FEB – Fresenius Environmental Bulletin Founded Jointly by F. Korte and F. Coulston Copyright © by PSP – Parlar Scientific Publications, Angerstr

FEB – Fresenius Environmental Bulletin founded jointly by F. Korte and F. Coulston Copyright © by PSP – Parlar Scientific Publications, Angerstr. 12, 85354 Freising, Germany. Production by MASELL Agency for Public Relations, Angerstr. 12, 85354 Freising, Germany All rights are reserved, especially the right to translate into foreign language. No part of the journal may be reproduced in any form- through photocopying, microfilming or other processes- or converted to a machine language, especially for data processing equipment- without the written permission of the publisher. The rights of reproduction by lecture, radio and television transmission, magnetic sound recording or similar means are also reserved.

Printed in GERMANY – ISSN 1018-4619

CONTENTS

ORIGINAL PAPERS

POLYCHLORINATED BIPHENYLS IN HUMAN MILK 342 FROM SLOVAK MOTHERS J. Petrík, B. Drobná, A. Kocan, J. Chovancová and M. Pavúk

ORGANOCHLORINE PESTICIDE RESIDUES IN BREAST MILK 349 FROM WOMEN LIVING IN TRIPOLI, LIBYA A. Elafi, M.Rahmani and El H. Abdennebi

DETERMINATION OF THE PHOTOCHEMICAL LIFE TIME OF 353 PESTICIDES IN AQUEOUS MEDIUM WITH SOLAR LIGHT S. Guittonneau, I. Konstantinou, C. Emmelin, T. Albanis and P. Meallier

PHOTOSTIMULATED DECOMPOSITION OF SOME SELECTED PESTICIDES 357 IN AQUEOUS PHASE IN THE PRESENCE OF AEROSOL CONSTITUENTS P.N. Moza, K. Hustert, E.A Feicht and A. Kettrup

BIOSORPTION OF COPPER (II) BY POLLENS OF THYPHA LATIFOLIA L. 363 Ü. Danis, Y. Kaya and Ö.F. Algur

EFFECT OF RESIDUAL ASHES ON CPMAS-13C-NMR SPECTRA OF 368 HUMIC SUBSTANCES FROM VOLCANIC SOILS P. Conte, A. Piccolo, B. van Lagen, P. Buurman and M. A. Hemminga

POLYCHLORINATED BIPHENYLS IN SEDIMENTS 375 IN EASTERN SLOVAKIA J. Petrík, A. Kocan, S. Jursa, B. Drobná, J. Chovancová and M. Pavúk

A NEW METHOD OF MEASURING 381 GASEOUS SEMI-VOLATILE ORGANIC COMPOUNDS (PAHS) BY SOLID PHASE MICRO EXTRACTION (SPME) L. Lassagne, V. Jacob, P. Desuzinges, F. Tripoli, P. Kaluzny, P. Baussand and P. Foster

OXIDATION OF MANGANESE (II) WITH AIR IN WATER TREATMENT 386 S. Aydin, N. Tüfekçi and S. Arayici

IMPROVEMENT OF LOW HYDRAULIC PERFORMANCE OF 392 SEA OUTFALLS USING BELL MOUTHEDPORTS IN THE BLACK SEA EXAMPLE S. Nemlioglu

DENITRIFICATION OF WASTEWATER IN A FIXED FILM REACTOR 396 H. Timur

PESTICIDE RESIDUES IN GROUND AND SURFACE WATER IN BULGARIA 401 Z. Bratanova and K. Vassilev

340 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

SHORT COMMUNICATIONS

INVESTIGATION OF HEAVY METAL POLLUTION OF 405 TRAFFIC IN KEMALPASA-TURKEY H. Arslan

SERUM COPPER AND ZINC CONCENTRATIONS OF PATIENTS 409 WITH RHEUMATOID ARTHRITIS FROM KAYSERI-TURKEY M. Soylak and M. Kirnap

MEASUREMENT OF RADON CONCENTRATION IN 411 NORTHERN GREECE GROUND WATER A. Savidou and N. Zouridakis

GUIDE FOR AUTHORS 416

INDEX 419

341 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

POLYCHLORINATED BIPHENYLS IN HUMAN MILK FROM SLOVAK MOTHERS

Ján Petrík1, Beáta Drobná1, Anton Kocan1, Jana Chovancová1 and Marián Pavúk2

1Institute of Preventive and Clinical Medicine, Bratislava, Slovak Republic 2Department of Epidemiology, University of Iowa College of Public Health, Iowa City, IA, USA

SUMMARY

Polychlorinated biphenyls (PCBs) were analyzed in The experimental data support opinion that developing 67 breast milk samples of primiparae from six districts of embryos and infants are more sensitive to harmful effects the Slovak Republic. Dietary habits, demographic and of these compounds than adults [20,21]. The period of other relevant data on mothers and their children were breast-feeding is considered as the most important one collected by self-administered questionnaires. The highest with respect to amount of PCBs transferred to a human -1 levels (mean = 1318 ng.g lipid basis, n = 12) of PCBs were body [16,22]. Determination of these contaminants in found in the Michalovce district where these chemicals mother milk is an appropriate non-invasive method of risk were produced. They were statistically significantly assessment of breast-fed children [4,23]. higher than levels in other districts (P = 8.8 E-6). The mean PCB daily intake of breast-fed infants was PCBs were manufactured from 1959 to 1984 in -1 -1 assessed to be 5.5 mg.kg b.w.d (S.D. = 3.1). Chemko Co., Strážske in the Slovak republic. Technical formulations under trademarks DELOR, HYDELOR and KEYWORDS: DELOTHERM were intensively used as heat exchange PCBs; polychlorinated biphenyls; human milk; daily intake fluids, but also as paint additives [24]. The presence of PCBs was confirmed in considerable amounts almost in all environmental compartments in Slovakia, including food INTRODUCTION and human biological matrices [20,25,26]. Although the limits [21] are not exceeded today [27], PCB levels in the PCBs are a group of 209 individual congeners with Slovak general population are still rather high [20,28-32]. different chlorine substitution patterns [1]. The higher chlorinated PCB congeners are predominantly present in the most of biological samples because of their lipophilic- MATERIALS AND METHODS ity, chemical stability, and low biodegradation. The main exposure route to humans is food consumption, mostly Breast milk was obtained from volunteers - primiparas milk products, meat and fish [2-7]. PCBs are classified as (singleton infant) without any serious metabolic disorders, probable carcinogens - group 2A [8]. The toxicity of pregnancy and delivery complications - during 1992-94 in individual congeners is structure dependent. The most six model areas (districts) of Slovakia (Figure 1). Milk toxic ones are those with no or just one chlorine atom in samples were collected in accordance with the WHO ortho position of the biphenyl structure that may assume guidelines [9,33] and stored in sealed precleaned PTFE glass planar configuration [9-14] and elicit dioxin-like toxicity. vials at -35 oC till analysis. Informed consent documents The toxicity lessens with decreasing planarity between were obtained from all mothers. A questionnaire was phenyl rings [6]. distributed to the participating mothers after the delivery, including questions on maternal education, age, height and Exposure to PCBs by ingestion was connected with weight before pregnancy and after delivery, smoking, dietary adverse effects to cardiovascular system, liver, and also habits including use of dairy products and fish consumption, with hyperpigmentosis, chloracne [7-17], and early-life occupational history, use of medicines, as well as newborns’ exposure to PCBs can have long term consequences: sex, weight, and height. Data on the time of residence in the growth, cognitive and psychomotor development delay, district and possible sources of environmental contamination neurologic and immunologic anomalies, effects to thyroid (industrial production, location of waste-disposal sites, heavy hormone metabolism, etc. [16,18,19]. traffic etc.) were also collected.

342 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

FIGURE 1 - A map of the Slovak Republic with the areas studied: Michalovce District (PCB production 1959-84), Trebišov District (possible influence of the former PCB production), Velký Krtíš District (PCB contamination of dairy products in 1989-92), town of Bratislava (industry, traffic), Nitra District (agriculture), Myjava Region (highland area, home-made food) - situation in y. 1992.

Sample preparation for analysis of PCBs The known amount of milk sample (50-100 mL) was Operating conditions: injection mode, splitless for weighed into a separatory funnel, and 10 mL 10% natrium 1 min.; column temperature held at 110 oC for 1.5 min. oxalate solution, 50 mL ethanol and 20 mL diethyl ether and then raised to 200 oC at 30 oC.min-1, to 300 oC at were added. The mixture was shaken vigorously for 2.5 oC.min-1; carrier gas, He; head column pressure, -1 1 min. Then, 30 mL hexane was added, and the mixture 200 kPa; makeup gas, N2 at 50 mL.min. . was shaken for 5 min. The organic phase was transferred into another separatory funnel. The aqueous phase was extracted twice with 30 mL hexane. The combined Each GC peak in sample chromatogram was identi- hexane phases were extracted twice with 20 mL distilled fied by accurate coincidence with retention time of an water, separated from aqueous phase, dried, concentrated, authentic standard. Identified peaks were checked by and weighed as lipid. HRGC/LRMS-SIM (GC HP 5890 in combination with MSD HP5790B) in problematic cases. A method of A small amount of lipid (ca 100-200 mg) was dis- external standard was applied for quantitation. In solved in 1 mL hexane and transferred onto a combined accordance with the internal QA/QC rules a solvent blank Florisil-Silica Gel column (0.5 g Florisil, 1 g 44% H2SO4 was run in every batch of samples (usually 10). This on Silica Gel, 0.5 g Florisil, 1 g anhydrous Na2SO4). blank was not allowed to overstep 1/20 of the level of Residues were eluted with 10 mL 10% (v/v) dichloro- compounds analyzed in the sample at the same dilution. methane in hexane. Eluate was collected into 50 mL flask, To confirm recovery of the analytical method a sample of rotary evaporated to a volume of ca 0.5 - 1 mL and blown pure olive oil with known amount of PCB congeners was up to dryness under a gentle N flow. Prior to injection to 2 used. The samples were diluted so that the peak area of the gas chromatograph the residues were redissolved in determined compound was in the middle of calibration known amount of heptane. curve. Individual congener was quantitated if retention Two micro liters of treated sample in heptane solution time fit the retention time in standard solution ± 6 s. was injected. Analyses were performed on HP5890 gas Arithmetic means were used to summarize the concentra- chromatograph equipped with 63Ni ECD. A DB-5 (J&W tions of PCBs in breast milk. The influence of different Scientific, USA) capillary column, 0.25 mm id × 60 m and covariates was examined by linear regression and analysis 0.25 mm film thickness was used for the separation. of variance (ANOVA) techniques.

343 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

RESULTS AND DISCUSSION

The characteristics of the mothers and their children Dividing all samples to the urban and rural subgroups from this study are presented in Table 1. Body mass index (Table 1) we did not find any significant difference (BMI) values before pregnancy and after delivery in the between PCB data (a = 0.05, P = 0.24). This is because rural population samples significantly exceeded BMIs in the rural part of the Michalovce district samples (n = 4, the urban samples (P = 0.0012, P = 0.005, a = 0.05). mean = 1542 ng.g-1) markedly exceeds “urban” values The fat content of rural mothers’ milk did not differ of other areas (n = 33, mean = 838.7 ng.g-1). markedly from urban samples. If the Michalovce district samples were excluded The mean PCB levels in breast milk samples meas- from the tested group a statistically significant difference ured on a lipid basis presented in Table show minimal (a = 0.05, P = 0.01) of the average PCB breast milk val- differences among PCB levels from individual sampling ues between urban (838.7 ng.g-1) and rural (639.6 ng.g-1) locations with the exception of the Michalovce district. locations was observed. The PCB levels in samples from the Michalovce The average weight of babies at the delivery was district (mean = 1318 ng.g-1) were statistically signifi- 3198 ± 439 g. There was no significant difference in the cantly higher than levels in other locations (a = 0.05, weights of urban- and rural- born children, neither nor it P = 0.034, P = 0.018, P = 0.012 and Brat Nit Treb was between males and females. PVKrt = 0.0003). Only the Myjava area shows statistically insignificant difference in relation to the Michalovce Also, there was no correlation between weight of in- sampling site. This fact is due to a low number of Myjava fants and PCB content in corresponding mother milk samples (n = 5), as the mean concentration of PCBs was samples (R2 = 0.0019, n = 66). similar to other locations (Table 2).

TABLE 1 - Characteristics of the mothers and their babies. Mean ± S.D. Variable Urban (n=42) Rural (n=25) Total (n=67) Maternal age, years 22.2 ± 3.1 21.3 ± 2.8 22.1 ± 3.0 BMI before pregnancy, kg.m-2 19.6 ± 2.4 22.3 ± 4.2 20.6 ± 3.3 BMI before delivery, kg.m-2 24.4 ± 2.4 26.8 ± 4.5 25.3 ± 3.4 Boys’ birth weight, g 3196 ± 401 3283 ± 583 3231 ± 471 Girls’ birth weight, g 3101 ± 389 3282 ± 433 3163 ± 400 Fat content of breast milk, % 3.2 ± 1.3 3.5 ± 1.0 3.3 ± 1.2 * -1 PCB concentration , ng.g lipid basis 908.5 ± 360.2 784.0 ± 497.1 862.1 ± 417.3 * sum of PCB-28, 52, 101, 105, 118, 138, 153, 156 and 180

TABLE 2 - PCB concentrations in breast milk from different model areas (districts) of Slovak women.

-1 Mean concentration, ng.g lipid basis Bratislava Michalovce Myjava Nitra Trebišov Velký Krtíš Total Compound (n=14) (n=12) (n=5) (n=9) (n=12) (n=15) (n=67) PCB28 4.2 22.8 1.6 27.8 3.1 5.4 10.6 PCB52 0.94 11.3 1.0 22.6 1.9 5.0 6.8 PCB101 1.4 5.0 0.80 3.6 1.9 2.7 2.7 PCB138(+163) 241 337 226 179 207 155 224 PCB153 347 479 324 245 322 227 324 PCB180 208 301 192 188 208 141 206 PCB105 8.5 24.9 5.7 6.0 4.2 2.8 8.8 PCB118 40.7 65.8 33.9 31.9 26.9 21.6 36.8 PCB156 45.5 71.2 36.1 40.9 38.5 26.4 43.3 SPCB (9 congeners) 897 1318 822 745 814 587 862

344 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

The elevated PCB levels in the Michalovce district The Czech milk samples contained almost the same are consistent with the overall contamination in this area PCB levels. The influence of common food, animal feed caused by former PCB production (Figure 2). According and goods during existence of the former Czechoslovakia to our estimation [20], approximately 700 kg a year of is obvious from the comparison of these concentrations. technical PCB formulations were discharged into the wastewater canal from the beginning of production in Intake of PCBs is the highest in the first weeks of 1959-1974. Mothers from this area were born in 1966-75, baby’s life, because mother milk is the only source of i.e. they spent more than half of their live in a period of a nutrients for newborns. Occasionally, infants can be large PCB leakage into the environment and an expansion exposed to significant levels of PCBs. In spite of that, the of production (Figure 2). benefits of breast-feeding in accordance with the current knowledge outweigh the potential risk of exposure to Although food contamination in this district does not PCBs [9,43]. The acceptable daily intake (ADI) for PCBs -1 dramatically exceed levels in other areas of Slovakia at was recommended [44,45] to be 1 mg.kg b.w.. If this value the present time, the situation was substantially different in is applied, the human milk PCB concentration in this -1 the past [20,31]. No significant changes in dietary habits study should not exceed 160 ng.g lipid basis. However, all of before pregnancy and after delivery were reported in the milk samples analyzed exceeded the ADI recom- the questionnaires and all mothers consumed a non- mended. Prachar et al. report the average daily intake -1 vegetarian diet. No occupational exposure was recorded. during breast-feeding 2.56 mg.kg b.w. in the area of the In spite of the high PCB levels determined in dairy Bratislava city [37]. According to that paper about 60% of products in the recent past [34,35], Vel’ký Krtíš mothers’ human milk samples contained amounts of PCBs exceed- milk samples reported the lowest mean PCB concentration ing the proposed ADI. We should note that multiparas, for -1 (587 ng.g lipid basis). whom chemical concentrations in milk are lower [4], The difference between already published results were also included in calculations in that study. -1 (mean = 1930 ng.g lipid basis) [35] and our study can be attributed, among others, to the time factor, and removal 750 mL milk consumption a day and current data of of sources of contamination. Regarding high elimination infant body weight, milk fat content, and PCB concentra- half time of PCBs [4,36] it is more probable that the tion for each individual milk sample were used for calcu- different analytical methods used, the way of quantitation lation of daily intakes (Table 3). A comparison of the and the selection of volunteers were responsible for the Michalovce district data and results from other areas differences. The contamination in this district was showed a significant difference (a = 0.05, P = 0.0033). observed only in proximity to a few cooperative farms. The calculated daily intakes depend upon several fac- Prachar et al. [37] reported mean PCB value (sum of tors. The true levels are probably slightly different be- 6 indicator congeners) in breast milk from Bratislava city cause the total PCB concentrations in these results are at approximately half the level in comparison with data expressed just as the sum of nine congeners (Table 2) and shown in Table 2. The mean PCB levels from 50 mothers the gastrointestinal absorption of PCBs which usually -1 -1 ranges from 66 to 96% [4,46,47], was not included. were 462 ng.g , and 210 and 243 ng.g lipid basis for the second and later deliveries, respectively. On the other hand, calculated intakes characterize just The average PCB exposure in the Slovak general short period of sampling and not the whole lactation population is much higher than the data reported from period during which the concentration of fat and other countries [38-42]. This also can be seen from the subsequently the PCB levels vary [4,43]. Assessed PCB findings of the second round of WHO-coordinated daily intake of former Yugoslavian infants ranged from exposure study on the levels of PCBs, PCDDs and PCDFs 0.75 to 1.95 mg.kg-1 [44]. Similarly, data reported from in human milk [33]. The concentrations of six indicator Wales reach on average 1.3 mg PCB.kg-1.d-1 [48]. Much -1 -1 PCBs (Figure 3) and mono- and non-ortho-substituted higher estimation (up to 26 mg.kg b.w..d ) was reported by congeners (Figure 4) in breast milk from an unpolluted Kredl et al. in the Czech Republic [49]. The level of PCBs area (Slovakia 1), represented by the Nitra district, breast milk contamination in this study shows were on the level comparable to maximal figures meas- considerably higher levels in Slovakia than in other highly ured in the European developed countries. The PCB lev- industrialized western countries. Because of small sample els in the polluted area - the Michalovce district size available, it was not possible to assess the trend in (Slovakia 2) were considerably higher (1015 ng.g-1). The infant exposure. Thus, continued monitoring and research discrepancy between Slovakia and other countries was of these compounds in mother milk is warranted to caused, in our opinion, presumably by the total environ- specify health effects derived from this relatively high ment contamination [27]. background PCB exposure.

345 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

FIGURE 2 Profile of PCB production during 1959-1984 in Slovakia. Tonne ! 2000 ! ! ! ! ! ! ! ! 1500

1000

! ! 500 ! ! ! ! ! ! ! ! ! ! 0 ! ! !

1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984

FIGURE 3 Levels of PCBs (sum of PCB-28, 52, 101, 138, 163, 180) in human milk samples from various countries (y.1993) [33].

Slovakia 1 489 Slovakia 2 1015 Czech Rep. 1 532 Czech Rep. 2 1069 Albania (n=2) 53 Hungary (n=2) 53 Russia (n=2) 150 Finland (n=2) 161 Ukraine (n=2) 214 Croatia (n=2) 219 Denmark (n=5) 236 Netherland (n=17) 253 Belgium (n=3) 281 Spain (n=2) 457 0 200 400 600 800 1000 1200 ng.g-1 (lipid basis)

FIGURE 4 Levels of mono- and nonortho-PCBs (expressed as WHO-TEQs12) in human milk samples (y.1993) [33].

Slovakia 1 21.5 Slovakia 2 43.3 Czech Rep. 1 19.8 Czech Rep. 2 42.4 Albania (n=2) 3.9 Hungary (n=2) 3.1 WHO-TEQ (1-ortho-PCBs) Russia (n=2) 13.4 WHO-TEQ (0-ortho-PCBs) Finland (n=2) 9.2 Ukraine (n=2) 23.1 Croatia (n=2) 14.6 Denmark (n=5) 14.9 Netherland (n=17) 20.4 Belgium (n=3) 23.3 Spain (n=2) 24.2 0 10 20 30 40 50 pg.g-1 (lipid basis)

TABLE 3 PCB daily intakes (DI) for the Slovak breast-fed infants from model locations.

-1 -1 DI, mg.kg b.w. .d (n) Bratislava Michalovce Myjava Nitra Trebišov Velký Krtíš Total Boys 6.8 (6) 8.3 (6) 7.8 (1) 3.8 (5) 4.8 (5) 3.6 (9) 5.4 (32) Girls 5.4 (7) 7.2 (6) 5.5 (3) 2.1 (4) 6.4 (7) 5.4 (3) 5.6 (30) Total 6.1 (13) 7.8 (12) 6.1 (4) 3.0 (9) 5.7 (12) 4.0 (12) 5.5 (62)

346 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

ACKNOWLEDGMENTS

This study was supported in part by the U.S.-Slovak [14] M. van den Berg, L. Birnbaum, A.T.C. Bosveld, B. Brunström, Science and Technology Joint Fund in cooperation with P. Cook, M. Feeley, P. Giesy, A. Hanberg, R. Hasegawa, S.W. the Ministry of Health of the Slovak Republic and the Kennedy, T. Kubiak, J.C. Larsen, F.X.R. van Leeuwen, A.K.D. U.S. Department of Health and Human Services under Liem, C. Nolt, R.E. Peterson, L. Poellinger, S. Safe, D. Project No. 94023. The authors acknowledge the assistance of hospital staff who collaborated on the study. Schrenk, D. Tillitt, M. Tysklind, M. Younes, F. Waern, T. Zacharewski, Environ. Health Perspect. 106, 775-792 (1998).

[15] Toxicological profile for selected PCBs (Aroclor 1260, 1254,

1248, 1242, 1232, 1221 and 1016). ATSDR/TP-92/16. US De- REFERENCES partment of Health and Human Services, Atlanta (1993).

[16] C. DeRosa, P. Richter, H. Pohl, D.E. Jones, J. Toxicol. Environ. [1] É. Dewailly, P. Ayotte, C. Laliberté, J-P. Weber, S. Gingras, Health, Part B 1, 3-26 (1998). A.J. Nantel, Am. J. Public Health 86, 1241-1246 (1996). [17] J.F. Brown, R.W. Lawton, M.R. Ross, J. Feingold, Chemos- [2] S. Safe, Crit. Rev. Toxicol. 21, 51-88 (1990). phere 23, 1811-1815 (1991) [3] S. Safe, Environ. Health Perspect. 100, 259-268 (1992). [18] Y.L. Guo, C.J. Lin, W.J. Yao, J.J. Ryan, C.C. Hsu, J. Toxicol. [4] M.R. Sim, J.J. McNeil, Am. J. Epidemiol. 138, 1-11 (1992). Environ. Health 41, 83-93 (1994). [5] Polychlorinated biphenyls and terphenyls. 2nd ed. EHC 140. [19] J.L. Jacobson, S.W. Jacobson, N. Engl. J. Med. 335, 783-789 WHO, Geneva (1993). (1996). [6] P.J.J. Sauer, M. Huisman, C. Koopman-Esseboom, D.C. Morse, [20] A. Kocan, J. Petrík, J. Chovancová, B. Drobná, S. Jursa, M. A.E. Smits-van Prooije, K.J. van den Berg, L.G.M.Th. Tuinstra, Pavúk, Exposure of population to PCBs and related compounds C.G. van der Paauw, E.R. Boersma, N. Weisglas-Kuperus, in risk area of Michalovce district. Final report, Project No. J.H.C.M. Lammers, B.M. Kulig, A. Brouwer, Hum. Exp. Toxi- 105/98-2.4., Institute of Preventive and Clinical Medicine, Bra- col. 13, 900-906 (1994). tislava (1999), (in Slovak). [7] E. Zuccato, S. Calvarese, G. Mariani, S. Mangiapan, P. Grasso, [21] Bulletin of the Ministry of Health of the Slovak Republic A. Guzzi, R. Fanelli, Chemosphere 38, 2753-2765 (1999). 44/9-13, 56-141 (1996) (in Slovak). [8] IARC Monographs on the evaluation of the carcinogenic risk of [22] Assessment of health risks in infants associated with exposure to chemicals to humans 18, Suppl. 7, Polychlorinated biphenyls PCBs, PCDDs and PCDFs in breast milk. Environmental Health and polybrominated biphenyls, Lyon (1987). Series 29, WHO Regional Office for Europe, Copenhagen (1988). [9] Levels of PCBs, PCDDs and PCDFs in breast milk. Environ- [23] R. Schoula, J. Hajšlová, V. Bencko, J. Poustka, K. Holadová, mental Health Series 34. WHO Regional Office for Europe, V. Vízek, Chemosphere 33, 1485-1494 (1996). Copenhagen (1989). [24] J. Hajšlová, M. Vávrová, PCB and other organic contaminants [10] R. Fischer, K. Ballschmiter, Fresenius Z. Anal. Chem. 335, in agriculture. Výživa a potraviny 3/91. Institute of Scientific 457-463 (1989). Information in Agriculture, Prague (1991), pp 52. (in Czech).

[11] S. Safe, Mutat. Res. 220, 31-47 (1989). [25] J. Chovancová, J. Petrík, A. Kocan, I. Holoubek, Toxicol. [12] U.G. Ahlborg, G.C. Becking, L.S. Birnbaum, A. Brouwer, Environ. Chem. 44, 73-80 (1994). H.J.G.M. Derks, M. Feeley, G. Golor, A. Hanberg, J.C. Larsen, [26] A. Kocan, J. Petrík, J. Chovancová, L. Neubauerová, M. A.K.D. Liem, S.H. Safe, C. Schlatter, F. Waern, M. Younes, E. Bezacinský, Chemosphere 23, 1473-1480 (1991). Yrjänheikki, Chemosphere 28, 1049-1067 (1994). [27] A. Kocan, B. Drobná, J. Chovancová, J. Petrík, E. Szabová, [13] U.G. Ahlborg, L. Lipworth, L. Titus-Ernstoff, Ch-Ch. Hsieh, A. Exposure of population to PCBs and related compounds in risk Hanberg, J. Baron, D. Trichopoulos, H-O. Adami, Crit. Rev. area of Michalovce district. Interim report, Project No. 105/98- Toxicol. 25, 463-531 (1995). 2.4., Institute of Preventive and Clinical Medicine, Bratislava (1998), (in Slovak).

347 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

[28] J. Petrík, J. Chovancová, A. Kocan, I. Holoubek, Toxicol. [44] R. Galetin-Smith, S. Pavkov, N. Roncevic, Bull. Environ. Environ. Chem. 34, 13-18 (1991). Contam. Toxicol. 45, 811-818 (1990).

[29] A. Kocan, J. Petrík, B. Drobná, J. Chovancová, Chemosphere [45] A.A. Jensen, Research Reviews 89, 104-107 (1983). 29, 2315-2325 (1994). [46] R. Duarte-Davidson, K.C. Jones, Sci. Total Environ. 151, 131- [30] A. Kocan, J. Petrík, B. Drobná, J. Chovancová, Organohalogen 152 (1994). Compounds 21, 147-151, (1994). [47] J. Ševcík, J. Lenícek, M. Cítková, M. Sekyra, E. Rychlíková, [31] A. Kocan, B. Drobná, J. Petrík, J. Chovancová, D.G. Patterson, Jr., Cs. Hyg. 31, 378-382 (1986). L.L. Needham, Organohalogen Compounds 26, 186-192 (1995). [48] R. Duarte-Davidson, S.J. Harrad, S.C. Allen, K.C. Jones, [32] A. Kocan, J. Petrík, J. Chovancová, B. Drobná, H. Uhrinová, Chemosphere 25, 1653-1663 (1992). Chlorinated aromatic compounds in the human organism from [49] F. Kredl, J. Svobodník, K. Khun, Cs. Hyg. 35, 149-163 (1990). selected model areas of the Slovak Republic. Final report, Pro- ject No. 93-535-03-32. Institute of Preventive and Clinical

Medicine, Bratislava (1995), (in Slovak).

[33] Levels of PCBs, PCDDs and PCDFs in human milk. Second round of WHO-coordinated exposure study. Environmental

Health in Europe 3, WHO Regional Office for Europe, Copen-

hagen (1996).

[34] Report on xenobiotics control results - year 1993. Center for

Evaluation of Xenobiotics Occurrence, Food Research Institute,

Bratislava, (1994) (in Slovak).

[35] H. Lengyelová, P. Miškovic, S. Korony, PCBs in human milk

in the middle Slovakia. In: Proceedings of Conference Foreign

Substances in Food, Tatranská Štrba, Slovak Republic, 82-84

(1991), (in Slovak).

[36] T. Yakushiji, I. Watanabe, K. Kuwabara, R. Tanaka, T. Kashi-

moto, N. Kunita, I. Hara, Arch. Environ. Contam. Toxicol. 13,

341-345 (1984).

[37] V. Prachar, M. Veningerová, J. Uhnák, J. Kovacicová, Che-

mosphere 29, 13-21 (1994).

[38] K. Norén, A. Lundén, Chemosphere 23, 1895-1901 (1991).

[39] H.R. Johansen, G. Becher, A. Polder, J.U. Skaare, J. Toxicol. Received for publication: November 15, 2000 Accepted for publication: March 15, 2001 Environ. Health 42, 157-171 (1994).

[40] C. Koopman-Esseboom, M. Huisman, N. Weisglas-Kuperus, C.G. van der Paauw, L.G.M.Th. Tuinstra, E.R. Boersma, P.J.J. CORRESPONDING AUTHOR Sauer, Chemosphere 28, 1721-1732 (1994).

[41] S. Georgii, G. Bachour, I. Elmadfa, H. Brunn, Bull. Environ. Ján Petrík Contam. Toxicol. 54, 541-545 (1995). Institute of Preventive and Clinical Medicine Limbová 14 [42] É. Dewailly, A. Nantel, J-P. Weber, F. Meyer, Bull. Environ. 833 01 Bratislava -SLOVAK REPUBLIC Contam. Toxicol. 43, 641-646 (1989). Email: [email protected] [43] K. Czaja, J.K. Ludwicki, K. Góralczyk, P. Strucinski, Arch. FEB – Vol 10/ No 4/ 2001 – pages 342 - 348 Environ. Contam. Toxicol. 36, 498-503 (1999).

