Environ Dev Sustain DOI 10.1007/s10668-014-9590-1 The speciation of cobalt and nickel at mine waste dump using improved correlation analysis: a case study of Sarcheshmeh copper mine Saeed Yousefi • Faramarz Doulati Ardejani • Mansour Ziaii Mohammad Karamoozian • Received: 4 June 2014 / Accepted: 9 October 2014 Ó Springer Science+Business Media Dordrecht 2014 Abstract Investigating the speciation of potentially toxic elements in mining waste dump systems provides valuable knowledge about the potential for transfer in the environment and the risks posed to mining sites. Sequential extraction analyses are common experiments, which are often used to determine the speciation of potentially toxic elements. However, there would be some drawbacks for using this experiment including labourintensive procedure, interferences of fractions, impractical for testing large numbers of samples from a heterogeneous environment and the inability to determine the individual minerals relevant to the corresponding fraction. The present paper is an attempt to determine the speciation of cobalt and nickel as potentially toxic elements in the waste dumps of Sarcheshmeh using improved correlation analysis. This method employed the cobalt and nickel contents together with the exact mineral contents which were classified according to the paste pH experiments for improving the correlation matrix. To achieve the aim of study, sixty samples were collected from two waste dumps at the Sarcheshmeh Copper mine in Kerman province of Iran. The result of proposed method showed that cobalt bound to hydroxysulphate minerals, muscovite and iron and manganese oxyhydroxide minerals and nickel is controlled by hydroxysulphate minerals, and manganese and iron oxyhydroxide minerals, as paste pH ranges ascend. Furthermore, at all paste pH ranges, pyrite was the main source of cobalt and nickel. These results were in agreement with the sequential extraction method and also previous experimental investigations, which confirms the performance of applied improved correlation analysis. S. Yousefi Department of Mining, Faculty of Engineering, University of Birjand, Birjand, Iran S. Yousefi (&)  F. Doulati Ardejani  M. Ziaii  M. Karamoozian Faculty of Mining, Petroleum and Geophysics Engineering, University of Shahrood, Shahrood, Iran e-mail: s.yousefi@birjand.ac.ir F. Doulati Ardejani School of Mining, College of Engineering, University of Tehran, Tehran, Iran 123 S. Yousefi et al. Keywords Acid mine drainage  Bioavailability  Sequential extraction analysis  Paste pH experiment 1 Introduction Cobalt (Co) and nickel (Ni) are members of the iron family or the ferrides which have similar behaviour from environmental point of view (Rankama and Sahama 1949). They are the 33th and 24th most abundant elements in the Earth’s crust, respectively (Cempel and Nikel 2006; Hamilton 1994). Cobalt is often found in association with nickel, silver, lead, copper and iron ores and occurs in mineral form as arsenides, sulphides and oxides. It is essential in trace amounts for humans and other mammals as it is an integral component of the vitamin B12 complex; however, larger concentrations will be harmful (Bruland et al. 1991). Nickel can exist in various ores such as sulphides, arsenides, antimonides and oxides or silicates. The most predominant forms are nickel sulphate and nickel oxides. Chalcopyrite, pyrite, pentlandite, garnierite, nicolite and millerite are considered as the chief sources of nickel (Soen 1964). Chronic exposure to cobalt and nickel may result in allergic reactions such as skin rashes, thyroid damage and internal organ cancers such as lung cancer, nose cancer, larynx cancer and prostate cancer (Abdul 2014; Adepoju-Bello et al. 2014). Most cobalt and nickel compounds released to the environment are adsorbed by sediment or soil particles and become immobile. However, acidic environment, such as polluted/contaminated soils near mining site or melting facilities, can lead to the remobilization of adsorbed metals and often rinse out to the surface and groundwater. Under this condition, the uptake by plants and animals is higher and accumulation in them may occur (Barałkiewicz and Siepak 1999). Therefore, most risk from cobalt and nickel are associated with the forms of them that easily accessible to the surface water and biologically available (bioavailable) to plants. Generally, the identification of the different chemical forms and phases relevant to the elements is called chemical speciation (Kumar et al. 2012). The speciation of toxic element has been investigated by sequential extraction (fractionation) analysis wherein they are categorized into several fractions based on a stepwise addition of reagents with increasing reactivity (Tessier et al. 1979). These fractions include soluble in water, exchangeable, reducible, oxidizable and residual. They are often attributed to the elements associated, bonded or adsorbed on hydroxysulphates, exchangeable sites in clay minerals or carbonates, in the iron and manganese oxyhydroxides, organic matter/sulphides and silicate phases, respectively (Violante et al. 2007). According to this categorization, the risk of toxicity is decreasing from ‘‘water soluble’’ to ‘‘residual’’ fractions (Khorasanipour et al. 2011a). Although sequential extraction analysis has been previously implemented in numerous projects to evaluate potentially toxic elements, it has, however, some shortcomings about complexity of procedure, interferences of fractions, inability in considering heterogeneity in a medium such as mine waste dumps and lack of exact determination of individual mineral as source or sinks terms. According to the variety of protocols, the procedural complexity is related to the wide range of possible processes, which leads to selectivity of different reagents, extraction time, solid-to-liquid ratios, agitation type, methods used for liquid/solid separation, test 123 The speciation of cobalt and nickel at mine waste dump sample mass and rinsing method (Hall et al. 1996; McCarty et al. 1998; Gleyzes et al. 2002). Hence, it will be a difficult task to compare the results obtained by different laboratories using various protocols. Consequently, different quantification errors are associated with sequential extraction analysis compared with single-step extraction, namely total extraction such as inductively coupled plasma mass spectrometry (ICP-MP) method (Villanueva et al. 2013). Ideally, extractants are designed to dissolve selectively one mineralogical phase of the initial material, but practically, according to the type of ore body, the other fractions are solved. This phenomenon can lead to interference of fractions and uncertainty of results (Gleyzes et al. 2002; Dold 2003). Due to being labour intensive, the application of sequential extraction is limited to a few number of samples. While waste dumps are huge in volume and have high