Thermochemistry of Compounds Formed during Fast Pyrolysis of

Oct 21, 2008 - CEA, Cadarache Research and DeVelopment Center, France. ReceiVed June 2, 2008. ReVised Manuscript ReceiVed August 25, 2008...
0 downloads 0 Views 216KB Size
Download PDF
Energy & Fuels 2008, 22, 4265–4273

4265

Thermochemistry of Compounds Formed during Fast Pyrolysis of Lignocellulosic Biomass Laurent Catoire,* Mohammed Yahyaoui, Antoine Osmont, and Iskender Go¨kalp ICARE-CNRS and UniVersity of Orleans 1C, AVenue de la Recherche Scientifique, 45071 Orleans cedex 2, France

Meryl Brothier, He´le`ne Lorcet, and David Gue´nadou CEA, Cadarache Research and DeVelopment Center, France ReceiVed June 2, 2008. ReVised Manuscript ReceiVed August 25, 2008

This paper deals with the gas-phase thermodynamic properties of organic compounds formed during biomass thermochemical energetic conversion. The standard enthalpies of formation at 298.15 K are determined by means of quantum chemistry calculations along with a protocol developed for general organic compounds. The resultant data, currently not available in the literature for most of these compounds, are critical to the modeling of pyrolysis, gasification, and combustion chemistry of biomass.

Introduction Nowadays, about 97% of the fuels consumed by cars, trucks, and airplanes are petroleum-based. Therefore, the need for new fuels will be more and more important in the forthcoming years. Several alternatives are possible. Among them, fuels from biomass are a credible alternative. First generation biofuels are bioethanol1,2 and biodiesel.3,4 We published studies dealing with the thermochemistry of biodiesel components obtained from vegetable oils.5-7 However, it is highly probable that first generation biofuels are not able to replace fossil fuels. Furthermore, the production of ethanol causes, or is expected to cause, damages to the environment such as deforestation and intensive use of pesticides, among others. Consequently, other fuels are needed. Second generation biofuels are obtained from other biomass than sugary or oleaginous plants, that is, from plants that are not cultivated (wood residues), from agricultural residues (sugar cane bagasse for instance), or from nonalimentary crops.8 Constituents of this biomass are cellulose, hemicellulose, pectic compounds, surface compounds, and lignin. Major constituents are cellulose, hemicellulose, and lignin, and the amounts of cellulose, hemicellulose, and lignin depend on the biomass. This biomass can be heated in the absence of oxygen for the * To whom correspondence should be addressed. E-mail: catoire@ cnrs-orleans.fr. (1) Arifeen, N.; Wang, R.; Kookos, I. K.; Webb, C.; Koutinas, A. A. Biotechnol. Prog. 2007, 23, 1394–1403. (2) Ogura, T.; Sakai, Y.; Miyoshi, A.; Koshi, M.; Dagaut, P. Energy Fuels 2007, 21, 3233–3239. (3) Lotero, E.; Liu, Y.; Lopez, D. E.; Suwannakarn, K.; Bruce, D. A.; Goodwin, J. G., Jr. Ind. Eng. Chem. Res. 2005, 44, 5353–5363. (4) Dagaut, P.; Gail, S. J. Phys. Chem. A. 2007, 111, 3992–4000. (5) Osmont, A.; Catoire, L.; Go¨kalp, I. Int. J. Chem. Kinet. 2007, 39, 481–491. (6) Osmont, A.; Catoire, L.; Gokalp, I.; Swihart, M. T. Energy Fuels 2007, 21, 2027–2032. (7) Osmont, A.; Yahyaoui, M.; Catoire,L.; Go¨kalp, I.; Swihart, M. T. Combust. Flame in press. (8) Pu, Y.; Zhang, D.; Singh, P. M.; Ragauskas, A. J. Biofuels Bioprod. Bioref. 2008, 2, 58–73.

