3.1 Yield of vegetables grown on PCB-contaminated soils
The influence of PCB contamination in different soils, neutral Chernozem and acidic Fluvisol, on vegetable cultivation was estimated from the biomass yield of plants, as shown in Table 3. The investigated level of spiked PCB contamination reflected the contemporary contamination in US residential soils in the range from 20 to 1700 µg/kg dw (Martinez et al., 2022). In this study, the total amount of dry radish biomass (16.2 g dw/pot) harvested from Chernozem was not statistically different (P < 0.05) from the dry weight of radishes planted on non-PCB-contaminated control Chernozem. The dryness of the respective harvested morphological parts of bulbs (surface layer and inner part) and shoots of contaminated Chernozem and the control treatment were comparable. Similar results for dry amounts of radishes cultivated on Fluvisol were obtained, as shown in Table 3. The same trends were also observed in the case of onion cultivation in these soils, where the total dry biomass ranged from 44.0 to 64.7 g dw/pot. The influence of the tested concentration of 7 PCBs (~ 1300 µg/kg dw), which was above the limit of PCBs (20 µg/kg dw) for Czech soils, had no adverse effect on radish or onion yield in Chernozem and Fluvisol treatments. This could be caused by the fact that radishes and onions usually grow mostly on the soil surface; therefore, contact with PCB compounds could be very limited. The negative effect of PCB contamination on the yield of carrot roots, shoots and total biomass (Table 3) was evident in the case of carrot cultivation on both tested soils in comparison to control soils. The yield of all carrot parts from PCB-contaminated Chernozem and Fluvisol treatments were statistically decreased (P < 0.05) compared to the respective control treatments. This could be because carrot roots usually grow deeper into the bulk soil, and contact with PCBs in the soil is longer during the vegetation period than in the case of radishes or onions. A similar reduction in plant growth and yield was observed by Javorská et al. (2011), who planted parsley and red beet on different soil types contaminated by PCBs with concentrations higher than 2100 µg/kg dw.
Table 3
Biomass dry weight (g dw/pot) of vegetables cultivated in PCB-contaminated soils.
Vegetable | Biomass (g dw/pot) | Chernozem | Fluvisol |
Control soil | PCB soil | Control soil | PCB soil |
Radish | Shoots | 9.1 ± 1.8a | 8.3 ± 1.7a | 7.7 ± 1.7a | 7.0 ± 1.4a |
Bulbs – surface layer | 5.1 ± 2.9a | 3.9 ± 2.1a | 4.2 ± 2.2a | 3.3 ± 1.6a |
Bulbs – inner part | 5.0 ± 1.3a | 4.0 ± 1.1a | 4.1 ± 1.0a | 3.3 ± 0.8a |
Total biomass | 19.3 ± 5.9a | 16.2 ± 4.8a | 16.1 ± 4.5a | 13.6 ± 3.7a |
Onion | Shoots | 21.5 ± 4.2a | 17.7 ± 2.9a | 17.3 ± 2.3a | 15.6 ± 2.2a |
Bulbs – surface layer | 13.2 ± 4.5a | 10.6 ± 3.4a | 11.3 ± 3.5a | 8.8 ± 2.6a |
Bulbs – inner part | 30.0 ± 5.5a | 24.1 ± 4.1a | 24.8 ± 5.2a | 19.6 ± 5.0a |
Total biomass | 64.7 ± 13.9a | 52.4 ± 9.4a | 53.4 ± 10.6a | 44.0 ± 8.9a |
Carrot | Shoots | 12.8 ± 3.2a | 9.4 ± 2.1ab | 10.3 ± 1.6ab | 9.3 ± 1.5b |
Roots – surface layer | 15.8 ± 3.8a | 12.5 ± 2.5ab | 12.3 ± 2.0ab | 9.6 ± 1.0b |
Roots – inner part | 65.9 ± 6.3a | 50.7 ± 5.0ab | 57.4 ± 2.4b | 44.9 ± 3.0c |
Total biomass | 94.5 ± 12.7a | 72.7 ± 7.4bc | 80.0 ± 5.5ab | 63.8 ± 4.9c |
All values represent the mean of four replications. Different lowercase letters within the same row indicate significant differences among the treatments (Tukey test; P < 0.05). |
3.2 Accumulation of PCBs in vegetables cultivated on PCB-contaminated soils
The surface layer, inner part of roots, and shoots were carefully evaluated for the content of 7 PCB congeners, and their sum was calculated, as shown in Fig. 1, which excluded the control treatments where PCBs were lower than the detection limit (< 1 µg/kg dw). All 7 tested indicator PCB congeners were found in all analysed morphological parts. From the obtained results, it was evident that higher PCB contents were found in the surface layer of the experimental vegetables (500–1400 µg/kg dw). The lowest content of the sum of 7 PCBs was found in the surface layer of onions (P < 0.05), but there was no significant difference (P < 0.05) between Chernozem and Fulvisol treatments. A slightly higher PCB content was found in the surface layer of radishes than in onions, and the highest PCB content was observed in the surface layer of carrots. Our results correspond to the findings of Javorská et al. (2011), who found a higher PCB concentration in peels than in the cortex of parsley or red beet cultivated Fluvisol and Chernozem. This could be related to the deeper growth in the bulk soil, longer contact with PCBs in soil during cultivation and the fact that carrot roots are known for higher lipid and carotene contents than onion or radish, where these lipophilic compounds, such as PCBs, are accumulated (Collins et al., 2006). In our study, PCB 28 was prone to accumulate in the highest amount compared to all individual PCBs. Other PCB congeners accumulated in vegetables with a similar level, and their content was not higher than 200 µg/kg dw. This could be explained by the octanol-water partition coefficient (Kow), which is used to predict the interaction between pollutants and cultivated plants. A higher uptake of PCB 28 than other congeners could be expected, as this congener had the lowest log Kow of all investigated PCBs (Whitfield Åslund et al., 2007).