heterogeneity from mineralogical, physical and geochemical point of view (Williams and Smith 2000), therefore, the small number of samples may cause less understanding about the mechanisms contributed to toxic elements speciation. The reason why this method is commonly applied in relatively homogeneous environments such as mine tailings, soil and sediment and rarely used for waste dumps (Leinz et al. 1999, 2000). Furthermore, sequential extraction can only determine the type of minerals containing elements where it is incapable of distinguishing the individual mineral responsible for immobilization of toxic elements. For example, sequential extraction can determine nickel as exchangeable element on clay minerals, but cannot determine which given mineral adsorb it. In addition to mentioned problems, the incomplete dissolution of some fractions, changes in pH of solvent, alteration on surface chemical characteristics and consequently developing the fresh surfaces can lead to the re-adsorption and the re-distribution of some elements (Gleyzes et al. 2002; Zimmerman and Weindor 2010). According to above-mentioned drawbacks, it is necessary to develop a method that can describe the speciation of potentially toxic elements. This subject highlights the bioavailability of potentially toxic elements which is very obligatory for environmental risk assessment of waste dumps as a basis of the future remediation programme. According to former investigations in Sarcheshmeh mine (Khorasanipour et al. 2012), cobalt and nickel are crucial elements with respect to environmental impacts associated with stream sediments. Therefore, the main aim of this study was to implement an improved correlation analysis in order to develop a general methodology for fractionation of cobalt and nickel in a mine waste dump environment. According to former investigations in the Sarcheshmeh mine site (Fig. 1a), dumps no. 19 and 31 (Fig. 1b) have high acid-producing potential (AP) (Alizadegan 2010; Khorasanipour et al. 2011b). Based on the modified Sobek method (Jambor 2003), net acid-producing potential (NAPP) values showed that these dumps can generate 107 and 92 kg H2SO4 per ton of waste, respectively (Alizadegan 2010). Because of the releasing of cobalt and nickel are a consequent of AP and acid mine drainage (AMD) generation, hence, the present study focuses on these two dumps for identifying the fractionation of them. Up to now, statistical methods were used in several environmental assessment studies such as soil (Facchinelli et al. 2001; Martı́nez et al. 2008; Zhang et al. 2008; Yalcin et al. 2010; Franco-Urı́a et al. 2009), sediment (Huang et al. 1994; Delvalls et al. 1998; Soares et al. 1999; Loska and Wiechuła 2003; Chaparro et al. 2011) and water resources (Zhou et al. 2008; Bu et al. 2010; Akbal et al. 2011; Akin et al. 2011). These methods have been widely developed for source apportionment between natural and anthropogenic contributions in industrial areas at the regional scale (Soares et al. 1999; Zhang et al. 2008; Zhou et al. 2008; Candeias et al. 2011; Wang and Lu 2011). However, no statistical study has yet 123 S. Yousefi et al. Fig. 1 Overview of the Sarcheshmeh mine area a geographical situation of Sarcheshmeh mine in Iran b a plan view of Sarcheshmeh mine complex c plan of the dump 31 accompanied by the location of trenches d cross section of dump 31 e plan of the dump 19 accompanied by the location of trenches f cross section of dump 19. Note that the vertical scale of the cross section is exaggerated for better visibility been focused to determine the speciation of toxic elements especially in waste dumps which is aggregate of anthropogenic activity in local scale. 2 Study area The Sarcheshmeh porphyry copper deposit is the biggest copper mine in Iran and one of the largest Oligo-Miocene deposits in the world. Sarcheshmeh mine is situated in southern 123 The speciation of cobalt and nickel at mine waste dump Iran at 30° N, 56° E, and about 160 km south-west of Kerman city (Fig. 1a). This mine is located in a semi-arid climate with a mean annual precipitation of 440 mm (Khorasanipour et al. 2012). Open-pit mining has been employed for more than 35 years, in the Sarcheshmeh area. The mine site consists of mining units, tailings dam, waste dumps, processing, melting and moulding plants. The Shour stream (Fig. 1b) is the major drainage in which mine water, acidic drainages from waste dumps, pilot and processing plants waste waters discharge to it. The Sarcheshmeh ore body, with dimensions of 2,000 m by 900 m, contains 1,200 million tons of ore with average grades of 1.13 % copper, 0.03 % molybdenum, 3.9 ppm silver and 0.11 ppm gold and a cut-off grade of 0.4 % copper (Waterman and Hamilton 1975). The development of mining activities in the region has produced in over 400 million tons of mining wastes. To minimize transportation costs, mining wastes are usually dumped in natural valleys near the mine. The Sarcheshmeh mine has 31 active and inactive waste dumps, some of which generate AMD, especially in the wet seasons. 3 Materials and methods 3.1 Sampling To achieve the aim of the study, six trenches with maximum 6.5 m in depth (A, B, C, D, E and F) were excavated in the above-mentioned waste dumps (Fig. 1c, e). Altogether, sixty samples were taken from the trenches in November 2011. To take representative samples, nearly 4 kg of waste material, sieved by screen 4 mesh, were collected at each sampling location. Samples were taken using a stainless steel device and stored in air-tight polyethylene plastic bags. The samples were then sent to the Central laboratory of the Sarcheshmeh Copper Complex for preparation and performing further processes required before chemical and mineralogical analyses. 3.2 Analytical method Total concentrations of cobalt and nickel were determined by inductively coupled plasma mass spectrometry (ICP-MS) method. Mineralogical studies including quantitative and qualitative analyses of primary and secondary minerals were carried out by X-ray diffraction (XRD) and studying thin and polished sections of the collected samples. XRD was qualitatively conducted by a Philips multipurpose X-ray diffraction system at the Iran Mineral Processing Research Centre (IMPRC). Mineralogical quantification was done according to Rietveld method (Rietveld 1993). This method can provide quantitative estimations even poorly crystallized minerals by having the chemical composition of samples such as ICP or X-ray fluorescence (XRF) data (Dercz et al. 2008). 