How can you prove that a particular piece of timber does not have illegal origins by analyzing only the chemical elements of the sample? And that coffee really is organic, or soybean is not from an area of deforestation? Agricultural engineer Elisabete Aparecida De Nadai Fernandes, responsible for the Radioisotope Laboratory of the Center for Nuclear Energy in Agriculture of the University of São Paulo (CENA-USP), in Piracicaba, has spent decades working on a technique based on principles of nuclear physics for analyzing foods and plants and seeks to answer these questions.
One of her projects, with funding from FAPESP, aims to develop means to authenticate the origin of Brazilian commodities using nuclear radiation, based on the direct measurement of their chemical elements. The method could become a tool for combatting fraud. “It is one thing to say that a sample did not come from an area of deforestation. Another is to demonstrate this by showing the intrinsic characteristics of the material, which may vary according to its origin,” says the researcher.
In the year marking 50 years since graduating from USP’s Luiz de Queiroz College of Agriculture (ESALQ), Fernandes, 72 years of age, has much to celebrate. In March, she was presented with the George Hevesy Medal Award 2024, the leading international award in nuclear and radioanalytical chemistry. In September, she won the Bunge Foundation Award in the Life and Work category, with the theme “Traceability in food production: Food safety, training, and the reduction of regional asymmetries.”
In this interview, the researcher describes how the neutron activation analysis technique she uses works. The technique is based on a discovery made in 1936 by Hungarian physicist and chemist George Hevesy, who was awarded the Nobel Prize in Chemistry in 1943, and German-Danish physicist Hilde Levi. They discovered that a type of rare earth element became radioactive after exposure to a source of neutrons. Fernandes recalls her experience examining foods contaminated by radioactivity from Chernobyl and the cesium-137 accident in Goiânia, both during the 1980s.
Could the technology you work with promote transparency in Brazilian supply chains for the export of agricultural commodities?
Neutron activation analysis enables the determination of the chemical elements present in a sample, whether it is from an animal, plant, or mineral, thereby obtaining each one’s chemical signature. Applying chemometrics [the science that interprets measurements from a chemical system or process] and artificial intelligence methods to these chemical composition data, we can distinguish the samples. A recent example is a project funded by the CNPq [National Council for Scientific and Technological Development] for the traceability of beef in Brazil. Using neutron activation analysis, we can differentiate between meat originating from five Brazilian regions, Paraguay, Uruguay, Argentina, and Australia. From the chemical composition and by using artificial intelligence, it was possible to differentiate the origin of these meats with almost 100% accuracy.
How does the neutron activation analysis technique work?
It consists of two fundamental steps. One is the activation of a sample of the material to be analyzed by neutron irradiation, producing radionuclides, the excited state of stable nuclides [the nuclei of atoms of the different chemical elements present], which are also called radioactive isotopes or radioisotopes. For this, the sample is exposed to a bombardment of neutrons in a nuclear reactor, which causes the formation of radioactive nuclei in the atoms of the exposed material. To return to their equilibrium state, the atoms emit radiation. Then comes the second stage of the analysis, when the characteristic radiation from these activated nuclei is detected as they transition to a less energetic state. The radioactive activity and the number of radionuclides of the sample gradually diminish. It is what is known as radioactive decay—and each chemical element presents differences in this process. This way, we determine the chemical elements of a sample without the need for its actual chemical processing. We only use principles of nuclear physics.
Does the sample need any preparation?
Yes. It is necessary to take it to a state of pulverization. First it is dried, either in an oven or by freeze-drying [a process of dehydration by sublimation]. Afterwards, it has to be transformed into powder. There are mills to do the grinding with the desired granulometry. The powdered sample is put into ultra-pure polyethylene capsules. It is then sent to the IPEN [Institute for Energy and Nuclear Research] research nuclear reactor in São Paulo. There, it is irradiated with neutrons, and the nuclides from the sample transform—transitioning to the excited state. After two or three days, the samples are released and return to CENA.
The cesium-137 accident that occurred in Goiânia put Brazil on the list of countries contaminated with radioactivity
And what happens there?
