Introduction

Unlike many compounds used in industry, medicine, and agriculture, heavy metals and metalloids cannot be degraded and are an integral part of the natural biogeochemical cycles [1]. Heavy industries such as the metallurgical and petrochemical ones, municipal wastewaters, fertilizers and pesticides, emissions from waste incinerators and cars, mining and construction are some of the anthropogenic activities that are increasing the dissemination and mobilization of heavy metals, inducing a change in their shape and rate of transport among the reservoirs of the ecosphere [2, 3]. These processes significantly contribute for the disruption, degradation, contamination, and pollution of the ecosystems, inducing a serious threat to the environment and public health. In particular, they might cause marginality of soils through the degradation of their quality—inducing the reduction of crop yields and the quality of agricultural products, desertification, and the loss of ecosystem services [4–7].

In order to remove heavy metals from soils and prevent their risks to the environment, animals, and human health, different physical, chemical, and biological approaches have been employed [3]. Conventional methods include excavation and landfill, soil washing, soil flushing, solidification, stabilization of electrokinetic systems, soil incineration, and in situ vitrification [3, 8, 9]. All these methods involve high costs of application, intensive labor, destruction or irreversible changes in soil properties, and disturbance of soil microorganisms. In particular, chemical approaches might also create secondary pollution problems [9, 10].

Phytoremediation, the use of plants and their associated microbes for soil, water, and air decontamination, is a solar-driven and environmentally suitable technology that can be employed for the restoration of soils contaminated with heavy metals [11–14], without compromising the other physical, structural, and chemical characteristics of the soil. This technology is cost-effective and economically attractive when compared with physicochemical technologies [15, 16], and when performed with perennial industrial crops, additional benefits might be generated for the soils under treatment. In fact, perennial crops show suitable characteristics for the phytoremediation process once plants display (a) rapid growth, (b) high biomass yields, (c) deep and extensive root systems, (d) known agronomic techniques, and (e) tolerance to contamination [10, 17, 18]. Besides, the use of perennial energy and fiber crops with genetic potential to tolerate, extract, and/or stabilize heavy metals offer owners and managers the possibility to associate soil decontamination and restoration with the production of biomass for bioenergy, fiber, and other products with economic value [7, 19–21]. Moreover, the production of perennial crops in contaminated soils avoids the land use conflict with food crops [7] and might contribute to the development of companies operating on new and green markets with several social advantages. Additionally, together with the reduction and mitigation of the risk posed by heavy metals for humans and ecosystems, new jobs in the restored land as well as markets for their products might be created in the region.

Miscanthus spp. and Arundo donax L. are lignocellulosic and robust perennial nonfood crops characterized by relatively high yields; greater vegetation cover that helps to prevent wind, water, and biological erosion; and greater performance under different types of soils and ecological conditions including contaminated soils [22–24]. These perennial energy crops show high water and nitrogen efficiencies; low susceptibility to pests and diseases; low need for pesticides and fertilizers; and deep, dense, and extensive root systems that are able to hold the soil and prevent groundwater and surface water contamination by controlling the leaching of contaminants through soil profile [25–28].

Therefore, the aim of this work was to study the tolerance and phytoremediation potential of A. donax L. and Miscanthus spp. to soils contaminated with heavy metals. This study was conducted in the framework of the project Optimization of Perennial Grasses for Biomass Production (OPTIMA, www.optimafp7.eu); EC-funded; and designed to tackle the challenges of establishing optimized production chains for the cultivation, conditioning, and use of perennial grasses on marginal land in the Mediterranean region. Under this umbrella, this research work aimed to study the phytoremediation response of giant reed to soils contaminated with zinc, chromium, or lead and the phytoremediation response of the following three Miscanthus genotypes: the species M. sinensis and M. floridulus and the hybrid M. × giganteus to zinc-contaminated soils, along 2 years. Several studies have already evaluated the effects of giant reed and Miscanthus spp. in soils contaminated with heavy metals. However, most of them cover only the first growing season or are limited in time (e.g., 21 days, 70 days, or 3 months [10]), focusing on the effect of a polimetallic-contaminated soil [25, 29, 30] or on the effect of contamination associated with irrigation with metal-rich waters [31, 32]. Therefore, this study intends to provide more information on the adaptation of giant reed and Miscanthus spp. to soils contaminated primarily from a single element, permitting also the comparison of both perennials for the zinc contamination. By prolonging the study for a 2-year period, the effect of the contamination on a longer term can also provide more insights on the tolerance and phytoremediation potential of A. donax L. and Miscanthus spp.

