Back to search

National Technical University of Athens (NTUA)

Sustainable conversion of biomass oxygenates: ab initio based model simulation

Abstract

dc:description

Biomass, as an abundant source for renewable energy, can provide a considerable contribution tofacing the global climate change and its impacts on humanity. The conversion of biomass intofuels, energy, specialty chemicals and industrial products is expected to increase in importancethe coming years due to the concern for sustainable development and environmentally friendlierproducts. Many of the most important biomass conversion processes, such as combustion,gasification and pyrolysis are based on free-radical chemistry. The optimization of suchprocesses requires the development of fundamental kinetic models that contain thousands ofreactions. Thermodynamic and kinetic parameters for all compounds involved are necessary andthe determination of all these parameters is a major challenge in kinetic modeling.It is impossible to acquire all these parameters needed from experimental data only, due to thereactive nature of radical intermediates. Computational chemistry offers a reliable andcomplementary tool to experiment, allowing to obtain accurate thermodynamic and kineticparameters for a variety of compounds. Its main drawback is the increasing computational costfor larger compounds. Alternatively, accurate data for larger compounds can be obtained byusing group contribution methods based on structure-property relations that link the corresponding parameter to structural functional subunits. These parameters can be derived usingcomputational chemistry methods for smaller compounds.Objective: The objective of this research is to develop consistent and accurate models for theprediction of the thermochemistry of oxygenate compounds and of the Arrhenius parameters andthe rate coefficients for some of the most important reaction families involved in the thermaldecomposition of oxygenates. The acquired thermodynamic and kinetic data and the groupadditive models developed will be applied in simulating the pyrolysis of methyl butanoate, acommonly used biodiesel model compound.Thermochemistry: Accurate thermodynamic parameters are necessary when developingreaction networks since they can be used for the determination of reaction enthalpies andreaction equilibria. In this work the accurate CBS-QB3 ab initio method was employed, withcorrections for one-dimensional hindered rotation (1D-HR), to determine thermodynamicparameters (standard enthalpy of formation, ΔfHo, standard entropy, So, and heat capacities, Cp)for 450 oxygenate molecules and radicals. From this dataset 157 group additive values (GAVs)for oxygenate molecules and radicals have been determined within Benson’s group additivitymethod. To improve the accuracy, 26 corrections for non-nearest-neighbor interactions areintroduced for oxygenate molecules and radicals. The vast majority of these values have neverbeen determined before. As an alternative to the group additive method for radicals, the hydrogen bond increment method (HBI) is used to introduce 77 new HBI structures and tocalculate their thermodynamic parameters. The developed group additive model was validatedby comparing model predictions with both experimentally obtained and ab initio calculated data,resulting in mean absolute deviations below 4 kJ mol-1for standard enthalpies of formation andbelow 4 J mol-1 K-1for standard entropies and heat capacities. Kinetics: Two types of reactions have been studied in this work, hydrogen abstraction reactionsand carbon radical additions with their reverse β-scissions involving oxygenates. Ratecoefficients in this study are determined using the conventional transition state theory (CTST).Ab initio calculations are performed by the CBS-QB3 method including corrections for onedimensionalhindered rotation around the forming/breaking bond in the transition state. Forhydrogen abstraction reactions also zero curvature tunneling corrections according to the Eckartscheme are included.For the modeling of kinetics, the group additive method developed at the Laboratory forChemical Technology (LCT) of Ghent University was used. With this method, Arrheniusparameters of a target reaction can be determined by adding perturbations to the correspondingparameters of a reference reaction. These perturbation terms refer to structural differencesbetween the transition states of the studied and the reference reaction, and are based upon groupadditivity for enthalpy and entropy. In this study it was shown that the inclusion of secondarycontributions is necessary to obtain an accurate model. For hydrogen abstractions, correctionsaccounting for the effect of resonance stabilization around the transition state were shown to benecessary for increasing the accuracy of the model. Four group additive models have beendeveloped in the framework of this work, which will be discussed hereafter. Hydrogen abstractions from oxygenates by carbon-centered radicals: A dataset has beendeveloped consisting of ab initio calculated Arrhenius parameters and rate coefficients for 103hydrogen abstraction reactions from oxygenates by carbon-centered