Design of alternative, energy-efficient processes for recovery of n-butanol from dilute (10 to 20 g/L) fermentation effluents is vital because of the high energy requirements of multi-stage distillation. We are investigating a new, energy-efficient process called "gel stripping" for recovery of n-butanol from fermentations via a column-based absorption process. The raw fermentation product is passed through a vertical column containing particles of a selectively absorbent polymeric gel material, which absorbs the alcohol component preferentially while draining excess water and other waste components from the system. The concentrated alcohol is subsequently recovered from the gel by one of several options without the need to feed large quantities of water to a multi-stage distillation unit.
One major challenge in gel stripping is to design an absorbent gel material that is both reasonably selective for the alcohol and able to swell to a high degree in dilute alcohol-water mixtures. There is usually a trade-off between selectivity and equilibrium swelling, however. In addition, the entropy of mixing of alcohol and water in the external solution is favorable enough to discourage the alcohol from entering the polymeric gel, even when the gel has a high affinity for pure alcohol. This problem is analogous to the problem faced in liquid-liquid extraction (LLE) operations in which the system must offer attractive selectivity without compromising the distribution coefficient for alcohol.
Groot et al. previously reported that Castor oil provides an attractive combination of high distribution coefficient for butanol (2.6), low distribution coefficient for water (0.00955), low toxicity, and low price ($1.2 per kg). However, Castor oil is an ineffective solvent for LLE because it is prone to forming stubborn emulsions and is subject to fouling by proteins. However, by forming a crosslinked, polymeric gel material from Castor oil, we find that its desirable separation characteristics are retained, while the resulting rubber-like material is unable to form an emulsion. A novel gel based on Castor oil was tested in the gel stripping process and permitted 95%+ removal of butanol from dilute (20 g/L) aqueous solutions, though additional work is needed to increase the absorption capacity of the material.
The second stage of gel stripping is the energy-efficient removal of the absorbed butanol from the gel. The most promising options are 1) extraction into a low-boiling liquid such as diethyl ether, 2) direct vacuum distillation from the gel, and 3) gas stripping. Current work in our group focuses on evaluating the strengths and weaknesses of these approaches and on evaluating the energy costs of each approach. Calculations suggest that the energy cost of recovering butanol by gel stripping might be reduced by more than 50 % compared to simple multi-stage distillation.
Without requiring sophisticated techniques or processing equipment, gel stripping potentially provides high separation efficiency with low energy cost. Advantages and limitations of gel stripping compared to other processes such as extractive fermentation, LLE, and pervaporation will be discussed.
 Groot, W. J.; Soedjak, H. S.; Donck, P. B.; Van der Lans, R. G. J. M.; Luyben, K. C. A.; Timmer, J. M. K. Butanol recovery from fermentations by liquid-liquid extraction and membrane solvent extraction. Bioprocess engineering, 1990, 5(5), 203-216.