**1. Introduction**

Mining extraction of heavy metals from sulfidic materials produces considerable levels of potential acidity which eventually, if not prevented and neutralized generate the so-called metalliferous acidic mine drainage, resulting in a potential mobilization of soluble heavy metals [1, 2]. The quantity of acid-forming minerals found in many mines of Central Mexico around the neo-volcanic axis [3, 4] and in Northern Mexico contains dominantly pyrite (FeS2), galena (PbS), sphalerite (ZnS), pyrrhotites (Fe1-xS), chalcopyrite (CuFeS2), arsenopyrite (AsFeS), bornite (Cu5FeS4), and many other metallic sulfosalts [3]. After oxidation, these minerals generate the H+ -producing redox and hydrolytic processes of the components (e.g., S, Fe, Mn, Zn, Cu, Pb, etc.) left behind in the mine tailings. Mine wastes, polluted- and pristine-soils must then be studied and chemically speciated [5], usually by applying simple or multiplesequential extraction procedures [5–8], this to fractionate the HM species according to their expected chemical interaction with the various solid phase compartments present in soils or mining wastes. Based on that, HMs can be grouped into one or several of the following categories: a) water-soluble (free metal ions, M2+, inorganic and organic metal complexes, ML, whether labile or not, etc.); b) exchangeable (non-specifically adsorbed); c) ligand extractable (bioavailable); d) acid-extractable (carbonate-precipitated); e) organically and sulfide bound (oxidizing fraction); f) chemisorbed on Fe-, and Mn oxides (reducible fraction) and g) lattice-retained (occluded or residual). Metal speciation and fractionation to assess any remediation strategy must then rely on finding the right methodology to evaluate which of the extraction procedures may serve as a metal available/toxic fraction indicator [5–8]. This will eventually assure that a metal-sensitive plant can grow once any other undesirable physicochemical characteristics are resolved, such as low pH values, low nutrient status, and high EC). The proposal must then include the addition of ameliorating materials (lime, phosphates, compost, gypsum) that both, help to abate the HM-phytoavailable levels, while serving to neutralize the acidified metal-polluted site, and to mend the growing media, so that plant regrowth will not be impeded. Once these two aspects are guaranteed, results must, all in all, meet the requirements of national or international standards and norms and pass the chosen specific bioassay so that the site can be considered agrostabilized [9–11]. Other necessary physical and chemical characterization procedures shall include the total metal content, [MT], original minerals identification, electrical conductivity, pH, HM-sorption-complexing affinities, etc. [12, 13]. The chapter addresses in Section 2, the relevant chemistry to be considered when researching HMs in mine tailings and polluted soils to chemically characterize wastes and pristine soils. The most commonly chemical speciation and fractionation terms and protocols used are discussed in terms of validating which one best adapts to the specific purpose. In Sections 3 and 4, two cases of study are presented: A) One including an example of how a single extraction procedure was used to study-remediate a Cd-, Cu-, Pb- and Zn- polluted mine waste from "Mina La Negra", at Zimapan, Hidalgo State, in Central Mexico, following a successful application of a methodology to: a) sufficiently abate the levels of phytoavailable-toxic HM-contents of the mine waste by adding lime, gypsum, phosphates or compost; b) ameliorate the waste material to create a "fertile" environment and; c) test a) and b) by applying and passing a well-established bioassay using an HM-sensitive plant, this to assure a successful regrow of vegetation; and B) Another, including a successful application of a sequential extraction protocol to evaluate the effect a spill of a Cu-mine into the Sonora River basin in Northern Sonora State, Mexico, on the bioaccessible levels of metal(loid)s such as Fe, Al, Cu, Mn, and minor amounts of As, Cd, Zn, Pb, and Cr.

### **2. Chemical speciation and extraction procedures**

It is well known that the chemical speciation of heavy metal(loid)s depends not on their total concentrations, but more on the form in which they are found in the environment [5]. Chemical forms depend not only on the reactions that control HM solubility [14–16], and thus on their availability [17–19] and toxicity to plants [20],

but, on their mobility and distribution in the environment [21], which in turn are dominated by their physicochemical interactions (complexation, redox-chemistry, and sorption) [14] with the different solid phase compartments (minerals lattices, oxides, organics, carbonates). Based on this, it is important first to discuss the relevant chemistry of these processes, and secondly, to which extent these will give a good indication of the availability/toxic levels to plants or other living organisms. Because many international standards [22] and national norms [23–28] base their thresholdlegal values on certain metal-extractabilities, and in few cases, on the performances of certain bioassays and biological tests [29–34], a good selection of extracting procedures and pertinent bioassays must be correlated to assure both, the successful remediation strategy followed, and the health of the terrain to sustain vegetation regrowth. Although several research papers and reviews have been published in the past decades regarding the importance of chemical speciation studies and sequential extraction protocols [5–8] to evaluate sediments, soils, biosolids, etc. [5] there is still a strong need to establish a sound methodology, in terms of their applicability, to examine the chemistry of HM-polluted mine wastes. To do so, it is important first then, to present a focused review of the relevant chemical aspects of the interactions implicated among the diverse metal(loid)s present, and the various solid phases found on these wastes.

