Article

Temperatures of Wood Particles in a Hot Sand Bed Fluidized by Nitrogen

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

The thermal history undergone by cylindrical beech wood particles, injected in a sand bed fluidized by nitrogen has been recorded. Experiments have been carried out by varying particle diameter (d = 2−10 mm) and bed temperature (Tr = 712−1107 K). The rate of volatile release becomes significant for temperatures above 625−650 K and is always completed for temperatures below 675−825 K. Devolatilization causes a strong reduction in the heating rate which, at the particle center and for the most intense reaction activity, is comprised between 0 and 25 K/s. For the conditions typical of fast pyrolysis (bed temperatures of 800−1100 K and particle sizes of 2−6 mm), the yields of char are 10−18% and the devolatilization times (corresponding to a conversion of 95%) 18−45s. Furthermore, in qualitative agreement with previous analyses carried out for coal particles, these are well predicted by an empirical power-law relation:  tv = 0.8e1525/Trd1.2 s, over the entire range of experimental conditions examined.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Besides the previous works, further experimental studies investigating biomass pyrolysis in fluidized beds are present in literature. Di Blasi and Branca [12] measured the temperature at the center of solid hardwood particles with a fixed thermocouple which is located in a fluidized bed. The influence of particle size and temperature were investigated. ...
... Therefore, four different heat transfer models by Prins [18] , Baskakov-Palchonok [4] , Agarwal [8] and Chao [19] were implemented and validated with experimental data ( Section 2 ). The adapted single particle model was compared to experimental data published by Di Blasi and Branca [12] , Reschmeier et al. [13] and Marato-Godino et al. [14] . Experimental data of wood pellet pyrolysis in a fluidized bed while varying the fluidization velocities can hardly be found in literature. ...
... For simulations using beech particles, the thermal conductivity of beech λ biomass = 0 . 14 W / ( m * K ) given by Di Blasi and Branca [12] was employed. The thermal conductivity for cardoon pellets was not found in literature, thus, a representative value of λ biomass = 0 . ...
Article
Full-text available
Four different models for heat transfer to the particles immersed in a fluidized bed were evaluated and implemented into an existing single particle model. Pyrolysis experiments have been conducted using a fluidized bed installed on a balance at different temperatures and fluidization velocities using softwood pellets. Using a heat transfer model applicable for fluidized beds, the single particle model was able to predict the experimental results of mass loss obtained in this study as well as experimental data from literature with a reasonable accuracy. A good agreement between experimental and modeling results was found for different reactor temperatures and configurations as well as different biomass types, particle sizes – in the typical range of pellets - and fluidization velocities when they were higher than U/Umf=1.5. However, significant deviations were found for fluidization velocities close to minimum fluidization. Heat transfer models which consider the influence of fluidization velocity show a better agreement in this case although differences are still present.
... This is done by measuring particle temperatures, and use these to evaluate heating rates, reaction temperatures, and devolatilization times. In a previous study [16] we found that FLiNaK gives significantly higher heating rates compared with fluidized sand bed [17] for cylindrical beech wood particles with d ≤ 4 mm at 500 °C. In the present work, we evaluate the effect of different salt mixtures (FLiNaK, (LiNaK) 2 CO 3 , ZnCl 2 -KCl, KNO 3 -NaNO 3 ) over a wider temperature range (400 -600 °C). ...
... [16] The characteristic points were originally proposed by Di Blasi and Branca in their study of pyrolysis of cylindrical beech wood particles (L = 20 mm, d = 2 -10 mm) in a hot sand bed (T = 534 °C) fluidized by nitrogen. [17] Maximum heating rate. This is measured right before any reactions occur. ...
... They are slightly, but not significantly, lower for ZnCl 2 -KCl. The values are comparable to a corresponding study in fluidized sand bed by Di Blasi and Branca [17], indicating that the heat transfer medium is of less importance to the reaction temperatures. For T 2 and T 3 there is only a weak dependence on reactor temperature, with values in the range 352 -386 and 404 -438 °C, respectively. ...
Article
Full-text available
The thermal behavior of wood particles in molten salt pyrolysis was investigated. Cylindrical beech wood particles (L = 30 mm, d = 3.5 mm) were pyrolyzed using different mixtures of molten salts (FLiNaK, (LiNaK)2CO3, ZnCl2–KCl, KNO3–NaNO3) over a temperature range of 400–600 °C. The temperature at the particle center was measured during the process, and used to evaluate heating rates, reaction temperatures and devolatilization times. A general observation was that beech wood is heated faster in fluoride and carbonate melts, but the differences diminish with increasing reactor temperatures. The highest heating rates at the particle center were observed in FLiNaK (46 – 56 °C/s). The effective pyrolysis temperature at which the main decomposition of cellulose and hemicellulose takes place showed a weak dependence on reactor temperature, but no significant difference between the heating media was discovered. The devolatilization time corresponding to conversion of 95% may be empirically correlated with the power law expression . Arrhenius plots were constructed to show the exponential dependence of temperature on the parameter A. The correspondingly low activation energies (13.3 – 27.4 kJ/mol) indicate heat transfer control during the decomposition process.
... Since the calculation of X fuel,G,v depends on the overall pyrolysis time τ v of the injected wood particles, τ v needs to be determined. In order to calculate τ v , a rather simple power law function as given in the literature is embedded in the modelling [34,41]: ...
... The values for n range from 1.5 to 1.7. Within this study, the exponent n is set to 1.5, as calculated from devolatilisation times for pre-dried wood particles (w H2O < 10 wt.%) and which appears to sufficiently agree with measured devolatilisation times from Jand and Foscolo as well as from Di Blasi [38,41]. ...
... The amount of resultant char can be calculated from these numbers, yielding a constant. However, it is known from literature that the resultant char after devolatilisation is a function of the particle geometry (size, shape) and the pyrolysis temperature [32,38,41,43,46,47,48]. For instance, Jand and Foscolo determined the amount of formed char employing spherical wood particles, while Di Blasi as well as Wang et al. used cylindrical biomass particles in their experiments [38,41,48]. ...
... Di Blasi and Branca [6] have measured the transient centre temperatures in long cylindrical wood particles (l=20 mm, d=2 to 10 mm) devolatilizing in a bed of sand fluidized with nitrogen and maintained at a bed temperature of 800-1100 K. They have determined the devolatilization time from the measured temperature and proposed a correlation for devolatilization time as a function of cylinder diameter and bed temperature. ...
... When that temperature is attained by the centre, the conversion is said to be complete. The model could successfully predict the experimentally measured centre temperatures of Di Blasi and Branca [6]. Also, the contours of isotherms, wood density, moisture content in the wood particle have been predicted. ...
... The experimental results have been predicted to an accuracy of 20%. The model too could predict the results of Di Blasi and Branca [6]. The study also showed that the external heat transfer coefficient has no significant influence on the devolatilization time beyond a value of 300 W/m 2 .K. Also, particle shrinkage did not influence the devolatilization time but has influenced the char yield. ...
Article
Full-text available
This paper presents a detailed review of experimental and modeling work carried out on devolatilization of different kinds of wood under fluidized bed combustion/pyrolysis conditions. Laboratory scale experimental studies as well as analytical, phenomenological and numerical modeling works have been reviewed and presented. It has been found that attempts to determine the kinetics in actual fluidized bed conditions have been carried out. There is no single devolatilization model that incorporates all the physical and chemical phenomena occurring during devolatilization. Moreover, the physical phenomenon of primary fragmentation has not been adequately incorporated in the models. Also, non-intrusive temperature measurement techniques need to be developed and demonstrated.
... La densité de flux maximale au point focal est de 670 kW/m 2 . Cette densité de flux est en accord avec celle rencontrée dans les lits fluidisés [Argawal, 1991] [Dyrness et al, 1992][Di Blasi et Branca, 2002] [Oka, 2004]. Les valeurs mesurées sont utilisées lors de la modélisation de la pyrolyse de la biomasse au four à image dans les conditions de densités de flux constantes ou modulées au cours du temps. ...
... En général, la coefficient global de transfert de chaleur est compris entre 200 et 1000 W/(m 2 K) dans les lits fluidisés [Agarwal, 1991] [Dyrness et al, 1992][Di Blasi et Branca, 2002] [Oka, 2004]. Le modèle permet de calculer l'évolution théorique de température de surface d'une particule de charbon dans le gazéifieur DFB du projet Gaya (température du lit T r = 850°C). ...
