当前位置:首页 >> 电力/水利 >>

Background values for evaluation of heavy metal contamination in sediments


Journal of Hazardous Materials 136 (2006) 111–119

Background values for evaluation of heavy metal contamination in sediments
M. Fukue a,? , M. Yanai a,1 , Y. Sato a,1 , T. Fujikawa a,1 , Y. Furukawa b,2 , S. Tani c,3
a

Marine Science and Technology, Tokai University, 3-20-1 Shimizu-Orido, Shimizu, Shizuoka, Japan b Faculty of Engineering, Nihon University, Tokusada, Tamuramachi, Kooriyama, Japan c National Institute for Rural Engineering, 2-1-6 Kannondai, Tsukuba, Japan Available online 18 January 2006

Abstract The contamination of sediments can be evaluated from comparison of the current concentration with the estimated background value. Although it is known that grain size affects background concentrations, there have been dif?culties in evaluating this effect. The speci?c surface area of sediments, which can be calculated from the grain size distribution, is introduced as a property index for correlation of background values. The results show that for sandy-to-sandy silt sediments, the background values of metals are expressed in terms of ?nes content, which can be correlated with the corresponding speci?c surface area. However, for silty clay or clayey silt sediments, which would have a high metal retention capacity, the level of the background may be constrained by the relatively low concentrations of metals in seawater. Finally, background levels of Zn, Cu and Pb are presented. ? 2005 Elsevier B.V. All rights reserved.
Keywords: Sediments; Heavy metals; Background; Fines content; Speci?c surface area

1. Introduction The background concentration of hazardous substances in sediments is an important factor for evaluating the degree of contamination. It has often been concluded that surface sediments were contaminated from the fact that deeper sediments show relatively low concentrations of pollutants, as shown in Fig. 1 [1]. However, this assumption does not hold if the grain size of sediments varies signi?cantly with depth, in which the background value also varies with depth, as shown in Fig. 2 [2]. This is because the background concentration of contaminants depends on many factors, including the particle characteristics, sorption/desorption capacity, and existing substances in nature. Background values were found to agree well with average values for gneisses rocks [3]. However, the background values for copper, zinc, and lead do not necessarily agree with values

Corresponding author. Tel.: +81 543 370921; fax: +81 543 349768. E-mail addresses: fukue@scc.u-tokai.ac.jp (M. Fukue), ysato@scc.u-tokai.ac.jp (Y. Sato), furukawa@civil.ce.nihon-u.ac.jp (Y. Furukawa), stani@nkk.affrc.go.jp (S. Tani). 1 Tel.: +81 543 34 0411, Fax: +81 543 34 9768. 2 Tel.: +81 249 56 8728, Fax: +81 249 56 8729. 3 Tel.: +81 298 38 7681, Fax: +81 298 38 7609. 0304-3894/$ – see front matter ? 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jhazmat.2005.11.020

?

previously obtained. Therefore, to allow an accurate estimation of background values, various techniques have been used. Fukue et al. [1] used carbonate content as a normalizing substance. On the other hand, Cobelo-Garc?a and Prego [4] obtained baseline ? relationships between the concentrations of iron and pollutants, whilst Din [5], Cortes¨ o and Vale [6] and Santschi et al. [7] used a aluminum (Al) to normalize the concentrations of heavy metals in sediments. Titanium has also been used as a normalizing element in the ?eld of geochemistry. All the techniques introduced here require data for non-contaminated sediment samples. Santschi et al. [7] mentioned that if sediment is richer in Al, it is composed of smaller-grained particles. This is because Al is a constituent of clay minerals, which can adsorb a greater amount of heavy metals than coarse particles. The fate of a pollutant depends on the surface activity factors of sediment particles, such as charge density, cation exchange capacity (CEC), speci?c surface area [8,9], and the equilibrium (natural) concentration of contaminants [10,11]. Since the particle characteristics of sediments are strongly in?uenced by the particle size, the grain size distribution may re?ect the true nature in practice. Many species of pollutants exist in nature and their concentrations vary widely. Therefore, it is necessary to estimate the background values to evaluate contamination. One easy and

