(192f) Characterization, Evaluation, and Fouling Analysis of Reverse Osmosis Membranes for Concentrating Sugars In a Lignocellulosic Biomass Hydrolysate

Gautam, A. - Presenter, South Dakota School of Mines and Technology
Hower, J. C. - Presenter, South Dakota School of Mines and Technology

In today’s era, with increasing fuel prices and greater dependence on petroleum resources, lignocellulosic biomass offers tremendous potential as an alternative renewable feedstock for fuels and chemicals.  An example of this is wood chips from Ponderosa Pine, which after pretreatment and enzymatic hydrolysis consists of a slurry containing fermentable sugars.  Unfortunately, due in large part to the limited solids loading capacity of many pretreatment and hydrolysis operations, the concentration of sugars is relatively low. Thus, concentrating the sugars prior to fermentation can lead to more efficient production and purification of renewable fuels and chemicals by allowing for higher titers in the fermentation broth.  In this study, wood chips from Ponderosa Pine, after pretreatment and  enzymatic hydrolysis,  contained approximately 20 g/L of glucose.  This hydrolysate was clarified with a 0.2 micron microfiltration and then fed to a reverse osmosis (RO) system for concentrating the sugar to over 100 g/L.  During the RO operation, severe fouling was observed on the membrane.  Thus, in order to develop improved membranes for processing the complex lignocellulosic hydrolysate, an understanding of the fouling behavior was needed.  Here we investigated the membrane fouling phenomena including operational variables (mixing versus no mixing), identification of components in the complex hydrolysate responsible for fouling, and properties of the membrane that contributed to greater observed fouling.   Different RO membranes were tested to measure the final concentration of the sugars (as well as other soluble and insoluble species) and the fouling characteristics of each of the membranes were studied using SEM, AFM, contact angle, BET and FTIR spectroscopy.  In addition, the molecular modeling software Charmm was used to theoretically evaluate the interactions between different membrane chemistries and the components present in a lignocellulosic enzymatic hydrosylate.  Molecular modeling was also used to guide design of membrane modifications that could be used to potentially improve performance.