Thesis
Ce travail s'inscrit dans le projet français de gazéification de la biomasse : le projet Gaya. C'est un vaste programme R&D partenarial coordonné par GDF SUEZ et soutenu par l'ADEME. L'objectif du projet Gaya est de développer une filière décentralisée de production de bio-méthane à partir de la gazéification de la biomasse selon un procédé thermochimique de deuxième génération. L'objectif de cette thèse est de réaliser un modèle de pyrolyse de biomasse représentatif des conditions du lit fluidisé de gazéification développé dans ce projet. Un pilote expérimental, le four à image, a été développé pour reproduire au mieux les conditions de chauffage d'un lit fluidisé à 850°C. Ce pilote permet de récupérer l'ensemble des produits de pyrolyse pour une analyse ultérieure. De là, les cinétiques des réactions de pyrolyse sont déterminées par modélisation des processus physico-chimiques et optimisation à partir des résultats expérimentaux. Le craquage des vapeurs primaires de pyrolyse éjectées de la particule de biomasse est étudié durant 300 millisecondes après leur éjection de la particule de biomasse. Ces expériences de craquage sont menées sur le montage expérimental combinant un réacteur tubulaire de pyrolyse et un réacteur parfaitement auto-agité de craquage. Le modèle développé permet de représenter la pyrolyse de la biomasse introduite dans le réacteur de gazéification
... Besides the detailed 1D models, empirical correlations have also been developed to estimate the devolatilization time of biomass particles [7,[21][22][23][24][25]. Most of the existing correlations primarily use a power law to correlate the devolatilization time with particle size with a with relatively low temperatures (< 1000°C) [21][22][23][24][25]. Empirical correlations are rarely reported for biomass devolatilization under pulverized flame conditions. ...
... Besides the detailed 1D models, empirical correlations have also been developed to estimate the devolatilization time of biomass particles [7,[21][22][23][24][25]. Most of the existing correlations primarily use a power law to correlate the devolatilization time with particle size with a with relatively low temperatures (< 1000°C) [21][22][23][24][25]. Empirical correlations are rarely reported for biomass devolatilization under pulverized flame conditions. ...
Article
Wood devolatilization experiments in a single particle combustor and comparison with a 1D devolatilization model were carried out to investigate the effects of wood particle properties and operating conditions on wood particle devolatilization time. The experiments were conducted with 3 mm spherical/cubic and 4 mm spherical particles at gas temperatures of 1200–1450 °C and oxygen contents of 0–4.4 vol%. Both experimental and modelling results showed that the devolatilization time increases linearly with particle density for raw, wetted, and torrefied wood particles. A sensitivity analysis done with the 1D devolatilization model showed that the biomass devolatilization time is sensitive to particle size, moisture content, gas temperature and particle density, and insensitive to volatiles fraction and gas velocity under the investigated experimental conditions. Using the same devolatilization kinetics, the 1D model could predict well the devolatilization time of different wood species with different particle size, density and moisture content. With this in mind, a simple correlation for devolatilization time has been developed based on the simulation data from the 1D model. The correlation uses a four-variable function with inputs of particle size, moisture content, gas temperature and particle density to determine the devolatilization time of biomass. Experimental devolatilization time found in literature could be predicted within ±25% for large particles (1–10 mm) under high temperature conditions (1000–1600 °C).
... The above kinetics do not account for transport limitations (e.g., associated with intra-particle heat and mass transfer); so, we expect that additional corrections to the predicted rates are needed to deal with differences in particle size. The impacts of transport were correlated from experimental data by Di Blasi and Branca (2003) and more recently summarized in detail by Di Blasi (2008). One of the most convenient correlations arising out of this work is an expression for estimating the devolatilization time summarized in equation 8 below. ...
... Other investigators [notably de Diego et al (2003)] included additional terms (shown below) to account for the effects of moisture. According to Di Blasi and Branca (2003), the time required for a particle to reach 95% of complete devolatilization, t v , is given by ...
... Temperature history of the fuel particle while within in a hot fluidized bed determines the devolatilization process taking place in the particle. Different methods are used to measure the fuel particle temperature in a hot fluidized bed, including photographic method [10,11], fusible wire ring [12], optical pyrometry radiation capturing by optical probe [13,14] or the digital camera [15] and thermocouple technique conducted for particles with different shapes including spherical [16][17][18], cylindrical [19,20] and cubic [21]. ...
... One of the few studies on biomass thermochemical conversion in a hot fluidized bed which recorded its temperature history using a thermocouple was done by Di Blasi and Branca [20]. They used beach wood particles in the shape of cylinders with diameters ranging from 2 mm to 10 mm and heights of 20 mm, placing them in a fluidized bed with bed material of calcined sand. ...
Article
Measuring the temperature of a fuel particle during the thermochemical process inside the fluidized bed is of great importance in order to obtain detailed knowledge of the conversion behavior of the fuel particle and so optimize the combustion performance. In this study a number of experiments were done in order to register the temperature and hydrodynamics of a biomass fuel particle in a fluidized bed during the devolatilization process of a biomass particle. The experiments were done in a 2D fluidized bed with a front transparent window in order to use the particle image velocimetry (PIV) method to obtain information on the hydrodynamics of the fuel particle and inert bed material inside the bed. A thermocouple was also used to measure the temperature of the particle during the conversion. Experiments were done at a fluidized bed temperature and fluidization velocity in the range of 350–450 °C and 0.2–0.6 m/s, respectively. The effect of the bed's temperature and fluidization velocity on the drying and devolatilization process of the biomass fuel particle was investigated. The results indicate that the bed's temperature and fluidization velocity have a significant effect on the mass and heat transfer between the fuel particle and the bed during the conversion process.
... The choice of salts were selected based on previously reported heat transfer characteristics for thermal processing of biomass [20]. In a previous study [21], we found that molten FLiNaK gives significantly higher heating rates compared with a fluidized sand bed [22] for beech wood cylinders with d ≤ 4 mm. In a subsequent study [23], both FLiNaK and (LiNaK) 2 CO 3 showed good promise as heat transfer media in fast pyrolysis due to high heating rates for beech wood cylinders (d = 3.5 mm) in the temperature range of 450 to 600 °C. ...
Article
Full-text available
A tubular electrostatic precipitator (ESP) was designed and tested for collection of pyrolysis oil in molten salt pyrolysis of milled beech wood (0.5-2 mm). The voltage-current (V-I) characteristics were studied, showing most stable performance of the ESP when N2 was utilized as inert gas. The pyrolysis experiments were carried out in FLiNaK and (LiNaK)2CO3 over the temperature range of 450-600 ℃. The highest yields of pyrolysis oil were achieved in FLiNaK, with a maximum of 34.2 wt% at 500 ℃, followed by a decrease with increasing reactor temperature. The temperature had nearly no effect on the oil yield for pyrolysis in (LiNaK)2CO3 (19.0-22.5 wt%). Possible hydration reactions and formation of HF gas during FLiNaK pyrolysis were investigated by simulations (HSC Chemistry software) and measurements of the outlet gas (FTIR), but no significant amounts of HF were detected.
... The heating rate of the biomass particles is of utmost importance for the pyrolysis process, impacting process control, product yields, and product quality [110,111,[118][119][120]. ...
Article
Full-text available
The social, economic, and environmental impacts of climate change have been shown to affect poorer populations throughout the world disproportionally, and the COVID-19 pandemic of 2020-2021 has only exacerbated the use of less sustainable energy, fuel, and chemical sources. The period of economic and social recovery following the pandemic presents an unprecedented opportunity to invest in biorefineries based on the pyrolysis of agricultural residues. These produce a plethora of sustainable resources while also contributing to the economic valorization of first-sector local economies. However, biomass-derived pyrolysis liquid is highly oxygenated, which hinders its long-term stability and usability. Catalytic hydrogenation is a proposed upgrading method to reduce this hindrance, while recent studies on the use of nickel and niobium as low-cost catalysts, both abundant in Brazil, reinforce the potential synergy between different economic sectors within the country. This review gathers state-of-the-art applications of these technologies with the intent to guide the scientific community and lawmakers alike on yet another alternative for energy and commodities production within an environmentally sustainable paradigm.
