(657b) The Importance of Reducing End Functionality in Oligosaccharide, Anhydrooligosaccharide and Cellulose Pyrolysis
AIChE Annual Meeting
2014
2014 AIChE Annual Meeting
2014 International Congress on Energy (ICE)
Thermochemical Conversion of Biomass IV: Pyrolysis
Thursday, November 20, 2014 - 8:55am to 9:20am
Introduction:
Fast pyrolysis of biomass offers the potential for generation of fuels and valuable chemicals from renewable resources in a relatively simple and scalable manner. In order to gain control over the product distribution obtained from fast pyrolysis of biomass, a deeper understanding of the sequences and mechanisms of the reactions occurring during fast pyrolysis is needed. Current fast pyrolysis methods generate product distributions which are the result of several steps of both primary and secondary reactions. In order to better understand the origins of the final pyrolysis product distribution we utilized a set of pyrolysis / tandem mass spectrometry methods (developed in house) in which the initial products can be detected and characterized.
Methods:
The pyrolysis products were formed via direct solid-solid contact on a resistively heated platinum ribbon that was heated at a rate of 1000 K×s-1 to 873 K. After they evaporated from the ribbon, they were quenched to near room temperature and ionized. Two ionization methods were used to ionize the pyrolysis products, positive mode Atmospheric Pressure Chemical Ionization (APCI) with ammonium attachment, and negative mode APCI with chloride attachment. These methods are complementary in that although many molecules can be ionized using both, there are a few exceptions, namely furan-related molecules, which were only ionized via protonation in the positive mode ammonium attachment method.
Results and Discussion:
Using a combination of a fast-heating pyroprobe and a linear quadrupole ion trap mass spectrometer, we have demonstrated that fast pyrolysis of cellulose yields only a few primary products, about half of which have not been reported in literature. These previously unreported pyrolysis products of cellulose contained several products with a ring-opened fragment attached to one or more intact glucopyranosyl rings. Additionally, pyrolysis of oligosaccharides of different chain lengths was performed (2 through 6 units). Remarkably, the major products observed upon fast pyrolysis of cellobiose were also observed for cellotriose, cellotetraose, cellopentaose and cellohexaose albeit with different relative abundances. In order to better understand the origin of the ring opened products observed from oligosaccharide pyrolysis, a range of different carbon labeled cellobioses were examined ([1-13C]glucopyranosyl-glucose, glucopyranosyl[1-13C]glucose, glucopyranosyl[3-13C]glucose, glucopyranosyl[5-13C]glucose). New mechanisms were proposed to explain the ring opened products observed from cellobiose pyrolysis based on these labeling results and were evaluated computationally using Density Functional Theory (DFT) performed at the M06-2X/6-311++G(d,p) level of theory. The newly proposed mechanisms represent a feasible low energy pathway to explain the formation of the ring opened products.
It was also observed that the relative abundance of products with an intact ring attached to a ring-opened fragment decreased with increasing chain length, which suggests that these products are associated with the ratio of terminal to total glucose units. The product distributions produced from pyrolysis of these oligosaccharides differs from cellulose in that they produce a higher relative abundance of products that are the result of ring fragmentation than cellulose. Furthermore, the largest product produced from cellotriose through cellohexaose with a relative abundance greater than 10% was cellotriosan. Inspired by this observation, fast pyrolysis of cellotriosan was also performed, which is noteworthy for its lack of a reducing end compared with standard glucooligosaccharides, which prevents tautomerization of the terminal unit into the ring opened form. Fast pyrolysis of cellotriosan resulted in a product distribution that was nearly identical to that from cellulose pyrolysis. Delineation of the fast pyrolysis reaction pathways and mechanisms for cellotriosan may prove extremely valuable for gaining control over the product distribution from fast pyrolysis of cellulose. Our results indicate that cellotriosan presents an excellent small-molecule surrogate for further experimental and theoretical studies on fast pyrolysis of cellulose.