AI-Driven Multimodal Sensing and Predictive Harvesting to Maximize Dual Lutein and Protein Yields in Chlorella vulgaris Biorefineries
Project Description
This research establishes a non-destructive, AI-driven bioprocessing framework to overcome a major industrial bottleneck in microalgae biorefineries: severe yield losses during the co-production of high-value lutein and functional protein from Chlorella vulgaris. While co-recovering both products from a single culture offers immense commercial potential, as addressing eye-health and functional food markets valued at over $0.5 billion and $1.0 billion, respectively, current facilities suffer from low operational efficiency. This gap exists because industry relies on arbitrary fixed-day harvesting and delayed, destructive analytical testing (HPLC/LC-MS). Lutein accumulation depends on complex, time-lagged nitrogen depletion kinetics, causing intracellular lutein to peak and rapidly decline well before cell biomass reaches its maximum. To resolve this trade-off, our project integrates non-destructive, real-time optical sensors to measure visible spectra, RGB color, and optical density, with nitrate depletion trends to feed predictive, time-series machine learning models. By continually analyzing these multimodal optical signals, the AI model will accurately forecast the precise transient window where volumetric lutein titer and recoverable protein yields are synergistically maximized. By replacing slow offline sampling with a scalable, low-cost sensor-AI architecture, this work transforms algal biomanufacturing from reactive monitoring to proactive, automated harvesting control.
Start Date
March 1, 2027
Postdoc Qualifications
Ideally, the candidate will have a strong background in algal cultivation, photobioreactors, chemistry, microbiology, AI/ML, and engineering design. Ph.D. candidates or graduates with a proven publication record are strongly preferred.
Co-advisors
Zhi Zhou, zhizhou@purdue.edu, School of Civil and Construction Engineering and School of Sustainability Engineering and Environmental Engineering, https://environbiotechnology.com/
Paul Brown, pb@purdue.edu, Department of Forestry & Natural Resources, https://ag.purdue.edu/directory/pb
Bibliography
Wang, C., M. Wang, M.X. Xie, L. Qi, M. Chen, X.S. Song, Z. Zhou, X. Shi, J. Yin, Y.a. Wei, M.X. Xu, L.P. Pan, A.-J.M. Miao, and L. Yang (2026). "Climate extremes intensify global lake eutrophication by increasing the stress resistance of harmful bloom-forming algae." Nature Communications (2026).
Lopez, A.M., S. Savage, and Z. Zhou (2026). "Artificial intelligence-enabled real-time monitoring of intracellular lipid dynamics in microalgae via optical sensors and kinetic early-warning differentiation." Bioresource Technology.
Lopez, A.M., S. Savage, and Z. Zhou (2026). "Novel machine learning unlocks high lipid productivity and resolves trade-offs in algal biofuel production." Renewable Energy 256: 123901.
Hughes, S., N. Shakelly, J.W. Sutherland, and Z. Zhou (2026). "Advancing biofuel economics through piggyback integration and earned profit sharing." Renewable Energy.
Lopez, A.M., Y. Choi, and Z. Zhou (2025). "Enhanced biomimetic algal lipid enrichment for improved biofuel production driven by non-stress viral lysis." Bioresource Technology: 133128.