... 촤 생성율은 촤 반응(R-4)의 활성화 에너지 값이 가장 높은 Chan 등의 모델이 11.5% 로 가장 낮게 나왔으며, 나머지는 15% 전후로 생성되었다. 촤의 생 성율은 사용되는 바이오매스의 특성에 따라 차이가 나지만, 일반적 으로 10~18% 범위로 보고되고 있다[41]. ...
Article
Full-text available
Devolatilization is an important mechanism in the gasification and pyrolysis of woody biomass, and has to be accordingly considered in designing a gasifier. In order to describe the devolatilization process of wood particle, there have been proposed a number of empirical correlations based on experimental data. However, the correlations are limited to apply for various reaction conditions due to the complex nature of wood devolatilization. In this study, a simple model was developed for predicting the devolatilization of a wood particle in a fluidized bed reactor. The model considered the drying, shrinkage and heat generation of intra-particle for a spherical biomass. The influence of various parameters such as size, initial moisture content, heat transfer coefficient, kinetic model and temperature, was investigated. The devolatilization time linearly increased with increasing initial moisture content and size of a wood particle, whereas decreases with reaction temperature. There is no significant change of results when the external heat transfer coefficient is over 300 , and smaller particles are more sensitive to the outer heat transfer coefficient. Predicted results from the model show a similar tendency with the experimental data from literatures within a deviation of 10%.
... Figure 24A: Schematic diagram of single wood pyrolysis experimental setup (Courtesy of Prof. Chun-Zhu Li, Curtin University) To study the temperature profile in a sample, it is necessary to make sure that the measuring thermocouples are always kept inside the biomass sample during the pyrolysis. Most of the pyrolysis studies are done using large particles (Bilbao et al., 1993;Chan et al., 1985;Di Blasi & Branca, 2003;Wang et al., 2005) and high temperature or high heat intensity, which results in the breakage of char after a certain time. The thermocouples measuring the temperature seldom reached more than the set pyrolysis temperature. ...
... The question to ask then is if the kinetics determined in a TGA at lower heating rate conditions can be used to model the high heating rate processes in industry. One thing that is important to remember is that even though the heating rate is high on the surface of the particle the in-ternal heating rate is not necessarily high [94,95], which suggests that, at least at lower process temperatures, TGA may give relevant information. From a scientific perspective, low heating rates are advantageous since they will decouple overlapping mass-loss rate peaks which enables separate analysis of the sub-processes. ...
... The products of the gasification of MSW are ash, oils, and gases, which are mainly carbon monoxide, hydrogen, carbon dioxide, and hydrocarbons [9]. Many researchers have investigated this process to evaluate the influences of operating parameters (i.e., temperature, steam-to-MSW ratio (STMR), residence time, feedstock particle size, addition of catalyst, etc.), types of feedstock, and gasifying agents on the gasification performance [12][13][14][15][16][17][18][19][20]. In order to develop an efficient and economic MSW gasification process, it is necessary to understand how these factors influence the gasification reactions, which can provide valuable information for the better design of the MSW gasification process. ...
Article
Full-text available
This work aims to understand the gasification performance of municipal solid waste (MSW) by means of thermodynamic analysis. Thermodynamic analysis is based on the assumption that the gasification reactions take place at the thermodynamic equilibrium condition, without regard to the reactor and process characteristics. First, model components of MSW including food, green wastes, paper, textiles, rubber, chlorine-free plastic, and polyvinyl chloride were chosen as the feedstock of a steam gasification process, with the steam temperature ranging from 973K to 2273K and the steam-to-MSW ratio (STMR) ranging from 1 to 5. It was found that the effect of the STMR on the gasification performance was almost the same as that of the steam temperature. All the differences among the seven types of MSW were caused by the variation of their compositions. Next, the gasification of actual MSW was analyzed using this thermodynamic equilibrium model. It was possible to count the inorganic components of actual MSW as silicon dioxide or aluminum oxide for the purpose of simplification, due to the fact that the inorganic components mainly affected the reactor temperature. A detailed comparison was made of the composition of the gaseous products obtained using steam, hydrogen, and air gasifying agents to provide basic knowledge regarding the appropriate choice of gasifying agent in MSW treatment upon demand.
... However, this is in a bit contrary to the observation made by Borah et al. [44], where the total volatile yield is invariable at a temperature above 727 C in conventional air environment. Also, Di Blasi and Branca [45] states that the effect of bed temperature on char yield becomes insignificant above 577 C when devolatilised under nitrogen environment. The low char yield results in the present study can be corroborated by the char yields obtained by Xu et al. [46] in oxy-steam environment. ...
Article
Chemical Looping Combustion (CLC) is one of the emerging technologies for carbon capture, with less energy penalty. The present way of using pulverized coals in a fluidized bed (FB)-CLC have limitations like loss of unconverted char and gaseous combustibles, which could be mitigated by use of coarser fuel particles. Devolatilization time is a critical input for the effective design of FB-CLC systems, primarily when large fuel particles are used. The present study investigates the devolatilization time and the char yield of three coals of two shapes, namely, two high ash Indian coals and a low ash Indonesian coal and a wood (Casuarina equisetifolia) in the size range of +8–25 mm, at different fuel reactor temperatures (800–950 °C) of a hematite based CLC unit. The devolatilization times of single fuel particles during CLC are determined using a visual method called ‘Color Indistinction Method’. Indonesian coal has the longest devolatilization time among the fuels, and biomass has the least. Increasing the bed temperature enhances the rate of volatile release, whereas this effect is less pronounced in larger particles. Devolatilization of Indonesian coal is found to be strongly influenced by the changes in operating conditions. With the decrease in sphericity, a maximum of 56% reduction in devolatilization time is observed for the +20–25 mm slender particles of Indonesian coals when compared to the near-round particles. The maximum average char yields at the end of the devolatilization phase for coal and biomass are about 55–76% and 16% respectively. Char yield in coal particles increases with an increase in particle size, whereas biomass particles show relatively consistent yield across all experimental conditions. Increase in bed temperature reduces the char yields of coal up to 12% and in biomass up to 30%. High volatile Indian coal is the most influenced fuel by the changes in fuels shape. A correlation for determining devolatilization time under CLC environment is presented, and it successfully fits most of the experimental values within ±20% deviation for coals (R² = 0.95) and within ±15% deviation for biomass (R² = 0.97).
... Maximum heating rates measured in the cylindrical wood particle (30 mm length and 3.5 mm diameter) were observed in FLiNaK salt (46-56°C/s). Maximum heating rate in wood particles of different diameters [20] was compared to heating rates measured in a fluidized bed reactor [21]. The authors observed that for wood cylindrical particles with a diameter smaller than 4 mm, the maximum heating rate in molten salt pyrolysis was significantly higher than that in a fluidized bed. ...
Article
Full-text available
The use of a binary mixture of solar molten salts (60 wt% NaNO3 and 40 wt% KNO3) as a heat transfer medium for the production of a solar fuel by the thermochemical conversion of biomass is investigated in the present paper. Thermochemical conversion can be a route for converting the surplus solar irradiation via the direct contact of nitrate molten salts and biomass into storable chemical fuel. Traditional fixed-bed pyrolysis and molten salts pyrolysis have been carried out under an inert atmosphere at a temperature of 500 °C. The composition of the permanent gases and the bio-oil produced has been analyzed along with the temperature profiles inside the reactor. Two distinctive pathways have been observed: an endothermic process in the case of traditional fixed-bed pyrolysis and an exothermic process in the case of molten salt pyrolysis. An attempt has been made to identify possible causes for such differences.
... Similarly, another study [95] also reported an improvement of 28.3% in charcoal production at the temperature of 900°C, by varying the particle size of hazelnut from 0.15 to 1.4 mm. Di Blasi et al. [102] observed that increasing particle size of biomass results in high rate of char production. During pyrolysis of beech wood in a fluidized bed reactor at the temperature of 807 K, an increase of 5 wt% was noticed while increasing particle size from 2 to 10 mm. ...
