Trace metals in sediments from the "Ria de Huelva"

Toxicological and Ent!ironmt'nldi Chemi ... rrv. VoL,,_ 3t<i2, pp, 27~-183 Reprints available directly rrom the puhtishcr Photocopying permitted by Ii\.'cnsc noly 1991 Gordon and Brea<:h Scieo<:c Publishers S.A. Printed jo lhe UnÍled Kingdom TRACE lVlETALS IN SEDIMENTS FROM THE "RIA DE HUEL V A" M. PEREZ, J. USERO and L GRACIA Departamento de Ingeniería Química y Ambiental. E.T.S.!.I. Avda. Reina Mercedes, s/n. 41012 Sevilla, Spain F. CABRERA Instituto de Recursos Naturales y Agrobiología de Sevilla, C.S.l.c. Apartado 1052. 41080 Sevilla, Spain The Ria de Huelva is an estuarine zone into which flow the Odiel and Tinto rivers, polluted by trace metals. Total contents of Al, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Pb, Cd and Hg were determined in the < 630 and < 63 pm fractions of bed sediments. Total trace metal contents in the coarsest fraction are lower Ihan in the finest, although both are highly correlaled for mosl of the metals, allowing an easy grain size elTect correclÍon. Total concentrations of Cu, Zn, As, Pb and Hg are very high, having enrichmenl faclor indexes rangíng from 35-111. Important proportions of these are in easily soluble and reducible physico-chemical forms, which are readily available for living organisms. KEY WORDS: Estuary, trace metal pollution, sediment, grain size elTect, melal specialion. INTRODUCTION The rivers Tinto and Odiel discharge into the Ría de Huelva, after flowing through an area known for its mining activities from times immemorial. The waters of both rivers are acidic (pH occasionally reaches values between 3 and 4), and contain large amounts of heavy metals and other trace metals from erosion and mining activity. In addition, the Ría de Huelva (formed by the union of the mouths or the rivers Tinto and Odiel and the Padre Santo Canal), is one of the most industrialízed areas of Southern Spain, and consequently receives the discharge of industrial and urban wastes. The high concentration of heavy metals in waters or lhe Ría de Huelva has negative effects on other activities of the region, parlicularly on the shellfish industry, which on occasion has been paralysed. The Ría de H uclva is an estuarine zone in which processes of flocculation and precipilation or melals take place beca use of the mixing of river and sea waters.1 ,2 Therefore, the rate or mosl of the dissolved and particulate metals discharged by rivers and waste outlets are sediments.3 ,4 Sediments are not only sinks of metals, but the latter can be released from them by chemical reaction and biological activity.3--6 Thererore, scdimcnl composition is often considered a good index for heavy metal and trace metal pollution of the aquatic environment. 3,4,7 275 276 M. PEREZ ET AL. Total element composition of a particular sediment, as determined by a total chemical analysis, is not indicative of its potential pollution. 6.8 Sediments tend to display marked metal variability, even within short distances, grain size being the most important factor causing this variability, as metals tend to concentrate in the fine fraction.3 ,4,7 Grain size effect can be minimized by different methods,4 the most effective being physical separation of a size range or fraction followed by a chemical analysis of the separate material. Since most of the metals are concentrated in the < 63 11m fraction and this is relatively easy to separate, numerous studies have already been made on this fraction.9 As sediment metals are distributed among a variety of physico-chemical forms which exhibit different chemical reactivities, the measurement of the total concentration of a particular metal pro vides little indication of potential interaction