trench 7,200 0°N, 18°W Pacific 5,210 18°N, 132°W Atlantic 4,150 27°N, 39°W Antarctic 3,970 50° S, 127tt°W Gulf of Trieste 30 46° N, 14°E Red Sea (Discovery Deep) 2,080 20° N, 38° E Baltic 220 57°N, 20° E grained marine sediments demonstrated a lack of correla- tion between radionuclide sorption and grain size. Uptake in these sediments (Figure 1) may be the result of poor sorting, as there were relatively large amounts of fine par- ticles in all the samples. On the other hand, the selective sorption of fine particles is not always directly proportional to the specific surface area. It depends on the specific min- erals and radionuclide involved (Table 1). Nevertheless, in Irish Sea sediments near the pipeline outlet, at Sellafield, the amount of radioactivity in different grain-size fractions was nearly proportional to the specific surface area of the grains (Jones, 1960; Jefferies, 1968). In sediments with large cation exchange capacities, as calculated from the mineral composition (Duursma and Eisma, unpublished), the radionuclides were somewhat more strongly sorbed (Figure 2). The sorption capacity for a spe- cific nuclide, calculated as the distribution coefficient,* did not on the average differ by more than a factor of four between the sediments with extremely high and extremely low cation exchange capacities. Increased carbonate con- tent is slightly correlated with decreased radionuclide up- take (Figure 3); in part, this can be explained as resulting from lower cation exchange capacities of the carbonate minerals. Sediments with higher iron content have larger distribu- tion coefficients (Figure 4). Iron in sediments may occur as concretions or particle coatings (Carroll, 1958), enhancing the formation of sorption complexes for 106Ru (Jones, 1960). Sorption of 59Fe, however, shows no obvious corre- lation with the total iron content of the sediment used in the experiments. Radionuclide Sequence Some radionuclides are more strongly sorbed to sediment particles than others; the specific sequence varies with dif- ferent sediments. In some Atlantic, Pacific, and Indian Ocean and Mediterranean Sea sediments, the sequence of sorption of radionuclides from seawater (from those weakly sorbed to those strongly sorbed) was (Duursma, 1969): 45Ca < 90Sr < U, Pu,i 37Cs < 86Rb < 65Zn < 59Fe, 95Zr/Nb, 54Mn< 106Ru< 147Pm. The radionuclides 144Ce and 60Co fall somewhere between 59 Fe and 147Pm in the ease with which they are sorbed. However, in the Irish Sea, 106Ru is less sorbed than, for *For the purpose of this discussion, some relevant terms are defined as follows: Distribution coefficient-A dimensionless ratip of the amount of bound radionuclide per unit volume of dry sediment to the amount of radionuclide per unit volume of seawater. The values presented are apparent-equilibrium values and are about 0.4 to 0.5 the value of similar coefficients defined on a weight basis. The larger the distribu- tion coefficient, the stronger the binding or sorption of the radio- nuclide to the sediment. Diffusion coefficient- A diffusion coefficient for a medium not in motion and in which the transport of radionuclides is caused by concentration gradients in the interstitial water; side reactions such as sorption are included.

Marine Sediments and Radioactivity 149 TABLE 1 Distribution of Sorbed Radionuclides and Total Radioactivity in Different Size Fractions in Three Types of Sediment" Sediments Size Fractions Distribution Coefficients for Each Size Fraction (value X 102) Origin f % Weight *>Sr 137Cs 106Ru 59Fe 65Zn ^Co 147Pm 54Mn 95Zr/Nb 144Ce Dutch >64 51.1 + 4.9 0.0 0.0 0.0 0.0 0.0 5.8 0.0 0.0 0.3 0.4 Wadden 32-64 21.7 + 4.5 0.0 1.6 3.1 76.0 0.0 3.7 0.0 1.4 42.0 3.3 Seaa 16-32 9.5 ±3.2 0.0 5.4 0.0 53.0 260.0 65.0 0.0 35.0 66.0 120.0 8-16 5. 6 ±2.0 0.0 12.0 7.3 370.0 490.0 59.0 0.0 8.0 1,040.0 950.0 4-8 8.4 ±1.9 0.0 16.0 5.2 510.0 380.0 430.0 320.0 480.0 1,220.0 540.0 <4 3.7 ±0.8 26.0 6.2 4.7 540.0 112.0 220.0 280.0 97.0 670.0 124.0 Mediterranean >64 0 - 5.2* - - - - - off 32-64 3.0±1.3 15.0 2.8 41.0 15.0 68.0 380.0 24.0 19.0 117.0 32.0 Monaco" 16-32 15.2±4.2 2.5 1.3 11.0 101.0 140.0 540.0 0.8 41.0 290.0 73.0 8-16 36.5 ± 8.6 7.6 1.4 22.0 118.0 150.0 730.0 4.5 61.0 150.0 82.0 4-8 39.1 ± 7.5 9.3 2.3 34.0 183.0 140.0 820.0 101.0 76.0 310.0 147.0 Sediments <4 6.2±1.9 5.9 0.5 7.8 63.0 97.0 140.0 130.0 23.0 160.0 41.0 Size Fractions Origin M % Weight Radioactivity MCi/g dry weight Irish Sea 100-200 26.0 2.0 X10-4 ^ Pipeline 50-100 27.4 3.0X10^ outlet, Sellafieldc 20-50 10-20 22.6 14.0 9.0 X1O-4 2.5 X10'3 ),• Specific Radioactivity = 1.3±0.4xl0-%Ci/cm2 4-10 6.1 4.6 X10-3 <4 4.0 6.2X10-3 J "The Dutch Wadden Sea and Mediterranean sediments were suspended in radionuclide-enriched seawater for one month (30 g sediment in 20 liters of seawater); size fractions were separated by sedimentation techniques (Duursma and Eisma, unpublished). 