Chromium and nickel distribution in soils, active river, overbank sediments and dust around the Burrel chromium smelter (Albania)

https://doi.org/10.1016/j.gexplo.2005.07.005Get rights and content

Abstract

Chromium ore was treated to produce ferrochromium from 1979 until 2000 in a smelter in Burrel, 35 km NE of Tirana (Albania). As a consequence, large amounts of solid waste, i.e. slags (about 9.106 m3) have been disposed next to the smelter, disfiguring the landscape. In an attempt to define contaminated sites, heavy metal content of the different sampling media have been compared with respective background samples.

In the study area, the determination of background values in soil samples is complicated due to the different geological substrates. Cr and Ni background concentrations in serpentinite-derived soils, west of the smelting plant, are markedly higher than in the Pliocene gravel/sandy soils, where the smelter is situated (Cr 2147 and 193 mg/kg, respectively; Ni 2356 and 264 mg/kg). These values are clearly lower than those encountered around the smelter. Average total Cr and Ni concentrations in soils around the smelter are 3117 and 1243 mg/kg, respectively. The highest concentrations of Cr (up to 2.3 wt.%), were recorded in samples taken near the smelting compartment within the industrial plant and next to the slags clearly indicating that the smelter forms a point source of Cr contamination. The Cr / Fe ratio is the best indicator to differentiate non-polluted (Cr / Fe Serpentinite soil: 130–390; Pliocene soils: < 130) from polluted areas (> 390 smelting nearby of the slags).

Cr and Ni values for local backgrounds in stream and overbank sediments were taken in the Mat river 6 km upstream and to the east of the smelter (268 and 430 mg/kg for Cr, and 306 and 604 mg/kg for Ni, respectively). Equivalent sediments taken from the Zalli i Germanit river, which drains the smelter area are respectively 816 and 1126 mg/kg for Cr and 1115 and 1185 mg/kg for Ni.

Dust samples, taken from the lofts of houses up to 2 km from the smelter, display high concentrations of Cr, Ni and Zn (average contents of 2899, 436 and 902 mg/kg, respectively). The later concentrations in the dust samples have been confirmed by mineralogical analysis where Cr-bearing mineral phases such as ferrochromium and chromium oxides, clearly relate to the activity of the smelter. Consequently, atmospheric deposition of dust particles forms a serious problem and can also be responsible for the elevated contents encountered in soil samples around the smelter.

All these data show that the degree of contamination caused by industrial activity of the Burrel Cr-smelter is severe, although no Cr(VI) was detected in soil water extractions nor in the surface or groundwater where concentrations were < 0.01 mg/kg.

Introduction

Chromium occurs as a trace element in many minerals, but the only mineral of commercial importance is chromite [(Mg,Fe)(Al,Cr,Fe)2O4]. Its simplified mineralogical composition is FeCr2O4. In chromite ore, the Cr2O3 oxide content varies between 15% and 65%. Estimated annual world consumption of chromium is about 12.5 million tonnes of which 85% are consumed by metallurgy, 8% by chemicals, and 7% by refractories (Darrie, 2001).

Albania has been an important chromite producing country, since major podiform chromite deposits occur mainly in the eastern ophiolite belt. It occupied the world's third place in chromium ore production until 1991, before the industry collapsed due to major political changes in the early nineties (Perron, 1995). The absence of any environmental legislation until 1990 and due to economical considerations, mining and industrial sites were severely polluted. However, no detailed data on the total pollutant content in different sampling media exists. Apart from the total content, element speciation is important, particularly for chromium, since it can exist in the environment in two stable oxidation states, namely Cr(III) and Cr(VI). Cr(III) is considered to be an essential trace element for the functioning of living organisms. According to Bartlett and Kimble (1976) chromium might control the glucose, known as glucose tolerance factor (CrGTF). Cr in soils and rocks occurs mostly as chromite which is extremely insoluble (Becquer et al., 2003). Cr(VI) is toxic and easily soluble and is an acute irritant of living cells (James, 1996). Skin contact induces allergies (Yassi and Niober, 1988). Furthermore it is carcinogenic to humans via inhalation (Whalley et al., 1999). Concentrations of Cr(VI), as low as 0.5 mg/kg in solution and 5 mg/kg in soils can be toxic to plants (Turner and Rust, 1971). Elevated Cr concentrations often are associated with other heavy metals such as Ni which in high concentrations might pose human and eco-toxicological concern.