### **2.1 Chemical speciation; definitions and concepts**

Although the term "heavy metal" could be imprecise and sometimes misleading, as stated by IUPAC [35], over the past four decades, the term "heavy metal" continues to be widely used and applied to a group name of metal(loid)s and that are associated with pollution and/or potential (eco)toxicity. Besides, legal regulations often refer to the term "heavy metal" not only not specifying which metals are included, but under which chemical basis it is assumed that HM and their compounds have (eco)toxic effects, or pose analogous physicochemical, biological, and toxicological properties. Thus, any new name, definition, or classification of metals would be better based on the chemical properties of metals, and such a categorization should reflect our understanding of the chemical basis of reactivity and toxicity, so that their toxic effects can be predicted. More appropriate terms seem to be sound if based on the relevant chemical behavior of the metal, such as those considered by the hard and soft acid and base Pearson's-Lewis theory [36–41], or the "s", "p", "d", and "f" character of metals, according to the orbitals used in the HM-bonding, or on their relevant periodic properties involved in the interaction (electronegativity, ionic potential, ionic radius, metal hydrolysis)[14] that governs their ultimate fate. Other names have been suggested [5–8, 42] presumably being more appropriate, such as potentially toxic elements, PTEs [42]. Yet, although the name heavy metal persists and continues to be used in literature, policies, and regulations, and we will utilize it, in this chapter, in correspondence with the title of the book, and following the general use and acceptance of many researchers in the literature, reference will be made to the chemical aspects of metal reactivity.

### *2.1.1 Chemical, functional, and operational speciation*

Several attempts have been made to clarify unambiguously the term chemical speciation (of heavy metals) in environmental and agricultural sciences [5, 7, 35, 42], especially when used to characterize the relevant chemistry of a specific element that can be toxic or nutrimental to living organisms. However, when a metal is distributed in the environment, several ways are used to describe its behavior in the environment, or its chemical form and activity, among other properties. In the following sections, we will focus not only on the metal chemical forms by itself, in terms of the phases where the metal is distributed (aqueous or solid), but unambiguously on the nature of the specific bonding involved (van der Waals forces, ionic or covalent bonds, inner or outer sphere complexes, etc.), and on the type of compartment that the specific HM occupies, according to its interaction with the solid phase, such as water-soluble, exchangeable (non-specifically adsorbed), chemisorbed (specifically adsorbed), chelate extractable (available), acid- extractable (carbonate-precipitated), organically +sulfide bound (oxidizable), Fe/Mn oxides (reducible), occluded-residual (latticeretained), or even more, on the ability to be taken up, ingested, bioaccumulated, etc., by living organisms. In a document on chemical speciation terminology published by IUPAC [43], it is recommended the term 'chemical species' for describing the form of an element defined as isotopic composition, electronic or oxidation state, and/or complex or molecular structure. This definition has been considered as inappropriate for most studies on solid materials such as soils, sediments, and other materials [5, 42] and transfers the selective extraction procedures relevant to such solid phases to the category of fractionation methodologies. Broader definitions to include soils and mine wastes, for instance, are defined in terms of the so-called functional or operational speciation. In this respect, the term may be better defined to identify, describe, and quantify the amounts of the species forms and phases present in each material [7, 42, 43]. Thus, a general consensus points toward adopting a definition that includes both, forms or phases, so that speciation seems to be better defined as: a) referring to a specific chemical compound or oxidation state in environmental samples, even though this type of speciation in which the precise chemical form of an element is measured is the most difficult to achieve, since very sensitive and selective analytical techniques are required; and; b) functional speciation, for specific usages as for terms like phytoavailable (plant-available species) or bio-accessible (for animals or humans), etc., or c) operational speciation, defined in terms of the extraction procedure utilized to refer to the physical or chemical fraction characterized (water-soluble, acidextractable, etc.) in soils, or mine wastes. Physical procedures to divide samples by particle size, or fractions separated by filtration, centrifugation, or dialysis, etc., are also considered an operational speciation [7, 42, 44], as the distinction between soluble and insoluble species is based on the ability to pass a sample through a 0.45 μm filter. In fact, many operational procedures are often used to fraction metals based on particle size: dissolved (<1 kDa), colloidal (1 kDa-0.45 μm), and particulate (> 0.45 μm).

