Introduction

Silicon (Si) is the second element on the earth's surface, being the values of silicic acid (H4SiO4) contained in the soil solution in the same range as other important nutrients for plants, such as potassium and calcium (Matichenkov et al. 1995; Matichenkov and Bocharnikova 2001; Hull 2004).

Silicon is taken up by roots from the soil solution in the form of monosilicic acid and is transported through xylem to the different tissues where it is finally deposited as amorphous silica (SiO2) or silicophytoliths (SiP) (Datnoff et al. 2001; Mitani and Ma 2005; Exley 2015; Ma and Yamaji 2015). Amorphous silica can constitute between 0.1 and 10% of the dry weight of higher plants, being the Poales one of the main silica-accumulating families, including species corresponding to the main world's food crops. Rice, maize, and wheat are the main contributors to global crop silicon fluxes (Curry and Perry 2007; Keller et al. 2012); considering, that wheat can produce between 342 ± 114 kg/h year−1 of SiP (Gocke et al. 2013).

Silicophytoliths produced in plant tissues can contain about 90% silica, 1–6% organic carbon, and traces of other components such as aluminium and iron (Parr and Sullivan 2011; Rajendiran et al. 2012).

Although Si is not considered essential for higher plants, there is increasing evidence that it has beneficial effects, such as promoting plant growth, enhancing photosynthesis, mitigating biotic and abiotic stresses, such as metal toxicity and nutrient imbalance, among others. (Datnoff et al. 2001; Guo and Gifford 2002; Liang et al. 2007; Van Bockhaven et al. 2012; Guo et al. 2013; Adrees et al. 2015; Hajiboland, 2017; Zhang, 2017). Because of these benefits, silicate fertilization has become a usual practice in many agroecosystems to improve productivity and ameliorate biotic and abiotic stresses (Marafon and Endres 2013; Artyszak 2018).

Heavy metals (HM) occur naturally in the Earth´s crust and are found with natural background concentrations; however, nowadays, it is also important to consider the anthropogenic emissions, which have led to an increase HM concentration in the environment (Alloway, 2012). Living organisms tend to accumulate HM within their tissues but at low levels to avoid their adverse effects (Gall et al. 2015). Particularly, plants are exposed principally through the aqueous phase of the soil (Plette et al. 1999) and these elements are captured by the roots (Khan et al. 2011). Heavy metal bioavailability is a function of total concentration together with the regulating effects of physicochemical and biological factors such as pH, organic matter content (Ernst 1996; Wu et al. 2006) and plant species genotypes (Orcutt et al. 2000), among others. Heavy metal contamination can affect many physiological processes (Benzarti et al. 2008; Walley and Huerta 2010), and the main plant strategies to avoid their toxic effects are: (1) exclusion, where the plant limits the entry of HM in the cells; (2) accumulation, in which HM taken by the cells are isolated through mechanisms such as chelation, compartmentalization, co-precipitation with the Si in different sites of the plant, among others (Ruley et al. 2004; Sharma and Dietz 2006).

Recently, some research has shown that the presence of Si in plants has led to enhance the tolerance to some adverse effects of HM, highlighting its potential role in mitigating them (Rizwan et al. 2015; Greger et al. 2016).

In Argentina, the soils of the Pampean plain hold one of the largest worldwide food productions and have increased their surface under direct sowing with a greater use of agrochemicals and fertilizers, associated with the simplification of rotations and a sharp increase in the area sown with wheat (Ministerio de Agroindustria de La Nación, 2018). As a result of these agricultural practices applied, the rate of soil degradation has increased during the last years (Alvarez et al. 2008; Álvarez et al. 2012; Demetrio 2012; Duval et al. 2015; Rodriguez et al. 2015).

Research on the biogeochemical cycling of silicon in the agroecosystems of the area comprises a few studies on different crops (Benvenuto 2017; Frayssinet et al. 2019), their associated soils, as well as their surface and groundwater systems (Borrelli et al. 2012; Martínez and Osterrieth 2013; Osterrieth et al. 2015).