3.3 Experiment 3.3.1 ASTM standard test method for determining the form of sulphur In environmental impact assessments of AMD, the determination of pyrite and secondary hydroxysulphate minerals originated from pyrite oxidation process is crucial. Due to a high 123 S. Yousefi et al. detection limit (approximately 2 %) and unfavourable efficiency related to poorly crystalline minerals, quantitative XRD cannot exactly determine negligible contents of pyrite, hydroxysulphate and iron oxyhydroxide minerals. Hence, a method introduced by ASTM (D 2492) (American Society for Testing and Materials 1986) was employed to determine such minerals. This method was based on two steps which were conducted for all samples. Step 1: diluted hydrochloric acid (HCl) was used to dissolve hydroxysulphate and oxyhydroxide minerals. It should be noted that diluted HCl cannot digest sulphide minerals. The obtained solution from this step was diluted to volume and analysed for total iron and sulphate (SO42-). Sulphate was measured by emission spectrometry which represented the hydroxysulphate minerals content. For convenience, sulphate content was transformed into sulphate sulphur concentration (Ss) by stoichiometric calculations. Total iron content of the solution was measured by atomic absorption spectrometer (AAS) which reflects the iron content in hydroxysulphate and oxyhydroxide minerals. Hydroxysulphate minerals usually have low concentrations in waste dumps. However, iron has a low or even no contribution in these minerals. For example, gibbsite, alunite and bluidite contain no iron in their formula, but butlerite and jarosite have maximum 27 and 33 % of iron, respectively. Thus, it can be reasonable to attribute total iron to iron oxyhydroxide minerals (Feo–h). Step 2: diluted nitric acid (HNO3) was added to residue material of step 1 in order to dissolve pyritic sulphur. Assuming that all iron content is in pyritic form, thus, pyrite content was stoichiometrically calculated from the iron concentration. For uniformity, pyrite content was transformed into pyritic sulphur concentration (Spy). 3.3.2 Paste pH Paste pH is a simple and inexpensive method to primarily estimate the presence of reactive carbonate or readily available acidity (Morin and Hutt 1997). It was determined by weighing 50 g of prepared sample and adding 50 ml of distilled water. After mixing for 5 s, the slurry was aged for 10 min. The electrode was then inserted into the slurry, and after stirring slightly, the pH was measured until a stable value was obtained. For convenience, the paste pH will be called p.pH in the following. 3.3.3 Sequential extraction test The selectivity of reagents for sequential extraction test has been a focus of criticism because a wide range of possible secondary phases are associated with waste dumps materials in sulphide deposits (Dold 2003). After reviewing sequential extraction schemes, especially those adapted to the specific mineralogy of porphyry Cu sulphide ores and evaluating the advantages and limitations of each protocol and reagents that were used for each fraction, a seven-step fractionation procedure was selected. This procedure was well developed by Khorasanipour et al. (2011b, c) in soil and sediment environment around the Sarcheshmeh mine. In this procedure, elements separate into seven geochemical fractions, as follows: water soluble, exchangeable, amorphous iron oxyhydroxides, crystalline iron oxides, manganese oxyhydroxides, sulphide and residuals. The performed procedure is listed in Table 1. One gram of air-dried sieved samples was treated successively according to the procedure pointed out in Table 1. After completing each step, the extracted phases were separated from the solid phase by centrifugation at 3,000 rpm for 30 min and the supernatant was then filtered through a 0.45-lm filter (ALBET, Nitrato Celulosa model) and 123 The speciation of cobalt and nickel at mine waste dump Table 1 Sequential extractions procedure applied for mine waste samples fractionation (Khorasanipour et al. 2011b, c) Fractions Procedure Water soluble 1 g sample into 50 ml deionizied H2O, shake for 1 h, at RT Exchangeable 50 ml, 1 M, NH4-acetate pH 5 shake for 2 h, at RT Mn oxyhydroxides 50 ml, 0.1 M NH2OH–HCl pH 2 shake for 2 h Amorphous Fe oxyhydroxides 50 ml, 0.2 M NH4-oxalate pH 3.3 shake for 1 h in darkness, at RT Crystalline Fe oxides 50 ml, 0.2 M NH4-oxalate pH 3.3 heat in water bath 80 °C for 2 h Sulphide Combination of KClO3 and HCl, followed by 4 M HNO3 boiling Residuals HNO3, HF, HClO4, HCl digestion RT room temperature, h hour stored at 4 °C until analysis. Finally, the obtained solution was analysed by ICP at the Central Laboratory of Sarcheshmeh Copper Complex. Reagents used for sequential extraction were of high purity and quality and include NH4-acetate, KClO4, HClO4, HCl and HNO3 (Merck, Darmstadt, Germany), HF (Fluka, AG, CH-9470 packed in Switzerland), NH2OH–HCl (Fluka, A Sigma-Aldrich Company, USA) and NH4-oxalate (manufactured by BDH limited, Poole, England). 3.4 Statistical analysis To perform a proper geochemical characterization of cobalt and nickel at the waste dumps, a correlation analysis was conducted using Pearson’s correlation coefficients. It describes the interaction or the level of linear association among pairs of variables that helps to better understanding of pollution signatures in industrial areas (Akbar et al. 2010). In this method, the data are treated and then interpreted into groups according to degree of correlation coefficients. These groups describe any trends or significant similarity between various variables. To form the correlation matrix, the cobalt- and nickel-containing minerals were selected as the variables. Cobalt and nickel may be present in the structure of coexisting pyrite (Boyle 1974). Hence, Spy that is representative for pyrite content was used in correlation analysis. The most common adsorbents of toxic elements, e.g. cobalt and nickel involve carbonates, organic matter, clay, iron and manganese oxyhydroxide and hydroxysulphate minerals (Kinniburgh et al. 1999; Adriano 2001; Covelo et al. 2007). All above-mentioned minerals were detected in the Sarcheshmeh waste dumps except carbonates and organic matter. Therefore, the correlation analyses was established between cobalt and nickel and observed minerals with source and sink behaviour. After data collection, correlation analysis was performed using SPSS software version 16 for windows. 4 Results and discussion 4.1 Mineralogical and geochemical analyses Primary minerals observed in the studied waste dumps as well as their descriptive statistical characterizations are described in Table 2. 