112

M. Fukue et al. / Journal of Hazardous Materials 136 (2006) 111–119

lished, whereby the grain size distribution of sediments can be expressed by a single constant value. This value is correlated to the background value of heavy metals in sediments. When the background, BG , is known, the degree of pollution, Pd , is de?ned as: Pd = C ? BG BG (1)

where C is the current concentration of a substance or element [1]. 2. Concept of ideal speci?c surface area The particles of actual soils or sediments show a wide range of shapes and sizes. With regard to size, they can be classi?ed as gravel, sand, silt or clay fractions. The characteristics of soil particles have been assumed to be a function of the speci?c surface area, because most reactions, such as sorption, desorption, and interactions between particles, occur on the surface of the particles. Since it is not easy to determine the experimental speci?c surface area, a calculated speci?c surface area is used in this study. This calculated value is considered to be one of the property indices. Soil particles have been assumed to be spherical in conventional grain size analysis. Therefore, it is not uncommon to use the speci?c surface area for particles assumed to be spherical. The calculated speci?c surface area is called “the ideal speci?c surface area” to distinguish it from an experimentally measured value. The speci?c surface area of a sphere is de?ned by: SS = S ms (2)

Fig. 1. Metal concentrations and constant background in a homogenous sediment pro?le.

economical method may be to use the grain size distribution, because it re?ects the characteristics of particles, such as the speci?c surface area, which may govern the electrical charge density, CEC, etc. [9]. For example, if the size of a particle is very small, it may be a clay mineral, which has a relatively high charge density and high CEC. These characteristics mostly result from the high speci?c surface area. To express the grain size distribution of soils, representative indices such as the clay content, coef?cient of uniformity, and grain size at 10% ?nes have been used [12]. These values are signi?cant when they are used in correlation with a particular property of soils or sediments. For example, the grain size at 10% ?nes has been used to predict the permeability of sands [13]. In this study, to characterize the grain size distribution, the concept of “an ideal speci?c surface area” is estab-

where SS is the speci?c surface area, de?ned as the summation of the surface area of the individual particles divided by the total mass of the sample, S is the surface area, and ms is the mass of a spherical particle. Furthermore, sediments are assumed to be composed of spherical particles. Since sediments consist of a large number and various sized particles, they are separated into groups, as shown in Fig. 3, and thus the speci?c surface area of

Fig. 2. Illustration of variable background concentrations with depth.

Fig. 3. Grouping of sediment particles for establishing ideal speci?c surface area.

M. Fukue et al. / Journal of Hazardous Materials 136 (2006) 111–119

113

sediments is given by: SS = αi SSi (3)

where αi is the weight percent of group i. 3. Evaluation of SS 3.1. Grain size distributions Various types of sediments were arbitrarily chosen to obtain the ideal speci?c surface area, SS, as shown in Fig. 4. Fig. 4a shows the results of the laser diffraction scattering method for particle-size analysis, which is based on two theories, i.e., Fraunhofer diffraction theory and Mie scattering theory [14]. The abrupt drops near 0.001 mm may result from disagreement between the two theories, because the analysis for particles smaller than 0.001 mm was carried out using the Mie theory. At present, there is no assurance that results using the Mie theory are correct, because there is no comparable analysis. There is another type of analysis that can be applied to a small amount of sample. Fig. 4b shows results obtained using X-ray extinction [15]. This method is applied to particle sizes greater than 0.001 mm. Therefore, distributions for smaller particles (colloid sizes <0.001 mm) were assumed by extending the curves obtained. In this case, the curves were smoothed, as

Fig. 5. Method for the determination of the mean grain size of groups.

shown in the ?gure. The reason why these analyses were used is explained below. Conventional grain size analysis requires approximately 50 g of sample, but a weight of only 4–6 g is enough for the X-ray method and 0.03–0.06 g for the laser method. When a sediment sample is obtained using a tube, the amount of sample available for grain size analysis is usually within several g. Both test methods are presented later. 3.2. Calculation of SS For calculation of SS, the sediment particles were assumed to be spherical. If a sediment sample is separated into 10 sequential groups of equal weight, the grain size distribution of the sediment can be obtained, as shown in Fig. 5. In this case, αi = 0.1 in Eq. (3). In Fig. 5, the speci?c surface area of each group, SSi , can be calculated using the mean grain size of each group. In this study, the mean grain size of a group was de?ned as grain size for average percentage ?nes. Because the mean grain size de?ned is based on weight percent, as well as speci?ed grain sizes, i.e., D10 and D60 used in geotechnical engineering [12], calculation using this value provides an accurate SSi . For the calculation of SSi , the density of particles was assumed to be 2.67 g/cm3 . After all the SSi values are obtained, SS can be calculated using Eq. (3). The SS values calculated are given with the ?ne contents (FC) in Table 1. Table 1 shows that the SS increases with increasing ?ne contents, where FC is de?ned as the content of clay-silt fractions smaller than 0.075 mm. 3.3. Ideal speci?c surface area and background values of heavy metals Fig. 6 shows relationships between SS and FC, using the grain size distribution curves shown in Fig. 4a and b. Two types of curves based on the experimental and smoothed grain-size distribution curves show different patterns resulting from the minimum grain size. The SS represented by the curve A –E corresponds to a smoothed grain size distribution, which is greater than the SS given by the curve A–D, based on the total experimental grain-size distribution. These SS values are examined later.

