Carcinogenic or not?
The broad-spectrum herbicide glyphosate is used all over the world in agriculture. Alongside farming, the chemical is also used for weed-killing in domestic gardens and in public and private spaces kept free from «vegetal invasion», such as railway tracks. Glyphosate has been used since the 1970s in pesticides and was hitherto thought to be harmless at typical levels of exposure. However, since the International Agency for Research on Cancer (IARC) – the specialized cancer-research agency of the WHO – found that glyphosate was «probably carcinogenic to humans» (Group 2A) in a report published in March 2015, the chemical repeatedly made headlines.1 Experts were then divided over whether glyphosate should be re-approved after the expiry of its EU market approval on June 30, 2016. This is because the European Food Safety Authority (EFSA) only recently arrived at the opposed conclusion that it is unlikely that glyphosate is genotoxic or poses a carcinogenic threat.2 The approval of glyphosate was initially extended by 18 months, but at the end of 2017 the question of whether glyphosate should remain in use in the EU will resurface.
Limit values for glyphosate in drinking water
Because chemicals used in farming can seep through the ground and into the ground water, limit values are in effect in some countries concerning the concentration of glyphosate in drinking water. For example, the US Environmental Protection Agency (EPA) forbids any concentrations that exceed the limit of 700 μg/L. In Canada, the maximum permissible concentration is 280 μg/L. Australia and the EU stipulate much lower limit values, at 10 μg/L and 0.1 μg/L, respectively.
Glyphosate and its metabolite AMPA (aminomethylphosphonic acid) are usually determined by HPLC with post-column derivatization and subsequent fluorescence detection (EPA Method 547), or alternatively by ion chromatography coupled with a mass-selective detector. The following will set out the initial results of the determination of glyphosate and AMPA in drinking water in the low μg/L range using ion chromatography (IC) with pulsed amperometric detection. The detection limits for glyphosate and AMPA previously attained with pulsed amperometric detection were around ≥ 50 μg/L . Given this improvement in terms of sensitivity, the method outlined here represents a promising approach to the screening of water and food samples for glyphosate and AMPA.
All determinations were performed with an IC system consisting of a 940 Professional IC Vario ONE with an IC Amperometric Detector and an 858 Professional Sample Processor for automatic sample injection (Figure 1). flexIPAD (FLEXible Integrated Pulsed Amperometric Detection) was used on a gold working electrode as a measuring mode in the amperometric detector. The flexIPAD mode is characterized by its special, multi-stage potential profile. In the determination of glyphosate and AMPA, this produces a stable signal over a longer period of time than with the three-stage potential profile of the regular PAD mode (pulsed amperometric detection). The profile of the potential curve produced in one measuring cycle in flexIPAD mode is presented in Figure 2.
Glyphosate and AMPA were separated on the high-capacity anion separation column Metrosep Carb 2 - 150/4.0. The caustic-soda–acetate eluent used contains 10 mmol/L sodium hydroxide and 300 mmol/L sodium acetate. Under these conditions, AMPA and glyphosate elute after 6.4 and 21.1 minutes, respectively.
The goal of this experiment is to investigate the separation of glyphosate and AMPA in the Metrosep Carb 2 separation column, as well as clarifying the detection using pulsed amperometry and its sensitivity. The Metrosep Carb 2 column is used mainly for separating and determining carbohydrates, sugar alcohols, alcohols, etc. Its high column capacity, combined with the high pH value of the eluent (which at approx. 10 is typical for sugar analysis), results in a large difference in retention time for AMPA and glyphosate. This is because, with a pH value of 10, all three acid groups are deprotonated in part of the glyphosate, meaning that it is partially present as a trivalent anion while the metabolite AMPA, which is missing the carboxyl group, is present as a divalent anion.
In order to accelerate the elution of glyphosate, a flow gradient is used: after AMPA elution at 6.4 minutes, the flow rate is doubled from 0.4 mL/min to 0.8 mL/min. This results in a retention time of 21 minutes for glyphosate. The chromatographic conditions are summarized in Table 1.
Figure 3 shows the chromatogram of the determination of AMPA and glyphosate under the conditions listed in Table 1. An aqueous standard solution was injected containing 10 μg/L each of both components. In order to investigate the suitability of the process for drinking water, tap water from Herisau (Switzerland) was analyzed and mixed with different amounts of AMPA and glyphosate. The concentrations and peak areas found are shown in Table 2.
The detection limits for both components were determined using the signal/noise (S/N) ratio, i.e., the ratio of the peak height to the baseline noise. At the detection limit, the S/N ratio is 3; with smaller values, secured detection is not possible. The detection limit found for AMPA was considerably lower than 1 μg/L, while the limit for glyphosate was approx. 1 μg/L. Figure 4 shows the chromatogram of the drinking water mixed with 2 μg/L glyphosate and AMPA.
For the first time, glyphosate and its primary metabolite AMPA were determined in drinking water in the low μg/L range using ion chromatography with pulsed amperometric detection (flexIPAD). This puts at our disposal a reliable and – compared with HPLC with a mass-selective detector – very inexpensive method for determining the glyphosate and AMPA content in water and foodstuffs. With a detection limit of approx. 1 μg/L, the adherence to limit values for glyphosate can be verified in the USA, Canada, and Australia, among others.