Materials and Methods

Experimental Setup

The 2-year pot experiment was conducted inside the Campus area of the Faculty of Sciences and Technology of the Universidade NOVA de Lisboa, from where the soil and rhizomes of A. donax L. were collected. The rhizomes of Miscanthus spp. were provided by the University of Catania. Each pot (0.06154 m2) was filled with 12 kg of soil sample (April 2012, control soil). The artificial contamination of soils was prepared by mixing the control soil with contaminated sludges. To test the single effect of Zn, contamination was done with an industrial sludge containing 3.5 % Zn (dry weight basis), provided by the CIRVER-SISAV, a waste treatment company placed in Chamusca. To test the single effect of Cr, contamination was done with an industrial sludge provided by the AUSTRA-CTIC association, placed in Alcanena, presenting an average of 8 g Cr kg−1 (dry weight basis), among other metals. Pb contamination, to test the single effect of this element, was done with an industrial sludge obtained from the “Sociedade Portuguesa do Acumulador Tudor,” a battery manufacturing company located in Castanheira do Ribatejo, near Lisbon, presenting 14 % Pb (dry weight basis). After soil preparation, each pot received two rhizomes (10 cm deep) [25, 33] and fertilization was applied, 3 g N m−2 (urea, 46 % N), 3 g N m−2 (nitrolusal, mixture of NH4NO3 + CaCO3, 27 % N), 17 g K2O m−2 (potassium sulfate, 51 % K2O), and 23 g P2O5 m−2 (superphosphate, 18 % P2O5) for both Arundo and Miscanthus. The urea was applied when plants reached approximately 40–50 cm height. The same NK fertilization was applied in the second year, when plants reached approximately 40–50 cm height, but not P once P fertilizer applied in the first year is enough for the growth of these perennial grasses for at least 10 years [23]. Pots without plants were also prepared to investigate the influence of the soil-biomass system versus soil system in the remediation of the contamination. The following two different concentrations of each heavy metal were tested to study the single effect of each element: (a) Zn 450 and 900 mg Zn kg−1 dry matter, Zn450 and Zn900, respectively; (b) Cr 300 and 600 mg Cr kg−1 dry matter, Cr300 and Cr600, respectively; and (c) Pb 450 and 900 mg Pb kg−1 dry matter, Pb450 and Pb900, respectively. The lowest and the highest concentrations tested correspond to the limit value and to twice the limit value for each element in soil according to the Portuguese Decree-Law [34]. Those thresholds correspond to the total concentration of the element in the soil. The phytoremediation response of Miscanthus spp. (three different Miscanthus genotypes—M. sinensis, M. floridulus, and the hybrid M. × giganteus) was tested on soils contaminated with zinc. The phytoremediation response of giant reed was tested on the single contamination of soils with zinc, chromium, or lead. Full irrigation was applied in all the pots to overcome water stress (950 mm), and each combination of species/heavy metal/level of contamination was independently evaluated in triplicate. At the end of each growing season (December 2012 and December 2013), the plants were harvested and the aerial productivity (g pot−1) and biomass quality (i.e., heavy metal content) were monitored for the December 2013 harvest. At the end of the experiment, belowground biomass (roots and rhizomes) was also collected and analyzed. Percolated waters were also collected at the end of the experiment before rhizome cutting.

