Sorption, mineralization and mobility of N-(phosphonomethyl)glycine (glyphosate) in five different types of gravel.
Monday 06th June 2005
Author: Strange-Hansen R, Holm PE, Jacobsen OS, Jacobsen C
Pest Management Science, 60, 570-578
METHODOLOGY
This study investigates the sorption, mineralization and mobility of glyphosate in gravel systems. Stock solutions of glyphosate were produced using [methyl-14C] Glyphosate. A diluted commercial formulation (Roundup Garden) was used for the column experiments.
Five types of gravel (those most commonly used for surfacing in Denmark) and a sandy agricultural reference soil were used in the experiments. These gravel types were Stoneflour (0-2 mm particle size), washed sand (0-4 mm), concrete gravel (0-8 mm), rounded gravel (2-8 mm) and coarse gravel (0-32 mm). After sampling, soil and gravel samples were stored in the dark at 5 ° C until use.
The six soils were characterized in terms of their water content, pH, water holding capacity, grain size distribution, and surface area. Oxalate-extractable aluminium and iron content were determined using atomic absorption spectrometry. Phosphate sorption ability was measured using spectrometry.
Gravel and soil samples were autoclaved every second day for a total of four sessions for all simultaneous experiments conducted on autoclaved controls. The samples were incubated at 30° C between autoclave sessions, and maintenance of sterility was not checked.
Sorption studies were conducted according to OECD guidelines, but used a soil:solution ratio of 2 rather than 5 to take into account the inhomogeneity of the gravel compared to soil. The analyses were carried out in triplicate. Air-dried substrate was uniformly mixed with calcium chloride solution. [14C] Glyphosate was added at 0.6 mg/kg and allowed to reach equilibrium. The substrate was centrifuged and a 1 mL aliquot of the supernatant was measured for radioactivity by liquid scintillation counting. Mass balance was used to calculate the amount of glyphosate adsorbed on the gravel, as the difference between the amount of glyphosate added and the glyphosate concentration in solution following equilibration. A linear relationship is assumed between glyphosate concentration in solution and glyphosate adsorption, to allow calculation of Kd (distribution coefficient between the concentrations in the two phases) using the following equation:
Cs = Kd.Caq
where Cs and Caq are the concentrations of glyphosate sorbed to the soils and in the solution, respectively.
Mineralization was determined by adding a mixture of [14C] glyphosate and unlabelled glyphosate to moist substrate and measuring the carbon dioxide production over 31 days. Effects of temperature during incubation were tested for by repeating the procedure at 3 temperatures.
Leaching studies were carried out for the six substrate types in columns. The substrates were packed into the columns at the densities recommended for road construction (larger columns were used for the coarse gravel to allow for the larger particle size). Two columns were produced for each substrate type, each exposed to a different precipitation pattern. [14C] Glyphosate mixed with Roundup Garden was applied to the top of the column at 2.4 kg glyphosate/ha (approximately equivalent to 1 m g/g). Over the course of six days, the columns were exposed to three precipitation events: column 1 was exposed on day 0, 2 and 6; column 2 was exposed on day 2, 4 and 6. The amount and intensity of precipitation was equal for all columns (20 mm over a 2 hour period for each event). The experiment was conducted at 20° C.
The effluent was collected through an outlet one day after each precipitation event. An aliquot of this effluent was then measured for radioactivity using liquid scintillation counting. Mineralization was measured during the experiment by trapping and measuring the [14C] carbon dioxide produced.
The experiment was run for nine days, after which the columns were divided into two or three parts (depending on size of the column): the top part (0-7 cm) and bottom/middle part (7-13 cm) and bottom part (13-20 cm). After thorough mixing of each section, the pH and water content were measured. Each section was air-dried and a sample was extracted for [14C] glyphosate/AMPA in the labile phase, and the fraction sorbed to iron and aluminium oxides and sodium carbonate. The sample was also analysed for residual [14C] compounds using liquid scintillation counting.
FINDINGS
Mineralization
The mineralization of glyphosate by indigenous soil micro-organisms varied between the five gravel types, ranging from 11 to 32% after 31 days of incubation at 30° C. In most cases, the most rapid mineralization occurred in the first four days. The smallest gravel size (Stoneflour: 0-2 mm particle size) showed slow, continuous, linear breakdown of glyhosate during the experiment. Mineralization of glyphosate in autoclaved gravel was lower, at 3(+2)%. Similarly, only 2% of the glyphosate added to the reference soil was mineralised, believed to be due to strong adsorption. It is suggested that adaptation of microbial communities may lead to increased mineralization with repeated exposure.
Incubation temperature was found to have a significant effect on the mineralization of glyphosate in the washed sand (0-4 mm). A temperature increase from 10° C to 20° C, and then to 30° C gave rise to an increased mineralization rate of 50 and 100%, respectively. However, there was no significant difference in the rates of mineralization at 20 and 30° C at the 5% significance level. There was no significant effect of incubation temperature on glyphosate mineralization in the stoneflour gravel (0-2 mm particle size). Glyphosate mineralization was more sensitive to temperature change in soil than in gravel.
Sorption
The Kd of glyphosate in the five gravels ranged from 62 to 164 litre/kg. The oxalate-extractable aluminium and iron oxides’ content ranged from 0.015 – 0.056% and 0.001 – 0.009%, respectively. The pH of the five gravels ranged from 9 to 9.4 – a high pH would be expected to decrease the sorptive strength. The higher Kd of the reference soil was believed to be due to the lower pH and higher content of oxalate-extractable aluminium and iron oxides.
Leaching
Over a nine day period, 4-7% of the added [14C] glyphosate was mineralized in the gravels and 0-1% in the reference soil.
In most cases the concentration of glyphosate in effluent was higher than the EU limit of 0.1 m g/litre in drinking water. The highest effluent concentration was found in the rounded gravel (2-8 mm), with the peak concentration being 1350 m g/litre after the first precipitation event. In the columns where precipitation was delayed for two days after application, the effluent concentration was lower in three of the five gravels and the reference soil.
Accumulated leaching (% recovery of the total amount of 14C applied after 9 days) was calculated. The rounded gravel (2-8 mm) and coarse gravel (0-32 mm) showed nearly one fifth of the glyphosate leached, compared to around 1% in the other gravels.
Most of the glyphosate in the columns was in the top layer (0-7 cm). The dissolved glyphosate appeared to flow through the remaining section(s) of the column, except for the rounded (2-8 mm) and coarse gravel (0-32 mm) where 40 and 16% of the glyphosate was detected in the lower 7-13 cm of the columns, respectively.
For the gravels tested (0-2mm, 0-4mm, 0-8mm, 2-8mm, 0-32mm and reference soil) theoretical effluent concentrations would be 11, 16, 6, 12, 10 and 3 m g/litre resectively.
CONCLUSIONS
Micro-organisms present in gravel substrate are able to mineralize glyphosate.
The capacity of gravel to adsorb glyphosate is relatively high.
Precipitation may cause as much as one fifth of applied glyphosate to leach from gravel.