Closing the mass-balance of pyroprobe research regarding biomass fast pyrolysis

Abstract

Presently, the majority of the worlds energy supply is dependent on fossil fuels. As these fuels area limited resource and have significant negative side effects, the way humanity uses these is unsustainable and other sources of energy have to be found. One of these sustainable energy sources is plant-derived biomass. However, biomass cannot be efficiently used in its natural form and needs to be converted to other products through processes such as pyrolysis and gasification. Pyrolysis will be the focus of this thesis, as fast pyrolysis of Miscanthus is studied for the BRISK 2 project at Process & Energy (TU Delft). However, the current setup used for pyrolysis research does not provide a complete mass-closure of the experiments, with only 80% of the total mass accounted for in the char, tar and gaseous yields. In order to improve this closure, four different methods were developed and investigated during this thesis. All of the pyrolysis experiments were performed in a Pyroprobe 5200. The pyrolysis parameters were comparable to fast pyrolysis, with a high heating rate of 600°C/s, a low residence time of 10 s and a Final Pyrolysis Temperature between 600°C and 1000°C. The objective of the first adjustment was to improve the purge flow stability, reduce the amount of blockages in the system and increase the accuracy of the experiments. The flow meter, controlling the purge flow, was replaced with a mass-flow controller and a tube in the setup was simplified and shortened. No further blockages were observed and the total mass-closure improved on average by 7.85 wt% over the temperature range of 600-1000°C, with an increase at every temperature. The second considered element was the homogeneity of the biomass. Two biomass feedstocks were investigated, the first feedstock consisted of inhomogeneous pellets grinded down to a size less than 80μm, while the second type was homogenized biomass of a size of less than 200μm. Both biomass types performed comparably to each other and better than the previous data collected with the setup. At 600°C the inhomogeneous biomass performed better, while at 700°C the homogeneous biomass gave a higher yield. The last two experimental series both focused on gravimetrically measuring the tars collected in the condenser of the setup, which was not gravimetrically measured before. The first of these methods removed the condenser in favour of a direct capture in a quartz trap. This gave a maximum yield increase of 1.14 wt% at 700°C. However, concerns of contaminating the gas mixture with tars arose, therefore this method was not investigated at higher temperatures. In the second technique, the 2ml of isopropanol in the condenser was evaporated from a glass Petri dish, leaving the tars behind. This method was explored over a larger temperature range (600-1000°C), yet only provided a maximum increase of 1.5 wt% at 1000°C, while giving significant fluctuations between experiments. Thus proving that neither of these methods is a suitable way of gravimetrically measuring the condenser.