348 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

ORGANOCHLORINE PESTICIDE RESIDUES IN BREAST MILK FROM WOMEN LIVING IN TRIPOLI, LIBYA

Amar Elafi1, Mohamed Rahmani2 and El Hassane Abdennebi3

1Food Science Department, Faculty of Agriculture, Al-Fateh, University, Tripoli, Libya. 2Départment de Sciences Alimentaires et Nnutritionnelles, IAV Hassan II, Rabat-Instituts, Rabat, Morocco. 3Département de Pharmacie, Toxicologie et Biochimie, IAV Hassan II, Rabat-Instituts, Rabat, Morocco.

SUMMARY

The presence of organochlorine pesticides in Libyan Both humans and animals are exposed to organochlorine human milk collected in two major Tripoli maternity pesticides from many different sources, but the most im- hospitals was investigated. The majority of the samples portant route of this exposure is the consumption of con- (76%) were contaminated with one or several of these taminated food. As these compounds are highly lipid compounds. The incidence of contamination varied soluble, they accumulate in adipose tissue of humans and from 5.4% for lindane to 56% for ß-HCH. The highest animals, may pass through the placenta and excrete in mean concentration was found for ß-HCH (240 ng/g fat) breast milk resulting in exposure of fetus and nursing and the lowest was that of op’-DDD (75ng/g fat). infants (5). The estimated daily intakes of some pesticides indicate that the detected levels of residues do not constitute any The present study reports the first data on the residue risk to babies’ health. levels of some organochorine compounds in breast milk from women living in Tripoli, Libya. The aim of this work is to evaluate the importance of breast milk as an KEYWORDS: Organochlorine residues - Human milk- Libya indicator of human exposure to these compounds and discuss the significance of these contamination to the public health, especially, to babies.

INTRODUCTION MATERIALS AND METHODS

Organochlorine compounds, such as hexachloroben- A total of 108 milk samples were collected at the zene (HCB), lindane, diclorodiphenyl-trichloroethane major maternity hospitals. The milk was collected from (DDT) enjoyed a widespread use in agriculture and public women at different ages and the first week of their health. These chemicals have become so globally distrib- delivery. Milk samples (about 15 ml) were sent to uted that they almost qualify to be called naturally occur- Hassan II Institute of Agriculture and Veterinary ring. They are considered to be among the most important Medicine of Rabat (Morocco), where they were kept pollutants due to their persistence in the environment and frozen at –20 °C until analysis. their possible harmful effects. Even though the use of organochlorine pesticides has been prohibited for decades The extraction of chlorinated compounds was accom- in many countries, these chemicals create some residue plished according to the method described by Veirov and problems in many living organisms (1). Their influence Aharonson (6), but with slight modification. It involves on broad range of reproductive disorder along with possi- isolation of pesticide residues from milk samples by ble carcinogenic effect is of considerable concern. liquid-liquid partitioning using concentrated sulfuric acid and petroleum ether. The acid cleaned petroleum ether In Libya, the use of this group of pesticides has been portion was passed through a glass column packed with permitted for about 15 years; from 1977 to 1990 (2), both silica gel. The column was eluted first with hexane to in agriculture and public health. Actually, most of these obtain the first fraction containing HCB and other related chemicals are banned from application by the Libyan pesticides and then eluted with a mixture of hexane and ministry of agriculture. However, some of them diethyl ether (75/25 % v/v) to collect the second fraction (e.g., lindane) are still used under restricted conditions. containing HCH-isomers, and DDT and its derivatives.

349 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

All the glass wear used were always washed in a solution The corresponding mean residue values for the other of 10 % nitric acid, and the different reagents employed compounds were 93 ng/ g for o,p’-DDE, 94 ng/ g for for the extraction were of analytical grade. d-HCH, 117 ng/ g for lindane, 122 ng/ g for p,p’-DDD, 125 ng/ g for p,p’-DDE and 143 ng/ g for a-HCH. The organochlorine pesticides were determined by a Hewlett Packard gas chromatograph equipped with an The average and maximum daily intakes of residues electron capture detector. Aliquots of 2 ml extracts were by the nursing infants from their breast milk are given injected into a glass column (300 mm ´ 20 id) packed in Table 2. For example, these values are 577.9 and with 1.5 % OV-17 on Chromosorb of 80-100 mesh. The 1467.2 ng/ kg for lindane, 2356.4 and 6955.5 ng / kg for standardized chromatographic conditions were, injector SHCH, 459.4 and 770.6 ng/ kg for o,p’-DDE. temperature 250 °C, oven temperature 200 °C and detector temperature 250 °C. DISCUSSION The identification of chlorinated compounds was accomplished using reference solutions of hexachloro- This is the first report dealing with chemical cyclobenzene (HCB), hexachlorocyclohexane (HCH) and contamination of human milk from women living in its isomers a, b, d and lindane, aldrin, pp’-dichlorodi- Libya. In other countries, similar investigations have phenyl trichloroethane (p,p’-DDT) and its derivative already been done and concerned residues of different (p,p’-DDD, o,p’-DDD, p,p’-DDE and o,p’-DDE). A compounds such as organochlorine pesticides, polychlori- Hewlett Packard 3390 integrator was used for the quanti- nated biphenyls and heavy metals (5,9-13). Results of the fication, and the identification procedure was based on the present work revealed that most (76 %) of the analyzed external standard method. samples were contaminated with pesticide residues. The high incidence of contamination is expected since, in The reproducibility of the results under addition to their persistence and wide occurrence, these the experimental conditions was tested by consecutive compounds have been just recently banned from use. analysis of standard solutions containing different However, the detected levels were very low in most of the concentrations of lindane. The linearity of the chromato- samples. As importantly as the ban is the recent introduc- graph was checked many times during our analysis using tion of pesticides in Libyan agriculture and related activi- small concentrations of fresh lindane, and the regression ties as compared to other countries (2). lines had correlation coefficients (r) greater than 0.96. Another possibility, which may explain this moderate Extraction recoveries and minimum detection limits contamination, is the large consumption of imported were determined by direct fortification of milk with manufactured products (milk and dairy products, meat solutions of known concentrations of pesticides and they etc.) which are likely to contribute to the total intake of varied from 44 % to 115 % and from 0.3 to 3 ng/ ml, pesticides. These products are supposed to be less con- respectively, depending on the nature of the organochlo- taminated, since legislation on manufacturing and use of rine compound. The estimation of daily intakes of or- these chemicals in their original countries is generally ganochlorine pesticides from breast-milk was based on very stringent and well enforced. In fact, chlorinated the average consumption of milk by babies of 130 grams compounds were restricted in the developed countries per kg body weight per day (7) and the mean fat content between 1970 and 1980, while they are still used in many of 3.8 % according to Ribadeau (8). developing countries. However, lindane remains licensed for use in a few developed countries such as Argentina, Italy and Spain and in many of the developing countries RESULTS like India, China and South America (14). This compound has extensive applications for wood preservation and Residues analysis demonstrated that most of the public health because it is considered to be the least per- studied milk samples (76 %) were contaminated with one sistent of the HCH isomers and has lower toxicity to or several of organochlorine compounds. The incidence of mammalia (15). this contamination varied from 5.4 % for lindane to about 56 % for b-HCH (Table 1). The occurrence of the other Our analysis revealed that lindane was the least de- compounds was 45 % for p,p’-DDE, 48 % for a-HCH, tected (5.4 %) compared to its b (56 %), a (48 %) and 10 % for d-HCH, 7.4 % for o,p’-DDE and 6.5 % for each d (10 %) isomers. The relatively low persistence of lin- of o,p’-DDD and p,p’-DDD. Mean values and range of dane and its ability to accumulate compared to b and organochlorine residues levels detected in human milk are a isomers (10) account probably for the observed low given in Table 1. The highest mean concentration was level. Furthermore, b-HCH is known to be persistent found for b-HCH (240 ng/ g on fat basis) and the lowest isomer (16) and the very slowly eliminated from the envi- was that for o,p’-DDD (75 ng/ g on fat basis). ronment as compared to lindane (17).

350 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

TABLE 1 - Occurrence and residue levels (ng/ g fat) of organochlorine pesticides in human milk ______Concentration (ng/g fat)

Pesticide Positive (%) Mean Range ______a-HCH 48 143 T - 323 b-HCH 56 240 T - 686 Lindane 5.4 117 T - 297 d-HCH 10 94 T - 399 o,p’-DDE 7.4 93 T - 156 p,p’-DDE 45 125 T - 561 o,p’-DDD 6.5 75 T - 148 p,p’-DDD 6.5 122 T – 426

______T= Traces

TABLE 2 - Estimated daily intakes (DI)* of some organochlorine pesticides and their comparison to accepted daily intakes (ADI)** ______DI ADI % ADI ______Pesticide Mean Max Mean Max a-HCH 706.4 1595.6 b-HCH 1185. 8 3388.8 d-HCH 464.4 1971.0 SHCH 2356.4 6955.5 8,000 29.4 86.9 Lindane 577.9 1467.2 10,000 5.8 14.7 o,p’-DDE 459. 4 770.6 p,p’-DDE 617.5 2771.3 o,p’-DDD 370.5 731.1 p,p’-DDD 602.7 2104.4 SDDTT 5000

______* ng/ kg body weight ** Recommended by FAO/WHO (20,21)

Our results also indicate that about 70 % of the ana- Comparing the level of insecticide residues in breast- lyzed samples contained HCH isomers (a, b, d and lin- milk recorded in the present study to those obtained in dane) while DDT derivatives (o,p’-DDE, p,p’-DDE and other countries, it appears that mean concentrations of p,p’-DDD) were detected in about 50 % of the samples. sum of HCH and sum of DDT are similar to those re- Similarly, the contamination levels for HCH and its iso- ported in Moroccan (7) and Swedish (13) human milk. mers (max 0.686 ng/ g) were higher than those of DDT However, the man level of these compounds are many compounds (max 0.561 ng/g). These results could times lower than the corresponding values detected in be interpreted by examining the available record on Indian (12) and Egyptian (19) human milk, countries with the previous uses of phytochemicals in Libya. extensive pesticide applications. The Ministry of Agriculture never registered DDT and The calculated daily exposures of Libyan infants for its derivatives for use, while HCH and some of other lindane, HCH isomers and DDT complex are relatively related isomers have been licensed for agricultural appli- similar to those reported by Benazzouz (7) for Moroccan cations since 1977. infants. According to our results, the child intakes of residues via human milk are below the accepted daily In the group of DDT, the highest concentration was intake (ADI) set by the WHO/FAO expert group (20,21). found for p,p’-DDE (0.561 ng/g). This observation is in The mean daily residue intake from Libyan human milk agreement with that reported by Alawi et al., (18) and contributes to 29.4 % of ADI for SHCH and to only Krunthacker et al. (11) in Jordanian and Yugoslavian hu- 5.8 % of ADI for lindane (Table 2). Consequently, these man milk, respectively, and may not be surprising since p,p’- values represent a minimum hazard to health even when DDE is the principal metabolite of DDT (10). maximum residue levels were taken into consideration.

351 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

REFERENCES

1. Kamarianos A, Iosifidou EG, Batzios C, Psomas IE, Kilikidis 16. Watson DH. Dioxins and other environmental chemicals. S (1999) Residues of organochlorine pesticides and PCBs in Watson DH (ed) Safety of chemicals in food chemical con- human adipose tissues in Greece. Fresenius Envir. Bull. 6: taminants. Ellis Horwood, Great Britain, p. 46. 383-389. 17. Herrera A, Arino AA, Conchello MP, Lazaro R, Bayarri S, 2. Taher M, Abozada E, Tabet T, Nagi MA (1978) Registered Perez C (1994) Organochlorine pesticide residues in Spanish and permitted pesticides under Libyan conditions. Principal meat products and meat of different species. J. Food Prot. 5: of plant protection. Edited by the Department of Agricultural 441-444 Information (Ed.), Ministry of Agriculture, Tripoli, Libya . 18. Alawi A, Ammari N, Al-Shuraiki Y (1992) Organochlorine 3. Shibamoto T, Bjeldanes LF (1993) Pesticide residues in food. pesticide contamination in human milk samples from women In: Shibamoto T, Bjeldanes LF (eds) Introduction to Food living in Amman Jordan. Arch. Environ. Contam. Toxicol. Toxicology, Academic Press, USA, p. 171. 23: 235-239.

4. Black WD (1994) Pesticide residues in food of animal origin. 19. Dogheim SM, El-Zarka M, Gad-Alla SA, EL-Said SA, EL- In: Crawford LM, Franco DA (eds) Animal drugs and hu- Saied S, Emel SY, Mohseen AM, Fahmy SM (1996) Moni- man health. Technomic Publishing, USA, p. 171. toring of pesticide residues in human milk, soil, water and food samples collected from Kafr El-Zayat Governorate. J. 5. Beacher OG, Sharre J, Utre R, Anuschka S, Brita RJ, S, Assoc. Off. Anal. Chem. 79: 1: 111-116. Rosland OJ, Hanson HK, Ptastieka J (1995) PCDF and PCBs in human milk from different parts of Norway and 20. Anonymous (1983) Codex Alimentarius. Codex maximum Lithunnia. J. Toxicol. Environ. Health. 46: 133-148. limits for pesticide residues. Vol. III. FAO/WHO, Rome.

6. Veirov D, Aharonson N (1980) Economic method for analy- 21. Anonymous (1987) Résidus de pesticides organochlorés dans sis of fluid milk for organochlorine residues at the 10 ppb les produits alimentaires. Rapport de la réunion conjointe level. J. Assoc. Off. Anal. Chem. 63: 3: 532-535. FAO/ OMS, 84.

7. Benazzouz I (1992) Etude comparative des niveaux de contamination des lait de femme, de vache et maternises par

les pesticides organochlorés. Thèse de 3 ème cycle. Biochimie. Univ. Mohammed V. Fac. Sci. Rabat, Maroc.

8. Ribadeau D (1983) La lait maternel. La recherche 14, 8-17.

9. Kanja LW Share NI, Maitai CK,, Lok P (1986) Organochlo- rine pesticides in human milk from different area of Kenya (1983-1985). J. Toxicol. Environ. Health. 19: 449-464.

10. Kahunye M, Frosline A, Maital CK (1988) Organochlorine pesticide residues in chicken eggs : A survey. J. Toxicol. Environ. Health 24: 543-550.

11. Krunthacker B, Kralj MTB, Reiner E (1986). Level of HCH, HCB, DDT, pp-DDT and PCBs in human milk from a conti- nental Town in Croatia, Yugoslavia. Inter. Arch. Occup. and Environ. Health, 69-74. Received for publication: April 26, 2000 Accepted for publication: March 15, 2001 12. Tanabe S, Gondaira F, Subramanan A, Ramash A, Mohan D, Kurmaran P, Venugopkelan K, Tabsukawa K (1990) Specific pattern of pesticide residues in human breast-milk from south India. J. Agric. Food Chem. 39: 899-903. CORRESPONDING AUTHOR

13. Vaz R (1993) Organochlorine contaminants in Swedish El Hassane ABDENNEBI foods of animal origin and human milk (1973-1992). Occur- rence, analysis, analytical quality assurance, levels, intake Département de Pharmacie and implications. Ph.D., Thesis. Uppsada. Toxicologie et Biochimie IAV Hassan II 14. Turnbull A (1976) Chlorinated pesticides. Chlorinated or- BP: 6202 ganic Micropollutants. Hester and Harrison (eds). The Royal Rabat-Instituts Society of Chemistry, Turpin Distribution Services Limited, Litchworth, UK., pp 113-135. Rabat-MOROCCO

15. Crosby NT (1991) Other contaminants. In Crosby NT (ed) E-mail: [email protected] Determination of pesticides and veterinary residues in food. Ellis Horwood Limited, Great Britain , p 117. FEB Vol 10/ No 4/ 2001 – pages 349 - 352

352 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

DETERMINATION OF THE PHOTOCHEMICAL LIFE TIME OF PESTICIDES IN AQUEOUS MEDIUM WITH SOLAR LIGHT

S. Guittonneau 1, I. Konstantinou 2, C. Emmelin 3, T. Albanis 2 and P. Meallier 3

1 LCME –ESIGEC – Université de Savoie – Campus Universitaire – Le Bourget du Lac cedex 2 Department of Chemistry, University of Ioannina, Ioannina, Greece 3 Laboratoire d’application de la chimie à l’Environnement - Université Claude Bernard, Villeurbanne Cedex.

Presented at the 10th International Symposium of MESAEP in Alicante, Spain, 2-6 Oct. 1999

SUMMARY

The aim of this paper was to determine the quantum accurate to measure the quantum yield with polychromatic efficiency of a serie of pesticides (s-triazine, anilide, thio- light (called also quantum efficiency) to permit an adequate carbamate, organophosphorus) and to evaluate their life estimation of the photochemical stability of a chemical in time in aqueous medium using sun light energy. The aqueous solution. This parameter is defined as the number of experiments for the determination of the quantum effi- molecules of pesticide photodegradated versus the summa- ciency were carried out in a solar light simulator equipped tion of the number of photons absorbed by the substance in with a xenon lamp. The life time was calculated according the studied range of wavelength. to the European guide line recommendation [1] with the sun energy from western Europe. The validity of the re- With polychromatic light, the experimental procedure sults obtained for the quantum efficiencies and the life is relatively simple, but the calculation is complicated. times are discussed in this study. A software was developed at the laboratory to calculate the quantum efficiency from the kinetic study of the irradiated compound and from the total amount of light energy ab- KEYWORDS: Quantum efficiency, life time, aqueous medium, sorbed during the irradiation [2]. This program of calcula- pesticides, solar irradiation, polychromatic light. tion was applied to a serie of molecules representing different functional groups of pesticides (anilide, triazine, organophosphorus and thiocarbamate). INTRODUCTION

The pollution of water by pesticides is of increasing concern and requires control studies in order to improve MATERIALS AND METHODS understanding of their evolution in the environment. To estimate the life time of these molecules it is necessary to Material take into account the global pathways of their degradation and interactions of the environment on their evolution Irradiations were performed in a « sun test » appara- processes. Among the parameters to be taken into tus (Heraeus) equipped with a xenon lamp which simu- account, the quantum yield is the one which will define lated solar UV radiation at wavelength longer than 270 the photochemical stability of the molecule. Its determina- nm. Incident photonic flux was determined by actinomet- tion is usually carried out in monochromatic light at a ric method using uranyl oxalate [3]. The total incident wavelength from the first absorption band of the molecule. light between 270 and 500 nm was 3.06 10-7 Einstein s-1. The photochemical reactor is a cylindrical vessel capped The majority of the pesticides have their maximum of with a quartz window and was placed under the xenon absorption in the UV region (l < 300 nm), but for most of lamp. The temperature was regulated at 20°C with water them, their absorption band continues above 290 nm, circulating in a cooling jacket. 50 ml of the pesticide which is relevant for a direct photodegradation of these solution were irradiated at an initial concentration of 10-5 compounds with solar light. To be close to the environ- to 10-4 mol l-1 depending on their solubility in the aqueous mental conditions and to take into account the variation of phase (optical path : 2.55 cm). All the solutions were the quantum yield with the excitation wavelength, it is more prepared in ultrapur water (Millipore, 18 MW).

353 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

Analytical methods

The pesticides were analysed by gas chromatography 1 1 1 (ECD detector) after extraction with n-hexane (5 ml t = = = l k h Pa 2 [3] sample were extracted with 2 x 2.5 ml of n-hexane). The h 2.3103 P Dl e analysis of the pesticides was carried out on a DB5 å o ,l l l1 column at 210°C for the oven temperature and 300°C for the injector and detector temperature. UV-visible Po,l: Incident sun light intensity at the wavelength l absorption spectra were performed on a Kontron spec- -2 -1 -1 trophotometer (Uvikon 930), gas chromatography, on a (Einstein cm s nm ) Dl : wavelength range (nm) Perkin Elmer apparatus (GC 2000). Quantum efficien- -1 -1 cies and life times of the pesticides were calculated with el : molar absorption coefficient (l mol cm ) 3 Photon software (Kontron). 2.3 10 : factor for conversion unit

The life time was calculated with the solar light energy Determination of the quantum efficiency published by Frank and Klöpfer [4] corresponding to those from western Europe (latitude 40 to 50° North). A computer program (Photon, Kontron), developed at the laboratory, was used to determine the quantum efficiency. This software takes into account the absorbance (Apest,l) and the photodegradation rate of the pesticide (Dnpest/Dt), the emission spectra of the light source (xenon lamp) and RESULTS AND DISCUSSION the actinometric data to determine the absolute incident photonic flux of the lamp at the different wavelength of UV spectra excitation of the pesticide (I ) : o,l Spectral characteristics of the 10 pesticides inves-

tigated in this study are reported in Table 1. For most

of these compounds, their maximum of absorption is in Dn pest the UV region (250 < l < 285nm) but their absorp- Dt max h = tion continues in the close UV confirming that these l2 [1] - Al compounds are relevant to a direct photodegradation by å Io ,l (1-10 ) l1 sun light. h : quantum efficiency

Apest,l : absorbance at the wavelenght l Quantum efficiencies

The quantum efficiency was calculated from equation [1] over the initial photolysis period corresponding Determination of the life time to less than 10% loss of the pesticide. The quantum In this study, the life time was calculated according to the efficiency and the corresponding life time of the 10 guide line published by the European community [1]. The pesticides studied are reported in Table 1. The compari- rate of the direct photodegradation of pesticide in aqueous son of the quantum efficiencies shows some differences medium is of first order and the life time is given by : for each pesticide group. For s-triazine an efficiency of 10-2 magnitude was found, 10-3 for anilines, molinate and two organophosphorus and 10-4 for methyl and ethyl 1 parathion. These values are in good agreement with t = k [2] those obtained with monochromatic light (l = 254 or t: life time 313 nm) published in the literature (Table 1) indicating a k : pseudo-first order rate constant relatively small dependency of the photoreaction with the wavelength. In addition, the systematic errors, which are related to the method of determination of the quan- The pseudo-rate constant depends on two parameters: tum efficiency, do not permit to observe differences from the photophysic properties of the molecule. The · The total number of photons absorbed Pa, which can principal advantage of the use of quantum efficiency be evaluated from Beer Lambert law with the incident instead of the quantum yield is the shorter duration of sun light intensity in the spectral range which coin- the experiment (irradiation) and the possibility to deter- cides with the light absorption of the pesticide. mine the quantum efficiency of substances with low · The quantum efficiency h determined with a poly- photoreactivity (weak absorption in close UV and/or chromatic irradiation. small quatum yield) [10].

354 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

TABLE 1 Quantum efficiencies and life times of the pesticides lmax Wavelength Quan- Quan- Life Pesticide Formular e max limit of absorp- tum tum time of the first tion effi- yield F t (h) absorption (e < 3 M-1 cm-1) ciency (ref) band (nm) h

Cl Propanil 285 nm 311 4 10-3 8 10-3 99 Cl NHC C2H5 800 M-1 cm-1 O [5]

Propachlor C H (CH3)2 -3 N C CH2Cl No maximum 330 4 10 - 105

Cl Atrazine 262 nm N N 320 1 10-2 3.7 10-2 50 2800 M-1 cm-1 (C H3)2CHNH N NHC 2H5 [6]

Cl Propazine 263 nm N N 295 2 10-2 4.8 10-2 27 2900 M-1 cm-1 (C H3)2CHNH N NH(C H3)2 [7]

SC H3 Prometryn 265 nm - N N 320 3 10-2 36 4800 M-1 cm-1 (C H3)2CHNH N NH(C H 3)2

Feni- CH3 trothion S 271 nm -3 -2 -1 -1 460 1 10 3.1 10 20 (CH3O)2P O NO 2 4600 M cm [8]

Méthyl S parathion 276 nm -4 -4 (C H 3O )2P O NO 2 397 6 10 1.7 10 128 7500 M-1 cm-1 [8]

Ethyl S parathion 276 nm -4 -4 (C 2H 5O )2P O NO 2 390 8 10 6.7 10 81 7600 M-1 cm-1 [9]

S Fenthion (CH O ) P O SCH 250 nm 3 2 3 312 7 10-3 3.5 10-2 31 1000 M-1 cm-1 CH 3 [8]

O -3 Molinate N C SC2H5 No maximum 290 2 10 - 143

355 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

Life time REFERENCES The life time of pesticide undergoing direct photolysis in water depends not only on the quantum 1. Test guide line – Phototransformation of chemicals in water – Umweltbundesamt – Berlin – September, 1990 efficiency and the absorption spectra of the compound but also on the light intensity of the spectral distribution of day light. For calculating the expected life time, 2. Moutié E., Guittonneau S., Prost S. and Meallier P. - Cin- the solar light energy published by Frank and Klöpfer quième rencontre du GRAPE, Grenoble 10 Juin, 1997. [4] were used. These data are mid-day averaged sun light intensities from western Europe. 3. Braun A.M., Maurette M.T. et Oliveros E. – Technologie photochimique – Presses polytechniques romandes, 1ère Edi- The life time obtained for the serie of pesticides tion, 1986. investigated in this study (table 1) varied between 20 h to about 150 h (continuous irradiation). These life 4. Frank R. And Klöpfer W. – Chemosphere, 10, 985, 1988. times are low because they are calculated at the upper layer of the aqueous phase and does not take into ac- 5. Faure V. et Boule P.– XXVIème congrès du GFP, Nancy count the penetration of the light into the water. Never- Vandoeuvre , 22-23 mai 1996. theless, these calculations can be used for a rough estimation of the persistence of a pesticide in water and 6. Khan S.U. and Gamble D.S. –– J. Agric. Food. Chem., 31, for a comparison of the life time of pesticides on a 1099, 1983. relative scale.

The real life time will necessitate to take into ac- 7. Wan H.B., Wong M.K., Mok C.Y. – J. Agric. Food chem., 42, 2625, 1994. count (i) the competitive absorption (absorption of other molecules or particles in the medium, screen effect), (ii) the reflection of the incident light, (iii) the 8. Durand G., Abad J.L., Sanchez-Baeza F., Messeguer A. and light penetration (depth of the water resource) and (iv) Barcelo D.– J. of Agric. Food Chem ; 42, 814, 1994. solar light intensity for a given geographical area. In any case, the real photochemical life time of pesticide 9. Mansour M., Feicht E. and Meallier P. – Toxic. and Environ. in natural water should be more important than those Chem., 41, 1264, 1993. found in this study. 10. ECETOC – Technical report – the transformation of chemicals in water : results of a ring test, Brussels, 13 june, 1984

CONCLUSION

In conclusion, the quantum yield or the quantum efficiency can both be used for the calculation of the life time of pesticides in the aqueous medium with a solar irradiation. The life times calculated in this study are only an estimation and should be used with care. To be able to determine the real life time, environmental parameters such as light intensity of the con- Received for publication: November 23, 1999 cerned region, depth of the water and absorption of the Accepted for publication: March 15, 2001 aqueous medium should be taken into account.

CORRESPONDING AUTHOR

S. Guittonneau LCME –ESIGEC Université de Savoie Campus Universitaire 73376 Le Bourget du Lac cedex – France

E-mail: [email protected]

FEB – Vol 10/ No 4/ 2001 – pages 353 - 356

356 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

PHOTOSTIMULATED DECOMPOSITION OF SOME SELECTED PESTICIDES IN AQUEOUS PHASE IN THE PRESENCE OF AEROSOL CONSTITUENTS

P.N. Moza*, K. Hustert, E.A Feicht and A. Kettrup

GSF- National Research Center for Environment and Health, Institute of Ecological Chemistry, Neuherberg, Germany

SUMMARY

The photodegradation of some selected pesticides Recently, comprehensive organic source profiles for known to be present in rain and precipitation in aqueous smoke from burning pine, oak, and synthetic logs in resi- solution has been examined in the presence of furfural dential fireplaces have been reported [19]. In order to and furfuryl alcohol (aerosol constituents). Irradiation understand the photochemistry of a pesticide in the atat 290 nm of isoproturon, mecoprop and aldicarb in the mosphere, knowledge of the effect of other organic car- presence of furfural led to increased degradation. bon present in air on its rate of photolysis is necessary. In Pirimicarb under the same experimental conditions this work as a model experiment we report the influence showed in regard to photodecomposition no significant of furfural and furfurylalcohol found in smoke on the difference with or without the presence of furfural and photodecomposition of isoproturon, mecoprop, aldicarb, furfuryl alcohol. Here direct photolysis plays a dominant atrazine, and pirimicarb in water. These pesticides are role. A number of photoproducts of pirimicarb and reported to be present in precipitation [20]. mecoprop have been identified.

MATERIALS AND METHODS KEYWORDS: Photodegradation, pesticides, furfural, furfurylalcohol, photoproducts Isoproturon [3-(4-isopropylphenyl)-1,1-dimethyl- urea], mecoprop [2-(4-chloro-2-methylphenoxy)propionic acid], aldicarb [2-methyl-2-(methylthio)propionaldehyde O-methylcarba-moyloxime], pirimicarb (2-dimethyl- INTRODUCTION amino-5,6-dimethylpyrimidine-4-yl dimethylcarbamate) and atrazine (2-chloro-4-ethylamino-6-isopropylamino- Pesticides are used world-wide to protect crops 1,3,5-triazine) were purchased from Riedel-de Haen against pests, weeds and fungi. It is reported that only (Pestanal). Furfuryl alcohol and furfural were obtained 1 - 3 % of an agricultural chemical reaches the site of from Aldrich. Aqueous solutions of the pesticides action [1]. Pesticides may be transformed by biotic or (2 mg/L) were prepared and 200 ml of each solution were abiotic processes and in addition are transported from the separately irradiated in a quartz vessel using a xenon lamp site of application by several processes and one among at 550 W/m2 (Suntest, Heraeus). The experiments were the many (run-off, leaching, transport on soil particles and carried out at 20o C by circulating cold air through the wind erosion) is volatilisation [2]. suntest apparatus with the help of a cooling machine (Yeti Moreover, loss of pesticides to the atmosphere may Plus). To investigate the effect of furfuryl alcohol on the result from air drift. A number of pesticides has been photodecomposition of the pesticides 10 ml of furfuryl found in precipitation throughout the world [3-18]. In alcohol was added to an aqueous solution (2 mg/L) of the Germany, several works have reported on the occurrence respective pesticide and 200ml of this solution was irradi- of pesticides in precipitation [3, 5, 6, 10, 13, 16-18]. ated in the suntest apparatus. Similar experiments were Pesticides released to the atmosphere may undergo carried out with the pesticides in the presence of furfural. photodecomposition which may be influenced by other Samples of the individual pesticides in aqueous solution chemicals present in the atmosphere. Combustion of with and without furfuryl alcohol and furfural were held wood in residential fire places emits primarily fine in dark as controls. At regular intervals of time 0.5 ml of organic carbon particles into the atmosphere. the irradiated solutions were taken for analysis.

