Simultaneous immobilization of metals and arsenic in acidic polluted soils near a copper smelter in central Chile
- 486 Downloads
Acidic and metal(oid)-rich topsoils resulted after 34 years of continuous operations of a copper smelter in the Puchuncaví valley, central Chile. Currently, large-scale remediation actions for simultaneous in situ immobilization of metals and As are needed to reduce environmental risks of polluted soils. Aided phytostabilization is a cost-effective alternative, but adequate local available soil amendments have to be identified and management options have to be defined.
Materials and methods
Efficacy of seashell grit (SG), biosolids (B), natural zeolite (Z), and iron-activated zeolite (AZ), either alone or in mixtures, was evaluated for reducing metal (Cu and Zn) and As solubilization in polluted soils under laboratory conditions. Perennial ryegrass was used to test phytotoxicity of experimental substrates.
Soil neutralization to a pH of 6.5 with SG, with or without incorporation of AZ, significantly reduces metal (Cu and Zn) solubilization without affecting As solubilization in soil pore water; furthermore, it eliminates phytotoxicity and excessive metal(oid) accumulation in aerial plant tissues. Addition of B or Z to SG-amended soil does not further reduce metal solubilization into soil pore water, but increase As solubilization due to excessive soil neutralization (pH > 6.5); however, no significant As increase occurs in aerial plant tissues.
Simultaneous in situ immobilization of metal(oid) in acidic topsoils is possible through aided phytostabilization.
KeywordsNatural zeolite Biosolids Seashell grit Metal phytotoxicity Ryegrass
Soils surrounding the Ventanas copper smelter in the Puchuncaví valley, central Chile, were heavily degraded due to continuous operation of the Cu concentrate processing facility in the absence of any environmental regulations from 1964 to 1992 (Environmental Resources Management 1993). Elevated emissions of sulfur dioxide (mean of 399 t day−1) to the atmosphere resulted in the occurrence of strong acid rain deposition with acute damage to local vegetation and soil acidification to a pH of 3.5–4.7 (ECUS 1996; Ginocchio 2000). Simultaneously, atmospheric fallout of metal(oid)-rich particulate matter (mean of 12 t day−1) resulted on enrichment of soils with metals (i.e., Cu, Zn, Pb, Cd) and As (ECUS 1996; Ginocchio 2000; Ginocchio et al. 2004). For example, total Cu (the main metal pollutant) and As contents measured in topsoils (0–30 cm depth) reached values up to 1,214 and 120 mg kg−1, respectively (ECUS 1996; González et al. 2008). As a result of the acid rain and metal/As pollution, plant richness and abundance were dramatically reduced, leaving soils exposed to wind and rain erosion. As a result, soils became deficient particularly in terms of both organic matter and macronutrient contents (Ginocchio 2000; Ginocchio et al. 2004).
Even though atmospheric emissions of pollutants from the processing facility were effectively controlled by 1999, several studies have demonstrated that metals/As remains in acidic topsoils (Ginocchio et al. 2004; Neaman et al. 2009). Therefore, in rainy months of the year (winter period), when plant development also occurs, metal solubility and toxicity to plants are expected to occur (Ginocchio 2000; Ginocchio et al. 2006). This may be favored by acidic soil conditions (Sauvé et al. 2000; Ginocchio et al. 2002) and the high percentage (56–67% w/w) of water-soluble salts (i.e., CuSO4, ZnSO4, PbSO4, As2O3) contained in dusts generated by base metal smelters (Skeaff et al. 2011). Indeed, it was demonstrated that soil acidification and the increase in soluble Cu concentration in topsoils explained 10% and 7%, respectively, of the reduction in plant abundance detected toward the Ventanas Cu smelter (Ginocchio 2000). Moreover, a laboratory assay demonstrated that Cu contained in smelter dust is more likely to be bioavailable and, thus, toxic to lettuce (Lactuca sativa L.) plants compared to other massive mine wastes (i.e., smelter slag, tailings) at equivalent total soil Cu concentrations when artificially spiked into agricultural soils (Ginocchio et al. 2006).
Currently, large-scale remediation is needed in the Puchuncaví valley to hold and reverse strong degradation of soils and to reduce environmental risks posed by metals/As accumulated in topsoils. However, this is not an easy task as feasible treatment options for As-polluted soils are limited (Lombi et al. 2004). In situ immobilization of metals in topsoils through aided phytostabilization has been recommended as a cost-effective method of soil remediation (Berti and Cunningham 2000; Mench et al. 2000, 2010). However, the remediation of soils co-polluted by metals and As through in situ aided phytostabilization has been less evaluated (i.e., Kumpiene et al. 2006; Mench et al. 2006; Lee et al. 2011).
