Distribution and bioavailability of manganese in soil in the vicinity of the „ Bužim “ abandoned mine

Soil samples from the industrial area in the town of Bužim, Bosnia and Herzegovina were analysed in order to determine their different manganese species. Samples were extracted from seven locations – at the manganese mine and the surrounding area. The paper aims to present the use of the sequential extraction method in determination of the specifi c distribution of Mn in soil, as well as in estimation of its origin, mobility and bioavailability in the sampling locations. Sequential extraction used here included determination of the amount of Mn in various soil fractions: the water-soluble fraction, exchangeable fraction, carbonate fraction, easily reduced fraction and the organic fraction. Additionally, it included manganese oxides or moderately reduced oxides, amorphous iron oxide, crystalline iron oxide and the residual fraction. It was determined that chemical properties of soil considerably affect the distribution of heavy metals within different soil fractions. The highest percentage of natural Mn was determined in the residual fraction (27.00%) at Popović polje, while the highest percentage of anthropogenic origin Mn was determined at Bućevci (57.00%) in the Fe-Mn oxides fraction. The highest near-total content of Mn was determined at Popović polje (20950.00 mg/kg). The highest percentage of natural Mn (27.00%) was determined in the same area. The highest percentage of Mn of an anthropogenic origin (57.00%) was determined at Bućevci.


INTRODUCTION
Interest in the geochemical pattern of soil contamination has increased within the last few decades, due to the fact that soils serve as a depot for metals in an environment (wARREN & BIRcH, 1987;LI & THORNTON, 2001;OREŠČANIN et al., 2003;MARTLEY et al., 2004).MAcKLIN (1992) provides a description of soil and sediment contamination by metals as a result of mining activities.
The research included analysis of water-soluble species, exchangeable species, carbonate species, easily reduced species, organic species, manganese oxides or moderately reduced oxides, amorphous iron oxide, crystalline iron oxide and the residual fraction of Mn in soil in order to determine how mining activity affected the amount and origin of Mn in soil.
The procedure of sequential extraction applied in this research is combined and modified according to the standardized procedure of sequential extraction in the European community Bureau of Reference (BcR), TESSIER et al. (1979TESSIER et al. ( , 1987)), AGNIESZKA & wITOLD (2004), MING et al. (2008), SILVEIRA et al. (2006) and other references dealing with the same issues (KAZI et al., 2005;PUEYO et al., 2008;DAV-IDSON et al., 2006;JOSE et al., 2006;KAASALAINEN & YLI-HALLA, 2003;ROUHOLLAH & FARZANEH, 2010).A modified procedure of sequential extraction includes the determination of Mn bound to particular soil phases in nine fractions: water-soluble fraction, exchangeable fraction, carbonate fraction, easily reduced fraction, organic fraction, manganese oxides or moderately reduced oxides, amorphous iron oxide, crystalline iron oxide and the residual fraction of Mn.The first three steps of sequential extraction provide further data on the bioavailability of metals (JAIN, 2004).

AREA OF INVESTIGATION
The research was conducted in the farthest northeast part of Bosnia and Herzegovina in the town of Bužim.Bužim has a population of about 20.000 inhabitants.It borders with the towns of Bosanska Krupa (east), cazin (south), Velika Kladusa (west and north) and the Republic of croatia (northeast).A well-known manganese mine in Bosnia and Herzegovina is located in Bužim, but has been out of use since the war in Bosnia and Herzegovina.
The area is characterised by slightly rounded and seldom angular hills.The lowest altitudes occur in the Bužimica river valley (150m).This is an area of mildly continental and mountain climate.The manganese deposit is placed in a synclinal structure which extends in a northeast-southwest direction.It occurs where the volcanogenic-sedimentary series overlie dolomites and dolomitic limestone.The deposit is dark red in colour, due to the high manganese content.Occasionally, greenish tufa and light grey kaolin rich marls can be observed.MOJIĆEVIĆ et al. (1984) determined the age of the sediments in the wider area of Bužim, having discovered that these are composed of Triassic, cretaceous, Palaeocene, Miocene, and Pliocene-quaternary structures.Figure 1. is a location and sampling scheme map of the collection area near the manganese mine in "Bužim" and the surrounding area.