production of liquid fuels or various chemicals.9,10 As biomass is pyrolyzed, chemical bonds in cellulose, hemicellulose, and lignin are broken to form solids, liquids, and gases. Some of these gases can be condensed. Therefore, quite complex liquid mixtures are obtained, as shown in Tables 1-3 for a bio-oil provided by BTG and obtained by flash pyrolysis of pine. Furthermore, the composition depends not only on the biomass pyrolyzed but also on the thermal treatment used (temperature, temperature rise rate, pressure, and residence time). There is, therefore, no typical composition for the liquids obtained. For flash pyrolysis of biomass, these raw pyrolysis liquids are named as bio-oil or biooil. Bio-oil can be stored and transported. It can be used in fuel boilers, some engines, and turbines, or it can be converted to syngas and then to transportation fuels.11 Most of the constituents of these bio-oils are identified but are generally not quantified.12-14 Headspace analysis of bio-oil is reported in ref 15. Table 1 reports names and structures of some of the constituents of bio-oil for which thermochemical data are not available. Table 2 reports names and structures of some of the constituents of bio-oil for which thermochemical data are available. This table also includes species belonging to the same chemical classes than the constituents of bio-oil. Although thermochemical data (standard enthalpy of formation in the gas(9) Bridgwater A. V., An Introduction to Fast Pyrolysis of Biomass for Fuels and Chemicals, Chapter 1. In Fast Pyrolysis of Biomass: A Handbook; CPL Press: 1999. (10) Diebold J. P., Bridgwater A. V., Overview of Fast Pyrolysis of Biomass for the Production of Liquid Fuels, Chapter 2. In Fast Pyrolysis of Biomass: A Handbook; CPL Press: 1999. (11) Gue´nadou D., Lorcet H., Brothier M., Gramondi P., Michon U., Onofri F., Vardelle M., Mariaux G., Gasification of Biomass by Thermal Plasma 7th High Temperature Air Combustion and Gasification International Symposium, Phuket, Thailand, 2008. (12) Meier D., Oasmaa A., Peacocke G. V. C., Properties of Fast Pyrolysis Liquids: Status of Test Methods, Chapter 7. In Fast Pyrolysis of Biomass: A Handbook; CPL Press: 1999. (13) Diebold J. P., A Review of the Toxicity of Biomass Pyrolysis Liquids Formed at Low Temperatures, Chapter 12. In Fast Pyrolysis of Biomass: A Handbook; CPL Press: 1999. (14) Wang, X.; Chen, H.; Luo, K.; Shao, J.; Yang, H. Energy Fuels 2008, 22, 67–74.

10.1021/ef800418z CCC: $40.75  2008 American Chemical Society Published on Web 10/21/2008

4266 Energy & Fuels, Vol. 22, No. 6, 2008 Table 1. Names and Structures of Selected Components in Bio-oila

Catoire et al.

Thermochemistry of Compounds from Pyrolyzed Biomass

Energy & Fuels, Vol. 22, No. 6, 2008 4267 Table 1. Continued

a

Units are kcal mol-1.

phase at 298.15 K, gas-phase standard heat capacities in a wide temperature range, gas-phase standard entropies in a wide temperature range) are available for some of these compounds (water, formic acid, acetic acid, 1,2-ethanediol, etc.), they are incomplete or unknown for most of them. The aim of this paper is to propose gas-phase thermochemical properties to allow some thermochemical calculations such

as heat of combustion, flame temperature, Chapman-Jouguet detonation velocity, amount of pollutants formed at thermodynamic equilibrium, etc., and the building of detailed chemical kinetic models to be implemented in codes devoted to the simulation of the high-temperature reactivity of biooil during gasification, partial oxidation, and other thermal processes. It is beyond the scope of this paper to propose a

4268 Energy & Fuels, Vol. 22, No. 6, 2008

Catoire et al.

Table 2. Validation of the Computational Method Used in This Study with Compounds Belonging to Bio-oil or to the Same Chemical Families than the Constituents of Bio-oil†

Thermochemistry of Compounds from Pyrolyzed Biomass

Energy & Fuels, Vol. 22, No. 6, 2008 4269 Table 2. Continued

† Units are kcal mol-1. References for this table are reported in the Supporting Information. a Compound considered in the determination/calibration test set of the method used in this study. b Bio-oil constituent. b Bio-oil constituent.

model fuel for bio-oil. This task is somehow premature because the composition of bio-oil is not sufficiently known. Bio-oil Composition The composition of biooil is complex. Species identification is reported in Table 3 for some of the more important

peaks. These compounds belong to the same chemical classes as the species reported in the literature. Many compounds reported in Tables 1 and 2 as bio-oil components have been identified. This work is devoted to bio-oil in general and not only to the bio-oil obtained from pine which is why the compounds of interest in this study and reported in Tables 1

4270 Energy & Fuels, Vol. 22, No. 6, 2008

Catoire et al.