Fig. 4. (a) Grain size distribution of typical marine sediments, analyzed using the laser diffraction scattering method. (b) Smoothed grain size distributions of marine sediments used to determine relationships between the speci?c surface area and ?nes content.

114

M. Fukue et al. / Journal of Hazardous Materials 136 (2006) 111–119

Table 1 Fine content and ideal speci?c surface area obtained from various grain size distributions Sediment Fine content, FC (%) SS for experimental curve (m2 /g) SS for smoothed curve (m2 /g) A 96.5 1.465 – A 95 – 24.98 B 88 1.212 – B 80 – 8.93 C 69 0.906 – D 54.5 0.548 – C 43 – 1.02 D 28 – 3.77 E 0 – 0.008

An example of calculation of the SS for individual groups is presented in Fig. 7, for which sample A was used for the calculation. Fig. 7 shows that the group of smallest particles has considerably higher SS in comparison to other groups with larger particles. This means that the SS value of sediments is governed by smaller particles and the SS value for larger particles can be neglected. Therefore, abrupt drops in grain size distribution obtained using the Mie theory give relatively small SS values. 3.4. Theoretical background concentration The intensity of reactions or phenomena on the surface of sediment particles depends on the area of the boundaries between the solid and seawater. The area of the boundaries can be expressed in terms of the speci?c surface area. Under a constant charge density and constant concentration of a substance, the amount of adsorption of an element may be assumed to be

proportional to the SS. Then, BGP ∝ SSact (4)

Introducing a proportionality constant kP , we can express this as: BGP = kP SS (5)

where kP ( g/m2 ) has dimensions of retained amount per unit area. If we can correlate SS with FC, a relationship between BGP and FC can be obtained. 4. Estimation of background concentrations 4.1. Speci?c surface area and plasticity index If the sorption capacity is proportional to the SS of particles, interactions between particles must be a function of SS. Fig. 8 shows correlation between SS and the plasticity index, IP , for various marine sediments. In general, IP varies with the interactions between particles, and is used for the classi?cation of soils in geotechnical engineering. Although some scattering is evident in Fig. 8, it is apparent that IP increases with increasing SS. Thus, SS can re?ect variations in the interactions between particles. 4.2. Measurements 4.2.1. Measurement of metal concentrations Various sediment samples from Seto Inland Sea and Osaka Bay, Japan, were obtained using a gravity tube corer and a grab sampler. The samples were dried at 110 ? C. Dry sediment samples were digested according to the US EPA 3051 guideline: 0.1 g of sediment were digested in 2 mL of 65% HNO3 and 0.6 mL of 48% HF in Te?on bombs using a microwave oven.

Fig. 6. Relationship between speci?c surface area and ?nes content.

Fig. 7. Speci?c surface area in relation to the total weight of sample and mean grain size.

Fig. 8. Relationship between plasticity and ideal speci?c surface area.

