Project Topics
Get Complete Project Material File(s) Now! »
HydroDeOxygenation of Pyrolysis Bio-oils
The gas phase HDO of lignin-derived pyrolysis bio-oil vapors, refers to high temperature (623-873K) chemical conversion to obtain oxygen-free molecules, and water as a by-product. It consists in several reactions, mainly hydrocracking, hydrogenation, dehydration, hydrogenolysis, demethylation (DME), demethoxylation (DMO) and decarboxylation.49 Pyrolysis coupled with online HDO process results in optimizing the heat integration, and reducing the problem of condensed bio-oil feeding.51,52 In 2000, as a brief literature review on HDO, Furimsky44 studied the kinetics and reaction networks of HDO, and concluded that more stable catalysts are needed to make fuel production more attractive. In 2007, Elliott45 reviewed the historical development of catalytic hydroprocessing of bio-oils, while Choudhary and Phillips46 reviewed the applications of HDO from an industrial perspective, using different bio-oil feeds, catalysts, and operating conditions. More recently, Bu et al.47 reviewed the HDO of phenols derived from lignin pyrolysis. The kinetics and reactions mechanisms of the process were investigated and analyzed using different model compounds and the performance of various conventional (CoMo/Al2O3 and NiMo/Al2O3) and noble catalysts were discussed. They concluded that for various techno-economic problems, future investigations should be performed on improving catalysts, optimizing catalytic conditions, and understanding reaction kinetics. He & Wang48 considered advances in HDO using various bio-oil models and catalysts (noble and non-precious metals, sulfides, nitrides, metal oxides…). They also studied the importance of the support, and proved that O from oxy-compounds is adsorbed on H in -OH of non-metal oxides (such as SiO2).
The HDO process includes two main deoxygenation routes, (i) the hydrogenation of the aromatic ring before C-O bond cleavage (HYD), or (ii) the direct C-O bond cleavage, Direct Deoxygenation (DDO) (see Figure 1.6). Most investigations in this field aim to promote the DDO route, in order to avoid hydrogenation of the aromatic ring, and therefore to reduce H2 consumption (which in consequence, decreases the working pressure and costs) and to produce more valuable aromatics (BTX). In addition to the enormous number of experimental studies in this field, theoretical Density Functional Theory (DFT) calculations are frequently used to study the adsorption of oxygenated molecules on the catalyst surface under HDO conditions, and the effect of inhibitors, for understanding the reaction pathways and mechanisms.
Catalysts and Supports for HDO Reaction
Several catalysts and supports have been tested for biomass conversion under HDO conditions;46– 49,65 Table 1.2 shows some examples under various operating conditions while Table 1.3 compares the conversion yield of guaiacol and selectivity of some catalysts. The choice of the appropriate catalyst depends on its activity and selectivity toward aromatics (DDO route), its deactivation rate under operating conditions and its suitability for environmental use. Noble metals showed a high activity but they are expensive and suffer from high deactivation and reactor clogging at high temperatures.34 Conventional metal sulfides are very active and selective toward aromatics but their stability depends on the H2S/H2O pressure ratio, which requires feeding with H2S, and in consequence poisoning the product.54 Low-price transition metals showed an optimal activity/selectivity ratio.
Adsorption mechanisms of oxygenated molecules on silica surface
Popov et al.68 studied the adsorption of phenolic molecules on oxide supports, which is the first step of HDO mechanism. Infrared spectroscopy investigations revealed that phenolic compounds interact mainly with the silica surface via H-bonds with silanols groups. Their study confirms that carbon deposition takes place on alumina support due to the presence of Lewis acid-base pairs on this surface, and that phenates are produced at high HDO operating temperatures. This study therefore suggests that silica support is a potential candidate with high stability.23 The adsorption of phenol, anisole and guaiacol on silica was studied by following the perturbations of the IR spectra of the Si-OH peak. Corresponding adsorption mechanisms are proposed (see Figure 1.11).
Performance of Fe@SiO2 for guaiacol HDO conversion
Olcese et al.24 studied the catalytic HDO conversion of guaiacol in a fixed-bed reactor using Fe@SiO2 catalyst synthetized by simple impregnation of commercial silica (Aerosil 130, Degussa) in an iron nitrate solution. They have studied the effect of operating conditions such as the reaction temperature, H2 partial pressure and catalyst contact time, and have compared their results with those obtained using a commercial cobalt-based catalyst. Guaiacol conversion yield without hydrogen was about 30%, while it became about 70% with a H2 partial pressure of 0.2 bar; however, for a higher H2 partial pressure (0.2-0.9 bar), HDO conversion remained almost unchanged, and the benzene yield increased slightly. As the reaction temperature increased, guaiacol conversion (activity) and benzene production (selectivity) increased; however, in the same time, the production of undesirable by-products such as CO and CH4 increased. Different contact times from 0.11 to 1.5 hours were tested, showing that at a contact time of 0.8 hr., guaiacol is totally converted with the highest phenol yield being 60 vol.%, and with the benzene yield being about 14 vol.%. At a contact time of 1.5 hr., 673 K and 90% H2, the guaiacol HDO conversion on Fe@SiO2 catalyst achieved a yield of about 74% with a BTX yield of 38%. Comparing these values with those obtained with Co-based catalyst, Fe@SiO2 catalysts show lower guaiacol HDO conversion but much higher BTX yield (selectivity) and lower coke deposition than Co-based catalysts, making Fe@SiO2 an attractive inexpensive environmental-friendly catalyst.