Sampling
Sample collection from the soil was conducted by means of the standard pedologic methods of soil sampling, as well as the open pit sampling with reference to ISO 10381 procedures (ISO, 2002).Sampling was conducted by means of the standard pedologic equipment and tools, while the special attention was given to prevention of sample contamination.
Sample collection occurred at the end of March 2010 at the manganese mine "Bužim" and the surrounding area and a total of 10 soil samples from different locations have been used.
Four samples were collected from a soil profile close to the mine at different depths, i.e. from four horizons in loose and solid part using Kopecky cylinders (Figure 2).clearly visible sequences of genetic horizons have been observed during the process of sample collection from the soil profile at a depth of 100 cm.The four horizons are distinguishable at a 0-26 cm, 26-49 cm, 49-69 cm and 69-100 cm deep.
A total of six ordinary loose samples were taken from the surrounding area of the mine at a depth of 0-25 cm.Locations of these samples were: Popović polje located 2.5-3 km north-west of the mine, Rusija and Bajraktarević polje (2 km to the south-west), Varoška rijeka also 2.5 km southwest of the mine, Radostovo (4 km to the south) and Bućevci 7 km south-east of the mine.These locations were chosen in order to determine the distribution of manganese in horizontal profile at the surrounding area in relation to the manganese mine.

Sample preparation and analysis
The soil samples were prepared for physico-chemical analyses according to the ISO 11464 procedure (ISO 1994b).
Soil samples used in the research were prepared and cleared of impurities (organic matter and visible shells), airdried, sifted with 2 mm hole diameter sieve, homogenised by the quartering method and sampled for laboratory analysis.Samples prepared in this way were used for physical and chemical analysis of soil.A fragment of prepared samples was dried in dryer at a temperature of 105 °c for an hour, afterwards powdered in agate mortar according to the ISO 11465 procedure (ISO 1993).Samples prepared in this way were used for determination of the near-total content of manganese in all samples, as well as for sequential extraction of manganese in the nine fractions.

Methods of analysis
Samples were analysed by means of the following methods: -Determination of the hygroscopic moisture content of the soil (Hy) according to the ISO 11465 procedure (ISO 1993), -The mineral part of the soil was determined by calcination at 500-600 °c 2 h (ČUSTOVIc & TVIcA, 2005), -Determination of granulometric soil texture was conducted according to the ISO 11277 procedure (ISO 1998).The texture of soil samples was conducted according to Ehwald (RESULOVIc & ČUSTOVIĆ, 2002), -Bulk density was determined according to the ISO 11272 procedure (ISO 1993), -Particle density was determined according to the ISO 11508 procedure (ISO 1998), -Total porosity (%) was calculated using the values of bulk density and particle density (PEKEČ et al., 2013).
-Amount of humus was determined by determination of organic carbon (c) by sulphocromic oxidation as prescribed by ISO 14235 (ISO 1998).A correction factor of 1.724 was used to calculate humus from organic c.
-Amount of carbonate in the soil was determined according to the ISO 10693 procedure (ISO 1995).
-Determination of soil pH value was conducted according to the ISO 10390 procedure (ISO 1994c), -Sum of the absorbed base cations and hydrolytic acidity were determinedby Kapenn's method (ČUSTOVIĆ & TVIcA, 2005) -The total adsorption capacity (T) was calculated from the total adsorbed basic cations (S) and total adsorbed H + ions (H).-The degree of base saturation was calculated from the proportions of base cations (S) relative to the total adsorption capacity (T) and was expressed as a percentage.
-Easily accessible phosphorus and potassiumin the soil, are determined through AL-method (Enger-Riehm-Domino) by ammonium lactate-acetic acid extractaction described by EGNER et al. (1960), -Near -total Mn content in the soil was determined by AAS after digestion of soil samples in a microwave oven, -Methods of Mn sequential extraction of soil samples are presented in Table 1.

