Current energy production is done in a way that cannot be sustained ultimately. Biomass is ideal as a sustainable alternative because it can be used in the existing energy infrastructure. For coal power plants verge grass is one of the products being considered for co-firing. The moisture content and its bioactivity currently prevent direct co-firing on a large scale. A very new and promising pre-treatment technique to counter this is wet torrefaction. Wet torrefaction does not evaporate the moisture, it allows a precise temperature and thus process control, high heat transfer and a combined washing out of unwanted salts. To predict the decomposition of verge grass during wet torrefaction a reaction model was set up. Hemicellulose and cellulose kinetic data was used from previous work on corn cob and pure cellulose. The model was tested with data from research on sugar maple wood meal decomposition. Wet torrefaction experiments showed that verge grass decomposes very rapidly in contrary to the model. Experiments done on xylan showed a very slow decomposition. Bagasse decomposition was very hard to monitor because very little decomposition products were detected. A possible explanation could be that grass is very young and thus has a low degree of polymerization. Xylan and bagasse on the other hand are already treated and therefore the toughest parts of the original material are being torrefied. Overall the biomass species had a very high solid mass loss. Verge grass and bagasse retained around 30% of the original mass and no solid residue was found with xylan. The solid fraction and liquid fraction that could be accounted for is lower than half of the original mass. Although additional tests are needed to accurately predict the decomposition of verge grass a design was made for a pre-treatment plant. This facility was regarded as a stand-alone facility and consists of several parallel CSTR’s and a heat exchanger. The needed calculations are performed to compare wet torrefaction with alternatives.