357 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

Quantitative analysis of the pesticides was carried out Pirimicarb upon irradiation under the same with a Hewlett-Packard HPLC instrument (series 1050) experimental conditions showed in regard to photodecom- equipped with a Machery-Nagel reversed-phase column position no significant difference with and without the (Nucleocil 5, C18, 25 x 0.4 cm i.d.). The chromatographic presence of furfurylalcohol and furfural (85-90%, Fig. 4). conditions are shown in Table 1. Here, direct photolysis plays a significant role in the degradation of the compound because pirimicarb absorbs UV-light at wavelengths > 290 nm. The course of the The substrates and their respective photodegradation -2 -1 products from the irradiated solutions were extracted reaction followed first order kinetics (k=1.1*10 min ), using solid phase extraction (Chromabond HR-P, M&N half-life in the presence of furfural and furfuryl alcohol columns). A Hewlett-Packard Model 5989A GC-MS was t½=0.95 h and 1.45 h respectively. The course of instrument at an ionisation potential of 70 eV was used to reaction for other pesticides was complex and calculation obtain the mass spectra. The GC conditions were as fol- of their half-lives under the experimental conditions was lows: a SGE capillary column (25m x 0.22 mm i.d) coated not possible. with a 0.1mm film of polysiloxane-carborane, injector temperature 250oC, carrier gas helium (0.4 ml/min), tem- Furfural accelerates the photodegradation of isoprotu- perature programme 85o-230oC, 8oC/min. The samples ron, mecoprop and aldicarb because it is a triplet sensi- were injected in splitless mode with 1 minute hold. tiser and the energy absorbed is transferred to chemicals under investigation leading to their degradation. Furfuryl alcohol, on the other hand, reduces the degradation rate of

isoproturon and aldicarb (Fig.1, 2). Furfuryl alcohol is RESULTS AND DISCUSSION known to act as a radical scavenger , the oxygen species if produced during the photolysis are being quenched thus Solutions of isoproturon, mecoprop, aldicarb and leading to the reduction of degradation rates. Further,it is atrazine (2 ppm, each) in deionized water were irradiated observed that furfuryl alcohol does not get oxidised to with simulated sun light (l ³ 290 nm, xenon lamp, sunt- furfural which in turn should have sensitised the reaction. est). The results of three such runs after 24 h of irradiation showed very low consumption (2-10%) of the compounds, whereas degradation of isoproturon, mecoprop The photochemical degradation of pirimicarb at and aldicarb when irradiated in the presence of furfural wavelengths l>290 nm is rapid and the photolysed solu- (10 ppm) under the same experimental conditions was 28, tion upon GC-MS analysis was found to be a mixture of 18 and 24% respectively (Fig. 1-3). Irradiation of these many products. Two main products m/e 167 and 181 compounds in the presence of furfuryl alcohol showed no (2 and 3; fig.5) have been identified. Here,hydrolysis of relevant increase in the degradation. Furfural is a triplet the carbamate moiety and oxidation of -NCH3 to -NCHO sensitiser and transfers the absorbed UV energy to the takes place. As expected these products were also identi- chemical under investigation which in turn undergoes fied when the substance was photolysed in the presence of degradation/transformation. On the other hand, irradiation furfural and furfuryl alcohol. In addition, two minor prod- of atrazine in the presence of furfural and furfurylalcohol ucts m/e 224 and 252 (4 and 5; fig.5) have been identi- leads to lower degradation (4 and 2% respectively) com- fied. These compounds are reported as photoproducts of pared to 7% in their absence (Fig.7). primicarb also [21].

TABLE 1 - HPLC analysis of the pesticides

Substance Mobile-phase Flow-rate UV-Detection Retention-time CH3CN/H2O ml/min nm min

Pirimicarb 60:40 0.4 245 8.9 Isoproturon 60:40 0.5 240 10.0 Mecoprop 60:40 0.4 245 3.6 Aldicarb 60:40 0.4 228 6.2 Atrazine 60:40 0.4 230 8.0

358 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

FIGURE 1 - Photodegradation of isoproturon (l> 290 nm)

100

95 90

85 Isoproturon 80 + Furfural 75 + Furfurylalcohol % remaining 70 65 60 0 5 10 15 20 25 time (h)

FIGURE 2 - Photodegradation of aldicarb (l> 290 nm)

100 95 90 85 Aldicarb 80 Furfural 75 Furfurylalcohol % remaining 70 65 60 0 5 10 15 20 25 time (h)

FIGURE 3 - Photodegradation of mecoprop (l> 290 nm) 110

100

Mecoprop 70 Furfural

60 Furfurylalcohol 0 1 2 3 4 24 time (h)

359 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

FIGURE 4 - Photodegradation of pirimicarb (l> 290 nm)

100 90 80

70 60 Pirimicarb + Furfural 50 + Furfurylalcohol 40 % remaining 30

20 10 0 0 1 2 3 4 time (h)

FIGURE 7 - Photodegradation of atrazine (l> 290 nm)

100

90 Atrazine + Furfural % remaining + Furfurylalcohol 85

80 0 5 10 15 20 25 time (h)

An aqueous solution of isoproturon (2 ppm) when ir- presence of furfural. In addition two more compounds radiated with UV-light (l>290 nm) yielded were identified. One showed a molecular ion m/e 196 and 4-isopropylaniline and 4-isopropylphenylisocyanate as the the presence of one chlorine atom. On the basis of the + + main photoproducts. These compounds are also reported fragments m/z 168 (M - CO), 153 (M - CO-CH 3) + among other degradation products in the literature [22]. 139 (M - CO - CHCH3 + 1) a possible structure (5, Fig.6) None of the products either determined by us or reported was assigned to this compound. in the literature could be identified when this compound This product could be formed by the intramolecular was irradiated in the suntest apparatus with or without the condensation of the parent molecule. The other compound presence of furfural and furfuryl alcohol. This result is showed a molecular ion m/e 184 and the presence of one attributed to the lower degradation rate of the chemical + chlorine atom, fragments at 142 (M - COCH3 + 1), under the experimental conditions. + 107 (M - COCH3- Cl + 1). On the basis of the fragments The irradiated solution of mecoprop on GC-MS we propose structure 4 (Fig.6) for this compound. analysis was found to be a mixture of many compounds. Photodegradation of both aldicarb and atrazine Two compounds, 2-methyl-4-chlorophenol and 2-chloro- (Fig.7) was very moderate under our experimental condi- 5-hydroxybenzaldehyde were identified as the main pho- tions and the amount of the products formed was insuffi- toproducts. Here CO-bond cleavage of the molecule and cient for structural elucidation. abstraction of the hydrogen atom from the solution takes place which leads to the formation of 2-methyl-4- In conclusion furfural accelerates the photodegrada- chlorophenol (2; fig.6). The second step is the oxidation tion of the aqueous solution of the investigated pesticides of the methyl group forming 2-chloro-5-hydroxy- and furfuryl alcohol has either no effect or retards the benzaldehyde (3; fig.6). These compounds were also degradation rate. It is presumed that such reactions can identified following photolysis of mecoprop in the also take place in the atmosphere.

360 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

FIGURE 5 - Photoproducts of pirimicarb

CH3 H3C OH

N N

N H3C CH3 2 CH O CH3 3 CH H C O C N 3 H3C OH 3 CH3 N N N N O N N C H C H C CH3 3 H 3 3 1

O CH3 CH3 H3C O C N O CH3 CH3 CH H C O C N 3 N N 3 CH3 4 N N N H C H 5 3 N O

H3C C H

FIGURE 6 - Photoproducts of mecoprop

Cl Cl CO

CH CH3 H C OH O 3 H3C 2 5 Cl COOH CH O H3C Cl CH3 Cl 1

HC OH H3C O O 3 OC 4 CH3

361 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

REFERENCES

1. J.R.Plimmer, Pesticide loss to the atmosphere. Amer. J. In- 16. J.Siebers, D.Gottschild, H.G.Nolting, 1994. Pesticides in pre- dustr. Med. 18, 461-466. cipitation in northern Germany,Chemosphere 28, 1559-1570.

2. W.F.Spencer and M.M.Cliath, 1990. Movement of pesticides 17. D.Quaghebeur, Pesticides in Flanders, 1995. Water 14, 121- from soil to the atmosphere, In: Long range transport of pes- 126. ticides. (Edited by D.A.Kurtz) Chap. 1. Lewis Publishers, 18. A.Geissler, H.F.Schöler, 1993. Atmospheric deposition of Chelsea, Michigan. pesticides in the Rhine/Sieg area (Germany). Vom Wasser 3. J.Scharf, R.Wiesiollek and K.Bächmann, 1992. Pesticides in 80, 357-370. the atmosphere. Fresenius J. Anal. Chem. 342, 813-816. 19. W.F.Bogge, L.M.Hildemann, M.A.Mazurak and G.R.Cass, 4. H.R.Buser, 1990 Atrazine and other s-triazines in lakes and B.R.T.Simoneit,. 1998. Sources of Fine Organic Aerosol. 9. in rain in Switzerland. Environ. Sci. Technol. 24, 1049-1058. Pine, oak and synthetic log combustion in residential fireplaces. Environ. Sci. Technol. 32, 13-2. 5. J.Siebers, D.Gottschild and H.G.Nolting, 1991. Investigation on pesticides and polyaromatic hydrobarbons in precipita- 20. R.Hüskes and K.Levsen, 1997. Pesticides in rain. Chemos- tions of Southeast Lower Saxony. Nachrichtenbl. Deut. phere 35 , 3013-3024. Pflanzenschutzd. 43, 191-200. 21. E. Romero, P. Schmitt, M. Mansour, 1994. Photolysis of 6. R.Herterich, 1991. Atrazine. Z. Umweltchem. Ökotox. 3, pirimicarb in water under natural and simulated sunlight con- 196-200. ditions. Pestic. Sci. 41, 21-26.

7. C.J.Schomburg, D.E.Glotfelty and J.N.Seiber, 1991. Pesti- 22. P.Dureja, S.Walia and K.K.Sharma, 1991. Photolysis of iso- cide occurrence and distribution in fog collected near Mon- proturon in aqueous solution. Toxicological and Environ. terey, California. Environ. Sci. Technol. 25, 155-160. Chem. 34, 65-71.

8. J.M.Zabik and J.N.Seiber, 1993. Atmospheric transport of organophosphate pesticides from California’s Central Valley

to Sierra Nevada mountains, J. Environ. Qual. 22, 80-90. 9. B.K.Nations and G.R.Hallberg, 1992. Pesticides in Iowa pre-

cipitation. J. Environ. Qual. 21, 486-492.

10. H.Schössner and A.Koch, 1991. Pesticides in rain, Forum Städte-Hygiene 42, 109-112.

11. J.C.Hall, T.D.van Deynze, J.Struger and C.H.Chan, 1993.

Enzyme immunoassay based survey of precipitation and surface water for the presence of atrazine, metolachlor and 2,4D. Environ. Sci. Health B 28, 577-598. Received for publication: January 19, 2001 Accepted for publication: March 15, 2001 12. M.Trevisan, C.Montepiani, L.Ragozza,. C.Bartoletti, E.Ioannilli and A.A. M. Del Re, 1993. Pesticides in rainfall

and air in Italy. Environ. Pollut. 80, 31-39.

13. Ch.Oberwalder, H.Gieße, L.Irion, J.Kirchhoff and K.Hurle, CORRESPONDING AUTHOR 1991. Pesticides in rainwater. Nachrichtenbl. Deut. Pflanzen- P.N. Moza schutzd. 43, 185-191. GSF- National Research Center for Environment and 14. K. Haraguchi, E.Kitamura, T.Yamashita, A.Kido, 1995. Si- Health Institute of Ecological Chemistry multaneous determination of trace pesticides in urban P.O.B.1129 precipitation. Atmos. Environ 29, 247-253. 85758 Neuherberg – GERMANY 15. O.Lode, O.M.Eklo, B.Holen, A.Svenson, A.M.Johnson, FEB Vol.10/ No.4/2001 – pages 357 - 362 1995. Pesticides in precipitation in Norway. Sci. Total Envi- ron. 161, 421-431.

362 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

BIOSORPTION OF COPPER (II) BY POLLENS OF TYPHA LATIFOLIA L.

Ü. Danis*, Y. Kaya** and Ö.F. Algur**

*Department of Environmental Engineering, Engineering Faculty, University of Atatürk, Erzurum, Turkey. **Department of Biology, Science and Art Faculty, University of Atatürk, Erzurum, Turkey.

SUMMARY

The removal of Cu (II) ions from aqueous solution by Typha latifolia L. is a perennial cosmopolitan flowering adsorption on nonliving dried pollens of Typha latifolia L. plant (15). Its male inflorescens have cylindrical heads was studied in batch system. Concentrations ranging from (spike-like) and, therefore, the collection of pollens of this 10 to 150 mg l-1 Cu (II) were tested and metal removal plant is very easy, and it is possible to collect abundantly. achieved nearly 96 %.The biosorption of Cu (II) ions by In this research we aimed to investigate the suitability of pollen was strongly affected by pH. Within the pH range pollen of T. Latifolia L. as a biosorbent material for re- of 4 to 6, the saturated uptake capacity for Cu (II) sorption movel of copper (II) from wastewater streams. was 6.23 mg g-1. In general, higher adsorptive uptake was observed at 20°C. The adsorption of Cu (II) on pollen appeared to follow the Langmuir adsorption isotherms. Results indicated that nonliving pollen have a high ad- MATERIALS AND METHODS sorptive capacity for copper. Materials The chemicals used in this study were all analytical grade KEYWORDS: Biosorption , copper removal, copper biosorption, and purchased from SIGMA. pollen, Typha latifolia

The biosorbent used in this study was pollens of

Typha latifolia L., a very common flowering plant found INTRODUCTION all over the world. It was collected from its natural habitats in Pasinler, Erzurum, Turkey and identified as The increase in industrial activities has intensified T.latifolia according to the identification keys of Davis environmental pollution and the deterioration of some (16). The pollens were sundried and stored in sterile dark ecosystems, with the accumulation of pollutants as heavy colored glasses. metals, synthetic compounds, waste nuclear liquids, etc.

Growing attention is being given to the potential health hazard presented by heavy metals to the environment. Methods Mining and metallurgical waste waters are considered to Copper (II) solution were prepared by diluting 1.0 g l-1 be the major sources of heavy metal contamination and of copper (II) stock solution. Stock solution was prepared the need for economical and effective methods for re- by dissolving weighed quantity of CuSO4.5H2O in distilled moval of metals has resulted in the development of new deionized water. separation technologies (1). Batch studies were carried out in Erlenmayer flask. Biosorption using the biomass of microorganisms is For this purpose, 0.75 g of suspended homogenized an effective and economical technology for the removal pollen solution was added to Erlenmayer flasks contain- and recovery of copper and other heavy metals from ing 150ml of copper (II) ion solution of known concentra- waste water streams (2, 3, 4, 5). Different types of tions (10, 15, 25, 35, 55 and 100 mg l-1). Initial pH was biomass have been investigated for the biosorption prop- adjusted with dilute H2S04 and Na0H. The flasks were erties of copper and other heavy metal ions. These include agitated at 100 rpm on a shaker (ROSI 1000) at desired bacteria (6), yeasts (7), bark (8), peat moss (9) fungi, (10, temperatures. After incubation, solution samples were 11, 12) and marine algae (12, 13). On the other hand, taken at definite time intervals and centrifuged at 5000 some synthetic resins, such as Dowex 50x8-200, are used rpm. The unadsorbed copper (II) ions in the supernatant as an adsorbent for Cu and other heavy metals (4). was determined spectrophotometricaly.

363 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

The metal uptake (q) for the construction of sorption Pollen also has some polymers such as cellulose, pectin isotherms was determined as follows: and cutin in their inner and outer membrane and these have also both carboxylic groups and some free radicals (14, 19). However, we have not enough information about q = V (Ci - Cf)/1000M the mechanism of copper adsorption by pollen. where q is the metal uptake (mg Cu g-1 of dried The equilibrium isotherms in Fig. 1 were analyzed by pollen), V is the volume of solution in the contact batch using the Langmuir equation, and the Langmuir constants flask (ml), Ci is the initial concentration of Cu (II) in the obtained from linear regression analysis are given in -1 solution (mg l ), Cf is the final concentration of Cu (II) in Table 1. It was reported that the Langmuir constant b can the solution ( mg l-1 ), and M is the mass of cells (g). serve as an indicator of isotherm rise in the region of lower residual metal concentration, which reflects the Langmuir sorption model was chosen for the estima- “strength” or “affinity” of the sorbent for the solute tion of the maximum metal uptake: (20, 21). As shown in Table1, b was found to increase as the incubation pH was raised and this implies that re- q = qmax .b. Cf / (1+bCf) moval of copper ions at high pH could be more complete than that at low pH. where b is the Langmuir constant. It was reported that Cu (II) adsorption by dead mi-

croorganisms is normally exothermic, thus the extent of adsorption generally increases with decreasing incubation RESULTS AND DISCUSSION temperature (22). The effects of incubation temperature on Cu (II) adsorption was examined and results in Fig. 2 It has been reported that biosorption capacities for suggest that optimal temperature for adsorption on to heavy metals are strongly pH sensitive (13, 17). There- pollen is 20 °C. fore, the effect of solution pH on copper biosorption was The equilibrium sorption of Cu (II) was rapidly studied and copper sorption isotherms for dried pollen at reached at varying initial incubating concentration at pH 6 different pH levels were shown in Fig 1. As shown in (Fig 3). The removal of the solute in about the first Fig 1, solution pH is also important factor in copper ad- 10 min of incubation was 85, 76 and 66 % of the bound sorption by pollen. A low pH of 3 resulted in markedly Cu (II) ions for the initial concentrations of 10, 15 and lower copper uptake. The Cu (II) adsorption capacity of 25 mg l-1, respectively. For the adsorption of copper (II) pollens increased with inreasing pH, to the maximum in ions to pollen, typical physical characteristics of adsorp- the pH range of 4 to 6. The maximum Cu (II) uptake by tion were observed. At the end of a rapid physical adsorp- -1 pollen at an optimum pH of 6 was 6.23 mg g , whilst the tion the equilibrium occurred within 15-20 minutes and removal achieved at pH 3 and 4 was lower. Adsorption at the adsorbed amount never changed for copper (II) ions. It pH above 7.0 was not carried out to avoid any possible was also reported that the character of rapid sorption is of interference from metal precipitation. significant importance in a continuous flow metal treatment system, since it enables optimum metal uptake over It was reported that the pH dependence of heavy short contact times in the process (21). metal uptake could be largely related to the various functional groups on the adsorbent materials and also on For comparison, the values of Qmax and b for copper the metal solution chemistry (5, 17). According to these adsorption on other biosorbents and the synthetic ion- investigators, weak acids such as carboxylic groups are exchange resin are given in Table 2. These values indicate functional groups involved in heavy metal biosorption that the pollen is quite a good biosorbent. It has relatively and these groups are protonated at low pH values. There- high affinity (high b value) to copper compared with other fore, at low pH values, the concentration of hydrogen ions biosorbents and synthetic ion – exchange resin, but lower effectively compete for binding sites, resulting in a re- than that of some other fungal biomasses. duced uptake capacity. However, Muraleedharan and According to our results we think that pollens are at- Venkobacher (18) reported that the free radicals present tractive in treatment of industrial wastewater because of consistently in the biosorbents are not taking part in the their capability to take up copper from a diluted copper copper uptake, and cell wall matrix, which has encom- solution. It was also reported that the copper level in passed and trapped these free radicals, opens up upon discharged wastewater has been regulated to as low as metal uptake, indicating its role in metal uptake. The 1-2 ppm and the performance of a biosorbent in low exposed cell wall matrix, thus, freely interacts with the concentration solutions is an important factor for evalua- metal, resulting in high removal rates and the free radicals tion of metal adsorbent (4). Our results showed that can be used as a probe for further identification of the sta- pollens have quite high adsorptive potential in low cop- ble cell wall components responsible for metal sorption. per concentrations.

364 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

FIGURE 1 - Copper sorption isoterms for pollen at different pH levels

7 pH 3 4 6 5 6 5 q (mg/g) 4

0 0 20 40 60 80 100 120 140 160 C f (mg/L)

FIGURE 2 - Effect of incubation temperature on copper biosorbent uptake by pollen

T( oC) 6 20 30 5 40 50

4 q (mg/g)

0 0 20 40 60 80 100 120 140 160 Cf (mg/L)

FIGURE 3 - Effect of contact time on the sorption of C (II) by pollen at pH 6

100

C0 (mg/L) Cu removal (%) 10 15 40 25 35 55 20 100 150

0 0 15 30 45 60 75 90 105 120 135 Contact time (min)

365 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

TABLE 1 - Langmuir constants of Cu (II) -pollen at 20 oC

2 pH Q max b r 3 0,70 0,130 0,87 4 4,67 0,107 0,99 5 5,78 1,470 0,99 6 6,23 2,044 0,99

TABLE 2 - Langmuir model parameters of copper adsorption on different adsorbents

Adsorbent Type q max b Refs Pollen A plant cell 6,230 2,044 This study P. chrysogenum White-rot fungus 3,905 3,215 (4) Dowex50x8-200 Synthetic resin 1,040 0,258 (4) A .spinosus Plant root tissue 0,206 1,270 (23) Gonoderma Mushroom 0,375 8,890 (24) S. fluitans Marine algae 1,700 0,356 (25)

REFERENCES

[1] Veglio F. and Beolchini F., Removal of metals by biosorption; [8] Freer J., Barza J.and Maturana H., Removal and recovery of a review. Hydrometallurg, 44, 301-316 (1997). uranium by modified Pinus radiata D. don bark. J.Chem. Tech. Biotechnol., 46, 41-48 (1989). [2] Volesky H., May H. and Holan Z.R., Cadmium biosorption by Saccharomyces cerevisiae. Biotechnol. Bioeng., 41, 826-829 [9] Chaney L.R. and Hundemann T.P., Use of peat moss columns (1992). to remove cadmium from wastewater. J. Water Poll. Control Fed., 51, 17-21 (1979) [3] Chang J. and Hong J., Biosorption of Mercury by the inacti- vated cells of Pseudomonas aeruginosa PU21(Rip 64). Bitech- [10] Tobin J.M., Cooper D.G. and Neufeld R.J., Uptake of metal nol. Bioeng., 44, 998-1006 (1994). ions by R. arrhizus biomass. Appl. Environ. Microbiol., 47, 821 (1984). [4] Sing C.and Yu J., Copper adsorption and removal from water [11] Fourest E., Chanal C. and Roux J.C., Iprovement of heavy by living mycelium of white-rot fungus Phanerochaete chryso- metal biosorption by mycelial dead biomass (R. arrhizus, Mu- sporium. Wat. Res., 32, 2746-2752 (1998). cor miehei and Penicillium chrysogenus). FEMS Microbial. Rev., 14, 325-332 (1994). [5] Yin P., Yu Q., Jin B. and Ling Z., Biosorption removal of cadmium from aqueous solution by using pretreated fungal [12] Volesky B., Advances in biosorption of metals: Selection of biomass cultured from starch wastewater. Wat. Res., 33, 1960- biomass types. FEMS Microbial. Rev., 14, 291-302 (1994). 1963 (1999). [13] Matheickal J.T. and Yu Q., Biosorption of lead from aqueous [6] Fry J.C., Gadd G.M. and Herbert R.A., Microbial cotrol of solutions by marine algae Ecklonia radiata. Water Sci. Tech- pollution. In Forty-Eighth Symposium of the Society for Gen- nol., 34, 1-7 (1996). eral Microbiology, University of Cardiff. Cambridge University Press, 59-80 (1992). [14] Yentür S., Bitki Anatomisi. 2th edn. Istanbul Üniversitesi Yayinlari, Istanbul, pp. 474-476 (1995). [7] Norris P.R. and Kelly D.P., Accumulation of metals by bacteria and yeast. Dev. Ind. Microbiol., 20, 299 (1979). [15] Forey P., Wild Flovers of The British Isles and Northern Eu- ropae. First edn., Singapore, Dragon’s World, pp. 193 (1992).

366 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

[16] Davis P.H., Flora of Turkey and the East Aegean Islands. 2th edn., Edinburgh, Univ. Press, pp. 559 (1984).

[17] Fourest E. and Roux J., Heavy metal biosorption by fungal mycelial by – product mecanisms and influence of pH. Apll. Microbiol. Biotechnol., 37, 399-403 (1992).

[18] Muraleedharan T.R. and Venkobachar C., Mechanism of biosorption of copper (II) by Ganoderma lucidum. Bitechnol. Bio- eng., 35, 320-325 (1990).

[19] Kaya Y. and Kadioglu A., Genel Botanik. 6th edn. Atatürk Üniversitesi, Fen-Edebiyat Fakültesi Yayinlari, Erzurum, pp. 111 (1996).

[20] Holan Z.R., Volesky B. and Prasetyo I., Biosorption of cadmium by biomass of marine algae. Biothec. Bioeng., 41, 819- 825 (1993).

[21] Zhao M., Duncan J.R.and Hille P.V., Removal and recovery of zinc from solution and electroplating effluent using Azolla filiculoides. Wat. Res., 33, 1516-1522 (1999).

[22] Özer A., Özer D., Ekiz H.I., Aksu Z., Kutsal T.and Çaglar A., The adsorption of iron (III) ions to Schizomeris leibleinii. Tr. J. Eng. Environ. Sci., 21, 183-188 (1997).

[23] Chen J.P., Chen W.R. and Hsu R.C., Biosorption of copper from aqueous solutions by plant root tissues. J. Ferment. Bio- eng., 81, 458-463 (1996).

[24] Muraleedharan T.R., Ryengar L. and Venkobachar C., Screen- ing of tropical wood-rotting mushrooms for copper biosorption. Apll. Environ. Microbiol., 61, 3507-3508 (1995).

[25] Leuszh L., Holan Z.R. and Volesky B., Biosorption of heavy metal ( Cd, Cu, Ni, Pb, Zn) by chemically reinforced biomass of marine algae. J. Chem. Tech. Biotechnol., 62, 279-288 (1995).

Received for publication: February 15, 2001 Accepted for publication: March 15, 2001

CORRESPONDING AUTHOR

Yusuf Kaya Department of Biology Science and Art Faculty University of Atatürk 25240 Erzurum – TURKEY E-Mail: [email protected]

FEB Vol 10/ No 4/ 2001 – pages 363 - 367

367 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

EFFECT OF RESIDUAL ASHES ON CPMAS-13C NMR SPECTRA OF HUMIC SUBSTANCES FROM VOLCANIC SOILS

P. Conte1, A. Piccolo1*, B. van Lagen2, P. Buurman2 and M. A. Hemminga3

1. Dipartimento di Scienze Chimico-Agrarie, Università di Napoli, Portici (Na) , Italy 2. Laboratory of Soil Science and Geology, Wageningen University, Wageningen, The Netherlands 3. Department of Molecular Physics, Wageningen University, Wageningen, The Netherlands

SUMMARY

Potentially paramagnetic elements such as Fe, Cu, Due to the natural variability of carbon atoms and and Mn, did not affect NMR sensitivity and T1r(H) values surrounding protons in an humic molecule, a single tCP of humic materials from volcanic soils, despite a large ash cannot provide the correct measurement for each carbon. content in the extracts. T1r(H) values increased with A retarded acquisition (long tCP) can prevent a quantita- increasing C/H ratios, thereby confirming a relationship tive observation of the fast relaxing nuclei, such as CH3- between the number of protons bound to carbons and and -CH2-, because of the early onset of their relaxation process of relaxation in solid humic materials. Differences mechanisms. Conversely, a short contact time can fail to completely measure the slow exciting nuclei, such as the in T1r(H) values were explained by molecular composition of the samples. Few carbon-bound protons and high quaternary carbons, since they are not yet excited when molecular rigidity in highly aromatic humic matter the acquisition process begins [2]. reduced the efficiency of the relaxation. Humic samples Conte et al. [3] showed that it is possible to obtain richer in polar carbons showed variable T1r(H) values due near quantitative NMR data and to overcome the prob- to different molecular rigidity conferred by intermolecular lems related to the choice of a correct contact time by hydrogen bonds. Not only the number, but also the applying a variable contact time pulse sequence (VCT) distribution of protons in humic molecules appeared to technique. The VCT experiments enable calculation of an affect T1r(H) values. optimum contact time (OCT), that accounts for the cross polarization rates of all the resonance signals, thereby minimizing errors in quantitative CPMAS 13C-NMR spectroscopy. Moreover, VCT experiments can also be KEYWORDS: volcanic soils, CPMAS 13C-NMR, cross-polarization used to obtain TCH and T1r(H) values for humic time, proton spin-lattice relaxation time, molecular motions. substances [1, 4, 5]. These values can provide an insight

on the molecular characteristics of humic substances in

the solid state [1, 2, 6]. In fact, the higher the T value, CH the lower is the number of protons directly bound to

INTRODUCTION carbons. On the other hand, the lower the T1r(H) value, the higher is the number of protons directly bound to carbons and/or the higher is the mobility in a particular Some of the factors affecting quantitative data from solid state NMR spectra are the cross polarization time molecular environment. (TCH), and the proton relaxation time in the rotating frame A source of error, which may affect the TCH and (T1r(H)). The former is a measure of the time needed to T1r(H) relaxation times is the presence of paramagnetic achieve the polarization transfer from protons to carbons, elements, such as Co2+, Ni2+, Mn2+, Cu2+, and Fe3+ in soils whereas the latter is a measure of the time needed for the or humic extracts [7]. Paramagnetism increases the rate spin-lattice relaxation of the protons [1, 2]. Both these of the relaxation mechanisms to such an extent that times affect the NMR signal intensity of the resonance the lowered and broadened NMR signals may apparently peaks. Signal intensity is also affected by contact time disappear. However, the paramagnetic ions have only (tCP), that is the time interval between NMR pulse and a short term effect since paramagnetism can only signal acquisition, and it is also related to the maximum operate within the distance of few bonds from intensity of a resonance peak. a paramagnetic center [4, 5, 8].