Method for determining the near-total Mn content in soil samples
Near-total contents of Mn in all soil samples were determined from their extracts (SILVEIRA et al., 2006).Sample extracts were processed by microwave digestion with the addition of 6 ml Hcl (36%) and 2 ml HNO 3 (65%) -microwave technique on a MARSXpress system (cEM) according to the EPA 3050 (USEPA, 1996).The microwave digestion programme included two phases: first, 8 minutes at 160°c, and 1200w; and second 10 minutes at 160°c, and 1200w.After completion of digestion and cooling, extracts were filtered (whattman type 50 filter paper) in glass bottles of 50 ml and diluted with distilled water to the level line (MOSSOP & DAVIDSON, 2003).Extracts prepared in this way were stored at 4°c until further analysis by AAS according to the procedures of ISO 11047 (ISO 1998).

Methodology for the sequential extraction of manganese from the soil
The defined steps and reagents used in the process of sequential extraction are shown in Table 1.
The extracts from the residual fractions were processed by microwave digestion using the same procedure as for determination of the near-total content of manganese in soil samples.Manganese concentrations in acidic residues were measured by Atomic Absorption Spectrophotometer (Zeeman, Varian), using a calibration curve method (LI et al., 2000).
3.3.3.Quality assurance and quality control (QA/QC) quality assurance (qA) of the research has been realised by means of the appropriate laboratory facilities, specialised work procedures, use of distilled water, high-purity chemicals, and measurements conducted on the equipment with satisfying performance (analytical scales, pH meter, AAS).
quality control (qc) has been realised by an analysis of the relevant blind experiments, calibration of the equipment using standardised solutions, and repetition of the analyses using the same methodology with reference to ISO 17025 procedures (ISO, 2005).

Physical and chemical soil analysis
The determined values of specific physical and chemical soil properties are shown in Tables 2 and 3. Research has shown that particle density decreases in a vertical soil profile with increasing depth, i.e. from 2.51 g/cm 3 in the 1 st horizon to 2.30 g/cm 3 in the 4 th horizon.Bulk density displays different values within horizons of the vertical soil profile.The average amount in the 1 st horizon is 1.46 g/cm 3 , 1.44 g/cm 3 in the 2 nd , 1.49 g/cm 3 in the 3 rd , and 1.41 g/cm 3 in the 4 th horizon.Based on the particle density and bulk density, it is possible to determine whether the soil is arable compacted soil.
Based on the research of soil porosity we determined soil with low porosity, with a total amount of pore space of 38.60% in 4 th horizon to 43.50% in 2 nd horizon.
The results of granulometric analysis indicate that samples from the 2 nd and 3 rd horizons of the vertical profile are similar in granulometric properties and can be regarded as a clay loam.Samples taken from the locations surrounding the manganese mine (Popović polje, Rusija, Bajraktarević polje, Varoška rijeka, Radostovo and Bućevci) at depth of 0-25 cm displayed similar granulometric properties, which are: sandy loaml, loam, loamy sand and silty sand.This can be regarded as being caused by intensive deep erosion.
The soils pH values range from moderately acidic to mildly alkaline, from pH=4.64 at the manganese mine location in the deepest horizon to pH=7.26 at the location of Varoška rijeka.At different horizons in vertical profile, pH value decreases from the 1 st to 4 th horizons.
All soil samples displayed lower pH values of Kcl than the pH values of H 2 O, especially in soils where pH value is usually high.Soils with delta pH higher than 0.10, are said to have geric properties, i.e. are very exposed to erosion and have low adsorption capacity or a low number of exchangeable bases (RESULOVIĆ, 1997).Delta pH had the highest value (pH=0.65)at the 4 th horizon (100 cm) at the manganese mine.
The amount of cacO 3 in all samples was lower than the limit set by the method, since the dissolution of soil samples was conducted by using diluted hydrochloric acid.Based on the results of research and values obtained for the level of base saturation (V), ranging from 73.58% at Popović polje to 99.47% at Varoška rijeka, it can be stated that all tested locations are Eutric cambisols.

