U and Th speciation in river sediments

U and Th speciation in river sediments A. Martinez-Aguirre3, M. Garcia-León3, M. Ivanovich* b ªFacultad de Fisica, Universidad de Sevilla, Apdo.1065, 41080 Sevilla, Spain bAEA Technology, Harwell Laborato,y, Oxon, OXll ORA UK Abstract A study of the distribution of sorne natural radionuclides in different chemical fractions of river bottom sediments has been carried out. Toe study has shown that the majority of the total U in sediments is located in the non-residual fractions, while Th is more likely to be present in the residual ones. This difference between U and Th reflects largely a much higher mobility of U relative to Th in surface and near-surface environments. Coprecipitation with amorphous ferromanganese oxyhydroxides is the main process of U and Th incorporation from the water column into the soil particles. Moreover, the distribution of the radionuclides and the analysis of the Th/U mass ratios in different chemical fractions of sediments has made an unequivocal connection between the enhanced U content in river sediments and the wastes discharged into the rivers by the operation of fertilizer industries. Keywords: Radionuclide pollution; Uranium (U); Thorium (Th); River sediments; Fertilizer industries l. lntroduction River sediments may be regarded as, at least, a temporary sink for much of the material which passes through the various aquatic chemical and biological cycles operating on the earth's surface. Thus, sediments become an environmental host for many of the waste products discharged by society. Toe effects of these man-made emissions, in sorne situations, can be sufficiently strong to aff ect highly the composition of the deposited * Corresponding author. sediments. Once a substance has been incor porated into a sediment, its ultimate fate depends on a number of very complex factors. An element may be considered to be locked permanently into a sediment component or it may later be released and take part in various biogeochemical reac tions. Toe fundamental distinctions that have been made in sediments geochemistry is that between material brought to the site of deposition in a solid form and that brought in a dissolved form. In that sense the distinction of elements between residual and non-residual types is particularly ap propriate to the investigation of radionuclide poi- '.,,:{� httion [1]. Thus, processes involved in the incor poration of elements from polluted waters results j�:;their location in the non-residual fractions and thus, they can usually be considered to be at least potentially mobile, i.e. environmentally reactive in the chemical and biological processes which occur in the sediments/interstitial water complex. On the other hand, elements located in residual frac tions can usually be considered, at least to a first approximation, to be immobile, i.e. environmen tally unreactive. Thus, the residence of an ele ment in a deposit may be only temporary, and their release into the ecosystem represents a po tential hazard. The extent to which this release occurs depends largely on the manner by which this element is bound to the sediment and on how particular kinds of bonds react to various physico-chemical conditions. The above considerations have led to the study of the distribution of uranium (U) and thorium (Th) in different fractions of sediments collected at the Odiel and Tinto rivers, which are located in Southwest Spain. Such sediments are clearly af fected by the wastes released by two fertilizer fáctories located in the left margin of the Odiel rlver as well as by the phosphogypsum piles stored at the right margin of the Tinto river [2]. 2. Samples and radiochemical procedures Sampling was carried out during the summer, 1989. Sediment samples were collected along the Odiel and Tinto river basins and stored in plastic b¡igs; then, at the laboratory they were dried and 1'9W.dered before the analysis. The location of the S!lmpling stations are shown in Fig. 1. Three sedi �ts from the Odiel river were studied: up stream from the fertilizer industries (S01), around o:n� of the fertilizer industries (S02) and down sfream from them (S03). In the Tinto river, sedi ments from two locations were studied: one ;;\IQUnd the phosphogypsum pile area (STl) and inother downstream from this zone near the con flu�nce with the Odiel river (ST2). Additionally, arióther sample (SOT) located downstream of the co'rtfluence of both rivers was studied. S:1bne sequential extraction of six sediment frac tio'ns (Table 1) has been implemented for each sediment sample (3,4]. The six sequentially leached fractions of the sediment are defined in Table 1. Fractions Fl -F4 are referred to as sec ondary phases expected to represent the non-re sidual part of the sediment, being either an end product derived from the river water/solid inter action and exchange processes or part of the solid capable of direct interaction with river water and thus sharing sorne geochemical and isotopic char acteristics with the liquid phase. Fractions F5 and F6 are termed residual phases and are not ex pected to necessarily interact on a short time-scale with the liquid phase, at least not under normal conditions found in the surface and near-surface environments. Once the fractions were separated, they were spiked with 236 U and 229Tb and the extraction of U and Th isotopes was carried out according to the methods given in [3,4]. U and Th activities were measured by a-spectrometry with surface barrier or ion implanted Si detectors. 3. Results and discussion Percentage distribution of U and Th in the different fractions (taking the content in the un fractionated sediment as 100% [2]) are shown in Figs. 2 and 3. With the exception of samples ST2 for U, the sum of the percentages in the different fractions agrees well with 100%. Analytical prob lems in chemical separations or heterogeneity in the sediment sample cannot be ignored. 3.1. Uranium Fig. 2 shows that U is mainly associated with non-residual fractions, particularly with F2 and F3 fractions. More than 40% of U is found asso ciated with the fraction F3 (an exception is sam ple STl with 30% in this fraction). This implies that the main process of U incorporation from the liquid phase (river water) into the sediments is the coprecipitation with amorphous ferroman ganese oxyhydroxides. This is a typical process observed in estuarine environments [5]. In sediment S02 (Fig. 2b), fraction F3 is still one of the main non-residual fractions containing >40% of the total U. However, in this sample a o N t � Phosphogypsum pi les *Fertlllzer factorles 3km 1 Fig. l. Map of the studied area showing the location of the sampling points, the fertilizer factories and the phosphogypsum storage area emplacements. high proportion of U (- 50%) is also found in fraction F6 (the primary mineral fraction). This sample (Fig. 1) was taken in the vicinity of a fertilizer factory which releases part of its solid wastes directly into the Odiel river [2,6]. Conse quently, this sediment is likely to contain material from the minerals used for the fertilizer produc- Table 1 tion. This material is expected to form a major part of fraction F6 of the sediment at this loca tion. Sample STl is also of sorne interest. At this location (Fig. 1) sorne material from the phosph gypsum storage area [2] is discharged directly into the Tinto river. Here, the largest fraction of total Definition of sediment fractions and reagents used for their extraction Fraction code Fl F2 F3 F4 F5 F6 Fraction Readily ion exchangeable Adsorbed ions in humic material and/or amorphous oxides Coprecipitated with amorphous ferromanganese oxyhydroxides Carbonates Crystalline ferromanganese oxyhydroxides Resistate (clay) minerals Reagent Magnesium chloride Sodium pyrophosphate Tamm's acid oxalate Sodium acetate Coffin's reagent .. u a 80....----------*U b 110---------.... �u e 80.-----------, .. u d*º .. u ·----------,, 80.-----------r 40 Fl Fl F3 F4 F5 F6 Fl Fl f'J F4 FS F6 Fl F2 FJ F4 F/J F6 Fig. 2. Percentage distribution of U in each sequentially extracted fractions of samples (a) S01, (b) S02, (e) S03, (d) STl, (e) ST2 and (O SOT. U resides in fraction F2 ( > 50%) followed by fraction F3 (- 30%). Thus, U is incorporated into the local sediment predominantly by adsorption from the liquid phase. A different chemical asso ciation in the wastes must be responsible for the 'anomalous' U distribution observed in this sta tion and related to the discharges from the phos phogypsum storage area [2, 7]. In Table 2, concentrations of U in µ,g/ g con tained in the main fractions of the sediments are shown. It is interesting to observe that sediment samples taken near the phosphate waste discharge