368 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

Notwithstanding the importance of paramagnetic FAs were then eluted from the column by a 1 M NaOH centers in the acquisition of NMR spectra of whole soil, solution, immediately neutralised, dialysed against water only few NMR studies [7, 9, 10] were performed in order and freeze-dried. A list of HAs and FAs extracted from to evaluate the minimum amount of Fe3+ (the most the different volcanic soils is reported in Table 1. abundant soil paramagnetic ion) that affects acquisition of NMR spectra. It was reported that unacceptable NMR Thermogravimetric and elemental analysis spectra are recorded when Fe/C ratio (w/w) in soil The ash and the moisture content of each humic samples is greater than 1. In the case of the whole soil, extract was obtained by thermogravimetric analysis NMR sensitivity can be enhanced if the paramagnetic 3+ (TGA) performed with a Dupont 900 apparatus. Moisture Fe is removed, i.e., by extraction with a solution and ash contents were obtained by measuring weight loss of 2% HF [11, 12]. at 105°C and at 750°C, respectively. Elemental analysis was conducted by using an Interscience EA1108 CHN The aim of this work was to evaluate the effect of elemental analyser. The elemental composition was paramagnetic ions on the acquisition of VCT experiments corrected by ash and moisture content. of humic substances from volcanic soils. Moreover, T1r(H) values were measured to obtain information on the Content of potentially paramagnetic elements molecular mobility of the different polar and apolar HAs and FAs were treated with HNO /HClO domains which constitute the arrangement of humic 3 4 solution at 250°C to oxidize their organic carbon and substances in volcanic soils. solubilize the elements contained in the ashes. The resulting solutions were analyzed with a Perkin-Elmer 3030B Atomic Absorption Spectrometer for their total content of MATERIALS AND METHODS Fe, Mn, and Cu which may potentially be in paramagnetic forms in the humic residual ashes. Soils Soil samples from three different volcanoes in Italy NMR spectroscopy and Costa Rica (Table 1) were collected. Italian soil CPMAS 13C-NMR spectra were obtained by a Bruker samples were from the volcanic areas near Naples and AMX400 spectrometer operating at 100.628 MHz on Rome. A surface horizon was sampled from a buried carbon-13. The rotor spin rate was set at 4500 Hz. Hapludand placed at 1100 m on the Monte Faito under a A recycle time of 1 sec and an acquisition time of present vegetation of Fagus selvaticus, and originated 13 millisec were used. The original NMR spectra are fully from ashes emitted by the nearby Vesuvius volcano near reported elsewhere [3]. According to Conte et al. [3], Naples (Italy). Two soils were sampled from the caldera all the NMR measurements were conducted with VCT of the ancient volcano of Vico, near Rome. Their surface pulse sequence. The contact times were varied from 0.010 horizons belonged to either a Typic Xerumbrept, at 300 m to 10 millisec. A line broadening (LB) of 50 Hz was used under a vegetation of Quercus cerris, or to an Allic Ful- to transform all the FIDs. NMR spectra were divided vudand, at 900 m under Fagus selvaticus. All soils from in different resonance intervals: 0-45 ppm (alkyl carbons), Costa Rica were sampled from the Turrialba volcano, and 45-60 ppm (C-O, and C-N carbons), 60-110 ppm were collected surface horizons (Ah) of two Hydrudands (C-O and anomeric carbons), 110-140 ppm (aromatic at an altitude of 950 and 2020 m, respectively, and of a carbons), and 140-200 ppm (carboxylic carbons). Fulvudand at an elevation of 3160 m. The soil at lower The areas of the 0-45 and 110-140 ppm regions were used altitude was under a tropical mountain rain-forest, while to calculate the hydrophobicity (HB), whereas those the soils at higher altitudes were under pasture. of the 45-60, 60-110 and 140-200 ppm regions were used to obtain the hydrophilicity (HI) of the humic extracts. Humic substances Humic (HAs) and fulvic (FAs) acids were extracted Reproducibility of NMR spectra from the volcanic soil samples (200 g) by common proce- In order to control the reproducibility of NMR dures using 500 mL of a 1 M NaOH and 0.1 M Na4P2O7 measurements, a mixture of HA and FA from the solution [13]. HAs were separated from FAs by treating H horizon of a podzolic soil of a Dutch inland sand dune the extract with HCl to pH=1. The flocculated HAs were was extracted with common procedures as reported earlier partially purified by first dissolving in 0.5M NaOH and [13]. Four different CPMAS 13C-NMR spectra were then precipitating again in 1M HCl. Further ash removal acquired for this humic material by using 0.8 ms as con- from HAs was achieved by shaking twice the humic mat- tact time, 2 sec as recycle time, and 13 millisec as acquisi- ter in a 0.25 M HF/HCl solution for 24 h and dialysed and tion time. The 0.8 millisec contact time was the OCT for freeze-dried. FAs were partially purified by adsorbing on this reference material [3]. The relative standard deviation a XAD-8 column to eliminate the soluble hydrophilic (RSD) of the different spectral areas indicated above impurities (carbo-hydrates and proteins). never exceeded 6%.

369 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

T1r (H) values The subsequent routine purification of HAs with a Equation (1) relates all the parameters affecting a 0.25M HF/HCl solution and of FAs with elution on a 13 XAD-8 resin were not sufficient to quantitatively remove CPMAS C-NMR experiment with the contact time (tCP) the co-extracted ashes. [1]. In this equation ItCP is the signal intensity at contact time tCP, and I0 is the signal intensity when the carbon relaxation is absent. Elemental analyses (Table 2) showed that HA1 had the highest C/H ratio. This may be attributed to its large -1 aromaticity (23%), as shown by the quantitative NMR I(t)= I0a [1-exp(-atCP/TCH)]exp(-tCP/T1r(H)] (1) data of HA1 (Table 3). The large aromatic character of HA1 can be ascribed to the buried condition of the soil where a=1-TCH/T1? (H), tCP is the contact time and I(t) is the signal intensity at t-th contact time value. from which it originated. The buried soil had not been in contact with fresh litter and was thus deprived of the polysaccharidic and peptidic features which are instead When T1r(H)>>TCH, equation (1) can be re-written as [1]: present in other humic extracts from recent soils. ItCP=I0 [exp(-tCP/ T1r(H)] (2) Furthermore, the anoxic condition of the buried soil that, in logarithmic form becomes: appears to have favoured the accumulation of aromatic rather than alkyl compounds [3]. Ln(ItCP)=Ln(I0)-tCP/ T1r(H) (3) Similar C/H ratios were shown by the HA2, HA3, HA5, and HA6 samples, though HA2 presented the largest A graph of Ln(ItCP) vs tCP is a straight line with Ln(I0) content of alkyl carbon (51%, Table 3) and HA3, HA5, as intercept and 1/ T1r(H) as slope. The T1r(H) values for the different resonance intervals indicated above were and HA6 differed in their content of alkyl carbon (20% of evaluated by using equation (3) at contact times varying HA3 against 30% of HA5 and HA6) and oxidized carbon (11% of HA against 30% of HA and HA ). However, between 2 and 10 ms. A general T (H) value for each 6 3 5 1r evaluation of the NMR experimental error (see Materials HA was, then, obtained by a mathematical average of the and Methods) indicated that the differences in the content relaxation times measured for the different resonance of oxidized carbon among the latter three extracts were intervals. The confidence interval was in the 2-11% range significant, whereas those in the content of alkyl carbon for all the T (H)s. 1r were not significant.

Also FA , FA , FA , and FA showed similar C/H ra- Washed sea-sand in CPMAS 13C-NMR experiments 1 2 4 5 tios and composition. Nevertheless, FA1, the fulvic acid Washed sea-sand was added to the humic material from the buried Hapludand (Table 1), showed the lowest when it was not enough to completely fill up the space content of oxidized carbon (4%) and the highest content inside the rotor. Since the intensity of the applied of carboxylic carbon (45%). As in the case of HA1, such a magnetic field on the extremities may differ between the small content of oxidized carbon can be ascribed to the inner and outer regions of the spin rotor, thereby leading buried condition of the soil from which FA originated. 13 1 to errors in quantification of CPMAS C-NMR spectra The discrepancy between the content of carboxylic carbon [4, 5], the HA/sea-sand mixtures were homogenized in a in FA1 and that in the other fulvic extracts, may be attrib- mortar before each NMR analysis. This procedure uted to the different cross-polarization and/or relaxation provided a homogeneous distribution of the organic time required to the C nuclei of FA1. In fact, due to a matter throughout the spin-rotor. possible high proton homogeneity or its ash content, the FA1 sample can cross-polarize or relax faster than the other humic and fulvic extracts. In the following sections RESULTS AND DISCUSSION the factors affecting the NMR parameters of all the humic extracts of this study are being discussed. HA and FA fractions Chemical properties and composition of the various humic fractions are given in Tables 2 and 3. There was Effect of paramagnetism a noticeable difference in purity among samples. The factors affecting the signal intensities in CPMAS 13 HAs from Costa Rica (HA5 and HA6) and the one from C-NMR measurements of humic substances are Vico (HA3) were the most purified samples, whereas all the number of protons in the organic matter, the internal other humic extracts showed an ash content ranging from molecular motions of the functional groups, and the 10% to 78% (Table 2). This large residual ash may content of paramagnetic ions in the whole soil. be attributed to the strong dispersive effect of pyrophos- The paramagnetic ions present in the residual ashes phate on the amorphous silicate material in these volcanic of humic extracts from volcanic soils may be particularly soils, which enabled high co-solubilization of ashes in large and, thus, affect the intensities of NMR signals the humic extracts [14]. during measurements.

370 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

TABLE 1 Origin and classification of soils, from which humic (HA) and fulvic (FA) acids were extracted Soils and altitude (m) Horizons Classification Extracts

Monte Faito (Naples), 1100 A Hapludand buried HA1, FA1

Caldera of Vico (Rome), 300 A1 Typic Xerumbrept HA2, FA2

Caldera of Vico (Rome), 900 A1 Allic Fulvudand HA

Turrialba volcano (Costa Rica), 950 Ah Hydrudand FA3

Turrialba volcano (Costa Rica), 2020 Ah Hydrudand HA4, FA4

Turrialba volcano (Costa Rica), 3160 Ah Fulvudand HA5

TABLE 2 Moisture (105°C), ash content (moisture free), and elemental analyses (moisture and ash free) of the humic and fulvic extracts Sample Moisture (%) Ash (%) C (%) N(%) H(%) C/H

HA1 7.24 14.84 54.88 3.36 2.60 21.1

HA2 8.34 10.03 53.67 4.26 4.86 11.0

HA3 7.21 3.17 54.72 4.25 5.23 10.5

HA4 7.07 1.68 55.99 5.02 5.38 10.4

HA5 9.24 0.82 55.93 2.93 4.62 12.1

FA1 3.82 77.50 42.36 3.11 3.01 14.1

FA2 13.40 13.12 47.43 3.06 3.49 13.6

FA3 13.87 37.78 63.47 3.96 4.65 13.6

FA4 11.69 15.21 48.43 4.77 3.64 13.3

TABLE 3 Distribution of Carbon content by solid state NMR of HAs and FAs Chemical shift ranges (ppm) Samples 0-45 45-60 60-110 110-140 140-200 HI (%) HB (%)

HA1 30 15 8 23 24 47 53

HA2 51 18 13 8 10 41 59

HA3 20 30 23 11 16 69 31

HA4 30 30 17 10 13 60 40

HA5 30 11 26 13 20 57 43

FA1 21 4 15 15 45 64 36

FA2 24 24 18 13 19 63 37

FA3 26 18 20 16 20 58 42

FA4 39 20 18 9 14 52 48

371 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

The effect of paramagnetism results in a decrease of T1r (H) values as measure of molecular mobility the TCH and T1r(H) values with a consequent peak In the absence of significant paramagnetic effect on broadening and reduction of the intensity of NMR signals. NMR measurements, the number of protons present in the Several works [7,9,10,11] reported that signal intensities humic molecules and their internal molecular motions decrease with increasing Fe/C (w/w) ratios in soils. become the molecular factors which must affect the In particular, it has been shown that good NMR spectra T1r(H) values. are obtained when Fe/C<1, while only reasonable spectra are recorded when Fe/C=1, and poorly significant spectra For rigid molecules, cross polarization is dominated are obtained when Fe/C»1 [10]. by the number of protons bound to the carbons. This implies that the larger the number of protons and/or the The effect of paramagnetism depends on both the faster the molecular motions, the better is the efficiency of concentration and nature of the paramagnetic species [8]. the proton spin-lattice relaxation and the shorter is the 3+ For this reason not only the effect of Fe , but also that of T1r(H) value. Moreover, the efficiency of the proton spin- other common soil ions such as Mn2+, and Cu2+ should be lattice relaxation depends on the homogeneity of the pro- investigated. Smernik and Oades8 showed that Mn2+ ton distribution over an organic molecule. Therefore, the 3+ 2+ is more effective than Fe and Cu , in the order, in T1r(H) values are progressively shorter with increasing affecting the signal intensities of CPMAS 13C-NMR homogeneity of proton distribution [15]. 2+ spectra. They showed that Mn decreased signal intensity In the case of humic and fulvic extracts of this study, by 77%, whereas Fe3+ and Cu2+, by 68% and, respectively we observed that HA1 has the highest aromaticity (23%) by 50%. These results agree with those of Skjemstad et al. (Table 3), thereby accounting for the highest molecular [11] who, generally, reported that when paramagnetic rigidity among the extracts. Consequently, the proton ions do not represent a substantial part of the inorganic spin-lattice relaxation times of HA1 should be mainly matrix (in either the whole soil or residual ashes) and the affected more by the number of protons than by their carbon content is ³3% (as in the case of the humic and 13 internal molecular motions. Since the number of protons fulvic extracts of this study), CPMAS C-NMR spectra in a highly aromatic material is smaller than those con- can be used for semi-quantitative analysis. Smernik and tained in a sample of less aromatic character, the effi- Oades also concluded that all paramagnetic ions had a 13 ciency of the proton spin-lattice relaxation must be lower short range effect, since only the C nuclei directly bound in the former than in the latter material. In the highly to the paramagnetic centers showed significant reduction aromatic HA1 (Table 3) the average T1r(H) is expected to of signal intensity [8]. be the longest of all other humic extracts. This is proved

by the findings of this study (Table 5). While the work of Smernik and Oades was conducted on soils which were artificially enriched with paramag- The HA2 extract has the largest content of aliphatic netic ions, the concentration of paramagnetic ions in carbons (51%) (Table 3), and thus, its large percentage of humic extracts of the present study was smaller than that CH3 and CH2 groups can account for the shortest T1r(H) reported by Smernik and Oades for their amended value (Table 5). In this case, the proton spin-lattice systems [8]. It can be thus assumed that the paramagnetic relaxation mechanisms are dominated by the large effect on spectra acquisition in this study should be lower. number of protons directly bound to the carbons rather than by molecular motions. This assumption is based on We measured the content of iron, copper, and the consideration that the aliphatic carbons may be mainly manganese in the humic and fulvic extracts, and used the arranged in large hydrophobic domains which can inhibit amount of carbon in the same extracts to obtain the total the internal molecular motions of HA2 components by metal-to-carbon ratios shown in Table 4. The Fe/C ratio steric hindrances, thereby limiting relaxation mechanisms. was always lower than 1, except for FA1 whose ratio HA3, HA4, and HA5 had a very similar chemical was 1.1. Since the Fe/C was generally £1 in all humic composition (Table 3), but their T1r(H) values resulted extracts, we assumed, according to previous literature very different (Table 5). HA showed the shortest proton [7, 9, 10, 11], that paramagnetic iron did not affect the 3 spin-lattice relaxation time, whereas the T (H) values sensitivity of our NMR measurements. However, the 1r were very similar for HA and HA . residual ashes of humic extracts also contained copper, 4 5 and manganese (Table 4), which may potentially be in As it has been mentioned above, T1r(H) is affected paramagnetic forms and affect NMR measurements. Our not only by the number of protons, but also by the internal results showed that Cu/C and Mn/C ratios were from one molecular motions15,16 which in turn, depend on the spatial or two orders of magnitude lower than the Fe/C ratio arrangement of the humic extracts. Because of the higher (Table 4). Based on the findings of Smernik and Oades hydrophilicity of HA3 (Table 3) we may assume that the [8] and Skjemstad et al. [11] we concluded that also the relative orientation of the polar groups in the solid state effect of manganese and copper was negligible in acquir- structures of HA3 may form a larger number of H-bonds 13 ing the CPMAS C-NMR spectra of this study. than those possible in the structures of HA4 and HA5.

372 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

TABLE 4 Content of Fe, Mn, Cu, and Cr calculated as percentage of the whole humic extracts. The metal-to-carbon ratios are obtained by using the carbon content (%) of the whole humic extracts Samples Fe (%) Fe/C Cu (%) Cu/C Mn (%) Mn/C x 10-3 x 10-3 x 10-3

HA1 1.5 35 0.011 0.26 0.12 2.8

HA2 1.4 33 0.0065 0.14 0.19 4.4

HA3 0.50 10 0.077 1.5 0.017 0.30 -2 HA4 0.071 1.4 0.0030 5.9x10 0.0077 0.10 -2 HA5 0.095 1.9 0.0009 1.8x10 0.011 0.20

FA1 8.4 1100 0.014 1.8 0.12 15

FA2 0.85 24 0.0038 0.10 0.012 0.30

FA3 0.38 12 0.0094 0.30 0.013 0.40

FA4 0.29 0.83 0.0069 0.20 0.0071 0.20

TABLE 5

T1r (H) values of the humic and fulvic extracts. The average values are obtained by averaging the relaxation times measured for the resonance signal at 172, 104, 72, 56 and 31 ppm

Samples Average T1r (H) (msec)

HA1 6080

HA2 2713

HA3 3526

HA4 5341

HA5 5322

FA1 4112

FA2 4761

FA3 3111

FA4 3638

For this reason the rates of molecular motion in HA4 Similar considerations apply to the results on fulvic and HA5 should be larger than for HA3. Based on these extracts. The higher hydrophilicity of FA1 and FA2 considerations we must expect that the T1r(H) of HA4 and compared to that of FA3 and FA4 (Table 3) may allow HA5 should be shorter than that of HA3. Conversely, the larger number of H-bonds in the former two fulvic acids experimental measurements showed an opposite behavior rather than in the latter ones. Larger steric hindrances (Table 5). We must then assume that the relaxation may then be established in FA1 and FA2 solid structures mechanisms should also depend on homogeneity of the because of their larger number of H-bonds, and are proton distribution in the humic samples and not only on likely to inhibit their internal molecular motions more the number of protons. This explanation may justify the than in FA3, and FA4. Therefore, longer T1r(H) values lower efficiency of the proton spin-lattice relaxation in should be expected for FA1 and FA2 than for FA3 and HA4 and HA5 than in HA3, thereby accounting for the FA4. Table 5 shows that the experimental T1r(H) values shorter T1r(H) value of HA3. confirm the expected behavior.

373 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

CONCLUSIONS REFERENCES

This work investigated, for the first time, the effect 1. Wilson, M.A., 1987, NMR Techniques and applications in of the residual ashes of humic and fulvic extracts on the geochemistry and soil chemistry, 1st ed., Pergamon Press, London. NMR measurements. Oxides-rich volcanic soils may contain paramagnetic ions which can persist in the 2. Piccolo, A., Conte, P., 1998, in P.M. Huang, N. Senesi, J. Buffle, Structure and surface reactions of soil particles, Wiley residual ashes of humic and fulvic materials after extrac- New York, pp 183-250. tion and routine purification. It is well known that paramagnetism affects the rate of the relaxation mechanisms 3. Conte, P., Piccolo, A., van Lagen, B., Buurman, P., de Jager, P.A., 1997, Geoderma, 80, 327-338. to an extent that may undermine the significance of NMR spectra. Our results showed that common puri- 4. Smernik, R.J., Oades, J.M., 2000, Geoderma, 96, 101-129. fication procedures provide humic and fulvic extracts 5. Smernik, R.J., Oades, J.M., 2000, Geoderma, 96, 159-171. containing a concentration of paramagnetic ions in the 6. Preston, C.M., 1996, Soil Sci., 161 (3), 144-166. residual ashes, that is low enough to allow reliable NMR measurements. Moreover, according to previous 7. Vassallo, A.M., Wilson, M.A., Collin, P.J., Oades, J.M., Wa- results4,5 addition of quartz to humic samples had the ters, A.G., Malcom, R.L., 1987, Anal., Chem., 59, 558-562. only effect to decrease the NMR signal-to-noise ratio. 8. Smernik, R.J., Oades, J.M., 1999, Geoderma, 89, 219-248. The use of quartz may, then, recommended when the 9. Arshad, M., Ripmeester, J.S., Schnitzer, M., 1988, Can. J. quantity of humic material is not sufficient to com- Soil Sci., 68, 593. pletely fill the rotor-spinner. 10. Preston, C.M., Schnitzer, M., Ripmeester, J.A., 1989, Soil Sci. Soc. Am. J., 53, 1442. 11. Skjemstad, J.O., Clarke, P., Taylor, J.A., Oades, J.M., New- Our results also showed that the use of variable man, R.H., 1994, Aust. J. Soil Res., 32, 1215-1229. contact time experiments3 enables to measure proton 12. Dai, K. H., Johnson, C.H., 1999, Geoderma, 93, 289-310. spin-lattice relaxation (T1r(H)) times which may be related to the humic and fulvic conformational structure 13. Stevenson, F.J., 1994, Humus Chemistry, Genesis, Composi- in the solid-state. Humic samples whose molecular tion, Reactions, 2nd ed., J. Wiley and Sons Inc., New York motions are inhibited by the formation of tight molecu- 14. McKeague, J.A., Schuppli, P.A., 1982, Soil Sci., 134, 265- lar associations17 stabilized by hydrogen or dispersive 270 (hydrophobic) bonding are likely to provide different 15. Alemany, L.B., Grant, D.M., Pugmuire, R.J., Alger, T.D., relaxation mechanisms than those which are more Zilm, K.W., 1983, J. Am. Chem. Soc., 105, 2133-2141. loosely associated. In fact, humic samples with faster 16. Alemany, L.B., Grant, D.M., Pugmuire, R.J., Alger, T.D., internal molecular motions provide T1r(H) values Zilm, K.W., 1983, J. Am. Chem. Soc., 105, 2142-2147. shorter than those measured for HAs and FAs with 17. Piccolo, A., Conte, P., 2000, Advances in Environmental Re- slower internal molecular motions. The results of this search, 3 (4), 508-521. study also implies that the proton distribution may not be homogeneous in humic samples having larger molecular motions, thereby providing T1r(H) values even longer than those theoretically expected.

Received for publication: December 27, 2000 ACKNOWLEDGMENTS Accepted for publication: March 15, 2001

The first author gratefully acknowledges a fellowship “Access to Large-Scale Facilities (LSF)” received from the European Union within the program “Human Capital CORRESPONDING AUTHOR and Mobility” that enabled experimental work at the Wageningen NMR Centre. A. Piccolo 1. Dipartimento di Scienze Chimico-Agrarie Università di Napoli via Università 100 80055, Portici (Na)-ITALY

E-Mail: [email protected] FEB – Vol 10/ No 4/ 2001 – pages 368 - 374

374 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

POLYCHLORINATED BIPHENYLS IN SEDIMENTS IN EASTERN SLOVAKIA

Ján Petrík, Anton Kocan, Stanislav Jursa, Beata Drobná, Jana Chovancová and Marián Pavúk*

Institute of Preventive and Clinical Medicine, Bratislava, Slovak Republic *University of Texas-Houston, School of Public Health, MPH Program in Dallas, Dallas, TX ,USA

SUMMARY

Surface water sediments were collected from an area Surface sediments contained from less than 1 to more around the former PCB production facility in the than 1500 mg PCBs.kg-1. Very high PCB concentrations - Michalovce District and in a comparison location in the up to 26.3 mg.kg-1 - were found in sediments of the Oder Stropkov District of Eastern Slovakia. PCB levels (sum of River estuary (Germany) and the Western Baltic Sea [4]. 9 congeners) ranged from 0.004 to 4100 mg.kg-1 of dry The contamination of sediments of the lake on the Eman weight (dw) in samples taken close to the plant, while the River (Sweden) originated from paper recycling and concentrations in the comparison area were in the range varied between 0.002 and 31 mg.kg-1 [5]. Contamination of 4.3-64 mg.kg-1dw. Samples from the vicinity of the of sediments from the Seine River basin (France) ranged PCB production facility reflect high amounts of accumu- from 0.05 to 26 mg PCBs.kg-1 [6]. The highest levels were lated PCBs even after 14 years after the ban of their pro- measured in Paris. The amount of PCB discharged into duction. The industrial effluent canal seems to be the the lower Hudson River after cessation of PCB discharges main source of PCB pollution of aquatic ecosystem in an has decreased from ~2 ton y-1 in the late 1970’s to adjacent area. The results of the analyses of individual 1 ton y-1 or less in the 1980’s [7]. The sediment PCB PCB congener patterns indicate high proportion of tri- and concentrations now vary from 0.08 to 1.41 mg.kg-1. tetraCB in the samples we investigated. The content of PCBs in sediments of Polish harbors’ and shipyards’ canals reached up to 1300 ng.g-1, while samples from water bodies without direct impact of urbaniza- tion and/or industrial activities showed much lower levels KEYWORDS: PCBs; polychlorinated biphenyls; sediment; fish 1.7-2.2 ng.g-1 [8]. Similarly, concentrations in sediments from water canals of Ho Chi Minh city (Vietnam) ranged -1 between 9.3-591 ng.g [9] Urban sediments were more contaminated than the ones from rural zones of the city. INTRODUCTION The same phenomenon was observed for DDTs. The average p,p’-DDE concentration was 29.1 ng.g-1 and -1 Polychlorinated biphenyls (PCBs) and organochlorine concentration of p,p’-DDT was 23 ng.g . pesticides in surface water are adsorbed on particulates, accumulate in sediments and subsequently concentrate in The environmental contamination with PCBs in the tissues of water organisms. Due to their possible mobili- Slovak Republic was caused mainly by the Chemko zation from sediments and accumulation in the food chain chemical plant located in the town of Strážske (the [1], and finally in the adipose tissue of humans, they rep- Michalovce District). 21481.8 tones of PCBs (trade marks resent a potential risk for human health. DELOR, HYDELOR, DELOTHERM) in total was manufactured between 1959-1984. It is estimated that about The impact of industrial activities on environmental 1600 tones of PCB waste products (mainly distillation pollution involving PCBs was investigated in many stud- residues containing highly chlorinated biphenyls and ies. Extremely high PCB levels, up to 66500 mg.kg-1, terphenyls, and PCDFs) were generated during the manu- were found in estuarine sediments as a result of the 31- facture [10]. Most of this waste is still stored in open-air year production of electrical capacitors filled with PCBs waste sites waiting for a safe and definitive disposal. in New Bedford Harbor (Massachusetts, USA) [2]. Use of Especially at the beginning of the production in 1960's, hydraulic fluids on the basis of Aroclor 1248 and Aroclor large amounts of PCBs were released into the environ- 1254 in paints’ production - between 1959 and 1971 was ment (mainly into watercourses and soil) due to poor the reason for the Sheboygan River (Wisconsin) massive technological measures. In addition, because of former contamination [3]. ignorance of local and state regulation authorities about

375 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

the environmental persistency of PCBs and their harmful A comparison system consisted of the Ondava River and effects on living organisms, almost no attention was paid the Velká Domaša water reservoir (surface area to PCB releases at that time. Besides PCBs, we also 15.1 km2, mean depth 25 m) in the geographically simi- measured levels of HCB and p,p’-DDE in this study. lar district of Stropkov (Figure 2, Table 1). Both the These pesticides were extensively used in Slovakia in the Zemplínska Šírava and the Velká Domaša Reservoirs past. DDT application in the former Czechoslovakia cul- serve flood control, irrigation and recreational purposes. minated in the 1960's and the use substantially decreased The bottom sediment samples were collected with the after 1970. In 1974, the production of DDT was banned, Ekman dredge in 1997-98. Approximately 5 cm of the however, its use continued till the depletion of stores. upper sediment core was sampled. In rivers and/or in HCB was used in agriculture as a selective fungicide from shallow waters, sediments were sampled manually with 1945. The use of HCB was banned in 1985 [11]. a scoop. Ambient air, room temperature dried (5 days) sediments, separated to fraction of 2 mm, were used for analysis. The sieved sediments were stored in the dark at MATERIALS AND METHODS room temperature in the pre-cleaned (acetone, n-hexane) glass PTFE sealed vials till analysis. Sampling sites FIGURE 2 The area of interest was the aquatic ecosystem sur- Sampling sites for sediments in the Stropkov District. rounding the former PCB producer: industrial effluent canal below the water treatment facility at the outlet of the factory - concrete bed with silt, effluent canal from the municipal water treatment plant (MWTP), merged effluent canal (Chemko + MWTP canals), the Laborec River, STROPKOV and the Zemplínska Šírava water reservoir (ZŠ, surface area 33.5 km2, mean depth 10 m) with filling and outlet canals (Figure 1, Table 1).

ONDAVA HILLS FIGURE 1 Sampling sites for sediments in the Michalovce District Olšavka StreamRiver Ondava

VIHORLAT VELKÁ DOMAŠA HILLS

né a k

i e V

ZEMPLÍNSKAŠÍRAVA

Sample preparation 13 Isotopically labeled internal standard (mixture of C12 labeled PCB-28, 52, 101, 105, 118, 138, 153, 180, 202, HCB, pp'-DDT and p,p’-DDE ) was added to aliquots of homogenized sediment samples, (2-10 g) which were then

Senné Fish Ponds Soxhlet extracted for 16 hrs. with 5% (v/v) diethyl ether in n-hexane.