with abiotic or biotic components present in the environment.6• 1o Partitioning of sediment-bound metals could be determined by thermodynamic calculations or by chemical speciation,6 The Ría de Huelva is presently subjected to a Corrective Plan for Control of Industrial Waste Disposal. Data on sediment characterization of the Ría de Huelva are scarce,11.12 thus this study has two main objectives: (i) to characterize sediments of the Ría de Huelva, specially in respect of metal contents and to establish a simple method to correct grain effects, and (ii) to determine metal distribution in the different chemical forms in which they are in the sediments, to estimate their readiness to dissolve and the degree of bio-availability. EXPERIMENTAL Sampling Bed sediment samples were taken using an impact drag from a boat, introduced into plastic bottles and frozen. Samples were collected in January 1989, in two sites of the Ría del Odiel (samples 01 and 02), in two sites of the Ría del Tinto (samples TI and T2), and in five sites of the Padre Santo Canal (samples CI to C5). Figure 1 shows the location of the sampling sites. Analysis of Samples Samples were homogenized and fractions < 630 I~m and < 63 11m were separated by sieving, dried at 60 'oC and ground to a fine powder. Fraction < 630 11m was used for the sediment characterization. Sediment pH values were measured in 1 M KCl at al: 4 ratio. Total carbon (TC) and total sulphur (TS) were determined by mean of a LECOS SC-132 autoanalyzer with an IR detector. Inorganic carbon (IC) was determined by the Bernard calcimeter method and the loss of weight by ignition at 500 oC for 2 h. Grain size distribution was determined by sieving for particles size > 63 11m and by mean a Sedigraph 5100 equipment for <63 11m. TRACE METALS IN SEDIMENTS 277 2000 3000 ¿OOO -~~~l INDUSTRIESL.J INDUSTRIAL RESIDUES' SOOO 4000 3000 2000 1000 01 02 C2 C3 C4 8000m¡IKg ~_.~---_ ..... '---~--:--------- 7000 6000 5000 4000 3000 2000 1000 01 02 Cl C2. C3 C4 CS Figure 1 Sampling localions and evolutÍon of the total metal contents throughout the Ría de Huelva. 278 M. PEREZ ET AL. Table 1 General characterístícs of the sediments oC the Ría de Huelva Sampling pH TC IC TS LW Gra;n size distribution site ()~ % ?/o ~¿ 630-200 ¡;m 200-63 p.m <63Jlm <2Jlm % % % 0/0 01 6.0 0.57 <0.1 0.55 2.85 71.2 15.6 13.2 2.3 02 6.2 1.51 <0.1 1.28 6.30 15.2 36.5 48.3 15.8 TI 5.8 1.58 <0.1 2.52 5.50 13.0 38.8 48.2 27.2 T2 6.8 1.92 <0.1 2.84 5.90 6.1 34.7 59.2 24.4 el 6.2 1.43 <0.1 3.33 4.30 19.0 51.5 29.5 13.6 e2 7.2 1.95 <0.1 1.92 7.25 18.0 27.1 54.9 27.1 e3 6.8 2.21 0.25 1.73 5.60 23.8 27.7 48.5 19.9 e4 6.2 2.17 0.45 2.76 7.55 26.2 6.5 67.3 31.2 e5 8.0 0.84 0.80 <0.1 1.05 95.1 3.9 LO <LO Total metal eontents were determined both in < 630 and < 63 Jlm fraetions digesting the sample with a mixture of hot eone. HN0 3-HCl (1: 3). Residues left after acid attaek of < 630 Jlm samples were washed with deionized water, dried and weighed. Analysis of Cu, Co, Cd, Zn, Fe, Al, Ni, er and Pb in solution was performed by AAS. eold-vapor and Hydride generation AAS techniques were used for Hg and As respeetively. Metal speeiation of < 630 Jlm fraetion was performed by the method of Tessier et al.!3 with slight modifieations. Five fraetions were obtained: (a) Fl: exehangeable metals. The sediment sample was extracted for 1 h with 1 M MgCI2 at pH 7; (b) F2: metals bound to carbonates or specifieally adsorbed. The residue from (a) was leaehed for 54 h with 1 M NaOAc adjusted to pH 5 with HOAe; (e) F3: metals bound to Fe-Mn oxides. The residue from (b) was extraeted at 96 oC for 6 h with 0.04MNH 2 0H.HCI in 25% (v/v) HOAe; (d) F4: metals bound to organic matter and sulphides. The residue from (e) was