Registration Dossier

Ecotoxicological information

Toxicity to terrestrial plants

Currently viewing:

Administrative data

Link to relevant study record(s)

Description of key information

The chronic NOEC for toxicity to plants is 999 mg Sb/kg dw (Smolders et al., 2007).

Key value for chemical safety assessment

Additional information

There are three studies available on Sb toxicity to plants that included analytical confirmation of the exposure concentrations. However, as stated above, only the study by Smolders et al. (2007), which resulted in a bounded NOEC of 999 mg Sb/kg dw, will be used when deriving PNECsoil.

Study 1

(Smolders et al., 2007) studied the toxicity of Sb to plants (lettuce emergence and growth, and barley root elongation) in Sb2O3amended soil, which had aged for 31 weeks in the field (starting 2006-12-19). The dose levels used were 3, 90, 322, 999, 2930 and 10119 mg Sb/kg dw (measured concentrations).

For the plant growth assay using lettuce (ISO 11269-2), the number of seedlings that emerged per pot and the shoot yield of the plants were determined. The soils were fertilized and pre-incubated for six days before twenty lettuce (Lactuca sativa) seeds were sown per pot. Four pots per concentration were used. The plants emerged after three days, and at day six the number of seedlings was counted and the plants were thinned out to five plants per pot. Fourteen days after 50% of the control plants (day 3) had emerged, plants were harvested and the fresh biomass was weighed.

According to guideline ISO 11269-2 from 1995, emergence should be sufficient to provide five healthy seedlings per pot in the control. In the study, six healthy seedlings were recorded in one control replicate, and 7-16 seedlings were recorded in three replicates, resulting in an arithmetic mean number of 10 seedlings per control pot. However, according to guideline ISO 11269-2 from 2005, the emergence should be at least seven healthy seedlings out of 10 planted seeds, which means that the lettuce emergence is not valid according to these newer validity criteria.

The lettuce assay resulted in a bounded NOEC of 2930 mg Sb/kg dw, a LOEC of 10119 mg Sb/kg dw and an EC10 of 4517 mg Sb/kg dw for lettuce biomass (shoot) yield, and an unbounded NOEC of 10119 mg Sb/kg dw for lettuce seed emergence.

To conclude, this means that the lettuce results (emergence and growth) from this study are valid and reliable according to the guideline under which it was performed, but the results on emergence are not compliant with the criteria in the most recent ISO guideline (from 2005). However, even if emergence in the control group had been 70 % (and the study therefore would have been valid according to the most recent ISO guideline), an effect on emergence would still not have been detected. Based on this, the results are useful since they provide a very strong indication that growth is a more sensitive endpoint than emergence for lettuce.

In the barley root elongation assay (ISO 11269 -1), barley (Hordeum vulgare) was pre-germinated for three days and four seeds were planted in each pot. Three replicates per Sb concentration were used. After five days of growth, intact roots were washed out of the soil matrix and the length of the longest root on each plant was recorded. The mean length of the longest root of all replicate samples per soil was determined. There are no validity criteria given in the guideline. However, a mean root elongation in the controls at test end of 9.1 cm with a relative standard deviation of 5.1% indicates that the results from the root elongation test are valid. The root elongation study resulted in a bounded NOEC of 999 mg Sb/kg dw, a LOEC of 2930 mg Sb/kg dw and an EC10 of 1931 mg Sb/kg dw.

The Sb pore water concentrations corresponding to the bounded NOECs of 999 and 2930 mg Sb/kg dw were 9.7 and 18.7 mg Sb/L respectively.

In order to answer the question of whether the soil amended with Sb2O3and aged for 31 weeks was fully reacted and equilibrated, the authors used modelling of the Sb2O3solubilization process in soil based on all previously available data. It was found that 31 weeks of ageing in the field was sufficient to have fully reacted and equilibrated soils at high doses, but not at low doses. For example, the model predicts an Sb pore water concentration of 26 mg Sb/L at 999 mg Sb/kg dw whereas the observed concentration in this study was 9.7 mg Sb/L.

As long as equilibrium is not reached the pore water concentration will gradually increase during the course of time. How much the pore water concentration will have increased, from the initiation of the study until termination, is not possible to know since the pore water concentration was only measured at the end of the study. However, the difference in pore water concentration between the initiation and termination of the bioassay is not considered to be of importance because the test started after 31 weeks of ageing and the bioassay continued only for another 5 d. As a consequence, a sufficiently constant toxicity pressure is considered to have been maintained during the bioassays.

During the 31 weeks ageing period a substantial part of the Sb2O3was transformed into soluble Sb. It is expected that further transformation of Sb2O3after this ageing period would be very slow and so a sufficiently constant toxic pressure is considered to have been obtained during these bioassays. Using Sb2O3amended soils avoids confounding effects of counter-ions or lowered pH, and therefore observed toxic effects can be attributed to the increasing Sb dose only. However, since not all Sb2O3had dissolved during the aging period used, the NOEC of 999 mg Sb/kg dw would underestimate the toxicity at complete transformation of Sb2O3into soluble forms. This is because the equilibrium pore water concentration was not reached at this test concentration during the study. Therefore, the NOEC is based on the porewater concentration (9.7 mg Sb/L), which is multiplied by the equilibrium solid:liquid distribution coefficient (Kd) for Sb in this soil. The Kd value for the soil used in the present study is 38 L/kg, which is the value observed for the Sb2O3amended soil aged for five years and for the soluble SbCl3added to soil (Oorts et al., 2005). The resulting NOEC after having performed this calculation is 370 mg Sb/kg dw (= 9.7 mg Sb/L x 38 L/kg).