Article
Full-text available
Oil palm industry generates different types of waste biomass in the form of oil palm empty fruit bunch (OPEFB), oil palm mesocarp fibers (OPMF), oil palm fronds (OPF), oil palm trunks (OPT), oil palm bark (OPB), oil palm leaves (OPL), and oil palm shell (OPS). These biomass wastes possess a great energy potential to be converted into biofuels, particularly bio-oil. Among all, the OPS have favorable physicochemical characteristics to be converted into bio-oil. Therefore, this paper mainly focuses to review the suitability of OPS as feedstock for bio-oil production compared to other oil palm biomasses. The physicochemical characteristics of the OPS, in terms of heating value, ultimate analysis, proximate analysis, and lignocellulosic composition, are presented and compared to those of the OPEFB, OPMF, OPF, OPT, OPB, and OPL. To illustrate further and signify the stability, the abovementioned properties of OPS bio-oil are also reviewed and compared to those of bio-oils produced from OPEFB, OPF, OPB, OPL, and petroleum fuels. The challenges and future prospects of OPS as a source of bio-oil are addressed and compared with other wastes of oil palm industry. Additionally, methods used for bio-oil production from oil palm industry biomass are discussed and illustrated in detail.
... 33 In a fluidized bed using steam at 650°C, a temperature at which devolatilization predominates over steam gasification, they realized 0.0 wt % charcoal yield with particles having a mean diameter of 0.287 mm that increased to 35 wt % with particles having a mean diameter of 1.09 mm. Likewise, Di Blasi et al. 64 reported a clear trend of increasing char yield with increasing wood particle size. When pyrolyzed in a fluidized bed reactor at 807 K, the char yield of beech wood increased by about 5 wt % as the particle size increased from 2 to 10 mm. ...
Article
The prosperity of Silicon Valley is built upon a foundation of wood charcoal that is the preferred reductant for the manufacture of pure silicon from quartz. Because ordinary pyrolysis processes offer low yields of charcoal from wood, the production of silicon makes heavy demands on the forest resource. The goal of this paper is to identify process conditions that improve the yield of charcoal from wood. To realize this goal, we first calculate the theoretical fixed-carbon yield of charcoal by use of the elemental composition of the wood feedstock. Next, we examine the influence of particle size, sample size, and pressure on experimental values of the fixed-carbon yields of the charcoal products and compare these values with the calculated theoretical limiting values. The carbonization by thermogravimetric analysis of small samples of small particles of wood in open crucibles delivers the lowest fixed-carbon yields, closely followed by standard proximate analysis procedures that employ a closed crucible and realize somewhat improved yields. The fixed-carbon yields (as determined by thermogravimetry) improve as the sample size increases and as the particle size increases. Further gains are realized when pyrolysis occurs in a closed crucible that hinders the egress of volatiles. At atmospheric pressure, high fixed-carbon yields are obtained from 30 mm wood cubes heated in a closed retort under nitrogen within a muffle furnace. The highest fixed-carbon yields are realized at elevated pressure by the flash carbonization process. Even at elevated pressure, gains are realized when large particles are carbonized. These findings reveal the key role that secondary reactions, involving the interaction of vapor-phase pyrolysis species with the solid substrate, play in the formation of charcoal. Models of biomass pyrolysis, which do not account for the impacts of sample size, particle size, and pressure on the interactions of volatiles with the solid substrate, cannot predict the yield of charcoal from biomass. These findings also offer important practical guidance to industry. Size reduction of wood feedstocks is not only energy and capital intensive; size reduction also reduces the yield of charcoal and exacerbates demands made on the forest resource.
Article
The pyrolytic behavior of a cylindrical packed bed of beech wood pellets, uniformly heated along the lateral surface, is studied. As the heating conditions are varied, the trends shown by the mass loss rate and the yields of the lumped product classes are in agreement with previous literature. Moreover, detailed measurements of the thermal field permit, for the first time, the observation at the central core of the bed, of sequential (a) exothermic, (b) endothermic, and (c) exothermic effects. The examination of the corresponding temperature ranges indicates that they are linked, in the order, to the (primary and secondary) degradation of (a) hemicellulose, (b) cellulose and lignin, and (c) lignin. The dependence on the heating conditions is also explained of the center temperature overshoot (with respect to the steady value) which attains maximum values around 90 K.
Article
The image furnace is used to mimic experimentally heating conditions representative of a fluidized bed biomass gasifier. The aim of this paper is to demonstrate the feasibility of heat flux density modulation with a new improved vertical image furnace which is based on the former horizontal setup. This is achieved through the comparison of the experimental and theoretical surfaces temperatures evolutions. A numerical model is solved in order to estimate the theoretical surface temperature evolution of a char particle immersed in a fluidized bed at 1123 K. The goal of the experiments performed with char samples is to reproduce the same temperature evolutions. The surface temperatures profiles are measured by pyrometry in different conditions (constant heat flux density, variation of the heat flux density over time) and compared to the modelling results. The experiments at constant heat flux density show poor agreement with the model, in comparison with experiments made with variable heat flux density. The best agreement is obtained when the heat flux density is decreased from 560 kW m−2 to 250 kW m−2 in 1 s after a period of 500 ms.
Article
Steam gasification is considered one of the most effective and efficient techniques of generating hydrogen from biomass. Of all the thermochemical processes, steam gasification offers the highest stoichiometric yield of hydrogen. There are several factors which influence the yield of hydrogen in steam gasification. Some of the prominent factors are: biomass type, biomass feed particle size, reaction temperature, steam to biomass ratio, addition of catalyst, sorbent to biomass ratio. This review article focuses on the hydrogen production from biomass via steam gasification and the influence of process parameters on hydrogen yield.
Article
Pyrolysis is carried out using beech wood for packed beds, consisting of pellets and particles of various sizes, and a block in the same shape and size as the beds. The different physical properties of the samples affect the heat- and mass-transfer rates and, consequently, the conversion time. However, the effects on the yields and composition of the lumped product classes are generally negligible. An exception is observed for the packed beds made of pulverized material, where re-evaporation and pyrolysis of the vapors, previously migrated and condensed in cold zones, result in reduced yields of some organic products (hydroxyacetaldehyde, levoglucosan, syringols) and an increased carbon content of the char. These effects are enhanced by successively smaller particle sizes and more severe heating conditions.
Article
The thermal decomposition of a cylindrical fixed bed consisting of agricultural residues (hazelnut shells, olive pomace, straw pellets) or softwood pellets, uniformly heated along the lateral surface, is investigated for heating temperatures in the range 473–800 K, and a comparison is made with results previously obtained for beech wood pellets. Although dependent on the external heating conditions, exothermic reaction heat effects are evident for all the biomasses, giving rise to maximum temperature overshoots of 225 K (hazelnut shells), 170 K (olive pomace), 78 K (straw), and 53 K (softwood pellets) (versus 86 K for beech wood pellets). For the first two materials and mild/moderate heating conditions, the entire bed volume experiences large temperature overshoots, so that the qualitative features of the pyrolytic conversion are those of a thermal runaway. Explanations are given, based on the different chemical and physical properties of the samples, for the different exothermicity level, and its implications in the practical application of torrefaction and pyrolysis are discussed.
Article
Hydrogen, the inevitable fuel of the future, can be generated from biomass through promising thermochemical methods. Modern-day thermochemical methods of hydrogen generation include fast pyrolysis followed by steam reforming of bio-oil, supercritical water gasification and steam gasification. Apart from the aforementioned methods, a novice technique of employing combined slow pyrolysis and steam gasification can be also engaged to produce hydrogen of improved yield and quality. This review paper discusses in detail about the existing hydrogen generation through thermochemical methods. It elaborates the merits and demerits of each method and gives insight about the combined slow pyrolysis and steam gasification process for hydrogen generation. The paper also elaborates about the various parameters affecting integrated slow pyrolysis and steam gasification process. Copyright © 2014 John Wiley & Sons, Ltd.