Table 3. Partial Listing of Species Identified in GC-MS Analyses Performed for This Studya

a Chromatographic conditions: column: DB-17 ms, 30 m × 0.32 mm ID, 0.25 µm film; oven: 40°C (30 min) to 250°C at 5°C/min (5 min hold); carrier: helium, 1 mL/min; detector: MS, m/z ) 50-350; injector: 1 µL, splitless, 250°C.

Thermochemistry of Compounds from Pyrolyzed Biomass

and 2 are partially from the literature12-15 and partially from the analyses presented in Table 3.

Energy & Fuels, Vol. 22, No. 6, 2008 4271 0 ∆f H 298.15K(g) ) 627.51(Ej + ZPEj + thermal corrections +

∑Ric*i ) (1) i

Computational Method The protocol proposed here is useful for molecules for which experimental data or previous higher level calculations do not exist. Three fundamental thermochemical properties are used in combustion: ∆fH°298.15K, Cp°T, and S°T, and generally these data are needed on a wide temperature range (from ambient up to 5000 K). Enthalpies of formation of species, in particular molecules, can be estimated quite conveniently using group additivity methods.16,17 However, these estimates are not always feasible due to the lack of some group values. It is the case for 2-acetylfuran, levoglucosan, 2(3H)-furanone, homovanillin, coniferyl alcohol, and 2(5H)-furanone in Table 1 and for 2,3-dihydrobenzofuran, benzofuran, and dibenzofuran in Table 2. This hampers the estimation of the thermodynamic data of radicals formed from parent molecules and can hamper the development of detailed chemical kinetic models. Furthermore, even if Cp and S can be estimated for only some compounds up to 1500 K using group additivity methods, there is a need for thermodynamic data up to 5000 K to allow thermodynamic analyses of bio-oil gasification in thermal plasma.11 Theoretical methods have been shown to be reliable tools to estimate the thermochemistry of many compounds.18 However, the most accurate of these methods, namely the G3 or G2 methods, are computationally too expensive for large molecules.19 These methods are also computationally too expensive when calculations for innumerable species are required. It is the case here for all the constituents of bio-oil and related radicals formed during bio-oil thermal treatment to obtain syngas, for instance. For this reason, a protocol with a small basis set is required. A comparative study of several existing approaches, including the semiempirical PM3 method, indicates that the B3LYP/6-31G(d,p) protocol appears to be a good compromise between numerical accuracy and expense. In previous studies,5-7 we presented a B3LYP/6-31G(d,p) method using the atomization approach to allow primarily the estimation of thermodynamic data for large molecules and large radicals for which highly accurate G2 or G3 methods are inappropriate. However, this method was also derived and validated with small species. The model employed in this paper includes zero-point energies and several atomic corrections. The method employed in the present study is derived based on compounds having wellcalibrated enthalpies of formation with uncertainty less than 1 kcal mol-1 for most of them. The compounds are general organic compounds from all the chemical families, including polyfunctional compounds. The atomic corrections ci* are determined by leastsquares fitting of the selected experimental gas-phase standard enthalpies of formation at 298.15 K. Calibration and validation sets for general organic compounds are given in the Supporting Information of ref 6. The C/H/O subset is considered here. The gas-phase standard enthalpy of formation of molecule j at 298.15 K can be determined from the following equation. (15) Meier, D., New Methods for Chemical and Physical Characterization and Round Robin Testing, Chapter 8. In Fast Pyrolysis of Biomass: A Handbook; CPL Press: 1999. (16) Domalski, E. S.; Hearing, E. D. J. Phys. Chem. Ref. Data 1993, 22, 805–1173. (17) Stein, S. E., Brown, R. L., Structures and Properties Group Additivity Model. Increments used in this model are from: Benson, S. W., Thermochemical Kinetics: Methods for the Estimation of Thermochemical Data and Rate Parameters, 2nd Ed.; John Wiley: New York, 1976. Additonal increments are from: Stein, S. E. NIST Standard Reference Database 25 In NIST Structures and Properties Database and Estimation Program; NIST: Gaithersburg, MD, 1991. The software used by this model was developed by: Stein S. E., Brown R. L., and Mirokhin Y. A. (18) Curtis, G. A.; Petersson, D. K.; Malick, W. G.; Wilson, J. W.; Ochterski, J. A.; Montgomery, M. J.; Frisch, J. Chem. Phys. 1998, 109, 10570–10579. (19) Curtiss, L. A.; Redfern, P. C.; Frurip, D. J., Theoretical Methods for Computing Enthalpies of Formation of Gaseous Compounds. In ReViews in Computational Chemistry; Lipkowitz, K. B., Boyd, D. B., Eds.; WileyVCH: 2000; Vol. 15, Chapter 3, pp 147-211.

