Transfer of perchlorate into different plant parts – Can perchlorate contaminations of soils explain perchlorate findings in vegetables and fruits?

Transfer von Perchlorat in unterschiedliche Pflanzenteile – Können Perchloratfunde in Obst und Gemüse über Perchloratkontaminationen im Boden erklärt werden?

Journal für Kulturpflanzen, 73 (11-12). S. 353–363, 2021, ISSN 1867-0911, DOI: 10.5073/JfK.2021.11-12.02, Verlag Eugen Ulmer KG, Stuttgart

Correspondence

Dr. Elke Bloem, Julius Kühn Institute (JKI) – Federal Research Centre for Cultivated Plants, Institute for Crop and Soil Science, Bundesallee 69, 38116 Braunschweig, Germany, email: elke.bloem@julius-kuehn.de

Submitted/Accepted

1 September 2021/2 November 2021

This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/deed.en). Dies ist ein Open-Access-Artikel, der unter den Bedingungen der Creative Commons Namensnennung 4.0 International Lizenz (CC BY 4.0) zur Verfügung gestellt wird (https://creativecommons.org/licenses/by/4.0/deed.de).

Materials and methods

Fertilization was split in two doses in case of Swiss chard and May turnip and major nutrient application was split in 10 doses in case of tomato. Spinach and radish with a shorter vegetation period received nutrients in one dose.

Perchlorate was applied at rates of 0, 5, 10, 100 or 500 μg ClO₄^-/pot. A perchlorate stock solution containing 5 g/L was prepared from NaClO₄ * H₂O. The stock solution was diluted 1:100 with water and from this solution the necessary amounts per pot were diluted with water to 50 mL before application. Swiss chard had 4–5 developed leaves, radish 2–3 leaves, May turnip 3–4 leaves and tomato 6 leaves when dissolved perchlorate was poured onto the substrate surface without contaminating the young plants. All treatments were performed in 5-fold repetition and the pots were randomized in the greenhouse.

Above-ground biomass of spinach was cut twice from the same plants. May turnip was harvested intermediately leaving three good developed plants per pot. Originally nine seeds were sown and the other plants that were grown were removed for the intermediate harvest. The other three species were harvested only once at ripening. In case of spinach and Swiss chard the above-ground biomass was harvested. Radish and May turnip plants were divided into leaf material and bulbs or beets, respectively. Bulbs and beets were washed with distilled water to remove substrate particles before the fresh weight was recorded. Six to nine plants (with eight plants on average) of radish were harvested as not all seeds germinated. Tomato plants were divided into stems, leaves and fruits. All tomato fruits per pot were combined, red and green ones.

Fresh yield was determined for all plant parts of all replicates of the vegetables separately, then the material was immediately shock-frozen in liquid nitrogen and later on ground under liquid nitrogen. Perchlorate was determined in the fresh plant material with an Agilent 1200 high pressure liquid chromatography (HPLC) system (Wilmington, DE, USA) which was connected to a 4000 Q-Trap LC/MS/MS system with an electro-spray ionization (ESI) source (Applied Biosystems MDS Sciex, Toronto, ON, Canada). Some selected fertilizers from an old fertilizer collection compiled between 1970 and 1990 (old Chilean caliche product and a fertilizer derived from Chilean caliche) and some commercially available fertilizers bought in 2014 were analyzed for perchlorate contaminations as well to answer the question if relevant contaminations with perchlorate can be expected in vegetables when fertilized with these fertilizers.

Fertilizers were solubilized in distilled water (1 g in 100 mL). 50 μL of isotopic internal standard (IS) (containing 40 μg/of NaClO₄) was added to an aliquot of 10 mL of this fertilizer solution and filtered via syringe filters into vials ready for the measurement at the LC/MS/MS.

