Scientists are using artificial intelligence to set a new world record for producing algae as a reliable, economic source for biofuel that can be used as an alternative fuel source for jet aircraft and other transportation needs. The team’s findings were published in Nature Communications

“The commercialization of algal biofuel has been hindered by the relatively low yield and high harvesting cost,” the senior author said. “The limited light penetration and poor cultivation dynamics both contributed to the low yield.”

Overcoming these challenges could enable viable algal biofuels to reduce carbon emissions, mitigate climate change, alleviate petroleum dependency and transform the bioeconomy, the author said.

The latest project utilizes a patented artificial intelligence advanced learning model to predict algae light penetration, growth and optimal density. The prediction model allows for continual harvest of synthetic algae using hydroponics to maintain the rapid growth at the optimal density to allow best light availability.

The method the team have successfully achieved in an outdoor experiment is 43.3 grams per square meter per day of biomass productivity, which would be a world record. The latest DOE target range is 25 grams per square meter per day.

Algae biofuel is regarded as one of the ultimate solutions for renewable energy, but its commercialization is hindered by growth limitations caused by mutual shading and high harvest costs.

“We overcome these challenges by advancing machine learning to inform the design of a semi-continuous algal cultivation (SAC) to sustain optimal cell growth and minimize mutual shading,” the author said.

The authors are using an aggregation-based sedimentation strategy designed to achieve low-cost biomass harvesting and economical SAC.

“The aggregation-based sedimentation is achieved by engineering a fast-growing blue-green algae strain, Synechococcus elongatus UTEX2973, to produce limonene, which increases cyanobacterial cell surface hydrophobicity and enables efficient cell aggregation and sedimentation,” the author said.

Scaling-up the SAC with an outdoor pond system achieves a biomass yield of 43.3 grams per square meter per day, bringing the minimum biomass selling price down to approximately $281 per ton, according to the journal article. In comparison, the standard low-cost feedstock for biomass in ethanol is corn, which is currently approximately $6 per bushel or $260 per ton. However, the process does not call for costly pre-treatment before fermentation. Corn must be ground and the mash must be cooked before fermentation.

https://www.nature.com/articles/s41467-021-27665-y

http://sciencemission.com/site/index.php?page=news&type=view&id=publications%2Fmachine-learning_5&filter=22