where Ri is the number of atom i in molecule j, and is the atomic correction for atom i. Ej and ZPEj denote, respectively, the electronic energy and zero-point energy, calculated using the Gaussian 98W and Gaussian 03 softwares. The units are hartree molecule-1 for Ej, ZPEj, and thermal corrections and are hartree atom-1 for c*i , 0 is in kcal mol-1. Table 4 gives the three whereas ∆f H 298.15K(g) atomic corrections used in this study. Calculations were carried out using the Gaussian 9820 and Gaussian 0321 computational chemistry programs. Three atomic corrections are derived for the molecules here: one for H-atoms, one for C-atoms, and one for O-atoms. This method has many advantages: (a) we do not distinguish between O-atoms. The correction to be applied to an O-atom in a carbonyl group is the same than the one to be applied to an O-atom in a hydroxyl group or in an ether bond. The same holds for C-atoms; the correction to be applied to a C involved in a single bond is the same as the correction to be applied to a C involved in double bonds. (b) It does not require group increments. (c) It can be applied in reasonable times to large molecules. (d) It is able to provide an accuracy better than 3 kcal/mol for compounds under consideration in this study (see Model Validation section). In this paper, we report data for the lowest energy conformer. Unscaled frequencies and moments of inertia are reported in the Supporting Information. A number of the molecules have one or several internal rotors, and Cp and S need to be corrected for these internal rotations. c*i

Model Validation The method developed and implemented in the present study is validated for a number of compounds, including almost all the chemical families and polyfunctional compounds. For the present study, compounds of interest are reported in Table 1. These compounds are monosubstituted furans (cyclic ethers); mono-, di-, and trisubstituted benzene; small carboxylic acids; hydroxyacetaldehyde; alcohols; ketones; phenols; guaiacols; syringols; and sugars (levoglucosan). Therefore, the method proposed here is validated with bio-oil components or with (20) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A. Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; HeadGordon, M.; Replogle, E. S., Pople, J. A., Gaussian 98, ReVision A.6; Gaussian, Inc.: Pittsburgh PA, 1998. (21) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.; Vreven, T., Jr.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Laham, M. A. Al; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C., Pople, J. A. Gaussian 03, ReVision D.01; Gaussian, Inc.: Wallingford CT, 2004.

4272 Energy & Fuels, Vol. 22, No. 6, 2008

Catoire et al.