6For 65Zn, another Mediterranean sediment was used with 3.3%, >64/u; 7.6%, 32-64 M; 49.2%, 16-32 ft; 24.9%, 8-16 M; 12.7%, 4-8 M; and 2.3%, <4M. ' i in. Irish Sea sediments were exposed in the field to radionuclides discharged near Sellafield (Jones, 1960). example, 95Zr/Nb (Preston, personal communication), a result that agrees with the figures given for the sediments mentioned in Table 1. Similar results have been obtained for freshwater sediments (Carder and Skulberg, 1964), ex- cept that i 37Cs was more strongly sorbed in fresh water. Details of the mechanisms controlling such sorption se- quences are still unknown; there is, however, an extensive literature on ion-exchange processes in soils and minerals (see Robinson, 1962, for a discussion and for references). The weak sorption of 90Sr and 45Ca on marine sediments suggests that these radionuclides are involved in exchange reactions with their stable isotopes, which are relatively abundant in seawater. For other nuclides, however, the concentrations of their stable isotopes in seawater are so low that it is unlikely that the stable forms of the elements involved play any significant role in ion exchange processes. At low concentrations, the concentration of the stable car- rier typically does not greatly influence the amount of sorp- tion of 60Co and 65Zn (Duursma and Bosch, 1970) indicat- ing that exchange reactions with other ions dominate over ion exchange with stable isotopes of the same element. Processes causing sorption of radionuclides on sediment particles seem to be affected by the chemical properties of the radionuclides involved and by the physicochemical and biological conditions of the sediment (pH, reduction- oxidation state, zeta-potential, bacterial activity) rather than by the bulk sediment composition. Jenne and Wahlberg (1968) studied the processes by which 90Sr, 60Co, and 137Cs are bound to Clinch River sediments downstream from the discharge from Oak Ridge National Laboratory. They found that 90Sr was associated with carbonates pre- cipitated in situ. The radionuclide 60Co was associated with Mn and Fe oxides, and 137Cs was taken up by incorpora- tion in the lattice of certain clay minerals. Similar mineral- related reactions, each one highly specific, probably control radionuclide uptake by marine sediments.

150 Radioactivity in the Marine Environment FIGURE 2 Distribution coefficients (K) for ten radionuclides and the sediments mentioned in Figure 1, as a function of calculated cation exchange capacity (*?), in milliequivalents per 100 grams (meq/100 g). The cation exchange capacity is calculated as follows (Duursma and Eisma, unpublished): n v-, P — ) * , 100 *—' 100 where n = % fraction <2 n, P = % clay mineral in the fraction of <2 M, and k = 100 meq/100 g for montmorillonite 25 meq/100 g for chlorite and illite 10 meq/100 g for kaolinite 0 meq/100 g for other minerals. Sr r105 -10' -I03 -I02 -10 IO•2 10•' 1 10 10O ION EXCHANGE CAPICITY (q) (MEQ/100 GRAM) ]O'i FIGURE 3 Distribution coefficients (K) for ten radionuclides and the sediments men- tioned in Figure 1, as a function of the total Ca + Mg content of the sediments (Duursma and Eisma, unpublished). 10'-. CD £ 10'- 10 Ru o. ""ttttt/Q. QP4*A\sssss\.\\. . '' 5 10 15 20 25 30 35 5 10 15 20 25 30 35 CALC1UM (%)

Marine Sediments and Radioactivity 151 I03- o o ' 03- m cc l- co '/ Sr(.) 12345 0 1 2 3 4 5 IRON (%) FIGUKE 4 Distribution coefficients (K) for ten radioisotopes and the sediments mentioned in Figure 1, as a function of the total Fe content of the sediments (Duursma and Eisma, unpublished). Equilibria In sorption experiments, the apparent equilibrium distribu- tion between seawater and sediment for different radio- nuclides is approached at different rates, depending on the nuclides involved. The nuclides 137Cs, 90Sr, and 65Zn ap- proach the apparent equilibrium much faster than do 59Fe and 95Zr/Nb. Cobalt-60,106Ru, 144Ce, 54Mn, and 147Pm exhibit apparent reaction rates intermediate between these extremes. Adsorption resulting from slow precipitation may cause the slow reaction (several days) for 59Fe and 95Zr/Nb. Sorption by chemical binding or ion exchange may cause the relatively high reaction rates observed (several hours) for 137Cs, 90Sr, and 65Zn (Duursma and Eisma, unpub- lished). The term equilibrium implies that the reactions are re- versible, but this is often difficult to demonstrate for radio- nuclides strongly held by sediments. The activity of radio- nuclides released to seawater may be so small that it is difficult to determine accurately, as in the experiments involving 144Ce- and 147Pm-labeled sediment and nonradio- active seawater. For 60Co, it was possible to demonstrate loss of the nuclide from the sediment (Figure 5). In this case, the sorption-desorption process is due to several quasi- reversible reactions in which the rate of nuclide loss is demonstrably slower than the rate of uptake. This is ex- plainable, in part, as resulting from a set of several processes, each with different rates, so that the overall reaction does not proceed at the same speed when it is reversed. 