Biomass Fast Pyrolysis Process Research Articles

Hybrid Biomass Fast Pyrolysis Process and Solar Thermochemical Energy Storage System, Investigation and Process Development

Hybrid Biomass Fast Pyrolysis Process and Solar Thermochemical Energy Storage System, Investigation and Process Development

A deep-learning model for predicting spatiotemporal evolution in reactive fluidized bed reactor

Detailed information of flow fields is of great significance for designing and optimizing multiphase flow systems. However, predicting spatiotemporal evolution of gas-solid flows using numerical simulation often requires a significant amount of computation and time. In this study, we proposed a 3D convolutional neural network for predicting reactive dense gas-solid flows. We first explored the design of model architecture and extensively evaluated the performance in terms of efficiency, accuracy, long-term prediction stability and generalizability for a non-reactive fluidized bed. Then we extended the method to a biomass fast pyrolysis process. The proposed model achieves real-time prediction, 3–4 orders of magnitude faster than CFD-DEM simulations. The surrogate model reasonably captures bubble-driven flow behaviors and effects of bubble on fast pyrolysis reactions. The predicted bubble characteristics, and time-averaged and RMS flow fields match well with the simulation results. Our approach exhibits excellent long-term stability and has good generalization capability to unseen fluidization velocities. To the best of our knowledge, this is the first time a neural network has been successfully applied to learn spatiotemporal evolution of reactive dense gas-solid flows.

Thermodynamic Evaluation of the Intermediate Liquid Compounds (ILC) from Biomass Fast Pyrolysis

Biomass fast pyrolysis process is a technology that converts renewable solids into a dense liquid. This study aims to apprehend the thermodynamic behaviour of the Intermediate Liquid Compounds (ILCs) observed during the biomass fast pyrolysis. The system studied was a closed system (20 mL) with air and a mixture solution of five components (Acetic acid (AA), hydroxyacetone (HX), phenol; furfural (FF) and methanol) at 90°C and under atmospheric pressure. The flash calculation was conducted at a given temperature and pressure. The vapor-liquid equilibrium compositions were determined combining equation of state and activity coefficient models, the Soave-Redlich-Kwong (SRK) equation of state coupled with Modified Huron-Vidal (MHV2) mixing rules incorporating the UNIversal Functionnal Activity Coefficient (UNIFAC) model. Theoretical calculations of vapor-liquid equilibrium compositions were experimentally validated by using a Head-Space GC-MS system. A quantitative agreement between simulated and measured concentrations in the liquid phase was achieved with this combined state-predictive model of SRK-MHV2-UNIFAC model; thus, confirming that it accounts well for the nonidealities.

Kinetic model for esterification of oleic acid catalyzed by a green catalyst in ethanol

Bio-oil produced from biomass fast pyrolysis process is an organic liquid that can be transformed into high quality biofuel through upgrading process. Bio-oil comprises a large amount of carboxylic acids, which cause its corrosiveness, hence, difficulty for storage and utilization. Esterification may be considered as an intermediate upgrading method to convert the acids in bio-oil into esters by reacting with alcohol. This work presents kinetic studies of the esterification of a bio-oil model compound in ethanol catalyzed by 12-tungstosilicic acid. Oleic acid was used to represent the main component of bio-oil derived from agro-residues. Conditions considered were the molar ratio of ethanol to oleic acid of 9.11:1, catalyst loading of 10 wt%, and reaction temperatures in the range of 35-75 °C. It was found that the pseudo-homogeneous kinetic model was suitable to represent the esterification of the bio-oil model compound. Average reaction order n = 1.9, activation energy Ea=50.1 kJ/mol, and pre-exponential factor A = 11.11×103 min−1 were obtained by fitting the kinetic model with experimental results.

Experimental study on fluidization, mixing and separation characteristics of binary mixtures of particles in a cold fluidized bed for biomass fast pyrolysis

To guarantee the uniform mixing of biomass and inert particles in the fluidized bed reactor for the full and complete reaction during the biomass fast pyrolysis process, a visualized cold fluidized bed experimental apparatus was set up to investigate the effects of static bed height, moisture content and particle diameter on the fluidization quality of single component particles, and the mixed volume ratio on the fluidization characteristics of binary mixtures. The results showed that the appropriate static bed height was significant for the stability of the fluidized bed. As the moisture content increased, the minimum fluidization velocity increased rapidly while the fluidization index continued to decrease. In addition, the fluidization performed on binary mixtures of larch/quartz sand-2 indicated that the same minimum fluidization velocity was the key parameter for the two kinds of particles to be fully fluidized simultaneously, avoiding the semi-fluidization phenomenon. The mixed volume ratio of larch (M = 10.3%)/quartz sand-2 and superficial gas velocity range were also the important factors affecting fluidization and mixing quality, and the optimal value were 1: 3 and 3–5 umf, respectively.