The internal standard was produced from an isotopic standard material of perchlorate (LGC Standards GmbH, Wesel, Germany) with a molecular weight of 130.44 g/mol (NaClO₄). The calibration curve was produced from NaClO₄ (NaClO₄ * H₂O, MW = 140.46 g/mol). For the calibration curve the quotient from the peak area of the sample (mass transfer from m/z: 99/83) and the peak area of the internal standard (m/z: 107/89) was recorded against the concentration. The concentration calculation of the samples is performed via formula [1] according to Anastassiades et al. (2016).

m_p: is the added mass of the internal standard (50 μL of IS containing 40 μg/mL NaClO₄ ≙ 2 μg NaClO₄)

m_s: is the added mass of IS (2) to the standards (50 μL of IS (2) containing 2 μg/mL NaClO₄ ≙ 0.1 μg NaClO₄)

SW: sample weight [g]

Fig. 1. Growth of May turnip, radish, Swiss chard and spinach in relation to perchlorate application [in μg ClO₄^-/pot] at harvest time.

Fig. 2. Growth of tomato plants in relation to perchlorate application [in μg ClO₄^-/pot] at harvest time.

Leafy vegetables like spinach and Swiss chard showed a very rapid uptake and transport of perchlorate into vegetative plant parts (Tables 2 and 3). Also the control plants
showed a low perchlorate contamination in leaves indicating to a possible contamination of the substrate which was not analyzed.

Spinach had taken up perchlorate nearly completely at the lower application levels by the first harvest. At the higher application rates of 100 and 500 μg ClO₄^-/pot some perchlorate was still taken up into the leaves at the second harvest. In Swiss chard (Table 3) the whole amount of perchlorate was taken up and transported into the leaf tissue within the growth period of 8 weeks.

Edible bulbs or beets of radish and May turnip take up perchlorate via the root system but perchlorate was transported very efficiently into the leaves (Tables 4 and 5). In bulbs of radish only at the highest application rate of 500 μg ClO₄^-/pot low concentrations of perchlorate could be detected and accounted for approximately 4% of the total applied amount. In beets of May turnip even lower amounts of perchlorate were detected at the highest application rate accounting for only 0.3% of the applied perchlorate. Therefore, it can be concluded that perchlorate was taken up by the growing crops and was not stored or deposited by the root system but was almost completely transferred into the aerial plant parts. The perchlorate distribution in different plant organs of tomatoes (Table 6) moreover indicate that perchlorate is retarded in the leaves and no transfer from leaves into fruits was observed. Also the stems contained only a low proportion of the applied perchlorate and the leaves contained 3–7 times more perchlorate than the stems at harvest.

Determination of the perchlorate content of modern and historical fertilizer samples (Table 7) revealed that old samples from Chile showed a very high contamination level while in most fertilizers bought in 2014 no perchlorate could be detected. Traces of perchlorate (below the limit of quantification) could be detected in some of the multi-component fertilizers. Only in two NPK fertilizers perchlorate was detected in traceable concentrations which were below the EU limit value for fertilizers of 50 mg ClO₄^-/kg.