Upon returning to their ground state, the samples emit radiation characteristic of the chemical elements contained in them. A spectrometer detects the gamma radiation and transforms this data into electrical signals that create a spectrum of counts as a function of the energy emitted. Depending on the matrix of the irradiated sample, it is possible to determine up to 40 chemical elements at once. The translation, or rather deconvolution of the gamma spectra and the calculation of the mass fractions of the chemical elements are performed using a computer package created by agronomy engineer Márcio Arruda Bacchi. It is a robust tool, especially because it does not require the sample to be dissolved, as is the case with the more common method. This is one of the advantages of neutron activation analysis. To dissolve it, acids are used, but it is not always possible to identify all the chemical elements present in the original sample in the resulting solution, because some of them are destroyed by the acids.
How is the differentiation made in the case of beef? Is there some chemical element in meat from Argentina that is not present in the Brazilian variety?
The chemical elements, in their majority, are the same in the meat from here and there, but there are those that vary according to the location where the animals were raised. They also differ in accordance to breed, diet, well-being while alive, and the type of slaughter. All of this is, in some form, recorded in the chemical composition of the meat. The fingerprint [chemical signature] of that sample will vary according to these characteristics. The distinction is not only based on the individual levels of the elements, but also on the relationship between them, which makes the fingerprint a powerful tool for determining the origin of meat. Throughout our project, we have published the elements that characterize each region and the models that allow the origin of the samples to be identified by the relationships between these chemical elements.
Is there any disadvantage to the neutron activation analysis technique?
I wouldn’t say it’s a disadvantage, but a nuclear reactor is required for the irradiation. Since we don’t have this equipment at CENA, we turned to IPEN. However, it has merits and high metrological rigor. In 2001, I took part in a working group created in Paris to demonstrate the measurement characteristics in pilot studies and comparisons promoted by the International Bureau of Weights and Measures [BIPM] and other organizations. Six years later, the technique was recognized as a primary measurement method by the Consultative Committee for Amount of Substance: Metrology in Chemistry and Biology [CCQM] of the BIPM. It is the highest category for an analytical method.
When did you begin working with this methodology?
Forty-two years ago. In 1982, we initiated a project to deal with a sugarcane problem in the state of São Paulo, when its harvest was still done manually with the help of front-end loaders. The method was designed to detect the presence of residues from soil in sugarcane processed by the plants. These residues ended up damaging the machinery. We created a traceability process and, using the neutron activation analysis technique, were able to determine the amount of soil residue in the sample. Scandium [a soft metal] was chosen as a tracer for soil in sugarcane. It was present in the 10 different types of soil in which the variety had been planted, but not in the sugarcane free of impurities. We wanted to investigate how far these residues could spread and the impact they might have. The traceability proved to be effective in evaluating the amount of earth adhered to the sugarcane, as well as the efficiency of the washing processes. Years later, in 1999, with a PhD thesis on the metrology of Brazilian coffees, we were able to differentiate coffees from organic production, conventional production, and the transition process [between the two systems] by the fingerprint of the samples. Cesium, cobalt, and rubidium were the chemical elements that stood out as tracers for this differentiation. At times they were more concentrated in one system, and at times in the other.
Who else works with this technique in Brazil?
There are researchers at IPEN and at the Center for the Development of Nuclear Technology [CDTN], in Belo Horizonte, but many have already retired. In 2023, I started a thematic project funded by FAPESP focused on the transparency and traceability of the supply chain for Brazilian commodities based on the direct measurement of isotopes and chemical elements. The team includes researchers from IPEN, Embrapa Southeast Livestock, from the INPA [National Institute of Amazonian Research], and from five other countries: Canada, Austria, the Netherlands, Belgium, and the Czech Republic. The commodities involved are beef, timber, and soybean. We have already been working on the project for one year and have managed to perform some analyses. An article focusing on the elemental profile of ipê species to support timber tracing in the Amazon rainforest has just been published in the journal Trees, Forests and People. Although everyone knows the different types of ipê—purple, yellow, white, etc.—it was not known that, due to the colors, there are variations in the chemical composition of the wood, let alone the differences between the various species of yellow ipê. We analyzed trees from two species and the chemical composition varied. It is incredible. This first study with samples collected from the Amazon highlighted that each species of native tree has its own chemical signature. Based on this work, we are producing a database with the fingerprints of different tree species from the Amazon. This database will be made available in publications, especially, and through other means, in accordance with confidentiality policies.
How did you participate in the analysis of foods contaminated by the Chernobyl disaster?