Fluidisation characteristics and inter-phase heat transfer on product yields in bubbling fluidised bed reactor

Fluidisation and mixing of sand and biomass particles have significant effects on product yields in bubbling fluidised bed reactors (BFBR). There is a lower limit for the velocity of fluidising media known as minimum fluidisation velocity. Similarly, an upper limit is defined for velocity where the bio-oil production is maximised – known as the maximum effective velocity (MEV). A computational fluid dynamics (CFD) simulation for biomass fast pyrolysis process in a 2-D lab-scale BFBR is developed to analyse the variation of MEV as the sand particle size varies. The model is first validated using the experimental data. Then, a parametric study is conducted for the carrier gas velocities in a range of 0.3–1.1 m/s where the sand particle sizes vary from 0.4 to 1 mm, and the biomass particles are in a range of 0.2–0.5 mm. Effects of the packing limit are also analysed. For this purpose, the different sand particle sizes and their corresponding MEV are tested at the sand packing limits of 45, 55, 65, and 75 mm in which the optimal packing limit is found to be 55 mm. A detailed thermodynamic study is performed with the focus on the sand particle size and packing limit affecting the heat transfer rates between phases. A decrease in required heat transfer rates is directly linked to an increase in bio-oil yield. It is observed that heat transfer between sand-biomass is much more efficient than the heat transfer between nitrogen-biomass, confirming the importance of the sand particles as the heat carriers.

A DEM modeling of biomass fast pyrolysis in a double auger reactor

Thermochemical conversion of biomass via fast pyrolysis is a proven pathway to product low-carbon crude bio-oils. In this research, an extended discrete element method (DEM) is proposed for simulating biomass fast pyrolysis reacting granular flows in a double auger reactor, in which particle hydrodynamics and interparticle heat transfer processes are involved and coupled with chemical reactions in solid particles. An adaptive time step algorithm is proposed to achieve a stable coupling between the integration of reaction ordinary differential equations and the DEM solver, and the algorithm is proven computationally efficient. A multi-component fast pyrolysis kinetics is adopted and its modeling accuracy is assessed by carrying out simulations of benchmark biomass pyrolysis experiments and comparing the prediction results with experimental data. The predicted product yields of bio-oil, char and non-condensable gas from the simulation of the biomass fast pyrolysis in the auger reactor are in satisfactory agreement with experimental measurements. The decomposition rates of biomass components in the reactor are revealed from the simulation and the pyrolysis number Py is calculated from the decomposition rate of biomass and the heat transfer coefficient. The Py number illustrates that the biomass fast pyrolysis process is limited by the heat transfer process at particle size of 2 mm.

A hybrid SVR-PSO model to predict a CFD-based optimised bubbling fluidised bed pyrolysis reactor

Comprehensive scrutiny is necessary to achieve an optimised set of operating conditions for a pyrolysis reactor to attain the maximum amount of the desired product. To reach this goal, a computational fluid dynamic (CFD) model is developed for biomass fast pyrolysis process and then validated using the experiment of a standard lab-scale bubbling fluidised bed reactor. This is followed by a detailed CFD parametric study. Key influencing parameters investigated are operating temperature, biomass flow rate, biomass and sand particle sizes, carrier gas velocity, biomass injector location, and pre-treatment temperature. Machine learning algorithms (MLAs) are then employed to predict the optimised conditions that lead to the maximum bio-oil yield. For this purpose, support vector regression with particle swarm optimisation algorithm (SVR-PSO) is developed and applied to the CFD datasets to predict the optimum values of parameters. The maximum bio-oil yield is then computed using the optimum values of the parameters. The CFD simulation is also performed using the optimum parameters obtained by the SVR-PSO. The CFD results and the values predicted by the MLA for the product yields are finally compared where a good agreement is achieved.

