An enormous number of sites around the world are contaminated to a lesser or greater degree, in part because many industries have left sites with untreated contaminated soil after ending their activities. Such sites are very costly to remediate; a common technique involves excavating the contaminated soil, removing it, and placing it in a landfill, in the so-called ‘dig and dump’. Other methods such as thermal and chemical remediation can also be applied. The remediation of sites with a high level of contamination is prioritized. However, many other sites with a medium or relatively low degree of contamination have been left untreated for decades. Thus, there is a need for a cheap and environmentally friendly method for remediating such sites, such as phytoremediation.
Although it takes a long time for phytoremediation to clean up a contaminated site, this method is outstanding from an economic perspective, especially when used for large areas [1
]. Phytoremediation is defined as a method “using green plants to remove, contain or render environmental contaminants harmless” [3
]. Only a few plant species or cultivars of a species have the ability to take up, stabilize or degrade certain contaminants, and thus be useful in phytoremediation [1
]. Moreover, the plants used in phytoremediation should be domestic in order to ensure that they will thrive in a given climate and to avoid invasion by alien species.
Most polluted sites contain a mixture of pollutants. Inorganic pollutants—such as heavy metals and arsenic—and organics—including polychlorinated biphenyls (PCBs) and polycyclic aromatic hydrocarbons (PAHs)—are common contaminants in many polluted soils. Specific plant species and varieties can be selected for the phytoremediation of various pollutants: some plants degrade organics well, while others show good performance in removing heavy metals [1
]. Plants can degrade organic pollutants by exuding degrading enzymes into the soil, enabling degrading microbes to degrade organics in the rhizosphere, or taking up organics and transforming them within plant cells [6
]. Plants can take up inorganics—such as metals and metalloids, which cannot be degraded—from the soil by their roots and accumulate them in plant tissue [7
]. Plants may also stabilize contaminants in the soil by binding them to soil particles, thereby decreasing their availability and toxicity; however, the contaminants are not removed from the soil in such cases [8
Willow, which is cultivated for bioenergy production, has been investigated as a possible phytoremediation crop since the 1990s [9
]. This species has been shown to take up and accumulate large quantities of zinc (Zn) and cadmium (Cd) [18
]; it also has high biomass production [19
]. In comparison with species such as poplar, sunflower and tobacco, willow shows the highest uptake capacity in the field [4
]. Some clones of willow (Salix pp.) used in short-rotation coppicing can extract more heavy metals from soil than other clones [9
]. It has been shown that the soil concentration of Cd, Zn and copper (Cu) decreases after cultivation with S. viminalis
] and S. caprea
]. The ability of Salix clones to remove PAH from gas work soil has been shown to vary [24
]. Vervaeke et al. [25
] demonstrated mineral oil reduction and PAH degradation in contaminated sediment after 1.5 years of treatment with willow in the field. Faubert et al. [26
] showed that S. miyabeana
cultivation facilitated the migration of heavy metals, PAHs and PCBs towards the plant roots due to the plants’ high transpiration rate.
In our earlier work, we found that S. viminalis
decreased the total Cd concentration in agricultural soil by up to 25% after 4 years of Salix cultivation [16
]. The aim of the present study was therefore to investigate whether S. viminalis
could decrease the total levels of PCBs, PAHs, heavy metals and arsenic (As) in the contaminated soil left at a site on which a mechanics workshop was previously situated for many decades. However, our intention in this study was not to achieve certain limit values or to remediate the whole surface.
The decrease in each of the contaminants at the site was followed over a period of 10 years to determine whether the rate of removal varied among the contaminants. The concentrations of the contaminants in the soil were analysed each year in a reference location with no S. viminalis and in a location with S. viminalis. An unpolluted location was chosen as the reference location in order to analyse the rate of removal of the contaminants each year without the presence of S. viminalis. In addition, to analyse the decrease in contaminants without the presence of S. viminalis, three contaminated locations without S. viminalis were monitored, starting at the beginning of the study and continuing for 7 years after the end of the study.
3. Results and Discussion
The total concentrations of the various contaminants in the soil before and after treatment with or without S. viminalis
are shown in Table 2
. After 10 years, all contaminants in the soil decreased significantly in the location planted with S. viminalis
). In contrast, the reference spot with low contamination and no S. viminalis
showed no significant changes (p
> 0.05) in the total concentrations of any of the contaminants analysed. Furthermore, the concentrations of the contaminants in the contaminated locations in the absence of S. viminalis
did not change significantly in the period 2005–2022 (Table 2
). These findings show that S. viminalis
influenced the decrease in the contaminants in the soil.
