Results and Discussion Low Temperature Matrix Isolation (LTMI) Solid Phase Spin Trapping 500 600 700 800 900 2.0080 2.0085 2.0090 2.0095 2.0100 p-CMA CA g value Temperature ( o C) Radicals from the Gas-Phase Pyrolysis of Lignin Model Compounds: p-Coumaryl and Cinnamyl Alcohols Lavrent Khachatryan 1 , Rubik Asatryan 2 , Mengxia Xu 1 , Barry Dellinger 1 1 Louisiana State University, Baton Rouge, LA 70803 2 Department of Chemical and Biological Engineering, University at Buffalo, SUNY, Buffalo, NY 14226 Abstract A number of pyrolysis studies on lignocellulosic materials show that the cinnamyl alcohols (CA) are primary products along with the main monolignols – p-coumaryl (p-CMA), coniferyl and sinapyl alcohols. Despite the content of the CA end-groups in lignin is relatively low, they are well involved in the pyrolysis reactions and can have large contribution in the overall process. Therefore, the pyrolysis studies of the model compounds are necessary to clarify the detailed mechanism of the lignin pyrolysis which, as research evidences, occurs through the radical mechanisms. Here we report the preliminary results on detection and identification of the intermediate radicals formed during the gas phase pyrolysis of p-CMA and CA over the temperature range of 500-900 o C. The intermediate radicals were trapped from the gas phase using low temperature matrix isolation (LTMI) and solid phase PBN (N-t-butyl-α-phenylnitrone)-spin trapping techniques in conjunction with the electron paramagnetic resonance (EPR) spectroscopy. Both experimental methods confirmed the dominance of the oxygen-centered radicals formed during the gas-phase pyrolysis of p-CMA and CA. These experimental findings are consistent with the theoretical DFT (density function theory) - calculations using Gaussian-03 quantum chemistry package. Introduction Lignin Radical Trapping Techniques Electron Paramagnetic Resonance (EPR) A technique to measure unpaired electrons Particularly useful for organic radicals Methods Experimental Theoretical Density functional theory (DFT) has been employed to calculate bond dissociation energies (BDE) and selected potential energy profiles for the p-CMA and CA to form the intermediate radicals using the Gaussian-03 quantum chemistry package. Electronic energies and zero point vibration energies were computed based on ground state geometries optimized at B3LYP/6-31+G(2d,p) level. Theoretical Calculation Conclusion References 1. Alder E., 1977. Lignin chemistry – Past, present and future. Wood Sci. Technol., 11, 169-218. 2. Dellinger B., Lomnicki S., Khachatryan L., et al., 2007. Proc. Combust. Inst., 31, 521-528. 3. Arroyo C.M., Kramer J.H., Leiboff R.H., et al., 1987. Spin trapping of oxygen and carbon-centered free radicals in ischemic canine myocardium. Free Radical Bio. Med., 3, 313-316. 4. Pryor W.A., Terauchi K.I., and Davis, W.H., 1976. Electron spin resonance (ESR) study of cigarette smoke by use of spin trapping techniques. Environ. Health Persp., 16, 161-175. Acknowledgements This work was funded by National Science Foundation grant (#1330311). Dr. Lavrent Khachatryan thanks Superfund Research Program (#2P42ES013648-03) and Dr. Mengxia Xu thanks RJ Reynolds’ Tobacco Company for partial support. Ruckenstein fund (UB) is also acknowledged by Dr. Rubik Asatryan for continuous support. Fig. 1. Typical structure of softwood lignin 1 p-CMA CA Fig. 2. LTMI (I) and solid phase spin trapping (II) methods Fig. 4. LTMI-EPR Setup Fig. 3. EPR Spectroscopy Fig. 5. Temperature dependence of radicals from gas-phase pyrolysis of p-CMA and CA Fig. 6. Annealing effect on EPR spectra of radicals from gas-phase pyrolysis of p-CMA (I) and CA (II) at 500 o C Fig. 7. EPR spectra of PBN spin adducts from gas-phase pyrolysis of CA at 800 o C   G 10 G   G Literature reported hyperfine splitting constants for adducts of PBN 3, 4 : RO: α N = 13.6 G, α H = 1.56 G; R: α N = 15.2 G, α H = 3.85 