The effect of by-products (CH4, CO, CO2 and water) was investigated on guaiacol conversion for iron-supported catalysts. These studies showed that water and CO have negative effects, as water inhibits the production of benzene and reduce the overall activity, and CO promotes coke formation due to the carburization of iron particles. While methane has negligible effects, CO2 has a positive effect as it reduces the catalyst deactivation.25 Coke deposits were observed near iron particles, which suggests that the reaction occurs at the metal-silica interface. The composition and type of iron species were identified using Mössbauer spectra before and after catalysis. The results showed the existence of residual Fe(III), Fe2O3, γ-Fe5C2 in addition to α-Fe particles. The comparison of different iron loads (5, 10 and 15%) showed that the guaiacol conversion is proportional to the exposed iron surface, with similar yields of aromatics and phenols. In order to understand the effect of silanol defects, the silica support was replaced with activated carbon. The activity was similar but no benzene or toluene were produced. The hydrotreatment of lignin pyrolysis vapors using Fe@SiO2 catalyst was reported, and the lignin to BTX process was simulated based on the kinetic model integrated in Aspen Plus as shown in Figure 1.
Applying DFT for Periodic Systems
The resolution of the Schrödinger equation for a crystal implies, in the Kohn-Sham method, the determination of an infinity number of mono-electronic wave functions; such a calculation is impossible. The plane wave expansion of KS wave functions is thus useful, since it takes the advantage of the crystal periodicity.
In the real space, the structure is characterized by a density function that determines how much matter exist at a given point. While in the reciprocal space, the structure is defined in a function of periodicity k are required to produce the same arrangement of atoms. Therefore, using Bloch theorem, each electron wave function , ( ) is written as the product of a plane wave ⃗⃗ ⃗ of vector (in the first Brillouin zone BZ) and a ⃗ crystal.
periodic function f having the same periodicity as the ⃗⃗ , ( ⃗) = f * ⃗⃗ ⃗ = ∑ ⃗ ,⃗⃗,⃗⃗⃗⃗ ∗ ⃗⃗+⃗⃗⃗⃗ ⃗ , ⃗⃗,⃗⃗⃗⃗ are coefficients for a chosen basis set. (23) Where ⃗ are reciprocal lattice vectors and The problem then returns to calculating a finite number of wave functions for an infinite number of ⃗⃗ . However, applying Bloch theorem, the calculation can be solved in a finite number of points represented by the first Brillouin zone. Thus, the convergence of the energy must be verified for the chosen points. On the other hand, it appears that the coefficients associated with the plane waves⃗⃗of great kinetic energy are markedly lower than those associated with plane waves of ,⃗⃗,⃗⃗⃗⃗ lower kinetic energy. This justifies the introduction of a cut in the base to a certain energy called − cut-off energy . In practice, it is also important to test the convergence of the system energy with the cut-off energy.
The properties of solids generally depend much more on valence electrons than on those of the heart. It thus seems more judicious to represent the electrons of the heart by an effective potential called pseudo-potential, instead of explicitly dealing with them. The notion of pseudo-potential makes it possible to modify the potential near the cores, so as to be affected by the core electrons while retaining the influence (by screen effect) of valence electrons.44 For periodic calculations, the pseudo-potentials US-PP method45 and the Projected Augmented Wave (PAW) approach developed by Blöchl46 are commonly used. PAW method consists in linking by a linear transformation the function of all electrons (electrons of valence + core) to a pseudo-function of valence.47 This method has yielded a good accuracy to calculation time ratio for various systems.
Table of contents :
Chap. 1 Bibliographic Review
1.1 Lignocellulosic Biomass
1.2 Lignin Valorization and Market Products
1.3 Lignin Pyrolysis and Hydrothermal Conversion
1.4 Hydrodeoxygenation of Pyrolysis Bio-oils
1.5 Reaction Mechanisms of HDO Model Molecules
1.5.a Conversion of phenol
1.5.b Conversion of guaiacol
1.6 Catalysts and Supports for HDO Reaction
1.7 Fe@SiO2 Catalyst
1.7.a Silica surface support
1.7.b Adsorption mechanisms of oxygenated molecules on silica surface
1.7.c Performance of Fe@SiO2 for guaiacol HDO conversion
1.8 DFT for Chemical Applications
1.8.a DFT for HDO studies
1.8.b DFT for adsorption and catalysis on silica models..