M. Fukue et al. / Journal of Hazardous Materials 136 (2006) 111–119

115

Chemical analyses of lead (Pb), copper (Cu) and zinc (Zn) were carried out using inductively coupled plasma-atomic emission spectrometry (ICP-AES) after digestion. 4.2.2. Grain size analyses For samples obtained using a tube sampler, the grain size distributions of samples were determined using laser and Xray methods [14,15] for the reasons stated earlier. The X-ray extinction method for grain size analysis utilizes Stokes’ law, as well as the conventional method. Therefore, the density of particles is required for the calculation of grain size and percentage ?nes. The settling velocity of sediment particles in a small cell (6 cm × 4 cm × 0.2 cm) ?lled with distilled water is measured using X-rays instead of light for the photoextinction method. The sample weight required is approximately 5 g. The percentage ?nes of the sample is calculated using the Lambert–Beer law for the relationship between the concentration of suspended solids and the intensity of X-ray extinction. However, this method cannot be used for carbon black, graphite and high-molecularweight compounds. The measurement instrument can be automatically controlled. On the other hand, the laser-diffraction scattering method utilizes the Fraunhofer diffraction theory for particle sizes greater than 0.001 mm, with the diffraction of particle re?ection expressed as a function of grain size. A laser is used to irradiate a small cell (5 cm × 3 cm × 0.3 cm) ?lled with sample solution mixed with sediment particles and distilled water. The weight of sample required is approximately 0.05 g. For particles smaller than 0.001 mm, the Mie scattering theory is used. According to this theory, the scattering of light is a function of the index of refraction of particles. The index of refraction for soil particles is approximately 1.08. 4.3. Metal adsorption and background concentrations Bay sediments often consist of a silty soil, which usually has FC greater than 80%. Fig. 9 shows typical metal concentration pro?les for two cores obtained from two site locations in Osaka Bay, Japan. The FC of the sediment samples is 95 and 98% through from the top surface to the bottom portion. Therefore, we usually assume that the minimum concentration of Zn for these sediment cores is approximately 130 mg/kg, and that an approximate thickness of 50 cm from the top surface is contaminated with Zn. This implies that the deeper sediments are not contaminated. This minimum value obtained can be used as the background value of zinc at these site locations. Similar background concentrations of Zn were obtained in Tokyo Bay [16] and on the coast of Galicia (northwest Spain) [3]. On the other hand, the background values of Zn for sandy sediments were 55 ± 11 mg/kg [4]. The sand content of the sediments varied approximately from 50 to 80%. In sediments, clay minerals are among the most important sorbents for metal cations. The adsorption of metals to clay particles can be explained by the Langmuir isotherm. According to the theory and experimental results [11,17], the adsorption of metals can proceed until the saturation of sorption occurs. This is one of the reasons why sediments can be contaminated by dis-

Fig. 9. Pro?les of Zn concentration for ?ne sediments from two locations in Osaka Bay.

charge of metals through human activities. In fact, contaminated ?ne sediments can still show adsorption capacity if they are put in highly concentrated metal solutions [18]. The amount of metal adsorption is strongly dependent on the adsorption capacity of the sediment particles, the concentration of the metal solution and the pH value of the solution [11,17]. The pH of ordinary seawater and marine sediments is within a narrow range, i.e., between pH 6.5 and 8.0. If the concentration of a metal solution is high, the amount of adsorption is also high. Therefore, the background level is determined by the natural metal concentration of soil water or seawater. Fig. 10 shows the general features of the amount of metal adsorbed under a wide range of equilibrium concentrations of metal solution based on the discussion above. In Fig. 10a, the amount of adsorption increases with increasing equilibrium concentration. Many studies have examined the characteristics of this type of adsorption theoretically and experimentally [11]. Under natural conditions, the equilibrium concentration of elements is relatively low, and more or less constant, indicated as a shaded zone in Fig. 10a. In this case, the higher the adsorption capacity, the higher is the amount of adsorption. However, if the equilibrium concentration is relatively low, the amount of adsorption for clayey soil may be constrained by the equilibrium concentration, as indicated by the shaded column in Fig. 10a, even if the soil has a high sorption capacity. If adsorption capacity is relatively low, this can constrain the amount of adsorption, as shown in Fig. 10a. The state can be described as “saturation of adsorption”. For coarse sediments under a certain equilibrium concentration, the amount of adsorption is relatively low because of the small speci?c surface area. Thus, the background level of metal for sediments can be interpreted from two aspects, as follows: (a) If a sediment has a relatively high adsorption capacity, BG is constrained by the metal concentration (equilibrium concentration) in water.

116

M. Fukue et al. / Journal of Hazardous Materials 136 (2006) 111–119

Fig. 10. Interpretation of background values based on the relationship between adsorbed metal concentration and equilibrium concentration in terms of sediment type. (a) Sorption characteristics of particles under a narrow range of natural equilibrium condition. (b) BG –SS relation.