Soil Characterization

The top 30-cm soil collected in the Campus area was characterized at the beginning of the experiments. Representative samples of soil were collected, dried in an oven at a temperature between 25 and 30 °C, and then sieved through a 2-mm mesh [35, 36]. Soil samples were analyzed for pH, electrical conductivity, cation exchange capacity (CEC), total organic matter, total N, total P, and extractable phosphates. The pH value (in H2O) was measured by a pH-meter micropH2001 Crison [36]. Electrical conductivity was measured by a conductivity meter MC226 Mettler Toledo [36]. CEC was measured by the Chapman method [37]. The soil organic matter was determined using the Walkley-Black method [38]. Total nitrogen content was measured using the Kjeldahl method [39, 40]. The digested sample obtained by the Kjeldahl method was also used for the determination of the total phosphorous content, measured by spectrophotometry [39]. Extractable phosphates were determined by spectrophotometry on the soil extracts obtained with 0.5 M NaHCO3 [41, 42]. Total metal content (K, Ca, Na, Mg, Zn, Cr, and Pb) of the soil was quantified by atomic absorption spectrometry (SOLAAR AA Spectrometer M Series, Thermo Electron Corporation) on the aqua regia digested samples, according to ISO 11466 [43]. After the artificial contamination of the soils to study the effect of the single contamination of Zn, Cr, and Pb on A. donax and the single contamination of Zn on Miscanthus spp., representative samples of soil were collected from each pot. After drying and sieving, the bioavailable fractions of Zn, Cr, and Pb in the soils were evaluated by atomic absorption spectroscopy on the extracts obtained with 0.05 M EDTA at pH 7.5 [44]. The total element content (Zn, Cr, and Pb) was also determined in the soil samples.

Plant and Percolated Water Analyses

At both harvests, aboveground biomass was separated into leaves, stems and, in the case of M. × giganteus and M. sinensis, also panicles. At the end of the experiment, belowground biomass was also collected and separated into roots and rhizomes. The plant tissues were dried at 70 °C, and the biomass production (aerial and belowground, g pot−1, dry basis) were determined by weight. After, the different plant fractions were ground and stored for subsequent analysis. Plant tissues were mineralized by dry combustion (550 °C), and the metal concentration (Zn, Cr, or Pb, depending on the source of contamination) was determined by atomic absorption spectrometry after dissolving the ash residue with nitric acid [45]. Percolated waters were also collected at the end of the experiment and analyzed in terms of their element content (by atomic absorption spectrometry, on filtered samples).

Mathematic Formulas

The tolerance index (TI) [46, 47],

$$ \mathrm{T}\mathrm{I}=\frac{{\mathrm{dry}\ \mathrm{aerial}\ \mathrm{biomass}\ \mathrm{weight}\ \mathrm{of}\ \mathrm{contaminated}\ \mathrm{plants},\ \mathrm{g}\ \mathrm{pot}}^{-1}}{{\mathrm{dry}\ \mathrm{aerial}\ \mathrm{biomass}\ \mathrm{weight}\ \mathrm{of}\ \mathrm{control}\ \mathrm{plants},\ \mathrm{g}\ \mathrm{pot}}^{-1}}, $$

was used to evaluate the susceptibility of the plants to the heavy metal contamination.

The modified accumulation index (mAI) [25],

$$ \mathrm{m}\mathrm{A}\mathrm{I}=\frac{{\mathrm{metal}\ \mathrm{accumulation}\ \mathrm{in}\ \mathrm{the}\ \mathrm{contaminated}\ \mathrm{plants},\ \mathrm{mg}\ \mathrm{pot}}^{-1}}{{\mathrm{metal}\ \mathrm{accumulation}\ \mathrm{in}\ \mathrm{the}\ \mathrm{control}\ \mathrm{plants},\ \mathrm{mg}\ \mathrm{pot}}^{-1}}, $$

was used to evaluate the ability of the plants to take up and accumulate a given element in larger quantities than customary.

The modified bioconcentration factor (mBCF) [25, 29, 48],

$$ \mathrm{mBCF}=\frac{{\mathrm{metal}\ \mathrm{concentration}\ \mathrm{in}\ \mathrm{the}\ \mathrm{plant}\ \mathrm{fraction},\ \mathrm{mg}\ \mathrm{kg}}^{-1}}{{\mathrm{bioavailable}\ \mathrm{metal}\ \mathrm{concentration}\ \mathrm{in}\ \mathrm{the}\ \mathrm{soil},\ \mathrm{mg}\ \mathrm{kg}}^{-1}}, $$

was used to evaluate the ability of the plants to take up and accumulate the metals into the aerial or belowground fraction of the biomass. The metal bioavailable content in the soil, determined through extraction with EDTA, represents the amount of contaminant potentially bioavailable for plant uptake. Thus, mBCF calculated can more realistically represent the capacity of metal transfer to plants [25, 29].