Near-total content of manganese
The near-total content of manganese for each sample location in the vertical profile are presented in Table 4, as well as for each sample location surrounding the "Bužim" man-   The amount of Mn was significantly decreased at the 4 th horizon (780.00 mg/kg), as displayed in Figure 3.
Figure 4. indicates that within the examination of all locations at a depth of 0-25 cm, the highest near-total content of 20950.00mg/kg of Mn, was measured at the area of Popović polje, while the lowest amount was measured at the area of Bućevci of 418.00 mg/kg.The high content of Mn in certain locations is probably due to the close proximity of the sampling areas to the mine.
Results indicate that, the sample locations contain extremely high concentrations of manganese, which can be related to the former mining activities in the area.

The distribution of Mn in different soil fraction determined from sequential extraction
Table 5. presents the Mn contents in the individual fractions as determined by the process of sequential extraction.The sum of all the examined Mn fractions displays relatively good correlation with near-total content of Mn at the same soil samples, while this type of correlation was not displayed at the area of Popović polje.According to ZEMBERYOVA et al. ( 2006) figures 5 and 6 display the extracted percentages of Mn among the examined fractions (in relation to the certified value of 100).MEERS et al. (2007) define the water-soluble fraction as being easily mobilised and a highly available fraction.In the first soluble fraction, the percentage of Mn in the vertical profile of horizons at the manganese mine location is lower than 1.00%, thus posing no threat according to the assessment code.The percentage of soluble fraction in soil samples from the area surrounding manganese mine at all sampling locations is lower than 1.00%, thus having no detrimental effect on the environment.
The exchangeable metal fraction is loosely bound to colloidal soil particles (AGNIESZKA & wITOLD, 2004).NOVOZAMSKI et al. (1993) proposed the use of cacl 2 as a reagent for evaluation of metal and nutrient bioavailability in air-dried soil sample.HOUBA et al. (1996HOUBA et al. ( , 2000) ) state the advantages of this reagent.In the second exchangeable fraction, the percentage of Mn in the vertical profile (sampled in the mine) is as follows: 2.00 % in 1 st horizon, 1.00% in the 2 nd and 3 rd horizon and 3.00 % in the 4 th horizon.The highest percentage of the exchangeable manganese fraction is obtained in the 4 th horizon, probably due to an increased ionic exchange which occurs as the soil depth and amount of cations in the soil increase.
The percentage of exchangeable fractions at locations surrounding the manganese mine (1st horizon) is as follows: <0.10% in Popović polje and Bajraktarević polje, 2.00% in Rusija and Varoška rijeka, 1.00% in Radostovo and Bućevci.The percentage of Mn in the exchangeable fraction (mobile metal fraction) at all sample locations, according to the assessment code, has no detrimental effect on the environment.
The carbonate fraction is regarded as an easily mobilised fraction.Extraction within this fraction was conducted by means of reagent 1M NaOAc solution, pH=5 adjusted with acetic acid in order to transfer most of the absorbed Mn into solution (BAEYENS et al., 2003), as the fraction is highly sensitive to changes of pH value (TESSIER et al., 1979).According to ROBBINS et al. (1984) the reagent releases 99.90% of the carbonate fraction and has minimum influence on other fractions.The percentage of carbonate fraction in vertical profile (sampled in the mine) of horizons ranges from 1.00% in 2 nd , 3 rd and 4 th horizon, to 2.00% in 1 st horizon, thus posing low risk to the environment according to the assessment code.The highest percentage of carbonate fraction in the 1 st horizon can be explained by the susceptibility of the fraction to changes of pH value (TESS-IER et al., 1979).consequently, it was observed that the highest percentage of carbonate fraction correlates with the highest measured pH value (5.53) at the 1 st horizon soil sample in the vertical profile in the area of the mine.
The percentages of carbonate fraction at the other sample locations surrounding the manganese mine (1 st horizon) are: <0.10% in Popović polje, 3.00% in Rusija, 1.00% in Bajraktarević polje, 12.00 % in Varoška rijeka, 3.00% in Radostovo and 9.00% in Bućevci.The highest percentage of the carbonate fraction was detected at the location of Varoška rijeka (12.00%) correlating with the highest measured pH value at the same location (7.26), which confirms the fact that the fraction is susceptible to changes of pH value.Moreover, it proves the sensitivity of the fraction to changes of pH-value and indicates that as the soil pH value increases, so does the percentage of carbonate and consequently of the bioavailable manganese fraction in the soil.The percentage of Mn in the carbonate soil fraction at the location of Varoška rijeka has a limited effect on the environment.
Easily reduced substances (Fe-Mn bound oxides) display moderate mobility, depending on the redox conditions of the environment.According to SILVERA et al. (2006) the redox potential of soil can significantly increase the solubility of Fe and Mn oxides.