As it was detailed above, Si fertilization has been applied in diverse countries and many works have been conducted in controlled experiments in laboratory, while field studies are still very scarce (Álvarez 2001; Guo and Gifford 2002; Ma and Takahashi 2002; Sasal et al. 2006; Alvarez et al. 2008; Walley and Huerta 2010; Guo et al. 2013; Greger and Landberg 2015; Chirkes et al. 2018). Particularly, in Argentina, the field studies are practically null (Frayssinet et al. 2019, 2021). Moreover, there is practically no information about the effect of silicon fertilization associated with HM on crops. The main objective is to evaluate the effect of two Si fertilizers in the SiP and HM contents of Triticum aestivum plants and HM contents of soils from an agroecosystem of Pampean Plain, Argentina. Considering the worldwide background on the possibility that Si can contribute to control the HM contents, it is proposed that an increase in the availability of Si in soils, due to Si fertilizers application, might affect the relation between Si and HM contents in soils and plants.

Materials and methods

Study area

A field study was carried out from July to December 2014, in the southeast of the province of Buenos Aires (38° 32′ 24" S; 58° 46′ 24" W), near the city of Necochea (Fig. 1), in Argentina. This study area being representative of a mid-latitude temperate plains environment.

Fig. 1
figure 1

Localization of the study area

Full size image

The climate is temperate with average winter temperature of 9 °C and reaching 26 °C in summer (Deluchi et al. 1995; Merlotto and Piccolo 2009). The water regime is subhumid-dry (Thornthwaite 1948) with mean annual precipitation of 700–800 mm (Quiroz 2009). During the period analyzed, a total amount of 672.5 mm of precipitation and average temperatures of 12 °C in July, rising 2 °C on a monthly average to reach 23 °C in December, were recorded (INTA 2014). The dominant soils are taxonomically classified as typical Phaeozem luvic (WRB 2014) which are characterized as very dark, well drained, non-alkaline and non-saline, with slope of 1%. The texture of the soil varies between sandy loam to clay loam, the average content of organic matter at superficial level of the agricultural soils of the zone is around 4%, the available silicon content is near 40 mg/L on average, and pH values near 6.5 (INTA 2010; Osterrieth et al. 2015; Benvenuto 2017).

Experimental design

Two independent essays with a complete randomized design were conducted on 15 m2 plots cultivated with wheat (Triticum aestivum, Baguette 801 Premium-Nidera Seeds, Intermediate-Long Cycle). The essays were carried out in a field with a typical history of use, involving hundreds of years with conventional tillage and over the last 20 years with direct seeding management producing mainly soybean, sunflower, wheat, corn and barley. In these experiments and before sowing the wheat, all the plots were fertilized with phosphorus (P) and nitrogen (N), the first one distributed randomly in a dose of 150 kg/h containing 46% of the available P (TSP-Helm®), and the second one as a controlled application of 200 kg/h of urea with 46% of the available N (CO(NH2)2-Pacifex®). Sowing took place on 9th July, while soil fertilization with silicon was made the following day (10th July) and finally harvesting was carried out on December 23rd.

Experiment 1

Two treatments were applied: (1) without silicon fertilization; (2) with application of a liquid silicon fertilizer on the soil (registered trademark Quicksol®, Diverse Enterprises, Texas, USA). This fertilizer contained the following components: Silicon (302 g/L = 36%), Calcium (1.76 g/L), Iron (1.55 g/L), Sodium (83 g/L), Potassium (3.3 g/L), Magnesium (5.66 g/L, Manganese (1%), Humic acids (1% p/v), Fulvic acids (1% p/v), Copper (1%) and Zinc (1%). The dose applied was 1 L per hectare.

Experiment 2

Two treatments were applied: (1) without silicon fertilization; (2) with application of powdered silicon fertilizer on the soil (registered trademark Silifix®, Siliground®, Argentina). The composition of the fertilizer was: CaO = 43.21%; SiO2 = 16.72%; FeO = 19.22%; MgO = 5.93%; MnO = 3.38%; P2O5 = 7.74%; Zn = 0.36%; S = 1.12%; and N = 2.8%. The dose of fertilizer applied was 300 kg/h.