The modified bioaccumulation factor (mBAF) [25],

$$ \mathrm{mBAF}\left(\%\right)=\frac{{\mathrm{metal}\ \mathrm{accumulation}\ \mathrm{in}\ \mathrm{the}\ \mathrm{plant}\ \mathrm{fraction},\ \mathrm{mg}\ \mathrm{pot}}^{-1}}{{\mathrm{bioavailable}\ \mathrm{metal}\ \mathrm{content}\ \mathrm{in}\ \mathrm{the}\ \mathrm{soil},\ \mathrm{mg}\ \mathrm{pot}}^{-1}}\times 100, $$

was used to evaluate the ability of the plants to remove and remediate the metal contamination in the soil.

The translocation factor (TF) and the modified translocation factor (mTF),

$$ \mathrm{T}\mathrm{F}=\frac{{\mathrm{metal}\ \mathrm{concentration}\ \mathrm{in}\ \mathrm{the}\ \mathrm{aboveground}\ \mathrm{plant}\ \mathrm{fraction},\ \mathrm{mg}\ \mathrm{kg}}^{-1}}{{\mathrm{metal}\ \mathrm{concentration}\ \mathrm{in}\ \mathrm{the}\ \mathrm{belowground}\ \mathrm{plant}\ \mathrm{fraction},\ \mathrm{mg}\ \mathrm{kg}}^{-1}}, $$
$$ \mathrm{m}\mathrm{T}\mathrm{F}=\frac{{\mathrm{metal}\ \mathrm{accumulation}\ \mathrm{in}\ \mathrm{the}\ \mathrm{aboveground}\ \mathrm{plant}\ \mathrm{fraction},\ \mathrm{mg}\ \mathrm{pot}}^{-1}}{{\mathrm{metal}\ \mathrm{accumulation}\ \mathrm{in}\ \mathrm{the}\ \mathrm{belowground}\ \mathrm{plant}\ \mathrm{fraction},\ \mathrm{mg}\ \mathrm{pot}}^{-1}}, $$

were used to evaluate the ability of the plant to translocate the metals to the harvestable aerial part. The calculus of TF was based on the method of Mattina et al. [49], and the calculus of mTF was based on the modifications introduced by Fernando [25], combining the biomass production with the biomass metal concentration. Plants with mAIs, mBCFs, and TFs greater than one (>1) have the potential to be used in phytoextraction [25].

Statistical Analysis

The statistical interpretation of the results was performed using analysis of variance (one-way and multifactorial ANOVA) (Statistica 6.0 program). LSD Fisher’s test was applied to separate means when ANOVA revealed significant differences. The results were presented as the mean ± standard deviation of triplicate analysis. The uncertainties of the TI and mAI results were calculated by the propagation of the deviation obtained in contaminated plants and control plants.

Results and Discussion

Soil Characterization

The results of the physical and chemical properties of the soil collected in the Campus and used in the experiments, before sludge addition, were depicted in Table 1. The soil used for the experiment was poor in both organic matter and basic chemical elements, such as N or P. The pH was slightly alkaline, and the concentration of zinc, chromium, and lead found in soil samples was low. Artificial heavy metal contamination was built from these initial properties. Table 2 shows the total and bioavailable content of Zn, Cr, and Pb in the control and artificial soils. According to Iqbal et al. [44], the extraction with 0.05 M EDTA can be used to assess the bioavailabilities of contaminants in soil and soil materials. Results obtained at the beginning of the experiment show that on the artificial contaminated soils, the bioavailable fractions of Zn, Cr, and Pb were, respectively, 36–39, 22–24, and 41–55 % of the total element content in the soils. Accordingly, these percentages reflect the amount of elements that may be absorbed or metabolized by ecological receptors or available for interaction with biological systems.

Table 1 Physical and chemical properties of soil collected in the Campus and used in the experiments, before artificial contamination
Full size table
Table 2 Control and artificial soil characterization
Full size table

Effects of Heavy Metals on the Biomass Production of A. donax and Miscanthus spp.