Effect of fractional condensers on characteristics, compounds distribution and phenols selection of bio-oil from pine sawdust fast pyrolysis

Bio-oil is a multicomponent mixture of more than 400 types of organic compounds, with high water content. Fractionation of bio-oil may be a more efficient approach for primary separation of bio-oil. In this work, to better understand the effect of fractional condensers on bio-oil yield, physicochemical characteristics, compounds distribution and phenols selection during biomass fast pyrolysis process, a semi-automatic controlled fluidized bed reactor biomass fast pyrolysis system with four-stage condensers was developed. Average temperatures of Condensers 1, 2, 3, 4 were 32.39 °C, 26.74 °C, 24.06 °C and 23.68 °C, respectively. And the bio-oil yields of Condenser 1, 2, 3, and 4 were 26.82%, 7.31%, 1.48% and 9.69%, respectively. Bio-oil collected from Condenser 4 had the lowest water content (9.68 wt%), the lowest acidity (pH = 3.67), and the highest HHV (29.2 MJ/kg). The highest relative contents of compounds collected from Condenser 1, 2, 3 and 4 were 1-(4-hydroxy-3-methoxyphenyl)-2-Propanone (6.95%), trans-Isoeugenol (6.63%), Creosol (5.28%), and trans-Isoeugenol (6.69%), respectively. Fractional condensers affected the compounds distribution, but it has a stronger effect on relative heavy compounds (molar mass > 250) and a weaker effect on relative light compounds (molar mass < 200). Fractional condensers were more conducive to the selection of phenols with relative yield of more than 30%. Phenols, acids and furfurans tended to distribute at higher temperature, while alcohols, ethers and hydrocarbons tended to distribute at relative lower temperature, but the difference was small. The research has provided a reference for the production of bio-oil.

Multi-fluid modeling biomass fast pryolysis with particle shrinkage model for complex reaction kinetics

In the gas-solid fluidized bed reactors for biomass fast pyrolysis, the diameter of particles normally shrink with the process of reaction, which not only affects the hydrodynamics such as segregation/mixing and entrainment behavior, but also influences the yield and composition of the products. Since the particle shrinkage model developed by our previous work (Powder Technology, 2016, 294: 43–54.) could be only applied for the simplified non-competitive single-component single-step reaction kinetics, in order to develop the advanced particle shrinkage model for complex pyrolysis reaction kinetics, three types of pyrolysis reaction kinetics considering the competitive reactions, the secondary cracking of tar, and the intermediate stage between virgin biomass and final products, respectively, were selected as the examples to illustrate the method for developing the relevant particle shrinkage models based on the mass conservation at the particle scale. Consequently, the biomass fast pyrolysis process in a bubbling fluidized bed reactor was simulated combing the particle shrinkage model and relevant pyrolysis reaction kinetics simultaneously. The influences of different particle shrinkage models and reaction kinetics on the flow and reaction behavior were revealed, and the predicted product yields were also compared with the experimental data.

Interaction characteristics and mechanism in the fast co-pyrolysis of cellulose and lignin model compounds

During biomass fast pyrolysis process, the interactions among biomass components will affect the pyrolytic products distribution. In this study, d-glucose and a β-O-4 type lignin model dimer (LMD, 1-(4-hydroxy-3-methoxyphenyl)-2-(2-methoxyphenoxy)propane-1,3-diol) were selected as the model compounds of cellulose and lignin. The interaction characteristics and mechanism during their fast co-pyrolysis process were investigated by combined pyrolysis–gas chromatography/mass spectrometry (Py–GC/MS) experiments and density functional theory (DFT) calculations. The Py–GC/MS results indicated that during fast co-pyrolysis process, the presence of LMD significantly decreased the formation of levoglucosan (LG) from d-glucose, while promoted the formation of linear carbonyls and furans. Meanwhile, the presence of d-glucose enhanced the decomposition of LMD to generate phenolic compounds. The DFT calculations revealed that d-glucose would interact with a homolysis radical of LMD to form a ten-membered ring transition state. The formed complex transition state changed the energy barriers of certain pyrolytic reactions of d-glucose and LMD, thus affecting the pyrolytic products distribution.

Numerical study of hydrodynamics with surface heat transfer in a bubbling fluidized-bed reactor applied to fast pyrolysis of rice husk

The present study investigates the hydrodynamics and heat transfer phenomena that occur during the biomass fast pyrolysis process. A numerical approach that combines a two-dimensional Eulerian multi-fluid model and the kinetic theory of granular flow has been applied to simulate the gas-solid flow in a bubbling fluidized-bed reactor. In this study, rice husk and quartz sand with specified properties were used as biomass and inert material, respectively. Our model was first validated the feasibility using previous findings, then an extensive parametric study was conducted to determine the effects of the major variables, especially the size of rice husk particles, on the flow distribution and the heat transfer between the phases. The concept of standard deviation attributed to the dispersion of solid volume fraction was used to calculate the intensity of segregation. The simulated results indicated that the mixing of binary mixture was strongly affected by different sizes of rice husk particles. The heat transfer occurring inside the fluidized bed was described by the distribution of solids temperature, the variation of surface heat flux and heat transfer coefficient. Both heat transfer quantities were observed to be dominant in the dense bed regions as they strongly depend on the solids concentration in the fluidized bed. The increasing inlet gas velocity promoted the mixing of solid particles, thus resulted in the effective heat transfer from wall to particles and between the particles.