The soils of both trials were analyzed after the end of the wheat cycle developed and the results showed dark colors, a predominantly silty loam texture, subangular block structure, hard, friable, slightly plastic and slightly sticky consistency; except for those treated with powdered fertilizer which showed a slightly sandier texture and a non-plastic consistency. The values of clay, silt and sand were on average: 12%, 55% and 33%, respectively, for all the plots analyzed, except the one treated with powder fertilizer, that showed nearly 9% of clay, 50% of silt and 41% of sand. These differences could be due to the coarser and non-mineral particles contained in the applied powder fertilizer (textural values: 6% clay, 47% silt and 47% sand). The values of Si measured in the soil solution ranged from 22.92 to 56.35 mg/L, increasing with significant differences from September to December in both experiments (p value = 0.0121) and also, within the liquid experiment in December, where the control plots had 15.95 mg/L more than treated ones (56.35 mg/L vs 40.40 mg/L, respectively; p value  = 0.0147) (values published in Frayssinet et al. (2019) being also available in Table B of Supplementary Material section of this work). Finally, the pH in the soil water solution (1:2,5) was measured, showing slightly acid to neutral values (6.04–6.81) and total organic matter (TOM) content was between 3.4 and 4.8% (Frayssinet et al. 2019), both properties being in agreement with previous reports for agricultural soils in the area.

Sample collection

Throughout the experiments, 3 wheat plants were harvested by treatment at the different phenological stages. Twelve complete plants were harvested at the tillering stage (phenological stage Z: 2.5 (Zadoks 1974) in September, and a further 12 complete plants were harvested at the ripening stage (phenological state Z: 9.9 (Zadoks 1974) in December (Fig. 2). Moreover, during field sampling, at each phenological stage, the number of plants per square meter were calculated to evaluate biomass. The calculated biomass per hectare was extrapolated from 4 plots (one of each treatment) of 0.5 m2 surface. In addition, 12 integrated soil samples from the first 0–15 cm were collected in each sample (3 replicates of the 4 treatments). In turn, each sample of each treatment belongs to a single sample composed of three subsamples, to obtain representative samples.

Fig. 2
figure 2

Left: illustrative drawing of the sampling design where the red points indicate the sampling places. Right: photographs of the crop at both sampling moments

Full size image

Heavy metals contents in soil

The content of HM [(Cadmium (Cd), Copper (Cu), Lead (Pb), Zinc (Zn), Manganese (Mn), Nickel (Ni), Chromium (Cr) and Iron (Fe)] were analyzed following the method described by Marcovecchio and Ferrer (2005), which includes the mineralization of the sample with a strong mixture of acids (HClO4:HNO3, 1:3) under controlled temperature (110 ± 5 °C). Then the residue was diluted in 0.7% (v/v) HNO3 up to 10 mL. The HM contents within the corresponding extracts were determined by Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES) (Perkin Elmer Optima 2100 DV) (La Colla et al. 2018).

Silicophytolith and heavy metal contents in plants

The plant samples were divided into leaves, stems, and spikes (only in ripening stage) and washed in ultrasound (Test-Lab, Ultrasonic TB-010) for 15–20 min. Then, they were dried in an oven at 60 °C for 24 h and weight (initial weight). Silicophytoliths were extracted through the calcination method proposed by Labouriau (1983). The samples were charred at 200 °C for 2 h, then boiled in 5 N HCl to dissolve the calcium crystals and finally ignited at 700–800 °C for 3 h. The ashes, compound mainly by silicophytoliths, were then weighted (final weight) and the amorphous silica content was calculated as final weight/initial weight and expressed as milligrams of silicophytoliths per gram of plant (Fernández Honaine et al. 2005). These final ashes were mounted on aluminium disks and coated with gold or palladium and observed under a scanning electron microscope (SEM), FEI (Quantas 200, LIMF-UNLP). The identification of the elements within the silicophytoliths was carried out by means of dispersive spectroscopy X-ray (Energy-Dispersive Spectroscopy (EDS), EDS-Silicon Drift Detector (EDS-SDD) Apolo 40, LIMF-UNLP).

The contents of HM in plant tissues (Cd, Cu, Pb, Zn, Mn, Ni, Cr and Fe) were also analyzed following the method described above by Marcovecchio and Ferrer (2005) and the contents in the corresponding extracts were determined by ICP-OES (Perkin Elmer Optima 2100 DV) (La Colla et al. 2018). The samples analyzed include leaves and stems from the tillering stage, while spikes were also included in the ripening stage. All HM analyses were performed in duplicate, with an uncertainty of < 15% based on one relative standard deviation of the replicates. Analytical grade reagents, reagent blanks and certified reference materials (CRM) were used to ensure analytical quality control. The analytical quality was tested against reference material (pepperbush (R.M. N °1)). The recovery percentages for all HM in their corresponding CRM were between 80 and 115%. The instrumental operating conditions of the ICP OES Optima 2100 DV (Perkin Elmer) with a Cross-Flow nebulizer together with the analytical method detection limit (MDL) for each metal (mg/g) are shown in Table A of Supplementary Material.