Yield performances of A. donax biomass grown in Zn-, Cr-, and Pb-contaminated soils and Miscanthus genotypes grown on Zn-contaminated soils are shown in Fig. 1. Results obtained for A. donax (Fig. 1a) showed that the different concentrations of Zn, Cr, and Pb on soils reduced biomass aerial production. However, the effect of Zn and Pb contamination was not significant (p > 0.05), as also the effect of the Cr300 soils. Giant reed showed similar aerial yield responses on Zn- and Pb-contaminated soils and under both tested concentrations. Similar biomass aerial production to Zn and Pb assays was also obtained in the Cr300 soils. In contrast, the higher Cr concentration in the Cr600 soils led to significant lower yields. The different Miscanthus genotypes showed different yield performances under the exposure to 450 and 900 mg Zn kg−1 in soil (Fig. 1b). M. × giganteus showed significantly higher yields than M. sinensis and M. floridulus. Yet, the reduction in yields due to the zinc toxicity effect was significant in this genotype. Contrarily, no significant effects due to zinc contamination were observed in M. sinensis and M. floridulus (even if in Zn900 soils, M. floridulus presents a trend for lower biomass production). M. sinensis showed the lowest yields, although differences to M. floridulus were not significant. A side by side comparison of A. donax in zinc-contaminated soils to Miscanthus spp. indicates that M. × giganteus was significantly (p < 0.05) more productive than giant reed and giant reed significantly more productive than M. floridulus and M. sinensis.

Fig. 1
figure 1

Aerial biomass production of Arundo donax L. in Zn-, Cr-, and Pb-contaminated soils (a) and aerial biomass production of Miscanthus spp. under Zn-contaminated soils (b). For each species, different lowercase letters indicate statistical significance (p < 0.05) between treatments

Full size image

Several authors reported tolerance to heavy metal contamination, namely, to Cd and Ni [24] and to Pb [50], by giant reed. However, the tolerance to heavy metal exposure is also governed by the element concentration in soil. Guo and Miao [50] observed only phytotoxic effects on giant reed when Pb concentration in soil was higher than 1000 mg kg−1. Below this concentration, no effects were observed on the shoot dry biomass production. Regarding M. × giganteus, Fernando and Oliveira [31] did not observed phytotoxic effects on M. × giganteus derived from zinc contamination. Nevertheless, the zinc concentrations tested by those authors were lower (150 and 300 mg Zn kg−1) than the ones tested in this work.

To evaluate the susceptibility of the plants to the heavy metal contamination, TI was calculated based on the plant’s aerial biomass production (Table 3). According to the results presented, the exposure of plants to the levels of heavy metals tested resulted in the reduction of biomass productivity and the TI was always below 1 (<1). According to Kabata-Pendias [3], Zn, Cr, and Pb in excess in soils contribute to damages on the photosynthesis apparatus and belowground organs, leading to retarded growth of the entire plant. Highest effects on the loss of biomass production, expressed as TI, were observed due to the Cr contamination in giant reed (losses of 30–40 %) and to the Zn contamination in M. sinensis (Zn450 induced a 30 % yield reduction) and M. floridulus (Zn900 induced a 40 % yield reduction). The remaining assays showed a yield loss of 20 %, except the Zn450 effect on M. floridulus, which presented only a 10 % loss in biomass production. However, in spite of the observed biomass production reduction, results show that both perennials are tolerant to the heavy metal contamination, once TI remains similar with increasing contamination (except M. floridulus). Anatomical changes are mechanisms that help to quench direct ionic metal toxicity, such as development of secondary sheath bundles and tissue scarification [50]. Yet, information on the interactions between the belowground organs of these perennials and its growing medium is still lacking, which might provide more insights on the tolerance mechanisms associated with these plants on contaminated soils.

Table 3 Tolerance index (TI), modified accumulation index (mAI), modified bioconcentration factor (mBCF), modified bioaccumulation factor (mBAF), translocation factor (TF), and modified translocation factor (mTF) of Arundo donax and Miscanthus spp. under heavy metal-contaminated soils
Full size table