area of the Odie) river (SOl, S02, S03 and ST2) have virtually identical U concentra tions in fractions F2 and F3 and these are notice ably higher than the rest of samples (STl and SOT). This difference is much clearer in the case of fraction F3. This can be taken as an evidence of the extension of the contamination of the Odie! river basin (even the lowest part of the Tinto river bed). Thus, the liquid wastes dis charged from the f ertilizer industries affect the river bed upstream and downstream from the industries, probably due to tidal movements of Table 2 U concentration in µ,g/ g in the main fractions of bottom sediments from the Odie! and Tinto rivers Fraction S01 S02 S03 F2 12.7 ± 0.8 12.1 ± 0.6 13.7 ± 0.9 F3 31.5 ± 3.6 37.5 ± 2.4 34.9 ± 3.5 F5 0.42± 0.04 2.86 ± 0.10 1.92 ± 0.06 F6 1.93 ± 0.07 40.8 ± 2.8 5.89 ± 0.15 the liquid discharges. In the case of the residual phases, it can be observed that samples S02 and S03 have higher U concentrations in fractions F5 and F6 than the rest of the samples. This differ ence is much clearer in the case of fraction F6 of sample S02. This sample is located in front of the fertilizer factories and presents a U concentration > 1 order of magnitude higher than the rest of samples. The U concentration in this fraction decreases downstream on the river bed. This can be taken as evidence that the solid particles from the fertilizer production characterised by a high U content in fraction F6 have reached stations S02 and S03 but are not sedimented further downstream in the estuary. 3.2. Thorium Fig. 3 shows the percentage distribution of Th (232 Th) in the six sequentially extracted fractions of sediments. It can be observed that Th has a higher association than U to residual fractions F5 and F6, always with > 30% in fraction F6. How ever, the non-residual fraction F3 appears to be another favoured fraction for Th. Thus, it seems that, as for U, coprecipitation with amorphous ferromanganese oxyhydroxides is the main process of Th incorporation from the river water to the sediment. Basically, we can make the same com ments as we did in the case of U. Thus, sample S02 presents a different percentage distribution than the rest of samples, with > 60% of Th in the mineral fraction F6. Moreover, fraction F2 does not seem to be a favoured fraction for Th. Only sample STl has -15% of Th in this fraction whereas the rest contain < 5% in the same frac tion. In Table 3, Th concentrations in µ,g/ g in the main fractions of the sediments are shown. It can be observed that, excepting sample S03, the Th ST1 ST2 SOT 8.69 ± 0.54 11.2 ± 0.8 6.86 ± 0.30 4.51 ± 0.26 33.3 ± 2.1 14.1 ± 0.7 0.35 ± 0.02 0.90± 0.04 0.92 ± 0.06 1.68 ± 0.13 2.77 ± 0.11 2.44 ± 0.14 content in the non-residual fractions F2 and F3 is quite similar in all stations. Station S03 has a higher concentration of Th in both non-residual fractions. This is more evident in the case of fractions F3, with - 1 order of magnitude higher concentrations in sample S03 than in the case of the rest of samples. It seems that the dissolved Th is quickly coprecipitated on the river bed. Thus, Th contained in the liquid discharges seems to affect only the sediments closest to the fertilizer industries. In the case of the residual fractions, samples S02 and S03 present higher concentra tions in both fractions than the rest of samples, whereas the rest have similar Th concentrations. Thus, it can be concluded that the river sediments near the fertilizer industries have incorporated Th derived from the liquid and solid wastes but it does not reach stations further upstream or downstream along the river. 3.3. Th / U mass ratios Interaction of non-residual fractions with solu tion is essentially reflected in the Th/U mass ratio. It is well known that Th is a particularly insoluble element in natural waters and it is usu ally found associated with solid matter. Thus, this ratio is typically much below unity in waters and above unity in solid particles. These facts can be clearly obtained by studying the values of such ratio in the main fractions of sediments from the Odiel and Tinto rivers which are presented in Table 4. In the case of non-residual fractions F2 and F3, and excepting fraction F3 of sample S03, the Th/U mass ratio is much below unity and typical of liquid phases. In fact, these values are similar to those found in water samples from the same rivers [2). Toe exception of phase F3 of sample S03 seems to be related to the fact that Th is , TII a 80,------------. ,n d 801------------, 20 Ft F2 F3 F4 F5 F6 ,Th b 80.