Improving green hydrogen production from Chlorella vulgaris via formic acid-mediated hydrothermal carbonisation and neural network modelling

Zita Gruber; Andras Jozsef Toth; Alfréd Menyhárd; Peter Mizsey; Mikołaj Owsianiak; Daniel Fozer

doi:10.1016/j.biortech.2022.128071

Improving green hydrogen production from Chlorella vulgaris via formic acid-mediated hydrothermal carbonisation and neural network modelling

Bioresour Technol. 2022 Dec:365:128071. doi: 10.1016/j.biortech.2022.128071. Epub 2022 Oct 17.

Authors

Zita Gruber¹, Andras Jozsef Toth¹, Alfréd Menyhárd², Peter Mizsey³, Mikołaj Owsianiak⁴, Daniel Fozer⁵

Affiliations

¹ Faculty of Chemical Technology and Biotechnology, Budapest University of Technology and Economics, Műegyetem rkp. 3., Budapest, Hungary.
² Department of Physical Chemistry and Materials Science, Laboratory of Plastics and Rubber Technology, Budapest University of Technology and Economics, H-1111 Budapest, Műegyetem rkp. 3. H. ép. I, Hungary.
³ Department of Fine Chemicals and Environmental Technology, University of Miskolc, Egyetem út, 3515 Miskolc, Hungary.
⁴ Department of Environmental and Resource Engineering, Quantitative Sustainability Assessment, Technical University of Denmark, Produktionstorvet, Building 424, DK-2800 Kgs. Lyngby, Denmark.
⁵ Department of Environmental and Resource Engineering, Quantitative Sustainability Assessment, Technical University of Denmark, Produktionstorvet, Building 424, DK-2800 Kgs. Lyngby, Denmark. Electronic address: danfo@dtu.dk.

PMID: 36257525
DOI: 10.1016/j.biortech.2022.128071

Abstract

This study investigates the formic acid-mediated hydrothermal carbonisation (HTC) of microalgae biomass to enhance green hydrogen production. The effects of combined severity factor (CSF) and feedstock-to-suspension ratio (FSR) are examined on HTC gas formation, hydrochar yield and quality, and composition of the liquid phase. The hydrothermal conversion of Chlorella vulgaris was investigated in a CSF and FSR range of -2.529 and 2.943; and 5.0 wt.% - 25.0 wt.%. Artificial neural networks (ANNs) were developed based on experimental data to model and analyse the HTC process. The results show that green hydrogen formation can be increased up to 3.04 mol kg^-1 by applying CSF 2.433 and 12.5 wt.% FSR reaction conditions. The developed ANN model (BR-2-11-9-11) describes the hydrothermal process with high testing and training performance (MSE_z = 1.71E-06 & 1.40E-06) and accuracy (R² = 0.9974 & R² = 0.9781). The enhanced H₂ yield indicates an effective alternative green hydrogen production scenario at low temperatures using high-moisture-containing biomass feedstocks.

Keywords: Combined severity factor; Gas formation; Green hydrogen; Hydrothermal carbonisation; Machine learning; Microalgae.

MeSH terms

Biomass
Carbon
Chlorella vulgaris*
Hydrogen
Neural Networks, Computer
Temperature

Substances

formic acid
Carbon
Hydrogen