Article
The aim of the present study is to develop a 1D modelling tool representing biomass gasification in a dual fluidized bed. This tool should be used with the objective to control and scale-up the process for different pilot scales and operating conditions. This implies that the modelling tool should be based on a phenomenological description but also that the calculations are not too much time-consuming. A one-dimensional steady state model was thus developed for the gasification reactor, combining a description of the biomass thermochemical conversion phenomena and of the hydrodynamic behaviour of the fluidized bed. The reactor is divided into two zones, namely: dense zone and freeboard zone. For the dense zone, the model assumes the existence of two phases –bubble and emulsion– with chemical reactions occurring in both phases (biomass devolatilisation, char gasification, water-gas shift). In the freeboard, a plug flow is assumed and the only reaction considered is water-gas shift. Mass balance is globally solved across the entire gasifier. The model allows calculating axial concentrations of the gas species in the reactor as well as the global yields. The parameters of the calculations are temperature of the gasifier, biomass composition, biomass and steam feeding rates. This article is protected by copyright. All rights reserved
Article
The pyrolysis of particles of glucomannan, the main component of softwood hemicelluloses, is investigated in a fluidized-bed reactor. A first set of experiments is carried out for temperatures in the range 530–690 K to determine yields and composition of products. The most abundant are char, water, and carbon dioxide. The condensable organic fraction mainly consists of acetic acid, formic acid, hydroxypropanone, hydroxyacetaldehyde, and furfuryl alcohol. The second set of experiments is made to determine the weight loss characteristics of small samples exposed in the expanded bed at temperatures of 503–593 K that are then used to develop one- or two-stage pyrolysis mechanisms.
Article
The influence of particle geometry and microstructure in fast pyrolysis of beech wood has been investigated. Milled wood particles (<0.08–2.4 mm) and natural wood cylinders (2–14 mm) with different lengths (10–50 mm) and artificial wood cylinders (Dp = 0.5–14 mm) made of steel walls, filled with small milled wood particles (<0.08–0.140 mm), have been pyrolyzed in a fluidized bed at 500 °C. From the results of the experiments, the influence of particle geometry and microstructure on char, gas, and pyrolysis oil yield and pyrolysis oil composition has been derived. The product yields of large cylinders with diameters of 6–14 mm are primarily determined by the outer diameter and resulting heating rate. The microstructure of these cylinders, being either natural channels or randomly packed small milled wood particles, has turned out to be much less important. For the smaller milled wood particles, the microstructure does have a profound effect on the product yields. The smallest particles (<0.140 mm), which consist only out of cell wall material and have lost their typical wood channel structure, show a clearly higher oil yield and lower char yield. It is postulated that the high pyrolysis oil yield can be explained by larger mass transfer rates of pyrolysis products from these smallest particles, as compared to mass transfer from particles containing channels.
Article
Cement production is highly energy intensive and requires large quantities of fuels. For both economical and environmental reasons, there is an increasing tendency for utilization of alternative fuels in the cement industry, examples being tire derived fuels, waste wood, or different types of industrial waste. In this study, devolatilization and combustion of large particles of tire rubber and pine wood with equivalent diameters of 10 mm to 26 mm are investigated in a pilot scale rotary kiln able to simulate the process conditions present in the material inlet end of cement rotary kilns. Investigated temperatures varied from 700 to 1000 °C, and oxygen concentrations varied from 5% v/v O2 to 21% v/v O2. The devolatilization time of tire rubber and pine wood were found to mainly depend on temperature and particle size and were within 40 to 170 s. Rate limiting parameters for char oxidation of tire rubber and pine wood were found to be bulk oxygen concentration, mass transfer rate of oxygen, raw material fill degree, raw material characteristics, and temperature. Kiln rotational speed only had a minor effect on the char oxidation when the raw material bed was in a rolling motion. Initial fuel particle size also influenced the char conversion time for pine wood char but had no influence on tire char conversion time, because the tire rubber crackled into several smaller char fragments immediately after devolatilization. The char conversion times were from 40 to 480 s for tire char and from 30 to 1300 s for pine wood char, depending on the conditions. Models for devolatilization and char oxidation of tire rubber and pine wood are validated against experimental results.
Article
Molten salt pyrolysis is a thermochemical conversion process in which biomass is fed into and heated up by a molten salt bath. Molten salts have very high thermal stability, good heat transfer characteristics, and a catalytic effect in cracking and liquefaction of large molecules found in biomass. In this study, the heat transfer characteristics of molten salts are studied by recording the thermal history of wood particles in molten salt pyrolysis. Experiments have been carried out with cylindrical beech and pine wood particles with constant length (L = 30 mm) and varying diameter (d = 1–8 mm) in a FLiNaK melt with a temperature of 500 °C. The thermal history at the particle center has been used to evaluate the reaction temperatures, the heating rates, and the devolatilization times. Results have been compared with a similar study in a fluidized sand bed. It is found that FLiNaK gives significantly higher heating rates for cylinders with d ≤ 4 mm. For larger cylinders, the process is dominated by heat transfer within the wood particle, and the heat transfer medium is of less importance. For the smallest cylinders (d = 1 mm), heating rates as high as 218 ± 6 and 186 ± 15 °C/s were observed for beech and pine wood, respectively. The average heating rate for wood cylinders until the main degradation takes place has been found to follow the empirical correlation β = (keff/ρ)103(24 + 390e–0.49d), and the total devolatilization time has been found to follow the empirical correlation tdev = ρ(0.146 e–keff – 1.09)d1.05.
Article
Share link ( https://authors.elsevier.com/a/1f9lP4x7R2cZbB - free access till 19 July 2022) Thermochemical conversion of larger biomass particles (thermally thick regime) toward high-end products still suffers from an unrevealed quantitative relationship between process and product parameters. The main issue relates to the influence of heating rate within the particle, critical conversion-wise but difficult to assess experimentally. Computational fluid dynamics (CFD) modelling may help, but first the model must prove its reliability to prevent error transfer to the results. This study aimed to provide an unbiased, state-of-the-art model constructed in a stepwise mode to investigate the heating rate’s distribution. Several datasets with broadly varying parameters from the literature were used for the development and validation since the reproduction of datasets would not bring novelty to solving the problem. Instead of the model's calibration to fit to the data, the parameters for each step-model were meticulously selected to match the experimental conditions. The stepwise development showed the best accuracy when the anisotropy and the heat sink drying sub-model were implemented. Moreover, using the Ranzi-Anca-Couce (RAC) scheme led to more accurate results than the Ranzi scheme. The comprehensive model was positively validated against a broad range of production parameters (pyrolysis temperature: 500°C - 840 °C, diameter of particles: 10 mm - 20 mm, shapes: cylinders and spheres). Investigation showed a pattern in volatiles release profiles and homogeneous heating rate distribution when particle size is below 4 mm. Despite basing the models on the literature’s data, the study includes novel and valuable insights for biomass conversion and constitutes a solid foundation for future development.
Article
Full-text available
Biomass energy is one of the most widely explored research fields in energy and environmental science. The major driver for biomass gasification research is to exploit low-cost feedstocks, to increase process efficiency, decrease installation and operational costs and socio-environmental effects. This work gives a holistic view of current research, development and deployment, and how we could move forward towards technologically, economically and socially acceptable biomass gasification technologies. We elucidate various areas and comparatively discuss various conventional gasification technologies, current developments, and challenges to boost gasification as a viable and environmentally sustainable technology.
Article
This study examines the effects of various pretreatments (size reduction, pelletization, hot water washing, torrefaction) on the pyrolysis of packed beds of hazelnut shells, exposed to moderate heating along the lateral surface, aimed at identifying the controlling or pre-dominant mechanisms for the high reaction exothermicity leading to thermal runaway. As long as the particles preserve their chief structural properties (roughly or finely crushed shells), the process dynamics are not altered by size variation. Instead exothermic effects are highly reduced or almost disappear for milled shells, even in the pelletized form, depending on the actual powder size. Hence secondary intra-particle reactions, owing to a peculiar scarcely porous microstructure, play a paramount role in the process exothermicity and thermal runaway. The displacement of the beginning of the reaction process at higher temperature, following sample washing, is enhanced by deepening the pretreatment. In this way sample conversion takes place in the presence of large spatial gradients, which hinder the occurrence of thermal runaway. Finally torrefaction, by partial or complete conversion of the more thermally labile chemical components, also eliminates the conditions which trigger and partly sustain the thermal runaway.
