Moscow: Scientists from Russia and India have developed a method for improving the production of biodiesel from the seeds of a poisonous tropical shrub, reported Sputnik News Agency.

According to the authors of the study, a method for processing the Barbados nut (Jatropha curcas) which is developed by training a deep neural network will cut labor costs and improve the quality of the subsequent biofuel.

Biodiesel is a long-chain fatty acid methyl ester that is derived from vegetable oil and animal fats. According to experts, it has several advantages as compared to petroleum-based diesel fuel, Sputnik News Agency reported.

Moreover, using such fuel leads to a reduction in exhaust gas emissions, unburned hydrocarbons, and particulate matter.

Adding to the advantages, biodiesel contains improved biodegradability and is non-toxic. It can also be used in any diesel engine or in blends.

According to scientists, the fundamental issues in producing biodiesel are high viscosity, poor volatility, and polyunsaturation (the presence of a hydrocarbon chain with two or more carbon-carbon double bonds).

According to Sputnik News Agency, scientists are using various chemical processes to solve them. One of that is transesterification which is a safe method of changing the physical and chemical properties of esters (fats, oils) by exchanging acid residues between fats, widely used by manufacturers of specialized fats in Europe and the former USSR.

Scientists from multiple institutes including, Don State Technical University (DSTU, Rostov-on-Don, Russia), the Federal Scientific Agroengineering Center VIM (FNAC VIM, Moscow, Russia) together with colleagues from the Department of Systemics, School of Computer Science, and Department of Applied Sciences (Chemistry), University of Petroleum and Energy Studies, and Dehradun India used deep machine learning to optimize the parameters of the interesterification process of vegetable oils derived from Barbados walnut seeds and predict the characteristics of the biodiesel produced.

This will improve the quality of the resultant fuel, the accuracy of forecasting its volume and reduce the labour costs of the production process, according to the scientists, reported Sputnik News Agency.

Vadim Bolshev said, "As a result of the work done, the most important parameters of the transesterification process were classified in terms of their maximum impact on the properties of biodiesel fuel."

He further said that a positive correlation was found between temperature and process time whereas, a negative correlation between methanol content and catalyst concentration.

"The highest possible ester yield (97 percent) was achieved using a reaction temperature of 60 degrees Celcius and a concentration of potassium hydroxide catalyst of 1 percent by weight. In addition, it was determined that a concentration of methanol in a ratio of 6:1 allows you to get the maximum amount of biodiesel,” said Bolshev.

According to the researchers, comparing the quality metrics of the resulting model with other machine learning models showed the supremacy of their design.

Moreover, the accuracy of predicting the volume and viscosity of the resulting biofuel was improved by 10 percent and 8 percent, respectively, as per the scientists.

"According to scientists, the study is particularly relevant for developing countries in Africa and South America, where the Barbados nut grows as a weed. It began to be cultivated commercially in India and China after specialists pointed to this plant as one of the best candidates for future biodiesel production," reported Sputnik News Agency.

Meanwhile, the Russian-Indian research team has planned to continue working to enhance the quality of forecasting of the developed deep learning model, as well as, training the model based on a different type of raw material.

Additionally, there is a need to conduct feasibility studies, life cycle assessments, energy and exergy assessments of the application of the model in the production of biodiesel fuel.
"As a result of the work done, the most important parameters of the transesterification process were classified in terms of their maximum impact on the properties of biodiesel fuel," said Vadim Bolshev.