(b) If a sediment has a relatively low adsorption capacity, the equilibrium condition is nearly saturated with respect to adsorption. Therefore, BG is strongly in?uenced by the adsorption capacity, which can be a strong function of grain size and its distribution. The grain size characteristics can be expressed by SS or FC. These can be interpreted by the two different curves shown in Fig. 10b. 4.4. Concentration and FC diagram In this study, FC is conveniently used instead of SS. This is because FC is usually obtained from the grain size analysis and because the concept of FC is familiar to scientists and engineers. For ?ne sediments, there are other factors more important than the grain size characteristics, as previously explained. Furthermore, if the background level is completely characterized, no further analysis is required. Zn concentrations and FC are shown in Fig. 11, with data for both contaminated and non-contaminated sediments. Since the background value should theoretically be a minimum value, evaluation of the background level is achieved by obtaining the lower limit of the scattering data. In this study, a theoretical approach was taken, using the relationship between BG and SS given by Eq. (5) and the relationship between FC and SS. The proportionality constant is determined from the best ?t of the lower limit. If we use the curve A –E in Fig. 6, a proportionality constant kP of 7.8 g/m2 is obtained. On the other hand, if the curve A–D in Fig. 6 is used, the best ?t gives a kP value of 62 g/m2 . The BG values calculated using the kP and SS values in Table 1 are indicated as solid and open squares in Fig. 11, corresponding to kP of 7.8 and 62 g/m2 , respectively. As seen in Fig. 11, the lines passing through the solid squares and open squares express almost the lower limit of plots, except for ?nes content higher than 90%. The lines obtained are called “background baseline (BB)” in this study. The BB of Zn is very small for sand or gravel and increases with increasing FC. The BB of Zn for FC between 20 and 60% is approximately 35 mg/kg. This range is slightly lower than the background

level for sandy sediments from the Galician Ria (NW Spain) [4]. Beyond a FC of 80%, the two curves start to part, as shown in Fig. 11. It seems that both lines deviate from the lower limit. This deviation may be due to the following reasons: (a) there might be an error in measurements of concentration or in calculation of the BB using the assumptions made in this study; (b) BG is constrained by adsorption under equilibrium concentrations, which can be explained by Langmuir adsorption; and (c) the Mie theory may overestimate the grain size of sediments.

Fig. 11. Deviation of Zn concentration and background level for various marine sediments. Solid squares: kP of 7.8 g/m2 using the smoothed grain size distributions, Open squares: kP of 62 g/m2 using the grain size distributions from the Mie theory.

M. Fukue et al. / Journal of Hazardous Materials 136 (2006) 111–119

117

Fig. 12. Deviation of Cu concentration and background level for various marine sediments.

Fig. 13. Deviation of Pb concentration and background level for various marine sediments.

Item (c) is more realistic, because divergence of the grain size distributions is apparent between the two theories used. Therefore, the smoothed grain size distributions were used for the following analyses. In this study, an empirical background level was de?ned for ?ne sediments with FC greater than ?80%, as indicated by the modi?ed line in Fig. 11. The modi?cation is reasonable, because from the interpretation made previously, the increase in BG with FC in this region should be lowered. If we accept the background level of Zn, the degree of pollution can be obtained using Eq. (1). Following a procedure similar to the case for Zn, background levels of Cu and Pb were determined, as shown in Figs. 12 and 13. Modi?ed lines based on the empirical results obtained from Osaka Bay, Seto Inland Sea and Tokyo Bay [16] are also given in Figs. 12 and 13. The results from Figs. 12 and 13 are summarized as follows: (1) Background levels based on the SS show good correlation with the actual data. (2) For ?ne sediments, modi?cation of BG for Cu and Pb is needed, based on the many reasons discussed. (3) The kP value for both Cu and Pb is 1.8 g/m2 . 4.5. Background values From Figs. 11–13, the background characteristics of sediments can be divided into four stages, as presented in Table 2. The sediments with BG values are divided into four types, i.e., coarse, coarse–medium, medium–?ne, and ?ne sediments. The boundaries between them are based on the BB shape. Note that the BG values presented in Table 2 are approximated.