123 123 Table 2 Descriptive statistics of primary minerals in the studied waste dumps (all variables are in terms of wt%) Mineral Quartz Pyrite Muscovite Illite Kaolinite Orthose Albite Chlorite Montmorillonite Epidote Number 60 48 60 56 19 29 48 48 11 5 Minimum 24 2 5 4 4 5 5 5 4 4 Maximum 60 19 23 11 9 9 46 28 15 8 Mean 42.5 7.11 12.57 7.04 6.15 7.28 14.69 11.90 6.36 6.02 SD 8.96 4.37 3.43 1.37 1.19 1.13 9.25 6.59 3.64 1.58 S. Yousefi et al. The speciation of cobalt and nickel at mine waste dump Table 3 Descriptive statistics of secondary minerals in the studied waste dumps (all variables are in terms of wt%) Mineral Butlerite Jarosite Gypsum Carfosiderite Bluidite Alunite Number 12 12 11 1 1 2 Minimum 5 5 5 6 7 3 Maximum 9 9 9 6 7 5 Mean 6.42 6.83 6.17 6 7 4 Std. D 1.16 1.47 1.03 – – 1.41 For all of the materials, the main minerals found by the XRD analysis were quartz (24–60 %) and muscovite (5–23 %) which were presented in all samples. Pyrite (2–19 %), illite (4–7 %), kaolinite (4–9 %), orthose (5–9 %), albite (5–46 %), chlorite (5–28 %), montmorillonite (4–15 %) and epidote (4–8 %) were also found present in some samples. The carbonate content which is the most major neutralizing agent not to be found in the samples. The XRD data and polished sections study also indicate that the main sulphide mineral was pyrite which was accompanied by small amounts of chalcopyrite and magnetite. At the Sarcheshmeh waste dumps, several secondary minerals were detected by XRD in some depths (Table 3). Moreover, amorphous iron oxyhydroxide minerals visually observed in waste dumps that did not detected by XRD due to being negligible and low level of crystallinity. Hence, they were measured in terms of (Feo–h) by ASTM standard test method. The concentrations of cobalt and nickel accompanied by paste pH and ASTM standard test results are shown in Table 4. In this table, cobalt and nickel concentrations are presented for each trench (A, B, C, D, E and F) from surface to bottom of each trench. The name of each sample point contains DS-(trench name) (sample number in the trench) that DS is the abbreviation of depth sampling. As can be seen in Table 4, there are no typical trends of cobalt and nickel variation, pH value and ASTM results with depth. This phenomenon is due to heterogeneity of material from physical and geochemical aspects which lead to the occurrence of different mechanisms related to AMD generation. The separation of these mechanisms is very essential for investigating the fractionation of cobalt and nickel. 4.2 Correlation analysis According to the limits established by Aguilar et al. Aguilar et al. (1999) for trace elements in soil of industrial areas, the concentration of maximum allowable, obligatory investigation and necessary treatment for cobalt (nickel) are 20 (40), 50 (80) and 300 (500) ppm, respectively. Cobalt and nickel concentrations in the Sarcheshmeh waste dumps samples range from 13.8 to 148 and 10 to 163 ppm, respectively (Table 4). Therefore, they exceed the maximum allowed, obligatory investigation thresholds and needs to study about their sources. To assess the important fractions of cobalt and nickel, first the database was configured. The database was comprised of XRD, ASTM D-2492 and ICP data such as the magnitude of clay minerals, iron in oxyhydroxide minerals (Feo–h), manganese, pyritic sulphur (Spy), sulphate sulphur (Ss) and total concentration of nickel and cobalt. In second step, the 123 S. Yousefi et al. Table 4 Chemical properties of the samples collected from paste pH, ICP and ASTM data 123 Sample Depth (m) Paste pH ICP (mg kg-1) ASTM (%) Co Ss Spy Ni Feo–h DS-A1a 0.2 3.92 163 1.84 10.74 3.72 DS-A2 0.5 2.43 55.4 112 1.54 3.49 3.86 DS-A3 0.8 3.16 81.2 89 1.83 7.47 3.40 DS-A4 1.1 3.30 93.4 58 1.62 3.70 4.43 DS-A5 1.5 4.36 146 58 7.06 8.41 2.38 DS-A6 2 3.79 23 29 0.85 0.34 2.71 DS-A7 2.5 3.86 51.2 33 0.75 1.49 2.55 1.70 112 DS-A8 3 4.04 83.5 22 0.50 0.20 DS-B1 0.2 3.13 43 38 1.50 1.19 3.23 DS-B2 0.5 5.93 75.4 44 0.16 1.20 4.84 DS-B3 0.8 6.19 57.3 44 0.22 1.30 4.83 DS-B4 1.1 6.26 64.3 47 0.26 1.19 5.99 DS-B5 1.5 6.40 78 44 0.21 1.30 4.67 DS-B6 2 6.26 70.7 43 0.28 1.71 5.01 DS-B7 2.5 6.00 61.1 44 0.27 1.49 5.67 DS-B8 3 6.06 64.2 50 0.26 1.43 7.02 DS-B12 5 2.88 31 36 1.34 0.80 4.84 DS-C1 0.2 6.00 54.7 40 0.26 1.46 4.31 DS-C2 0.5 6.34 53.8 45 0.20 1.33 4.30 DS-C3 0.8 5.97 46 42 0.21 1.13 3.83 DS-C4 1.1 6.00 69 48 0.23 1.70 5.36 DS-C5 1.5 5.99 62.8 47 0.24 1.55 5.33 DS-C6 2 6.21 56.1 44 0.17 1.32 3.99 DS-C7 2.5 6.08 80.6 60 0.25 1.31 7.81 DS-C8 3 6.30 82.1 65 0.37 1.46 9.31 DS-C9 3.5 6.38 55.6 39 0.21 0.91 4.57 DS-C10 4 6.38 81.7 63 0.37 1.35 8.47 DS-C11 4.5 6.40 69.8 51 0.43 1.65 6.88 DS-C12 5 5.99 62.6 49 0.47 1.37 6.42 DS-D4 1.1 4.24 31.1 28 0.52 1.43 3.78 DS-D5 1.5 5.40 57.3 37 0.51 1.87 4.38 DS-D6 2 5.68 48.7 38 0.47 1.57 4.21 DS-D7 2.5 6.30 44 36 0.33 1.58 4.09 DS-D8 3 3.99 25.6 25 1.08 0.75 4.17 DS-E1 0.2 3.84 44.6 17 1.47 3.05 6.01 DS-E2 0.5 5.97 75.6 22 0.55 1.23 7.93 DS-E3 0.8 4.57 53.4 23 0.49 1.48 4.87 DS-E4 1.1 6.01 53.8 21 0.49 0.78 6.46 DS-E5 1.5 6.68 68.3 21 0.48 0.69 6.73 DS-E7 2.5 5.61 56.1 24 0.31 1.23 6.05 DS-E8 3 6.20 42.9 10 0.42 3.98 3.21 DS-E9 3.5 6.34 41.5 12 0.87 6.10 4.19 The speciation of cobalt and nickel at mine waste dump Table 4 continued Sample Ss sulphate sulphur, Spy pyritic sulphur, Feo–h iron in oxyhydroxide minerals a Sample name is summarized from abbreviation of depth sampling (DS) by adding trench name and sample number Depth (m) Paste pH ICP (mg kg-1) ASTM (%) Co Ni Ss Spy Feo–h DS-E10 4 6.37 40.1 11 0.36 5.48 2.89 DS-E11 4.5 6.39 34 11 0.63 4.82 3.15 DS-E12 5 4.19 31.3 12 0.57 5.23 3.06 DS-E13 5.5 4.42 42.3 13 0.56 5.67 3.13 DS-E14 6 4.13 41 13 0.47 6.05 2.33 DS-E15 6.5 6.17 57.9 11 0.39 6.50 2.46 DS-F1 0.2 2.94 31.3 16 1.26 5.80 3.00 DS-F2 0.5 3.75 47.8 15 0.36 7.31 1.90 DS-F3 0.8 3.72 57.2 15 0.66 6.67 2.13 DS-F4 1.1 3.84 17.8 10 0.45 0.15 2.19 DS-F5 1.5 3.88 13.8 12 0.45 0.23 2.96 DS-F7 2.5 3.63 21.6 13 0.90 0.20 3.39 DS-F9 3.5 6.19 48.6 24 0.37 0.14 7.42 DS-F10 4 5.75 67.8 24 0.37 0.16 7.32 DS-F11 4.5 4.36 41.7 16 0.51 0.13 5.83 DS-F12 5 4.17 26.6 16 0.61 0.18 6.10 DS-F13 5.5 4.90 21.7 17 0.45 0.13 5.08 DS-F14 6 4.17 43.3 23 0.48 0.21 6.77 Table 5 Correlation matrix for all data Co 0.24 Feo–h 0.32 0.67 0.62 0.18 0.19 Ni 0.05 -0.24 -0.60 0.38 0.33 -0.48 -0.39 0.27 0.42 Spys -0.18 -0.02 -0.13 -0.11 -0.17 0.04 Ms -0.05 -0.09 -0.12 0.01 0.07 0.15 0.21 Ill -0.16 -0.05 -0.16 -0.14 0.08 -0.05 0.32 0.19 Kln -0.20 -0.15 -0.32 -0.25 0.02 0.22 0.08 0.04 -0.29 Chl 0.18 0.09 -0.05 0.28 0.30 0.04 -0.22 0.10 -0.13 -0.10 Mn Ss Mnt Feo–h iron in oxyhydroxide minerals Ss sulphate sulphur, Spy pyritic sulphur, Ms muscovite, Ill illite, Kln Kaolinite, Chl chlorite, Mnt Montmorillonite correlation coefficients were dually calculated between all variables. Table 5 gives the correlation coefficient matrix for above-mentioned variables. According to the results, there were insignificant correlation coefficients between cobalt and nickel and other variables. It indicates that the correlation matrix (Table 5) is not effective to speciation of cobalt and nickel with this data structure. This may be due to the interference of different geochemical conditions caused by weathering and relevant geochemical processes such as release, mobility and adsorption of these elements. To solve the problem, it might be reasonable to classify the data based on a criterion which can limit the geochemical 123 S. Yousefi et al. Table 6 Correlation values between cobalt and variables at different p.pHs p.pH Ni Feo–h Mn Ss Spy Ms