Statistical analysis

A two-sample t-test was performed in all the variables analyzed using the Infostat® software package (Di Rienzo et al. 2016), to determine the relation between Si and HM contents for each experiment throughout the study. The significance level was set at 5% for all analyses.

Results

Heavy metal contents

Soil

The HM content decreased from September to December and did not show significant difference between treatments for both analyses, being low on all plots analyzed, both in soil solution (less than 0.4 µg/L) and in the solid phase (less than 29 mg/g) (Table 1).

Table 1 Mean ± standard deviation of heavy metals (HM) content in soil solution (µg/L) and HM in the solid phase of the soil (mg/g) per treatment (n = 3) through time
Full size table

Plants

The biomass values were similar for both experiments and increasing significantly over time, rising from nearly 290 kg/h dry weight in the tillering stage, to almost 7540 kg/h dry weight.in the ripening (p value < 0.0001) (Table 2).

Table 2 Mean values of the wheat biomass measured (kg/h dry weight) per treatment through time
Full size table

The SiP values were similar in both experiments, ranging from 38.42 to 86.11 mg/g on average (Fig. 3) and increasing over time (p value  < 0.0001). The only case in which the SiP content has shown significant differences was in December for the powdered plots, with control having more SiP than the treated ones (p value  = 0.0242). The HM content was low in all samples analyzed, decreasing from September to December (376–168.87 µg/g on average, respectively) (p value  < 0.0001) (Fig. 3).

Fig. 3
figure 3

Average of heavy metals (HM) and silicophytoliths (SiP) content per treatment (n = 3) in the aboveground biomass of wheat plants through time. Triangles (∆) indicate significant differences between treatments (p values < 0.05) and asterisks (*) between dates for both variables analyzed (p values < 0.0001) within the two essays.

Full size image

The deposition of SiP and HM among the different organs was evaluated at the ripening stage, so their distribution in leaf, stem and spike samples was analyzed (Fig. 4). A strong relationship between SiP and HM was observed in the different organs of the plant. All the elements studied were mainly found in the leaf, followed by the spike and finally the stem. Significant differences were found for SiP contents between control and treated leaves within the powder experiment (p value  = 0.0065).

Fig. 4
figure 4

Average of heavy metals (HM) and silicophytoliths (SiP) contents per treatment (n = 3) in December, among the different plant organs. Asterisks (*) indicates significant differences only for SiP content between control and treated leaves of the powder experiment (p values < 0.05)

Full size image

Moreover, the content of HM within the spike in the ripening stage was analyzed, since it contains the grain that is the edible part of the crop (Table 3). The HM in order of abundance were Fe > Mn > Zn > Ni > Cu > Cr > Cd, while Pb was below the detection limit of the method applied. Significant differences were found for Cr content only between the treated and control spike of the second (powder) experiment.

Table 3 Mean ± standard deviation of each heavy metal (HM) and silicophytoliths (SiP) contents for each treatment (n = 3), in December, at the spike organ discriminated per element analysed
Full size table

Finally, to study the elemental composition of SiP, a study was carried out using SEM and EDS analyses (Fig. 5). Different typical morphotypes of SiP were found in the ashes of the control and treated wheat leaves and spikes, showing different degrees of silicification, ranking from totally to barely (Fig. 5a–c, g). Analyzing through EDS with n = 5, for the first morphotypes of silicification (Fig. 5a–c) the contents of Si, Oxygen (O) and Carbon (C) were variable in the control plant (58–71% Si; 19–36% O; 9–13% C) (Fig. 5e); in some articulated elongated morphotypes (AEC and AD) and in hair cells of treated plants, contents of Aluminium (Al), Calcium (Ca), Potassium (K) and Sodium (Na) (6–12% Al; 1.1–3% Ca; 1.7–6% K; 0.2–1.5% Na) were observed (Fig. 5d, f). Meanwhile, it was observed that some of the short epidermal cells (SEC) were incompletely silicified and formed by nanospheres (0.08 µm on average) (Fig. 5g–i). These nanospheres were analyzed by EDS with a semiquantitative analysis of n = 7 and showed that in addition to the Si, O and C content, they also had low contents of Al, Fe, K, Ca, Na, Chlorine (Cl), and Magnesium (Mg) (35–70% Si; 7–43% O; 0.6–30% C; 2–14% Al; 1.7–7.2% Fe; 1.4–7% K; 1–6% Ca; 0.5–2% Na; 0.7–2.8% Cl; 0.4–1.8% Mg) (Fig. 5j–l).