Computational fluid dynamics modelling of biomass fast pyrolysis in fluidised bed reactors, focusing different kinetic schemes

The present work concerns with CFD modelling of biomass fast pyrolysis in a fluidised bed reactor. Initially, a study was conducted to understand the hydrodynamics of the fluidised bed reactor by investigating the particle density and size, and gas velocity effect. With the basic understanding of hydrodynamics, the study was further extended to investigate the different kinetic schemes for biomass fast pyrolysis process. The Eulerian–Eulerian approach was used to model the complex multiphase flows in the reactor. The yield of the products from the simulation was compared with the experimental data. A good comparison was obtained between the literature results and CFD simulation. It is also found that CFD prediction with the advanced kinetic scheme is better when compared to other schemes. With the confidence obtained from the CFD models, a parametric study was carried out to study the effect of biomass particle type and size and temperature on the yield of the products.

Modeling and simulation of biomass fast pyrolysis in a fluidized bed reactor

This paper presents a phenomenologically based mathematical model for biomass fast pyrolysis process in a bubbling fluidized-bed. The model was developed in transient state, is one-dimensional and is based on the two-phase theory. A semi-global reaction mechanism in two stages was used, considering the primary formation of products and the secondary reactions of the vapors. Also, population balances for the density/temperature distribution of the particles and distribution of the bubble size were proposed. The model can predict the temperature of the phases, the distribution and yields of the products, the heating rate of particles, and the residence time of gases. Finally, a solution algorithm for the model was proposed, which was programmed in MATLAB 7.0, being able to find a good fit between simulation results and the experimental data.

A numerical study on biomass fast pyrolysis process: A comparison between full lumped modeling and hybrid modeling combined with CFD

This study focuses on process modeling and simulation of a biomass fast pyrolysis system. For the simulation of the biomass fast pyrolysis process, two types of simulation models were developed: lumped model and hybrid model. Employing the above models, the effect of reaction temperature on reaction rate and final product yields were analyzed. It was found that the hybrid model exhibited a peculiar characteristic of displaying multiphase reacting flow occurring in fluidized bed. This behavior of hybrid model could have attributed for the difference in product yields of the models (hybrid and lumped). For the yields of the tar and NCG, hybrid model prediction was very consistent with the experimental results than the lumped model. However, for char yield, the results of both the lumped model and the hybrid model were close to that of experimental results.

A CFD study of biomass pyrolysis in a downer reactor equipped with a novel gas–solid separator-II thermochemical performance and products

A Eulerian–Eulerian CFD model was used to investigate the fast pyrolysis of biomass in a downer reactor equipped with a novel gas–solid separation mechanism. The highly endothermic pyrolysis reaction was assumed to be entirely driven by an inert solid heat carrier (sand). A one-step global pyrolysis reaction, along with the equations describing the biomass drying and heat transfer, was implemented in the hydrodynamic model presented in part I of this study (Fuel Processing Technology, V126, 366–382). The predictions of the gas–solid separation efficiency, temperature distribution, residence time and the pyrolysis product yield are presented and discussed. For the operating conditions considered, the devolatilisation efficiency was found to be above 60% and the yield composition in mass fraction was 56.85% bio-oil, 37.87% bio-char and 5.28% non-condensable gas (NCG). This has been found to agree reasonably well with recent relevant published experimental data. The novel gas–solid separation mechanism allowed achieving greater than 99.9% separation efficiency and <2s pyrolysis gas residence time. The model has been found to be robust and fast in terms of computational time, thus has the great potential to aid in future design and optimisation of the biomass fast pyrolysis process.