Ill Kln Chl At all p.pHs 0.62 0.24 0.32 0.05 0.33 -0.18 -0.05 -0.16 -0.20 2 \ p.pH \ 4 0.79 0.18 0.12 0.65 0.77 -0.25 0.20 -0.22 -0.13 0 4 \ p.pH \ 5 0.87 -0.49 -0.29 0.62 0.51 -0.47 -0.24 -0.29 5 \ p.pH \ 6 -0.13 0.66 0.25 0.02 -0.22 0.75 -0.32 0.04 -0.21 6 \ p.pH \ 7 0.8 0.71 0.78 -0.51 -0.66 0.11 -0.21 0.20 -0.49 The significant correlation are highlighted Feo–h iron in oxyhydroxide minerals, Ss sulphate sulphur, Spy pyritic sulphur, Ms muscovite, Ill illite, Kln kaolinite, Chl chlorite, Mnt montmorillonite Table 7 Correlation values between nickel and variables at different p.pHs p.pH Co Feo–h Mn Ss Spy Ms Ill Kln Chl -0.25 At all p.pHs 0.62 0.18 0.19 0.38 0.27 -0.11 0.01 -0.14 2 \ p.pH \ 4 0.79 0.22 0.13 0.71 0.60 -0.16 0.07 -0.52 0.02 4 \ p.pH \ 5 0.87 -0.23 -0.25 0.62 0.39 -0.39 0.03 -0.22 -0.45 5 \ p.pH \ 6 -0.13 -0.55 0.63 -0.44 0.33 0.10 0.25 0.41 -0.51 6 \ p.pH \ 7 0.8 0.71 0.78 -0.59 -0.66 0.04 -0.12 0.25 -0.59 The significant correlation are highlighted Feo–h iron in oxyhydroxide minerals, Ss sulphate sulphur, Spy pyritic sulphur, Ms muscovite, Ill illite, Kln kaolinite, Chl chlorite, Mnt montmorillonite processes that affect the fractionation of cobalt and nickel. It is expected that pH is a suitable criterion for separation of geochemical processes. Hence, the data was categorized based on p.pH experiments, resulting in four classes, and then, the correlation matrix was calculated for each class, separately. The p.pH classes included (2 \ p.pH \ 4), (4 \ p.pH \ 5), (5 \ p.pH \ 6) and (6 \ p.pH \ 7). Correlation values between cobalt and nickel and the variables at different p.pHs classes are shown in Tables 6 and 7. As can be seen in Tables 7 and 8, at p.pHs below 4, cobalt and nickel revealed significant correlations with pyritic sulphur (Spy) (r = 0.77 and 0.60, respectively). It confirms the former investigator’s belief, in Sarcheshmeh area, that proved pyrite is the main source of cobalt and nickel (Aftabi and Atapour 1997, 2007). However, cobalt and nickel indicated only a negligible correlation with Spy at p.pHs higher than 4, while pyrite existed at all p.pH values. The cause of this behaviour of cobalt is schematically shown in Fig. 2. As can be seen in Fig. 2a, at p.pH \ 4, pyrite becomes smaller with a decrease in their concentrations caused by oxidation process. Simultaneously, the released cobalt is removed from the medium due to high mobility at this p.pH. Therefore, a significant correlation between Co and pyrite is detected. At p.pH [ 4 (Fig. 2b), as pyrite shrinks in size, cobalt remains in the medium, and consequently, the correlation coefficient between cobalt and pyrite drops to insignificant values (Table 7). This reason is also valid for nickel which has the pyritic origin. Considering former study in tailing and sediment of Sarcheshmeh area (Khorasanipour et al. 2011a), hydroxysulphate minerals were the main phases which contain cobalt and nickel. In accordance with this fact, cobalt and nickel inhibited significant correlation with 123 The speciation of cobalt and nickel at mine waste dump Table 8 Results obtained from seven operationally defined chemical fractionation methods: F1(water soluble fraction), F2 (Exchangeable fraction), F3 (Mn oxyhydroxides fraction), F4 (amorphous Fe oxyhydroxides fraction), F5 (Crystalline Fe oxide Fraction), F6 (sulphide fraction) and F7 (Residual fraction) Sample DS-B12 DS-F5 DS-E3 DS-E7 p.pH 2 \ p.pH \ 4 2 \ p.pH \ 4 4 \ p.pH \ 5 5 \ p.pH \ 6 Fraction Elements (mg kg-1) and recovery values (%) Co Ni F1 20.5 23.5 F2 BDL BDL F3 BDL BDL F4 BDL BDL F5 BDL BDL F6 5.4 4.2 F7 BDL 2.5 Sum 25.9 30.2 Bulk 31 36 Recovery (%) 84 84 F1 8.5 6.6 F2 BDL BDL F3 BDL BDL F4 BDL BDL F5 BDL BDL F6 3.2 3.4 F7 BDL BDL Sum 11.7 10 Bulk 13.8 12 Recovery (%) 85 83 F1 14 7.4 F2 2.8 BDL F3 BDL BDL F4 BDL BDL F5 BDL BDL F6 19.2 7.9 F7 3 4.5 Sum 39 19.8 Bulk 53.4 23 Recovery (%) 73 86 F1 4 BDL F2 8.7 BDL F3 4 3.5 F4 12 BDL F5 2.1 BDL F6 17.8 15.4 F7 4 2.9 Sum 52.6 21.8 Bulk 56.1 24 Recovery (%) 94 91 123 S. Yousefi et al. Table 8 continued Sample p.pH Fraction Elements (mg kg-1) and recovery values (%) Co DS-B4 BDL blow detection limit 6 \ p.pH \ 7 Ni F1 BDL BDL F2 3.7 BDL F3 18.9 16.4 F4 3 BDL F5 BDL BDL F6 36.2 25.4 F7 BDL BDL Sum 61.8 41.8 Bulk 64.3 47 Recovery (%) 96 89 Fig. 2 Schematic model of pyrite grain representing association between Co and Spy a p.pH \ 4 b p.pH [ 4 Ss at p.pH lower than 5 (Tables 6, 7). Tables 6 and 7 also show the adsorbent role of iron and manganese oxyhydroxide. They control the concentrations of cobalt and nickel, especially manganese oxyhydroxide that indicated highly correlation coefficient at p.pH values greater than 6 (r = 0.78 for cobalt and nickel). This finding is consistent well with the results obtained by Peigneur et al. (1975) and Egozy (1980). They introduced iron and manganese oxyhydroxides as the most important adsorbents of cobalt and nickel. Therefore, one can say that cobalt and nickel bounded to iron and manganese oxyhydroxides at p.pH values between 5 and 7. According to Table 7, cobalt and nickel showed a similar behaviour, at all p.pH values, except between 5 and 6 (r = -0.13). This discrepancy in the behaviour of cobalt and nickel may be due to the adsorption of cobalt on the muscovite surface (r = 0.75) which occurred at this p.pH range. This phenomenon caused to the depletion of cobalt from iron oxyhydroxide surface which led to negatively moderate correlation between Co and Feo–h (r = -0.55). For further enhanced viewing, a summary can be presented for speciation of cobalt and nickel in the Sarcheshmeh waste dumps, as follows: at p.pH lower than 4, hydroxysulphate minerals contained cobalt and nickel. At p.pH ranged between 4 and 5, once again 123 The speciation of cobalt and nickel at mine waste dump hydroxysulphate minerals were involved. At p.pH between 5 and 6, muscovite (as a clay mineral) and iron oxyhydroxide minerals were responsible for fixing cobalt and manganese oxyhydroxide minerals adsorbed nickel. At p.pH between 6 and 7, manganese and iron oxyhydroxide minerals controlled the cobalt and nickel concentration. At all p.pH, pyrite was the main provider of cobalt and nickel in mine waste dump of Sarcheshmeh. 4.3 Validation of the proposed method To verify the applicability of proposed method, the seven-step sequential extraction was carried out for five samples (DS-B12, DS-F5, DS-E3, DS-E7 and DS-B4) relevant to each p.pH range. For quality control, blank samples (containing reagent but no samples) were also taken by using the same reagents in equal quantities as described in the procedure throughout the experiments (Table 1). The analytical precision for each extraction step and the overall procedure were tested by subjecting seven duplicate samples to the applied fractionation procedure. Likewise, the accuracy of the chemical fractionation method could be evaluated from the elemental recovery after whole extraction. The recovery percentage for each element in each waste sample was calculated using Eq. 1: Recoveryð%Þ ¼ ½ðF1 þ F2 þ F3 þ F4 þ F5 þ F6 þ F7Þ=Mtotal   100 ð1Þ where F is the chemical fraction, and Mtotal is the bulk concentration of each element. The extractable concentrations of cobalt and nickel in each fraction are shown in Table 8. The total concentration and recovery values of the elements in samples are also shown in this table. The extracted percentage values of the elements with respect to the sum of the seven fractions are presented in Fig. 3. 