extraeted at 85°e for 5 h with 30% H202 adjusted to pH 2 with HN0 3 and then at room temperature with 3.2 M NH 4 0Ae in 20% (v/v) HN0 3; (e) F5: residual metals. The residue from (d) was digested with a mixture of cone. HCI-HN03 (3: 1). RESUL TS ANO DISCUSSION General Characteristics of Sediments Sediments of the Ría de Huelva are moderately aeidie or neutral, exeept one at the end of the Padre Santo Canal (sampling e5), near open sea, whieh have pH 8 (Table 1). Inorganie carbon, le, contents are very low, in many samples being be10w the deteetion limit of the mcthod (Table 1). Sediment organic matter is estimatcd by the total carbon, Te, and the los s of weight on ignition, L W, (Table 1). Both parameters follow a similar trend throughout the sampling area. Organie matter eontents are high, typieal for a TRACE MET ALS IN SEDIMENTS 279 sedimentary deposit,14 except for 01 and C5, located In the border of the estuarine zone. Total sulphur, TS, values (Table 1), follow a similar trend to TC and LW, with lowest values in sites 01 and C5, although a peak is observed in Cl. Grain size distribution (Table 1) shows that sand content (630--63¡tm) for all the samples is greater than 33 %. On the other hand clay content ( < 2¡tm ) is greater than 13.6 % except for 01 and C5. Therefore these sediments can be classified between sand (sample C5) and clay loam (sample C4). Total Trace Metals in Sediments Table 2 shows total trace metal concentrations in both < 630 ¡tm and < 63¡tm fractions. It can be observed for all the metal s that total concentrations in the < 630 ¡tm fraction are lower than those in the < 63 ¡tm fraction. This is a result of the grain effect, as trace metal s tend to accumulate in the fine fraction.3•4. 7 In fact, it can be noticed that differences between values in those fractions are generally greater in the coarsest sediments (01 and C5), and lower in the finest (C4). Total values in both fractions are linearly correlated for all the elements, except for Co and Ni (Table 2), the significance levels for Cr, Mn, Fe, Cu, Zn, As, Pb, Cd and Hg (P<O.01) being higher than for Al (P<0.05). Correcting the values in the < 630 ¡tm on the basis of the percentage of dissolved sample, DS, (Table 2), the new values are al so linearly correlated with those in the < 63¡tm fraction, except for Ni. The new correlation coefficients are somewhat higher than the previous ones, except for Fe and As, their significance levels remaining egual, except for Fe which decreases to P < 0.1. Therefore, total trace metal content, except Ni, in the < 63¡tm fraction can be estimated by the content in the < 630 ¡tm fraction through the corresponding linear eguations whose slopes and intercepts are shown in Table 2. For Co this estimation improves correcting the value for the < 630¡tm fraction by the percentage of dissolved sample. Levels of all the trace metal s in the < 63 ¡tm fraction, except those for Al and Mn, are generally higher than the values in the average fossil shale 15 (Tables 2 and 3). The enrichment factor index, EFI, of trace metals in the sediments of the Ría de Huelva was calculated using the mean values of each trace metal and the average compositions of fossil shales 15 (Table 3). High EFI were found for Fe, Zn, Cu, Pb, As, Hg and Cd, indicating heavy pollution of anthropogenic origino For Cr, Ni and Co, the EFI were close to one. EFI values les s than one were found only for Al and Mn. Low EFI values for Al and Mn can be explained because of the incomplete recovery of the HNOrHCI digestion. Treating the remainder of the residues from the HNOrHCI attack with a mixture of conc. HF-HNO r HCI0 4, it was found that the former attack dissolved only a mean of 75 and 72 % of Al and Mn