HEAVY METALS POLLUTION IN TO THE MARANO LAGOON IN NORTHERN ITALY

A. Piacenti*, M. Ferrini*, F. La Marca*, L. Piga°

* Department of Chemical, Materials, Raw Materials and Metallurgical Engineering

University of Rome “La Sapienza”, Rome, Italy

° National Research Council, Rome, Italy

Corresponding author: F. La Marca <floriana@imagemp.ing.uniroma1.it>

Abstract

This paper concerns the environmental pollution in the most northern lagoon of Europe, halfway the cities of Venice and Trieste, close to the Adriatic Sea. As a consequence of industrial and natural pollution from near mercury bearing region of Idria, a very high level of heavy metals content has been revealed. Samples have been taken concerning sediments, water and lagoon vegetation. Chemical analyses have shown the existence of diffused mercury and chromium pollution. It has been demonstrated that mercury is transferring from sediments to the trophic chain, until fish and at least to the men. The distribution of pollution agents in the lagoon is pointed out utilizing geostatistical methods in order to localize sediments to be removed for remediation purpose of sea roads into the lagoon.

A strategy for sediments removal by selective drainage of the lagoon has been pointed out.

Introduction

The Marano-Grado lagoon is situated in the wetlands system extending along the northern Adriatic coast, and occupies a surface of 160 km² (32 km long, 5 km wide), confined westward by Tagliamento river delta and eastward by Isonzo river delta (Figure 1). Geological studies have highlighted as sediments mean size distribution decreases gradually from west toward east, and from inner toward outer harbor, by moving away from the terrigenous source (Marocco, 1995).

Industrial settlements are located in the surrounding area, particularly chemical industries for cellulose production and tanneries, while fishing and valliculture activities are developed in the lagoon. Environmental studies on the Marano-Grado lagoon revealed a significant pollution from heavy metals, especially mercury and chromium. River Cormor is considered the main chromium pollution source: the metal concentration at river mouth is about 160 ppm. This pollution is mainly due to the effluents from industrial plants for electro-plastics and tanneries around the river basin. Referring to mercury pollution, contamination decreases from east toward west (from Marano lagoon toward Grado lagoon) and from hinterland toward the port entrance (mercury concentration varies from 15¸20 to 0.6 ppm). Furthermore, there is a close relation between the metal content in sediments and their fine size distribution. The Idria cinnabar deposit long the Isonzo basin in Slovenia, outside Italy, has been identified as the main pollution source (Biester et al., 1999; Hines et al., 1999). 3500 years of mining exploitation and in particular the last 500 years of intense mining activity, ended about twenty years ago, have produced a huge soil and sediments contamination. As a complement of the mine pollution, a mercury spill from Torviscosa cloro-soda plant occurred for 35 years, estimating on amount of 186,000 kg. This event caused a potential contamination equal to 11.2 g/m², as resulting from the pollution level in the sediments in the area between the Marano channel and the industrial plants close to the Aussa-Corno river.

In 1996, according to a provisional annual mass balance in the Trieste Gulf, it has been estimated that the Isonzo river should discharge about 1,780 kg /year; the 93% should be deposited and 74% of this fraction should be settled. Deep sediments as mercury interceptor/re-distributor in environment are the main contamination sources in the Trieste Gulf (Širca et al, 1999). Methylmercury production process is favored by reducing conditions (internal areas with low hydrodinamics) and by organic matter presence (abundant in the eastern part of the lagoon) (Bloom and Lasorsa, 1999). The 100% bioavailable methylimercury comes into contact with deep waters and transfers in the marine food chain. A positive correlation between mercury level in sediments, in algae and in bentonic organisms has been proved (Brambati, 1996). The metal accumulation increases by proceeding in the upper tropic levels (biomagnification) reaching its maximum in the final consumers (human beings included). It has been shown that some cerebral malfunctions, as senile dementia, memory disorders and Alzheimer's disease are in connection with mercury and its vapors accumulation and retention due to a fish based feeding (Khatoon et al., 1989; Wenstrup et al., 1990). The methilymercury minimum dose proposed by the World Health Organization for a critical exposure is 0.003¸0.007 mg/kg*d for adults, reduced to 0.001 mg/kg*d for pregnant women to avoid neonatal neurological disorder (IPCS, 1991). Mercury concentration values, found in fishes catch in the lagoon, are near to the warning level. Studies of the Italian National Institute for Feeding and of the Italian National Agency for Environment demonstrate that all the people in the Marano area present mercury level in the hair higher than the average level (1.8¸27.8 ppm, being average level @6.4 ppm). In this contest, the disposal of dredging mud for lagoon boat navigation is a relevant question (Gentilomo, 1998; Ringeling, 1998). Contamination level in most of the area under investigation is above the admissible limit imposed by Regional and European legislation on soil quality. It is necessary to reclaim sediments to prevent lagoon pollution. Until now, the toxicological situation is under control, but new antropic contributions and/or river Isonzo floods could increase the mercury flow in the lagoon and cause a contamination extent.