376 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

Interferences were cleaned up from extracts on an As a consequence of water flow from industrial canal to H2SO4/silica column. After elution, residues of analytes the Laborec River and the inflow canal to ZŠ Reservoir, were fractionated on a basic alumina column. Dichloro- we observed high levels of PCBs in waters, sediments and methane-hexane fractions were analyzed by high resolu- fish. PCB levels in sediments and in waters of the La- tion gas chromatography with mass selective detection in borec River below the junction with the merged canal are, a single ion monitoring mode (HRGC/LRMS-SIM) for respectively, 20- to 190-times, and 10-times higher than the presence of PCBs, HCB and p,p'-DDE. A DB-5 (J&W the levels above this junction (Figure 1, Table 1). Scientific, USA) capillary column, 0.25 mm ID × 60 m and 0.25 mm film thickness was used for the separation. Regarding the time trends of contamination, it can be argued, in agreement with published data [14, 18, 19], that older ZŠ sediment layers contain much higher Individual PCB congeners and organochlorine pesti- levels of PCBs than those presented in Table 1. The cides were quantified by a modified isotope dilution processes of adsorption and the following sedimentation method [10. 12]. Calibration curve of individual com- can immobilize PCBs in water system for relatively long pounds was constructed on 4 concentration levels and time [20, 21]. Significant amounts of PCBs in water checked periodically in accordance with internal QA/QC sediments act as a reservoir from which they are re- rules. The native PCB congeners without labeled equiva- leased to the environment. Accumulated PCBs can thus lents in internal standards were quantified on the basis of contaminate waters for long time if this is not remedi- 13 relative response factors to the C12 labeled congeners in ated by compensatory clean-up activities. From this the same retention window monitored [10]. point of view the Chemko effluent canal (length about 4 km), the Laborec River below the Chemko chemical plant and the ZŠ Reservoir very likely represent a large deposit of PCBs. With regard to our results (Table 1),

we assume the presence of approximately 5-10 tones of RESULTS AND DISCUSSION PCBs accumulated in sediments of these water bodies. This correlates with the value of about 0.7 tones of tech- PCB levels in sediments from the Michalovce District nical formulations, which was estimated to be dis- were substantially higher than levels in samples from the charged directly into the common waste water system Stropkov District. As presented in Table 1, sediments every year between 1959 and 1974 [10]. with high content of organic matter from the merged waste canal (see Figure 1, S06/20) act as a dominant PCB While PCB levels in sediment samples from the La- source even 14 years after cessation of the PCB produc- borec River basin ranged between 0.011 and 9.7 mg.kg-1, tion. This finding is supported by the high PCB levels in that in sediments from the control Stropkov District var- surface water sampled at the same time. The concentra- ied only between 0.004 and 0.064 mg.kg-1. HCB levels in tions in water samples from this canal were significantly -1 all sediments ranged from <0.04 to 2.5 ppb dw. Observed higher (1070 and 2830 ng PCB.L ) than levels in water levels of DDT/DDE reached up to 79 ppb dw. There were canal from the municipal WTP of the city of Strážske -1 no significant differences observed in levels of or- (64 ng.L ). ganochlorine pesticides between contaminated and the comparison area except the waste water canals. The content of PCBs in the sediment from the Chemko effluent canal was much lower (48 mg.kg-1) than The consequences of water contamination in the in sediments from the merged canal (4100 and Michalovce District can be manifested in the PCB levels 1900 mg.kg-1). This is because of the low content of or- in fish. Fish from contaminated Michalovce waters con- ganic carbon in the industrial effluent canal sediments. It tained about hundred times higher PCB levels (lipid ad- is known that sediments with low organic content can justed) than those caught in Stropkov (Figure 4). High adsorb much lower amounts of PCBs than samples with concentrations result from high level of bioaccumulation high organic matter portion [13]. Concentrations and PCB of the persistent compounds in fish [20, 22-24]. The sub- profile in sediments generally show high similarity with stantially higher levels of PCBs in predacious fish in data on production and usage of these compounds comparison to planktonic and/or benthic fish species [14-17]. Patterns of PCBs in sediments and surface wa- (Figure 4) can be explained as due to the exposure of the ters, respectively, are very similar to the proportion of former to contaminated food chain [25, 26], i.e. benthic produced amounts of DELOR 103 (equivalent of Aroclor organisms. An average PCB congener pattern in fat of 1242) - 14140.5 tones and DELOR 106 (Aroclor 1260) - 117 individual fish caught (pooled to 31 samples) is 4380.7 tones ( Figure 3). shown in Figure 5.

377 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

TABLE 1 Sampling sites, date and measured concentrations [µg.kg-1] of analytes of interest in sediments from the Michalovce and Stropkov Districts (Results are not organic carbon adjusted)

Sample Sampling site description S PCB HCB pp’DDE Date S-01 Zemp. Šírava 1 (west edge) 1700 1.1 15 X/97 S-30 -“- 3000 <0.2 8.9 VI/98 S-02 Zemp. Šírava 2 (middle edge) 2300 0.1 19 X/97 S-31 -“- 3900 <0.7 11 VI/98 S-03 Zemp. Šírava 3 (east edge) 2200 <0.07 20 X/97 S-32 -“- 1200 <0.4 9.1 VI/98 S-04 Filling canal to Z.Š. (Zbudza village) 9400 2.0 79 X/97 S-16 -“- 2600 <0.7 3.0 VI/98 S-15 Outlet canal from Z.Š. 220 <0.3 2.2 X/97 S-22 -“- 600 <1.1 5.0 VI/98 S-05 Laborec River above the Strážske city 54 0.31 7.0 X/97 (Krivoštany village) S-17 -“- 51 <2.7 4.1 VI/98 S-07 Laborec River (1) below the Strážske city (Vola 9700 <0.5 54 X/97 village) S-24 -“- 830 <0.9 2.3 VI/98 S-25 Laborec River (2) below the Strážske city (Las- 1100 <0.9 7.0 VI/98 tomír village) S-08 Canal Cierna voda (Blatné Revištia village) 260 <0.1 14 X/97 S-21 -“- 290 <0.25 6.9 VI/98 S-09 Senné fish ponds (Inacovce village) 95 <0.06 3.0 X/97 S-23 -“- 11 <0.4 2.2 VI/98 S-18 Effluent canal from Chemko Co. 48000 <0.6 3.5 VI/98 S-19 Canal from municipal WTP Strážske 1600 24 45 VI/98 S-06 Merged canal (Chemko effluent + municipal 4100000 120 36 X/97 WTP canals) S-20 -“- 1900000 72 40 VI/98 S-10 Velká Domaša 1 (Turany n/Ondavou village) 12 1.1 15 X/97 S-28 -“- 7.8 <0.3 1.6 VI/98 S-11 Velká Domaša 2 (Bžany village) 9.5 0.50 9.8 X/97 S-29 -“- 4.3 2.5 8.2 VI/98 S-12 Ondava River (1) (river estuary to V. Domaša) 64 0.70 6.2 X/97 S-27 -“- 41 <0.8 1.0 VI/98 S-13 Ondava 2 (Nižná Olšava village) 5.7 <0.04 2.6 X/97 S-26 -“- 23 <0.2 5.8 VI/98

378 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

FIGURE 3 Individual PCB congener profile in technical formulations of DELOR 103 and DELOR 106 (mixture 3:1) and an average profiles in sediment and water from the Zemplínska Šírava Reservoir.

relative weight %

25 DELORs WATER 20 SEDIMENT 15

#28 #52

#101 #118 #138 #153 #156 #170 #180

P5CB

T3CB T4CB

D2CB H6CB H7CB O8CB

FIGURE 4 PCB levels (the sum of 28, 52, 101, 118, 138, 153, 156, 170, and 180 congeners) in lipids from fish caught in some watercourses in the districts of Michalovce and Stropkov.

224 375 292

Michalovce District Stropkov District

1.5 Planktonic+Benthic Species 5.2 Predacious Species 2.9 Average

0 100 200 300 400 PCB Concentration [mg/kg, lipid basis]

FIGURE 5 An average PCB profile of fish from the Laborec River and ZŠ reservoir relative weight % 20

#28 #52

#101 #118 #138 #153 #156 #170 #180

P5CB

T3CB T4CB

D2CB H6CB H7CB O8CB .

CONCLUSIONS

The content of PCBs, HCB and p,p’-DDT + p,p’- Concentrations of PCBs in samples from contaminated DDE in upper layer sediments of water system near the area were 10 to 200-times higher than levels in the com- former PCB production facility (1959-84) was investi- parison area. Extremely high concentration - 4100 mg.kg- gated in this study. We have chosen the Stropkov District 1 in dry sediment - was found in the industrial canal from - the Ondava River and recreation basin Velká Domaša - the former PCB producing factory. This attests to the as a comparison area. The analyses showed considerable relatively large amounts of PCBs accumulated in sedi- contamination in the location of the former PCB producer ments of the Laborec River, and mainly in the Zemplínska even 14 years after the cessation of the manufacture. Šírava Reservoir, which are directly connected to the

379 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

waste water system of the former PCB-producer. The levels of contamination of sediments and waters of the Ondava River, Velká Domaša basin and/or the Laborec 13. M. Chevreuil, M. Blanchard, M.J. Teil, A. Chesterikoff, Wat. Res. River above the junction with the industrial canal, and 32, 1204-1212 (1998). also in waters without direct influence of contaminated 14. R. Bossi, B. Larsen, G. Premazzi, Sci. Total Environ. 121, (77-3) water circulation are similar to the levels in drinking wa- 1992. ter basins and can be considered as background levels 15. K. Masahide, J. Falandysz, B. Brudnowska, T. Wakimoto, Orga- nohalogen Compounds 39, 331-335 (1998). [27]. The presence of persistent organochlorine pollutants 16. J.D. Jeremiason, S.J. Eisenreich, J.E. Baker, B.J. Eadie, Environ. in these water bodies are the consequence of long range Sci. Technol. 32, 3249-3256 (1998). transport processes and dry or wet deposition. The levels 17. M. Bazzanti, S. Chavarini, C. Cremisini, P. Soldati, Environ. Int. of organochlorine pesticides in the Michalovce District 23, 799-813 (1997). were not significantly higher than levels in the Stropkov 18. G. Sanders, K.C. Jones, J. Hamilton-Taylor, Environ. Sci. Technol. District. Contamination of the environment with these 26, 1815-1821 (1992). compounds is based mainly on the extensive use of p,p’- 19. K. Sugiura, Chemosphere 24, 427-432 (1992). DDT and HCB in agriculture till 1974, and 1985, respec- 20. Toxicological profile for selected PCBs (Aroclor-1260, 1254,1248, tively [28]. 1242, 1232, 1221 and 1016). ATSDR/TP-92/16, U.S. Department of Health & Human Services, Public Health Service, U.S. Gov- ernment Printing Office, 209 pp (1993). ACKNOWLEDGMENTS 21. J. Lohse, Mar. Pollut. Bull. 19, 366-371 (1988). 22. H. Geyer, I. Scheunert, F. Korte, Regul. Toxicol. Pharmacol. 6, 313-347 (1986). The authors acknowledge the assistance of P. Hucko, PhD 23. A.C. De Kock, D.A. Lord, Chemosphere 17, 2381-2390 (1988). from Water Research Institute in Bratislava for his help in collecting of samples from the Zemplínska Šírava Reser- 24. M.J. Kennish, B.E. Ruppel, Estuaries 19/2A, 288-295 (1996). voir. 25. A. Subramanian, S. Tanabe, H. Hidaka, R. Tatsukawa, Environ. Pollut. Ser. A 40, 173-189 (1986).

26. S. Jursa, J. Chovancová, A Kocan, J. Petrík, B. Drobná, Contami- nation of fish and games with polychlorinated biphenyls in the de- REFERENCES partment of Michalovce. In: Proceedings of Conference Foreign Substances in Food, Tatranská Štrba, Slovak Republic, 98-99 1. M. Salizzato, V. Bertato, B. Pavoni, A.V. Ghirardini, P.F. Ghetti, (1999), (in Slovak). Environ. Toxicol. Chem. 17, 655-661 (1998). 27. P. Hucko, Occurrence of organic substances in waters and sedi- 2. J.L. Lake, R.J. Pruell, F.A. Osterman, Marine Environ. Res. 33, 31- ments of selected water reservoirs of Slovakia. In: Proceedings of 47 (1992). Conference Hydrogeochemistry ’98, Dept. of Hydrogeology, Fac- ulty of Natural Sciences, Comenius University, Bratislava, Slovak 3. M. David, L.M. Brondyk, W.C. Sonzogni, J. Great Lakes Res. Republic, 51-62 (1998), (in Slovak). 20(3), 510-522 (1994). 28. A. Kocan, J. Petrík, J. Chovancová, B. Drobná, H. Uhrinová, I. 4. D. Dannenberger, R. Andersson, C. Rappe, Mar. Pollut. Bull. Holoubek, K. Magulová, D. Sulovec, Z. Keppertová, K. Spišáková, 34/12, 1016-1024 (1997). Pollution of ambient air by perzistent organic pollutants in the Slo- 5. G. Bremle, P. Larsson, T. Hammar, A. Helgée, B. Troedsson, vak Republic [Report]. Slovak Hydrometeorological Institute, Bra- Water Air Soil Poll. 107, 237-250 (1998). tislava, Slovak Republic, 119 pp (1994), (in Slovak). 6. M. Chevreuil, M. Blanchard, M.J. Teil, A. Chesterikoff, Wat. Res. 32/4, 1204-1212 (1998). 7. H. Feng, J.K. Cochran, H. Lwiza, B.J. Brownawell, D.J. Hirschberg, Mar. Environ. Res. 45/1, 69-88 (1998).

8. M. Kawano, J. Falandysz, B. Brudnowska, W. Tadaaki, Or- ganochlorine Compounds 39, 331-335 (1998). 9. P.K. Phuong, C.P.N. Son, J-J. Sauvain, J. Tarradellas, Bull. Envi- Received for publication: February 19, 2001 ron. Contam. Toxicol. 60, 347-354 (1998). Accepted for publication: March 15, 2001 10. A. Kocan, B. Drobná, J. Chovancová, J. Petrík, E. Szabová, Expo- sure of population to PCBs and related compounds in risk area of the Michalovce District. Interim report, Project No. 105/98-2.4., Institute of Preventive and Clinical Medicine, Bratislava, Slovak Republic, 113 pp (1998), (in Slovak). CORRESPONDING AUTHOR 11. Summary of Controls on Prioritised Persistent Organic Pollutants. Report on Stage 2, AEA/CS/RCEC/16419001/Issue 1, AEA Tech- Ján Petrík nology Consultancy Services, Harwwell, Oxfordshire, United Institute of Preventive and Clinical Medicine Kingdom, 60 pp (1993). Limbová 14 12. D.G. Patterson, Jr., S.G. Isaacs, L.R. Alexander, W.E. Turner, L. 833 01 Bratislava, Slovak Republic Hampton, J.T. Bernert, L.L. Needham, Method 6: Determination of specific polychlorinated dibenzo-p-dioxins and dibenzofurans in blood and adipose tissue by isotope dilution high-resolution mass E-Mail: [email protected] spectrometry. Environmental carcinogen method of analysis and exposure measurement, p 299-341. U.S.Department of Health and Human Services, Centers for Disease Control, Atlanta (1991). FEB – Vol 10/ No 4/ 2001 – pages 375 - 380

380 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

A NEW METHOD OF MEASURING GASEOUS SEMI-VOLATILE ORGANIC COMPOUNDS (PAHS) BY SOLID PHASE MICRO EXTRACTION (SPME)

Laurent Lassagne*, Véronique Jacob, Pierrick Desuzinges, François Tripoli, P. Kaluzny, Patrick Baussand and Panayotis Foster

GRECA (Groupe de Recherche sur l’Environnement et la Chimie Atmosphérique), Grenoble, FRANCE

Presented at the 10th International Symposium of MESAEP in Alicante, Spain, 2-6 Oct. 1999

SUMMARY

A methodology based on the SPME technique was They are most frequently encountered as the by-products developed for the quantitative measurement in air of incomplete combustion of organic materials, such as of semi- volatile organic compounds (PAHs). The coal, fossil fuels, woods and tobacco products (8, 9, 10). investigated PAHs were the seven ones which can be So they are mainly emitted from fossil-fuel-fired power found in air gaseous form, according to their vapour pres- plants, refuse incinerators, and, more generally, from sures i.e. naphthalene (Nph), fluoranthene (Ft), ace- industrial processes (stationary sources), and diesel auto- naphthene (Ace), fluorene (Fl), phenanthrene (Phe), mobile engines (mobile sources) (11, 12, 13, 14, 15, 16). pyrene (P), and acenaphthylene (Act). A standard gaseous generation based on sublimation equilibrium was developed in order to calibrate the PDMS (polymethyldi silox- According to their vapour pressures, the PAHs are ane) 7µm fibre. The relation between the equilibrium considered differently by American and European constant K and temperature T was determined for each instances. For the American ones, organic compounds are studied compound. Thus, two different behaviours were considered as volatile if their vapour pressures are higher observed: one for naphthalene, which shows a linear rela- than 0.13 Pa: it means that, the PAH volatility limit is tionship, and another for the other PAHs, which show a then situated around the molecules with 16 carbon atoms broker linear relationships. Nevertheless, the determined (approximately the pyren). For the European instances, constants can be used to compute the concentration of any the PAH volatility limit is different (Pv ³ 10 Pa) and only gas phase at any temperature. naphthalene is considered as a volatile compound (17). According to their vapour pressures, PAHs are present in the atmosphere either in gaseous phase or associated with particles (18, 19, 20, 21). In France, the PAHs measure- KEYWORDS: ment must be realised according to the French standard Semi-volatile compounds, PAH, standard generation, SPME, XP X 43-329 (April 1995) (22). However, this procedure equilibrium constant is very long and hazardous to implement mainly because of the numerous risks of product loss and contamination due to the numerous steps and manipulations.

INTRODUCTION In this work, we present a new methodology based on the Solid Phase Micro Extraction, SPME, for the Polycyclic Aromatic Hydrocarbons (PAHs) are measurement of gaseous PAHs. SPME is a relatively extremely dangerous pollutants for health (carcinogenic, recent technique, which is a viable alternative to sample mutagenic, …) (1,2,3,4), even if their atmospheric airborne compounds as well as it has shown to be a viable concentrations are very low (5). They are now widely alternative to extract a multitude of target analytes from a known, and sixteen of them are classified as prioritary myriad of aqueous systems and their corresponding head- pollutants by the US EPA (Environmental Protection spaces (23, 24, 25, 26). Sampling air with SPME provides Agency) (6). They are included in the POP (Persistant a significant advantage in comparison with traditional Organic Pollutant) list (7). methods, such as active sampling and whole air sampling.

381 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

SPME uses no solvent and it provides the analytes The adopted principle for this standard realisation is based with extremely sensitive tool for sampling and analysis on the sublimation equilibrium between the solid PAH (27, 28). SPME is based on an equilibrium process which and the gas phase. The gaseous concentration in the occurs between a polymeric fibre and a matrix (29, 30). closed container is then governed only by the reactor The compounds are absorbed by the fibre and, at equilib- temperature. These gaseous concentrations may be theo- rium, the concentration of an analyte in the fibre coating retically calculated using the thermodynamic properties of is directly proportional to the concentration in the matrix. each compound (31, 32).

Figure 2 shows an example of concentration to In this study, we developed a methodology to calibrate the temperature profiles determined using thermodynamic fibre in order to analyse semi-volatile PAHs in air. properties of each analytes. These curves were used to

determine the concentration of the 7 focussed PAHs in the

reactor at fixed temperatures. EXPERIMENTAL SECTION The SPME fibre was exposed to the standard atmos- SPME device: The SPME fibre coating was 7 µm of pheres in order to determine the equilibrium time and polydimethylsiloxane (PDMS), from Supelco. The fibres calibrate the fibre. were conditioned for at least 3 hours at 300°C before the first use. Fibres are used until loss of reproducibility. Gas chromatography Usually, no more than 50 injections were made with each A Chrompack CP 9000 gas chromatograph was used for fibre. all the analyses. The instrument was equipped with

a split/splitless injector, a flame ionisation detector (FID) Standards: The pure compounds were from Fluka and a Spectra Physic integrator. The column was a Chemika, the purities were 99% for naphthalene (Nph), 15 meters SE54 capillary with an inside diameter of 98% for fluoranthene (Ft), 99% for the acenaphthene 0.32 mm and a film thickness of 0.12 µm. The carrier gas (Ace), 90-95% for the fluorene (Fl) and 97% for the was nitrogen at a pressure of 45 kPa. The detector tem- phenanthrene (Phe), 98% for pyrene (P), 86% for perature was 300 °C. SPME injections were manually acenaphthylene (Act). Standard atmospheres of PAH done through the split/splitless injector during 4 minutes were generated in a 2 litre glass container equipped with at 300°C, i.e. the optimum desorption time and tempera- a septum for the introduction of the SPME syringe ture of the SPME fibre for the studied PAHs. (Figure 1).

FIGURE 1 - Reactor scheme

SPME syringe

Temperature control Light protection

Thermostated bath Solid PAH

Fig.1 Reactor scheme

382 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

FIGURE 2

0,013

0,011

0,009

0,007

0,005

Concentration (g/l) 0,003

0,001

-0,001

-0,003 250 270 290 310 330 350 370 390 410 Temperature (K) Ace Fl Phe

RESULTS AND DISCUSSION

Equilibrium time Where Cfibre is the concentration of the analyte in the Absorption/time profile studies were carried out in fibre and Cair is the concentration of the analyte in the air. order to determine the equilibrium time for each analyte The knowledge of these K values is vital to determine at different temperatures (i.e. concentration) with the unknown concentrations of target analytes in air. This 7 µm PDMS. The equilibrium time was determined by quantification would not otherwise be possible without exposing the SPME fibre to the standard atmospheres extensive external calibration, which is a hazardous during increasing time until the mass of analyte absorbed operation for semi-volatile compounds because of the in the fibre becomes constant. Figure 3 shows the hardness to create standard atmospheres of different representative absorption/time profiles for the tested ana- concentrations at low temperature. lytes for three temperatures: 40°C, 70°C and 80°C. The reactor volume is large enough in order to consider that As the equilibrium is affected by the temperature, the gas phase concentration was constant during the the relation between the K constant and temperature exposition time. Indeed less than 0.07% was absorbed in should be determined in order to know the concentration the fibre during one exposition run. The equilibrium of an unknown sampled and analysed atmosphere at time did not seem to be affected neither by the any temperature. temperature nor by the concentration of the gas phase. Thus, we chose 60 minutes, which is the equilibrium time Figure 4 shows the relationship between log(K) and of the less volatile compound. the inverse of temperature for, respectively, naphthalene and acenaphthylene. We can observed two different behaviours. Only naphthalene gives one linear relation 1 Equilibrium constant Figure 4 (log K = 3.58* - 7;43). The K value provides a measure for the distribution T of an analyte from gas to the PDMS fibre. In air, the For the other PAHs, there is an interuption in the slope K coefficient for an analyte can be describedat fixed tem- wich occurs at approximatly 65°C. At present no perature by the Nernst distribution law, satisfactory explanation could be given for this; it may be

linked to the fiber itself rather than to the PAHs properties. C fibre K = C air

383 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

3500000

3000000

2500000

2000000

1500000

Chromatographic Areas 1000000

500000

0 0 10 20 30 40 50 60 70 80 90 100 Exposition Time (min)

Nph Act Ace Fl Ph Ft P

FIGURE 3

Naphtalène

4,5

4,3 y = 3,581x - 7,4322 2 4,1 R = 0,9747

3,9

3,7

3,5

Log K 3,3

3,1

2,9

2,7 2,5 2,8 2,9 3 3,1 3,2 3,3 3,4

1/T (1/K)(*1000) Acénaphtylène 5,5

y = 3,36x - 6,1426 R2 = 0,9796 4,5

y = 0,8511x + 1,3401 R 2 = 0,2915

log(K) 4

3,5

3 2,7 2,8 2,9 3 3,1 3,2 3,3 3,4 1/T (*1000) FIGURE 4

384 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

CONCLUSION [18] Le Cloirec, P. in Les composés organiques dans SPME with 7 µm PDMS is a suitable method for l’environnement: Lavoisier, Techniques et documents, 1998; the determination of PAHs in the air. Besides, the chapitre 1 et chapitre 3 quantification is possible if the dependence of K on [19] Junge, C. E. Adv. Environ. Sci. Technol. 1977, 8 (1), 7-25 temperature is known. [20] Lane, D. A.; Mc Curvin, D. M. A. A personal computer data The methodology developed in this work for the base for the chemical physical and thermodynamic properties of generation of gaseous standards of semi-volatile and polycyclic aromatic hydrocarbons. In : Polynuclear Aromatic Hydrocarbons: A decade of progress (Cooke, M.; Dennis, A. J.: non-volatile compounds allows to determine the K Eds.), Battelle Press, Colombus, OH, 1987 to temperature relationship. [20] Pupp, C.; Lao, R. C.; Murray, J. J.; Pottie, R. F. Atmos. Envi- SPME provides numerous advantages compared with ron. 1974, 8, 915-925 the traditional sampling techniques (French standard [21] Murray, J. J.; Pottie, R. F.; Pupp, C. Can. J. Chem. 1974, 52, XP X 43-329) and should be further investigated 557-563 considering its simplicity, rapidity, accuracy, low con- [22] AFNOR, XP X 43-329, 1995 tamination risks and low cost. [23] Louch, D.; Motlagh, S.; Pawliszyn, J. Anal. Chem. 1992, 64, 1187-1199

[24] Zhang, Z.; Pawliszyn, J. Anal. Chem. 1993, 65, 1843-1852 REFERENCES [25] Llompart, M.; Li, K.; Fingas, M. J. Chromatogr. A 1998, 824 (1), 53-61 [1] Lee, M. L.; Novotry, M. V.; Bartle, K. D. Analytical Chemistry of Polycyclic Aromatic Compounds, Academic Press : New [26] Chai, M.; Pawliszyn, J. Environ. Sci. Technol. 1995, 29, 693- York 1981, 50-73 701 [2] Sawicki T. Polynuclear Aromatic Hydrocarbons: Mechanism, [27] Zhang, Z.; Yang, M. J.; Pawliszyn, J. Anal. Chem. A 1994, 66, Methods and Metabolism; Batelle Press: Colombus, OH, 1985, 844-853 1-47 [28] Gorecki, T.; Boyd-Boland, A.; Zhang, Z.; Pawliszyn, J. Can. J. [3] Villemin, D.; Cherqaoui, D.; Mesbah, A. J. Chem. Inf. Comput. Chem. 1996, 74, 1297-1308 Sci. 1994, 34, 1288-1293 [29] Saraullo, A.; Martos, P. A.; Pawliszyn, J. Anal. Chem. 1997, 69 [4] Howard, J. B.; Longwell, J. P.; Maar, J. A.; Pope, C. J.; Busby (11), 1992-1998 W. F. Jr.; Lafleur, A. L.; Taghizadeh, K. Combustion and [30] Arthur, C. L.; Potter, D.; Buchholz, K.; Motlagh, S. Pawliszyn Flame 1995, 101, 262-270 LC-GC 1992, 10, 656-661 [5] Menichini, E. The Science of the Total Environment 1992, 116, [31] Physical and thermodynamic properties of pure chemicals, 109-135 Design Institute for Physical Property Data – American Insti- [6] Compendium of methods for the determination of toxic organic tute of Chemical Engineers, part. 5, 1995 compounds in ambient air; Environmental Protection Agency, [32] Stephenson R. M. and Malanowski S., Handbook of the Ther- Washington, DC, 1988, EPA 600/4-89/017 modynamics of Organic Compounds, Elsevier, 1987. [7] Communiqué de presse ECE/EN/97/2

[8] National Academy of Science Particulate Polycyclic Organic

Matter; National Academy Press: Washington, DC, 1973

[9] Bjørseth, A.; Ramdahl, T. in Handbook of Polycyclic Aromatic Hydrocarbons: Emission Sources and Recent Progress in Ana- Received for publication: February 23, 2000 lytical Chemistry; Bjørseth, A.; Ramdahl, T., Eds.; Marcel Accepted for publication: March 15, 2001 Dekker: New York, 1985; 1-20

[10] Mahanama, K. R. R.; Gundel, L. A.; Daicey, J. M. J. Environ. Anal. Chem. 1994, 56, 289-309 CORRESPONDING AUTHOR [11] Kirton, P. J.; Ellis, J.; Crisp, P. T. Fuel 1991, 70, 1383-1389 [12] Dambrine, E. Ciments, Betons, Platres, Chaux 1993, 800, 37- Laurent Lassagne 40 GRECA (Groupe de Recherche sur l’Environnement [13] Menichini, E.; Monfredini, F. Chemosphere 1997, 35 (10), et la Chimie Atmosphérique) 2389-2404 IUT Chimie 39 Boulevard Gambetta, [14] Siebert, P. C.; Alston, D. R.; Jones, K. H. Environmental Pro- 38000 Grenoble - FRANCE gress 1991, 10 (1), 1-12 [15] Benner, B. A.; Gordon, G. E.; Wise, S. A. Environ. Sci. Thech- Fax : +33(0)4 76 85 15 26 nol. 1989, 23 (10), 1269-1278 E-mail : [email protected] [17] Adams, R. E.; James, R. H.; Farr, L. B.; Thomason, M. M.; Miller, H. C.; Johnson, L. D. Environ. Sci. Technol. 1986, 20 (7), 711-716 FEB – Vol 10/ No 4/ 2001 – pages 381 - 385

385 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

OXIDATION OF MANGANESE (II) WITH AIR IN WATER TREATMENT

S. Aydin, N. Tüfekçi and S. Arayici

Istanbul University, Faculty of Engineering, Environmental Engineering Department, Avcilar, ISTANBUL-TURKEY

Presented at the 10th International Symposium of MESAEP in Alicante, Spain, 2-6 Oct. 1999

SUMMARY

Manganese is the second most common constituent The quantity of manganese has great importance for after iron of impounded water and many well waters. drinking waters and process waters. It causes undesirable It causes difficulties in public supplies and in industrial taste and aesthetic problems in drinking waters. In public supplies. In this study, the oxidation of Manganese(II) is supplies, it causes difficulties such as staining of clothes, studied in batch reactors in which the concentrations "black" residues on plumbing fixtures and incrustation of of manganese(IV) was in the range 0-300 mg/L. This mains. In industrial supplies, it causes severe economic study has demonstrated the catalytic effect of MN(IV) on losses through discoloration of products, specks in the Mn(II) oxidation by air to the MN(IV) levels of finished paper, textile, food and beverage products, and 100 mg/L. A quadratic equation has been obtained to reduction of pipeline carrying capacities. Manganese is determine the catalytic reaction rate constant, kcat, as a not known to cause any health problems, and the above function of Mn(IV). conditions are limited to existence of manganese content by 0.05 mg/L in drinking waters, and by 0. 1 mg/L or less in industrial waters.

Methods available for manganese removal from water KEYWORDS: Autocatalytic effect, batch reactor, catalytic effect, supplies can normally be classified as high pH, oxidation Manganese oxidation or combination of the two. Manganese removal methods generally require the use of strong oxidizing agents such as oxygen, potassium permanganate, chlorine, hypochlorite, chlorine dioxide or ozone [2]. The oxidation of manganese by air alone is usually a slow process, unless the pH is raised above neutrality. However, the INTRODUCTION oxidation rates are much higher at elevated pHs than they are at pH values less than 8.0. Aeration as sole means of Manganese is the second most common constituent manganese(II) oxidation is not commonly practiced, after iron found in both surface and ground water, but because the process is kinetically slow and pH dependent. predominantly in the latter. It is a macro nutrient essential The oxidation rate is considered acceptable at pH above to life. Generally, trace amounts of metals as manganese 9.5 [3, 4]. The removal of manganese from aqueous and iron are present in fresh waters from weathering solution by oxidation with air at pH of 9.0 has previously of rocks and soils. Manganese can be present in natural been shown to be autocatalytic in nature [5]. The rate waters in concentrations exceeding 10 mg/L, in truly equation for the conversion of Mn(II) to MN(IV) includes dissolved, colloidal and suspended forms, e.g. Mn(II), an autocatalytic term, in addition to a term proportional Mn(IV). Soluble manganese (Mn(II)) is typically found to the reduced metal ion concentration. However, Pankow in all water supplies, with higher concentrations usually and Morgan [6] also concluded that an auto-oxidative present in water obtained from groundwater sources or reaction does not take place between MN(II) and MnO2 (s) reservoir hypolimnions [1]. at pH values below 9.0.