compounds belonging to the same chemical classes as bio-oil components. The corresponding validation is given in Table 2 and is discussed in the Results and Discussion section. The accuracy provided by the protocol presented in this paper is estimated to be about 3 kcal/mol. This is based on results obtained in the frame of the calibration and validation procedures of the protocol. Calibration and validation sets comprise more than 600 species: 311 for the calibration set and 342 for the validation set for C, H, O, Cl, S, N, F, and Br atoms. For the calibration set, the mean absolute deviation is 2 kcal/mol and the mean deviation is about -0.3 kcal/mol. For the validation set, the mean absolute deviation (MAD) is about 2.5 kcal/mol and the mean deviation is -0.15 kcal/mol. For C/H/O and C/H species, that is, the species considered in this study, the MAD and the mean deviation are known to be better. This is logical because the experimental accuracy for CHO species is generally better than the accuracy reported for Cl- or S- containing species. Both calibration and validation sets comprise numerous molecules containing one or more OH groups, CdO groups, methoxy groups, or carbon-carbon double bonds. The occurrence of systematic biases for specific classes of compounds (nitrates, highly strained polycyclic alkanes) are possible and have been examined elsewhere. For general C/H/O compounds, the model does not seem consistently biased for a particular group. Results and Discussion Calculated standard enthalpies of formation at 298.15 K for gas-phase compounds are reported in Table 1 for compounds for which no experimental data are reported. Calculated data are compared with experimental data in Table 2. For compounds given in Table 2, the calculated data agree with the experimental data. The corresponding MAD is 1.1 kcal mol-1 and the mean deviation is -0.2 kcal mol-1. Maximum absolute deviations of about 3 kcal mol-1 are observed for 2-furaldehyde, 4-methyl guaiacol, and benzophenone. For benzophenone, the determination is unique, therefore complementary experimental work is needed to assess the existing experimental value. This discrepancy is quite surprising for 2-furaldehyde because it is a monosubstituted derivative of furan for which experimental and calculated data are equal (see Table 2). Furthermore, experiments and calculations compare well for 2-furfuryl alcohol, a monosubstituted furan, and 2,5-dimethylfuran, a disubstituted furan. Reference17 provides an estimate of -34.1 kcal mol-1 using a group additivity method, but the same method provides for 2,5-dimethylfuran an estimate of -25.0 kcal mol-1, which is about 6 kcal mol-1 too high compared to the experimental data. Another uncertainty lies in the value of the standard enthalpy of vaporization at 298.15 K of 2-furaldehyde. In fact, the gas-phase experimental enthalpy of formation is obtained indirectly according to: ◦ ◦ ◦ ∆fH298K(g) ) ∆fH298K(l) + ∆vapH298K

(2)

with a ∆vapH°298K value of 12.1 ( 0.1 kcal mol-1.22 If one considers the gas-phase enthalpy of formation at 298.15 K (-36.9) and the liquid-phase enthalpy of formation at 298.15 K (-47.4) both estimated according to Domalski and Hearing16 for 2-furaldehyde, the resulting standard enthalpy of vaporization at 298.15 K is equal to 10.5 kcal mol-1, that is, 1.6 kcal mol-1 lower than 12.1. Therefore, the corresponding experimental gasphase standard enthalpy of formation at 298.15 K may also be 1.6 kcal mol-1 lower than reported and equal to -37.6 ( 1.0 (22) NIST Chemisty WebBook, http://webbook.nist.gov/chemistry/.

kcal mol-1. Things are, in fact, more complicated. It appears that there is no data reported for the standard enthalpy of vaporization of 2-furaldehyde at 298.15 K. Three data are reported in the literature:23 11.5 kcal mol-1 at 344 K, 10.7 kcal mol-1 at 372 K, and 11.4 kcal mol-1 at 380 K. Unfortunately, these data are not consistent since the standard enthalpy of vaporization increases when the temperature decreases according to the following equation; T

∆vapH(298.15K) ) ∆vapH(T) +

∫

(Cp(l ) - Cp(g)) dT (3)

298.15

because Cp(l ) > Cp(g). Therefore, the value at 372 K cannot be lower than the value at 380 K, and the value at 344 K cannot be about equal to the value at 380 K since Cp(l ) - Cp(g) ≈ 15.5 cal mol-1 K-1 at 298.15 K. Therefore, it is difficult to conclude unambiguously concerning the actual value of the gas-phase standard enthalpy of formation at 298.15 K for 2-furaldehyde. Nevertheless, the experimental -36 ( 1 and the calculated -39.6 ( 3.0 are consistent with a value of -37 kcal mol-1. For 4-methyl guaiacol, the calculated value is 3.1 kcal mol-1 too high compared to the experimental data. The experimental data is unique, and a confirmation is needed. The experimental data is obtained indirectly according to eq 2 by considering a ∆vapH°298K of 16.9 kcal mol-1. This data is not reported in the exhaustive reference compilation of enthalpies of vaporization of Chickos et al.23 and needs, therefore, confirmation. The estimated value according to ref16 for the enthalpy of formation of gas-phase 4-methyl guaiacol is -67.6 kcal mol-1. This estimate is closer to the calculated data than to the experimental data. For all the other species, no important differences are noted between experimental data and calculated data. In some cases, the comparisons between experimental data and calculated data lead to interesting conclusions. For instance, two data are available in the literature for the gas-phase standard enthalpy of formation at 298.15 K of dibenzofuran: 11.3 ( 1.122 and 13.2 ( 0.1 kcal mol-1.24 NIST Chemistry Webbook22 reports the first one only, although the second one is generally considered as more reliable.25 The method proposed in this study provides a value of 13.7 kcal mol-1, which is very close with the recommended value 13.2 ( 0.1. For phenol, the absolute difference between experimental data (-23.0) and calculated data (-0.9) is within 3 kcal mol-1. However, an experimental value of -20.4 kcal mol-1, very close with the calculated data, is also reported in the literature.26 Calculated data are generally consistent with estimated data obtained using group-additivity methods.17 As discussed previously in the Computational Method section, the approach used here is much more broadly applicable than the group additivity method because it uses just a single empirical correction for each atom type, rather than a separate empirical value for each group. For most of the compounds, no experimental measurement is reported for Cp and S, except for some compounds at only one temperature, generally 298.15 K. One of the most successful (23) Chickos, J. S.; Acree, W. E., Jr. J. Phys. Chem. Ref. Data 2003, 32, 519–878. (24) Chirico, R. D.; Gammon, B. E.; Knipmeyer, S. E.; Nguyen, A.; Strube, M. M.; Tsonopoulos, C.; Steele, W. V. J. Chem. Thermodyn. 1990, 22, 1075–1096. (25) Dorofeeva, O. V.; Iorish, V. S.; Moiseeva, N. F. J. Chem. Eng. Data 1999, 44, 516–523. (26) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin, R. D.; Mallard, W. G. J. Phys. Chem. Ref. Data 1988, 17, supplement No. 1.