In the case of multielemental polluted soils, particularly those co-polluted with metals and As, successful in situ stabilization depends on the combination of critical elements in the soil and the choice of proper amendments (U.S. EPA 2007; Kumpiene et al. 2008). For example, lime and other alkaline materials are known to reduce metal solubility in polluted soils (Lombi et al. 2003; Ruttens et al. 2006; Lopareva-Pohu et al. 2011) while organic matter is known to strongly adsorb Cu cations (i.e., Ruttens et al. 2006; Bes and Mench 2008). However, amelioration of soluble soil metals by increasing soil pH to neutral values may result in increased solubilization of As and potential bioaccumulation of this element through food webs (Sauvé et al. 1997; Adriano 2001). Therefore, appropriate soil amendments and rates of application have to be carefully defined for simultaneous in situ immobilization of metals and As in polluted soils. The U.S. EPA (2007) has suggested to adjust soil pH between 5.5 and 6.5 to decrease metal bioavailability/phytotoxicity in metal-polluted soils when As is also present. Addition of organic matter, medium rates of phosphorous (P) fertilizer, and/or sorbents such as zeolites, charcoal, and high-surface area iron oxides have been also suggested as effective sorbents for both anions and cations (U.S. EPA 2007; Kumpiene et al. 2006, 2008). Therefore, the main objective of the present study was to evaluate, under laboratory controlled conditions, the efficacy of some local available organic (i.e., municipal biosolids) and inorganic (i.e., zeolites and seashell grit) amendments to reduce the solubility of both metals and As, and thus phytoxicity and accumulation into aerial plant tissues, when added into polluted topsoils surrounding the Ventanas copper smelter. Results of the present study will allow the development of a much more focused rehabilitation program at the field scale.
2 Materials and methods
2.1 Study soils
General physicochemical characteristics of study soils (polluted and reference sites) collected at the Puchuncaví valley, central Chile
Los Maitenes (polluted soil)
Reference site (reference soil)
Density (g cm−3)
EC (dS L−1)
CEC (cmol+ kg−1)
Organic matter (%)
Total N (%)
Available N (mg kg−1)
Available P (mg kg−1)
Available K (mg kg−1)
Total Cu (mg kg−1)
Total Zn (mg kg−1)
Total Cd (mg kg−1)
Total Pb (mg kg−1)
Total Fe (%)
Total As (mg kg−1)
To reduce metals and As solubilization in polluted soils of Los Maitenes, a range of local available amendments were selected. Seashell grit (SG) from Sociedad Contractual Minera Aceituno, a natural calcium carbonate-rich material, was used for acidity neutralization as several mines exist in coastal areas of north-central Chile. Laboratory protocols for calculating the acid–base account from polluted study soil were used to determine the appropriate application rate to increase soil pH from 5.6 to 6.5, as suggested by the U.S. EPA (2007) when excess As is also present in a metal-polluted soil. Application rate was 0.5% (w/w), dry weight basis (d.w.).
Class B fresh biosolids (B; 76% of water content) from El Trebal plant of Aguas Andinas S.A. (central Chile), the primary organic solid product yielded by municipal wastewater treatment processes (Haering et al. 2000), were selected as organic amendments as they are readily available at free cost. Fresh biosolids came directly from the centrifuge at the end of the anaerobic stage of the municipal water treatment process. Biopolymers are used for dewatering of biosolids during centrifugation process; no alkaline materials are added into biosolids. Metals having elevated concentrations in selected biosolids were Cu (664 mg kg−1, d.w.) and Zn (1,366 mg kg−1, d.w.) while As levels were 18 mg kg−1 d.w. Application rates were defined considering the maximum rate (50 t ha−1, d.w.) allowed for degraded soils according to the Chilean guidelines and twofold this rate (100 t ha−1, d.w.), as metal-contaminated sites (primarily hard rock mining sites) in the USA have been restored with biosolids added at rates of about 25 to 100 t ha−1 and higher (U.S. EPA 2007).
