Glyphosate residues in coffee beans

Brazil stands out as the world's largest producer and exporter of coffee, with production in the coffee year 2024, including the sum of the species of Coffea arabica (Arabica coffee) and Coffea canephora (robusta+conilon), reached a physical volume equivalent to 54,21 million processed 60kg bags. Produced in an area of ​​1,88 million hectares. Coffea arabica totaled the equivalent of 39,59 million 60kg bags, which correspond to 73,03% of national production. Coffea canephora, reached the equivalent of 14,61 million 60kg bags, corresponding to 26,97% of the sum of the total Brazilian harvest of the two species.

Coffee farming stands out as a high-value-added economic activity that generates income and jobs even on small properties and mountainous terrain. To remain competitive, producers need to invest in productivity, cost control and good agricultural practices.

One of the main challenges is weed control, which can impact productivity by more than 40%. Although there are alternatives such as mowing, glyphosate is still the most widely used herbicide and stands out for its ability to control broad- and narrow-leaf species and for its relatively low toxicity and favorable cost-effectiveness.

To maintain sustainable practices, a limit of three glyphosate applications annually was established: the first after the start of the rainy season in October, the second in November-December and the third in March.

In southeastern Brazil, coffee plants go through different stages throughout the year: budding (August-September), flowering and fruiting (October to March), hilling (April-May) and harvesting (June-July). Glyphosate applications, with doses between 480 and 2.880 g of active ingredient/ha (between 1,0 and 6,0 L/ha of the commercial product Roundup), target weeds emerging between and within crop rows, from seeding to fruiting.

After hilling, it is advisable to apply residual pre-emergent herbicides to limit weed germination before coffee harvesting.

However, due to cost considerations, many farmers use glyphosate to clear the area and facilitate harvesting. This is because the last application of glyphosate should be aligned with the end of the rainy season (March). Post-hilling applications fall within the 15-day safe re-entry intervals established by the National Health Surveillance Agency (Anvisa) for this herbicide.

Application technology is essential to ensure the efficient and safe use of agricultural pesticides. This includes everything from the correct selection of equipment and spray nozzles to training of applicators. Backpack sprayers are common on small farms, while tractor-mounted sprayers dominate large areas. Incorrect application can cause waste, contamination and risks to the environment.

Recent reports, such as that of Pro-Huma, show how Brazil applies these products in different states and production scales, highlighting the importance of correct management for the sustainability of Brazilian coffee farming.

One of the major concerns about the use of herbicides is their environmental behavior. When a chemical is applied, a large part of it is retained in the target weeds, but a fraction may reach the coffee plant and some ends up in the soil. The disappearance of a pesticide in the environment involves both physical processes and chemical, photochemical or biological transformations, which depend on the type of soil and the climatic conditions of each region.

The development of more precise chemical analysis methods and equipment allows for the administration of safer laws. One of the major concerns of Brazilian authorities, which regulate the registration of these products in Brazil, is in relation to the level of residues in food and soil, requiring the registering companies that these tests be specifically carried out in Brazil, for each specific use (ANVISA, 2019).

Maximum Residue Limit (MRL) is defined as the maximum amount of residue of a pesticide or related substance officially accepted in food, as a result of appropriate application at a specific stage, from its production to consumption, expressed in parts (by weight) of the pesticide or related substance or residues per million parts of food (by weight) or ppm or mg/kg. The term tolerance, used in the USA, is in most cases synonymous with MRL. In another definition, it is the maximum concentration of a pesticide residue (expressed as mg/kg), recommended by the Codex Alimentarius Commission as legally permitted in foodstuffs and/or animal feed (WHO, 1997).

The magnitude of residues in food is therefore a very complex issue. The final residue level of a herbicide depends on the method of use, the quantities applied, the number of applications in the harvest, and the processes of absorption, translocation, assimilation, exudation, metabolism and degradation by plants. These, in turn, depend on the crop species, the physiological maturity of the plants, and the environmental conditions (relative air humidity, ambient temperature, luminosity and soil moisture).

The amount of pesticide residues on coffee beans is influenced by several factors, including application methods (spray equipment, protective devices, nozzles, pressure and volume) and timing relative to the phenological stage of the crop. For example, when glyphosate is applied close to harvest, when the berries are ripe or dry, there is a high risk of increased residues on coffee beans because the berries are no longer undergoing physiological maturation processes. In addition, coffee harvest coincides with a period of low rainfall, making it difficult for the herbicide to leach and degrade, raising the possibility that glyphosate concentrations in coffee beans may exceed the maximum residue limit (MRL).

Anvisa has set an MRL for glyphosate in food at 1,0 mg/kg, which is equal to or lower than standards in countries such as the United States and Japan. However, coffee importers, particularly in Europe, have stricter glyphosate MRLs, with a maximum of 0,1 mg/kg (European Food Safety Authority, 2019), which is ten times lower than Brazil’s MRL. Furthermore, scientific studies on glyphosate residues in Brazilian green coffee beans are scarce.

Within this vision, a group of researchers from FEAGRI-UNICAMP (LL Foloni) and UNESP – FCA (ED Velini, CA Carbonari and RAde La Cruz) and the Institute of Biosciences (JD Rodrigues and EO Ono) – Botucatu Campus, proposed to carry out an experiment to determine the levels of glyphosate residues in green coffee beans, considering different methods of herbicide application (mechanical or manual equipment with and without protective devices, directed to the soil and the lower third of the coffee tree), using various types of spray tips (nozzles) and flow rates, and different stages of beans - from green coffee to dry beans - and at different sampling times after herbicide application.