### *2.1.2 The heavy metal-solid phase interaction: Extraction protocols and fractions*

It has been long recognized that knowledge of both, the elemental composition of the solid materials (sediments, soils, biosolids, mine tailings, etc.), and the total concentration of HMs present in the environment, are of little use in assessing the availability and toxicity to biota. For these reasons, chemical tests have relied more on measurement of extractable or "labile" fractions of these potentially toxic elements. Such tests, however, have provided little basis to relate HM-extractability in mine wastes, to the chemical forms that can be toxic to organisms and deleterious to the environment. To evaluate and plan a remediation strategy for an HM-polluted site, a fundamental understanding of the processes that control heavy metal solubility and availability to biota is needed. Thus, the relevant physicochemical aspects of the

interactions between HM and the solid phases present in the mine waste deposits must be reviewed to successfully correlate their extractability with plant tolerance, so that both, the requirements imposed by standards and norms are fulfilled, and the site can be agriculturally stabilized to allow revegetation. Chemical interactions among HMs and solid phases in unaltered minerals of soils, and in altered mine wastes left behind after ore exploitation, pose and/or create very diverse physical and chemical conditions that influence speciation, such as 1) ionic strength-electrical conductivity of media, 2) presence of dissolved organic matter and complexing ligands, 3) pH and potential acid-forming equilibria, 4) redox potential, Eh-*pe*-values, 5) hard/soft-acid/ base character of the metals (charge, ionic radii, ionic potential, metal hydrolysis, etc.), and 6) reaction kinetics. The combined action of these factors, plus those of the edaphogenetic characteristics of the original soil, the biogeochemical processes occurring, and climate will favor the formation of different metal species, resulting in an innocuous/toxic HM species with a higher or lower bioavailability [45]. Thus, analyzing metal concentrations of aqueous or solid phases is not sufficient to determine its toxicity as biological effects [46], so that, chemical knowledge might provide a more effective diagnosis of environmental conditions [47]. Several biological factors that may influence the bioavailability of metals, include the route of exposure, the mechanism of sequestration and transport of metals by organic ligands, and the exposed organism [48]. According to Rainbow and Luoma [49], in metal ecotoxicology, the term bioavailability, corresponds, both, to the metal that is available for capture by a living organism and is integrated into the metabolic processes, and the fraction of the concentration of the metal that is absorbed and/or adsorbed by the organism. The assimilated fraction may then interact with receptors and physiological sites, causing toxic effects [49]. In the following sections a brief review of the relevant chemistry that will otherwise tilt the balance toward a given single or sequential extraction procedures to speciate HM in mine wastes and soils, will be discussed. Based on Tessier et al. [6], and other studies [5, 42, 50] and protocols [7, 8], the most common soil- and mine waste species and phases to be single or sequentially extracted may include: a) the soil- and mine waste-water-soluble fraction; b) the exchangeable/nonspecifically adsorbed species; c) the acid-extractable (carbonate+specifically chemisorbed species) phase; d) the (oxidizable) organically complexed+sulfide metal species; e) the (reducible) hydrous Fe/Mn oxides fraction; and f) the residualoccluded and strong acid-extractable species.

### **2.2 The water-soluble fraction and solution speciation**

In the soil, sediment, and mine waste solutions, the chemical speciation of heavy metals (HM) must consider the solvation process and complexes formed with organic and inorganic ligands [51]. In solution, ligands can form inner- or outer sphere complexes with metal cations [52]. Information on solution speciation is required for predicting bioavailability because the free metal ion, M2+, is the most toxic species for biota and the most reactive one that interacts with the solid phases [53]. Depending on the metal, the free ion may be the dominant species (Cd2+ (aq), Zn2+ (aq), etc.) or may account for only a minor fraction of the total metal in solution [54], due to the formation of stable metal complexes forms (CuL, PbL, etc.). Metals in soil solution may be present as M2+ (aq), or as kinetically labile/non-labile metal- complexes, with inorganic (Cl�, SO4 <sup>2</sup>�, CO3 <sup>2</sup>�, etc.), or organic ligands, or associated with mineral colloids [52, 53]. Thus, the analytical evaluation of the free metal activity, (M2+)-value, is an important step in the process of chemical speciation of pollutant metals in aqueous

solutions [38, 55–58]. However, in the past, the experimental determination of (M2+) was restricted due to several limitations regarding sensitivity and selectivity of the applied methodologies. Traditionally, the strict evaluation of (M2+) was only possible using solid-state metal ion-selective electrodes (M-ISE), but some other analytical techniques have also been used to estimate 'free ion' concentrations of metals in solution (Donnan dialysis, resin exchange methods, and chromatographic techniques [54, 59], voltammetry (e.g., Anodic Stripping Voltammetry, ASV [56], or Cathodic-Stripping Voltammetry)). However, HM ion fraction determinations, for instance, might become unreliable if concentrations in solution are below the detection limits (e.g., Donnan dialysis, resin exchange methods) necessary for the use of very sensitive techniques such as ASV [55, 56]. Measurements with ion-selective electrodes (ISEs) in natural samples may be affected by the fouling of the electrode by organic matter [60]. The use of ISEs under large Cl� concentrations are also not advisable, and other interferences may occur, resulting in an overestimation of the free metal ion concentrations [61]. Although during the last decades, much progress has been made in reducing the detection limits of ISEs [62]. Most studies to measure trace metal speciation in soil solutions have been on Cu2+, for which detection limits of 10�<sup>13</sup> M or even smaller have been reported [63]. Voltametric techniques, specifically ASV, have been successfully used to determine not only the concentrations of labile metal species from a current measured in solution as a metal is dissolved or released from a Hg-electrode, but the (M2+)-value too. Although analytical problems have been claimed regarding overlapping of stripping peaks, adsorption of surface-active organic compounds on the Hg-surface inhibiting the metal deposition, or formation of insoluble intermetallic compounds that affect peak size and position, other studies showed that these inconveniences can be overcome, so that free Cd2+, Cu2+, Pb2+, and Zn2+ activities [55, 56] can be measured in natural polluted soil solution samples.

