The main objective of this project is to investigate the continuous extraction capabilities of microbubbles in a fermentation reactor operated at 60-65C, to improve cellulosic biofuel production. One of the main issues pertaining to fermentation of sugars to alcohol is the decline in performance of fermentative organisms at high product concentrations, due to the inhibitory effects of the product on the producing organism. This is particularly true with thermophilic bacteria which grow at relatively high temperatures (50-70C). However, some of these bacteria are particularly well suited to growth on renewable, lignocellulosic feedstocks, so an effective way to continuously remove the alcohol from the fermentation broth would make the lignocellulose to ethanol process more economic. In previous studies we have already shown that at high gas flow through rates, using normal (mm) sized bubbles, ethanol can be continuously stripped from the fermentation broth, so removing its inhibitory effects. However, the gas flow rates required are far too high to be practical. When using a gas to strip material from a liquid, or to deliver material (eg oxygen) from a gas to a liquid the most important feature for determining the mass transfer rate is the ratio of bubble surface area to volume. For the same volume of gas, smaller bubbles will have a higher surface area than larger bubbles and should therefore be more effective at stripping ethanol from a solution. However, smaller bubbles could potentially be more damaging to the bacteria, so their overall benefits cannot be assumed. In this project we will develop devices to allow continuous microbubble generation and extraction in a small scale bioreactor, demonstrate its effectiveness in simulated mixtures containing ethanol but no cells and finally investigate its effectiveness for continuous ethanol extraction from fermentations containing bacteria growing at 60-65C. The effects on the bacteria will be monitored and conditions (temperature, bubble size etc) modified to achieve optimal performance.