UC Berkeley

Production of biofuels at an industrial scale is a challenge that must be addressed for a green, sustainable future. One of the major goals to achieve in order to successfully manufacture biofuels in large-scale is in situ product recovery of the biofuels. This is an important issue for producing biofuels via biological pathways and also via chemical pathways. In fermentation, in situ product recovery is crucial because of product inhibition. Product inhibition is severe enough to terminate fermentation at around 20 g/L of product concentration, thus limiting the productivity and resulting in high separation costs as well as high operation costs due to batch processing. In chemical reactions for producing biofuels, in situ product removal is important in minimizing the formation of side products in the reaction, which also limits the productivity.

We approach this challenge by using pervaporation, a membrane-based separation method. We have designed PDMS-derived block copolymers, which are novel materials for this application, for in situ product recovery of biofuels by pervaporation. We aim to study the physical properties of these block copolymer membranes to apply them in fermenters and chemical reactors for product recovery.

Here, we first studied the structure-property relationship of PDMS-derived block copolymer membranes. The block copolymer that we have designed self-assembles into various morphologies when solvent-cast under different conditions. Comparing the morphologies of the membranes to the permeabilities of the membranes allowed us to understand the effect of morphology on permeation. The lamellar structure was the most detrimental to the permeability of the membrane; it resulted in a five-fold decrease in biofuel permeability and a three-fold decrease in biofuel selectivity. The reason for this decrease was found to be originating from the diffusion step in the permeation process.

Next, the effect of support layer resistance was studied by measuring the permeabilities of membranes of different thicknesses and by direct imaging. In order to maximize the flux of block copolymer pervaporation membranes, using an additional porous membrane layer is inevitable. However, pore penetration of the block copolymer into the porous membrane results in a dramatic increase in support layer resistance. This explains the permeability decrease with decreasing membrane thickness, and by assuming a certain pore penetration layer thickness, we were able to successfully use the resistance model to fit the permeability data. In addition, we succeeded in visually confirming pore penetration of the block copolymer via transmission electron microscopy.

The PDMS-derived block copolymer membrane was also applied in an in situ pervaporation setup attached to an ongoing acetone-butanol-ethanol (ABE) fermentation. We were able to successfully demonstrate a pervaporative-fermentation experiment where the ABE removal rate of the block copolymer membrane was matched to the rate of production in the fermenter. This resulted in a semi-continuous mode of operation for 109 hours.

Finally, crosslinked block copolymers were studied for the application of in situ pervaporation during chemical reactions to produce biofuels. These reactions are operated above the temperatures that normal polymer membranes can withstand. Thus, we studied the possibility of forming crosslinks in the non-transporting block of the block copolymer to enhance the heat tolerance to the block copolymer membranes. We were able to form crosslinks within the polyethylene domain of a polyethylene-b-polydimethylsiloxane-b-polyethylene membrane, and discovered that the crosslinks enhanced the temperature stability of the membrane without hindering permeability.