4.3.1 P.pH between 2 and 4 Two samples DS-B12 (p.pH = 2.88) and DS-F5 (p.pH = 3.88) were selected for the validation of the proposed method relevant to p.pH between 2 and 4. Chemical fractionation analyses (Fig. 3) showed that cobalt and nickel were dominantly hosted in water soluble fraction. This fraction was incorporated 79.2 % (23.5 mg kg-1) and 72.6 % (6.6 mg kg-1) of cobalt and 77.6 % (23.5 mg kg-1) and 66 % (6.6 mg kg-1) of nickel. The concentration in water soluble fraction is often attributed to the elements associated with hydroxysulphate minerals which are readily soluble and can release metals, acidity and sulphate into the water (Lottermoser 2007). Therefore, it was well consistent with significant correlation between cobalt and nickel, and sulphate sulphur (Ss) at p.pH between 2 and 4 (see Tables 6, 7). The remaining of cobalt and nickel were mainly associated with sulphide minerals (fraction 6). According to Fig. 3 (Table 8), 20.8 % (5.4 mg kg-1) and 27.4 % (3.2 mg kg-1) of cobalt, and 13.9 % (4.2 mg kg-1) and 34 % (3.4 mg kg-1) of nickel, for DS-B12 and DS-F5, respectively, were extracted during this fraction. As mentioned in Sec. 4.1, pyrite was the most abundant sulphide mineral in the studied waste dumps which can be accounted as a major source of releasing cobalt and nickel. This finding verifies the consistency of fractionation and the result of the proposed method which showed significant correlation between cobalt and nickel, and Spy. Nickel was negligibly detected in residual fractions which can be due to re-sorption and re-distribution errors (Gleyzes et al. 2002). 123 S. Yousefi et al. Fig. 3 Percentage of cobalt (a) and nickel (b) extracted in each step of the chemical fractionation procedure F1: water soluble F2: exchangeable F3: Mn oxyhydroxides F4: amorphous Fe oxyhydroxides F5: crystalline Fe oxides F6: sulphide and F7: residuals 4.3.2 P.pH between 4 and 5 A sample, named ‘‘DS-E3’’, was selected to verify the proposed method results, among p.pH between 4 and 5. Under this condition, fractionation analyses results showed that 49.2 % (19.2 mg kg-1) of cobalt and 33.9 % (7.9 mg kg-1) of nickel were originated from sulphide minerals. As discussed earlier in Sec. 4.2 and Fig. 3, the proposed method can only detect the sulphide fraction at p.pH values between 3 and 4, and at other p.pH ranges, the relation between cobalt and nickel by pyritic sulphur (Spy) were missed. Besides sulphide fraction, water soluble fraction contained 35.9 % (14 mg kg-1) of cobalt and 37.4 % (7.4 mg kg-1) of nickel which were in full accordance with significant correlation coefficients between cobalt and sulphate sulphur (Ss) (r = 0.62), and nickel and sulphate sulphur (Ss) (r = 0.62) as pointed out in Table 6 and 7. The above-described p.pH, little amount of cobalt 7.7 % (3 mg kg-1) and nickel 22.7 % (4.5 mg kg-1) were found in residual fraction which can be attributed to resorption and re-distribution errors. A little of cobalt 7.2 % (2.8 mg kg-1) was present as exchangeable fraction. Exchangeable cobalt was not detected in correlation matrix. The reason why this fraction was missed could be because it was not generalized in all samples related to the p.pH between 4 and 5. 4.3.3 P.pH between 5 and 6 Sample ‘‘DS-E7’’ was fractionated to enhance the performance of proposed method at p.pH between 5 and 6. The fractionation pattern of target elements at this p.pH (Fig. 3) exhibited that cobalt was less associated with water soluble (7.6 %, 4 mg kg-1), manganese oxyhydroxides (7.6 %, 4 mg kg-1), crystalline iron oxides (4 %, 2.1 mg kg-1) and residual (7.6 %, 4 mg kg-1) fractions. Figure 3 also showed that nickel has lower incorporation with residual (13.3 %, 2.9 mg kg-1) fraction. Unlike expected, the proposed method was not capable to detect these minor fractions because the contents are negligible and probably not exist in all samples at this p.pH range. Considering Fig. 3, cobalt was significantly accompanied by exchangeable (16.5 %, 8.7 mg kg-1), amorphus iron oxyhydroxides (22.8 %, 12 mg kg-1) and sulphide (33.8 %, 17.8 mg kg-1) fractions. These results can verify the reliability of result obtained by proposed method so that it showed significant correlations between cobalt with muscovite as a absorbent mineral (r = 0.75) and manganese oxyhydroxide minerals (r = 0.66) (Table 6). The dominant fraction for nickel (Fig. 3) involves manganese oxyhydroxides 123 The speciation of cobalt and nickel at mine waste dump (16.1 % 3.5 mg kg-1) and sulphide (70.6 %, 15.4 mg kg-1) fractions. This result was also in accordance with significant correlation coefficient between nickel and manganese (r = 0.63) (Table 7). Like other p.pH domains except p.pH between 2 and 4, sulphide fraction was not detected by proposed method in this p.pH. 4.3.4 P.pH between 6 and 7 To verify the proposed method results at p.pH between 6 and 7, sample ‘‘DS-B4’’ was fractionated by p.pH 6.26 (Table 4). The fractionation scheme (Fig. 3a) revealed that cobalt, in minor content, was accompanied by exchangeable (6 %, 3.7 mg kg-1) and amorphous iron oxyhydroxide (4.9 %, 3 mg kg-1) fractions. Similar to other p.pH ranges, these fractions were missed due to minor content and lack of generality in all samples relevant to the considered p.pH range. The major contents of cobalt and nickel were released by manganese oxyhydroxides (30.6 %, 18.9 mg kg-1 of cobalt and 39.2 %, 16.4 mg kg-1 of nickel) and sulphide (58.6 %, 36.2 mg kg-1 of cobalt and 60.8 %, 25.4 mg kg-1 of nickel) fractions. The presence of cobalt and nickel in manganese oxyhydroxide fraction confirms the reliability of proposed method by showing significant correlation between cobalt and nickel, and manganese. Although, like other p.pH, the correlations between cobalt and nickel, and pyrite (Spy) were not found significant in the proposed method. 