The experiments were carried out at the Santa Adelina farm (22°5.554' S and 48º45.7' W), in Bariri, central region of the State of São Paulo, at an altitude of 439 meters, in a Red Latosol with a clayey texture, in adult crops, 2,5 m high, with spacing of 3,2 m X 0,8 m.

In the experimental area, normal crop treatments were carried out, and two applications of glyphosate of 2,0 L/ha were made in October and another in February. Nine treatments with glyphosate were evaluated (Roundup Original DI), using mechanical and manual backpack sprayers. The treatments were distributed in plots measuring 16 m wide (5 rows) and 50 m long, in a completely randomized design, with herbicide application on both sides of the three central rows.

All treatments were carried out after the street planting phase, which is outside the rainy season and the recommended period for glyphosate applications in coffee crops. Most treatments were applied on April 2, 2020, except for T7 and T8, carried out on May 17, 2020. The dose applied in treatments (1 to 8) was 1.850 ge a./ha, and (9) 4.625 ge a./ha (respectively 5,0 and 12,5 l/ha of the commercial product; Roundup Original DI).

The nine treatments were applied, as summarized below:

T-1.- Mechanized application – Tractor sprayer with protected boom (PH 200) equipped with four nozzles, two TK-VS-03 and two 8003. Application height - 30-40 cm.

T-2.- Application with a manual backpack sprayer with a protected TK VS 02 tip (Napoleon hat), directed towards the soil to avoid reaching the coffee tree skirt, with a spray volume of 497 L/ha, and application height varied from 40 to 50 cm.

T-3.- Application with manual backpack sprayer with TK VS 02 tip - unprotected, directed towards the ground to avoid reaching the coffee tree skirt, with the same volume of spray as before and at the same heights.

T-4.- Application with manual backpack sprayer - reboleira type with AI 02 tip, and application height varied from 40 to 50 cm, in the space between the edges of the trees and the ground, avoiding contact with the skirt of the coffee plants.

T-5.- Application with a manual backpack sprayer, directed to the lower third of the coffee trees, the height varied from 50 to 70 cm, simulating real scenarios where unintentional errors or inattention by the applicator may occur, with AI 02 tip, with a spray volume of 180 L/ha.

T-6.- Application with manual backpack sprayer directed to the lower third, the height varied from 50 to 70 cm, simulating real scenarios where unintentional errors or inattention of the applicator may occur, reaching the lower 1/3 of the coffee plantation skirt, with TK VS 02 tip, with a spray volume of 497 L/ha.

T-7.- Application with manual backpack sprayer with protected TK VS 02 tip (Napoleon hat), with a spray volume of 497 L/ha. The application height varied from 40 to 50 cm, in the space between the edges of the trees and the soil, avoiding contact with the skirts of the coffee plants, 15 days before harvest.

T-8.- Application with manual backpack sprayer with TK VS 02 tip – unprotected), with a spray volume of 497 L/ha. The application height varied from 40 to 50 cm, in the space between the edges of the trees and the soil, avoiding contact with the skirts of the coffee plants, 15 days before harvest.

T-9.- Application with a manual backpack sprayer, directed to the lower third of the coffee trees, directly and purposefully reaching the lower third of the coffee plantation skirt with a TK VS 02 nozzle, with 2 ½ X the dose, with a spray volume of 180 L/ha.

Harvest time - For each treatment, on each sampling date, four replicates were collected (200 mL plastic containers filled with coffee beans). Depending on the sampling date, which ranged from zero (beginning of the experiment) to harvest, 60 days later (60DAT), the ripeness of the coffee beans varied from green, red (ripe) and dry coffee.

Following the methodology for working with residues, the samples, immediately after collection, still in the field, were stored in a thermal box with ice during transport (±2 h) before being stored at -18 °C until the time of analysis. The conventional method for detecting glyphosate residues in soil and coffee beans uses high-performance liquid chromatography (HPLC).

In this experiment, high-performance liquid chromatography and mass spectrometry were performed using a liquid chromatograph coupled to a mass spectrometer (LC-MS/MS). These equipments can detect glyphosate concentrations as low as 0,01 μg/mL (microgram per gram of plant tissue), meeting the strict requirements imposed by green coffee importers.

Thus, the glyphosate content was determined in micrograms per gram of dry plant tissue (μg/g) based on the herbicide concentrations found in the samples. As a result, the glyphosate residues detected in the coffee beans varied according to the type of application, nozzle or tip used, with or without protection, spray volume used and time elapsed between application and sampling.

Analysis of the data obtained from the coffee samples, as shown in the following graph, shows that:

• The lowest levels of glyphosate residues in coffee beans were detected in applications that used a protective device on the spray bar, a protected bar on the mechanical sprayer, and a Napoleon hat on the manual backpack sprayer with a TK VS 02 tip, that is, in treatments T1, T2 and T7. The protected application ensured that glyphosate residues remained below the MRLs.

• The levels of glyphosate residues in coffee beans were insignificant, respecting the MRLs established by national and international regulatory authorities in the correct application of doses of 5,0 l/ha of the commercial product.

• Roundup, using mechanical and manual equipment, with a protective device to ensure that the herbicide does not come into direct contact with the lower third of the coffee plants.

• In contrast, applications made without the use of a protective device, using conventional nozzles (TK-VS-02, high flow and low impact) or anti-drift (AI11002, low flow induced by air), but reaching the lower third of the coffee plants, resulted in high levels of glyphosate residues in the coffee beans. Under these conditions, despite the gradual decrease in residue levels over time, glyphosate concentrations exceed the MRLs permitted by Anvisa and international standards, even when applications are made 60 days before harvest.

* By Luiz Lonardoni Foloni, agricultural engineer