Best-performing Bacillus strains for microbiologically induced CaCO3 precipitation: Screening of relative influence of operational and environmental factors

• Published in: Journal of biotechnology Vol. 350; pp. 31 - 41
• Main Authors: Šovljanski, Olja, Pezo, Lato, Grahovac, Jovana, Tomić, Ana, Ranitović, Aleksandra, Cvetković, Dragoljub, Markov, Siniša
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 20.05.2022
• Subjects: algorithms; Alkalophilic bacteria; Bacillus; Bacteria; Biomineralization; Biomineralization capacity; bioremediation; biotechnology; calcite; Calcium Carbonate; Chemical Precipitation; Experimental design; MICP; neural networks; prediction; Soil; soil bacteria; Sporosarcina; Sporosarcina pasteurii; urea; Alkalophilic bacteria; Biomineralization capacity; Sporosarcina; Experimental design; MICP; Bacillus
• ISSN: 0168-1656; 1873-4863; 1873-4863
• DOI: 10.1016/j.jbiotec.2022.04.002

The microbiologically induced calcite precipitation (MICP) can be an emerging approach that could tap onto soil bacterial diversity and use as a bioremediation technique. Based on the concept that bacteria with biomineralization capacity could be effective CaCO3 inductance agents, this study aimed to evaluate the simultaneous influence of 11 operational and environmental factors on the MICP process, for the first time. Therefore, Bacillus muralis, B. lentus, B. simplex, B. firmus, and B. licheniformis, isolated from alkaline soils, were used in the selection of the best performing bacterium compared with a well-known MICP bioagent Sporosarcina pasteurii DSM 33. Plackett-Burman's experimental design was labouring to screen all independent variables for their significances on five outputs (pH value, number of viable cells and spores, amount of urea and CaCO3 precipitate). According to experimentally obtained data, an artificial neural network model based on the Broyden-Fletcher-Goldfarb-Shanno algorithm showed good prediction capabilities, while differences in the relative influences were observed at the bacterial strain level. B. licheniformis turn out to be the most potent bioagent, with a maximum amount of CaCO3 precipitate of 3.14 g/100 mL in the optimal conditions. •Screening of best-performing bacteria with biomineralization capacity was showed.•Plackett-Burman design involves ten operation parameters and one dummy factor.•Five variables were elaborated in the experimental design.•Developed ANN model showed good prediction capabilities.•Bacillus licheniformiswas the best bacterium using multi-objective optimization.