386 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

The oxidation of manganese (II) with air can be MATERIALS AND METHODS explained by the equations: The oxidation of Mn(II) was studied in a batch reac- k tor of one liter volume under constant pH, temperature Mn(II) + O2 ? MnO2 (s) and O2 concentration. The experimental setup is illus- trated in Fig.1. The reaction vessel was intensely mixed k1 using IKA-WERK Rn 20.n type of mixer. NaHCO3 was Mn(II) + MnO2 (s) + O2 ? 2 MnO2 (s) added into the distilled water in order to obtain a solution with an alkalinity equal to 0.02 M. Air was passed into Since the reaction product is solid, the autocatalysis is the solution using fine bubble diffusers. The pH of the heterogeneous and the exact kinetic analysis of the solution was controlled by adding 0.1 N NaOH and 0.1 N reaction is difficult. Nevertheless, at constant pH and H2SO4 and measured by JENWAY model 3040 type pH under constant partial pressure of oxygen, an autocatalytic meter with sensitivity of ±0.001 pH unit. The dissolved rate expression of the type was shown to be obeyed dur- oxygen levels were monitored using WTW Oxi 538 oxy- ing the initial stages of the oxidation. gen meter. Constant temperature (28 °C) was maintained by immersing the reaction vessel into a water bath. Mn(II) - d[Mn(II)] stock solution was prepared by dissolving MnSO4.H2O = k [Mn(II)]+ k1 [Mn(II)] [Mn(IV)] (1) (manganese sulphate monohydrate) in 1 liter demineral- dt ized water. The samples taken at pre-decided times as measured from the start of the experiments were immedi- kcat = (k + k1 [Mn(IV)]) (2) ately filtered and acidified after filtration with 2 ml HNO3. Filtration through a 0.45 µm membrane filter is an The aim of study is the removal of manganese(II) accepted procedure defined in Standard Methods (3010A) by oxidation with air, and to precipitate as MnO2 (s) for the determination of dissolved manganese. Residual from the waters with high manganese content. manganese(II) concentrations were determined by AAS The oxidation of manganese (II) is studied in batch analysis [7]. The detection limit for the AAS manganese reactors in which the concentrations of manganese (IV) measurement was 0.015 mg/L. All experiments were was in the range 0-300 mg/L. conducted at 28 °C and 9.5 of pH.

FIGURE 1 - The experimental setup

387 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

FIGURE 2 - Variation of log Mn(II) with time different pH values (Mn(II), 10 mg/L; MN(IV), 0 mg/L; temperature, 28 °C; alkalinity, 2x10-2 eq/L; pO2, 0.21 atm)

FIGURE 3 - Variation of log Mn(II) with time different initial Mn(II) concentrations -2 (pH, 9.5; temperature, 28 °C; alkalinity, 2x10 eq/L; PO2, 0.21 atm)

FIGURE 4 - Variation of log Mn(II) with time different initial MN(IV) concentrations -2 (pH, 10.0; temperature, 28 °C; alkalinity, 2x 10 eq/L; PO2, 0.21 atm)

388 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

RESULTS AND DISCUSSION

Homogeneous system The reaction time was reduced to about 12 min when The experimental results with initial Mn(II) 100 mg/L of MN(IV) was added initially into the reactor. concentration of 10 mg/L and with varying pH values Thus, the reaction time is fairly shortened with increasing between 9.0 and 10.0 are given in Fig. 2. These plots initial MN(IV) concentrations. k and kcat values obtained indicate that increasing pH causes decreasing reaction through the experiments in homogeneous and heterogene- time in oxidation of Mn(II) with air. Reaction time in ous systems, respectively, are given in Tables 1 and 2. which the oxidation is completed, t, for pH=9.0 was found to be 318 min. The corresponding t value for pH=9.5, pH=9.6, pH=9.8 and pH=10.0 were found as Comparison of Figs. 3 and 4 indicates that similar 158 min, 96 min, 65 min and 49 min, respectively. catalytic oxidation rates are obtained irrespective of the way of MN(IV) addition. In other words, approximately same catalytic oxidation effect has been obtained, whether The results of the experiments obtained for 5, 10 and MN(IV) was formed as a result of Mn(II) oxidation or 25 mg/L concentrations are given in Fig. 3. No MN(IV) MN(IV) was initially added. The plot of kcat versus initial was initially added in these experiments. Linear MN(IV) concentration is given in Fig. 5. As it can be seen relationship between the Mn(II) concentration and time from this figure, kcat is linearly increasing with on a semi-logarithmic plot indicated the first order increasing initial MN(IV) concentration up to 100 mg/L. kinetics in line with the results given in the literature. This means that there is a catalytic effect of MN(IV) Reproducibility of the results was good in the on the Mn(II) of oxidation at MN(IV) concentrations homogeneous systems. up to 100 mg/L. But there is no catalytic effect of MN(IV) on the Mn(II) of oxidation at MN(IV) concentra- When initial MN(II) concentration was increased, tions after this value. keeping all other experimental conditions constant, the reaction was accelerated due to the MN(IV) formed dur- Applying the curve fitting techniques to the data of ing oxidation. Average homogeneous reaction rate con- Fig. 5., the following equation is obtained between the stant, k, for initial Mn(II) concentration of 5 mg/L was kcat and MN(IV) concentration: found to be 0.009988 min-1. The corresponding k value for [Mn(II)]o = 10 mg/L and [Mn(II)]o = 25 mg/L was found as 0.01264 min-1 and 0.02128 min-1, respectively. kcat = 0.0011 [Mn(IV)]o + 0.0335 (3) And, t for Mn(II)=25 mg/L was found to be 112 min. The corresponding t values for Mn(II)=10 mg/L and

Mn(II)=5 mg/L were 157 min and 170 min, respectively. in which [Mn(IV)]o is in mg/L. Oxygenation rate is terms of kcat and Mn(11) is lt is evident from Fig. 3. that the oxidation is much faster with initial Mn(II) concentrations of 25 mg/L and d[Mn(II)] 10 mg/L than the oxidation rate at initial Mn(II) = - kcat [Mn(II)] (4) concentration of 5 mg/L. The acceleration of the dt oxidation rate at high initial Mn(II) concentrations has been attributed to the formation of substantial amounts of MN(IV) in the reaction medium during oxidation. Assuming MN(IV) concentration is constant, integration of equation 4 yields

e-kat.t [Mn(II)] = [Mn(II)]o (5) Heterogeneous System The results of the experiments with initial Mn(II) in which [Mn(II)] is initial Mn(II) concentration. Or concentration of 10 mg/L and with varying MN(IV) con- by substituting the equation 5 in place of kcat, [Mn(II)] centrations are given in Fig. 4. Catalytic rate constant, concentration as a function of the reaction time is given kcat, was calculated from the slopes of the lines on semi by the following equation: logarithmic plots of MN(II) and time. It is evident from these Figs. that the MN(II) oxidation is in accordance -{0,0011.[Mn(IV)]o+ 0,03351}.t with the first order kinetics. Time needed for the comple- [Mn(II)] = [Mn(II)]o e (6) tion of the reaction was about 49 min when no MN(IV) was present in the reactor.

389 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

TABLE 1 - Average k values in homogeneous systems

pH [Mn(II)]o [Mn(IV)]o k reaction time t* (mg/L) ( mg/L) (min-1) (min)

9.0 10 0 0.00630 318 9.5 10 0 0.01264 158 9.6 10 0 0.02084 96 9.8 10 0 0.03096 65 10.0 10 0 0.04077 49 9.5 5 0 0.00998 170 9.5 10 0 0.01264 158 9.5 25 0 0.02128 112 * completion of oxidation

TABLE 2 - Average kcat values in heterogeneous systems

pH [Mn(II)]o [Mn(IV)]o kcat reaction time t* (mg/L) (mg/L) (min-1) (min)

10.0 10 0 0.04077 49 10.0 10 10 0.04317 45 10.0 10 25 0.06263 32 10.0 10 50 0.06998 29 10.0 10 100 0.14848 12 10.0 10 200 0.07078 28 10.0 10 300 0.06969 29 * completion of oxidation

FIGURE 5 - Change in kcat with [Mn(IV)]o)

1 - kcat, 1 min

390 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

CONCLUSIONS REFERENCES

This study has demonstrated the catalytic effect of 1. Chapman, D., (1992), Water Quality Assessments, First Edi- Mn(IV) on the Mn(II) oxidation by air to the MN(IV) tion, Chapman&Hall, Cambridge levels of 100 mg/L. A quadratic equation has been obtained to determine the catalytic reaction rate constant, 2. Knocke, W. R., Hoehn, R. C., Sinsabaugh, R. L., (1987), Us- kcat, as a function of Mn(IV). For very low MN(IV) ing Alternative Oxidants To Remove Dissolved Manganese concentrations the oxidation rate is dominated by From Waters Laden With Organics, J. AWWA 79, 5, 75-79 the homogeneous reaction whose rate constant, k, -1 has been determined as k=0,0335 min for pH = 10.0, 3. Aydin, S., (1998), Oxidation of Manganese(II) in Contact temperature = 28 °C, alkalinity = 2x10-2 eq/L and Aeration Systems, Master science thesis, Istanbul University PO2 =0.21 atm. 4. Coughlin, W. R., Matsui, I., (1976), Catalytic Oxidation of Aqueous Mn(II), J. Catalysis 41, 108-123

5. Kessick, M. A., Morgan, J. J., (1975), Mechanism of autoxi- ACKNOWLEDGEMENTS dation of manganese in aqueous solution, Env. Sci. Tech. Vol. 9. No. 2., pp. 157-159 This work was supported by the Research Fund of the University of Istanbul (Project number B-283/200899). 6. Pankow, J. F., Morgan, J. J., (1981), Kinetics for the Aquatic Environment, Env. Sci. Tech. Vol. 15. No. 11, pp. 1306- 1313

7. APHA-AWWA-WPCF, (1989), Standard Methods for the Examination of Water and Wastewater, 17th edition, New York

Received for publication: October 11, 1999 Accepted for publication: March 15, 2001

CORRESPONDING AUTHOR

S. Aydin Istanbul University Faculty of Engineering Environmental Engineering Department 34850 Avcilar, Istanbul – TURKEY

Phone: ++90 212 5911921 Fax: ++90 212 5911997 E-Mail: [email protected]

FEB – Vol 10/ No 4/ 2001 – pages 386 - 391

391 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

IMPROVEMENT OF LOW HYDRAULIC PERFORMANCE OF SEA OUTFALLS USING BELLMOUTHED PORTS IN THE BLACK SEA

S. Nemlioglu University of Istanbul, Faculty of Engineering, Department of Environmental Engineering, Avcilar - Istanbul, Turkey

Presented at the 10th International Symposium of MESAEP in Alicante, Spain, 2-6 Oct. 1999

SUMMARY PORT SHAPE CONVERSION METHOD

The operation of marine outfalls is connected with Port shape is added to calculations with diffuser their high hydraulic performance. However, it could be hydraulic discharge coefficient CD. In marine outfall encountered at low hydraulic performance, especially in diffuser, rate of flow from one port q and CD are some old outfall systems, which were designed as sharp expressed as follows [8]: edged ports’ diffuser pipes. When these sharp edged old systems are converted to the bell mouthed ports their hydraulic performance could be increased more than their usual performance. Due to the increase of flow capacity of systems their service life would be higher than their initial one. In this study some sharp edged ports’ outfall systems were designed according to various flow rates, then the same systems were converted to the bell mouthed ports’ systems, and, the differences of flow rates of the systems were studied theoretically. In projecting of diffuser pipes Black Sea parameters were used. The results, which were taken after changing shape of the ports, were where d is diameter of port, g is acceleration due to analyzed according to the increase of outfall systems` gravity, E is total head, n is port number (counting from service lifes and conversion of the sharp edged ports to offshore end), V is mean pipe velocity. the bell mouthed ports in outfall systems were discussed.

It is presumed that the discharge coefficient for KEYWORDS: Marine outfall, sharp edged port, bell mouthed port, diffuser, project life, Black Sea negligible velocity of approach

? 0 is 0.61 for a sharp edged port in a thin pipe wall thickness (SETWT) and INTRODUCTION 0.91 for a bell mmouthed port (BM).

Many of the sea outfall systems with sharp edged ports are used in order to dispose municipal waste water. Thin pipe wall thickness is expressed as t/d < 0.5 Because the sharp edged port has a low hydraulic (t: Pipe wall thickness) [5]. It is seen that port shape can performance, these outfalls have a specialty that their cause big differences in the diffuser hydraulic [1, 7]. In flow capacity could be increased under suitable this paper, it is aimed to take advantage of the property of conditions, when the shape of the port is converted into a increasing rate of diffuser flow Q, by using ports with more efficient one. It has been determined that project life higher CD values. To achieve this, 12 outfalls with of an outfall could be increased up to 17 years through SETWT diffusers had been designed with tp = 30 years theoretically researching of a conversion between two project life, different flow capacity (Q30), port diameter types of sharp edged ports [6]. This case shows that and initial total head (EI) values. Designed diffusers' conversion of outfall from sharp edged ports into bell variation of hydraulic parameters have gained after mouthed ports, which are much more efficient type of the conversion of ports shape into BM type. Data ports, could increase project life of the system. are given in Tables 1 and 2.

392 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

In Table 1, N is the total port number of a diffuser, In population estimation calculations, population EN is the total head of a diffuser, Vmax is maximum increase rate was used as p = 1 with logarithmic increase velocity in diffuser pipe, uav is average velocity in all of method in both before and after the conversion. The diffuser ports. In Table 2, ? is the difference before and values of E1 were accepted because of irrelevant condition after conversion. Parameters, which were used in diffuser of wave height of Black Sea. designation, are given as follows: After the conversion of port shape, EN values had · Diffuser type: increased a little as shown Table 1. While Q30 values Ports opened in horizontal axis of pipe altematively increase, ?Q values increase either. After the conversion, · Port distance: 1 = 3 m percentages of flow increase rates are near each other, as · Darcy-Weisbach's friction factor: f = 0.03 shown Fig. 1. Average value is ?Q (%) = 48.4. Average project life is ?t = 39.9 years. ?t distribution is given in · Slope of diffuser: J = 0.00 p p Fig. 2. V values exceeded 0.60 m/s £ V £ 0.90 m/s · max (total area of ports/cross sectional area of diffuser max limit range. On the other hand, the biggest value is V = pipe) = 0.55 max 1.33 m/s and this value exceeded 0.50 m/s £ V £ 1.25 m/s · Narrow range velocity limits: 0.60 m/s £ V £ 0.90 m/s the limit range very little. Average values are (used for first designation) [3] ?V = 0.40 m/s and ?V .(%) = 48.1. Velocity in · Wide range velocity limits: 0.50 m/s £ V £ 1.25 m/s max max port (u) values were affected by E . After the conversion, (used for after port shape conversion checking) [2] 1 average velocity in port values are u = 3.98 m/s, · Acceleration due to gravity: g = 9.81 rn/s2 av Aua,=1.30 m/s, ?uav %) = 48.3 for E1 = l m and uav = Diffuser flow values were calculated by the way 5.63 m/s, ?uav = 1.83 m/s, ?uav (%) = 48.1 for E1 = 2 m. of project populations (y30) accepted as 100,000, 200,000 Difference of velocity in port distributions are given in and 300,000, and the rate of waste water flow per individ- Fig. 3. Variation of port diameter didn't affect remarkably ual accepted as 250 L/person/day [4]. on diffuser hydraulic parameters, except N.

FIGURE 1 FIGURE 2 Difference of flow rate after the conversion Difference of project life after the conversion

FIGURE 3 Difference of velocity in port after the conversion

393 © by PSP Volume 10 – No 4. 2001 Fresenius Enviromental Bulletin

TABLE 1 - Before and after port shape conversion outfalls' hydraulic parameter values

Port Project No E1 N Sets of data for diffusers with Sets of data for diffusers with dia- population sharp edged ports bell mouthed ports (m) meter y30 d Vmax uav EN Q tp Vmax uav EN Q tp project flow (m) (m/s) (m/s) (m) (L/s) (year) (m/s) (m/s) (m) (L/s) (year) Q30 (L/s) 100,000 1 1 14 0.85 2.68 1.066 295.0 31 1.25 3.98 1.143 437.7 71 289.4 2 2 10 0.85 3.80 2.042 298.7 33 1.26 5.64 2.092 443.1 72 0.1 200,00 3 1 28 0.89 2.71 1.109 595.8 32 1.33 4.06 1.240 894.0 73 578.7 4 2 20 0.86 3.82 2.070 599.1 33 1.27 5.68 2.155 892.2 73 300,00 5 1 41 0.89 2.73 1.139 880.0 31 1.33 4.14 1.308 1332.9 73 868.1 6 2 29 0.88 3.82 2.092 870.8 30 1.32 5.71 2.202 1300.7 70 100,00 7 1 4 0.80 2.67 1.009 335.0 44 1.17 3.92 1.020 492.8 83 289.4 8 2 3 0.61 3.80 2.006 358.5 51 0.91 5.63 2.013 531.3 91 0.2 200,00 9 1 7 0.84 2.65 1.018 584.3 30 1.23 3.89 1.038 856.4 69 578.7 10 2 5 0.80 3.79 2.011 596.0 32 1.19 5.52 2.025 881.3 72 300,00 11 1 11 0.84 2.66 1.027 917.5 35 1.23 3.89 1.058 1344.0 73 868.1 12 2 8 0.80 3.79 2.018 952.4 39 1.19 5.61 2.039 1407.0 78

TABLE 2 - Data of hydraulic parameter increase

No ?Vmax ?Vmax ?uav ?uav ?EN ?EN ?Q ?Q ?tp ?tp (m/s) (%) (m/s) (%) (m) (%) (L/s) (%) (year) (%) 1 0.40 47.1 1.30 48.5 0.077 7.2 142.7 48.4 40 129 2 0.41 48.2 1.84 48.4 0.050 2.4 144.4 48.3 39 118 3 0.44 49.4 1.35 49.8 0.131 11.8 298.2 50.1 41 128 4 0.41 47.7 1.86 48.7 0.085 4.1 293.1 48.9 40 121 5 0.44 49.4 1.41 51.6 0.169 14.8 452.9 51.5 42 135 6 0.44 50.0 1.89 49.5 0.110 5.3 429.9 49.4 40 133 7 0.37 46.3 1.25 46.8 0.011 1.1 157.8 47.1 39 89 8 0.30 49.2 1.83 48.2 0.007 0.3 172.8 48.2 40 78 9 0.39 46.4 1.24 46.8 0.020 2.0 272.1 46.6 39 130 10 0.39 48.8 1.73 45.6 0.014 0.7 285.3 47.9 42 131 11 0.39 46.4 1.23 46.2 0.031 3.0 426.9 46.5 38 108 12 0.39 48.8 1.82 48.0 0.021 1.0 454.6 47.7 39 100

CONCLUSIONS REFERENCES

As a result of converting port shape theoretically 1. Berkün, M. and Nemlioglu, S., Effects of Internal Flow Dif- from the sharp edged port in pipe with thin wall, diffuser fuser Hydraulic Parameters and Characteristics on Outfall Design, TMMOB Civil Engineering Xlllrd Technical Con- flow, for that reason, project life, and velocity in port are gress, 499-512, (in Turkish), Ankara, Turkey, 1995. considerably increased. On the other hand, total hydraulic head is increased a little and velocity in the diffuser pipe 2. Danish Isotope Centre, Diffuser Design, Lecture Submitted at exceeded a little the limit values. Increasing the capacity the WHO Training Course on Coastal Pollution Control, In- of diffuser flow in large amounts extends the project life ternal Report No. 24, Copenhagen, Denmark, 1971. of outfall system about 40 years after the conversion, if the environmental criteria are acquired. Because of in- 3. Grace, R. A., Marine Outfall Systems, 1st Edn., Prentice-Hall creasing the total head a little, in many cases, the project Incorp., Englewood Cliffs, New Jersey, USA, 1978. life needed no supplementary energy sources to increase the project life. As the data of Black Sea's unsuitable 4. Martz, G., Siedlungswasserbau Teil 2, Kanalisation, WIT 18, wave height are taken into account in this paper, this case 1980. shows that this method can be applied to the seas which have data similar to that of the Black Sea. Besides, 5. Miller, D. S., Internal Flow, BHRA, Cranfield, Bedford, UK, beeause the conversion increases velocity values in ports 1971. in lange amounts, it will also increase the initial dilution in great amounts. As the velocity values in pipe exceed 6. Nemlioglu, S., Improvement Hydraulic Performance of Ma- the wide range velocity limits even a little, it is very im- rine Outfall with Sharp Edged Ports by the Method of In- portant to obey the narrow range velocity limits. As using creasing Pipe Wall Thickness, TMMOB 1st National Shore port shape conversion method in the outfalls, which are Engineering Symposium, 203-206, (in Turkish), Samsun, Turkey, 1996 a. designed by selecting the maximum velocity in pipe, in high value increases the velocity value very much, this method might not be suitable in the outfalls like this. 7. Nemlioglu, S., Effects of Internal Flow Diffuser Hydraulic Parameters and Characteristics on Outfall Design, Master of Science Thesis, Black Sea Technical University, Institute of In existing outfall, because of the difficulty to form Science, (in Turkish), Trabzon, Turkey, 1996 b. bell mouthed port in the bottom of the sea, changing the diffuser pipe of the outfall and keeping the main pipe line 8. Rawn, A. M., Bowerman, F. R. and Brooks, N. H., Diffusers in its place is the only possibility be applied. Changes that for Disposal of Sewage in Sea Water, Sanitary Engineering have to be carried out on the outfall should also be dura- Div. ASCE 86 (SA2), 65-105, 1960. ble. For that reason, it is necessary to control whether the pipe is durable/suitable enough to make the conversion or not.

In conclusion, the port shape conversion method increases the project life without affecting much the original Received for publication: October 11, 1999 flow capacity and the port diameter and could be used Accepted for publication: March 15, 2001 when the diffuser in marine outfall is designed.

CORRESPONDING AUTHOR

S. Nemlioglu University of Istanbul ACKNOWLEDGMENTS Faculty of Engineering Department of Environmental Engineering This work was supported by the Research Fund of the 34850 Avcilar - Istanbul, TURKEY University of Istanbul (Project number: B-287/200899). FEB – Vol 10/ No 4/ 2001 – pages 392 - 395

DENITRIFICATION OF WASTEWATER IN A FIXED FILM REACTOR

H. Timur

TUBITAK MRC Energy Systems and Environmental Research Institute, Gebze-Kocaeli-TURKEY

Presented at the 10th International Symposium of MESAEP in Alicante, Spain, 2-6 Oct. 1999

SUMMARY

Denitrification of high nitrate containing synthetic Any of the last three substances can be released as a wastewater was tested in a bench-scale upflow fixed bed gaseous end product, of which N2 form is preferred. It is filter. Molasses was used as carbon source. Maximum known that the type of microorganisms, the pH, and denitrification rates were achieved at COD/NO3-N ratios COD/NO3-N ratio are important factors influencing the about 5-6 independent of hydraulic loading rate and influ- type of end products (1). Cells' transition between ent NO3-N concentrations. Nitrate and COD surface aerobic and anoxic conditions enhance the formation removal rates increased with increasing surface and of these intermediates (2). However, N2 appears to be hydraulic loading. Results suggested that the quantity of the major product formed by mixed cultures used in COD used for nitrate removal is independent of the load- wastewater treatment. ing rate up to a certain limit. Nitrate removal efficiencies varied in the range of 71-99% depending on the operating Biological denitrification may occur by both hetero- conditions. Denitrification followed half-order kinetics. trophic and autotrophic bacteria. Heterotrophic denitrifi- Calculated half order reaction rate constants were in the cation is very efficient in nitrate removal provided that range of 0.013-0.034 g1/2 m-1/2hr-l. adequate amounts of electron donor are available. In some situations organic carbon may be insufficient and supplied as external carbon source. In literature organic materials such as methanol, acetic acid, glucose, glycerol and methane or inorganic materials such as hydrogen and KEYWORDS: sulfur have been used as electron donor. Denitrification, fixed-bed filter, kinetics, molasses, wastewater To minimize the quantity of electron donor used and amount of sludge produced, electron donor should be one with a low yield. Since methanol is a low yield single- carbon compound it has been widely used as an external electron donor (3). For practical use internal carbon INTRODUCTION (raw sewage carbon and endogenous carbon) is the second most important carbon source. If carbon rich A large number of chemical industrial wastewaters industrial wastes are available they might be used contain high amounts of nitrogenous compounds. as carbon sources. Discharges containing nitrogen may stimulate the growth of algae and rooted aquatic plants and may interfere with Studies for post-denitrification of wastewater beneficial uses of water resources. Industrial wastewaters treatment plant discharge using methanol (4) or methanol mainly contain nitrogen in the form of ammonia, nitrate and ethanol (5) as external carbon source have been and nitrite. Nitrification is used to convert ammonia to reported. Tam et al. (6) studied the effects of exogeneous nitrate, which does not reduce the mass of nitrogen to be carbon sources of acetate, methanol and glucose on discharged. Biological denitrification is an attractive denitrification and found that acetate was the most option, because nitrate is efficiently removed by efficient and effective source. Canziani and Bonomo (7) conversion to nitrogen gas. This conversion is proceeded used acetate as electron donor for post-denitrification of in this sequence: textile effluent and reported over 90% NO3-N removal. Andreasen et al. (8) used the hydrolysis product of - - NO3 à NO2 à NO à N2O à N2 primary and activated sludges as electron donor.

Flere and Zhang (9) used sulfur and limestone Daily operation and sampling period autotrophic denitrification process for denitrification of The filter was sampled twice a day and seven days a nitrate-contaminated surface water in a pond. Electron week. The parameters that were measured were NO3, NO2 donor requirements are usually expressed in terms and COD of influent and effluent. COD analyses were of COD/NO3-N ratio. This ratio is dependent upon done by the potassium dichromate ferrous ammonium the electron donor used and specific design. For optimum sulfate method according to AWWA (11). Nitrate and denitrification, optimum temperature is 40 °C and nitrite analyses were done colorimetrically using a soluble oxygen concentration should not exceed 1-2 mg/1 two-channel Technicon AA2 Model Auto Analyzer. for fixed-bed reactor and 0.5 mg/1 for suspended The methods of analyzing were similar to those given in bed reactor (10). Standard Methods (11) and by Strickland and Parsons (12). The influent flow rate was adjusted in order to main- In the framework of this study the following have been tain the desired hydraulic retention time (HRT). The given priority for evaluation of denitrification process in results are the averages which were obtained once steady fixed film system: state regime was achieved.

· Determination of the optimum COD/N ratio · Monitoring the system performance at various organic and hydraulic loading conditions · The use of molasses as carbon source (ed) for denitrifiction processes and investigation of the RESULTS AND DISCUSSION reaction kinetics In the first phase of the study at certain hydraulic retention times (HRT) various strength of feed solution was introduced into the filter to evaluate the optimum MATERIALS AND METHODS COD/NO3-N ratio. The results indicated that maximum denitrification rates were achieved at COD/NO3-N ratios Fixed bed reactor of about 5-6 independent of hydraulic loading rates and influent NO -N concentrations (Fig. 1). COD removal In experimental studies an upflow fixed-bed reactor made 3 rates increased with increasing COD/NO -N ratios. from cylindrical Plexiglas column with a height of 75 cm, 3 and an internal diameter of 19 cm was used. Total liquid volume was 17 litres. A gas liquid separation device was To examine the relationship between COD/N ratio provided to the top of the column. The filter was filled and relative COD removal rate, rCOD/rN was plotted with l-shaped plastic Pall rings giving 3.6 m2 of total vs influent COD/N ratio. As seen in Fig. 2, COD removal surface area. The void volume of the filter was 15.4 litres with respect to nitrogen removal increased with with a corresponding void fraction of 90%. increasing COD/N ratio, indicating that at high COD/N ratios COD removal processes become dominant and Feed solution the filter works as COD removing unit. It is reported in During the experimental studies potassium nitrate was literature (13) that when methanol, spent sulphite liquor, used as nitrate source and molasses as carbon and energy yeast, and whey were used as carbon sources the optimum source. Synthetic feed solutions were prepared with tap COD/NO3-N ratios for maximum denitrification were water adding appropriate amounts of technical grade 2.5/1, 2.9/1, 2.8/1 and 9.65/1, respectively. Both Figs. show that optimum COD/NO -N ratio is strongly KNO3 and NaH2PO4 (0.013 g P/g NO3-N) and molasses. 3 No alkalinity was supplied throughout the study. dependent on the substrate used as organic carbon source.

Environmental conditions In the second phase of the study the HRT ranges of The filter was operated at room temperature and under 2-11.2 h and influent NO3-N concentrations of 47-308 mg -1 anoxic conditions. The filter was covered with NO3-N 1 were selected as the main operational aluminum folio to avoid algae growth in the reactor. parameters. COD/N ratio has been maintained between 4.6-6.8 throughout this phase. The influence of hydraulic Start-up procedure loading rate (Q/A) on nitrate removal rate is shown in The filter was seeded with activated sludge taken from a Fig. 3 for three ranges (47-65 mg/l, 102104 mg/l, and domestic wastewater treatment plant. No feeding was 273-308 mg/l) of influent NO3-N concentrations. introduced, but 20 1 d-1 of recycling was maintained until the gas production, then batch feeding was started by Nitrate removal rate increased with increasing hydraulic adding diluted feed solution. The start-up was completed loading rate and this effect was more evident at high at the end of the 30 day. influent nitrate concentrations (Fig. 3).