### *2.2.1 The free metal activity measurement*

Traditionally, the strict evaluation of (M2+)-values was only possible using solidstate metal ion-selective electrodes (M-ISE), but except perhaps for Cu2+, no M-ISE has adequate sensitivity and specificity for evaluating trace (free) metals in solutions where many metals coexist. It is well known that the M-ISE for Cd2+ and Pb2+, for instance, respond similarly to both metals; hence, Cd and Pb interfere with each other [64, 65]. Aguirre et al. [55, 56] developed a robust method to determine (M2+)-values of Cd, Cu, Pb, and Zn by ASV. The method was tested using metal-buffer solutions to control (M2+), by complexing metals with weak, medium, and strong ligands, and varying pH, total metal aqueous concentration (10�<sup>6</sup> –10�<sup>7</sup> M), metal–ligand ratios of 1:20 M, and 0.010 M acetate medium. For the studied metals, Cd, Cu, Zn, and Pb, the agreement was found among theory, experimental ASV measurements, and (M2+), predicted by using a speciation chemical equilibria program and stability constants reported in the literature. Good agreement was found between the theory and calculated (M2+), and between experimental ASV results and calculated (M2+). Free metal activities in the order of pCd ≤ 12, pCu ≤ 18, pPb ≤ 10, and pZn ≤ 9 were measurable under the established experimental conditions. Results (not shown) also revealed good agreement between Cu-ISE and Cu-ASV when measuring the free Cu2+ activity in aqueous extracts of four soils. Values of soil-(Cu2+) measured were in the order of 10�<sup>5</sup> to 10�<sup>9</sup> . The calibration curves for each metal were prepared according to speciation calculations of the metal–ligand–pH equilibrium systems (M–L–pH) in 0.1 M acetate as an indicator ligand, since fulvic acids in soils contain appreciable amounts of carboxylic compounds

of low molecular weights, such as acetates, oxalates, and citrates, among others. Synthetic solutions of final concentrations of total metal [MT]=1 � <sup>10</sup>�<sup>5</sup> M, total ligand [LT] =2–<sup>4</sup> � <sup>10</sup>�<sup>4</sup> M, and pH values in the 4.0–7.5 range were tested. The calibration curves (ΔEp)c versus log(M2+) were generated, with (ΔEp)c being the displacement of the peak potential due to metal complexation, MLp, and log(M2+) is the logarithm of theoretical chemical activity. Calculations were made with the MINEQL+ program [66, 67], using stability constants reported in the literature [55, 56, 68], corresponding for each of the metals Cd, Cu, Pb, and Zn, in their aqueous free form (M2+). The parameter (ΔEp)c was calculated with eq. (1)

$$\left(\left(\Delta E\_p\right)\_C = \Delta E\_p + \frac{RT}{nF} \ln \frac{i\_{D,ML}}{\Phi\_{ML\_p}} = \frac{RT}{nF} \ln \left(M^{2+}\right)\_b \tag{1}$$

where ΔEp is the observed experimental value of the displacement in half-wave potential due to complexation, i.e. the difference in half-wave potentials of the complex, MLp minus that of the free metal (in acetate), (Ep)c -(Ep)M2+, ф is the sensitivity of the determination, obtained from the linear calibration curve (not shown) of iD,M2+ vs. (M2+)std, (μA per units of chemical activity, (A uaq�<sup>1</sup> ): Substitution of the common R, T, and F values, and converting *ln* to *log*, gives eq. (2):

$$\left(\Delta E\_p\right)\_C = \mathfrak{B}\mathcal{I} \times \log \left(\mathcal{M}^{2+}\right)\_b \tag{2}$$