G RO+ ArCO 2 ∙: α N = 13.9 G, α H = 2.00 G Fig. 8. Simple decomposition pathways for pyrolysis of p-CMA and CA 900 o C 20 G 800 o C p-CMA 800 o C 700 o C 600 o C CA 500 o C 500 o C 60 s 20 G p-CMA 40 s 10 s Initial CA 180 s 20 G 60 s 30 s Initial 0 20 40 60 2.0080 2.0085 2.0090 g-value g value Annealing Time (seconds) p-CMA 0.05 0.06 0.07 0.08 signal intensity Signal Intensity (a.u.) 0 60 120 180 2.006 2.007 2.008 2.009 2.010 g-value g value Annealing Time (seconds) CA 0 2 4 6 8 signal intensity Signal Intensity (a.u.) Note: Fig. 5 reveals the following trends: I) p-CMA showed similar anisotropic EPR signal throughout the whole temperature range (500-900 o C ), which resembled that from CA at the high reaction temperature (800 o C ). II) EPR signal intensity from p-CMA pyrolysis increased with increasing reaction temperature while CA showed the opposite trend. III) EPR signals from pyrolysis of both p-CMA and CA showed high g- values (2.0085-2.0095) that increased largely with temperature. (I) (II) (III) (I) (II) (I) (II) Note: Fig. 6 reveals the following trends: I) With the increased annealing time, the EPR signals from p-CMA pyrolysis changed slightly; g-value and signal intensity decreased from 2.0089 to 2.0082, and from 0.066 to 0.060 (a.u.) , respectively . II) With the increased annealing time, the EPR signals from CA pyrolysis changed significantly; g-value and signal intensity decreased from 2.0086 to 2.0061, and from 6.1 to 0.14 (a.u.) , respectively . Note: The hyperfine splitting constants (α N = 13.8 G and α H = 2.18 G ) obtained for the adducts of PBN in this study were closer to the reported values for oxygen centered radicals than for carbon centered radicals. 3370 3375 3380 1 st Derivative Curve Magnetic Field Strength (G) Aborption Curve g value 2 H p-p g = hv/H 1) A similar, anisotropic EPR signal was detected from the gas- phase pyrolysis of p-CMA and CA in the entire temperature region from 500 to 900 o C. 2) Cryogenic experiments (LTMI EPR) revealed that EPR signals from gas-phase pyrolysis of both p-CMA and CA showed high g- values (>2.008) indicating the dominance of oxygen centered radicals. 3) The predominance of oxygen centered radicals in the reaction has also been demonstrated by solid phase spinning trapping experiments using PBN. 4) Theoretical calculations confirmed the dominance of oxygen centered radicals at higher temperatures. 5) A low-energy pathway explains the high yield of one of the major products, cinnamyl aldehyde detected experimentally during the pyrolysis of CA. Note: According to the calculations, the lowest energy decomposition pathways for CA are C9-H and C9-OH bonds scissions, resulting formation of carbon centered radicals. The high energy demanding process C9O-H bond scission may occur at high pyrolysis temperature resulting formation of oxygen centered radicals. 3 rd largest source of natural polymer after cellulose and hemicellulose Most abundant natural poly- meric aromatics Randomly linked phenyl- propane units, varies with plant species and growth conditions Three monolignols: p-CMA, coniferyl and sinapyl alcohols Experiments were set up according to Fig. 2. Before each experiment, the system was purged by pure nitrogen for 45 min to remove the oxygen. The sample was first heated in the preheater and then pyrolyzed at 500-900 o C in the reactor. The radicals trapped by liquid nitrogen or as PBN spin adducts were detected by EPR spectroscopy. 500 600 700 800 900 0 2 4 6 8 p-CMA CA Signal Intensity (a.u.) Temperature ( o C) Note: Theoretical calculations revealed a novel low-energy pathway explaining the major experimental fact – the dominance of O-centered radical even at lower temperatures. This radical (with high g value) may be a precursor for formation of one of the major products, cinnamyl aldehyde detected experimentally. Fig. 9. Low energy H-migration channel to form an terminal oxygen centered radical