FIGURE 1 - Effect of COD/NO3-N ratio on nitrate and COD removal rate -1: HRT: 3.4-3.9 h; Inf. NO3-N: 39-56 mg/l. -2: HRT: 3.4-3.9 h; Inf. NO3-N: 11 0- 1 15 mg/l. -3: HRT: 2.2-2.5 h; Inf. NO3-N: 47-63 mg/l

FIGURE 2 - Effect of COD/N ratio on relative removal rates

COD/N

FIGURE 3 - Effect of hydraulic loading rate on nitrate removal

Hydraulic loading rate (l m-2 d-1)

FIGURE 4 - COD removal versus NO3-N removal

FIGURE 5 - Calculation of half-order reaction rate constant

NO3-N removal rate (g m-2 d-1)

REFERENCES

The relationship between nitrate and COD removal is 1. Hanaki, K., Hong, Z., and Matsuo, T. (1992). Production of shown in Fig. 4. The consumption of molasses (expressed nitrous oxide gas during denitrification of wastewater. Wat. Sci. Tech. 26, No. 5-6, pp. 1027-1036. in terms of COD) increased linearly with nitrate removal -2 -1 2. Otte, S., Grobben, N. G., Robertson, L. A., Jetten, M. S. M., up to 7-8 g NO3-N m d . The slope of 4.06 represents and Kuenen, J. G. (1996). Nitrous oxide production by Alca- the average observed stoichiometric ratio, which is com- ligenes faecalis under transient and dynamic aerobic and an- parable with the value of 3.79 found by Canziani and aerobic conditions. Appl. Env. Microb. 62, pp. 2421-2426. Bonomo (7), who used sodium acetate as carbon source in 3. Environmental Protection Agency (EPA) (1975). Process de- a sand bed filter. sign manual for nitrogen control. Office of technology transfer, Washington, D.C. Data from lab-scale tests have been worked out in or- 4. Balley,W., Tesfaye, A., Dakita, J., McGrath, M., Daigger, G., Benjamin, A., and Sadick, T. (1998). Wat. Sci. Tech. 38, No. der to check whether the overall process kinetic was zero 1, pp. 79-86. order (complete penetration) or 1/2 order (partial penetra- 5. Aspegren, H., Nyberg, U., Andersson, B., Gotthardsson, tion). Half order reactions proceed at a rate proportional S.,and Jansen, C. (1998). Post denitrification in a moving bed to the square root of substrate concentration: biofilm reactor process. Wat. Sci. Tech. 38, No. 1, pp. 31- 38. 6. Tam, N. F. Y., Wong, Y. S., and Leung, G. (1992). Effect of 1/2 exogenous carbon sources on removal of inorganic nutrient ra = k1/2a S by the nitrification-denitrification process. Wat. Res. 26 (9), or pp. 1229-1236. 7. Canziani, R., and Bonomo, L.(1998). Biological denitrifica- ln ra = 0.5 ln S + ln kl/2a tion of a textile effluent in a dynamic sand filter. Wat. Sci. Tech. 38, No. 1, pp. 123-132. -2 -1 -3 where: ra = reaction rate; g m d , S = g m ; 8. Andreasen, K., Petersen, G., Thomsen, H., and Strube, 1/2 -1/2 -1 k1/2 = half order reaction rate constant, g m d R.(1997). Reduction of nutrient emission by sludge hydrolysis. Wat. Sci. Tech. 35, No. 10, pp.79-85. 9. Flere, J. M. and Zhang, T. C. (1998). Sulfur-based autotro- The nitrogen removal rate vs bulk NO3-N phic denitrification pond systems for in-situ remediation of concentration is given in Fig. 5 (ln-ln plot). Straight lines nitrate-contaminated surface water. Wat. Sci. Tech. 38, No. with slope of 0.5 give an intercept of ln k1/2. It can be seen 1, pp. 15 -22. from Fig. 5 that in the concentration range of this study 10. Henze, M., and Harremoes, P. 1977). Biological denitrifica- only half order reaction is observed. The calculated tion of sewage. A literature review. Prog. Wat. Tech. 8, Pergamon Press. half order reaction rate constant was in the range of 1/2 -1/2 -1 1/2 -1/2 -1 11. AWWA (1989). Standard Methods for the Examination of 0.013 - 0.034 g m hr (2.96-8.06 mg 1 hr or Water and Wastewater, 17thed. 0.049-0.134 mg1/2 1-1/2 min-1). In literature the half order 12. Strickland, J. D. H. and Parsons, T. R. (1972). A practical reaction constant in a down-flow filter with methanol as handbook of seawater analysis, 2nd ed., Bull. Fish. Res. Board carbon source and crushed granite of 2.5 mm as inert Can. 167, pp. 1/2 -1/2 -1 support was reported as 0.35-0.77 mg l min (14), 13. Skrinde,J. R. and Bhagat, S. K. (1982). Industrial wastes as and 0.0135 g1/2 m-1/2 hr-1 in a CSTR with hydroalcoholic carbon sources in biological denitrification. Jour. WPCF 54. mixture as carbon source (7). 14. Harremoes, P. and, Riemer, M. (1975). Report on Pilot Scale Experiment in downflow filter denitrification. Denmark Tech. Univ. Environ. Eng. Dept.

CONCLUSIONS Received for publication: October 11, 1999 Approximately one month was required to develop an Accepted for publication: March 15, 2001 optimum level of denitrifying population within the column. This corresponds to literature data when methanol was used as carbon source. Through the comparison of CORRESPONDING AUTHOR this study with literature it can be easily said that molasses is a good carbon source for denitrification process. In H. Timur laboratory scale studies we faced the abnormal scaling TUBITAK MRC problem with molasses, which from time to time caused Energy Systems and Environmental Research clogging in the gas liquid separator. In large scale applica- Institute tion this effect should be considered and necessary ar- P. O. Box 21 rangements should be made to overcome this problem. 41470 Gebze-Kocaeli-TURKEY

FEB – Vol 10/ No4/ 2001 – pages 396 -400

PESTICIDE RESIDUES IN GROUND AND SURFACE WATER IN BULGARIA

Z. Bratanova and K. Vassilev

National Centre of Hygiene, Sofia, Bulgaria

SUMMARY

Data on the occurrence of pesticide residues in 17 ground water sources (43 samples) providing drinking 68 surface and ground water sources are presented. water supply for 600 000 persons. Forty eight non- Residues of 15 pesticides have been found in 19 % of the drinking water bodies were examined (120 samples) in- examined 176 water samples. The most positive findings cluding 6 of the biggest Bulgarian rivers (river Iskar, were related to atrazine, 13 % and lindane, 10%. The Ogosta, Jantra, Vit, Maritza and Struma), as well as the contamination levels ranged from 0,01 to 0,1 µg/l for Bulgarian part of river Danube, lakes, irrigation dams, atrazine and from 0,01 to 0,06 µg/l for lindane. The drilling wells etc. One hundred and fifty six samples were 5 cases of incidental contamination with high levels of examined for organochlorine pesticides (HCH – a, ß, ? pesticides, due to unregulated storage and disposal of isomers, aldrin, alachlor, dieldrin, pp-DDT, op-DDT, pesticides are discussed. pp-DDD, endrin, heptachlor, metolachlor, hexachloroben- zene, heptachlor epoxide), 66 - for trifluralin, 69 - for triazines (atrazine, propazine, simazine, prometryne, terbutryn), 42 - for phosphororganic pesticides (feni- KEYWORDS: Pesticide residues, water, Bulgaria trothion, diazinon, methyl parathion, chlorpyrifos, tetra- chlorvinphos, dimethoate, dichlorvos, trichlorfon), 22 - for pyrethroids (cyhalothrin, cypermethrin, perme- thrin, fenvalerate), 20 - for 2,4-D. Gas chromatography INTRODUCTION with electron capture detector (GC/ECD) and nitrogen phosphorus detector (EC/NPD), Gas chromatography The “Integrated Program on Environment and Health mass spectrometry (GC/MS) and High performance in Bulgaria”, a project supported by WHO, Bilthoven liquid chromatography (HPLC) were applied. The division, was completed in 1997. The main object of this limits of sensitivity of analytical methods vary project was to collect information on the occurrence of from 0,001 to 0,01 µg /l. pesticide residues in surface and ground water in the country. This paper presents the compiled data obtained in this project as well as more recent data collected up to now. The data are collected from our own sources (National Centre of Hygiene) and have not been published RESULTS AND DISCUSSION previously in journals. Presence of one or more pesticides was established in 19 % of the samples (176/34). The number of positive MATERIALS AND METHODS samples is relatively greater among non-drinking waters (21 %) in comparison with contamination of drinking The present investigation includes data for pesticides waters (16 %). No positive samples in surface drinking in 68 water bodies in Bulgaria (rivers, wells, irrigation water sources were found (13/0). The content of pesti- waters and dams for drinking water). One to four samples cides in positive samples originating from ground sources were examined for each water body (a total of 176 sam- of drinking water did not exceed 0,1 µg /l. The satisfac- ples for the period 1993-1999). Twenty drinking water tory water quality from the three dams intended for drink- sources were examined (56 samples) including 3 dams for ing water supply can be explained on the fact that their drinking water supply for a population of 1,8 million water catchment areas include highland territories with people (21 % of the total Bulgarian population) and small antropogenic influence.

On the other hand, there are principally greater possi- Lindane was detected in 15 samples from different bilities for biotic and abiotic degradation of the pollutants non-drinking water bodies in a concentration range of in surface waters compared to ground waters. The exist- 0,01-0,06 µg/l (Table 1). The presence of lindane in 10 % ing legal procedures for strong preliminary control at of the samples is relatively unexpected considering the approval of new drinking water sources, as well as limita- prohibition for sale and use of this insecticide in Bulgaria tions and/or prohibitions for pesticide usage on the terri- since 1991. Probably, this finding is due to nonpoint tory around drinking water sources ensure relatively pollution of the particular water bodies. Some other au- lower extent of contamination of waters intended for thors found lindane in surface waters in countries from human consumption. Eastern Europe – Slovak Republic (9), Croatia, Slovenia, Romania, Moldova, and Ukraine (10). The percentage rate of positive samples in ground and surface non-drinking waters is almost equal (21 and Lindane belongs to the group of so-called global or- 22%). The relatively higher number of positive samples ganochlorine pollutants that circulate for years, even (predominantly rivers) is probably due to the lack of spe- decades, through the environment. From the persistent cial protective measures for surface waters. Still water organochlorine compounds lindane has the poorest ability bodies, especially rivers sometimes, are used for non- for adsorption and the highest volatility, that define its regulated deposition of pesticide residues, washing of greater migration rate in the environment – by evapora- agricultural equipment etc. The data of the analysis are tion and secondary deposition by precipitation in the presented in Table 1. surface waters.

The results are divided into 2 groups – one from As for the classic representative of organochlorine different monitoring programmes and the other one - insecticides – DDT and its metabolites, the present work found at incidents. It is clear (Table 1) that atrazine (69/9) shows clear trend for decreasing of these substances in and lindane (156/15) were the most often found pesticides the hydrosphere of Bulgaria. During the 70’s DDT and at monitoring. Atrazine showed the highest rate of its degradation products were found regularly in positive samples (13 %). Triazines were detected in amounts between 0,023-0,410 µg/l (12), during the 80’s, ground drinking water sources of some of the biggest 0,013-0,150 µg/l; at present DDT is found only towns in the country – such as Plovdiv, Pleven etc., at incidental point pollution, and DDE – in rare cases although the detected levels did not exceed the value of (176/2) in non-drinking waters (Table 1). That remark- 0,1 µg/l (1), recommended by the European directive. able reduction of the residues for the most persistent Symtriazines, though are not highly persistent in the insecticide is an expectable consequence from 30 years environment, are often detected in ground waters. prohibited use of DDT in Bulgaria (banned in 1969). Degradation time (DT 50) for atrazine in soil is usually 1-2 months, but under ground water conditions DT 50 is over 200 days (2). Other pesticides belonging to different chemical groups – organophosphorous pesticides (dichlorvos, Some other authors (3-5) reported similar data for trichlorfon), 2,4-D, metolachlor and pyrethroids presence of higher concentrations of triazines in ground (fenvalerate), were found in single samples most often in water. Riparbelli, C. et al. (6) found 11 500 positive sam- the case of incidents. As a rule, relatively higher amounts ples (> 0,1 µg/l) from a total of 113 000 samples, and of pesticides, including banned ones, were found in the registered most often atrazine (up to 50 µg/L) and benta- case of incidents. The table presents the results of 30 zone. A. Grohman et al. (7) found that out of 19 samples samples taken in 5 cases of incidental pollution of water of drinking water contaminated with pesticides above bodies with high amounts of pesticides. 0,1 µg/l on the territory of the former East Germany, 10 contained atrazine. The analysis of the incidents shows that usually A. D. Carter (8) found that in Great Britain 96 % of non-regulated deposition of the pesticide residues took contaminated drinking water samples (> 0,1 µg/l) place – deposition in deserted wells or landfalls, de- contained 5 herbicides: atrazine, simazine, diuron, stroyed or incinerated pesticide stores, or criminal isoproturon and mecoprop. Obviously, because of the deposition of banned formulations or such with expired very long period of usage of triazines (atrazine was in- term. According to our earlier review (13) only in volved in 1957) and their long degradation time in ground Northern Bulgaria there are 25 stores containing water, the persistence of triazines is a real problem for banned pesticides and non-labeled substances, thus hydrosphere nowadays. presenting a constant risk for the environment.

TABLE 1 Concentrations of pesticides found in Bulgarian waters

Pesticides found in monitoring Pesticide found in case of programmes accidents LOD Pesticide N° samples / Concentration Range/ N° sam- Concentration Detection limit N° positives Mean Value, ples Range (mg/l) / Method (mg/l) (mg/l) Lindane 156/15 0.01-0.06 / 0.03 4 0.01-0.14 0.001/ GC/ECD 0.01/GC-MS DDT 156/0 - 7 0.06-1.63 0.001/ GC/ECD 0.01/GC-MS DDE 156/2 0.001-0.003 7 0.13-0.57 0.001/ GC/ECD 0.01/GC-MS Heptachlor 156/2 0.001-0.004 - - 0.001/ GC/ECD 0.01/GC-MS Hexachloro- 156/4 0.01-0.04 / 0.02 - - 0.001/ GC/ECD benzene 0.01/GC-MS Trifluralin 66/0 - 2 700-1000 0.01 / GC/ECD

2,4-D 20/1 0.01 - - 0.01 / GC/ECD 0.01 / HPLC Atrazine 69/9 0.01 – 0.1 / 0.03 3 0.9-2000 0.01/ GC/NPD 0.01 HPLC Simazine 69/1 0.02 3 0.1-1120 0.01/ GC/NPD 0.01 HPLC Propazine 69/0 - 3 4.8-115 0.01/ GC/NPD 0.01 HPLC Metolachlor 42/4 0.08-6.4 / 2.3 - - 0.001/ GC/ECD 0.01/GC-MS Dimethoate 42/0 - 1 8.2 0.05 / GC/NPD

Dichlorvos 42/1 6.2 - - 0.05 / GC/NPD

Trichlorfon 42/2 0.4-1.8 - - 0.05 / GC/NPD

Fenvalerate 22/1 9.0 - - 0.05 / GC/ECD

CONCLUSIONS REFERENCES

The data on the presence of pesticides in Bulgarian [1] Council Directive 98/83/EC on the quality of water intended for waters after 1990 were available only to a limited extent. human consumption. Due to the comparatively high cost of pesticide analyses the monitoring has not been maintained on a regular basis. With [2] The Pesticide Manual, Tenth edition, BCPC, 1994, 51-52. a few exceptions the monitoring programs summarized here were carried out on random basis in case of accidents or [3] Albanis, T. A., Hella, D. G., Pesticide concentrations in Louros upon request. However, we believe that in spite of its limita- River and their Fluxes into the Marine Environment., Interna- tions the study provides useful evidence on the most fre- tional Journal of Environmental Analytical Chemistry, 1998. quently occurring types of pesticides and contamination levels in the Bulgarian waters during the last years. [4] Pionke, H. B., Glofelty, D. E., Lucas; A. D., Urban, J. B., Pesticide Contamination of Groundwater in the Mahantango Creek Watershed, Pennsylvania, USA, J. Environ. Qual., 17, 1988, 76-84.

[5] Dellavedova, P., Sesana, G.,Ferre, P., Bersani M., Riparbelli, C.,Maroni, M., Atrazine groundwater contamination in an in- tensive agricultural area west of Milan – Italy., Period 1986- 1994. Proseedings of the X Symposium Pesticide Chemistry – Piacenza, 1996, 669-673.

[6] Riparbelli, C., Scalvini, C.,Bersani, M., Auteri, D., Azimonti, G., Maroni, M., Salamana, M., Carreri, V., Groundwater contamination from herbicides in the region of Lombardy – Italy., Period 1983 – 1993. Proceedings of the X Symposium Pesticide Chemistry – Piacenza, 1996, 559-563.

[7] Von Grohman, A., Winter, W., Ottenwalder, H., Mogliche Beeintrachtigungen des Trinkwassers in den neuen Landern durch Pflanzenschutzmittel., Bundesgesundhbl., 12, 1994, 496- 502.

[8] Carter, A.D., Pesticides in soil and water., Pesicide Outlook., 4, 1993, 23-28.

[9] Bratanova, Z., Kovacicova, J., Gopina, G., A review of the occurrence of pesticides in water of the river Danube and its

tributaries. Fresnius Env. Bull., 7, 1998, 495-501.

[10] Veningerova, M., Pracha, V., Kovacicova, J., Uhnak, J., Ana- lytical methods for the determination of organochlorine compounds. Application to environmental samples in the Slovak Republic. Journal of Chromatography A. 774, 1997, 333-347.

[11] Gitzova, S., Water pollution by organochlorine pesticides

residues and methods for their determination, Dissertation, 1976, Sofia.

CORRESPONDING AUTHOR [12] PHARE PROJECT 95010000, Environmental Proggrrame for the Danube River Basin, Danube Regional Pesticide Study, 1997. Z. Bratanova/ K. Vassilev National Centre of Hygiene 15, D. Nestorov 1431 Sofia- BULGARIA Received for publication: July 17, 2000 Accepted for publication: March 15, 2001

FEB – Vol 10/ No 4/ 2001 – pages 401 - 404

INVESTIGATION OF HEAVY METAL POLLUTION OF TRAFFIC IN KEMALPASA-TURKEY

Hakan Arslan

Department of Chemistry, Faculty of Arts and Sciences, Mersin University, Mersin, Turkey

SUMMARY

Trace metal concentrations of the soil samples Kemalpasa is located in west Anatolia. It is 15 km away of Kemalpasa-Turkey region traffic were investigated from Izmir. It has a surface area of 190 km2 and had to determine Pb, Ni, Cd, Mn, and Zn content by flame a population of 43400 in 1995. The city which is famous atomic absorption spectrometer. The concentration of Pb, for its cherry and olive gardens, is the second largest Ni, Cd, Mn, and Zn were found at all places in the range industrial region of Turkey with 43 food and 167 of 13-375, 30-145, 1.1-5.8, 218-348, and 27-108 mg/g, industrial factories. respectively. A good correlation was found between There are no publications so far about the determina- the number of cars or total number of vehicle and tion of trace elements originating from traffic pollution of the metal contents. Kemalpasa-Turkey. In this study, the level of the heavy metals in soil samples collected from ten stations in Ke- malpasa-Turkey were determined by flame atomic absorption spectrometry (FAAS). The relationship between metal contents and traffic intensity was calculated. KEYWORDS: Pollution, Trace Metals, Soil, Traffic

EXPERIMENTAL

Sampling: A total of nine street dust samples were col- INTRODUCTION lected from low and high density traffic roads in Kemal- pasa during the period of April to June, 2000. Sampling The trace element contents of soil samples have been was performed according to detailed guidelines men- investigated for various purposes such as agricultural tioned by Somer and Aydin [1], which consisted of soil studies and different pollution studies [1-4]. Soil samples dust collected from both sides of a 10-15 m length of the are among important materials to be tested during investi- road. They were dried at 60 oC and sieved through a 170 gations of environmental pollution originating from traf- mesh sieve. Samples were then taken by the coning and fic. Traffic is one of the sources of emission of toxic quartering methods. heavy metals such as Pb, Ni, Cd, Mn and Zn. The largest part of environmental lead in cities originates from ex- Preparation to the analysis: The soil samples were dried at hausts of motor vehicles in traffic. Lead is added to gaso- 110 oC for 3h and ground to pass through 200 mesh sieve line as the organic tetraalkyllead additives: tetrame- and homogenized for analysis. 1.0 g of soil samples were thyllead, tetraetyllead, and mixed alkyls triethyl- digested in 15 mL concentrated hydrochloric acid and methyllead; diethyldimethyllead and etyltrimethyl-lead 5 mL of concentrated nitric acid at room temperature, [5]. Motor oils also contain nickel in trace levels [6]. o then heated to 95 C. After the evolution of NO2 fumes had ceased, the mixture was evaporated nearly to dryness The source of nickel in the street dust has been re- on a sandbath and mixed with 20 mL solution containing ported as corrosion of the metallic part of the cars in the 1%(v/v) HCl and 1%(v/v) HNO3. Then, the resulting traffic [7,8]. The source of manganese from traffic has mixture was filtered through a Gouche crucible and the been reported as tyre-wear [7]. Cadmium and zinc exist in insoluble fraction of silicates was determined. The accumulators of motor vehicles or in carburettors, as clear solution was used for an FAAS measurement after alloys [9]. They are released as combustion products. dilution to 50 mL [10,11].

RESULTS AND DISCUSSION Measuring condition: All the samples were analyzed by Perkin Elmer Model Before starting sample analysis the accuracy and pre- 3110 Flame Atomic Absorption spectrophotometer. At- cision of digestion procedure were tested to determine omization was performed by Acetylene/Air flame with a recoveries of the metal ions. Sample was analyzed both flow rate of 2.5 L/h for Acetylene and 6 L/h for air. Two with and without spiked standard containing a mixture of types of sources from Cathedon and Perkin-Elmer were the examined metals of different amounts. The results are employed to excite the elements. shown in Table 1.

The procedure given in experimental was applied to Preparation of standard solutions: All the standard solutions were prepared from analytical the sample collected from Kirazli Kahve Square Station. grade compounds of Merck Company. For each element, No losses of Pb, Ni, Cd, Mn, and Zn was found during six standard solutions of different concentrations in the the digestion due to the high recovery rates of ³96%. A preliminary test involving seven replicate digestions linear range were prepared in 2M HNO3, the optimum linear concentration range for the measurement [12]. on one soil sample for these metals produced relative The calibration curves were prepared for each of the standard deviations of about 4-11%. Using the elements investigated. The least square fitting was procedure, the heavy metal contents under investiga- employed to get the best line in the linear range of the tion in soil samples were determined and the results are calibration lines. presented in Table 2.

TABLE 1 - Analytical performance of digestion Metal Recovery, Relative standard deviation, % (s/x)* Pb 99 0.054 Ni 97 0.038 Cd 96 0.098 Zn 98 0.105 Mn 97 0.045 *N=7

TABLE 2 - Heavy metal concentrations in surface soil of different traffic volume location in Kemalpasa-Turkey (April-June 2000)

No. of vehicles No. of cars Concentration. x±t*s/ÖN. mg/g Location 07.00-19.00h 07.00-19.00h Pb Ni Cd Zn Mn 1. Nift Mountain 0 0 13.2±0.6 29.8±1.3 1.08±0.10 26.7±3.0 217.5±9.1 2. Emniyet Square 5021 2541 115.9±6.0 74.4±2.8 1.18±0.08 51.8±5.3 260.2±14.5 3. Kirazli Kahve 5701 3624 151.1±9.2 66.8±2.6 4.07±0.39 51.6±5.1 332.7±12.4 Square 4. Izmir Road 9034 3806 172.8±7.3 87.4±3.8 2.00±0.14 40.6±3.5 313.4±13.1 (Olmuksa) 5. Akdeniz Kimya 8975 4522 375.2±16.1 101.3±4.8 4.65±0.22 105.3±8.4 271.8±14.3 Cross 6. Izmir Road 9123 3300 146.0±8.3 114.2±5.4 2.81±0.31 83.2±7.6 320.1±13.2 (Omya Maden.) 7. Ankara Road 10927 6733 302.1±18.4 144.7±7.4 5.84±0.76 108.1±10.6 332.9±12.2 8. Ansica Road 5231 2800 143.1±5.6 77.4±3.2 2.67±0.34 33.7±4.0 310.2±13.6 9. Sütcüler Road 4956 3690 152.1±8.8 125.8±4.8 2.73±0.25 46.2±4.0 348.3±14.8 10. Meyhane 8790 3608 Bogazi Square 164.4±7.4 57.9±2.5 3.26±0.36 34.7±4.2 277.1±13.5 *Uncertainty at 95% confidence level (N=6)

TABLE 3 - Correlation between vehicle types and metal concentrations Vehicle types Correlation coefficient, r Pb Ni Cd Zn Mn Cars 0.837 0.806 0.858 0.725 0.664 All vehicles 0.763 0.661 0.677 0.666 0.508

TABLE 4 - Coefficient correlation data between metal concentrations Pb Ni Cd Zn Mn Pb 1.000 Ni 0.641 1.000 Cd 0.825 0.602 1.000 Zn 0.825 0.728 0.752 1.000 Mn 0.320 0.724 0.499 0.268 1.000

Nift Mountain was selected as the control station due This shows that Cd and Zn pollution is not only of traffic to having zero traffic volume. If the heavy metal concen- origin. Soil is a chemically, physically and biologically trations in the roadside soil dusts were compared with the complex system, whose constituents are constantly un- metal contents in the location 1, it is seen that the metal dergoing changes due to weather conditions, concentrations of the dusts are higher than the control geographical location and human activities such as traffic, samples. This shows that traffic volume has a strong in- industrial activity and agricultural [14,15]. Soils in fluence on the metal pollution of the roadside soil dusts. industrialised areas and motorways that have high traffic The highest vehicle number (911 vehicles/h) is at Ankara volume have become contaminated with traces of heavy Road Station. The highest metal concentrations were metals [1,16-18]. Cadmium especially is used in the found in this station. The concentrations of Pb, Ni, Cd, electroplating industry, mining industry, dye productions, Mn, and Zn were found at all places in the range pipe and iron plating industries and agricultural 13-375 mg/g, 30-145 mg/g, 1.1-5.8 mg/g, 218-348 mg/g, applications in Kemalpasa and in this way cadmium id and 27-108 mg/g, respectively. emitted to the environment. A similar results was found by Yalcin et al. and Bereket et al. [2,9]. There was a good correlation between the number of cars and the metal contents (Table 3). According to statis- tical calculations, at the 95% confidence level, the The heavy metal values of soil samples in Kemalpasa correlations between the investigated metal contents and were approximately similar to other studies on urban soil the number of cars or total vehicles number is significant samples [19,20]. [13]. As it was expected, the good correlations have been observed in Pb concentrations, because of leaded petrol used in the vehicles. A linear regression correlation test was performed to investigate correlations between metal concentrations of Cadmium and zinc concentrations were obtained 1.1 our present samples. The values of the correlations’ coef- and 27 mg/g, respectively, at the location 1 where the ficients between metal concentrations in the soil samples traffic volume was zero. The mean value of Cadmium from Kemalpasa are given in Table 4 (p=0.95). Our pre- and zinc in soils world-wide, is <1 mg/g and 15-25 mg/g, sent correlation data have shown good agreement with the respectively [3]. results found by Tatsumi et al. [21].

REFERENCES

[1] Somer G., Aydin H.: 1985, Analyst, 110, 631. Received for publication: August 1, 2000 Accepted for publication: March 15, 2001 [2] Yalcin N., Aydin A.O., Sevinc V.: 1989, Doga TU Müh. ve Cev. D., 13, 429.

[3] Demir M.: 1986, PhD. Thesis, Inönü University, Malatya-TR.

[4] Balci A., Küçüksezgin F.: 1994, Chimica Acta Turcica, 22, 97.

[5] Harrison R.M. and Laxen D.P., Lead Pollution, Chapman and CORRESPONDING AUTHOR Hall, London, 1981. Hakan ARSLAN [6] Yakitlar ve Yaglar, Petrol Ofisi Pub., Ankara, 1980. Department of Chemistry Faculty of Arts & Sciences [7] Akhter M.S., Madany I.M.: 1993, Water Air Soil Poll., 66, 111- Mersin University 119. Mezitli-Çiftlikköyü Kampüsü, 33342 Mersin, TURKEY [8] Fergusson J.E, Kim N.D.: 1991, Sci. Total Environ, 100, 125- 150. Phone: +90-532-707 31 22 Fax: +90-324-361 00 47 [9] Bereket G. and Yücel E.: 1990, Doga-Tr. J. of Chemistry, 14, E-mail: [email protected] 266-271. FEB Vol.10/ No 4/2001 – pages 405 - 408 [10] Allen S.E, Chimshow H.M., Chemical Analysis of Ecological Materyals, John Willey, New York, 1974.

[11] Avila A.K., Curtius A.J.: 1994, J. Anal.Atom. Spect., 9, 543.

[12] Black C.A., Methods of Soil Analysis, Prentice-Hall, Engle- wood Cliffs, 1965.

[13] Alpaut O., Kimyacilar için Pratik Hesap, Hacettepe University Publ., Ankara, 1984.

[14] Chutke N.L., Ambulkar M.N., Singh V.K.N., Garg A.N.: 1994, Fresenius J. Anal. Chem., 350, 723-726.

[15] Nurnberg H.W., Pollutants and their ecotoxicological significance, Wiley, Chichester, 1985.

[16] Warren R.S., Birch P.: 1987, Sci. Total Environ., 59, 253-256.

[17] Ramlan M.N., Badri M.A.: 1989, Environ. Tech. Lett., 10, 435- 444.

[18] Jones K.C.: 1991, Environ. Poll., 69, 311-325.

[19] Francek M.A.: 1992, Environ. Poll., 76, 251.

[20] Nakos G.: 1982, Plant and Soil, 66, 271.

[21] Tatsumi Y., Yoka K., Ikedo A.: 1983, Jap. J. Ecol., 33, 293- 303.

SERUM COPPER AND ZINC CONCENTRATIONS OF PATIENTS WITH RHEUMATOID ARTHRITIS FROM KAYSERI-TURKEY

Mustafa Soylak1 and Mehmet Kirnap2

1Erciyes University, Faculty of Art and Science, Department of Chemistry, Kayseri- TURKEY 2Erciyes University, Faculty of Medicine, Department of Physical Medicine and Rehabilitation, Kayseri- TURKEY

SUMMARY

Serum copper and zinc levels of 10 male and 33 fe- The mean serum copper and zinc concentrations in the male patients, aged 26 to 65 years, with rheumatoid healthy people given by Lapitaj et al. [10] were in the arthritis from Kayseri-Turkey were determined by flame range of 0.5-1.5 mg/l and 0.7-1.2 mg/l, respectively. atomic absorption spectrometry (FAAS). The mean levels of copper and zinc in rheumatoid arthritis were found to In the present study, the variations in the copper and be 1.03 ± 0.37 mg/l and 1.19 ± 0.41 mg/l, respectively. zinc content of serum of the patients with rheumatoid The levels of copper and zinc in the control samples were arthritis living in Kayseri-Türkiye with sexe and age found to be 1.23 ± 0.16 mg/l and 1.66 ± 0.33 mg/l, groups were determined by flame atomic absorption spec- respectively. The zinc and copper levels in the serum of trometry. The correlations between the concentrations of control group were significantly higher than patients with copper and zinc were investigated. rheumatoid arthritis. A positive correlation was found between copper and zinc levels of rheumatoid arthritis.