Heavy Metal Concentrations in A. donax and Miscanthus Genotypes

Table 4 presents the heavy metal concentrations in the aboveground and belowground fractions of A. donax and Miscanthus spp. For all the species and fractions of the plant, differences were statistically significant between treatments, and higher concentration was obtained with increasing levels of metals in the soil. The same behavior for both perennials was also observed by several authors as reviewed in the works of Nsanganwimana et al. [10, 51]. However, the distribution between aboveground organs and belowground organs differs depending on the studied element. Regarding zinc concentration in plant organs, no significant differences (p > 0.05) were observed between above and belowground fractions, either for giant reed and Miscanthus genotypes. This indicates that zinc is easily mobilized within the plant, being transported from the belowground organs to the aboveground organs. Kabata-Pendias [3] indicates that this mobility can be associated with the bounding of Zn to light organic compounds in xylem fluids. According to Kabata-Pendias [3], normal zinc content in herbaceous plants ranges between 6 and 126 mg Zn/kg (dry matter). Total Zn content of the biomass in the different species lies in the range referred by Kabata-Pendias [3]. No significant differences (p < 0.05) were also observed among species in terms of zinc concentration, indicating that the accumulation behavior was similar. Regarding Cr and Pb, in the artificially contaminated soils, Cr and Pb concentrations in the belowground fraction were higher than in the aerial fraction. Those results are also consistent to other studies that relate the translocation of Pb and Cr from belowground organs to the aerial fraction as limited [3]. Consequently, both metals concentrated mainly in the belowground organs, apparently because of the propensity of Cr3+ and Pb pyrophosphate to bind to cell walls [3]. Yet, the Cr and Pb concentrations in both aerial and belowground fractions of giant reed from contaminated pots are similar or higher than what is commonly found in plants [3], thus showing ability from giant reed to accumulate those metals.

Table 4 Heavy metal concentration (mg kg−1, DW) in the aboveground and belowground fraction of Arundo donax and Miscanthus spp
Full size table

Bioaccumulation of Heavy Metals by A. donax and Miscanthus Genotypes

Bioaccumulation of zinc, chromium, and lead in the root, rhizome, leaves, and stems of A. donax was depicted in Fig. 2a. The increment of the Zn, Cr, and Pb concentrations on soil increased their accumulation in plant components, mainly in the hypogeal part, especially Cr and Pb.

Fig. 2
figure 2

Total metal accumulated in Arundo donax L. under contaminates soils (a) and in Miscanthus spp. under Zn-contaminated soils (b). In Fig. 2a experiment, different lowercase letters indicate statistical significance (p < 0.05) between treatments for each element studied. In Fig. 2b experiment, different lowercase letters indicate statistical significance (p < 0.05) between treatments for each genotype. Error bars indicate the standard deviation associated with the sum of the metal accumulated in the different biomass organs (n = 3)

Full size image

The ability to accumulate metals in the aboveground organs is a positive indicator of the potential capacity to a certain plant species to serve as a phytoremediation plant [52]. Our results show that the accumulation occurs mainly in the roots and rhizome. Chromium is slightly available to plants and not easily translocated for aerial plant components; consequently, it is concentrated mainly in roots [3]. Our study confirms the lower chromium accumulation by the plant (compared with zinc) as well as lower translocation for aerial components. The same was observed for Pb. Although Pb occurs naturally in all plants, it seems that it does not play any essential role in their metabolism, and if it is necessary for plants, its concentration level of 2–6 μg/kg should be sufficient. Normally, a small percentage of its content in soil solution is available to plants and its translocation from roots to tops is greatly limited [3]. The values obtained for Pb bioaccumulation are the lowest of the heavy metals tested in this experiment. Regarding Zn accumulation by giant reed, our study confirms that Zn is easily available to plants and easily transported to the aerial components. Higher accumulation was observed for this metal, compared to Cr and Pb, and higher accumulation was also observed in the aerial organs of the plant.

Bioaccumulation of zinc in the root, rhizome, panicles, leaves, and stems of Miscanthus spp. was depicted in Fig. 2b. As observed for giant reed, the increment of the Zn concentrations on soil increased their accumulation in plant components, mainly in the hypogeal part (Fig. 2b). However, the differences in mg pot−1 among species reflect the differences in biomass production, once concentration of Zn was similar for the different Miscanthus genotypes. Higher yield performances were the main vehicle for higher Zn phytoextraction. Higher Zn accumulation was observed for M. × giganteus, lower accumulation was observed for M. sinensis, and M. floridulus in between. If we compare the accumulation of Zn between giant reed and M. × giganteus, the higher accumulation observed for the Miscanthus genotype is also a result of the higher biomass production (Fig. 2a, b). According to Leung [53], the species under study can be considered as indicators once; according to Fig. 2, the contaminant accumulation in their tissues reflect metal level in soil.

mAI, mBCF, mBAF, and TF for A. donax and Miscanthus Genotypes

But the main question of this work is which is, in practice, the potential of both perennials for phytoremediation of heavy metal-contaminated soils?