From the results of PCBs in the total biomass, it was obvious that radishes and carrots planted on Fluvisol had a higher ability to accumulate individual PCBs than from Chernozem. This is in concordance with a study by Javorská et al. (2011), who investigated the fate of 7 PCB congeners in Fluvisol from different localities in the Czech Republic planted with parsley or red beet. This could be related to the agrochemical properties of the experimental soils (Table 1). Our Fluvisol contained 83% sand and 1.1% soil organic matter, which is a much lower amount of organic matter than that in Chernozem (2.9%). Therefore, it could be expected that highly chlorinated PCB congeners (PCB 118, PCB 138, PCB 153 and PCB 180) have a greater affinity for soil organic matter and are more persistent in the environment than the lower ones, which are also considered to be more degradable (Backe et al., 2004). Due to this fact, selected PCBs were probably not firmly adsorbed on soil organic matter of Fluvisol, and they were able to be obtained in the soil solution, being more bioavailable than in the Chernozem soil.
A much lower PCB content was found in the inner part of the vegetables in comparison to the respective vegetable peels (Table 3). Tested PCBs were also found in the shoots of vegetables, but their content was on the edge of the detection limits of PCBs (Table S5). Low amounts of PCBs found in the inner parts of roots indicated that the possible uptake of PCBs by aboveground biomass is very limited. This could be supported by the calculated translocation factors shown in Table 4. All translocation factors were lower than 1, indicating a poor ability to take up PCBs from contaminated soil. Kacálková and Tlustoš (2011) also found only low chlorinated PCB 28, PCB 52 and PCB 101 in shoots of maize, sunflower and poplars (45 − 135 µg/kg dw) in a field experiment with very low bioaccumulation factors (0.03 − 0.09). Nevertheless, the bioaccumulation factors of carrots in this study were close to 1, indicating the potential to accumulate PCBs in higher amounts. Similar PCB distributions to our study have been found in crops (turnip, sweet potato and broccoli) grown around the waste-dismantling area in China (Liu et al., 2020).
Table 4
Bioaccumulation and translocation factors of total PCBs in vegetables cultivated on PCB-contaminated soils.
Vegetable | Factor | Chernozem | Fluvisol |
Control soil | PCB soil | Control soil | PCB soil |
Radish | Bioaccumulation | – | 0.49 ± 0.01a | – | 0.69 ± 0.01b |
Translocation | – | 0.01 ± 0.00a | – | 0.03 ± 0.01b |
Onion | Bioaccumulation | – | 0.46 ± 0.02a | – | 0.46 ± 0.03a |
Translocation | – | 0.01 ± 0.00a | – | 0.01 ± 0.00a |
Carrot | Bioaccumulation | – | 0.87 ± 0.04a | – | 0.95 ± 0.02b |
Translocation | – | 0.02 ± 0.01a | – | 0.03 ± 0.00b |
All values represent the mean of four replications. – not estimated, as the content of PCBs in biomass was below the limit of detection. Different lowercase letters within the same row indicate significant differences between treatments (Tukey test; P < 0.05) |
Unfortunately, from the data obtained in the current experiment, it is not clear whether some PCB degradation can occur in plants. The low PCB accumulation in the inner part of the vegetables indicated that the PCBs were only adsorbed on the root surface. This means that individual PCBs are not able to enter the epidermal plant cells of roots, thereby suppressing symplast/apoplast transfer of PCBs in plant roots. Furthermore, from the obtained results, the possible translocation of PCBs from plant roots to their shoots was not evident. Therefore, the contamination of shoots by PCBs could be caused by the volatilisation of PCBs directly from contaminated soil during the vegetation period of vegetables, as suggested by Salihoglu and Tasdemir (2009). This is supported by the results of Wang et al. (2023), who used clover in their phytoremediation of PCB-contaminated soil and detected no PCBs in either the roots or the aboveground parts of the clover. The low PCB concentrations in the soil suggest that the removal of PCBs was mainly due to rhizosphere microorganisms. In general, root exudates can induce the desorption of soil residual PCBs and secondary metabolites of root exudates induce the biphenyl catabolic pathway to degrade PCBs in soil.