Thesis
Full-text available
Les procédés de pyrolyse et gazéification permettent de convertir la biomasse en différents vecteurs énergétiques (carburants, électricité, chaleur). Ces procédés thermochimiques pourraient ainsi contribuer à la diminution des émissions de GES, mais leur mise en œuvre à l’échelle industrielle reste lente. En plus des difficultés réglementaires et économiques, il existe un manque de prévisibilité de leurs performances technique, énergétique et environnementale. La simulation fine des réacteurs thermochimiques pourrait ainsi contribuer à l’émergence de ces filières bio-sourcées, mais elle se heurte à la variabilité de la biomasse ligno-cellulosique et à la complexité des phénomènes rencontrés. Ce mémoire expose les travaux entrepris par l’auteur sur cette problématique multi-échelle et se focalise en particulier sur la modélisation de la pyrolyse à l’échelle de la particule, qui est le phénomène commun à tous les réacteurs thermochimiques. Les modèles développés sont confrontés à des résultats expérimentaux obtenus sur des bancs d’essais originaux (four à image, micro-lit fluidisé, lit fluidisé 5kg/h).
Article
In this paper, the effect of CO2 gasification on the oxy-fuel combustion of Zhundong coal char was studied by reactive molecular dynamic (ReaxFF-MD) method combined with experiments. The structural representation of Zhundong coal char was constructed based on experimental analysis. ReaxFF simulations of char combustion in O2/N2 and O2/CO2 environments at various conditions were carried out to investigate the effect of CO2 on the conversion of char. The flat-flame entrained flow experiments were also performed to compare with the ReaxFF MD simulations. The results showed that CO2 reduced the diffusion rate of O2 and thus inhibited the oxidation rate of char. However, the total carbon consumption of char was found to be increased by the CO2 gasification reaction, especially at low O2 concentration at high temperatures. The contributions of CO2 gasification to the total carbon consumption of char at 3500 K were 51.38%, 34.74% and 19.88% in 5%, 10% and 20% O2 concentrations, respectively. The activation energy of CO2 gasification was determined as 250 kJ/mol in the temperature range of 3000–3500 K. Finally, the detailed and dynamical description of CO2 gasification reaction pathways were revealed in atomic scale. CO2 molecule was initially adsorbed on the active carbon site, and then the CO bond broken resulting in the formation of CO.
Article
Cubes and spheres of spruce wood have been prepared, with a fine thermocouple inserted to measure the temperature at their centre. Individual particles were immersed rapidly in a bed of sand (mean size ∼0.2 mm), which was fluidised by nitrogen and held at a fixed temperature up to 700°C. The rising temperature measured at a particle's centre yielded the effective value of the particle's thermal diffusivity. The temperature response showed evidence of at least two endothermic decomposition reactions, which corresponded to the pyrolysis of fine particles of the wood in a thermogravimetric analyser (TGA). However, the wood undergoing thermal decomposition in a fluidised bed at 500°C revealed at least one exothermic step at the very end of heating. After being heated this way in a hot bed fluidised by nitrogen, the particles of char formed by spruce wood had very much the same size and shape as the original piece of wood before being heated. These new particles of “char”, formed whilst being heated in a hot fluidised bed, were cooled in a stream of nitrogen and returned to the fluidised bed for re-heating without any complications from pyrolysis. The rise in the char's central temperature with time gave an unambiguous value for the thermal diffusivity of the char. It is clear that volatile matter leaving a particle of wood reduced the rate of heat transfer between a hot fluidised bed and the centre of a devolatilising particle. Also, the time for complete pyrolysis was proportional to the square of the characteristic size (r0) of the spruce being heated. In addition, the time for pyrolysis was proportional to Tbed3±1, so that for a cube of spruce tpyr = 2.9 ± 0.3 × 10¹⁵r02/Tbed3 in seconds. Photographic evidence confirmed that devolatilisation of particles of spruce larger than ≈ 2 mm in a fluidised bed follows a shrinking core model and is accordingly controlled by internal heat transfer.
Article
Chemical Looping Combustion (CLC) is one of the promising fuel conversion technologies for carbon capture with low energy penalty. Devolatilization is an important physical phenomenon occurring during solid fuel CLC. Devolatilization behaviour influences fragmentation, combustion rate, emission, and particulates generation in fluidized bed CLC (FB-CLC), thus a critical input for its design. Existing visual techniques for determining devolatilization time cannot be applied in CLC conditions because of its flameless combustion nature. In the present study, a new, simple and quick technique termed as ‘Colour Indistinction Method’ (CIM) is proposed for the determination of devolatilization time (τd) in FB-CLC, where the end of devolatilization is inferred from the disappearance of fuel particle in a hot fluidised bed. Single particle devolatilization studies in FB-CLC are conducted to determine the devolatilization time using CIM for two types of fuels viz. coal and biomass (Casuarina equisetifolia wood), of size range 8-25 mm and 10-20 mm respectively, at three different fuel reactor bed temperatures (800, 875 and 950 ⁰C), and at one fluidization velocity. The proposed technique is validated in three ways namely, (i) the measurement of residual volatiles present in char by Thermo-Gravimetric Analysis (TGA), (ii) mass loss history of the fuel during devolatilization and (iii) diagnostics using particle centre temperature measurements. The results of CIM experiments, in terms of degree of error involved, are compared with an established Flame Extinction Technique (FET) and a more accurate particle centre temperature method. The amount of volatiles released during devolatilization, as determined by CIM is 91.3 % for coals and of 98.9 % for biomass. These values compare very well with the results of the established FET, in which the volatile release is 90.7% for coal and 99.1 % for biomass samples. The devolatilization times determined using CIM is in line with particle centre temperature measurements with an acceptable error range of -7.57 to +3.70 %. The proposed CIM is successful in establishing the devolatilization time of different fuels in CLC conditions and can also be applied in other flameless combustion conditions.
Article
Gasification of coconut shell (CS) and palm kernel shell (PKS) is conducted in a batch type downdraft fixed-bed reactor to evaluate the effect of particle size (1e3 mm, 4e7 mm, and 8e11 mm) and temperature (700, 800, and 900 C) on gas composition and gasification performance. The response surface methodology integrated variance-optimal design is used to identify the optimum condition for gasification. Gas composition, which is measured using the biomass particle size of 1e11 mm at 700e900 C, are 8.20e14.6 vol% (H 2), 13.0e17.4 vol% (CO), 14.7e16.7 vol% (CO 2), and 2.82e4.23 vol% (CH 4) for CS and 7.01e13.3 vol% (H 2), 13.3e17.8 vol% (CO), 14.9e17.1 vol% (CO 2), and 2.39e3.90 vol% (CH 4) for PKS. At similar conditions, the syngas higher heating value, dry gas yield, carbon conversion efficiency, and cold gas efficiency are 4.01e5.39 MJ/Nm 3 , 1.50e1.95 Nm 3 /kg, 52.2e75.9%, and 30.9e56.4% for CS, respectively, and 3.82e5.09 MJ/Nm 3 , 1.48e1.92 Nm 3 /kg, 59.0e81.5%, and 33.0e57.1% for PKS, respectively. Results reveal that temperature has a greater role than particle size in influencing the gasification reaction rate.
Article
Potato plant waste pyrolysis is investigated by means of a cylindrical packed-bed reactor exposed along the lateral surface to heat fluxes of 20.6-33.2kW/m2 (corresponding to heating temperatures, Ts, around 530-770K). After an abrupt transition from torrefaction, a pyrolytic runaway regime is established (570K≤Ts≤620K), characterized by temperature overshoots up to 300K and actual conversion times only 4-15% of the pre-heating times. A uniform radial temperature profile is first associated with a trigger point positioned at the bed center and the propagation of a progressively thick thermal wave towards the lateral surface and then with one or more trigger points and the almost simultaneous conversion of the entire bed. Subsequently the trigger point moves to the lateral surface and a thermal wave propagates towards the bed center. The latter heating dynamics are also qualitatively preserved for Ts>620K when, in the presence of increasing radial gradients, a self-controlled pyrolysis regime is established with the reaction heat mainly exploited to pre-heat the bed
Article
Computed tomography (CT) was used to scan the samples obtained from devolatilization in a turbulent fluidized bed combustor (FBC). The evolution of three-dimensional shrinkages at different residence time was measured and visualized successfully without destroying the sample. The effects of the bed temperature, and the fuel species on the mass conversion, devolatilization time, and shrinkage characteristics were also investigated. Four common Chinese wood types, poplar, Chinese fir, oak, and teak were used. The results show that the shrinkage and mass conversion increase with increasing residence time. In addition, the devolatilization time and final tangential and radial shrinkages decrease with increasing bed temperature. The final longitudinal shrinkage shows an inverse trend compared with the other two shrinkages. The tangential shrinkage is slightly greater than the radial shrinkage and much greater than the longitudinal shrinkage for all tests. Poplar wood as a kind of cork, which has the highest volatile matter content among the four woods, showed the maximum final shrinkages.