The phytoremediation potential of these grasses can be associated with the capacity of the plants to adsorb on the belowground organs, the heavy metals in the soil, stabilizing and immobilizing the contaminant, that otherwise could be leached. But, the phytoremediation potential can be also associated with the release of exudates and enzymes to soil by the rhizosphere that can stimulate the remediation of pollutants, by the stabilization/immobilization of metals. In effect, the radicular fraction of the biomass releases CO2 to the soil, which, dissolved in the water solution of the soil and in alkaline pH, can precipitate the metal, as carbonates [25]. The phytoremediation potential of these grasses can be also associated with the stimulation of the stabilization/immobilization of metals by fungi or other microorganisms in the soil-rhizome-root interface [25, 53]. However, in terms of treating contaminated soils, the phytoextraction perspective is the most interesting, both from an environmental point of view and from an economic standpoint. Indeed, the processing of the extractor plant may represent a form of economic recovery.

Considering the phytoextraction perspective, the metal accumulation can be defined in several ways, (a) presence of larger quantities of a given element than customary, which can be quantified by the mAI; (b) accumulation of a given element in concentrations higher than the growth medium, which can be quantified by the mBCF and the mBAF; and (c) ability to accumulate a given element in the stems and leaves above the accumulation by the belowground organs, which can be quantified by the TF and the mTF. Table 3 presents the results obtained for mAI, mBCF, mBAF, TF, and mTF of A. donax and Miscanthus spp. under heavy metal-contaminated soils.

The ratio between a certain heavy metal accumulated into the biomass of giant reed or Miscanthus genotypes from contaminated soils compared with the same metal accumulated into the control biomass is represented by mAI. For all the species, elements, and the different fractions of the plants, the results obtained were superior to one (>1), indicating that both grasses are able to take up and accumulate a given element in larger quantities than customary. Only in the assay with Cr300, the aboveground biomass showed a mAI inferior to one (<1). According to this parameter, both grasses show potential to be used in phytoextraction. Except for A. donax in the presence of Zn contamination, all the results show that mAI for the belowground is higher than for the aboveground biomass. And with the exception of M. floridulus, mAI increased with increasing contamination. The highest accumulation index was obtained with M. × giganteus in Zn900 soils. In this case, the belowground organs of this grass were able to accumulate 20 times more Zn than control plants. Regarding giant reed, higher mAIs were obtained for Pb with the belowground organs.

Regarding the mBCF values, all the results were below one (<1), suggesting that the heavy metal transfer from the growing medium to plant is limited. These results were also reported by Kacprzak et al. [54] for M. × giganteus and Zn contamination, by Jin and You [55] who tested M. sinensis under heavy metal-contaminated soils by mine tailings and by several other authors referenced in the works of Nsanganwimana et al. [10, 51]. Higher transfer was observed for Zn, followed by Cr and then by Pb, in agreement with the results presented by Nsanganwimana et al. [10, 51], mostly due to the highest mobility associated with zinc [3]. Nevertheless, these results may reflect also the characteristics of the growing medium. The slightly alkaline pH of the soil contributed to lessen the mobility of the metals in the soil solution and the highest calcium content of the soil (Table 1) may also have interfered with the metal accumulation by plants, as shown by Kabata-Pendias [3] for Zn with increasing Ca/Zn. The mBAF can be calculated to evaluate how much of the contamination can be removed from the soil by the biomass. According to the results presented in Table 3, the harvestable biomass (the aerial fraction) removes less than 1 % of the contaminant from the soil per year. This means that the removal of the contamination from the soil by these plants represents a slow process, representing a constraint as reported by others [20, 56]. Still, when contamination does not represent an imminent danger to environment or humans, these crops offer the possibility to generate income from contaminated soils, overtaking the limitation associated with the time required to eliminate the contaminant.