**Figure 1** shows: a) at upper left, the calibration working curves to estimate the free metal activities of Zn, Cd, Pb and Cu in solution; b) at upper right, the information of the selected ligands and pH used to generate specific levels of (M2+)-values, calculated by Mineql+, for each metal–ligand system, besides the relevant parameters used for calculations of (ΔEp)C; c) at lower left, current-potential curves containing five voltammograms generated for each metal, under the predicted (M2+)-values (from left to right, respectively) for Zn, Cd, Pb, and Cu, of 9.7 � <sup>10</sup>�<sup>6</sup> , 7.1 � <sup>10</sup>�<sup>6</sup> , 8.2 � <sup>10</sup>�<sup>7</sup> , 5.9 � <sup>10</sup>�<sup>7</sup> , and 1.5 � <sup>10</sup>�<sup>7</sup> in 0.01 M acetic acid, pH = 4.5; and d) at lower right, voltammograms of the real samples, [M2+]ac-OM: 20% Soil:80% mine waste solution. From left to right, voltammograms correspond to Zn, Cd, Pb, and Cu, respectively. Free metal activities measured under low (5%) and high (20%) doses of compost added to mixtures (w/w) are presented in red and green, respectively. Levels of metal measured were Zn = 3 � <sup>10</sup>�<sup>6</sup> (red); Cd = 2.4 � <sup>10</sup>�<sup>10</sup> (red); Pb = 2.4 � <sup>10</sup>�<sup>8</sup> (red); and Cu = 3.5 � <sup>10</sup>�<sup>15</sup> (green) and 4.4 � <sup>10</sup>�<sup>7</sup> (red).

### *2.2.2 The total HM soluble fraction*

As mentioned before, an important step in the process of chemical speciation of pollutant metals in aqueous solutions is the analytical evaluation of the (M2+)-value [57, 58]. Although this parameter helps in the assessment and remediation of polluted sites, it is well known that a fraction of the sorbed metals may also contribute to the bioavailable fraction by replenishing into the solution, part of the exhausted ions that plants take up from the solution. Extractable fractions, i.e., the exchangeable and the readily acid-soluble precipitates (e.g., sulfates, carbonates, etc.), can also substantially contribute to the nutrition of plants, as well as a small fraction of the metals nonspecifically adsorbed by organic matter and the Si, Mn, Fe and Al-oxides. Thus, although the free metal ion is the most toxic of metal species, its determination is not the only important one when evaluating the phytoavailable-toxic HM levels. Thus, the

### **Figure 1.**

*a) Upper left linear graphs show (ΔEp)c-log(M2+) calibration curves for Cd, Cu, Pb, and Zn as free metals, (M2+)aq, obtained with eq. 2. (M2+) was calculated using Mineql+; b) upper right columns show calculated log(M2+)-values, measured ASV-ΔE½-values (as conditioned by ligands and pHexp) and log-(ΔEp)c graphs; c) Voltammograms at lower left, show experimental peak-current curves for Zn, Cd, Pb, and Cu, respectively, for five free Zn2+, Cd2+, Pb2+, Cu2 activities, corresponding to 9.7* � *<sup>10</sup>*�*<sup>6</sup> , 7.1* � *<sup>10</sup>*�*<sup>6</sup> , 8.2* � *<sup>10</sup>*�*<sup>7</sup> , 5.9* � *<sup>10</sup>*�*<sup>7</sup> , and 1.5* � *<sup>10</sup>*�*<sup>7</sup> , for each metal, in 0.01 M acetic acid pH 4.5; d) lower right voltammograms show current-potential curves for [M2+]ac-OM-treated mixture 20%-soil:80%-mine waste solutions. Peak currents in red and green represent (M2+)-values measured under low (5% w/w) and high (20%) compost doses added to mixtures. Metal activities were Zn = 3* � *<sup>10</sup>*�*<sup>6</sup> ; Cd = 2.4* � *<sup>10</sup>*�*10; Pb = 2.4* � *<sup>10</sup>*�*<sup>8</sup> ; and Cu = 4.4* � *<sup>10</sup>*�*<sup>7</sup> (in red) and Cu = 3.5* � *<sup>10</sup>*�*<sup>15</sup> (green). For ASV-conditions, see [55, 56].*

exchangeable-, acid-soluble-, and chelate-extractions, must be considered too. Thus, in the water-soluble fraction both, the (M2+)-value, and the total soluble metal must be evaluated when studying HM-polluted mine wastes.

### **2.3 Extraction procedures for solid phase-bound-heavy metals**

One of the most widely used protocols to extract HMs sequentially was proposed by Tessier et al. [6]. Elements were separated into five "operationally" defined fractions: exchangeable; acid-soluble (carbonates); reducible (Fe/Mn oxides); oxidizable (organic matter); and residual. Other authors have referred differently to similar fractions, and even suggesting different order of sequence (chelate extractable, sulfide-associated, etc.), and even modifying concentrations, reaction times, separation procedures, etc. (BCR [7], modified BCR [8]; Geological Society of Canada (GCS)-procedure [69–71]). Based on these, diverse fractions can be visualized to include most HM-containing phases. Although many attempts to unify terms and criteria have been published, the most popular protocols and concepts will be reviewed in terms of chemical relevance to be applied to HM-polluted sites.