Article
Temperature profiles inside a large pyrolysing particle were studied and are reported in this paper. Mallee trunks of similar diameter from the same tree were used to prepare cylindrical samples with 40 mm length. A fluidised-bed reactor was used to pyrolyse the large particles. The temperature profiles inside the particles were recorded during pyrolysis to allow the calculation of corresponding heating rate profiles inside the particle. The effects of moisture were studied by pyrolysing some particles with 15 to 20% moisture content. The temperature profiles obtained from the pyrolysis of dry and wet samples have been compared to identify the possible effects of moisture on the temperature profiles. A possible change in the thermal conductivity of the wood was identified around 100 °C, which caused a peak in the heating rate profile. Some possible exothermic peaks were observed at around 325 °C and 425 °C. A peak in the heating rate profile at around 200 °C in the case of the pyrolysis of wet particles was believed to be related to the changed 3-D macromolecular structure of the biomass in the presence of moisture. Some yields of tar and char along with other analytical results were presented to support our observations on the temperature profiles. Our results indicate that moisture can potentially alter the overall pyrolysis reactions and product distribution, in particular through changes in the 3-D macromolecular structure of biomass.
Article
Biomass thermochemical conversion holds great promise for producing biofuels and will play a determining role in displacing petroleum-based fuel consumption toward renewable sources. Empirical approaches have shown severe limitations in their capability to understand and control the conversion processes. However, without the ability to accurately predict and optimize thermochemical conversion performance, large-scale commercialization of these systems is severely compromised. In this context, Computational Fluid Dynamics (CFD) appears as an essential tool to better comprehend the complex physical and chemical processes involved, paving the way toward efficient control and design strategies. After a brief description of the numerical models needed to simulate biomass gasification and pyrolysis, the contributions of CFD to process design and optimization are detailed. Finally, the state of the art in terms of numerical models for the dense, reactive particulate flows typically found in conversion processes are reviewed. Shortcomings of existing CFD simulations, especially in terms of validation and predictability, are examined; and directions for future research based on the progress of CFD in other fields are suggested.
Article
This article presents a novel air-blown bubbling fluidized bed device that has the ability to sample gas and bed materials at various axial positions during the gasification experiments. The reactor was operated with olivine, as bed material, and Miscanthus, a biomass rich in potassium and silica, and thus prone to bed agglomeration. The comparison of gas and char axial profiles along the bed allows a better understanding of the biomass gasification: it shows in particular that O2 consumption and CO2 production at the bottom of the bed are mainly due to char oxidation, even if few pyrolysis gases may also be produced and oxidized near the grid. Regarding bed defluidization, the agglomerate fraction is followed by taking bed samples during the operation: it is shown that the rate of agglomeration is linear while defluidization signs appear when the agglomerate fraction reaches 6% near the grid. Small agglomerates are observed on the top of the bed, whereas big agglomerates are segregated near the grid. The SEM-EDX analysis shows that the layer that sticks olivine particles together does not strictly correspond to biomass ashes melt: it contains also particles and atoms that come from the erosion of olivine and the stainless steel wall.
Article
Reaction-induced overheating during the pyrolysis of lignocellulosic material has been reported by various authors, but it is still one of the less understood aspects of the process. This Review outlines the experimental results for a mixed kinetic-transport control in which exothermicity is displayed clearly. The thermal conditions and the feedstock properties that enhance these events are discussed. Intraparticle activity, at the microscopic level, of homogeneous and heterogeneous reactions of vapor-phase tars is the main cause of the strong exothermicity, which is more significant for hardly porous materials. Pyrolytic (thermal) runaway is also sometimes observed for fixed-bed pyrolyzers. This is initiated by the fast release rate of tarry vapors at low temperatures (in the presence of catalytically active ash and significant contents of extractives and hemicelluloses) over wide spatial extensions. Then, the acceleration of the sample heating rate triggers the degradation of the other chemical components.
Article
Kinetic models of beech wood decomposition, based on the three main pseudo-components, hemicellulose (one or two steps), cellulose (one step) and lignin (one or two steps), are applied for the first time to predict differential thermogravimetric (DTG) and differential scanning calorimeter (DSC) curves available from the literature. For the conditions that minimize the activity of secondary reactions, it is found that while the DTG curves are already well predicted by the extensively used three-step model, at least an additional step is required for the pseudo-lignin decomposition for a good prediction of the DSC data. The pseudo-cellulose decomposition is a globally endothermic process with the corresponding heat varying from about 601 to 528 J/g (of volatile evolved), as the number of reaction steps is increased from three to five. Pseudo-hemicellulose decomposition also takes place endothermally with reaction heats of 245 J/g (one step) or 321 and 226 J/g (two steps). Instead pseudo-lignin decomposition occurs with remarkable exothermicity corresponding to −923 J/g (one step) or −728 and −635 J/g (two steps).
Article
Packed-bed experiments are conducted to examine the effects of moisture evaporation on the global exothermicity of hazelnut shells pyrolysis. Percentages typically reached from ambient conditioning (9-12 wt %, db) do not modify the features of the pyrolytic runaway, with respect to bone-dry samples, apart from the obvious temporal delay. In fact, consequent to moderate external heating, moisture evaporation and solid decomposition take place sequentially. Instead, higher moisture contents lower the display of the reaction exothermicity, owing to the simultaneous occurrence of the two processes at different spatial positions. The interaction between the two is also highly dependent on the thermal severity of the external heating. Finally artificial humidification by wetting small-sized particles (versus steaming), applied to achieve moisture contents above those of ambient conditioning, modifies the sample chemical properties, so giving only a qualitative valence to the measurements.
Chapter
Thermo-chemical decomposition of biomass to bio-energy via pyrolysis is a complex process. Several pyrolysis models have been proposed for predicting the yields of desired components as a function of operating conditions. These models, however, have not considered the overall effect of process parameters and hence, are not capable of accurately predicting product yields with variation in operating conditions. Consequently, there is requirement for developing comprehensive multi-scale models for studying the combined impact of various parameters during biomass conversion. In this study, a detailed particle scale model has been developed by coupling two-stage reaction mechanism with transport phenomena to account for the combined impact of different parameters on the conversion process. Simulations have been conducted for validating this model and analysing the effect of operating temperature and particle size variation on the biomass conversion time. Based on results, it has been concluded that both particle size and operating temperature affects the rate of biomass decomposition and it is required to optimally choose these parameters with other operating conditions for getting complete conversion of biomass.
Article
This chapter deals with solid fuels and sorbent conversion during fluidized bed combustion and gasification. The conversion of solid fuels is discussed with reference to fuel properties, fuel devolatilization and volatiles conversion, char combustion and gasification reactions, mechanisms controlling char conversion rate, regimes and models of char conversion, and char particle temperature. As regards the second topic, the discussion is focused on calcium-based sorbents for in-situ desulphurization, calcination and sulphation processes, reaction models, controlling mechanisms, reactivation by water- and steam-hydration of spent sorbents and fluidized bed ash properties. Other sorbent conversion processes, such as calcium looping technique, are also mentioned.
Article
The pyrolysis characteristics of two Chinese coals, two biomass materials, and their blends were investigated by both experimental and numerical methods. Single particles of the coal and biomass were prepared for the pyrolysis experiment through grinding and pressing, while the blended particles were made by mixing the coal and biomass powder with different ratios before the pressing. Sample particles pyrolyzed in a single-particle reactor system, with the time history of the particle temperature and mass recorded. The analysis of the measured pyrolysis data of the coal, biomass, and coal–biomass blends indicate the absence of a synergistic effect between the coal and biomass pyrolysis. A numerical method coupling the chemical percolation devolatilization (CPD) model with a particle energy equation was employed to analyze the pyrolysis process. The model prediction agreed well with the experimental data for different particle diameters, fuel types, and blend mixing conditions. The fact that the co-pyrolysis of blended coal–biomass particles is well-predicted by the simple addition of the individual pyrolysis characteristics of its components also corroborates the lack of synergistic interactions. These findings will be useful for the co-combustion modeling of coal–biomass blends.