Fe and Mn oxides have a great effect on control and mobility of other metals in the environment.In the reduced fraction (Fe-Mn bound oxides), the percentages of Mn in the vertical profile (sampled in the mine) of individual horizons are: 1 st horizon 21.00%, 2 nd horizon 17.00%, 3 rd horizon 15.00% and 4 th horizon 31.00%.The highest percentage of  Mn in the vertical profile was determined at the 4 th horizon (31.00%), which could be due to an increased redox potential there, leading to an increased solubility of the fraction.The highest percentage of Mn in this fraction correlates with the lowest measured pH value at the 4 th horizon.
Percentages of the reduced fraction at the locations surrounding the manganese mine (1 st horizon) are: Popović polje and Rusija 11.00%, Bajraktarević polje 29.00%, Varoška rijeka 48.00%, Radostovo 19.00% and Bućevci 57.00%.The percentage of the reduced fraction at all sample locations is relatively high, due to numerous natural and anthropogenic sources of Mn over the area.
The highest percentage of the easily reduced fraction is determined in the area of Bućevci (57.00%), which can be explained by manganese oxidation resulting from mining activity there (NADASKA et al., 2009).The fact that Mn dominates in the fraction of Fe-Mn oxide at the area of Bućevci indicates they share a common origin.
According to HAUNG et al. (2007) Mn bound to the organic fraction displays moderate mobility, although it can be increased due to the oxidation of organic matter.The solubility of the Mn fraction bound to organic matter, precedes those of Mn oxides, amorphous and crystalline iron oxides.This enables organic phase degradation, providing better conditions for the extraction of fractions bound to Mn (AL-VAREZ et al., 2006).
The percentage of the organic fraction in the vertical profile (sampled in the mine) of individual horizons is: 1 st horizon 52.00%, 2 nd horizon 47.00%, 3 rd horizon 49.00% and 4 th horizon 29.00%.The percentage of the organic fraction at sample locations surrounding the "Bužim" manganese mine (1 st horizon) is: Popović polje 19.00%, Rusija 49.00%, Bajraktarević polje 30.00%,Varoška rijeka 9.00%, Radostovo 44.00% and Bućevci 10.00%.Due to a high affinity of bonding to organic matter (RESULOVIĆ, 1997) Mn is dominant in this fraction with the highest percentage (52.00%) in the 1 st horizon of the vertical profile (sampled in the mine).
The percentage of manganese oxides or moderately reduced oxides in horizons of vertical profiles (sampled in the mine) is: 1 st horizon 8.00%, 2 nd horizon 6.00%, 3 rd horizon 7.00% and 4 th horizon 5.00%.The percentage of this fraction at other locations surrounding the manganese mine (1 st horizon) is: Popović polje 11.00%, Rusija i Bajraktarević polje 9.00%, Varoška rijeka and Bućevci 4.00%, Radostovo 6.00%.The highest percentage of the fraction was measured at the Popović polje location (11.00%), which can be related to relatively low pH-values of the location (5.23) (SILVEIRA et al., 2006).
The percentage of Mn bound to the amorphous iron oxide fraction in the horizons of the vertical profile (sampled in the mine) is:1 st and 4 th horizon 3.00%, 2 nd horizon 4.00% and 3 rd horizon 5.00%.At locations surrounding the manganese mine (1 st horizon) it is as follows: Popović polje 10.00%, Rusija 5.00%, Bajraktarević polje 15.00%, Varoška rijeka and Bućevci 2.00% and Radostovo 5.00%.The highest percentage of this fraction is measured in the area of Bajraktarević polje (15.00%).ScHwERTMANN (1991) proved the efficacy of 6 M Hcl used for dissolution of the crystalline iron oxide fraction.In his research, the author stated that nearly 60.00% of Fe dissolves with 6M Hcl solution (SILVIERA et al., 2006).The percentage of Mn bound to the crystalline iron oxide fraction in individual horizons of vertical profile (sampled in the mine) is determined as: 1 st horizon 4.00%, 2 nd , 3 rd and 4 th horizon 8.00%.At locations surrounding the manganese mine (1 st horizon) it is as follows: Popović polje 22.00%, Rusija 8.00%, Bajraktarević polje 10.00%, Varoška rijeka 16.00%, Radostovo 6.00% and Bućevci 8.00%.The highest percentage of the fraction was measured in the Popović polje (22.00%).
The highest percentage of Mn in this fraction correlates with the highest percentage in manganese oxides and moderately reduced oxides fraction, which can be due to the size of their ionic radii and isomorphic substitution of Mn and Fe (SILVEIRA et al., 2006).
According to SUTHERLAND et al. (2000), metals extracted in non-residual fractions indicate an anthropogenic origin, and metals extracted in the residual fraction indicate a natural origin (FORSTNER, 1983;BLAScO et al., 2000;JAIN, 2004;RAMIREZ et al., 2005).PRIcA (2011) states that the residual fraction cannot be mobilised from geological material and is not bioavailable.
A rather high percentage of Mn at the aforementioned locations is probably due to extremely strong bonds of Mn and the soil solid phase.The highest percentage of the fraction is determined at the area of Popović polje (27.00%).The highest percentage of Mn in this fraction in the area correlates with the crystalline iron oxide fraction and Mn oxide and moderately reduced oxide fraction.