The results of the current study clearly showed a transfer of soil applied perchlorate into growing plants. Perchlorate is highly mobile and is taken up un-controlled with other nutrients to be transported via the transpiration stream into leaves. Estrada et al. (2017) could show that perchlorate was transported from solution into plants similar to nitrate and 4-fold more than chloride. The data of radish and May turnip clearly indicate that there is no retention in the roots. May turnip, which is a typical beet root contained less than 1% of the applied perchlorate in the beets and it could be only detected at the highest application rate of 500 μg ClO₄^-/pot. In all other treatments the perchlorate content in the beets was below the lower limit of detection (LLD). Radish as a hypocotyl bulb contained a slightly higher perchlorate proportion of 3–4% of the applied amount in the bulbs and perchlorate was detected at 100 and 500 μg ClO₄^-/pot. The data indicate a very efficient transfer of perchlorate into the leaves. Evaluation of the intermediate harvest of May turnip showed that 44 days after sowing (bulb growth had not yet started) the total amount of applied perchlorate was already taken up into the growing plants. The tomato results (Table 6) underline that leaves contained the highest perchlorate content followed by stems and no transfer into the fruits occurred, when perchlorate was applied once at the beginning of the vegetation period. At the highest perchlorate level (500 μg ClO₄^-/pot) a considerable proportion of 12% of perchlorate was detected in the stems but more than 80% was transferred into the leaves. In the leafy vegetables spinach and Swiss chard this transfer was even more pronounced. In case of Swiss chard perchlorate was taken up and transferred into the leaves to 100% within 56 days from planting to harvest. In spinach the percentage of uptake at the highest dose (500 ClO₄^- μg/pot) was slightly lower with 70% of the total amount at the first harvest but the absolute concentration in the plant material was higher with 7.5 μg/g FW in comparison to 3.5 μg/g FW in Swiss chard (Fig. 3) because of the lower biomass production. Swiss chard produced a much higher mean biomass of 156 g FW/pot in comparison to spinach with 44 g FW/pot at the first harvest which might explain the higher total uptake in comparison to spinach. Calderón et al. (2020) observed a higher perchlorate concentration in spinach in comparison to chard, too, when the same amount of perchlorate was available in the soil. Accordingly, spinach has the potential to accumulate perchlorate to a high concentration in the leaves which is of relevance as high amounts of spinach can be consumed and spinach is a typical vegetable in children’s nutrition.

Fig. 3. Concentration and transfer of perchlorate at the highest application rate of 500 μg ClO₄^-/pot into different plant parts of the investigated crops. In case of May turnip a great perchlorate proportion was bound in the intermediate harvest explaining the lower percentage of perchlorate transfer. The concentration valu­es are given in μg ClO₄^-/g FW and the transfer factor from soil into the plant is given in %.

A high accumulation of perchlorate in leafy vegetables such as spinach, lettuce and chard was reported in other studies as well (Ha et al., 2011; Calderón et al., 2017; 2020). It was observed that perchlorate application in solubilized form as fertigation caused an even higher perchlorate level in spinach than the same amount applied with mineral fertilizer in solid form (Calderón et al., 2020).

Plant organ specific perchlorate uptake of a variety of crops was also reported by other authors (Yu et al., 2004; Tan et al., 2004). Jackson et al. (2004) could show that in accordance with our results tomatoes contained significantly higher perchlorate contents in the vegetative plant parts compared to the fruits. Leafy vegetable such as lettuce, salad and spinach very often showed the highest contamination level with perchlorate when different plant species were compared (reviewed by Bloem & Panten, 2014).

The European Commission has published perchlorate levels for vegetables and fruits for intra-Union trade to protect consumer health (European Commission, 2020). For vegetables and fruits in general this level was set to 50 μg ClO₄^-/kg FW, for leafy vegetables to 500 μg ClO₄^-/kg FW. Foods for infants and young children have much lower reference values with 10 to 20 μg ClO₄^-/kg FW. In the present study these limit values were exceeded in case of spinach and Swiss chard when 100 μg ClO₄^-/pot or more was applied and in radish bulbs with an application of 500 μg ClO₄^-/pot.

Modern fertilizers contain much lower perchlorate concentration than former fertilizers produced from Chilean saltpeter (Table 7). In the present study the highest value was found in a NPK multicomponent fertilizer with a contamination of 45.8 mg ClO₄^-/kg. This fertilizer contained 15% N. To answer the question if such fertilizers represent a risk to contaminate vegetables above the reference values set by the European Commission (2020), a simple worst case scenario was calculated based on a field-grown marketable spinach yield of 250 dt/ha with 144 kg N/ha bound in the total vegetative biomass (400 dt/ha) (data from Feller et al., 2011). If we would need to fertilize the whole N-amount without taking any soil resource into account the NPK fertilizer containing 45.8 mg ClO₄^-/kg and 15% N would deliver 43,968 mg ClO₄^-/ha and could result in a maximum concentration of 1.1 mg ClO₄^-/kg FW in the marketable spinach portion under the assumption that the whole amount of perchlorate was taken up. This value would be above the maximum admissible perchlorate level of 0.5 mg ClO₄^-/kg FW for spinach. In case of the NPK fertilizer containing 20.6 mg ClO₄^-/kg and 18% N, the concentration would be slightly below the limit with a value of 0.41 mg ClO₄^-/kg FW of spinach. The limit for infant formulations would be exceeded with both fertilizers. Such worst case scenario conditions we usually can find under protected farming conditions in glasshouses where soils are more used as a substrate and less like a living soil with the poten­tial to mineralize and supply nutrients. Crops like spinach are often grown under protected conditions. Under such conditions perchlorate does not leach and remain in the soil also with leaf fall and crop residues that stay in the soil after harvest.

