Date of Award

Spring 1-1-2014

Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Robert H. Davis

Second Advisor

John Pellegrino

Third Advisor

Christine Hrenya

Fourth Advisor

Mark Stoykovich

Fifth Advisor

Richard Noble


Microalgae are often cited as a sustainable source of energy due to their ability to fix CO2 as biomass through photosynthesis. At present, however, the large material and energy inputs required to grow and process the cells using conventional methods make these processes infeasible from both an economic and environmental perspective. We have proposed a novel alternative to the conventional, batch processing scenarios, in which secreted metabolites from living cells are harvested using staged, pressure-driven membrane separations. These separations include submerged microfiltration (MF) to separate product-containing growth medium from the whole algal cells, ultrafiltration (UF) to remove small colloidal contaminants such as bacteria and large macromolecules, and nanofiltration (NF) to fractionate the secreted product--in this case, small reduced-carbons (Mw < 350 g/mol)--from aqueous electrolyte. By keeping the cells alive, this approach may overcome the large nutrient inputs that make conventional methods unsustainable.

The initial research focus was on fouling and fractionation with porous, MF and UF membranes. We found that bacteria and large macromolecules are present in the algal suspensions, and can be fractionated with a nominal 5 μm MF membrane that retains the whole cells and allows smaller colloidal species to pass through. A dialyzing growth reactor provides support for this concept, and was shown to reduce bacterial contamination by ~50% compared to a control growth reactor that lacks membranes. Fouling mitigation strategies for the subsequent contaminant removal step were also addressed using selection of the membrane and operating conditions.

We also assessed the use of dense, NF membranes to fractionate the product from aqueous electrolyte. Mass transport was modeled using the solution-diffusion model and ternary separation factor plots. These plots provide a unique way to characterize the fractionation properties that are intrinsic to different membranes. After an initial, broad screening of commercial and novel membranes was conducted, temperature-variation studies allowed us to propose a novel strategy to harvest n-butanol produced by Clostridium pasteurianum while retaining its glycerol carbon source and nutrient electrolytes. Finally, surface-patterned NF membranes were investigated as a means to mitigate fouling and were found to have improved performance compared to their flat counterparts.