The Chernobyl nuclear accident [in Ukraine, which was part of the former Soviet Union at the time] took place in April 1986, when I was taking a course in Vienna. In Europe, clouds of radioactive material formed. A large quantity of powdered milk from countries such as France, Austria, and Sweden was contaminated and should have been discarded. But part of the product went to Ireland where it was given new packaging and was exported. This milk arrived in Brazil. When I was back in Piracicaba, working with analytical techniques capable of detecting radionuclides, I purchased imported milk, because I had small children at the time. It was sold in the supermarket for a higher price. To our dismay, it contained radionuclides released in the precipitation of the nuclear cloud. We traced these products and discovered that they were contaminated by radiation from Chernobyl.
What happened as a result of that discovery?
Nothing. My PhD advisor, Epaminondas de Barros Ferraz, currently 89 years old, called a minister at the time, who I prefer not to name, and warned them of what was happening. With the news of possible contamination, principals of public infant schools and daycare centers brought bags with powdered milk here to CENA for analysis. I spent several nights analyzing the samples. All that milk was contaminated. Professor Epaminondas warned the authorities, but the product was already on the market—and its sale was not prohibited. We did not reach any conclusions about the effects of the contaminated milk on the bodies of the children who drank it.
The number of female researchers at CENA is really low. We are just five women in a group of 29 people
Did your group work on the case of the cesium-137 leak in Goiânia?
Yes, the largest radioactive accident in Brazil took place in September 1987. Recyclable material pickers inadvertently opened a source of cesium in an abandoned radiotherapy clinic. Over 200 people came into contact with the white powder from the capsule. Professor Epaminondas was summoned by the National Nuclear Energy Commission (CNEN). When he got to the location, he noticed that children were picking mangos from a tree under which the capsule had been broken. Predicting the risk to which the public was exposed, he requested that the area be cordoned off and samples of the fruit and leaves of that mango tree be collected. I analyzed this material at CENA. When we put the mango in the gamma radiation detector, an enormous photopic peak corresponding to cesium-137 was observed. It was raining heavily when the capsule was handled. Part of the material that came out of the capsule flowed over the ground, reaching the roots. The plant absorbed it, and because a mango tree like that one has enormous evapotranspiration, a lot of contaminated water circulated. The cesium would exit through the leaves, rise into the atmosphere, and settle again on the leaves and mangoes. By the time professor Epaminondas arrived at the site of the accident, many people could have consumed the fruits.
What other consequences did the accident cause?
It affected the country, since Brazil entered a list of countries contaminated by radioactivity. The cesium-137 radionuclide has a physical half-life of 30 years. In order for it to no longer be considered present in nature, five half-lives are recommended. In other words, 150 years.
So, Brazil is still on the list.
Yes, it will remain on it until the middle of the next century, since only 37 years have passed thus far. The export sector was heavily affected because purchasing countries, regardless of the incident being limited to a region of Brazil, could demand a certificate or report confirming the absence of radioactivity. The first export commodity to be questioned was coffee. Then, it was sugar. Brazilian commodities had to comply with these foreign trade requirements. We have already issued over 100,000 certificates. The revenue from these services has been enabling investments in laboratory infrastructure, equipment, and grants for undergraduate and graduate students, and is important for funding research and development activities in nuclear analytical techniques, metrology, and artificial intelligence applied to agriculture and environmental studies. We continue issuing these certificates, since Brazil remains on the list of countries affected by radiation.
Do the products pass through CENA just to certify possible radioactivity?
A certificate of origin is also required in the exportation documentation. They can be issued by some institutions based on documents, which are susceptible to fraud. For this reason, the traceability project for commodities using neutron activation was developed. It is one thing to say that a sample did not come from an area of deforestation. It is another to prove, based on some intrinsic characteristic of the material, that it really did not. We are pioneers in Brazil for this certification.
You have received several awards in the past year. How do you feel?
I have to take a deep breath, because I am not used to it. I have always worked timidly and without great visibility, publishing, presenting work abroad, doing fellowships. During the 50th anniversary graduation ceremony at ESALQ, my colleagues told me: “You were always the best student in class.” So, I agreed: “I think I deserve it.” I have more contacts and recognition abroad than in Brazil.
Why?
I have run this laboratory since 1988 when professor Epaminondas retired. When it was him, everything went well. But then it was a woman in charge. It is an arduous battle to remain in the position and show your colleagues that you are capable, regardless of gender. It has been like that to this day. The number of female researchers at CENA is really low. Five of the staff are women in a group of 29. It is something that has been around for a long time and, unfortunately, persists.