This simplified calculation reveals that mineral fertilizers even when they contain much lower perchlorate contaminations than in the last century are still able to contaminate vegetables above maximum admissible levels. Lower perchlorate limit values in fertilizers are the only measure to protect consumers from too high perchlorate contaminations in vegetables. The calculation has shown that already 50 mg ClO₄^-/kg of fertilizer can be too high especially under protected farming conditions when leafy vegetables were grown and the whole amount is taken up by the crops. Moreover, as it is not warranted that the regular intake of low amounts of perchlorate is harmless it must be the target to prevent contaminations of foodstuffs with perchlorates almost completely.

Additionally, it seems possible that, next to mineral fertilizers, other sources can lead to perchlorate contaminations of fruits and vegetables. In the presented study it was shown that no transfer of perchlorate from leaves into tomato fruits occurred (Table 6). Contrastingly, considerable amounts of perchlorate in tomato fruit were detected in other studies. For example, El Aribi et al. (2006) reported values up to 62.8 μg ClO₄^-/kg FW and Hepperle et al. (2013) up to 400 μg ClO₄^-/kg FW in tomato fruits. There are two possible explanations for this discrepancy. Firstly, perchlorate in the current study was applied in one dose at the beginning of the experiment and not with each fertilizer application. According to the UGA (2017) field grown tomatoes are fertilized up to seven times during the vegetation period and if these fertilizers contain perchlorate the plants get perchlorate permanently over a longer duration. Commercial hydroponic greenhouse tomato productions constantly provide fresh nutrient solution to the tomato plants up to an 11 months cropping period. Final conclusions about the possibility that perchlorate can be transported into the fruit cannot be drawn from the present experiment without continuous perchlorate application. Secondly, vegetables and fruits in the reported studies could be contaminated through other sources. According to BfR (2018) cleaning water containing disinfectants is another possible perchlorate contamination source. Chlorine and nowadays predominantly chlorine dioxide (ClO₂) is used as chemical disinfection technology to protect fruits and vegetables and to increase shelf-life. ClO₂ has a broad antimicrobial efficiency and can be used in aqueous and gaseous form to sterilize fresh products (Deng et al., 2020). Moreover, ClO₂ has the potential to preserve the quality of vegetables and fruits by alleviating the browning of fruits and by reducing water losses during storage (Sun et al., 2017; Chen et al., 2018). It is possible that perchlorate is formed as a by-product from chlorine containing disinfectants (BfR, 2018).

Although the presented data clearly revealed that perchlorate-contaminated fertilizers have the potential to contaminate vegetables considerably with perchlorate, other possible sources should be carefully screened as well. For example, high perchlorate findings in organically-grown tomatoes were explained by a possible perchlorate contamination of vinasse, which was produced from conventionally-grown sugar beets (Hinsch, 2013). Against the background of the present results this explanation seems not very reasonable: sugar beets are typical beet roots like May turnip and it seems not very likely that they behave different and accumulate perchlorate in their beets which can end up in the vinasse fertilizer product. Nevertheless, a contamination from other sources cannot be excluded such as from factory water for cleaning the beets. When investigating the origin of such soluble and highly mobile contaminants in foodstuffs it is necessary to be open-minded and to investigate all possible pathways to guarantee a safe food production and to detect and eliminate possible contamination pathways.