Methodology

Sediment, water and plant samples have been collected from 20 sampling stations in March 1999 as a care of the University of Rome “La Sapienza” in collaboration with the “A.S.S. Bassa Friulana” of the Friuli Venezia Giulia Region, in order to determine the lagoon contamination level and to suggest a treatment process for decontaminating the polluted dredging mud (Figure 1). Sediment sampling has been realized by a Vanveen bucket, collecting thus the superficial layer that is involved in mobilization phenomena. Furthermore, water, plant, algae and root samples have been gathered whereas present. In fact, plants, by their nature, accumulate heavy metals in tissues.

a)

b)

Figure 1: Marano-Grado lagoon hydrology (a) and sampling stations (b) by University of Rome “La Sapienza”.

A size distribution analysis has been applied to sediments by wet screening in three classes: >1 mm, <1 mm >38 μm and <38 μm (Table I). The finer class (<38 μm) is particularly interesting, because of the high concentration in contaminants, specially heavy metals. It is also a proper size easy cycloning for separation from sands, preceding further treatment processes or control analyses.

Chemical analyses have been performed on each sample to determine the content in heavy metals, particularly referring to mercury ad chromium. A microwave batch Milestone-1200 has been applied to contaminant determination in sediments and plants (root and aerial portion). The chromium concentration has been determined by inductively-coupled plasma atomic emission spectroscopy (ICP-AES spectroscopy), adopting the Perkin-Elmer spectrophotometer PLASMA 400, while mercury concentration has been determined by hydride generation atomic absorption spectrometry. Results have been compared with the available data related to previous sampling campaigns (1987-1997). Geostatistical analysis has been performed in order to carry out a mercury distribution model in the lagoon area, considering the limits imposed by environmental rules.

Table I: Size distribution and chemical analysis results, concerning the sediment samples from Marano lagoon.

	Size class <38 mm			Total
Samples	Weight	Hg	Cr	Hg
	%	ppm	ppm	ppm
MA1	89.21	2.20	31.71	1.97
MA2	87.45	2.51	37.27	2.20
MA3	90.35	2.21	41.08	2.15
MA4	92.83	4.05	43.61	3.90
MA5	47.71	3.57	31.29	1.88
MA6	25.73	2.19	40.37	1.83
MA7	87.44	3.14	44.85	2.99
MA8	5.91	2.65	222.59	0.90
MA9	23.21	5.62	70.61	2.82
MA10	3.05	1.83	203.91	1.98
MA11	66.02	6.04	55.57	4.60
MA12	28.17	10.68	54.79	4.44
MA13	63.76	8.92	48.42	6.22
MA14	76.30	6.80	57.12	5.21
MA15	58.98	8.95	46.53	5.28
MA16	12.76	9.46	49.29	2.85
MA17	31.42	12.27	45.12	5.18
MA18	24.35	10.04	42.88	3.86
MA19	0.65	12.37	69.66	2.02
MA20	6.14	15.12	47.26	2.34

A scheme for remediation process for dredging mud and polluted sediments is proposed in Figure 2. It includes the following ten operations: (i) dredging of sediments by sludge pump, installed on dredger; (ii) separation of the size class >1 mm by Humphrey spirals, installed on dredger; (iii) separation of the size class <38 mm, by cycloning; (iv) control of the mercury grade of the size class <1 mm >38 mm and disposal by application of Venice Protocol; (v) handling of the size class <38 mm by centrifugal pump; (vi) thickening and filtration of the size class <38 mm; (vii) characterization of thickened mud; (viii) thermal treatment of thickened mud (if necessary), (ix) mercury removal and recovery from roasting gas and (x) landfilling.

The mud dredging involves 1,100,000 m³ of materials. The volume of materials to be reclaimed has been evaluated on the risk map, considering the areas with Hg concentration higher than the admissible limit, for 10 cm in depth, for an amount equal to 5,900,000 m³: in total 7,000,000 m³ (Table II).

Table II: Materials balance of the polluted sediments into the Marano lagoon.

Material	Sediments from mud dredging		Sediments to be reclaimed		Total
	m^3	t	m^3	t	m^3	t
to be removed ^(1)	1,100,000	1,650,000	5,900,000	8,850,000	7,000,000	10,500,000
to be reclaimed ^(2)	308,000	462,000	1,121,000	1,681,500	1,429,000	2,143,500
after treatment ^(3)		441,672		1,607,514		2,049,186
to be disposed (dry)		1,208,328		7,242,486		8,450,814
to be disposed (wet) ^(4)	805,552	1,530,548	4,828,324	9,173,816	5,633,876	10,704,364

⁽¹⁾ dry specific weight: 1.5 t/m³;

⁽²⁾ weight of the size class <38mm: 28% for sediments from mud dredging and 19% for sediments to be reclaimed;

⁽³⁾ weight loss after mercury desorbing is equal to 4.4%;

⁽⁴⁾ wet specific gravity: 1.9 t/m³.

A risk map has been defined after recognition of the areas of the lagoon on the basis of Hg contamination levels (OECD, 1995): for Hg <0.5 ppm, remediation is not necessary, sea disposal is provided for dredging mud, near shallow water or deep areas; for Hg 0.5¸2.0 ppm, remediation is not necessary and treatment for size class <38 μm of dredging mud, disposal for size class >38 μm according the admissible limits of Venice Protocol is provided; Hg >2.0 ppm, remediation is necessary, treatment for size class <38 μm of dredging mud and sediments to be reclaimed, disposal for size class <1 mm >38 μm according the admissible limits of Venice Protocol is provided (prudently, this fraction may be treated too).

Thermal desorption process consists of: (i) pre-heating at the temperature of 120 °C for materials drying and water vapor and other volatile compound removal, (ii) heating for mercury removal, (iii) condensation and recovery of metallic mercury, (iv) activated carbon treatment of waste waters deriving from pre-heating and (v) percolation of gaseous effluents deriving from the heating process throughout activated carbon column before release into the atmosphere (Figure 3).

Figure 3: Remediation process scheme.	Figure 4: Thermal desorption process scheme.

The thermal desorption process has been tested on a reconstituted sample of the fraction <38 μm, by a Carbolite MTF 12/25/250 oven. To improve the process efficiency iron sulfide (FeS) and metallic iron (Fe) have been utilized as catalysts, as proposed in recent studies (Matsuyama and Akagi 1999).

Volatile compounds, mercury and water vapors were carried away by a constant airflow of 5 l/h through the oven. The elementary mercury was condensed as hydroxide in two soda baths downstream, to retain acidic gases generated in the heating phase (Russel, 1996). 6 tests have been performed at different temperature levels, from 300°C to 500 °C, and reagents (no reagents, adding 0.055 g of FeS and adding 0.100 g of Fe for a 10-g sample). The duration of pre-heating and heating processes has been set to 2 h in each test. After treatment, chemical analyses have been performed on treated samples to determine residual content in mercury.

Table III: Thermal desorption results carried out on a reconstituted sample of the fraction <38 μm.

Reagents	Temperature	Residual Hg	Hg removal
	°C	ppm	%
-	300	1.97	77.91
-	500	1.12	87.44
FeS	300	0.51	94.28
FeS	500	<0.50	>97.20
Fe	300	0.83	90.70
Fe	500	<0.50	>97.20

Results and discussion

Size distribution analysis demonstrates as the mean size of the sediments decrease in direction East-West and South-North. The weight percentage of the size fraction <38 mm is about 28% inside channels (average value equal to 19% in the lagoon) because of fine material transport by tidal current. As previously specified, heavy metal content in sediment increases by decreasing particle size (see Table I). This is due to the phenomena of adhesion/adsorption on fine particles surfaces of metals, coating carbonate particles with an organic film. The maximum mercury concentration level is 6.22 ppm in sediments and 15.12 ppm in size class <38 mm. Chemical analyses results show a different heavy metals content in roots and leaves of aquatic plants, indicating their behavior in pollutants accumulation around root apparatus. Mercury content in plants is then lower than in sediments, however the Hg content results higher than the average value measured for lagoon plants, equal to about 0.5 ppm. Chemical analysis on water samples and supernatant fractions (recovered after size class <38 μm settling), indicates a content in metals suitable with imposed limits of Italian legislation, referring to pollutants concentration for effluents immission in superficial water body. Thermal desorption tests highlight reagent efficiency (Table III, Figures 5 and 6). In fact, mercury removal is improved by adding Fe or FeS as reagents, at both temperatures (300 °C and 500 °C).

The suggested procedure is able to face the problems related both to sediments remediation and dredging mud processing. The process comprehends waste waters and gaseous effluents treatment.

Figure 5: Hg removal by thermal desorption.	Figure 6: Residual Hg after thermal desorption.

Conclusions

Past and present samplings (sediments, water, plants) and related chemical analyses indicate a diffuse pollution from mercury and chromium in sediments of the Marano-Grado lagoons, coherent with the previous data reported in literature. Geostatistical analysis allows to distribute contamination on a risk map of the lagoon. Contamination due to mercury is particularly interesting, since transferred to feed chain, involving human health.

A remediation strategy should be pointed out for sediments, able to capture and extract the pollutants. Treatment of sediments from dredging operations is also necessary. Polluted areas are mainly close to industrial settlements on Cormor river mouth (chromium) and on Aussa-Corno river mouth (mercury). A high concentration in mercury has been found also close to Isonzo river mouth, still transporting mercury tailings of ceased mining and metallurgical activities located in the high basin of the river.

Proposed treatment process allows to reduce mercury content from 8.9 ppm to values lower than 0.5 ppm, corresponding to a mercury removal higher than 97.2%. The volume of material to landfill is then estimated to be decreased from 7,000,000 m³ to 5,600,000 m³.

References

Biester H., Gosar M., Müller G. (1999), Mercury speciation in tailings and sediments of the Idrja mercury mine. 5^thInternational Conference Mercury as a global pollutant, Rio De Janeiro.

Bloom N.S., Lasorsa B.K. (1999), Production and release of methyl mercury as a result of marine sediment dredging activities. 5^thInternational Conference Mercury as a global pollutant, Rio De Janeiro.

Brambati (1996), Metalli pesanti nelle lagune di Grado e Marano. Regione Autonoma Friuli-Venezia Giulia.

Falnoga, Horvat M., Mandic V., Stegnar P. (1999), Mercury in Human autopsy samples from Idrija residents and retired mercury workers. 5^thInternational Conference Mercury as a global pollutant, Rio De Janeiro.

Gentilomo M. (1998), Il trattamento e la destinazione finale dei materiali dragati – aspetti generali e casi particolari. IA Ingegneria Ambientale Vol. XXVII n. 7-8:342-351.

Hines M.E., Horvat M., Faganeli J., Bonzongo J-C.J., Barkay T., Major E.B., Scott K., Bailey E.A., Warwick J.J., Lyons W.B. (1999), Mercury biogeochemistry in the Idrija river, Slovenia from above the mine into the Gulf of Trieste. 5^thInternational Conference Mercury as a global pollutant, Rio De Janeiro.

IPCS (International Programme on Chemical Safety) (1991). Environmental Health Criteria 118: Inorganic Mercury. World Health Organisation, Geneva.

Khatoon S., Campbell S.R., Haley B.E., Slevin J.T. (1989), Aberrant guanosine triphosphate-beta-tubulin interaction in Alzheimer’s disease. Ann. Neurol., 26:210-215.

Marocco R. (1995), Sediment distribution and dispersal in northern Adriatic lagoons (Marano and Grado paralic system). Giornale di Geologia, Vol. 57 n. 1-2.

Matsuyama, Akagi H. (1999), Basic research on the treatment of mercury-contaminated soil at moderate temperatures using iron sulfide. 5^thInternational Conference Mercury as a global pollutant, Rio De Janeiro.

OECD (Organisation for Economic Co-operation and Development) (1995), Risk Reduction Monograph no. 4: Mercury. Paris.

Ringeling R.H.P. (1998), Handling of contaminated dredged material. Delft University Press.

Russel J. (1996), Environmental Engineering Sourcebook. EPA.

Širca, Horvat M., Covelli S., Faganelli J., Rajar R. (1999), Mercury fluxes and mass balance in the Gulf of Trieste. 5^thInternational Conference Mercury as a global pollutant, Rio De Janeiro.

Wenstrup, Ehmann W.D., Markesbery W.R. (1990), Trace element in balances in isolated subcellular fractions of Alzheimer’s disease brains. Brain Res., 533:125-31.