radicals. Based on these datafifteen group additive values (ΔGAVos) are obtained for the forward and the reverse direction ofthe reaction accounting for primary and secondary contributions and also nine corrections forcross-resonance stabilization of the transition state. The resulting group additive model has beenvalidated upon an ab initio dataset showing mean absolute deviation smaller than 0.250 log(m3mol-1s-1) and 1.5 kJ mol-1for pre-exponential factors and activation energies, respectively, for the forward direction of the reaction at 300 K. Comparing the group additive predictions with aset of 33 experimental rate coefficients shows that the mean factor of deviation between groupadditively predicted and experimentally obtained rate coefficients is only 2.2.Hydrogen abstractions from oxygenates by hydrogen atom: The same approach is adopted forhydrogen abstractions by hydrogen atoms, and kinetic parameters for 60 reactions have beendetermined. From this set of reactions, fifteen group additive values for Arrhenius parameters forthe forward and the reverse reaction have been obtained; these ΔGAVos account for primary andsecondary contributions. The accuracy of the model was validated upon an ab initio and anexperimental validation set, showing that rate coefficients can be predicted with a meandeviation factor of 2 and 3, respectively.Hydrogen abstractions from oxygenates by oxygen-centered radicals: This reaction familyinvolves two types of transition states, that is, the abstraction of a hydrogen atom linked to acarbon or an oxygen atom by an oxygen-centered radical, which correspond to transition statesof the O--H--C and O--H--O types, respectively. A total set of 118 reactions was studied andkinetic parameters have been obtained by ab initio calculations at temperatures 300-2500 K.Based on this set of reactions, 43 ΔGAVos are determined for primary and secondarycontributions, along with 37 corrections accounting for cross-resonance effects in the transitionstate. The developed group additive model has been validated upon an ab initio validation set including 25 reactions and an experimental validation set of 61 rate coefficients with a meanfactor of deviation of ~3 both for the two validation sets.Carbon-centered radical additions to oxygenates and β-scissions of oxygenates: A database ofArrhenius parameters and rate coefficients for 66 carbon radical additions to oxygenates andtheir reverse β-scissions of oxygenates was developed covering temperatures from 300 to 2500K. Based on these data, 32 group additive values are determined accounting for primary and secondary contributions. In the absence of experimental data in literature, an ab initio validationset is used to compare with the corresponding group additively predicted values, resulting in amean factor of deviation of 3 both for the radical addition and the β-scission reactions. To thebest of our knowledge ΔGAVos for these reaction families have never been determined before,whereas experimental data for this type of reactions are very scarce.Simulation of methyl butanoate (MB) pyrolysis: The program Genesys, developed at the LCT,allows an automated construction of reaction networks for hydrocarbons, oxygenates and severalother compounds. The calculated ab initio data and the group additive values determined forthermochemistry and kinetics have been implemented in the corresponding Genesys databases.The reaction network generated by Genesys allows to study the thermal decomposition of largeoxygenate compounds. In this study the pyrolysis of methyl butanoate was simulated using abinitio based group additivity for the reactions between larger compounds combined with anexperiment-based seed mechanism for the thermal decomposition reactions of the smallercompounds. The results of this simulation for MB conversion and the majority of the productyields agree well with the corresponding available experimental values, which shows theusefulness of the constructed group additive models for use in large reaction networks.Conclusion: In this study group additive models have been developed for an accurate prediction of thermodynamic parameters for oxygenate compounds and kinetic parameters for reactionsbetween oxygenates. The Genesys software, developed at the LCT, was successfully used togenerate a reaction network for the pyrolysis of methyl butanoate. Based on the agreement in theMB conversion and the majority of the product yields to experimental data, the usefulness andaccuracy of the developed group additive models for the simulation of the thermaldecomposition of oxygenate compounds has been shown.

Degree

thesis:*
Grantor dc:publisher
National Technical University of Athens (NTUA)
Year dc:date
2015

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Paraskevas, Paschalis
  • Παρασκευάς, Πασχάλης

Subjects

dc:subject × 14

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/43243

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Paraskevas, Paschalis; Παρασκευάς, Πασχάλης. Sustainable conversion of biomass oxygenates: ab initio based model simulation. National Technical University of Athens (NTUA), 2015. http://hdl.handle.net/10442/hedi/43243