### *2.3.1 The exchangeable and chemisorbed fractions*

Heavy metals extracted in the exchangeable fractions comprise both, inner and outer sphere adsorbed species. Whereas the outer sphere weakly adsorbed metal species include those retained on the solid surfaces by relatively weak electrostatic interactions (e.g., van der Waals forces) that can be released by ion-exchange processes, those metals strongly sorbed (chemisorbed and precipitated), are retained covalently by (inner-sphere-) complex interactions. Reagents used for these purposes include mostly the rather strong Mg2+ ion-exchange capacity. The most popular reagents used for these extraction procedure are MgCl2, Mg(NO3)2, CaCl2, Ca(NO3)2, KNO3, KCl, NH4Cl, CH3COONH4, CH3COOH (see Section 2.3.2), and Ba(NO3)2, among others. These reagents do not attack organic matter, silicates, or metal sulfides [6, 72], although some dissolution of carbonates has been reported [6]. Slight decrease in pH has also been reported during the extraction [73], most probably because heavy metals may displace chemisorbed-H<sup>+</sup> ions (salt effect), or polymeric Al-ions [14] which might hydrolyze leading to a partial dissolution of carbonates and manganese oxide fractions [14, 16]. Extraction with acetate salts, particularly NH4OOCCH3, has also been used frequently in soil studies. Divalent cations, in general, are more effective than monovalent cations in ion-exchange processes, but K<sup>+</sup> and NH4 <sup>+</sup> promote the replacement of chemisorbed metal ions in the interlayer exchange sites of some clay minerals (illite and vermiculite). Acetate ions are slightly more stable than chloro-metal-complexes reducing the readsorption and precipitation of the extracted metals and limiting pH variations because of the buffering capacity of the solution [72]. Other reagents showing similar properties have also been used, such as nitrate salts (to avoid complexation) or calcium salts (Ca2+ being sometimes more effective than Mg2+ or NH4 <sup>+</sup> in removing exchangeable ions, but showing precipitation risks with, e.g., sulfates or phosphates). Results obtained with most of these reagents have shown a good correlation with plant uptake [74]. Permanent charge sites of layer silicate clays also retain metal cations by nonspecific electrostatic forces and, in the absence of conditions that would favor metal hydrolysis (e.g., high pH), divalent (M2+) and trivalent (M3+) transition- and HM-cations show typical ion-exchange behavior on layer silicates [14]. Several studies have confirmed this for ions such as Cu2+, Co2+, Ni2+, Mn2+, etc. which retain their inner hydration sphere offering direct support for the involvement of electrostatic forces only [51]. The strength of metal bonding then, should only depend on charge, ionic radii, and hydration properties of the cation. Thus, the ionic radii series for M2+-ions seem to apply:

$$\text{Ca}^{2+} > \text{Sr}^{2+} > \text{Ca}^{2+} > \text{Mg}^{2+} > \text{Hg}^{2+} > \text{Cd}^{2+} > \text{Zn}^{2+} > \text{Mn}^{2+} > \text{Fe}^{2+} > \text{Co}^{2+} > \text{Ni}^{2+} > \text{Cu}^{2+}.$$

Based on this sequence, it results clear the usefulness of using Mg2+-ions [6] to exchange HM-divalent ions from the nonspecific adsorption sites on clay minerals and other solid phases present on soils (Fe, Mn, Al, and Si oxides and organic matter) [50]. Other ions such as K+ [75–77], NH4 <sup>+</sup> [78–81], Ca2+ [78, 82], Ba2+ [76], and even H+ (from CH3COOH, [8]) have also been used with this purpose. However, chances of precipitation of Ca2+ and Ba2+ with specific anions (e.g., CO3 <sup>2</sup>�, PO4 <sup>3</sup>�, SO4 <sup>2</sup>�) must be considered. Increasing the concentration of K<sup>+</sup> and NH4 <sup>+</sup> (to 1 M–2 M), lowering the concentration to 0.01 M for CaCl2, buffering pH, and adding complexing ions for Ba2+ may avoid overestimating this fraction. Transition and HMs in soils, when present at trace levels, are retained largely in non-exchangeable forms [14–16]. Schemes for complete metal extraction require extreme treatments, including the oxidative

degradation of organic matter and dissolution of Fe and Mn oxides [83, 84]. Even the preferential adsorption of polymeric hydroxy-metal cations by layer silicates would not seem to account for the stability of these sorbed form of metals. Hydr-(oxides) of Si, Al, Fe, and Mn, as well as amorphous aluminosilicates offer surface sites for HMchemisorption. According to McBride [14], evidence for the formation of surfacemetal bonds includes; a) a stoichiometry of 2H<sup>+</sup> ions released for each M2+ ion adsorbed [85]; b) a high degree of specificity shown by Al- and Fe-oxides [86], humic substances for some metals; c) changes in the surface charge properties of the oxide as a result of adsorption [38, 58], this last effect attributed to the increased surface positive charge developed by chemisorption. A generally accepted affinity series for the specific adsorption of HMs by solid phases present in soils and sediments relates directly to their increasing ability to form hydroxy complexes (metal hydrolysis). The expected order of adsorption would then be Hg > Pb > Cu> > Zn > Co > Ni > Cd [87]. Whereas this series correlates well, but not identical, with goethite and hematite, however, several authors have reported different affinity sequences [14, 50, 86, 88]. These sequences indicate that oxides and organic fractions adsorb preferentially Pb, Cu, and Zn, as compared with Cd, Ni, and Co [89, 90]. **Tables 1** and **2** show the relative adsorption selectivity of solid phases for metal ions, and metal ions preferences for adsorption, respectively, if based on their chemical properties. Predicted metal affinity sequences based on their chemical properties are shown in **Table 2**.


