USP solution can reduce losses in ethanol production

The Research Centre for Greenhouse Gas Innovation (RCGI) at the University of São Paulo (USP) is completing a highly relevant Research and Development project within the Nature Based Solutions (NBS) program, coordinated by Professor Carlos Alberto Labate from the Luiz de Queiroz College of Agriculture (Esalq-USP). This is an innovation that uses mass spectrometry and artificial intelligence to optimize the identification of fermentation contaminants, enabling the reduction of efficiency losses in ethanol production. The technology has potential for application in several industries.

The project, funded by Shell Brasil through the ANP's R&D&I clause, is based on mass spectrometry techniques to develop a new methodology for detecting contaminating bacteria in the ethanol production process from sugarcane. To this end, the researchers use Maldi-TOF (Matrix-Assisted Laser Desorption/Ionization Time-of-Flight), a device widely used in the healthcare sector for microbiological diagnostics.

According to Labate, “in hospital settings, Maldi-TOF rapidly identifies the microorganism responsible for the patient’s infection, allowing the medical team to act quickly and effectively. We are expanding this concept to industry, developing methods that allow Maldi-TOF to identify microorganisms present in industrial settings with similar speed and accuracy.”

The new methodology has the potential to significantly reduce the time needed to identify contaminants compared to current methods, allowing plants to respond faster and act more precisely in combating contamination, optimizing the consumption of antimicrobials and inputs. “Microbial contamination is one of the main causes of reduced yield and productivity, and effective control of it is essential to ensure industrial efficiency,” adds Professor Labate.

AI and automation

One of the major innovations brought about by the project is the integration of artificial intelligence (AI) into the analysis process. Maldi-TOF is currently working to identify isolated microorganisms. Researchers are working on models that will enable the identification of multiple microorganisms in a single analysis, reducing the complexity, time and cost of the technique. “This is the first step in the development of automated control systems. In the future, AI could not only detect the contaminant, but also suggest the most effective corrective measures. This would bring even greater efficiency and reduce response time at plants,” says Labate.

In addition to benefiting ethanol plants, the technology developed by RCGI also has the potential to be applied in other industrial sectors. The production of food, beer and meat, for example, also faces challenges with contamination by microorganisms. The same technology can be adapted to control these processes, ensuring greater safety in contamination control and production efficiency.

The RCGI project has the support of two important players in the energy sector, Shell Brasil and Raízen, and is scheduled to be completed in May 2025.