Fig. 5
figure 5

Scanning electron microscopy (SEM) images and EDS analyses of silicophytoliths of wheat plants. a Silicophytoliths of treated leaf, b Silicophytoliths of control leaf. c Silicophytoliths of spike. d–f EDS spectra of the surface marked in green as E1–E3 within the different silicophytoliths. g Short silica cells of leaf. h Detailed image of g looking at the nanospheres of the SEC morphotype. i Image detail of (h) showing the areas where the EDS spectra marked in green as E4–E6 within the SEC morphotype were made. j–l EDS spectrum mentioned in the set of nanospheres marked in i. Silicophytoliths morphotypes: TP Trapeziform polylobate cell, R Rondel, HC unciforms and acicular hair cell, BH base of hair, SEC short epidermal cells, P papillae, AEC articulated epidermal cells, AD articulated dendritic of spike, T Trapeziform (ICPT, 2019)

Full size image

Discussion

The contents of heavy metals and silicon in soils in relation to Si fertilization in the present work are similar to the ones observed for agricultural soils worldwide (Gocke et al. 2013; Seyfferth et al. 2013) and for the mollic epipedons of similar areas within the southeastern Pampean plain (Osterrieth et al. 2015; Benvenuto 2017) (Table 1).

The results in our experiments showed that the HM contents in the soil (both solid and in solution) decreases over time, although these decreases were not statistically significant (Table 1). On the other hand, a previous work in the same experiment showed that the Si content in the soil solution increases along time, suggesting an opposite pattern between these two elements (Frayssinet et al. 2019). This inverse relationship might be explained due to some HM complexation promoted by silicon (Greger and Landberg 2015) and/or some HM desorption followed by a leaching process (Meharg 2011). Future studies that include sites without vegetation and/or sites with groundwater analysis may shed light on the relevance of this process. Another explanation to this dynamic may be related to the plant development, associated with an increase in the rate of HM absorption as it grows, or the retention of HM within the SiP, the roots or in the rhizosphere environment (Lux et al. 2020; Alloway 2012; Adrees et al. 2015).

As it has been reported worldwide, the silicification process in plants is an irreversible phenomenon and therefore accumulator species produce more biomineralizations of Si as they grow (Motomura et al. 2002; de Melo et al. 2010; Fernandez Honaine et al. 2013, 2017). In this work, the results obtained have supported this pattern, as the content of SiP had increased from September to December (Fig. 3). Moreover, in the mentioned figure, a significant difference could be observed in December for the essay with liquid fertilizer, between the fertilized plants (86.1 mg/g) and the control ones (79.2 mg/g), and this pattern, of higher amorphous silica accumulation by the treated wheat, might explain why the control plots were having significantly higher content of available Si available in the soil solution than the treated ones (Frayssinet et al. 2019).

In addition, some studies have shown a positive impact of the Si in soil solution promoting the uptake of some nutrients by the plant (Hernandez-Apaolaza 2014; White et al. 2017), which could support the idea of the Si leading to an increase in biomass that would help the dilution process, while protecting the crop from the toxic effects of HM (Greger et al. 2016; Rizwan et al. 2016). It could be hypothesized that the decrease of HM content observed in our plants (Fig. 3) is explained by a dilution effect due to the increase in biomass, as in this study a statistically significant increase of 7257.32 kg/h dry weight between tillering and ripening phases was observed (290.44–7547.76 kg/h dry weight, respectively) (Table 2). It is worth mentioning that there was no type of fertilization that notoriously improved this process, since the values observed in Table 2 were very similar for both trials.