0^642 60CO <C x>(VALUES AFTER I WEEK UPTAKE) I/ .X T0TAL SUSPENSI0N (Iml) FIGURE 5 Desorption of 60Co as a function of time from Mediterranean sediment, previously suspended for one week in seawater (stirring) containing °"Co, and then resuspended in seawater containing no ''"( o (Duursma and Bosch, 1970). - TREND T0 0RIGINAL \ D1STRIBUTI0N C0EFFICIENT' SEDIMENT (Img in 1m1) 10 15 TIME (DAYS) 20

152 Radioactivity in the Marine Environment Experiments on the uptake of 45Ca by sediments show a stepwise set of reactions (Duursma and Bosch, 1970). The first step is relatively rapid, occurring in the first few sec- onds or minutes of contact. The final step is much slower- measurable only after days or weeks-and apparently in- volves diffusion into the sediment particles. Furthermore, there is a suggestion of an intermediate step that requires from a few minutes to a day for completion. This interme- diate reaction may result from migration of the radionuclide in small capillaries between broken ends of crystal lattices or from diffusion through thin layers of organic matter coating the particles (Chave, 1965), or both. Anomalies Under different experimental conditions, Duursma and Bosch (1969), using the same radionuclide and the same sediment, found that it was often easier to obtain reproduc- ible results with one radionuclide than with another. Specif- ically, 60Co sometimes gives different results when dif- ferent methods are used, a situation not encountered with other radionuclides. Even using the same sediment, the amount of 60Co uptake depends on whether the sediment was in contact with quiet (unstirred, thin layer, or settling) seawater or suspended in rapidly stirred water. Each experi- ment came to apparent equilibrium states that were dis- tinctly different (Figure 6). Changing the experimental conditions afterwards from quiet to stirred caused the de- sorption of some 60Co and the establishment of another apparent equilibrium. However, a change from stirring to quiet conditions did not result in increased sorption of the nuclide. Hence, in using experimental data to predict sorp- tion processes in nature, it is necessary to consider whether the experimental conditions approximate the expected field conditions and whether the history of the water-sediment systems was similar prior to the introduction of the radio- nuclide. Anoxic Systems Thus far, the discussion has been concerned only with sedi- ments under oxygenated conditions. Anoxic conditions (complete oxygen depletion) are rare in open-ocean water (Richards, 1957) but moderately common in interstitial water of sediments, especially those deposited in highly productive nearshore areas. Consequently, we must consider radionuclide behavior under anoxic conditions. Experiments using Black Sea sediment under oxidizing and reducing con- ditions show that there is little change in the distribution coefficients of many radionuclides (Table 2). Typical 60 Co 0--0 T0TAL (S0LUTI0N *SEDIMENT) • 'S0LUTI0N (Im1) X X SED1MENT (I mg) + +BLANK FILTER * a s n o Ul 6000- (e) * S* x • x x l • 3 2000- AI z cc 10000- U OL o- ~* & .9. -9 9"9_9— = 9 — 9 £ 6000. z o 2000- Az 11- 10000- 3 £ > t- ^—-^ n Jkv*-*-x •*--=—=? A /" o j^s < 2000- B 1 234567 5 10 15 (WEEKS) TIME (DAYS) FIGURE 6 Results of a seven-week experiment in which Mediter- ranean sediment was exposed under various conditions to seawater containing "^Co, followed by a 19-day experiment in which the conditions were changed. Note the relatively rapid adjustment to the new conditions in experiment B, but not in experiment A1 (Duursma and Bosch, 1970). TABLE 2 Distribution Coefficients of Sorption for 10 Radioisotopes as Determined under Oxygenated and Anoxic Conditions in Black Sea Sediment" Radionuclide (As Chloride) Distribution Coefficients Oxygenated* Anoxicc *>Sr 1.2X102 4.7 X103 l37Cs 1.4X103 3.1 X103 106Ru (nitrate) 3.4 X104 9.0 X103 59Fe 1.2X104 4.0 X103 65Zn 3.0 X103 3.2 X104 «>Co 6.0X103 1.4 X104 147Pm 4.8 X104 5.2X103 MMn 4.6 X104 4.3 X103 95Zr/95Nb 1.2X104 9.0 X103 144Ce 9.0 X104 1.2X104 "Duursma and Eisma, unpublished data. 6Dissolved oxygen concentration: 5 ml O2/liter. cHydrogen sulfide concentration: 2-3 mg 1 1 ,S, liter