respectively, while for Cr the recovery was 90%, and for the rest of the elements was of the order of 99 %. Figure 1 shows the evolution throughout the Ría de Huelva of the contents of those trace metal s which showed highest EFI in the sediments. The six metals considered follow a similar trend, with two maxima near to the industrial areas A and B. The first, close to T2, can be attributed both to the river-estuarine Table 2 Total contents of trace metals in sediments. CaJcu1ated values were determined on the basis of the 630 Jl.m dissolved samp1e, DS (see text). Slope, intercepts, and correlation coeflicients, r, are for the linear corre1ation between the values for 63 Jl.m vs. 630 Jl.m or for 63 Jl.m vs. calcu1ated values. Sampling DS Al Cr Mn Fe Co Ni sites 01 02 TI T2 C1 C2 C3 C4 C5 Slope Intercept p 01 02 TI T2 CI C2 C3 C4 C5 Slope Intercept p % 28.7 43.3 630Jl.m 63Jl.m % 2.1 4.8 3.7 5.1 43.4 2.5 46.1 3.2 38.3 2.0 42.4 3.4 4.2 4.3 4.1 5.3 4.1 5.6 33.3 2.3 51.2 3.8 11.8 0.3 0.5952 2.9640 0.7613 Cal. 630 Jl.m 63 Jl.m mglKg 7.3 8.5 45 105 118 195 5.8 84 6.9 125 5.2 56 8.0 119 6.9 63 7.4 135 2.5 19 130 215 105 195 120 195 Cal. 630Jl.m 63Jl.m mglKg 157 273 280 495 330 415 194 192 272 195 146 132 281 290 189 280 264 330 161 100 255 265 240 684 340 410 Cal. 630Jl.m 63Jl.m % 976 762 7.3 12.1 7.9 10.3 442 11.2 423 10.2 345 10.0 684 9.3 841 7.5 645 11.9 847 0.6 15.5 14.8 14.5 13.0 13.2 15.4 0.4061 0.7344 0.8094 0.6527 0.3700 0.8058 1.8320 -22.54 -22.19 23.13 116.4 -1.5460 0.7676 0.9683 0.9786 0.8396 0.8834 0.8435 Cal. 630Jl.m 63Jl.m mglKg 25.4 18.2 15 30 17 24 25.8 23 22.1 23 26.1 15 21.9 20 22.5 19 23.2 30 5.1 4 35 28 30 25 31 32 Cal. 52 39 81 50 39 47 57 59 5 0.4235 0.5143 0.2166 3.795 5.1430 0.6175 0.3700 630Jl.m 63Jl.m mglKg 30 71 24 76 35 28 30 25 31 80 15 lOO 100 88 86 96 96 Cal. 105 55 81 61 78 59 93 156 127 0.5666 0.0456 - 15.43 85.08 0.3387 0.\391 <0.05 <0.05 <0.01 <0.01 <0.01 <0.01 <0.01 <0.1 17.89 0.8067 <0.01 Cu Zn As Pb Cd Hg 630 Jl.m 63 Jl.m mglkg Cal. 630 Jl.m 63 Jl.m mglKg Cal. 630 Jl.m 63 Jl.m mglKg Cal. 630 Jl.m 63 Jl.m Cal. mglkg 630 Jl.m 63 Jl.m mglKg Cal. 630 Jl.m 63 Jl.m Cal. mglKg 645 2120 2248 1150 2300 1495 2400 3453 1965 2550 1930 3380 4447 2830 4220 2875 4770 6236 2790 4280 1825 1975 1785 2440 3400 4765 1480 2600 3370 3310 3980 3170 4658 2190 3260 5360 2050 3430 4766 3330 25 212 58 4007 730 1410 4538 1050 1460 6521 1900 2630 6052 2300 3330 3864 1600 5165 1320 6156 1450 6504 1800 492 19 2160 1950 2300 2500 2544 690 1770 2425 1165 1700 4378 2260 3740 4989 3570 5670 4178 1565 3113 1785 4354 1250 3516 2120 161 II 2050 2830 2260 3000 2404 2691 1.8 5.4 7.0 10.8 5207 16.9 27.7 7744 14.2 22.7 4086 3.5 4210 6.2 3754 4.6 4141 8.1 93 0.4 5.6 9.4 6.8 10.7 0.7579 0.6095 0.8445 0.8298 0.7658 0.6122 0.6447 0.7700 0.6223 0.0868 0.9871 -585.2 503.4 0.9165 0.9243 <0.01 <0.01 -586.0 -138.7 0.9107 <0.01 0.9109 <0.01 -179.5 -39.75 -54.60 -435.00 0.9742 0.9056 0.9670 0.9705 <0.01 <0.01 <0.01 <0.01 <0.01 6.3 6.0 20.6 20.9 16.2 21.1 31.7 48.7 38.9 16.6 23.5 38.2 30.8 17.3 29.7 37.5 9.1 5.7 14.6 14.4 13.8 10.3 15.8 18.1 3.4 0.3 0.7439 -1.l430 0.9898 0.9824 9.566 0.8851 <0.01 <0.01 11.6 22.7 18.5 25.7 14.9 33.7 30.9 35.4 2.5 0.5413 5.293 0.8949 <0.01 w 00 o TRACE METALS IN SEDIMENTS 281 Table 3 Trace elements in the average fossil shale 15 and mean values and Enrichment factor index, EFI, in sediments of the Ría de Huelva Element Average Mean values EFl shale Ría de Hue/va mg/Kg mg/Kg Al 80000 46507 0.6 Fe 47000 133769 2.8 Mn 850 353 0.4 Zn 95 3329 35 Cr 90 158 1.8 Ni 68 89 1.3 Cu 45 3244 72 Pb 20 2879 144 Co 019 29 1.5 As 13 2217 171 Hg 0.4 23 58 Cd 0.3 12 40 boundary efTect and to the industrial waste discharges. The same is not observed in 02 probably because Odiel river water is less polIuted by trace metals. Minimum values are found in sampling site C5, in the estuary-sea border, where deposition, precipitation, coprecipitation, etc. of trace metal s have already finished. Trace Metals Speciation Figure 2 shows mean distribution of the trace metals in the five physico-chemical fractions obtained by the sequential extraction procedure. 