### **Table 1.**

*Heavy metal affinities for some soil fractions (adapted from Ross [50]).*


### **Table 2.**

*Chemical properties determining metal adsorption selectivity on soils, sediments, and mine wastes solid phases [16, 37].*

**Table 2** shows the expected preferences of adsorption on solid phases of soils, sediments, and mine wastes, of free metal ions, based on the relevant chemical properties that could determine at first instance, the selectivity of solid phases for the metal ions. These are charge, electronegativity, ionic radii [16] which together with charge potential (z/r) or ionic potential (Z<sup>2</sup> /r), and the Pearson's hardness parameter, Hpvalue [36, 39, 91], o Softness Y-value [16, 37] (polarizability and hardness-softness), directly influence the covalent-ionic character of the adsorbed-adsorbate interaction and the relative affinity of adsorption for each metal [92].

### *2.3.2 Acid-soluble fraction*

The acid-soluble fraction attacks mainly acid-active solid phases, releasing HMs such as Mn and Cd, which usually co-precipitate with carbonates. This procedure attacks solid phases that become soluble at pH ≈ 5. A buffered acetic acid/acetate solution is used (0.1–1 M, pH 2–5). The HM fraction recovered under these conditions not only may come from coprecipitates with carbonate minerals but from parts of specifically adsorbed metals on clay surfaces and edges, organic matter, Fe/Mn oxyhydroxides [72], and some sulfosalts of lead, PbSO4 [93], amorphous Fe-sulfides and Fe associated with pyrrhotite [94]. This reagent releases some trace metals remaining on the specifically adsorbed sites that would other way escape the extraction in previous steps [10]. Although large proportions of total Mn are frequently found in these extracts [95], Tessier et al. [6] concluded that Fe2+ and Mn2+ were not coming from a partial attack of FeIII/MnIV oxides but from Mg/Ca carbonate coprecipitates [96, 97], and/or from Mn chemisorbed at calcite surfaces. To get a complete carbonate dissolution, a 0.5 M (pH 4.74) acetate solution can be used [98]. Complexing agents such as EDTA, are used to extract HM ions bound to organic matter too. This acid-soluble extraction procedure if used under sequential extraction protocols should be applied before the oxidation of organic matter [99].

### *2.3.3 The Fe and Mn hydrous oxides: The reducible fraction*

Iron and Mn oxides are excellent HMs-adsorbents. By controlling the reaction Eh and pH, dissolution of metal-oxide phases can be achieved [72]. The most successful reagents to extract the total amount of metal ions associated with these oxides use both, a reducing reagent, and a ligand to retain released ions in a soluble form, the efficacy of the reagent is determined by its reduction potential and the ability to attack Fe and Mn crystalline oxyhydroxides [72]. This dissolution can take place in one to three steps, to separate amorphous and crystalline Mn and Fe-oxides. Hydroxylamine (NH2OH), oxalic acid (H2C2O4), and dithionite (Na2S2O4) are the most used reagents.

a. Hydroxylamine (Eh° = 1.87 V) can dissolve different metal oxides, depending on pH, concentration, extracting time, and temperature. To differentiate the various Fe-oxides, warm NH2OH solutions can be used at pH 2. Acetic acid or HCl is preferred over HNO3 to avoid readsorption problems [100], taking advantage of complexing properties of ions such as Cl� or CH3COO�. A complete dissolution of amorphous Fe-oxides has been reported [101], skipping the attack of the crystalline phases. Other authors preferred NH2OH/CH3COOH solutions for better extraction yields than NH2OH�HCl in HNO3 [102]. Simultaneous extraction of Mn-Fe-oxides can be achieved with 0.02–0.04 M NH2OH solutions in 25% CH3COOH, at high temperatures (96–100°C). Tessier

et al. [6] found total dissolution of Fe-reducible fractions within 6 h. However, the protocol seemed insufficient for Fe extraction [97, 103, 104] when Fe content is high [103, 105], for which an additional Fe-specific step is advised [97, 104, 105]. Total Mn- and amorphous Fe-oxides, and partial dissolution of crystalline Fe-oxides, can be reached at low pH (1.7) and high NH2OH�HCl concentrations (0.25 M).