Catalytic Cracking of C2–C3 Carbonyls with Vacuum Gas Oil

The fluid catalytic cracking (FCC) route could be considered for the co-processing of biomass-derived pyrolysis oil with petroleum-derived vacuum gas oil (VGO) in a typical refinery unit by integrating it with a biomass fast pyrolysis process. This paper aims to study the effect of co-processing pyrolysis oil representative model compounds (C2–C3 carbonyls such as hydroxyacetone and glycolaldehyde dimer) with VGO in a FCC advanced cracking evaluation (ACE-R) unit, using an industrially available FCC equilibrium catalyst (E-CAT). The blending ratios of C2–C3 carbonyls and VGO were varied from 5:95, 10:90, 15:85, and 20:80. The reactor was operated in close to industrial FCC process at a temperature of 530 °C and atmospheric pressure with a catalyst-to-oil ratio of 5.0. The blended FCC feedstock and their liquid distillates were structurally characterized by 1H and gated decoupled 13C NMR techniques. The average structural parameters like branchiness index, substitution index, average length of alkyl chains, and fraction of aromaticity per molecule was obtained from NMR data. A linear increase in FCC conversion was observed with increase in hydroxyacetone blending ratio with VGO from 5% to 20%, which may be due to an increase in propylene and a decrease in heavy cycle oil. Increasing the glycolaldehyde blending ratio from 5% to 10% increased the FCC conversion, while the conversion decreased at ratios of 15% and 20%. The 1H NMR and 13C NMR data indicated that there is an increase in the amount of monoaromatics with increases in the amount of hydroxyacetone, whereas in the case of glycolaldehyde, the monoaromatics increased for blending ratios up to 10% and decreased upon further increases in blending ratio. The equilibrium FCC catalyst is capable of cracking oxygenates presents in pyrolysis oil such as hydroxyacetone and glycolaldehyde toward the production of liquefied petroleum gas (LPG) range products. Furthermore, a scheme has been proposed for the processing of pyrolysis oil in petroleum refinery units.

Modeling Effects of Operating Conditions on Biomass Fast Pyrolysis in Bubbling Fluidized Bed Reactors

This numerical study investigates the effects of operating conditions on the product yields of a biomass fast pyrolysis reactor. A numerical approach that combines a multifluid model and pyrolysis reaction kinetics was applied to simulate the biomass fast pyrolysis process in a bubbling fluidized-bed reactor. The model was first validated using experimental data, and a parametric study was conducted. Operating parameters were varied to characterize their effects on the final product yields. For the reactor studied, it was found that the maximum tar yield could be obtained by maintaining both the wall temperature and the inlet temperature of nitorgen at approximately 800 K. The inlet velocity of nitrogen at about 0.6 m/s also produced favorable results. Simulations indicated that the optimal biomass particle diameter and the feeding rate were 900 μm and 1.92 kg/h, respectively, for tar production. Larger sand particle diameter and deeper initial sand bed also favored the tar yield.

Effects of CO2 on biomass fast pyrolysis: Reaction rate, gas yields and char reactive properties

The effect of CO2 introduction in a biomass fast pyrolysis process at 850°C was investigated. It was found that CO2 impacts the final gas yield and composition, and the char yield and properties. Introducing CO2 in the pyrolysis medium alongside nitrogen enhanced CO production as a result of homogeneous and heterogeneous reactions of CO2 with gases, tars and char. The char yield was lower compared to a reference char yield in pure nitrogen. The char obtained in a CO2-containing atmosphere has its surface area increased nearly sixfold and has a chemical composition different from that of chars obtained in N2 atmosphere. However, the reactivity of the two chars towards H2O, CO2 and O2 was almost the same. Temperature-programmed oxidation experiments on both chars – those obtained in pure nitrogen and those obtained in a CO2-containing atmosphere – revealed quite different oxidation profiles and peak temperatures. Taken together, these results tend to confirm that CO2 is impacting the biomass fast pyrolysis process. In the light of these results and of the literature findings, we propose a mechanism illustrating the role of CO2 during fast pyrolysis of biomass.

Hydrodynamic behavior in a moving granular bed filter for modeling on char separation during the biomass fast pyrolysis process

Moving granular bed filters (MGBFs) are the promising approach to improve the quality of pyrolysis oil by removing char particles from fast pyrolysis vapor streams. To minimize the commercial MGBF system, the inner MGBF structures were modified, resulting in differential granular media velocities between inlet and outlet gas stream. The modified structure was applied to increase the removal efficiency under the same superficial gas velocity. The experiments were conducted at room temperature and the designed parameters included the insert length and angle of inserted plates in the bottom of MGBF. In this study, the insert plates were used to distinguish the MGBF into two particle flow streams in the same section. The section area was the function of particle response angle, insert length, insert angle, and angle of the MGBF boundary plate. An accurate equation for determining the mass flow rate of the bed material particle and section area was proposed in this study. In addition, cold model filter tests were conducted on differential designs for three sets of flow fields. These results show that differential flow field designs can effectively enhance filtration efficiency. Under the same filtration pressure decline or loss, the filtration efficiency can be increased from the baseline of 95.4% to 99.9%.