13 As had been observed in previous studies, 3,4,16 each elements exhibits a distinctive partitioning pattern among the fractions. AlI the trace metals, except Zn, Cd and Cu, present the highest pro portio n in the residual fraction F5. Fraction F5 contains mainly primary and secondary minerals, which may hold trace metals within their crystal structure, therefore these metals are not expected to be released into solution over a reasonable time span under the conditions normalIy encountered in nature. 13 Copper shows a tendency to accumulate in fraction F4, as had been found by other authors,4 while Zn and Cd accumulate in F3 and Fl respectively. In these sediments another important proportion of trace metals is associated to Fe-Mn oxides, F3, as expected for estuarine sediments, from which they can be easily released by lowering pH and changes in the redox conditions. It is interesting that, although the proportion of trace metals in Fl is not high, except for Cd and Zn, Fl, the "exchangeable phase" is rather important as it represents very loosely bound trace metals and may regulate and/or reflect the composition of surface water. 3 For those sediments in which the content of caicium carbonate is negligible (Table 1), F2 represents specificalIy adsorbed trace metals which can be dissolved at pH 5. Neither is the proportion of trace metals in F2 high except for Zn. However, the sums of Fl plus F2 (which appears on the 0% 56.5 Ic~ B ~ §J 8 EJ §J ~ IA~ FRACTION 1 FRACTION 2 FRAcnON 3 FRACTION 4 Figure 2 Mean values of the specialÍon of trace metals in the sediments of the Ría de Huelva. 06c# ~ FRACTION 5 E '" 00 '" 283 M. PFRFZ /óT AL left of Ihe upper part 01' eat:h bar in Figure 2), rcat:h wnsiderable values (35.5-1 OA '\,) for some metals, c.g.: Zn, ed, Ph and Cu, which is extremely important beeause this reprcscnts thc proporlion or trace mClals readíly available ror living orgallisl11s. RejácNCt'.I' 1. 1. D. Burton, t;asllIarine Chemistry (Academic Press Inc., London. 1976), Chapo 1, pp. 1-36. 2. P. S. Liss. Easttwr;ne Chemistry (Academic Press Inc., London, 1976), Chapo 4, pp. 93-129. 3. U. Forsner. Metal Pol/utíon in the Aquacic Environment (Springer-Verlag, New York, 1983), Chapo E. pp. 197-270. 4. V. Salomons and U. Forstner, Meta/s in the Hydrocycle (Springer-Verlag, New York, 1984). 5. V. Salomons, Env, Technol. Lea 6,315 (1985). 6. A. Tessier and P. G. C. Campbell, Hydrobiologia 149,43 (1987). 7. A. 1. Horowitz and K. A. Elrick, Chemical and Biological Characterization of Sludges, Sediments, Dredge Spoíls, and Drilling Muds. ASTM STP 976 (American Society for Testíng and Materials, (988), pp. 114-128. 8. M. Verloo and A. Cottenie, Aled. Fac. Landouww. Rijkuniv. Genl, SO, 47 (1985). 9. U. Forstner and W. Salomons, Env. Technol. Let!. 1,494 (1980). 10. G. E. Batley, Aust. J, Mar, Freshw, Res. 38, 591 (1987), 11. X. Tomás, J. Obiols and L. Peiró, Afinidad 40,413 (1983). 12, D. Cortés, Uf Seminario de Química Marina (Servicio de Publicaciones, Universidad de Cádiz, 1987), pp. 83-91. 13. A. Tessier, P. G. C. CampbelI and M. Bísson, Ana/. Chem, 51, 844 (1979). 14. M. A. Rashid, G. Vilks and J. D. Leonard, Chem. Geo/. 15, 83 (1975). 15, K. K. Turekían and K. H. Wedepohl, Bull. Geo/, So, Am, 72, 175 (1961). 16. A. Tessíer, p, G. C. Campbell, J. C. Auclair and M. Bisson, Can. J, Fish. Aquat, Sci. 41, 1463 (1984).