5 Conclusions The speciation of cobalt and nickel was well performed using improved correlation analysis. This method provided comprehensive information about speciation of cobalt and nickel in the mine waste dumps. It was so suitable to deal with heterogeneity in mine waste dump which can be substituted instead of sequential extraction test. The methodology considered in this study was simple, fast and more exact (about assign the given mineral in the corresponding fraction) than the sequential extraction test. The results of proposed method indicate that paste pH values allow to control on the speciation patterns of cobalt and nickel so that according to ascending paste pH values, cobalt was related with hydroxysulphate minerals, muscovite and iron and manganese oxyhydroxide minerals, respectively. Likewise, nickel concentration was controlled by hydroxysulphate minerals, manganese and iron oxyhydroxide minerals, by ascending paste pH value of samples. Furthermore, pyrite was the major origin of the cobalt and nickel at all paste pH values. Therefore, the bioavailability of cobalt and nickel directly depended on the stability of corresponding minerals. Consequently, from an environmental point of view, in samples by low paste pH values, hydroxysulphate minerals can be responsible for releasing cobalt and nickel into environment. These results were well consistent with the sequential extraction results especially in the dominant fractions which have generality in all samples of corresponding paste pH range. Acknowledgments The authors appreciate the cooperation of the Research and Development Division of the Sarcheshmeh Copper Complex for financial support and access to sampling and analysis facilities. References Abdul, A. M. (2014). Analysis of environmental pollution in Sundarbans. American Journal of Biomedical and Life Sciences, 2(5), 98–107. 123 S. Yousefi et al. Adepoju-Bello, A. A., Oguntibeju, O. O., Onuegbu, M. T., Ayoola, G. A. A., & Coker, H. A. B. (2014). Analysis of selected metallic impurities in soft drinks marketed in Lagos, Nigeria. African Journal of Biotechnology, 11(20), 4676–4680. Adriano, D. C. (2001). Trace elements in terrestrial environments. Berlin: Springer. Aftabi, A., & Atapour, H. (1997). Geochemical and petrological characteristics of shoshonitic and potassic calcalkaline magmatism at Sarcheshmeh and Dehsiahan porphyry copper deposits. Kerman, Iran, Research Bulletin Isfahan University, 9, 127–156. (in Persian). Aguilar, J., Dorronsoro, C., Fernández, E., Fernández, J., Garcı́a, I., Martı́n, F., et al. (1999). Soil pollution by a pyrite mine spill in Spain: Evolution in time. Environmental Pollution, 132(3), 395–401. Akbal, F., Gürel, L., Bahadır, T., Güler, İ., Bakan, G., & Büyükgüngör, H. (2011). Water and sediment quality assessment in the mid-Black Sea coast of Turkey using multivariate statistical techniques. Environmental Earth Sciences, 64(5), 1387–1395. Akbar Jan, A., Ishaq, M., Ihsanullah, I., & Asim, S. M. (2010). Multivariate statistical analysis of heavy metals pollution in industrial area and its comparison with relatively less polluted area: A case study from the city of Peshawar and district Dir Lower. Journal of Hazardous Material,. doi:10.1016/j. jhazmat.2009.11.073. Akin, B. S., Atıcı, T., Katircioglu, H., & Keskin, F. (2011). Investigation of water quality on Gokcekaya dam lake using multivariate statistical analysis, in Eskisehir, Turkey. Environmental Earth Science, 63, 1251–1261. Alizadegan, A. (2010). Geochemical and mineralogical characteristics of the waste dump, from the economic and environmental aspect in Sarcheshmeh porphyry copper mine, Kerman, Iran, Msc. Thesis, Tehran university, Geology Faculty, Iran, (in Persian). American Society for Testing and Materials (ASTM); (1986). Standard test method for forms of sulphur in coal (D 2492-2484). In Annual book of ASTM standards: gaseous fuels; coal and coke, Sec. 5, vol. 5.05, (pp. 354–358), United States: ASTM International, West Conshohocken. Atapour, H., & Aftabi, A. (2007). The geochemistry of gossans associated with Sarcheshmeh porphyry copper deposit, Rafsanjan, Kerman, Iran: Implications for exploration and the environment. Journal of Geochemical Exploration, 93, 47–65. Barałkiewicz, D., & Siepak, J. (1999). Chromium, nickel and cobalt in environmental samples and existing legal norms. Polish Journal of Environmental Studies, 8(4), 201–208. Boyle, R. W. (1974). The use of major elemental ratios in detailed geochemical prospecting utilizing primary halos. Journal of Geochemical Exploration, 3, 345–369. Bruland, K. W., Donat, J. R., & Hutchins, D. A. (1991). Interactive influences of bioactive trace metals on biological production in oceanic waters. Limnology and Oceanography, 36(8), 1555–1577. Bu, H., Tan, X., Li, S., & Zhang, Q. (2010). Water quality assessment of the Jinshui River (China) using multivariate statistical techniques. Environmental Earth Sciences, 60(8), 1631–1639. Candeias, C., da Silva, E. F., Salgueiro, A. R., Pereira, H. G., Reis, A. P., Patinha, C., et al. (2011). The use of multivariate statistical analysis of geochemical data for assessing the spatial distribution of soil contamination by potentially toxic elements in the Aljustrel mining area (Iberian Pyrite Belt, Portugal). Environmental Earth Sciences, 62(7), 1461–1479. Cempel, M., & Nikel, G. (2006). Nickel: A review of its sources and environmental toxicology. Polish Journal of Environmental Studies, 15(3), 375–382. Chaparro, M. A., Chaparro, M. A., Rajkumar, P., Ramasamy, V., & Sinito, A. M. (2011). Magnetic parameters, trace elements, and multivariate statistical studies of river sediments from southeastern India: A case study from the Vellar River. Environmental Earth Sciences, 63(2), 297–310. Covelo, E. F., Vega, F. A., & Andrade, M. L. (2007). Simultaneous sorption and desorption of Cd, Cr, Cu, Ni, Pb, and Zn in acid soils II. Soil ranking and influence of soil characteristics. Journal of Hazardous Materials, 147, 862–870. DelValls, T. Á., Forja, J. M., & Gómez-Parra, A. (1998). The use of multivariate analysis to link sediment contamination and toxicity data to establish sediment quality guidelines: An example in the Gulf of Cadiz (Spain). Ciencias Marinas, 24(2), 127–154. Dercz, G., Oleszak, D., Prusik, K., & Paja, K. L. (2008). Rietveld-based quantitative analysis of multiphase powders with nanocrystalline Ni Al and Fe Al phases. Reviews on Advanced Materials Science, 18, 764–768. Dold, B. (2003). Speciation of the most soluble phases in a sequential extraction procedure adapted for geochemical studies of copper sulphide mine waste. Journal of Geochemical Exploration, 80, 55–68. Egozy, Y. (1980). Adsorption of cadmium and cobalt montmorillonite as a function of solution composition. Clays and Clay Minerals, 28(4), 311–318. Facchinelli, A., Sacchi, E., & Mallen, L. (2001). Multivariate statistical and GIS-based approach to identify heavy metal sources in soils. Environmental Pollution, 114(3), 313–324. 