Thermochemistry of Compounds from Pyrolyzed Biomass

procedures for obtaining fundamental vibration frequencies is to use scaled frequencies computed at the B3LYP/6-31G(d) level of theory. Unscaled B3LYP/6-31G(d,p) vibration frequencies, from which heat capacities and entropies can be derived, are reported as Supporting Information. Some bio-oil components given in Table 2 are not considered in the Supporting Information because their thermochemistry is available in Burcat’s thermodynamic tables.27 The entropy and heat capacity can be calculated following the harmonic oscillator approximation. This approximation is usually the one used in the combustion chemistry field but, at high temperature, errors can result if the harmonic oscillator approximation is used for low-frequency modes that represent hindered internal rotations or free internal rotations.28,29 Because the partition functions for a harmonic oscillator and for a free or hindered rotor are quite different, some corrections to the heat capacity and entropy are needed. Normal vibration modes corresponding to internal rotations can be identified by using the procedure of Ayala and Schlegel28 as implemented in Gaussian 03. Conclusions A protocol has been derived and validated for the predictions of the thermodynamic data needed for thermochemical calcula(27) Alexander Burcat and Branko Ruscic Ideal Gas Thermochemical Database with updates from Active Thermochemical Tables ftp://ftp. technion.ac.il/pub/supported/aetdd/thermodynamics (accessed 15 July 2008). Mirrored at http://garfield.chem.elte.hu/Burcat/burcat.html (accessed 15 July 2008). (28) Ayala, P. Y.; Schlegel, H. B. J. Chem. Phys. 1998, 108, 2314– 2325. (29) Katzer, G.; Sax, A. F. J. Phys. Chem. A 2002, 106, 7204–7215.

Energy & Fuels, Vol. 22, No. 6, 2008 4273 Table 4. Atomic Corrections ci* Used in the Present Model atom

ci* (hartree atom-1)

Ha Ca O

0.581896 38.115345 75.150410

a These atomic corrections are not recommended for highly strained polycyclic alkanes (tricyclic, tetracyclic, etc. alkanes). These atomic corrections are recommended for other classes of alkanes.

tions and for the building of detailed chemical kinetic models to simulate the high-temperature reactivity of bio-oil. This research showed that there are a number of the constituents for which thermochemical properties were incomplete, in question, or completely unknown. This is not surprising for large molecules such as coniferyl alcohol, for instance, but is more surprising for small species such as hydroxyacetaldehyde. The level of theory, hybrid density functional theory with a modest basis set plus empirical corrections, is found to be appropriate for the size of the molecules and the current state of knowledge (quite minimal) of their thermochemistry. Acknowledgment. We would like to thank the French Research National Agency (ANR) for its financial support and the French Agency for Environment and Energy Management (ADEME) for its coordination in the frame of the GALACSY project. Supporting Information Available: Additional materials mentioned within the text are separately provided. This material is available free of charge via the Internet at http://pubs.acs.org. EF800418Z