Study 2

In the second study, Oorts et al. (2005) studied the toxicity of Sb to plants (lettuce shoot yield) in freshly spiked soil using Sb2O3(measured concentrations 0.6, 12.4, 34.5, 67, 124, 422, 897, 1804 mg Sb/kg dw) or SbCl3(measured concentrations 0.6, 10, 43.2, 73.1, 159, 384, 836, 1741 mg Sb/kg dw), and in five year aged Sb2O3spiked soil (measured concentrations 0.4, 5.8, 28.4, 71.7 and 116 mg Sb/kg dw.). In addition, a plant growth test was also performed in the field in the aged Sb2O3spiked soil.

In order to study the potential effect of the counter ion (chloride) resulting from spiking with SbCl3, soil was also spiked with CaCl2at equivalent chloride doses using identical preincubation and spiking procedures. Soil solution from all Sb amended soils was extracted four weeks after spiking by centrifugation and analysed using ICP within two days after extraction. Soil solution was also analysed with respect to pH and electrical conductivity. 

The lettuce shoot yield test was based on ISO 11269-2. Twenty lettuce seeds (Lactuca sativa cv.) were sown in three replicate pots for each treatment. Following emergence, seedlings were thinned to five plants per pot. Plants were harvested 24 days after sowing. At harvest, shoots were cut just above the soil surface and dried at 70 ºC for at least 48 hours, and the dry matter yield was recorded.

In addition to the plant test performed in the laboratory, a field study using aged Sb2O3spiked soil was also performed, using 18 day old pre-cultivated plants. Four plants were planted on each container with aged soil (control, and the originally added concentrations 10, 50 and 250 mg Sb/kg). After an exposure period of two months the plants were harvested (2005-09-15). Above ground biomass and root weight were measured for each plant. Edible parts were selected and oven dried at 50 ºC. Dry plant material was ground and used for Sb analysis. Total metal concentrations were determined by boiling nitric acid digestion and subsequent analysis with inductively coupled plasma - optical emission spectroscopy (ICP-OES).

The use of freshly spiked Sb2O3resulted in an unbound NOEC of 1804 mg Sb/kg dw (nominal added concentration of 2000 mg Sb/kg dw).

The use of freshly spiked SbCl3resulted in a NOEC of 43 mg Sb/kg dw (nominal concentration of 50 mg Sb/kg dw).

The use of five-year aged aged Sb2O3spiked soil in a laboratory test resulted in dose-response effects of Sb, and a NOEC of 28 mg Sb/kg dw (nominal concentration, 50 mg Sb/kg dw). Oorts et al.(2005) did not consider that this result could be correct and found that the container with the highest concentration was located next to a hedgerow that had been treated with an herbicide (CANYON, a mixture of glyphosphate, diuron and diflufenican). The second highest concentration resulted from mixing soil from the highest dose with a lower dose, and this had therefore also been potentially contaminated with the herbicide. Thus, the effects observed in the two highest doses may be confounded due to the presence of an herbicide.

Another test using the aged Sb2O3spiked soil in a field study with 18 day old precultivated plants resulted in unbounded NOECs of 116 mg Sb/kg dw for both lettuce shoot yield and root yield. However, the experimental set-up for the lettuce grown in the laboratory and in the field differed in several important aspects, which is why it is difficult to draw any conclusions based on the difference in toxicity (e. g. the field tests were initiated with 18 day old pre-cultivated plants while seeds were used in the laboratory tests).

Since the original test design for the aged Sb2O3spiked soils tested in the laboratory included four doses and a control, removing the two highest doses results in two doses and a control. The highest remaining dose is 28 mg Sb/kg dw. The evaluation of the data using a step-down approach results in a NOEC of 5.8 mg Sb/kg dw (nominal added concentration of 10 mg Sb/kg dw). There are, however, several concerns about this value: (i) there are only two doses above the control (of which the lowest is the NOEC and the other is a LOEC), (ii) the NOEC value resulting from SbCl3is 43 mg Sb/kg (for which the observed effect is attributed to the effect of the chloride), (iii) the Sb pore water concentration, which is used as a simplified measure of toxic pressure, was far below the critical pore water concentration in the freshly spiked soils (the largest pore water concentration in the aged soils was 0.7 mg Sb/l, whereas the lowest LOEC in freshly SbCl3spiked soils was 3.6 mg Sb/l), and (iv) the maximum internal Sb concentration in the shoot for the five-year aged Sb2O3spiked soils (1 mg Sb/kg dw) was far below the value at which toxicity was observed in soil freshly spiked with Sb2O3or SbCl3(>10 mg Sb/kg dw).

Study 3

Hartley et al. (1999) investigated the effects of antimony exposure on Scots pine (Pinus sylvestris). In two experiments plants were either exposed to a natural soil that had been contaminated by a chemical accident or to control soil that was amended with antimony tartrate.

The natural soil was contaminated with multiple metals (Cd, Pb, Zn, Sb and Cu) and so the effects cannot be attributed solely to antimony. For this reason, this part of the study was considered unreliable.

In the second part of the experiment antimony tartrate was added directly to a pot containing the 6 month old seedlings, so there was no equilibration time. In addition, although the method states that soil concentrations were measured at 6, 7, 8 and 9 month harvests the results of these measurements are not reported in the paper. No effects on shoots or roots were observed at the single nominal concentration of 0.3 mg Sb/kg tested. For these reasons, this study is not considered reliable.