Considering the HM deposition pattern among the different organs, some authors have mentioned that they were mainly deposited in the leaf and lastly in the grains (Khan et al. 2011). In the case of silicophytoliths, some research, have shown that they are mostly accumulated in leaves, stems and spikes (Datnoff et al. 2001; Ma and Yamaji 2015; Tubana and Heckman 2015). The same pattern of HM and SiP total contents distributions was found in our results, being the highest within the leaves, followed by the spikes and lastly by the stems (Fig. 4). Indeed, looking at Fig. 4 may be useful to explain the statistically significant difference observed in Fig. 3, given the SiP values found in the leaves of the plots treated with powdered fertilizer, which also showed statistically significant differences (Fig. 4), highlighting which of all the organs analyzed may be related to the difference seen. This statistically significant differences might be explained according to Frayssinet et al. (2019), since by losing Si available in the soil solution the plant is possibly less likely to absorb it and allocate it to the SiP production. Moreover, and delving into the HM contents of the spike, that contains the edible part (Table 2), it can be seen that the organ contains HM levels under the threshold values stablished by Food and Agriculture Organization (FAO) and World Health Organization (WHO) (Alimentarius 2006), and therefore meets the standards to be introduced in the market.

There is another process that could explain this strong relationship in the amounts of HM and SiP accumulated within the analyzed organs, but for which there are not many studies performed. In fact, there are only a few studies that agree with the idea that SiP co-precipitates with HM (He et al. 2013; Collin et al. 2014; Keller et al. 2015; Adrees et al. 2015; Bhat et al. 2019), but recent studies have revealed that several different cations are present in silica biomineralizations (Neumann and zur Nieden 2001; Parr and Sullivan 2011; Kamenik et al. 2013; Wu et al. 2014), which would contribute to the isolation of metals. Moreover, we focalized on the strength relation between SiP and the total content of HM in aboveground tissues of the plant and we observed that the contents of SiP and HM have a very similar distribution pattern among the different organs analyzed (Fig. 4), which could reinforce the idea that HM is retained and/or occluded in the silicophytoliths. Indeed, the high magnification analysis with SEM and EDS (Fig. 5), allowed to detect different elements within the SiP. Inside the short epidermal cells that were barely silicified, an internal structure made up of very small nanospheres (0.08 µm) with a large specific surface area were found, probably favouring the adsorption of different chemical elements (Cl, Na, K, Ca) in addition to HM (in this case Fe and Mg). However, it is important to note that while the presence of other HM could not be detected, possibly due to the high detection limit of the applied methodologies, it cannot be completely ruled out that SiP are incorporating some of the HM that are within the plant, and this important aspect will be analyzed in the future.

Finally, one can also consider the hypothesis that the HM is retained not only in the SiP within the aboveground organs but also in the roots of this crop or in its rhizosphere environment, probably explaining the low HM contents found in the soil of this work (Table 1). Furthermore, as many authors have shown, rhizofiltration is the main process used by plants to avoid the toxic effects of HM present in soils (Sharma and Dietz 2006). In addition, there are studies that reported a reduction in Cd uptake after Si fertilization, which is attributed to some reduction in root–shoot translocation and changes in compartmentation within the plant cell (Neumann and zur Nieden 2001; Liang et al. 2007). Moreover, Lux et al. (2003) have suggested that the silica deposition in the endodermis and pericycle of the roots was responsible for HM stress tolerance. These effects may be happening in this work, as seen with our results both in the soil and in the crop and will prevent the spread of metals through the food chain, aspects that become important when an edible crop as wheat, that is a silica-accumulating specie, is involved (Khan et al. 2011; Alloway 2012).

Conclusion

This is the first study in the Pampean plain region regarding the relation between silica, silicophytoliths and heavy metal contents within the soil and the aboveground organs in a typical crop representative of a mid-latitude temperate plains environment. At a general level, an inverse relation between HM and Si was observed in the soils and within the aboveground organs of the plant. According to the fertilizers used, the best source of Si appears to be the liquid fertilizer, as it has shown an increase in the SiP contents and a lower HM content in the organ containing the edible part of the crop; issues that did not occur with the powdered fertilizer. Moreover, on the aboveground organs of plants, there is a strong relation between SiP and HM, that is explained by the images of SEM and EDS analysis, suggesting that SiP can retain some elements within its nanospheres.

Although further studies at the cellular level are needed to elucidate some hypotheses, the results obtained here suggest that Si and SiP can avoid the HM adverse effects within the soil–plant system. This work indicates that this crop, that is a silica-accumulating specie, might be useful as a sustainable strategy to ameliorate the chemical quality of soils, crops, considering the beneficial effects of the Si biogeochemical cycle with certain health and safety conditions.