4.1 Conclusions
The main conclusions of the present study are as follows:
1) The ash content in the feedstock plays a major role in pyrolysis. It has been shown by the relative behaviour of pinewood and of agro-biomasses that ashes, mainly alkali metals, catalyse pyrolysis reactions to decrease the bio-oil production and yield extra water, which decreased drastically the bio-oil quality:
2) The bio-oils of the agro-biomasses present higher amounts of nitrogen as a result of their initial composition;
3) High water contents (> 30 wt.%) causes phase separation of bio-oils;
4) The O/C ratio analysis for bio-oils and chars suggested that oxygen settled in the mineral matrix of initial feedstock followed to char rather than for bio-oil, which resulted in a char more oxygenated and a bio-oil less oxygenated for the agro-biomasses when comparing to those of pinewood;
5) The char from the agro-biomasses present higher ash contents and lower energy densities than that of pinewood;
As result,
6) Pinewood has shown the more consistent results with the highest yield of bio-oil (51 ± 0.5 wt.%) and the lowest yield of char (24 ± 1.5 wt.%). The bio-oil produced from pinewood met the specifications of the ASTM standard (D 7544-12) for the measured properties, and has potential to be used as a direct liquid biofuel in industrial burners equipped to handle these types of fuels. Though, its utilization as a transport liquid fuel would just to be possible with an upgraded to reduce its considerable oxygen content [78,79,90].
7) Agro-biomasses have pyrolysed into bio-oils with low energetic content (apart from olive bagasse), due to the higher water content, and with significant nitrogen amounts (3.1 wt.%
for olive bagasse). The combustion implications with so high water contents discard these agro-biomasses as a potential feedstock to pyrolyse into a direct fuel liquid unless they are upgraded [15,78]. Their significant nitrogen content would require an appropriate emission control. Furthermore, the low conversion yields related to such bio-oils in the present work (low as 31 ± 1.8 wt.%) may not justify their production with pyrolysis.
8) The non-homogeneity (biphasic) of the bio-oils from agro-biomass is a high challenge to their use as fuel, however, is an opportunity for recovering added-value by-products with particular properties [13,35]. Particular studies [63,64] have concluded promising potential for the non-aqueous phase of bio-oil.
9) The resultant char from pinewood has shown the highest heating value with 27.2 MJ/kg, a considerable value comparable with those of solid fuels ranging from lignite to anthracite [105], suggesting its potential to be used as solid fuel (e.g. in the form of briquettes or in char-oil water slurry [106]). The chars obtained from the agro-biomasses with higher ash contents and lower energy densities may be used in the preparation of active carbon when its pore structure and surface are appropriate [107].
4.2 Recommendations for future work
The feeding system has shown some limitations namely dealing with the low “fluidity” of biomass in the feeding column, which led to blockage of biomass in many cases and the sabotage of the test. Furthermore, bigger amounts of biomass sample to pyrolyse led to bigger cases of blockage.
A new configuration for the feeding system (e.g. hopper and screw feeder) should be useful to pyrolyse bigger quantities of biomass and to permit a continuously operation for the reactor that simplifies the experiments in many aspects.
A preliminary project of the screw, attending fast pyrolysis features, would be an extra aspect to optimize results. The material and geometry of the new screw would have to permit a consistent transport of the particles, an efficient and rapid heating rate with a good escape of the hot vapours. A more rapid heating rate would be achievable if the screw was heated internally through an incorporated resistance.
This would be an interesting point to keep in mind for a new design.
The pre-heating of nitrogen to considerable temperatures was an experimental fact shown in other works and has been pointed out as a possible reason to decrease the performance of the reactor, as far as it may increase thermal losses. So, it would be interesting to pre-heat nitrogen to higher temperatures and analyse its effect on the results. A new design of the main pipe with a more efficient separation of products at its end would also be an improvement to achieve better results. The char must be removed as soon as possible from the reaction to avoid secondary reactions. This would be achievable through a possible cyclone connected directly to the main pipe that would separate the solid phase from the gas/vapour phase. A thermal blanket at 400 ºC applied around the cyclone till the condensation stage would avoid the pre-condensation of vapours.
In regard to the condensation, it would be interesting to analyse the influence of other type of condensers on the results since condensation is crucial in any pyrolysis process. As said before, the condensers used in the present work showed a compromising behaviour since the trial tests and may have induced bigger water content, in relative terms, in the bio-oil sample.
After the optimization process, it would be interesting to achieve oils with better quality and higher conversion yield, and test the influence of important process parameters such as temperature, residence time, pre-heating of nitrogen and velocity of the screw. The study could be also extended to other biomasses. As the ash content in the agro-biomass had a drastic role on the quality and yields of their bio-oils, it would be interesting to compare the present results with the possible results of pre-
treated olive bagasse, wheat straw and rice husk with no ash content in order to better quantify the effects of ash in pyrolysis.
A chemical characterization of the bio-oils through a GC-MS analysis would further identify their major chemical compounds and assess their potential as chemical feedstock. A porosimetry analysis of the chars would also assess their potential as industrial adsorbents.