During the 10 years of treatment with S. viminalis
, the total concentrations of the various metals and metalloids in the soil decreased by 21–87% (Table 2
). The percentage of each contaminant that was removed after 10 years of treatment was as follows: Cr, 21%; As, 30%; Cd, 54%; Zn, 61%; Cu, 62%; Pb, 63%; and Ni, 87%. The lowest decrease was found for As and chromium (Cr). These elements are commonly found in complexes with oxygen; in the available soil fractions, they occurred in the forms of arsenate and chromate anions, which can be taken up by plants. A greater decrease was found for Cd, Zn, Cu and lead (Pb), all of which were removed to a similar extent by the plants. Nickel (Ni) was removed to an even greater extent from the soil by S. viminalis
. In the mobile phase, these elements exist in cationic forms, which can be taken up by the plant. Thus, the data indicate that it may be more difficult to remove anions than cations from the soil with S. viminalis
No good reason has been found to explain the low removal of Cr and As, in comparison with the other metals. However, one possible explanation is that, at high pH, Cr and As anions have higher mobility in the soil solution than at lower pH, while the opposite holds true for cations. The plant root exudate may influence the soil pH and thereby affect the availability of metals to plants in the rhizosphere soil [28
]. S. viminalis
may have acidified the soil and thus increased the solubility of the metal cations [30
], while simultaneously precipitating the Cr and As anions [31
Although this was not the focus of this work, the decrease in the total concentration of the metals caused the concentrations to reach under the limit values (Table 1
) in some cases. For example, the concentration of Cu decreased from 294 to 111 mg kg−1
during 10 years of treatment with S. viminalis
), and reached the less sensitive land-use limit value (see Table 1
) after just 2 years.
The analysed organic substances also decreased in the soil under S. viminalis
cultivation (Table 2
). During the 10 years of treatment, PAHs decreased by 20–73%, depending on the substance. It has been shown that PAHs with more rings and more complex ring structures are less able to degrade and be removed from the soil by plants [33
]. However, in this work, 47% of a PAH with two aromatic rings—namely, naphthalene—was removed. Even more of phenanthrene—a PAH with a three-ring structure—was removed, at 73%. Regarding PAHs with four rings—namely, chrysene and pyrene—25% and 54% were removed, respectively (Table 2
, Figure 2
). In comparison with chrysene, pyrene has a more complex structure with more bindings to carbon (Figure 2
). Thus, it could be expected that pyrene would be more difficult to degrade and remove than chrysene; however, that was not the case in this study.
The sum of the PCBs decreased by 53% (Table 2
). PCBs consist of two aromatic rings; unlike the PAH naphthalene, the rings are connected with just one C–C binding (Figure 2
). Therefore, based on C-bindings alone, PCBs ought to be easier to degrade than naphthalene. However, no pattern was observed in terms of the differences in the removal of the two substances. On the other hand, PCBs differ from PAHs in terms of their chloride content, with various PCB congeners having different numbers of chloride and chloride binding sites (Figure 2
). Chlorine atoms bind to carbon with bindings that have only a few natural degraders [34
]. Nevertheless, this study showed that chloride-containing molecules can be removed from soil (Table 2
). Based on the data collected in this study, it is not possible to know whether the PAHs or PCBs were degraded or taken up by the plants, as such information lies beyond the scope of this research. The data show no signs of transformation between different types of PCBs or PAHs.
The rate of removal differed among the contaminants. Most of the elements showed a decrease in their removal, with a plateau in the curve after 5 years of treatment with S. viminalis
). However, Cr was an exception; even after 10 years, its concentration in the soil seemed to decrease at the same rate. Regarding the organics (Figure 4
), no plateau in the curve was observed for the sum of PCBs; however, specific substances such as PCB-153 and PCB-180 did exhibit a plateau after 2 and 4 years, respectively. Among some of the PAHs (i.e., naphthalene, phenanthrene and pyrene; Figure 4
), there was a slight decrease in the removal after 5 years of S. viminalis
cultivation. The removal of chrysene, the sum of the carcinogenic PAHs and the other PAHs decreased drastically after 2 years. This decrease in the removal after a couple of years may be caused by root growth and by roots interfering with each other during uptake, the bioremediation rate of bacteria decreasing, the contaminant reaching a depletion zone of removal, or biomass growth slowing down due to the depletion of nutrients.