123 The speciation of cobalt and nickel at mine waste dump Franco-Urı́a, A., López-Mateo, C., Roca, E., & Fernández-Marcos, M. L. (2009). Source identification of heavy metals in pastureland by multivariate analysis in NW Spain. Journal of Hazardous Materials, 165(1), 1008–1015. Gleyzes, C., Tellier, S., & Astruc, M. (2002). Fractionation studies of trace elements in contaminated soils and sediments: A review of sequential extraction procedures. Trends in Analytical Chemistry, 21, 451–467. Hall, G. E. M., Vaive, J. E., Beer, R., & Hoashi, M. (1996). Selective leaches revisited, with emphasis on the amorphous Fe oxyhydroxide phase extraction. Journal of Geochemical Exploration, 56, 59–78. Hamilton, E. I. (1994). The geobiochemistry of cobalt. The Science of the Total Environment, 150, 7–39. Huang, W., Campredon, R., Abrao, J. J., Bernat, M., & Latouche, C. (1994). Variation of heavy metals in recent sediments from Piratininga Lagoon (Brazil): Interpretation of geochemical data with the aid of multivariate analysis. Environmental Geology, 23(4), 241–247. Jambor, I. L. (2003). Mine waste mineralogy and mineralogical perspectives on acid–base accounting. In J. L. Jambor, D. W. Blowes, & A. I. M. Ritchie (Eds.), Environmental aspects of mine wastes: Short course series (Vol. 31, pp. 117–145). Québec: Mineralogical Association of Canada. Khorasanipour, M., & Aftabi, A. (2011). Environmental geochemistry of toxic heavy metals in soils around Sarcheshmeh porphyry copper mine smelter plant, Rafsanjan, Kerman, Iran. Environmental Earth Science, 62, 449–465. doi:10.1007/s12665-010-0539-x. Khorasanipour, M., Tangestani, M. H., & Naseh, R. (2011a). Application of multivariate statistical methods to indicate the origin and geochemical behaviour of potentially hazardous elements in sediment around the Sarcheshmeh copper mine, SE Iran. Environmental Earth Science, 66(2), 589–605. Khorasanipour, M., Tangestani, M. H., Naseh, R., & Hajmohammadi, H. (2011b). Hydrochemistry, mineralogy and chemical fractionation of mine and processing wastes associated with porphyry copper mines: A case study from the Sarcheshmeh mine, SE Iran. Applied Geochemistry, 26, 714–730. Khorasanipour, M., Tangestani, M. H., Naseh, R., & Hajmohammadi, H. (2012). Chemical fractionation and contamination intensity of trace elements in stream sediments at the Sarcheshmeh porphyry copper mine, SE Iran. Mine Water Environment, 31, 199–213. Kinniburgh, D. G., Van Riemsdijk, W. H., Koopal, L. K., Borkovec, M., Benedetti, M. F., & Avena, M. J. (1999). Ion binding to natural organic matter: Competition, heterogeneity, stoichiometry and thermodynamic consistency. Colloids Surfaces A: Physicochemical and Engineering Aspects, 151(1–2), 147–166. Kumar, A., Ramanathan, A. L., Prabha, S., Ranjan, R. K., Ranjan, S., & Singh, G. (2012). Metal speciation studies in the aquifer sediments of Semria Ojhapatti, Bhojpur District, Bihar. Environmental Monitoring and Assessment, 184, 3027–3042. doi:10.1007/s10661-011-2168-6. Leinz, R. W, Sutley, S. J., & Briggs, P. H. (1999). The use of sequential extractions for the chemical speciation of mine wastes, In Tailing and Mine Waste 99- Proceedings of 6th International Conference on Tailings and Mine Waste (pp. 555–561), Colorado: Fort Collins. Leinz, R. W., Sutley, S. J., Desborough, & Briggs, P. H. (2000). An investigation of the partitioning of metals in mine wastes using sequential extractions. In Proceeding from 5th International Conference on Acid Rock Drainage (pp. 343–356). Littleton, Colorado: Society for Mining, Metallurgy, and Exploration. Loska, K., & Wiechuła, D. (2003). Application of principal component analysis for the estimation of source of heavy metal contamination in surface sediments from the Rybnik Reservoir. Chemosphere, 51(8), 723–733. Lottermoser, B. G. (2007). Mine wastes. Characterization, treatment, environmental impacts (2nd ed.). Berlin, Heidelberg: Springer. Martı́nez, J., Llamas, J. F., De Miguel, E., Rey, J., & Hidalgo, M. C. (2008). Soil contamination from urban and industrial activity: Example of the mining district of Linares (southern Spain). Environmental Geology, 54(4), 669–677. McCarty, D. K., Moore, J. N., & Marcus, W. A. (1998). Mineralogy and trace element association in an acid mine drainage iron oxide precipitate; comparison of selective extractions. Applied Geochemistry, 13, 165–176. Morin, K. A., & Hutt, N. M. (1997). Environmental geochemistry of mine site drainage: Practical theory and case studies. Vancouver: MDAG Publishing. Peigneur, P., Maes, A., & Cremers, A. (1975). Heterogeneity of charge density distribution in montmorillonite as inferred from cobalt adsorption. Clays and Clay Minerals, 23, 71–75. Rankama, K., & Sahama, T. H. (1949). Geochemistry. Chicago: The University of Chicago. Rietveld, H. M. (1993). The Rietveld method. London, UK: Oxford University Press. 123 S. Yousefi et al. Soares, H. M. V. M., Boaventura, R. A. R., Machado, A. A. S. C., & Esteves da Silva, J. C. G. (1999). Sediments as monitors of heavy metal contamination in the Ave river basin (Portugal): Multivariate analysis of data. Environmental Pollution, 105(3), 311–323. Soen O., 1964, The occurrence of nickel-arsenides and nickel-antimonide at IgdluÌ nguaq. In the IliÌ maussaq alkaline massif, South Greenland, Copenhagen, Denmark: C. A. Reitzels Forlag, ASIN: B0007J9RLW. Tessier, A., Campbell, P. G., & Bisson, M. (1979). Sequential extraction procedures for the specification of particulate trace metals. Analytical Chemistry, 5, 844–855. Villanueva, U., Raposo, J. C., & Madariaga, M. (2013). A new methodological approach to assess the mobility of As, Cd Co, Cr, Cu, Fe, Ni and Pb in river sediments. Microchemical Journal, 106, 107–120. Violante, A., Huang, P. M., & Gadd, G. M. (Eds.). (2007). Biophysico-chemical processes of heavy metals and metalloids in soil environments. Hoboken: Wiley. Wang, H., & Lu, S. (2011). Spatial distribution, source identification and affecting factors of heavy metals contamination in urban–suburban soils of Lishui city, China. Environmental Earth Sciences, 64(7), 1921–1929. Waterman, G. C., & Hamilton, R. L. (1975). The Sarcheshmeh porphyry copper deposit. Economic Geology, 70, 568–576. Williams, T. M., & Smith, B. (2000). Hydrochemical characterization of acute acid mine drainage at Iron Duke mine Mazowe, Zimbabwe. Environmental Geology, 39, 272–278. Yalcin, M. G., Tumuklu, A., Sonmez, M., & Erdag, D. S. (2010). Application of multivariate statistical approach to identify heavy metal sources in bottom soil of the Seyhan River (Adana), Turkey. Environmental Monitoring and Assessment, 164(1–4), 311–322. Zhang, C., Wu, L., Luo, Y., Zhang, H., & Christie, P. (2008). Identifying sources of soil inorganic pollutants on a regional scale using a multivariate statistical approach: Role of pollutant migration and soil physicochemical properties. Environmental Pollution, 151(3), 470–476. Zhou, J., Ma, D., Pan, J., Nie, W., & Wu, K. (2008). Application of multivariate statistical approach to identify heavy metal sources in sediment and waters: A case study in Yangzhong, China. Environmental Geology, 54(2), 373–380. Zimmerman, A. J., & Weindor, D. (2010). Heavy metal and trace metal analysis in soil by sequential extraction: A review of procedures. International Journal of Analytical Chemistry,. doi:10.1155/2010/ 387803. 123