CONCLUSIONS
According to the research conducted at the "Bužim" manganese mine and the surrounding area, the following conclusions can be drawn: According to pH values the soils are moderately acidic to mildly alkaline, thus being considered as Eutric cambisols.The results of sequential extraction indicate that chemical soil properties significantly affect Mn distribution in the different soil fractions.The highest percentage of bioavailable Mn was measured at the "Bužim" manganese mine (in 1 st fraction) and in Popović polje (in 2 nd fraction), since there are numerous natural and anthropogenic sources of Mn at these areas.The percentage of Mn soluble fraction is less than 1.00% at all locations, thus having no detrimental effect on the environment.The highest percentage of exchangeable manganese fraction (3.00%) was measured in the 4 th horizon of the vertical profile, probably due to an increased ionic exchange which occurs as depth increases, thus having no detrimental effect on the environment according to the assessment code.The highest percentage of the carbonate fraction (12.00%) was determined at the location of Varoška rijeka correlating with pH=7.26.The percentage of Mn in the carbonate fraction at the location of Varoška rijeka moderately affects the environment.
The highest percentage of natural Mn was determined in the residual fraction (27.00%) at Popović polje, while the highest percentage of anthropogenic origin Mn was determined at Bućevci (57.00%) in the Fe-Mn oxides fraction.The highest near-total content of Mn was determined at Popović polje (20950.00mg/kg).The highest percentage of natural Mn (27.00%) was determined in the same area.The highest percentage of Mn of an anthropogenic origin (57.00%) was determined at Bućevci.The results are possibly due to mining activity which took place there in the past.
Further research shall be directed towards additional techniques such as X-ray analysis, which is highly important in the research of mineral phases enabling the determination of any correlation between certain chemical forms (species) of Mn at the aforementioned soil areas.

Figure 2 :
Figure 2: Layout of soil horizons in the soil profile.

Figure 3 :
Figure 3: Near-total content of Mn at the mining area.

Figure 4 :
Figure 4: Near-total content of Mn in the surrounding area.

Figure 5 :
Figure 5: Distribution of Mn at the mining area using the sequential extraction procedure.

Figure 6 :
Figure 6: Distribution of Mn in the area surrounding the mine using sequential the extraction procedure.

Table 1 :
Sequential extraction of Mn from soil.

Table 2 :
Physical properties of soil.

Table 3 :
Chemical properties of soil.

Table 4 :
Near-total content of Mn for each sample location.

Table 5 :
the amount of Mn in certain fractions after sequential extraction procedure (mg/kg).