Natural zeolite (Z), ZORMET® of Andayen Ltda., was used as potential sorbent of excess free cations (i.e., Cu, Zn, Cd, and Pb). It corresponds to a mixture of clinoptilolite, huelandite, and mordenite, with particle size less than 75 μm, and cation exchange capacity of 80–100 cmol+ kg−1. Application rate (10% w/w) of natural zeolite was determined based on previous studies at similar sites in the literature (i.e., Mench et al. 1998; Moirou et al. 2001). Iron-activated zeolite (AZ) has an increased capability of As adsorption when compared to natural zeolite (Onyango et al. 2003); therefore, a batch of the natural zeolite was activated with a solution of 0.2 M of iron sulfate according to the method described by Bonnin (2000). As a result, total Fe content of activated zeolite was five times higher than in natural zeolite (42,061 and 8,248 mg kg−1, respectively). Application rate of Fe-activated zeolite (4.2% w/w) was defined considering the total As concentration (51 mg kg−1) measured in the polluted study soil (Los Maitenes).
2.3 Experimental assay
Experimental treatments to evaluate the efficacy of inorganic (seashell grit, natural zeolite, Fe-activated zeolite) and organic (biosolids) amendments for in situ immobilization of metals and arsenic in polluted topsoils of the Puchuncaví valley, central Chile
Seashell grit (SG) (%)a
Biosolids (B) (%)a
Natural zeolite (Z) (%)a
Fe-activated zeolite (AZ)(%)a
Experimental pots were kept in a plant growth room under controlled conditions (23 ± 2°C; light intensity, 103 ± 14.3 μm s−1 m−2; 12:12-h light to dark photoperiod) during 15 weeks and randomly relocated once a week. The soil moisture of every pot was daily checked, and deionized water was added when necessary (no leaching) to maintain each substrate at 70% field capacity. At the beginning of the fifth week, all pots were sown with 0.6 g of seeds of Lolium perenne L. var. nui (perennial ryegrass), as this species has been frequently used to test metal(oid) toxicity in polluted soils (i.e., Arienzo et al. 2004). Irrigation was increased up to 90% field capacity in those treatments with elevated plant yield by the end of the assay.
During the 1st, 3rd, and 14th week of the assay, aliquots of soil pore water (5–7 mL) were taken from three experimental pots per treatment with Rhizon® Soil Pore Water Samplers (5-cm length; Rhizosphere Research Products, Wageningen, The Netherlands) following the methodology described by Vulkan et al. (2000). A composite sample per experimental treatment and week was produced in order to have enough volume for the analysis. Soil pore water samples were kept in acid-washed polyethylene plastic vials (15 mL) and characterized as described below.
By week 15, aerial plant biomass was harvested per pot. Shoots were washed with deionized water, air-dried at 45°C to a constant weight, and weighed. Shoot Cu and As contents were determined as described below. Pots were then longitudinally cut in halves for root cover determination. A 1 × 1-cm grid was overimposed on the exposed soil surface of one pot half, and the number of intersection points effectively contacted by roots from total intersection points was registered. Root cover (in percentage) was then calculated from the frequency of contacts.
2.4 Analytical determinations of experimental substrates, pore water samples, and plant tissues
For determining general properties of experimental substrates, 1-kg samples were dried in a forced air-drying cabinet at 30°C and then sieved through a 2-mm nylon sieve. Texture, pH, electric conductivity (EC), cation exchange capacity (CEC), organic matter (OM) and available nitrogen (N), P, and potassium (K) were determined after the procedures of the USDA methods manual (USDA 2004).
Samples of experimental substrates and dried ryegrass shoot samples were subjected to microwave acid digestion (1,200 Mega Microwave System, Milestone) and extraction (U.S. EPA method 3051 (U.S. EPA 1995) and USDA method 3050 (USDA 2004)) prior to trace element determinations. Every digestion batch had one blank, one standard reference material sample, one duplicate sample, and one quality control sample. Digested samples were analyzed for total Cu, Zn, As, and Ca contents by flame atomic absorption spectrometry (FAAS, AAnalyst 300; Perkin-Elmer) under U.S. EPA methods (U.S. EPA 1995). The quality assurance and quality control criteria were satisfied when the measured parameter of the standard reference material (Loam-B, catalog no. CRM-LO-B, High-Purity Standard, USA; SRM 1573a tomato leaves or SRM 1570a spinach leaves, National Institute of Standards and Technology, USA) and the quality control sample (a previously characterized soil sample with a known concentration of metals) differed by no more than 5%.
Determinations of pH, EC and free ionic copper (Cu2+) were done in collected pore water samples up to 10 min after sample collection with a combination pH electrode (Sensorex 120C), a conductivimeter (Schott Gerate CG858), and a Cu-ion selective electrode (Orion, model 9629 BN); the Cu-ion specific electrode was calibrated using a diaminoacetic acid solution (Rachou et al. 2007). From Cu2+ values the activity of the free Cu2+ ions or pCu2+ values were calculated according to pCu2+ = −log [Cu2+]. Every pore water sample was then divided into two aliquots and fixed with either a 65% solution of HNO3 Suprapur (Merck) for elemental (Cu, Zn, As, Na, Ca, and Mg) determinations or with concentrated H3PO4 Suprapur (Merck) for dissolved organic carbon (DOC) determinations. DOC was determined by combustion with the U.S. EPA method 415.1 (Keith 1996) in a carbon analyzer (Tekmar Dohrmann Carbon Analyzer, model Apollo 9000). The calibration standard solution used was 1 mg L−1 as carbon of dipotassium phthalate for organic carbon. As reference material ION-20 was used (National Waters Research Institute). Total Cu, Zn, As, Na, Ca, and Mg were determined by FAAS (AAnalyst 300; Perkin-Elmer) under U.S. EPA methods (U.S. EPA 1995), as described above. The sodium adsorption ratio (SAR) was calculated from pore water concentrations of Na, Ca, and Mg (in miliequivalents/liter) according to the equation SAR = Na/√[(Ca + Mg)/2] (USDA 2004). The SAR was calculated to characterize salt-affected soils.
2.5 Statistical analyses
Factorial analysis of variance (ANOVA) was used for testing the effects of amendments and time since amendment addition on pore water chemistry (pH, EC, and pCu2+). Fisher’s least significant difference (LSD) was used as a posteriori test (α = 0.05). One-way ANOVA and Fisher’s LSD (α = 0.05) were used to contrast plant variables (i.e., shoot biomass, root cover, Cu and As content in shoots) under experimental treatments (Zar 1984). Linear simple correlations (Pearson’s r) were used to evaluate statistical relationships among soil chemical variables and plant variables (Draper and Smith 1988).
To elucidate the experimental treatments that allowed proper transformation of pore water chemical parameters of the polluted soil (C−) to those of the reference soil (C+), a principal components and classification analysis (PCCA) was performed (Jongman et al. 1995). Pore water chemical parameters (pH, EC, SAR, DOC, pCu2+, total dissolved Cu, total dissolved Zn, and total dissolved As) determined by week 14 were first reduced to uncorrelated variables, and then both chemical parameters and experimental treatments were classified in order to highlight their relationships.
When needed, data were corrected for non-normality using logarithmic transformations (x = log10(x + 1); Zar 1984) for the valid application of the parametric analyses described above. All statistical analyses were performed on Statistica software, version 6.0 (StatSoft Inc., U.S.A.).
3 Results and discussion
3.1 Physicochemical characteristics of experimental substrates
General physicochemical parameters of experimental substrates at the beginning of the assay (week 1)
FC100% (g water)
PWP (% w.w.)
Available macronutrients (mg/kg)
Total metal (mg/kg)
Addition of biosolids into SG-amended soil increased FC, PWP, OM content, and available levels of N and P in experimental substrates, in a dose-dependent form, but not CEC (Table 3). These results confirm previous findings. Even though biosolids have been used for their liming and metal cation sorbent properties (i.e., Brown et al. 2003; U.S. EPA 2007), they have been most frequently used to provide essential nutrients (such as N and P) and to rebuild soil organic matter content (i.e., Haering et al. 2000; Brown et al. 2003). Benefits directly associated with improved organic matter content in soils are enhanced water infiltration and moisture-holding capability, aggregation, aeration, and nutrient supply for plant growth (i.e., Wallace and Terry 1998; Stofella and Kahn 2001). All of these parameters are relevant for improving the quality of contaminated and degraded soils under semiarid Mediterranean climate that is characteristic of the Puchuncaví valley. Besides these positive effects, biosolids increased EC of experimental substrates, in a dose-dependent form (Table 3). Biosolids have high EC values (14 dS m−1), thus explaining their impact on this parameter when added to soil at high loading rates (i.e., Santibáñez et al. 2007; Verdugo et al. 2011). This effect is aggravated under pot cultivation conditions, where soil salinization problems frequently occur, particularly when no water lixiviation is allowed from experimental pots, as in this study. Therefore, the washout effect of salts due to natural precipitation under field conditions does not occur in laboratory experimental conditions.
Incorporation of zeolites into SGB-amended soils increased CEC of experimental substrates (Table 3), as expected due to its high CEC (80–100 cmol+ kg−1). However, it resulted on excessive increases in PWP values (>19% w.w.), particularly at a rate of 100 t ha−1 (Table 3), which may limit plant growth under semiarid field conditions.
3.2 Pore water chemistry of experimental substrates
Temporal evolution of pH, electric conductivity (EC) and copper activity (pCu2+) in pore water samples taken from experimental substrates
Week × treatment
Pore water chemistry of experimental substrates by the end of the experimental assay (week 14)
Soil acidity neutralization with inorganic liming agents (i.e., CaCO3, Ca(OH)2, CaO) is an effective traditional practice in acidic agricultural soils (Simon 2005). However, industrial alkaline by-products (i.e., fly ash, red mud, furnace slag) have been also considered for neutralization of acidic metal-polluted soils (i.e., U.S. EPA 2007; Kumpiene et al. 2008; Lee et al. 2011). Biosolids may also have soil neutralizing capabilities (Brown et al. 2003; U.S. EPA 2007), but several factors can modify the pH of the substrate in long-term experiments, such as the microbial breakdown of organic matter and the nitrification process, among others. In the present study, application of high rates of biosolids to SG-amended soils resulted in a significant decrease of pH values in pore water in time, which agrees with results reported by Tsadilas et al. (1995) and Santibáñez et al. (2008).
In terms of pore water EC, incorporation of Z, B, or their mixture into SG-amended soils significantly increased this parameter when compared to control soils (C+ and C−; Table 4). In the case of zeolite (SGZ and SGAZZ), EC dramatically decreased from weeks 1 to 9, reaching values below 2 dS m−1 by week 14 and similar to control soils (C+ and C−) and SG-amended soil. On the other hand, even though pore water EC on B- and BZ-amended soils tended to slightly decrease from weeks 1 to 14, this parameter remained significantly high throughout the assay with values above 4 dS m−1 (Table 4). Soil salinization is expected after application of biosolids, particularly at high doses, as biosolids have high EC values (i.e., Brown et al. 2003; Santibáñez et al. 2008; Verdugo et al. 2011).
The pCu2+ on C− was 1.3 to 1.5 times lower than in the C+, thus meaning that activity of free Cu ions (Cu2+) is much higher in C− soil than in C+ soil (Table 4). Furthermore, pCu2+ on control soils (C+ and C−) significantly increased from weeks 1 to 9 and then remained constant until week 14 (Table 4), a phenomenon explained by metal aging in soils (Lock et al. 2006; Rooney et al. 2007), as no leaching occurred from experimental pots. Incorporation of all amendments significantly increased pCu2+ values by week 1 to values similar or even higher than C+, thus effectively and rapidly decreasing the concentration of free Cu2+ ions in pore water (short-term equilibration time). However, they vary in terms of their efficacy to reduce the concentration of free Cu2+ ions in pore water (Table 4), being the order of treatments from low to high efficacy SG = SGAZ < SGZ = SGAZZ = SGB = SGBAZ < SGBZ = SGBAZZ. Addition of SG to C− reduced the activity of free copper ions measured in soil pore water to levels even lower than C+, an expected result as activity of free metal cations decreases as soil pH increases (Bourg 1995). Addition of Z and/or B to SG-amended soils further reduced free Cu2+ ion values, as they have high cation exchange capacities due to negative charged surface (Sauvé et al. 2000; Simon 2005; Janos et al. 2010).
By the end of the assay (week 14), those treatments where Z was added to both SG- and SGB-amended soils resulted in marked increases in the SAR (Table 5) of experimental substrates. The SGZ treatment also increased pore water DOC (Table 5), a phenomenon explained by dissolution of soil OM due to both soil sodification and the slightly alkaline conditions (pore water pH of 7.74 by week 14; Table 4) produced by zeolite. Furthermore, a dose-dependent increase of DOC resulted from incorporation of biosolids to the polluted study soil (Table 5). This finding has been described on biosolid- and other organic-amended soils (i.e., Ruttens et al. 2006) and hard rock mine wastes (i.e., Santibáñez et al. 2008; Verdugo et al. 2011). In some cases, increased amount of soluble organic ligands can increase metal mobility in the substrate (i.e., Santibáñez et al. 2008; Neaman et al. 2009; Verdugo et al. 2011), a phenomenon that was not detected in the present study; indeed, no significant correlations were found among pore water DOC and total dissolved Cu (r = 0.18), total dissolved Zn (r = −0.30), and pCu2+ (r = 0.35).
In the case of total dissolved metals (Cu and Zn), addition of all amendments and their mixtures resulted in reduced concentrations by week 14 when compared to concentrations present in the soil C−. The exception was the treatment SGZ for total dissolved Cu where an increase occurred; this result may be explained by both the elevated SAR values ((25.3 meq L−1)0.5; Table 5) and DOC values reached in this particular treatment. The effective increase in both pore water (Table 4) and soil pH (Table 3) values after amendment additions may account for detected reductions in total dissolved metal concentrations (Table 5) and free Cu2+ ion activities (Table 4) in soil pore water, as largely demonstrated by others when acidic metal-polluted soils or acidic hard rock mine wastes are neutralized (i.e., Bourg 1995; Sauvé et al. 2000; Ginocchio et al. 2002, 2009; Simon 2005; U.S. EPA 2007). In the present study, a significant negative correlation (r = −0.85, P < 0.05) occurred between total dissolved Zn and pH in pore water of experimental treatments while a significant positive correlation was found between pCu2+ and pore water pH (r = 0.95, P < 0.05) by week 14.
By week 14, total dissolved As showed increases from 1.2 to 7.2 times in all amended soils when compared to the soil C−, with the exception of treatments SG and SGAZ (Table 5). This response was significant and positively correlated to soil pore water pH (r = 0.56, P < 0.05) and also occurred in treatments where AZ was incorporated into experimental soil with other amendments. Increased As solubilization has been also demonstrated on amended Cu mine tailings (Verdugo et al. 2011) and on beringite amended As-enriched soils near an As smelter in Belgium (Mench et al. 2006), particularly when substrate neutralization reached pH values higher than 6.5. Therefore, soil neutralization has to be carefully controlled to the target soil pH of 6.5, as suggested by U.S. EPA (2007). In the present study, most treatments reached pHs higher than 7.0 (Table 3).
3.3 Plant responses to experimental substrates
Shoot and root parameters of ryegrass plants showed the same trends to experimental treatments (Fig. 2); indeed, shoot biomass was significantly and positively correlated to root cover (r = 0.92, P < 0.05). Addition of SG, AZ, Z, and B to C− resulted in significant increases in both shoot biomass production and root cover, reaching values even higher (125–280%) than in soil C+ (Fig. 2). The same positive responses on ryegrass productivity have been demonstrated for other inorganic and/or organic amendments incorporated into highly polluted soils of the Puchuncaví valley (i.e., Stuckey et al. 2009; Goecke et al. 2011). Exceptions were treatments SGB50Z, SGB50AZZ, SGB100Z, and SGB100AZZ where no significant differences of shoot productivity and root cover with C− were found (Fig. 2). Soil sodification and elevated contents of both free Cu ions and total dissolved As in pore water samples resulting from addition of Z to SGB-amended soils (Tables 4 and 5) could explain these results. Specifically, SAR values above 12 (meq L−1)0.5, as the values resulting from these experimental treatments, restrict plant growth due to elevated concentrations of exchangeable sodium (Na+ toxicity to plants); besides this, in nonsaline sodic soils, with pH lower than 8.5, Ca2+ and/or Mg2+ concentrations are often nutritionally inadequate (Arshad 2008).
Shoot biomass and root cover were significantly and negatively correlated to shoot Cu concentrations (r = −0.70 and −0.67, respectively; P < 0.05) and to shoot As concentrations (r = −0.60 and −0.53, respectively; P < 0.05). Therefore, experimental treatments that significantly reduced shoot biomass and root cover (C−, SGB50Z, SGB50AZZ, SGB100Z, SGB100AZZ) also showed significant increases in both shoot Cu and As concentrations (Fig. 3), and thus resulted in metal toxicity. This adverse effect was reflected in typical metal toxicity visual symptoms, such as stunted growth with chlorotic and reddish leaves. Shoot Cu and As concentrations of ryegrass plants grown in all the other experimental treatments did not differ from values of the soil C+ and were below 50 and 4 mg kg−1, respectively (Fig. 3).
The typical shoot Cu concentration range in perennial ryegrass is 7.3–13.4 mg kg−1, and 20 mg kg−1is often considered the threshold of excessive shoot Cu content in plants (excluding highly sensitive and tolerant species; Kabata-Pendias and Pendias 1992). However, the majority of plant species can accumulate more than 20 mg kg−1 of Cu when growing in soils with elevated Cu levels (Kabata-Pendias and Pendias 1992). The phytotoxic threshold for shoot Cu concentration in ryegrass is approximately 21–22 mg kg−1 d.w. and is relatively independent of growing conditions (Adriano 2001). In the present study, the nonamended polluted soil (C−) resulted in shoot Cu concentrations higher than 200 mg kg−1 (Fig. 3), a value well above the phytotoxic threshold of 21–22 mg kg−1.
The incorporation of B into SG-amended soils did not resulted in any further significant increase in shoot biomass (Fig. 2), even though available macronutrient contents, among other soils factors, were clearly improved (Table 3). The significant increase in pore water EC values resulting from incorporation of biosolids into SG-amended soil may explain this result. Although soil salinization did not occur (EC values <4 dS m−1, Arshad 2008), perennial ryegrass has been found to be sensitive to soil salinity (Marcar 1987). On the other hand, root cover was significantly reduced by incorporation of B into SG-amended soils (Fig. 2), as expected; this effect was probably due to increased pore water EC values (Table 5) but also due to a change in the “foraging” strategy of the plant root system when exposed to improved availability of macronutrients in soils (Table 4). It is well established that in low-nutritional substrates, such as the soil C−, plants show long and thin roots as resources are mainly allocated to root growth due to the need to “forage” for nutrients (Ginocchio 1994). However, under increased level of nutrients in the substrate, plants may allocate more resources to the development of aerial structures (i.e., Noble and Marshall 1983; Ginocchio 1994), thus showing short and thick roots.
Even though the question is often raised whether results from pot experiments are applicable to events in the field (i.e., Friesl et al. 2006; Mench et al. 2006), the present study adds further knowledge on the importance of not only using proper amendments but also for controlling key chemical characteristics of resulting soils, such as pH, for reaching the goal of simultaneous in situ immobilization of metals and As in polluted soils.
We are aware that results from pot assays are not equivalent to field trials when assessing the efficacy of soil amendments; however, from the results of the present study, we can now plan better field trials in polluted soils of the Puchuncaví valley. Furthermore, long-term field studies will be also needed in order to better assess chemical mechanisms in amended soils under field conditions, ecological consequences, and potential remobilization of metals and As, among other factors.
Effective in situ immobilization of metals and As in polluted soils of the Puchuncaví valley is possible through the incorporation of some specific amendments, at least in the short time (14 weeks) and under controlled laboratory conditions. Specifically, soil neutralization to a pH of approximately 6.5 with seashell grit is enough for reducing metal solubilization without affecting As solubilization into soil pore water. Furthermore, it eliminates phytotoxicity and excessive metal(oid) accumulation in aerial plant tissues, allowing adequate plant yields.
Contrary to expected, addition of iron-activated zeolite to seashell grit-amended soil does not further reduce As solubilization into soil pore water, but it is effective for further reducing As uptake and translocation to aerial plant tissues. However, its costs may be elevated for large-scale field applications. Identification of locally available low-price, iron-rich amendments, such as the fly ash generated by a thermoelectric plant located next to the Ventanas copper smelter, may be a feasible alternative, but this has to be evaluated.
Addition of biosolids and/or natural zeolites to seashell grit-amended soil does not further reduce metal solubilization into soil pore water, at least at the rates used in the present study, but increased As solubilization into soil pore water. However, no significant As increase occurs in aerial plant tissues. Therefore, particular care has to be taken with excessive soil neutralization (pH > 6.5) when using a mixture of rather alkaline materials as amendments, to avoid excessive arsenic solubilization in soil, translocation to aerial plant tissues, and accumulation through food webs.
Biosolids are effective for increasing water retention capability, organic matter content, and available concentrations of N and P of the polluted soil of Los Maitenes, all positive factors for improving plant productivity, particularly under Mediterranean semiarid climate field conditions. However, these positive effects are likely counteracted by the soil salinization produced by this amendment, at least under laboratory conditions (no leaching allowed) and at high doses (100 t ha−1 of biosolids). In the case of natural zeolites, soil sodification may result when incorporated to neutralized soils that have been simultaneously amended with biosolids, limiting plant productivity. Soil salinization problems may, however, be controlled in field trials with excess irrigation.
In conclusion, for simultaneous in situ immobilization of metals and As in polluted soil of the Puchuncaví valley through aided phytostabilization, we suggest neutralization of topsoil to a pH up to 6.5 with an alkaline material, such as seashell grit, and simultaneous addition of biosolids, or any other organic matter-rich material, to improve physical and nutritional limiting conditions of soils. Metal phytotoxicity will be eliminated, and accumulation of metal(oid) in aerial plant tissues will be avoided.
The present study was funded by yhe FONDECYT grant project 1085005. The authors would like to thank Paola Arata of Aguas Andinas S.A., Luis Cerda of Andayem Ltda., and Eduardo Barbieri of Sociedad Contractual Minera Aceituno for providing the amendments.
- Adriano DC (2001) Trace elements in terrestrial environments. Biogeochemistry, bioavailability, and risk of metals. Springer, New YorkGoogle Scholar
- Berti WWR, Cunningham SD (2000) Phytostabilization of metals. In: Raskin I, Ensley BD (eds) Phytoremediation of toxic metals—using plants to clean up the environment. Wiley, New York, pp 71–88Google Scholar
- Bonnin D (2000) Method of removing arsenic species from an aqueous medium using modified zeolite minerals. United States Patent. 10pGoogle Scholar
- Bourg ACM (1995) Speciation of heavy metals in soils and groundwater and implications for their natural and provoked mobility. In: Salomons W, Forstner U, Mader P (eds) Heavy metals. Problems and solutions. Springer, Berlin, pp 19–31Google Scholar
- Draper NR, Smith H (1988) Applied regression analysis. John Wiley, New YorkGoogle Scholar
- ECUS (1996) Trace metal distributions in soils of the Puchuncaví Valley near the Ventanas Copper Smelter, Region V, Chile. Technical Report. ECUS, SheffieldGoogle Scholar
- Environmental Resources Management (1993) Environmental project. Ventanas region, Chile. Vol. 1: final report. Environmental Resources Limited, LondonGoogle Scholar
- Ginocchio R (1994) Morfología dinámica: ¿una estrategia de adquisición de nutrientes en plantas herbáceas perennes? Rev Chil Hist Nat 67:121–127Google Scholar
- Ginocchio R, Sánchez P, de la Fuente LM, Camus I, Bustamante E, Silva Y, Urrestarazu P, Torres JC, Rodríguez PH (2006) Agricultural soils spiked with copper mine wastes and copper concentrate: implications for copper bioavailability and bioaccumulation. Environ Toxicol Chem 25:712–718CrossRefGoogle Scholar
- González I, Muena V, Cisternas M, Neaman A (2008) Acumulación de cobre en una comunidad vegetal afectada por contaminación minera en el valle de Puchuncaví, Chile central. Rev Chil Hist Nat 81:279–291Google Scholar
- Haering KC, Daniels WL, Feagley SE (2000) Reclaiming mined land with biosolids, manures and papermill sludge. In: Barnhisel RI et al (eds) Reclamation of drastically disturbed lands. Monograph #41. American Society of Agronomy, Madison, pp 615–644Google Scholar
- Kabata-Pendias A, Pendias H (1992) Trace elements in soils and plants. CRC, Ann. ArborGoogle Scholar
- Klute A (1986) Methods of soil analysis. Physical and mineralogical methods, vol. 1. American Society of Agronomy, WisconsinGoogle Scholar
- Lopareva-Pohu A, Verdin A, Garçon G, Lounes-Hadj Sahraoui A, Pourrut B, Debiane D, Waterlot C, Laruelle F, Bidar G, Douay F, Shirali P (2011) Influence of fly ash aided phytostabilization of Pb, Cd and Zn highly contaminated soils on Lolium perenne and Trifolium repens metal transfer and physiological stress. Environ Pollut 159:1721–1729CrossRefGoogle Scholar
- Mench M, Vansgrosveld J, Lepp N, Edwards R (1998) Physico-chemical aspects and efficiency of trace element immobilization by soil amendments. In: Vangronsveld J, Cunningham S (eds) Metal-contaminated soils: in situ inactivation and phytorestoration. Springer, Berlin, pp 151–182Google Scholar
- Richards LA (1985) Diagnóstico y rehabilitación de suelos salinos y sódicos. Manual de Agricultura Nº 60. 6ta. Edición. Editorial Limusa, MexicoGoogle Scholar
- U.S. EPA (1995) Test methods for evaluating solid waste: physical/chemical methods, 3rd edn. Office of Solid Waste, The Environmental Protection Agency, Springfield, EPA 530/ SW-846Google Scholar
- U.S. EPA (2007) The use of soil amendments for remediation, revitalization, and reusue. EPA 542-R-07-013. Office of Superfund Remediation and Technology Innovation (OSRTI). EPA/National Service Center for Environmental Publications, CincinnatiGoogle Scholar
- USDA (2004) Soil survey laboratory methods manual. Soil Survey Investigations Report Nº 42, version 4.0, Lincoln: National Soil Survey Center, Natural Resources Conservation Service, United States Department of AgricultureGoogle Scholar
- Wallace A, Terry RE (1998) Handbook of soil conditioners. Substances that enhance the physical properties of soil. Marcel Dekker, New YorKGoogle Scholar
- Zar JH (1984) Biostatistical analysis. Prentice-Hall, Englewood CliffsGoogle Scholar