### *2.3.4 The organic matter and sulfides: The oxidizing fraction*

Heavy metals interact in many forms, not only with organic matter, humified materials, and living organisms, in soils and sediments, but with organic detritus or sulfides of some old mine wastes deposits that may have sustained HM-hyper-accumulating plants [50, 72]. In freshly deposited mine wastes, although the content of

humic substances can be limited, part of the original sulfidic material may remain, so that the levels of oxidizing fraction can be high. Under oxidizing conditions, organic materials and sulfides tend to be degraded, leading to the release of sorbed metals. So, oxidizing reagents such as H2O2 (E° = 1.77 V) or NaClO (E° = 0.90 V), and pyrophosphate ions, are frequently used in fractionation studies to extract HMs associated with organic matter and sulfidic materials. Thus, since some oxidizing agents simultaneously oxidize organic matter and sulfides, this step is more commonly named as the "oxidizing fraction".


### *2.3.5 The residual fraction*

Primary and secondary minerals containing metals in the crystalline lattice constitute the bulk of this fraction. Its destruction is achieved by digestion with strong acids, such as HF, HClO4, HCl, and HNO3.

### **2.4 A note on sequential extraction schemes**

Sequential extraction protocols are very useful experimental tools for special cases where complete characterization and HM speciation studies are required. However, in cases where a single bioavailable-toxic fraction is required, these classical extraction procedures are of less use if applied in sequence. Nevertheless, established methodologies may become more instrumental if used as single extraction methods for evaluating HM-fractions that could correlate well with plant responses when exposed to limiting or excessive concentrations of essential trace (e.g., Cu, Mn, Zn, etc.), or toxic (Pb, Cd, etc.) metals. These HM-fractionation schemes, such as those of Tessier (five steps) [6, 111], BCR (four steps) [5, 7, 112], or modified BCR (three steps) [9] serve more to evaluate the potential mobilization of metals in polluted soils, sediments, and mine wastes, where pH fluctuations, extreme potential leaching conditions, or high-risk assessment studies that might foresee floodings and other effects of severe dispersion vectors that can affect specific environments. To reduce the complexity of the procedures but maintain similar outcomes, a three-step scheme has been proposed for HMpolluted soils, sewage sludge, and for studying sulfur in soils [111, 112] and hence sulfidic mine wastes. This three-step procedure uses: acetic acid (step 1), hydroxylamine (step 2), and hydrogen peroxide (step 3). The scheme was then applied for a certification of a sediment reference material (CRM 601), and that allowed it to be validated [113]. Sequential extraction procedures are applied not without presenting several experimental and theoretical problems, mainly due to the lack of selectivity of reagents [94, 98, 114, 115], readsorption and redistribution of metals during the extraction [94, 98, 116] sample pretreatments [73, 94, 100, 103–105], and general methodological associated methods. Regarding incomplete dissolution of some phases and changes in pH can lead to controversial results regarding readsorption and the redistribution of some metals. Many authors have reported that Cu, Zn, and Pb redistribute on Fe-oxides or on humic substances [94, 117] whereas others [118] stated that redistribution was significant only for high metal contents. Carbonated species of the various metals with different solubilities KsPbCO3 = 10–13.1; KsMnCO3 = 10–9.3; and KsZnCO3 = 10�10, will show incomplete dissolution during this step, and an overestimation of the HMs extracted in the reducing fraction appear too, especially Pb [94, 105, 116] showed the lack of selectivity of these schemes toward sulfides and organic compounds. Extraction of OM by oxidative agents is also unsatisfactory because refractory compounds are not completely destroyed, and sulfides are also oxidized.

From this review, it appears that all the reagents used in the various schemes have advantages and disadvantages and there is no ideal reagent or an ideal protocol for general use. Therefore, the choice of procedure must be related to a definite objective, considering the nature of the sample.

### **2.5 Mining wastes and the functional extraction procedures selected**

Mining of Pb-Zn-Cu ores commonly generates mine wastes rich in Pb, Zn, Cu, and Cd. Some of these tailings contain pyrite-rich materials which produce not only strong

acidity when oxidized (pH values <2) but cause emissions of Zn, Pb, and Cd at levels which can cause adverse effects in terrestrial environments. It has been reported that strongly acidic Zn-rich mine wastes cause severe Zn phytotoxicity [20] and can prevent all plants from surviving on the soil. There is evidence of Zn phytotoxicity, potential Cd risk to humans if tobacco, or edible plants, are grown on contaminated soils, and Pb risk to children, if exposed to road and/or house dust [119]. Although there has been important progress in risk assessment strategies for soil metals, and research on methods to remediate Zn, Cd, and Pb polluted soils and sediments, by in situ treatments or by adding amendments (e.g., phosphates, compost, biochar, biosolids, lime-rich wood-ashes, etc.), which reverse phytotoxicity of Zn and Pb risk [11, 119], there is still a strong need to find sound methodologies to remediate HMpolluted mine wastes. The following sections will present examples of such methodologies to handle this type of HM-polluted terrestrial environment.