Results obtained for TF indicate that zinc transfer, for both perennials, are not encouraging once they show that zinc content in the aboveground organs are similar or lower than the concentration of this element in the belowground organs. Similar findings were observed by Kacprzak et al. [54] with M. × giganteus and Jin and You [55] with M. sinensis and by Nsanganwimana et al. [10, 51]. Results also indicate that giant reed translocated zinc in higher amounts than chromium or lead, for the reasons stated in “Heavy metal concentration in A. donax and Miscanthus genotypes” section. Regarding Cr and Pb, TFs are much lower than one (<1) as giant reed accumulates these metals more in the hypogeal components. Other studies [10] report similar findings for those metals. Nevertheless, some studies [52, 57, 58] report TFs above one. However, those results were obtained in specific conditions and should not be generalized. As heavy metal translocation to the harvestable plant components is the important feature for the selection of an extractor plant [59], mTFs were calculated to combine the biomass production with the metal concentration. Results are, nonetheless, even more discouraging once the amount transported to the aerial components was much lower than the amount stored in roots and rhizomes, per pot. The ability to immobilize ions in the hypogeal components can be related to the metal tolerance [3], and the mycorrhizal association in the dense belowground organs, that provides a metal excluder barrier, is one of the survival strategies associated with these grasses [53].

Heavy Metal Content in Percolated Waters

The dense hypogeal system associated with perennial crops, such as giant reed or Miscanthus spp., allows it to stabilize the heavy metals on the soil-root system. Results obtained showed that in the experiments with Miscanthus spp., no significant differences were observed between species, treatments, and pots with and without plants. Average zinc content in the percolated waters was 0.10 mg dm−3, and the range of results varied from not detectable to 0.83 mg dm−3 (in a control pot with M. floridulus). In the giant reed experiments, heavy metal content in percolated waters did also not differ significantly between treatments and pots with and without plants. In the Zn assay, average zinc content was 0.045 mg dm−3 and the range of results varied from not detectable to 0.18 mg dm−3; in the Cr assay, average chromium content was 0.020 mg dm−3 and the range of results varied from not detectable to 0.07 mg dm−3; and in the Pb assay, average lead content was 0.015 mg dm−3 and the range of results varied from not detectable to 0.07 mg dm−3. These results are lower than the limit values for irrigation water (Zn, 2.0 mg dm−3; Cr, 0.1 mg dm−3; and Pb, 5.0 mg dm−3 [60]) and thus do not pose a risk to the ecosystems. Although differences to pots without plants were not noticeable, results indicate and corroborate that these two crops are able to retain, stabilize, and control heavy metal leaching. Use of perennials for the prevention of groundwater contamination was also extensively reported in the work of Barbosa et al. [28].

Conclusions

Results obtained in the conducted experiments showed that production of A. donax was not significantly affected by heavy metal contamination, except under the 600 mg Cr kg−1 soil. Concerning Miscanthus genotypes, the zinc contamination studied did not affect the production of M. sinensis and M. floridulus and reduced significantly, by 20 %, the production of M. × giganteus. However, in spite of this yield reduction, this genotype was the most productive. Analyses made on the percolated waters suggest that giant reed hypogeal system promoted the phytostabilization of Zn, Cr, and Pb on soil, as leaching was prevented. The same can also be mentioned for Miscanthus genotypes under Zn-contaminated soils, as Zn content in the percolated waters did not represent a threat to the ecosystems. Metal concentration in the biomass reflected soil metal concentration, and therefore, these grasses were considered as indicators. However, results indicate that these grasses have limited capacity to extract metals from the growing medium and to transport them from the belowground organs to the harvestable biomass. But, regardless of the reduced mBCF and mTF, these species can be considered as interesting candidates for phytoextraction favored by the mAIs observed and the high biomass produced (especially giant reed and M. × giganteus). The possibility to reduce metal concentration from contaminated soils by means of phytoremediation using these perennials is of great interest and importance to maintain environmental quality standards and to improve soil quality with relative low costs. Moreover, the reduced metal content in the harvestable biomass can be also looked as beneficial, once contamination is not influencing the biomass quality (in terms of the metal content). Still, more studies are necessary to clarify the mechanisms associated with the absorption and translocation of heavy metals in these perennials, its relation with the growing medium, and also the interaction of these contaminants with other elements.