Photosynthesis of microalgae in outdoor mass cultures and modelling its effects on biomass productivity for fuels, feeds and chemicals

Abstract

Practical applications of microalgae include uses as feedstocks for sustainable fuels, feeds, nutraceuticals, and chemicals; for the treatment of wastewater and industrial flue gas; and for use as recombinant protein expression systems. Improvements in biomass yields are required for commercially viable and sustainable algal technologies, particularly for low value commodities such as biofuels. Although several algae exhibit higher growth rates than terrestrial crops, with theoretical upper limits of 8-10% photon conversion efficiency (PCE), solar-illuminated outdoor microalgal production systems typically have a PCE far below this and even below those achieved under laboratory conditions. The most intractable limiting factor is poor light distribution through the mass culture, where photoinhibitory light at the surface and dark zones deeper in the culture reduce photosynthetic productivity. Moreover, microalgae have evolved a suite of photoacclimation and photoregulation mechanisms to cope with rapid (seconds to minutes) light fluxes in natural environments, yet these processes remain poorly understood in well-mixed photobioreactors. The aim of this thesis study was to combine an experimental and theoretical approach to: 1) investigate and understand the photosynthetic response of algae under mass culture conditions; 2) develop a mathematical model which could accurately predict light-to-biomass productivities under a broad range of design scenarios for outdoor production systems; and 3) identify key biological and design parameters to maximise biomass yields. In the first part of the study, a light-to-biomass model was developed that could be easily applied to rapidly assess biomass yields of different algal strains under a range of design scenarios. This model predicts temporal and spatial irradiance at the reactor surface and through the mass culture with inputs of local solar radiation data, local coordinates, system properties and the optical properties of the culture. Local growth rates are then coupled to local light intensities within the reactor based on a static Haldane growth model derived from empirical growth-irradiance curves. Growth rates are integrated over space and time to compute areal and volumetric productivities. Validation of the model was attempted by conducting batch harvest experiments in a laboratory scale photobioreactor matrix which simulated diurnal light cycles representative of three ‘typical’ days of solar radiation in Brisbane. The strains used were Chlamydomonas reinhardtii and its truncated light-harvesting antenna mutant, tla1, reported to possess improved optical properties. Initially, the model did not satisfactorily predict growth under batch cultivation for the three different incident light conditions, overestimating productivity under low light days and overestimating on high light days. Subsequent adjustment of the model to include photoacclimative changes in the optical properties of the cell and a re-fit of the growth parameter values provided a more accurate match to the experimental data. Model simulations were performed to compare productivities of C. reinhardtii, tla1 and a fast-growing strain (Chlorella sp. 11_H5) in open pond and flat panel reactors (FPRs) under various optimal operating concentrations and, for FPRs, various spacing distances between adjacent panels and orientation. One limitation of the simple model is that it does not incorporate physiological adaptation of algae to the variable light conditions experienced in bioreactors during mixing cycles. The second part of the study investigated the effects of photoacclimation on the productivity of C. reinhardtii under fluctuating light cycles which simulated cells mixing in dilute (low-density; LDFluc) and dense (high-density; HDFluc) outdoor mass cultures. Fluctuating light cycles were compared to non-fluctuating light regimes of the same average irradiance (LDAvg and HDAvg) to discriminate between total light dosage and light regime. It was shown that HDFluc cells that spent a large portion of time in the dark (0.5 of the duty cycle) became low-light acclimated, even though cells under non-fluctuating light of the same irradiance (HDAvg) become high light acclimated. The main phenotypes of low light acclimation were a two-fold increase in pigment concentration; reduced maximum energy-dependent NPQ (qE) attributed to low expression of the light harvesting complex stress protein LHCSR3; and higher NPQ under low light, possibly due to enhanced cyclic electron flows. It was found that these responses were maladapted to mass culture conditions, resulting in a three-fold lower biomass accumulation efficiency in comparison to cells under non-fluctuating light of the same average irradiance. This suggests that excess photon absorption during high light periods of the cycle, and an inability to safely dissipate them via NPQ may have contributed to photodamage, subsequently increasing metabolic costs for repair. In contrast, the small dark fraction (0.1 of the duty cycle) of LDFluc led to high light acclimated cells and a slightly higher biomass accumulation efficiency than LDAvg. Based on these findings, a more mechanistic model was proposed in an attempt to describe the underlying dynamic photoacclimation and regulation processes that occur in diel light cycles and under an evolving batch culture. Finally, the suitability of the two different modelling approaches presented as well as possibilities for improvement of productivity in mass culture conditions are discussed.