The objective of this study was to elucidate primary and secondary reactions of cellulose pyrolysis, which was accomplished by comparing results from a micro-pyrolyzer coupled to a GC–MS/FID system ...and a 100g/hr bench scale fluidized bed reactor system. The residence time of vapors in the micro-pyrolyzer was only 15–20ms, which precluded significant secondary reactions. The fluidized bed reactor had a vapor residence time of 1–2s, which is similar to full-scale pyrolysis systems and is long enough for secondary reactions to occur. Products from the fluidized bed pyrolyzer reactor were analyzed using a combination of micro-GC, GC–MS/FID, LC–MS and IC techniques. Comparison between the products from the two reactor systems revealed that the oligomerization of leglucosan and decomposition of primary products such as 5-hydroxymethyl furfural, anhydro xylopyranose and 2-furaldehyde were the major secondary reactions occurring in the fluidized bed reactor. This study can be used to build more descriptive pyrolysis models that can predict yield of specific compounds.
•Indirect and direct contact heat exchange was compared for recovery of bio-oil.•Direct contact heat exchange was almost seven times faster than indirect contact.•Direct contact heat exchange using ...liquid nitrogen recovered 20% more levoglucosan.•Direct contact heat exchange using water recovered 15% more heavy ends.
This study investigates whether the rate of cooling of pyrolysis vapors affects the composition of the resulting bio-oil. Pure cellulose was pyrolyzed in a laboratory-scale fluidized bed reactor at 500 °C and the bio-oil collected in either an indirect contact heat exchange (conventional water-cooled condenser system) or a direct contact heat exchange (liquid quench) system developed in our laboratory. The liquid quench system was estimated to achieve a seven-fold increase in cooling rate compared to the water-cooled condensers. Direct contact cooling in the quench system also eliminated temperature gradients experienced by films of bio-oil running down the walls of the water-cooled condensers. The combination of these two factors helped reduce secondary decomposition of primary pyrolysis products, especially anhydrosugars such as levoglucosan. The quench system increased the yield of levoglucosan by over 20% while minimally effecting yield of other compounds.
The concept of direct contact cooling was applied to a pilot-scale, lignocellulosic biomass pyrolysis plant using water as a more practical quench media than liquid nitrogen. As with the liquid nitrogen quench, the water flashed to gas while the heavy ends of the bio-oil condensed to liquid. The quench vessel was operated above the dew point of the water to assure that it left the vessel as gas along with produced water and light ends of bio-oil, which were recovered in a condenser as an aqueous phase. In pyrolysis experiments with red oak, the quench vessel increased the yield of heavy ends by 15% compared to conventional condensers. These results encourage the design of bio-oil recovery systems that can rapidly quench products to achieve high yields and improve the quality of bio-oil.
This study evaluated the effect of thermophysical properties of heat carriers on the performance of a laboratory-scale auger reactor. Heat carriers tested included stainless steel shot, fine sand, ...coarse sand and silicon carbide. The results showed similar organic yield and composition of bio-oil among the heat carriers when pyrolyzing red oak. Significant differences in yields of reaction water, char and non-condensable gases were observed. It was also found that residual carbon contributed to as high as 20 wt% of total char yield and attrition of heat carrier as high as 7% on a mass basis were present after as little as 2 h of operation. Tradeoffs between physical performance, material cost, and product yields may exist when selecting heat carrier materials for pyrolysis of biomass in an auger reactor.
•Evaluated effect of heat carrier thermophysical properties in an auger pyrolyzer.•Organic bio-oil yields and composition were similar for all heat carriers.•Residual carbon accounted for as high as 20 wt% of total char yields.•Attrition of sand heat carrier reached as high as 7 wt%.
Although upgrading bio-oil from fast pyrolysis of biomass is an attractive pathway for biofuel production, nitrogen (N) and mineral matter carried over from the feedstock to the bio-oil represents a ...serious contaminant in the process. Reducing the N and ash content of biomass feedstocks would improve process reliability and reduce production costs of pyrolytic biofuels. This study investigated: (1) How does switchgrass harvest date influence the yield, N concentration (N), and ash concentration of biomass and fast pyrolysis products? and (2) Is there a predictive relationship between N of switchgrass biomass and N of fast pyrolysis products? Switchgrass from five harvest dates and varying N from central Iowa were pyrolyzed using a free-fall reactor. Harvestable biomass peaked in August (8.6 Mg ha
−1
), dropping significantly by November (6.7 Mg ha
−1
,
P
= 0.0027). Production of bio-oil per unit area mirrored that of harvested biomass at each harvest date; however, bio-oil yield per unit dry biomass increased from 46.6 % to 56.7 % during the season (
P
= 0.0018). Allowing switchgrass to senesce lowered biomass N dramatically, by as much as 68 % from June to November (
P
< 0.0001). Concurrently, bio-oil N declined from 0.51 % in June to 0.17 % by November (
P
< 0.0001). Significant reductions in ash concentration were also observed in biomass and char. Finally, we show for the first time that the N of switchgrass biomass is a strong predictor of the N of bio-oil, char, and non-condensable gas with
R
2
values of 0.89, 0.94, and 0.88, respectively.