5. Practical application 5.1. Sediments from Seto Inland Sea A contaminated sediment core was obtained from Kasaoka Bay in the Seto Inland Sea in Japan. A 140-cm-long core was taken by driving a stainless steel core tube into sediments. The tube has a length of 2.0 m and a sectional area of 100 cm2 [19]. The top 30 cm of the core was cut into 3-cm sections, and each section was kept in a clean glass container until testing. In the laboratory, the grain size distribution and heavy metal concentrations of each section were determined. The grain size distribution of the samples was obtained using the techniques described above. Elemental analyses were performed by ICP spectrometry. The sample digestion method has already been described. The sediments consisted of silty clay to clayey sand. The corresponding FC values varied from 95.3 to 55.5%. The water content of silty clay, i.e., the top sediments, was 183% and the ignition loss was 9.9%. The speci?c gravity of particles was 2.58. Sediments in this depth range are usually contaminated,
Table 2 Background levels estimated in this study Name FC (%) BG (mg/kg) Zn Coarse Coarse to medium Medium to ?ne Fine <20 20–60 60–90 >85 <35 35 35–100 100–150 Cu <8 8 8–14 10–30 Pb <8 8 8–13 10–30

118

M. Fukue et al. / Journal of Hazardous Materials 136 (2006) 111–119

Fig. 14. Estimation of background values for a sediment pro?le in Seto Inland Sea.

but the degree of pollution cannot be evaluated without the corresponding background concentration. Evaluations of the contamination are presented in Fig. 14. The background values for Zn, Cu and Pb were estimated using the ?ne contents and the relationships between FC and BG shown in Figs. 11–13, respectively. The BG value for Cu varied from 7 to 30 mg/kg, which is similar to values for Pb. Comparison of current concentrations and BG values shows that all the sediment samples were contaminated, as shown in Fig. 14. The lowest degree of pollution, de?ned previously, is 0.4 for both Zn and Pb, while the highest degree of pollution is 6.6 for Pb at a depth of 19.5 cm. 5.2. Clayey sediments In this section, clay content is examined as an index property for the contamination of ?ne sediments. A sediment core sample obtained from Osaka Bay showed a very high FC, i.e., 98%, but the clay content varied with depth, as shown in Fig. 15; the clay content varied from approximately 30 to 70% with depth, and the lower sediments contain more clay fractions. Therefore, it might be expected that the lower sediments would contain more heavy metals. However, the results obtained show the reverse trend, as demonstrated in Figs. 16 and 17. This is because of contamination of the shallower sediments. Fig. 16 shows the deviation of Cu concentration with clay content. If we take a constant and lower limit, a BG value of approximately 30 mg/kg can be obtained, as indicated in Fig. 16. This value is comparable to that for FC of 98% in Fig. 12. It is important to note that this BG value is provided for a FC of almost 100%.
Fig. 15. Clay content with depth for a core sample obtained from Osaka Bay.

Fig. 16. Concentration of Cu according to clay content for Osaka Bay sediments.

M. Fukue et al. / Journal of Hazardous Materials 136 (2006) 111–119

119

Fig. 17. Concentration of Zn according to clay content for Osaka Bay.

A similar situation can be cited for the case of Zn, as shown in Fig. 17. The Zn BG value for this ?ne sediment was assumed to be 150 mg/kg, which is also obtained from Fig. 11. Therefore, the results from ?ne sediment core samples indicate that the BG value near FC of 100% is considerably lower than that expected from the BB value, which was calculated using the speci?c surface area. This probably results from constraint by the equilibrium concentration, as explained above. 6. Concluding remarks In evaluating the contamination of sediments, background values of metals have often been considered to be constant. On the other hand, normalization techniques using different elements have been used to estimate the background value. In this study, a technique for estimating background values of some heavy metals in marine sediments is proposed using the grain size characteristics, i.e., ?ne content and speci?c surface area. The examination shows that the background value, BG , is expressed as a function of the ideal speci?c surface area, SS, de?ned in this study. The SS is correlated to ?ne content, FC. Finally, BG values of Zn, Cu and Pb are de?ned in terms of FC, dividing the sediments into four types, i.e., coarse, coarse–medium, medium–?ne, and ?ne sediments. The BG of Zn varied from almost zero to 150 mg/kg with FC. The BG values for Cu and Pb varied from almost zero to approximately 30 mg/kg with FC. The background values of other metals, such as cadmium, arsenic and chromium, can be determined with procedures similar to those proposed in this study. References
[1] M. Fukue, T. Nakamura, Y. Kato, S. Yamasaki, Degree of pollution for marine sediments, Eng. Geol. 53 (1999) 131–137.