Ferrochromium, slags as well as gas–dust emissions are the final by-products of the smelter and obviously form a potential point source for pollution (see for example Macklin and Dowsett, 1989, Maskall et al., 1995). Furthermore, in the study area the slags have been used in local road construction causing a widespread contamination. Areal contamination can also be due to spreading of wind blown dust particles, originating directly from the smelter chimneys or from slag dumps. Finally, additional spread of contaminants can occur along rivers which drain the industrial areas (Wolfenden and Lewin, 1977, Swennen et al., 1994).

The aim of this study is to map the areal distribution particularly of Cr and Ni, but also of associated heavy metals (Zn, Co, Cu) in different sampling media (i.e. soils, river and overbank sediments as well as dust) around the Burrel industrial site. To be able to fully assess the degree of contamination, local background concentrations in the different sample media were analyzed. Attention had to be paid to the fact that two geological substrates dominate the study area. Serpentinite rocks form the hilly substrate west of the smelter while Pliocene gravel/sand occur in the plain around the smelter site and downstream of this potential contamination source. In addition, possible atmospheric contamination has to be taken into account. To define local background values, soil and substrate samples were analyzed above and within serpentinite and Pliocene gravel/sand at a distance 3–8 km from the smelter. Additionally active stream sediments as well as overbank from several locations upstream of the smelter in an area with similar geological substrate as encountered near the smelter were analyzed. Overbank samples were especially taken since, according to Ottesen et al. (1989) and Swennen et al., 1994, Swennen et al., 1996, samples taken at depth can record pristine or at least pre-industrial values within specific catchment areas.

Section snippets

Overall setting and Cr smelting industry

Albania is situated in the western part of the Balkan Peninsula (i.e. east of Adriatic and Ionic Seas). About 50% of the territory has an elevation situated between 200–1000 m, while the rest corresponds to lowland areas (e.g. Periadriatic Depression). Its climate is typically Mediterranean, with moderate and wet winters and very hot and dry summers (ASHRSH-QSGJ, 1990).

Albania is made up of different geological units, which are grouped in the Albanides (Fig. 1A). In the central and eastern part

Deduction of local background concentrations

Since geochemical signatures in soils and sediments are lithology-dependent (Hawkes and Webb, 1962, Levinson, 1974), first local background values will be discussed.

Lithogeochemical data of the Serpentinite and Pliocene substrata and overlying soils are given in Table 1. Due to the limited number of samples analyzed it is impossible to report on local threshold values. Average concentrations for Cr and Ni in the studied Serpentinite rocks are 1625 and 1515 mg/kg, respectively. These values are

Conclusions

The data gathered from soils, active river and overbank sediments and dust samples around the Burrel smelter, shows that in all sampling media a clear signal can be deciphered reflecting pollution by the industrial activities. However the study also showed that there is a need to define local background concentrations. This is of especial importance in evaluating the soil geochemical signatures, otherwise soils developed upon ultramafic rocks could be wrongly classified as polluted.

Acknowledgments

We acknowledge the assistance of Dany Coetermans from the Fysico-chemische Geologie during the laboratory work as well as Prof. Geoffrey King whom comments helped to improve the English. The comments of the reviewers helped in improving the quality of the manuscript. The research was realized with the financial support of the Katholieke Universiteit of Leuven, East European Initiatives, Project 3E000659.

References (34)

  • G. Darrie

    Commercial extraction technology and process waste disposal in the manufacture of chromium chemicals from ore

    Environmental Geochemistry and Health

    (2001)
  • J.M. Davies et al.

    Mortality from respiratory cancer and other causes in United Kingdom chromate production workers

    British Journal of Industrial Medicine

    (1991)
  • M. Gosar et al.

    Mercury in soil and attic dust as a reflection of Idrija mining and mineralization (Slovenia)

    Geologija

    (2001)
  • H.E. Hawkes et al.
  • ISPTK

    Njohja e gjendjes dhe përmirësimet e mundëshme teknologjike në sistemin e pluhurkapjes nga furrat

  • B.R. James

    The challenge of remediating chromium contaminated soil

    Environmental Science and Technology

    (1996)
  • J. Japenga et al.

    Heavy metal and organic micropollutants in floodplains of the river Waal, a distributary of the river Rhine, 1958–1981

    Netherlands Journal of Agricultural Science

    (1990)
  • Cited by (0)

    View full text