MATERIALS AND METHODS KEYWORDS: Serum, copper, zinc, rheumatoid arthritis, control group, Re-distilled deionised water from a quartz apparatus flame atomic absorption spectrometry (FAAS), Kayseri –Turkey was used. All analytical reagent grade acids and other chemicals used in this study were obtained from Merck, Darmstadt, Germany. The stock solutions (1000 mg/l) of zinc and copper were prepared by dissolving their nitrates INTRODUCTION in diluted nitric acid. Prior to preparation of standard solutions aliquots of stock solutions were further diluted. Copper and zinc are essential micronutrients for our life. They are part of many metalloenyzimes. Body must The blood samples were collected from 43 patients have zinc and copper to remain healthy. Zinc deficiency with rheumatoid arthritis and 32 people without rheuma- may result from inadequate dietary intake, impaired ab- toid arthritis in November 1998 from Department of sorption, or excessive excretion. Too little copper in the Physical Medicine and Rehabilitation of Erciyes Univer- body can actually lead to disease. Copper is involved in sity Research Hospital in pre-washed (in the order with the functioning of the nervous system, and in maintaining detergent, doubly de-ionized distilled water, diluted the balance of other useful metals in the body such as zinc HNO3 and doubly de-ionized distilled water) polyethylene and possibly other body functions [1-5]. tubes.

The determination of trace element contents of vari- The serum samples were centrifuged at 1500 rpm for ous biological fluids such as urine, plasma, serum etc. 10 min. The samples were separated and stored at -20 oC from the patients with rheumatoid arthritis have been until analyzed. The serum samples used for analysis were performed [6,7]. Also the levels of copper and zinc in thawed and mixed just before assay. Copper and zinc serum are important indicator for rheumatoid arthritis concentrations in the samples were determined by using a [8,9]. In these studies, low serum Zn and high serum Cu Perkin Elmer 3110 model flame atomic absorption spec- levels have been reported for patients with rheumatoid trometer (FAAS) after five-fold dilution, according to arthritis in comparison to control groups. procedure given by Tietz and Soylak et al. [11,12].

RESULTS AND DISCUSSION ACKNOWLEDGEMENTS

Serum copper and zinc concentrations, determined in Mustafa Soylak would like to thank TÜBITAK (the Sci- 43 patients with rheumatoid arthritis, and thirty-two entific and Technical Research Council of Türkiye) for healthy subjects as control group, by flame atomic providing Perkin-Elmer 3110 AAS used in this study. absorption spectrometry (FAAS) are given in Table 1.

TABLE 1 - Copper and Zinc levels in Serum Samples (mg/l) REFERENCES

Concentration, x±SD [1] Soylak M, Elçi L, Dogan M, Analytical Letters 26,1997-2007 (1993). Control Rheumatoid Arthritis [2] Rosenstein ED, Caldwell JR, Rheumatic Diseases Clinics of Copper 1.23 ± 0.16 1.03 ± 0.37 North America, 25, 929-930 (1999).

Zinc 1.66 ± 0.33 1.19 ± 0.41 [3] Soylak M, Dogan M, Su Kimyasi, Erciyes Üniversitesi, Yayin x±SD : Average ± Standard Deviation No: 104, Kayseri (2000).

[4] Saraymen R., Soylak M., Narin I, Fresenius Environmental Serum zinc concentrations of patients with rheuma- Bulletin 7, 403-405 (1998). toid arthritis from Kayseri/Turkey were found to be sig- [5] Saraymen R., Soylak M., Elçi L., Dogan M., Fresenius Envi- nificantly lower than that of the normal, healthy controls ronmental Bulletin 6, 694-698 (1997). (mean (SD) 1.19 (0.41) v 1.66 (0.33) mg/l). These results confirmed the lower zinc level of patients with rheuma- [6] Christensen JM, Pedersen LM, Acta Pharmacol. Toxicol. toid arthritis observed by Honkanen et al [8,9]. The level (Copenh.) 59, 399-402 (1986). of copper in the serum of rheumatoid arthritis and control [7] Pedersen LM, Christensen JM, Acta Pharmacol. Toxicol. group were found to be in the range of 0.65-1.90 mg/l and (Copenh.) 59, 392-395 (1986). 1.00-2.51 mg/l, respectively (Table 1). The mean levels of the copper found in the present study for patients with [8] Honkanen V, Pelkonen P, Mussalo-Rauhamaa H, Lehto J, Westermarck T, Clin. Rheumatol. 8, 64-70 (1989). rheumatoid arthritis was also lower than that in the control group. The results did not confirm the higher copper values [9] Honkanen V, Konttinen YT, Sorsa T, Hukkanen M, Kemppinen for patients with rheumatoid arthritis observed by Hon- P, Santavirta S, Saari H, Westermarck T, J Trace. Elem. Electr. kanen et al [8,9]. The levels of copper and zinc in the serum Health Dis. 5, 261-263 (1991). of both control and rheumatoid arthritis were within the [10] Lapitajs G, Greb U, Dunemann L, Begerow J, Moens L, Ver- limits of that determined by Lapitaj et al. [10]. The varia- rept P, International Lab. 5, 21-27 (1995). tions in the concentrations of copper and zinc of the serum with sexes were also investigated. The levels of copper in [11] Tietz NW (Editor) Textbook of Clinical chemistry, Blanke RV, Decker WJ, Analysis of Toxic Substances, Saunders males (n=10) and females (n=33) with rheumatoid arthritis Company, Philadelphia (1986). were 1.06 ± 0.39 mg/l and 1.02 ± 0.34 mg/l, respectively. Zinc concentrations in the serum of males and females with [12] Soylak M, Saraymen R., Narin I, Dogan M., Ulusal Çinko Semp. 1, 571-575 (1997). rheumatoid arthritis were found to be 1.26 ± 0.44 mg/l and 1.15 ± 0.40 mg/l, respectively. No significant difference has been found between the serum copper and zinc levels of male and female patients with rheumatoid arthritis. Received for publication: May 19, 2000 The average zinc and copper concentrations in the se- Accepted for publication: March 15, 2001 rum samples from Kayseri/Turkey withpect to age were also examined. The concentrations of zinc and copper in the serum samples of the patients with rheumatoid arthritis CORRESPONDING AUTHOR increased with age until 26-35 (n=10) age group, then a decrease was observed in 36-50 (12) and 50+ (n=21) age Mustafa Soylak groups. The highest zinc and copper concentrations were Erciyes University found as 1.34 ± 0.12 mg/l and 1.21 ± 0.20 mg/l, respec- Faculty of Art and Science tively in the 26-35 age group. The correlations between Department of Chemistry copper and zinc concentrations in the serum samples of 38039 Kayseri- TURKEY patients with rheumatoid arthritis were investigated. There was a positive correlation (r= 0.482, 95%) between serum E-mail: [email protected] copper and zinc levels. FEB Vol 10/ No 4/ 2001 – pages 409 - 410

MEASUREMENT OF RADON CONCENTRATION IN NORTHERN GREECE GROUND WATER

A. Savidou and N. Zouridakis

National Centre for Scientific Research “Demokritos”, Athens- Greece

Presented at the 10th International Symposium of MESAEP in Alicante, Spain, 2-6 Oct. 1999

SUMMARY

In this study a Liquid Scintillation Counting (LSC) However, some activities like showering or dish-washing system is used to obtain preliminary concentrations may result in greater radon concentrations than this ratio of radon (222Rn) in ground water samples from Migdonia suggests for short periods, especially in poorly ventilated valley. The Migdonia valley, located NE of the city areas of the home (such as bathrooms). The objective of of Thessaloniki, is an E-W graben. The E-W faults are this survey was to describe the distribution of waterborne large, some km in length and still active, being responsi- radon concentrations in Migdonia ground water supplies ble for great damage during the 1978 earthquake. Water used by the rural population. Previous studies of the area samples were drawn from municipal supplies, wells and showed that the concentrations of U238 in the ground were bore- holes. The measured 222Rn concentrations are up elevated [5]. to100 Bq/L. The U.S. Environmental Protection Agency (EPA) has proposed a maximum contaminant level (MCL) of radon in drinking water of 11 Bq/L (300 pCi/ L). A typical concentration of radon in ground water is 40 Bq/L, but KEYWORDS: sometimes elevated concentrations of some kBq/L in Radon, 222Rn, Groundwater, Liquid Scintillation ground water are reported. The Migdonia basin encompasses a 600 km2 valley with the lakes of Koronia and Volvi and the surrounding mountains that drain into the valley. The population of the INTRODUCTION area is estimated to be 20.000, distributed over 10 vil- lages. The water used by the rural population is from the 222Rn (radon) is an established lung carcinogen. bore-holes of the area. The inhalation of short-lived decay products of radon The lower strata of the Migdonia basin are conglom- accounts on average for about one half of the effective erates, sandstones, clay-sands and red-beds. The upper dose equivalent from all natural sources of radiation [1]. strata consists of clay, sand and gravel intercalations and 222Rn can become airborne from water, entering the home less of conglomerates. The maximum thickness of the from sources like showers, dishwashers, and boiled water sediments is 500 m. from cooking. Exposure to waterborne radon may occur Two types of aquifers are located in the porous sedi- by both ingestion (drinking 222Rn-containing water) and ments of the Migdonia basin: inhalation (breathing 222Rn-gas which has been released by household water). Inhalation is the most important · shallow (5-20m) phreatic aquifer. route for radon, as concerns the contribution to the annual · several deep (30-300m) confined aquifers, effective dose equivalent. It has been estimated that hosted within the sands, gravels and sandstones of 10 Bq/L of 222Rn in water contributes about 1 Bq/m3 of the basin. There are also thermal water in the area. 222Rn to the indoor air [2-4].

MATERIALS AND METHODS

Samples were drawn from wells, bore-holes, at the At laboratory, a 20 mL syringe is filled with 10 mL of the vicinity of the graben as well as from aqueduct tank and appropriate elution cocktail (OPTI-FLUOR, Blend of house taps on two sampling dates one in March and one long chain alkylbenzenes with scintillators PPO and bis- in June. All the samples have been measured by the LSC MSB) and 10 mL of water from the bottom of the sam- system for 50 min, the third day after sampling. The pling vial is drawn into the syringe. The mixture is then background rate for water is 30 counts/min. The detection transferred to a liquid scintillation vial and shaken for limit (2s) of the method for the parameters is 5 Bq/L. about 5 seconds before being set aside for equilibration to take place. After three hours the equilibrium between

radon and its daughters is established and the elution Liquid Scintillation Counting Method to Measure Radon is more than 95 % complete. The counts per minute ob- For sampling, 40 mL glass vials sealed with a Teflon tained in the Liquid Scintillation Counting after three disc are used to prevent leakage of radon from water. hours yielded the radon concentration. Each decay of The water sampling was done with minimum of aeration; radon in water results in 5 detected counts: three alpha the vials were filled to the brim without leaving any particles and two beta particles. The activity of radon in trapped air. The reason is that radon prefers to be in air water is given by the actual rate minus background rate rather than in water, and a 1cc bubble of air will have 4 divided by the number of detected particles per decay. times the radon concentration as in 1 cc of water. This The background counts have been obtained using 10 cc of will give an apparent radon concentration which is lower cocktail and 10 cc of water drawn from municipal tap than the true one. several months before..

TABLE I - Results of Rn-222 measurements in water samples from Migdonia valley The values indicated by asterisks are from the same point of sampling at different sampling date.

Village Sampl/ Date Bq/L March 1999 Bore-hole (30 m) 6 Bore-hole 19 Community Tap 47* I June 1999 Well (3m)

RESULTS AND DISCUSSION

Of the 22 points of sampling (Table I) examined during this study, 70 % of the radon concentrations were above the limit proposed by EPA.

The results from the three points of sampling, that were examined at both sampling dates, revealed no significant differences in radon concentrations. Samples from the same area revealed wide variations. These were in accordance with other studies [9,10].

The results of this study suggest that the area is inter- esting for more detailed investigations over a wider period of time and also the study of other radionuclides of the 238 U series would be of interest

REFERENCES

[1] United Nations Scientific Committee on the Effects of Atomic Radiation. Sources and effects of ionizing radiation, 1993 report to the General Assembly, New York. United Nations sales publication E.94.IX.2, 1993. Page 62-81.

[2] Gesell, T.F. and Prichard H. M.. Natural Radiation Environ- ment III. CONF-780422 (1980).

[3] Hess, C.T., Michel J., Horton T.R. et al. Health Phys. 48: 553- 586 (1985).

[4] Nazaroff , W.W.; Doyle, S. M.; Nero, A. V.; Sextro, R.G.. Health Phys. 52: 281-289 (1987).

[5] Anagnostakis, M.J; Hinis, E.P.; Simopoulos, S.E. and Angelo- poulos’ M. G.. Environment International, Vol. 22; S3- S8;1996.

[6] Zouridakis N.. DEMO 92/3G, 1992. National Centre for Scien- tific Research “Demokritos”, Athens, Greece, 153 10 Ag.Paraskevi Attikis POB 60228.

[7] Zouridakis N.. DEMO 95/9G, March 1995. National Centre for Received for publication: December 13, 1999 Accepted for publication: February 15, 2001 Scientific Research “Demokritos”, Athens, Greece, 153 10 Ag.Paraskevi Attikis POB 60228.

[8] Savidou A., Zouridakis N.. DEMO 99/2G, June 1999. National Centre for Scientific Research “Demokritos”, Athens, Greece, 153 10 Ag.Paraskevi Attikis POB 60228. CORRESPONDING AUTHOR

[9] Drane, W. K. ;York, E. L.; Hightower, J. H. and Watson, J.E.. Health Physics 73(6): 906-911 (1997). A. Savidou and N. Zouridakis National Centre for Scientific Research “Demokritos” 153 10 Ag. Paraskevi, Athens- GREECE [10] Otwoma, D. and Mustapha, A. O.. Health Physics 74(1): 91-95 (1998). FEB – Vol 10/ No 4/ 2001 – pages 411 - 413

FEB- Editorial Board FEB- Advisory Board

Editor-in-Chief: Environmental Analytical Chemistry: K. Ballschmitter, D - K. Fischer, D - R. Kallenborn, N Prof. Dr. H. Parlar D.C.G. Muir, CAN - R. Niessner, D - K. Bester, D Institut für Lebensmitteltechnologie und Analytische Chemie

TU München -85350 Freising-Weihenstephan, Germany Environmental Biology: E-mail: [email protected] D. Adelung, D - A. Görg, D - F. K. Kinoshita, U.S.A G.I. Kvesitadze, GEOR - A. Reichlmayr-Lais, D R. Viswanathan, D

Co-Editors: Environmental Management: F.J. Carlin, Jr., U.S.A - E.B. Fitzpatrick III, U.S.A Environmental Analytical Chemistry: O. Hutzinger, D - L.O. Ruzo, U.S.A - U. Schlottmann, D Prof. Dr. W. Fresenius P.J.M. Weusthof, NL Institut Fresenius GmbH, Im Maisel 14, 65232 Taunusstein, Germany Environmental Chemistry: J.P. Lay, D - J. Burhenne, D - S. Nitz, D Dr. D. Kotzias D.L. Swackhammer, U.S.A. - R. Zepp, U.S.A. Commission of the European Communities, Joint Research Centre Environmental Toxicology: Ispra Establishment, 21020 Ispra (Varese), Italy F. Bro-Rasmussen, DK - F. Coulston, U.S.A. - H. Frank, D H. P. Hagenmeier, D - D. Schulz-Jander, U.S.A. H.U. Wolf, D Environmental Biology:

Prof. Dr. A. Piccolo Università di Napoli “Frederico II”, Dipto. Di Scienze Chimico-Agrarie Via Università 100, 80055 Portici (Napoli), Italy

Prof. Dr. G. Schüürmann UFZ-Umweltforschungszentrum, Sektion Chemische Ökotoxikologie Leipzig-Halle GmbH, Permoserstr.15, 04318 Leipzig, Germany

Environmental Chemistry: Managing Editor: Prof. Dr. M. Bahadir Dr. G. Leupold Institut für Ökologische Chemie und Abfallanalytik Institut für Chemisch-Technische Analyse und Chemische TU Braunschweig Lebensmitteltechnologie, TU München Hagenring 30, 38106 Braunschweig, Germany 85350 Freising-Weihenstephan, Germany E-Mail: [email protected] Prof. Dr. M. Spiteller Institut für Umweltforschung Universität Dortmund Editorial Chief-Officer: Otto-Hahn-Str. 6, 44221 Dortmund, Germany Selma Parlar PSP- Parlar Scientific Publications Angerstr.12, 85354 Freising, Germany Environmental Toxicology: E-Mail: [email protected] www.psp-parlar.de Prof. Dr. H. Greim Istitut für Toxikologie und Umwelthygiene Production & Marketing Chief Manager: Lazarettstr. 62, 80636 München, Germany Max-Josef Kirchmaier MASELL-Agency for Marketing & Communication Prof. Dr. A. Kettrup and Public Relations GSF-Forschungszentrum für Umwelt und Gesundheit GmbH Angerstr.12, 85354 Freising, Germany Ingolstädter Landstraße 1, 85764 Neuherberg, Germany E-Mail: [email protected] www.masell.com

Environmental Management: Dr. H. Schlesing Secretary General Abstracted/ Indexed in: EARTO, Rue de Luxembourg,3, 1000 Brussels, BELGIUM Biology & Environmental Sciences, C.A.B. International, Cambridge Scientific Abstracts, Chemical Abstracts, Prof. F. Vosniakos Current Awareness, Current Contents/ Agriculture, IBIDS data- T.E.I. of Thessaloniki, Applied Physics Lab. base, Information Ventures, Research Alert, Science Citation P.O. Box 14561, 54101 Thessaloniki, Greece Index (SCI), SciSearch, Selected Water Resources Abstracts.

FEB - ORDER FORM

Yes, I wish to order FEB starting with Volume 10/ 2001 Delivery Address (if different)

Name:...... o Printed journal inclusive site-wide access to FEB online version Position:...... 250 EURO plus postage/ handling (Germany 35 EURO/ Europe 55 EURO/ Organization:...... International 90 EURO) Adress:...... o Online journal ...... 200 EURO ......

Minimum subscription period: 1 year E-Mail:...... Cancellation must be generally effected 3 months before end of subscription period. o Will be paid in full Methods of Payment o 50% will be paid (postage/ handling full) because… o Please send me a pre-payment invoice o Member of MESAEP/ SECOTOX o contributor from developing country o I enclose a cheque made payable to PSP

Value Added Tax Special-Offer to Subscribers- back issues at reduced rates! In certain circumstances we may be obliged to charge Subscribers in 2001 are entitled to receive back issues Value Added Tax (VAT) on sales to other EU member at reduced rates (while stocks last). countries. To avoid this, it is therefore essential to provide (6 double issues/ less 50%:Euro 125,00 + postage and handling). us your VAT number if you have one.

To take advantage of this offer please tick the boxes below when ordering a subscription to FEB for 2001. I am not registered for VAT o o 1996 o 1997 o 1998 o 1999 o 2000 o My VAT number is ......

Signature:......

Date: ...... Invoice address

Name: ...... Please complete this form and return to: Position:...... FEB – Fresenius Environmental Bulletin Organization: ...... c/o PSP – Parlar Scientific Publications Adress: ...... Angerstr. 12 - 85354 Freising – GERMANY

...... Tel: ++ 49 (0) 8161 48420 ...... Fax: ++ 49 (0) 8161 484248 E-Mail: [email protected] E-Mail:......

FEB - GUIDE FOR AUTHORS

General

FEB accepts original papers (research reports as well as Short descriptions of the authors, presentation of their theoretical accounts), review articles, short communica- groups and their research activities (with photo) should tions, research abstracts on all aspects of the environmental together not exceed 1 typewritten page. sciences including instrumentation, chemical, physical, and biological methods of analysis, computer based techniques, Short research abstracts should report in a few brief sen- chemometrics, structure elucidation, process control, auto- tences (one-forth to one page) particularly significant mation and robotics, industrial applications, quality assur- findings. ance and laboratory accreditation and book reviews. The journal is also devoted to the development of strategies for Short articles by relative newcomers to the chemical in- solving environmental problems. novation arena highlight the key elements of their Master and PhD-works in about 1 page.

Acceptance or nonacceptance of a contribution will be Book Reviews are normally written in-house, but sugges- decided, as in the case of other scientific journals, by a tions for books to review are welcome. board of reviewers.

Papers are processed with the understanding that they have not been published before (except in form of an Preparation of manuscript abstract or as apart of a published lecture, review or thesis); that it is not under consideration for publication Dear Authors, elsewhere; that its publication has been approved by all co-authors, if any, as well as- tacitly or explicitly- by the FEB is available both as printed journal and as online responsible authorities at the institute where the work has journal on the web. You can now e-mail your manu- been carried out and that, if accepted, it will not be pub- scripts with an attached file. Save both time and lished elsewhere in the same form, in either the same or money! To avoid any problems handling your text, another language, without the consent of the copyright please follow the instructions given below. holders. When preparing your manuscripts have the formula KISS ( Keep It Simple and Stupid) in mind. Most word processing programs such as MS-Word offer a Language lot of features. Some of them can do serious harm to our layout. So please do not insert hyperlinks and/or Papers must be written in English. Spelling may either automatic cross-references, tables of contents, refer- follow American (Webster) or British (Oxford) usage but ences, footnotes, etc. must be consistent. Authors who are less familiar with the 1. Please use the standard format features of English language should seek assistance from proficient your word processor (such as standard.dot colleagues in order to produce manuscripts that are for MS Word). grammatically and linguistically correct. 2. Please do not insert automatisms or secret link-ups between your text and your figures and tables. These features will drive our graphic department sometimes mad.

3. Please only use two fonts- for text or tables Size of manuscript “Times New Roman” and for graphical presentations “Arial”. Review articles should not exceed 30 typewritten pages. 4. Stylesheets, text, tables and grafics in shade In addition up to 5 figures may be included. of grey

5. Turn on the automatic language detection in Original papers must not exceed 14 typewritten pages. In English (American or British) addition up to 5 figures may be included. 6. Please - check your files for viruses before you send them to us!! Short-Communications should be limited to 4 typewritten pages plus not more than 1 illustration. Thank you very much!

Structure of manuscript

1) Title page 8) Acknowledgements

The first page of the manuscript should contain the fol- Acknowledgements of financial support, advice or other lowing items in the sequence given: kind of assistance should be given at the end of the text under the heading "Acknowledgements". The names of A concise title of the paper (no abbreviations) funding organisations should be written in full.

The names1 of all authors with at least one first name spelled out for every author. 9) References

The 1names of University with Faculty, City and Country Responsibility for the accuracy of references rests with of all authors. the authors. References are to be limited in number to those absolutely necessary. 2) Summary References should appear in numerical order in brackets The second page of the manuscript should start with an and in order of their citation in the text. They should be abstract that summarizes briefly the contents of the paper grouped at the end of the paper in numerical order of (except short communications). Its lenght should not appearance. Abbreviated titles of periodicals are to be exceed 150-200 words. The abstract should be as used according to Chemical or Biological Abstracts, but informative as possible. An extended repetition of the names of lesser known journals should be typed in full. paper's title is not considered to be an abstract. References should be be styled and punctuated according to the following examples:

3) Key words

Below the Summary up to 6 key words have to be pro- ORIGINAL PAPERS: vided which will assist indexers in crossindexing your article. 1. AUTHOR, N.N. AND AUTHOR, N.N. (Year) Full title of the article. Journal and Volume, first and last page.

4) Introduction BOOK OR PROCEEDING:

This should define the problem and, if possible, the frame 2. AUTHOR, N.N. AND AUTHOR, N.N. (Year) Title of the contribution. In: Title of the book or proceeding. Volume of existing knowledge. Please ensure that people not (Edition of Editor-s, ed-s) Publisher, City, first and last page working in that particular field will be able to understand the intention. The word length of the introduction should DOCTORAL THESIS: be 150 to 300 words. 3. AUTHOR, N.N. (Year) Title of the thesis, University 5) Material and Methods and Faculty, City

Please be as precise as possible to enable other scientists UNPUBLISHED WORK: to repeat the work. Papers that are unpublished but have been submitted to a journal may be cited with the journal's name followed by "in press". How- ever, this practice is acceptable only if the author has at least re- 6) Results ceived galley proofs of his paper. In all other cases reference must be made to "unpublished work" or "personal communication". Only material pertinent to the subject must be included. Data must not be repeated in figures and tables.

7) Discussion and Conclusion

This part should interpret the results in reference to the problem outlined in the introduction and of related obser- 10) Corresponding Author vations by the author/s or others. Implications for further studies or application may be discussed. A conclusion The name of the corresponding author with complete should be added if results and discussion are combined. postal address and E-mail address.

Tables Offprints

Every table should be numbered in Arabic numerals in the Precondition for publishing: sequences in which they occur. They are to be included in A minimum number of 25 Offprints must be ordered and the manuscript. Every table must begin with a caption that prepaid. starts with, for example, "Table 2”. The caption must explain precisely the contents of the table. The table itself They are purchased at cost price [1-4 pages = 6 €; 5-8 must be written so that it can be read and understood pages = 10 €, every further page 1 €; additionally postage/ without reference to the text. Every column and every line handling (Germany 10 €, Europe 15 €, International 20 €) of a table must be labeled unambiguously and indicate and VAT in Germany and EU member countries (if you units wherever data are reported. References to a table are do not have a VAT-No.)]. Granting of 50% discount for to be handled in the same way as references to the text members of MESAEP, SECOTOX, contributors from (see Section References). Footnotes to a table should be developing countries, and students. 25 offprints will be indicated by lower-case letters in parentheses and typed provided free of charge only to subscribers of FEB. directly under the table.

Figures Copyright

The figures should be numbered consecutively in Arabic The articles published in this journal are protected by numerals in order of mention in the text. Every figure copyright. All rights are reserved, especially the right to must be accompanied by a legend that begins with, for translate into foreign language. No part of the journal may example, "Figure 4". be reproduced in any form- through photocopying, microfilming or other processes- or converted to a machine language, especially for data processing equipment- with- Photographs out the written permission of the publisher. The rights of reproduction by lecture, radio and television transmission, Black-and-white photographs are to be submitted in TIF- magnetic sound recording or similar means are also re- format (shade of gray) or as JPEG black-and-white- served. format (shade of gray). Glossy prints with soft contrasts are also acceptable.

SI metric system Abstracted/ Indexed in:

SI units are to be used for all data (exceptions: L, g, bar, Biology & Environmental Sciences, C.A.B. International, h, ppm, ppb, ppt), e.g. c (NaOH) = 0,1 mol L-1. Greek/ Cambridge Scientific Abstracts, Chemical Abstracts, unusual symbols/ abbreviations should be defined in the Current Awareness, Current Contents/ Agriculture, IBIDS text at their first occurrence. database, Information Ventures, Research Alert, Science Citation Index (SCI), SciSearch, Selected Water Re- Submission of manuscript sources Abstracts.

The manuscripts should be sent directly to:

PSP Publishing, Angerstr.12, 85354 Freising GERMANY.

Email: [email protected]

Authors are requested to submit manuscripts in electronic form (as an E-Mail attachement). Electronic manuscripts eliminate the need for re-keying and thereby introduction of new errors. They must be in exact journal format and identical to the final hard copies. The manuscript should be saved in the native format of the word processor used (please use Microsoft Word). Authors should keep copies of everything submitted.

SUBJECT INDEX

A P aqueous medium 353 PAH 381 Autocatalytic effect 386 PCBs 342, 375 pesticides 353, 357 B pesticide residues 401 batch reactor 386 photodegradation 357 bell mouthed port 392 photoproducts 357 biosorption 363 pollen 363 Black Sea 392 pollution 405 Bulgaria 401 polychlorinated biphenyls 342, 375 C polychromatic light 353 catalytic effect 386 project life 392 control group 409 proton spin-lattice relaxation time 368 copper 409 Q copper biosorption 363 quantum efficiency 353 copper removal 363 CPMAS 13C-NMR 368 R cross-polarization time 368 radon 411 D rheumatoid arthritis 409 222Rn 411 daily intake 342 denitrification 396 S diffuser 392 sediment 375 E semi-volatile compounds 381 serum 409 equilibrium constant 381 sharp edged port 392 F soil 405 fish 375 solar irradiation 353 fixed-bed filter 396 SPME 381 flame atomic absorption spectrometry 409 standard generation 381 furfural 357 T furfurylalcohol 357 trace metals 405 G traffic 405 groundwater 411 Typha latifolia 363 H V human milk 342, 349 volcanic soils 368 K W Kayseri -Turkey 409 wastewater 396 kinetics 396 water 401 L Z Libya 349 zinc 409 life time 353

Liquid Scintillation 411

M manganese oxidation 386 marine outfall 392 molasses 396 molecular motions 368

organochlorine residues 349 subject-index

AUTHOR INDEX

A M Abdennebi, El H. 349 Meallier, P. 353 Albanis, T. 353 Moza, P.N. 357 Algur, Ö.F. 363 Arayici, S. 386 N Arslan, H. 405 Nemlioglu, S. 392 Aydin, S. 386 B P Baussand, P. 381 Pavúk, M. 342, 375 Bratanova, Z. 401 Petrík, J. 342, 375 Buurman, P. 368 Piccolo, A. 368

C R Chovancová, J. 342, 375 Rahmani, M. 349 Conte, P. 368 S D Savidou, A. 411 Danis, Ü. 363 Soylak, M. 409 Desuzinges, P. 381 Drobná, B. 342, 375 T Timur, H. 396 E Tripoli, F. 381 Elafi, A. 349 Tüfekçi, N. 386 Emmelin, C. 353 V F van Lagen, B. 368 Feicht, E.A. 357 Vassilev, K. 401 Foster, P. 381 Z G Zouridakis, N. 411

Guittonneau, S. 353

H Hemminga, M.A. 368 Hustert, K. 357

J Jacob, V. 381 Jursa, S. 375

Kaluzny, P. 381 Kaya, Y. 363 Kettrup, A. 357 Kirnap, M. 409 Kocan, A. 342, 375 Konstantinou, I. 353

L Lassagne, L. 381 author-index

420