[2] M. Fukue, Y. Kato, T. Nakamura, S. Yamasaki, Heavy metal concentration in bay sediments of Japan, ASTM Spec. Tech. Pub. 1293 (1995) 58–73. [3] E. Carral, R. Villares, X. Puente, A. Carballeira, In?uence of watershed lithology on heavy metal levels in estuarine sediments and organism in Galicia (north-west Spain), Mar. Pollut. Bull. 30 (1995) 604– 608. [4] A. Cobelo-Garc?a, R. Prego, Heavy metal sedimentary record in a Gali? cian Ria (NW Spain) background values and recent contamination, Mar. Pollut. Bull. 46 (2003) 1253–1262, www.elsevier.com/locate/marpolbul. [5] Z. Din, Use of aluminum to normalize heavy metals data from the estuarine and coastal sediments of the strait of Melaka, Mar. Pollut. Bull. 24 (1992) 484–491. [6] C. Cortes¨ o, C. Vale, Metals in sediments of the Sado Estuary, Portugal, a Mar. Pollut. Bull. 30 (1995) 34–37. [7] P.H. Santschi, B.J. Presley, T.L. Wade, B. Garcia-Romero, M. Baskaran, Historical contamination of PAHs, PCBs, DDTs, and heavy metals in Mississippi River Delta, Galveston Bay and Tampa Bay sediment, Mar. Environ. Res. 52 (2001) 51–79, www.elsevier.com/locate/marenvrev. [8] P. Alumaa, U. Kirso, V. Petersell, E. Steinnes, Sorption of toxic heavy metals to soils, Int. J. Hyg. Environ. Health 204 (2002) 375– 376. [9] R.N. Yong, A.M.O. Mohamed, B.P. Warkentin, Principles of Contaminant Transport in Soils, Developments in Geotechnical Engineering, Vol. 75, Elsevier, 1992. ¨ [10] O. Yavuz, Y. Altunkaynak, F. G¨ zel, Removal of copper, nickel, cobalt u and manganese from aqueous solution by kaolinite, Water Res. 37 (2003) 948–952. [11] J. Markiewicz-Patkawska, A. Hursthouse, H. Przybyla-Kij, The interaction of heavy metals with urban soils: sorption behaviour of Cd, Cu, Cr, Pb and Zn with a typical mixed brown?eld deposit, Environ. Int. 31 (4) (2005) 513–521. [12] K. Terzaghi, R.B. Peck, Soil Mechanics in Engineering Practice, John Wiley & Sons, New York, 1948. [13] A. Hazen, Physical properties of sands and gravels with reference to their use in ?ltration, Report for the Massachusetts State Board of Health, 1892, p. 539. [14] Y. Furukawa, T. Fujita, T. Kunihiro, M. Fukazawa, Investigation of particle size distribution of soil using a particle size analysis equipment automated by laser and its applicability to soil samples, Div. 3 JSCE J. 687 (56) (2001) 219–231 (text in Japanese). [15] Y. Furukawa, T. Fujita, T. Kunihiro, H. Tsuchiya, Y. Saito, Investigation of particle size distribution of soil using a particle size analysis equipment automated by X-ray method, J. Jpn. Geotech. Soc. 40 (2) (2000) 127–133 (text in Japanese). [16] E. Matsumoto, The sedimentary environment in the Tokyo Bay, Geochemistry 17 (1983) 27–32 (text in Japanese). [17] H.B. Bradl, Adsorption of heavy metal ions on soils and soils constituents, J. Colloid Interface Sci. 277 (2004) 1–18. [18] M. Fukue, M. Yanai, Y. Takami, S. Kuboshima, S. Yamasaki, Containment, sorption and desorption of heavy metals for dredged sediments, in: K. Adachi, M. Fukue (Eds.), Clay Science for Engineering, A.A. Balkema, Rotterdam, 2001, pp. 389–392. [19] M. Fukue, Y. Sato, M. Yanai, M. Nakamura, S. Yamasaki, Rehabilitation of contaminated marine sediments in relation to living things, in: Geoenvironmental Engineering, Geoenvironmental Impact Management, Thomas Telford, 2001, pp. 198–203.


赞助商链接
相关文章:
更多相关标签: