3. Results and discussion
Blends of switchgrass and P3HB were pyrolyzed in a series of several experiments that are summarized in Table 1. In most experiments the blends contained 90% switchgrass and 10% P3HB, with the exception of experiment 8 where the proportion of P3HB was
increased to 25%. Also, in one instance, lime (CaO) was added as a 5 wt% loading to the mixture to explore its potential in assisting P3HB thermal decomposition (experiment 3). The blends were pyrolyzed at temperatures of 350 to 375 °C. Two particle size cuts (fine and coarse) of the P3HB particles were used along with three different carrier gas flow rates. The latter enabled the testing of the effect of the residence time of the P3HB particles in the reactor, with the lower flow rate providing more residence time. Residence times are estimated to be 0.12 s at 60 L/min N2 flow and 0.09 s at 75 L/min [10]. For comparison, two control experiments with switchgrass alone were also included, one at traditional fast pyrolysis conditions (480 ̊C) and one at the mild pyrolysis conditions (375 ̊C) employed for the P3HB/switchgrass blends. The distribution of the post-pyrolysis material in solid, liquid and non-condensable gas fractions is also provided in Table 1. Actual recoveries are provided along with values corrected for material retained in the system by a non-linear optimization mass balance model, as described in Section 2[22]. Compared with the fast pyrolysis of switchgrass (480 ̊C) which produced 62.3% liquid, the yield of liquid (bio-oil) in all of the mild pyrolysis experiments was found to be lower (Table 1). For the mild, lime-free pyrolysis of switchgrass/P3HB blends, greater liquid yields (40 to 48 wt%) were achieved compared with the mild pyrolysis control experiment containing switchgrass alone (36%). Conversely, the yield of solid products (bio-char) from the switchgrass/P3HB blends (9.6 to 24.4%) was increased compared to the fast pyrolysis of switchgrass alone (9.8%) but decreased relative to the mild pyrolysis of switchgrass alone (∼30%). These results suggest that both the use of lower temperatures, which are necessary to prevent further degradation of crotonic acid, and the presence of P3HB and/or crotonic acid in the reaction mixture influence the product distribution. Use of pyrolysis temperatures lower than 450 °C for biomass pyrolysis has previously been shown to result in decreased bio-oil and increased bio-char production [12]. The total yield of gaseous products is calculated to be higher in the experiments with the switchgrass/P3HB blends, compared to the mild and fast pyrolysis control experiments. This could be due to the increased acidity of the environment resulting in increased decarboxylation and decarbonylation type reactions of the biomass derived materials. The composition of the gases produced with the P3HB/switchgrass blends showed an increase in the ratio of CO2/CO compared with the control fast pyrolysis of switchgrass (Table 2) but similar to the mild pyrolysis control. Decomposition of crotonic acid into CO2 and propylene[23] could also contribute to the increased gas yield, but propylene was not observed. However, its concentration may be too small to detect by the GC method employed since it is diluted with the N2 fluidizing gas. Crotonic acid decomposition could however not produce enough gas to explain the difference
between the gas yield in the switchgrass fast pyrolysis control experiment and the experiments with the P3HB-switchgrass blends suggesting that indeed the increased acidity of the environment does help to convert some condensable range products to permanent gases.