3.2 Dissipation of PCBs from non-planted and vegetable-cultivated soils
The relative removal of individual PCB congeners and their sum in soils and the contribution of plants to the total dissipation of PCBs from experimental soils are shown in Fig. 2. The highest relative removal of PCBs from soil was observed in PCB-contaminated Fluvisol with carrots, accounting for 27.2%. This removal of the total sum of PCBs was significantly (P < 0.05) higher than in other realised treatments in the range from 12 to 17%. As mentioned above, this observation could be related to the soil type and the physical and chemical properties of the experimental soil. The total sum of PCBs (Fig. 2) also indicated that PCBs could be removed and taken up by selected plants more easily from acidic soil with a low pH reaction, especially together with a low amount of soil organic matter (Table 1).
From the individual PCBs, the removal of PCB 28 was higher than that of other PCB congeners with a higher molecular weight. This could be linked to a higher bioavailability of PCB 28 in tested soils due to the ability to be dissolved in solution in a higher portion, which is further consumed by roots. The removal of PCB 28 from soil planted with carrots was nearly 7%, and there was no significant difference between the removal from Chernozem and Fluvisol. The remaining PCB congeners were less removed < 5% with no differences at P < 0.05.
As indicated from the relative dissipation of PCBs in soils (Fig. 2), there were no significant differences (P < 0.05) between tested treatments, including the control soil. This means that the root vegetables of the planted treatments were not able to enhance the ability of soil microorganisms to degrade these compounds in comparison to the soil without plants. Nevertheless, the total soil PCB removal was in the range of 11.5–14.6%, which could not be considered negligible. This relatively high removal could be caused by the higher activity of soil autochthonous microorganisms involved in PCB degradation, which was probably enhanced by fertilisation prior to the establishment of the experiment. Similar behaviour of PCBs in soil to the current study has been reported by Javorská et al. (2009), who found that the depletion of PCBs in the fertilised rhizosphere of rape was significantly (P < 0.05) higher in the Fluvisol by 32 − 38% with lower organic matter content than in Chernozem. Nevertheless, the remaining PCB content in soil was still high and could be hazardous for the whole biota in the environment, as reported by Xu et al. (2019a). Soil microorganisms can be boosted for the reduction of PCBs in the soil by biochar amendment, as it could be used as the medium of PCB adsorption to reduce plant uptake and PCB accumulation in soil microorganisms (Valizadeh et al., 2021; Silvani et al., 2019). Dissipation of PCBs in soil can also be accelerated by the addition of organic matter, such as sludge from wastewater treatment plants (Xu et al., 2019b). The abundance and potentially the most successful PCB-degrading bacterial genera (Bacillus, Streptomyces, Ramlibacter and Paenibacillus) in different soil types was reported by Xu et al. (2020c). The most promising strategies for degrading PCBs in soil are graphene oxide-assisted bacterial agent amendment (Li et al., 2023) and pyrophosphate-chelated Fenton-like reaction (Sun et al., 2020) for soil remediation.
The overall contribution of plant uptake to the relative removal of total PCBs from soil is shown in Fig. 2. The best results were obtained in soils planted with carrots compared to radishes, onions and bare soils. The contribution of plants to the total removal of PCBs was 2% at maximum, which was achieved in treatments where carrots were cultivated. Only PCB 28 removal of 3.6% was slightly higher from Fluvisol than in the case of Chernozem. The removal of the remaining individual PCBs from neutral and acidic soils was almost the same. The soil organic matter together with the time of PCB ageing and other soil physicochemical properties diminish the bioaccessibility for strengthened adsorption; therefore, less accommodation of PCBs in soil pore water can be expected (Ti et al., 2018; Shen et al., 2019). Based on the data obtained from the conducted experiment, the root vegetables are not able to accumulate PCBs in a relatively high amount. Therefore, according to Public Notice No. 53 (2012) about the chemical requirements of foodstuff and other materials for health protection in the Czech Republic, the cultivation of root vegetables at a contamination level of nearly 1500 µg/kg dw in Chernozem and Fluvisol soils can be considered safe.