----------- % Tlo e. ao---------- Fl F2 F3 F4 F5 F6 ,T11 C 80.-----_.;;..----....... ..,n f 80...---------- FI F2 F3 F4 FS F6 Fig. 3. Percentage distribution of Th in each sequentially extracted fractions of samples (a) S01, (b) S02, (e) S03, (d) ST1, (e) ST2 and (0 SOT. Table 3 Th concentration in µ.g/g in main fractions of bottom sediments from the Odie! and Tinto rivers Fraction S01 S02 S03 STl ST2 SOT F2 0.79 ± 0.08 0.74 ± 0.06 1.31 ± 0.08 1.10 ± 0.12 0.59 ± 0.05 0.70 ±0.05 F3 7.71 ± 0.51 3.97 ±0.22 38.4 ± 4.0 2.02 ± 0.14 5.05 ±0.39 5.88±0.36 F5 1.89 ± 0.13 1.41 ± 0.07 6.22 ±0.33 0.68±0.08 1.44 ± 0.11 0.96±0,07 F6 4.83 ± 0.22 12.7 ± 1.1 15.8 ± 0.80 4.36± 0.31 4.21 ± 0.33 6.61 ±0.63 Table 4 To/U mass ratio in main fractions of bottom sediments from the Odie! and Tinto rivers Fraction S01 S02 S03 F2 0.062 ± 0.007 0.061 ± 0.006 0.096 ± 0.009 F3 0.245 ± 0.032 0.10 ± 0.009 1.10 ± 0.16 F5 4.5 ± 0.5 0.493 ± 0.030 3.24 ± 0.20 F6 2.50 ± 0.15 0.311 ± 0.034 2.68 ± 0.15 quickly coprecipitated on the surface of sus pended matter which are deposited downstream at such a point and not further downstream. It is also interesting to observe that this ratio is much lower in fraction F2 than in F3. This could be interpreted by the fact that Th is not easily adsor bed onto solid particles compared with U. In fact all stations present similar values of this ratio in fraction F2, independently of the concentration of U and/or Th in the fraction. In the residual fractions F5 and F6 the Th/U mass ratio for all samples, except sample S02, has values above unity and close to the typical value for near surface rocks of 3.5 [8]. Toe exception of both residual fractions of sample S02 confirms the U enrichment relative to Th typical of the phos phate mineral utilised in the fertilizer production [7]. 4. Conclusions Toe U and Th distribution in different frac tions of bottom sediments from the Odiel and Tinto rivers has provided much information on their environmental behaviour. Under estuarine conditions U and Th occur mainly in the non-re sidual fractions although Th tends to reside more in residual fractions than U. These diff erences between U and Th reflect largely the much higher solubility (and hence mobility) of U related to Th in surface and near-surface environments. Toe sediment fractionation and analysis presented here has made an unequivocal connection ST1 ST2 SOT 0.127 ± 0.016 0.053 ± 0.006 0.102 ± 0.009 0.448 ± 0.040 0.152 ± 0.015 0.417 ± 0.032 1.94 ± 0.25 1.60 ± 0.14 1.04 ± 0.10 2.60 ± 0.27 1.52 ± 0.13 2.71 ± 0.30 between the enhanced U content and respective Th/U mass ratios in river sediments close to the fertilizer factories and the wastes discharge point in the Odiel and Tinto rivers. References [1] R. Chester and S.R. Aston, The partitioning of trace metals and transuranics in sediments, in Techniques for Identifying Speciation in Aquatic Environments, IAEA, Vienna, 1981, pp. 173-193. [2] A. Martinez-Aguirre, M. Garcia-León and M. lvanovich, Toe distribution of U, Th and 226Ra derived from the phosphate fertilizer industries on an estuarine system in Southwest Spain, J. Environ. Radioactivity, 22 (1994) 155-177. [3] M. Ivanovich and R.S. Harmon, Uranium Series Dise quilibrium: Applications to Earth, Marine and Environ mental Sciences, Clarendon Press, Ox:ford, UK, 2nd edn, 1992. [4] A. Martinez-Aguirre, Radiactividad natural en diverses compartimentos naturales de Andalucía, PhD thesis, Seville University, Spain, 1991 (in Spanish). [5] E.R. Sholkovitz, Flocculation of dissolved organic and inorganic matter during the mixing of river water and seawater, Geochim. Cosmochim. Acta, 40 (1976) 831-845. [6] A. Martinez-Aguirre and M. Garcia-León, Natural ra dioactivity enhancements by human activities in rivers of the South of Spain, J. Radioanal. Nucl. Chem. Lett., 155 (1991) 97-106. [7] A. Martinez-Aguirre, M. Garcia-León and M. Ivanovich, Identification and effects of anthropogenic emissions of U and Th on the composition of sediments in a river/estuarine system in Southem Spain, J. Environ. Radioactivity, 23 (1994) 231-248. [8] J.J.W. Roger and J.A.S. Adams, Handbook of Geochem istry, Wedepohl Springer Verlag, Berlin, 1969.