1.9 Synthesis of Porous Silica Materials
1.9.a Surfactants and metallosurfactants
1.9.b Mesoporous silica materials
1.9.c Sol-Gel process
1.9.d Corporative Self-Assembly (CSA) mechanism
Chap. 2 Theoretical and Experimental Methods
Part. 1 Density Functional Theory
2.1.1 Before Density Functional Theory
2.1.2 Early DFT Approximations
2.1.3 Hohenberg-Kohn Theorem
2.1.3.a First theorem
2.1.3.b Second theorem
2.1.4 Kohn-Sham Equations
2.1.5 Exchange-Correlation Functionals
2.1.5.a Local density approximation (LDA)
2.1.5.b Generalized gradient approximation (GGA)
2.1.5.c Meta-GGA and hybrid functionals
2.1.6 Applying DFT for Periodic Systems
2.1.7 Dispersion Correction Methods
Part. 2 Silica-Supported Catalysts
2.2.1 Synthesis of SBA-15 Silica Material….
2.2.2 Synthesis of Supported Metal Catalysts
2.2.2.a Chemical impregnation
2.2.2.b Co-precipitation
2.2.2.c Deposition-precipitation
2.2.2.d Microemulsions
2.2.2.e Other methods
2.2.3 Synthesis of Single Atom Catalysts (SACs) using Metallosurfactants Templates
2.2.3.a Reminders on SACs
2.2.3.b SACs using metallosurfactant templates
Part. 3 Hydrodeoxygenation Process
2.3.1 Introduction
2.3.2 Characterization Techniques
2.3.2.a Gas chromatography μGC
2.3.2.b Mass spectrometer MS….
2.3.2.c Gas chromatography mass spectrometry GCMS
Chap. 3 Atomistic Description of Phenol, CO and H2O Adsorption over Crystalline and Amorphous Silica Surfaces for Hydrodeoxygenation Applications
Abstract
3.1 Introduction
3.2 Materials and Methods
3.2.1 Models of Silica Surfaces
3.2.1.a β-cristobalite surface
3.2.1.b [001] α-quartz surface
3.2.1.c Amorphous surfaces
3.2.2 Computational Methods
3.3 Results and Discussions
3.3.1 Phenol Adsorption Modes on Silica Surfaces
3.3.1.a Crystalline silica surfaces
3.3.1.b Amorphous silica surfaces
3.3.2 Competitive Adsorption of Inhibiting Molecules
3.4 Conclusions
References
Chap. 4 Synthesis of Metal@silica Catalysts
Part. 1 Conventional Synthesis of Catalysts
4.1.1 SBA-15-like Silica Support
4.1.2 Synthesis of Fe, Cu and Fe-Cu@silica Catalysts by the Impregnation Method
4.1.3 Synthesis of Fe@SiO2 Following the Co-precipitation with Decomposition of Urea (DPU)
Part. 2 via Mixed Micelles
4.2.1 Physico-Chemical Characterization of Mixed Micelles Solutions
4.2.2 Small Angle Neutron Scattering (SANS) Experiments
4.2.3 Magnetic Studies
4.2.4 Conclusions
Part. 3 Imprinting Isolated Single Iron Atoms onto Mesoporous Silica by Templating with Metallosurfactants
Abstract
4.3.1 Introduction
4.3.2 Materials and Methods
4.3.2.1 Chemicals
4.3.2.2 Preparation of Silica Materials
4.3.2.3 Characterization Methods
4.3.2.3.a Nitrogen sorption measurements
4.3.2.3.b X-ray diffraction
4.3.2.3.c Elemental analysis
4.3.2.3.d Spectroscopic measurements (ATR and UV-Vis)
4.3.2.3.e Transmission electron microscopy and elemental cartography
4.3.2.3.f Total scattering X-ray measurements and PDF analysis
4.3.2.3.g Solid-state nuclear magnetic resonance
4.3.2.3.h Magnetic measurements
4.3.2.3.i Density functional theory calculations
4.3.3 Results and Discussions
4.3.3.1 Synthesis of Silica Materials
4.3.3.2 Structural and Textural Characterization of Fe@SiO2(r) Materials
4.3.3.3 Chemical Characterization of Fe@SiO2(r) Materials
4.3.3.4 Characterization of Single Atom Fe@SiO2(r) Materials
4.3.3.5 DFT Calculations
4.3.4 Conclusions
Chap. 5 Hydrodeoxygenation Catalytic Tests
5.1 Introduction
5.1.1 Catalytic Materials
5.1.2 Reaction Conditions
5.1.3 Conversion, Activity, Aromatics Selectivity, Carbon Yield, and Carbon Selectivity
5.1.4 Blank Test (empty reactor)
5.2 Catalytic Performance of Fe@SiO2(10) Single Atom Catalysts (SACs)
5.3 Catalytic Performance of Conventionally Synthetized Catalysts
5.4 X-ray and Mössbauer Measurements of Conventionally Synthetized Catalysts
5.5 Conclusions
Conclusions and Perspectives
Annexes
Annex 1. Dispersion Forces (vdW) Contribution
Annex 2. Structural Properties of Pristine SBA-15-like Material
Annex 3. Thermogravimetric Analysis (TGA)
Annex 4. Magnetic Measurements
4.1 Substraction of the Diamagnetic Signal
4.2 Low Temperature Spin Crossover
References