Article
CO2 gasification of coal char may play an important role in oxy-combustion environments with flue gas recirculation (FGR), but its effect on the overall reaction rate has not been clearly understood. To give clarity to the likely impact of CO2 gasification on the oxy-combustion of pulverized coal chars, burnout simulations of coal char particles were carried out, adopting apparent char reactivity and a single-film model that includes the Stefan flow effect on mass and energy transfer. Three oxygen concentrations (21%, 30%, and 5% O2), representing air, oxy-fuel, and oxygen-deficient combustion environments were simulated. A new experimental approach was used to directly measure the CO2 gasification rate of a subbituminous coal char at high temperatures and atmospheric pressure. The measured gasification rate is somewhat higher than previous measurements. The simulation results show that the endothermic gasification reaction reduces the char particle temperature and thereby reduces the oxidation rates. However, due to the contribution of the direct gasification reaction on carbon consumption, the char burnout time and the carbon consumption were improved. The gasification reaction has a greater influence on the char burnout time and the relative carbon consumption in an oxygen-deficient environment and on the drop of particle temperature in an oxygen-enriched environment (for a given gas temperature). In addition, the influence of the gasification reaction on char combustion increases as the gas temperature increases and as the particle size increases. Further, it was observed that the impact of the gasification reaction is dependent on the presumed kinetic rate, which highlights the importance of using reliable kinetic parameters in simulations. Based on the present results, it is important to include the gasification reaction by CO2 when simulating char combustion in oxy-fuel combustion environments.
Article
Full-text available
SUMMARY To date, fluidized bed combustion (FBC) has been widely accepted as a clean, efficient and flexible technique for the conversion of various solid fuels. The study described in this thesis is focussed on the combustion behaviour of a single carbon particle, as they can be found dispersed in the inert bulk material of a fluidized bed during its continuous operation as a coal combustor. Preceeding to the main research work done, a global exploratory study of single coal particle behaviour was performed through visual observation of a two dimensional fluidized bed construction at elevated temperatures. Topics of this study were coal particle fracture, mixing with the bed, tar release, evolution, ignition and combustion of volatiles, and ignition of char. Several new phenomena have been observed. The single carbon particle combustion behaviour itself is known to be determined by the following chemical / physical processes: a. oxygen transfer from the fluidized bed to the burning particle b. intraparticle transport of oxygen c. the intrinsic rate of the heterogeneous combustion reaction and its stoichiometry d. conversion induced transformations of the carbon particle structure e. heat release from the burning particle to the fluidized bed. The structure of the programme of work for this thesis has been to design specific experiments and then investigate seperately the most important aspects of the processes listed above. The experimentally determined values of the parameters involved were used as input data for a newly developed combustion model. Finally, model predictions have been compared with results of single particle combustion experiments. An experimental study has been carried out with respect to oxygen transfer within the fluidized bed (process (a)) and involved individual vaporisation of different sized naphthalene spheres immersed in fluidized beds of different materials. The mass exchange rate between the fluidized bed and a freely moving single particle could thus be determined by measuring the spheres weight loss as a function of time, although at a much lower temperature than applied for FBC. The heat exchange rate between the fluidized bed and an immersed particle (process (e)) determines, together with the combustion rate, the difference in temperature between a burning carbon particle and the fluidized bed. It has been measured by recording the heating rate of different sized cold silver and graphite spheres in the interior of a hot fluidized bed. A combustion model for a porous spherical carbon particle has been developed which accounts for the influence of the local intraparticle degree of conversion (process (d)) on the heterogeneous reaction rate and the diffusive permeability. The model allows the prediction of the combustion rate, burning particle overtemperature, density/diameter change and burnout time as a function of the fluidized bed temperature, the carbon particle diameter and its average conversion. Necessary input data for the model are the carbon reactivity and effective diffusivity together with values for the external mass and heat ransfer coefficients. Graphite has been used as a carbon particle model material for this thesis work. To allow a proper comparison with experimental results, the reactivity of graphite had to be determined by the combustion of pulverized particles in a thermobalance (process (c)), and the resulting reaction rate expression to be introduced in the model. Experimentally determined values of the solid phase diffusivity of graphite (process (b)) have been estimated from data available in literature. Combustion of individual graphite spheres has been realized in a fluidized bed column of O. 1 m diameter for two different bed temperatures. The combustion 2 rate could be followed by continuous measurement of the CO and co2 concentrations in the flue gas. The burning temperature of a seperate series of graphite spheres was recorded by thermocouples embedded at the spheres centre. Comparison between experimental results and model predictions provides a clear understanding of the combustion mechanism. The illustration of general trends by model simulations is satisfactory. A number of the combustion results were used to evaluate mass and heat transfer coefficients. The agreement with results of the afore mentioned direct measurements is excellent. The combustion rate of a single carbon particle under mass and/or heat transfer limited conditions can be predicted accurately. This, however, is not the case if the chemical reaction rate becomes an important factor (e.g. for small particles), because highly coal specific parameters are involved then. Moreover, internal burning of a carbon particle may not occur uniformly and the smooth field assumption, used for model calculations, is possibly not valid.
Thesis
Full-text available
Thesis (D. of Ingeniør)--Norwegian University of Science and Technology, 1996. In the first part of this thesis, experimental and modeling work on the pyrolysis of biomass under regimes controlled by chemical kinetics are presented. The pyrolysis of three Scandinavian wood species (birch, pine, pineroot and spruce); five different celluloses, and hemicellulose and lignin isolated from birch and spruce have been studied by thermogravimetry. The celluloses exhibited sharp, single DTG curves which could be well described by a single first order reaction model. The kinetic analysis gave activation energies between 210 and 280 kJ/mole for all the celluloses. The thermograms of the wood species revealed different weight loss characteristics which can be attributed to their different chemical composition. In the kinetic analysis, a model of independent parallel reactions was successfully used to describe the thermal degradation. The main advantage of this modeling approach is that the in situ decomposition of cellulose in wood can be studied. A mean activation energy of 260 kJ/mole with a standard deviation ±16 kJ/mole was calculated. In the second part of this thesis, experimental and modeling work on the pyrolysis of biomass under regimes controlled by heat and mass transfer are presented. Particles (D=20 mm, L=30 mm) prepared from the same wood species were one-dimensionally heated. The effect of heating conditions (heat flux, grain orientation) on the product yields distribution and reacted fraction was investigated. The experiments show that heat flux alters the pyrolysis products as well as the intraparticle temperatures to the greatest extent. A comprehensive mathematical model which can simulate drying and pyrolysis of moist wood is presented. The thermal degradation of wood involves the interaction in a porous media of heat, mass and momentum transfer with chemical reactions. Heat is transported by conduction, convection and radiation and mass transfer is driven by pressure and concentration gradients. The simulation of these processes involves the simul¬taneous solution of the partial differential, conservation equations for mass, energy and momentum with kinetic expressions describing the rate of reaction. By using three parallel competitive reactions to account for primary production of gas, tar and char, and a consecutive reaction for the secondary cracking of tar, the predicted intraparticle temperature profiles, ultimate product yields distribution and reacted fraction agreed well with the experimental results.
Article
Drying and devolatilization from coal particles and the mass transfer processes in the immediate vicinity of burning char particles in a fluidized bed consisting of predominantly smaller non-combusting particles are reviewed. Single particle studies with well defined external heat/mass transfer conditions are thought necessary to refine the current understanding of drying, devolatilization and volatile matter combustion local to the coal particle. An empirical model and a theoretical model are considered to most accurately reflect the mass transfer data. It is considered important to include in theoretical models dealing with various aspects of coal combustion (drying, devolatilization, heat/mass transfer, char combustion) the two-phase nature of the bed and the possibility that the burning particle spends a significant proportion of its 'life' in the bubble phase. Further careful experimentation is necessary to refine mechanistic understanding.
Chapter
The biomass or commonly available lignocellulosic materials could be converted to different types of fuel and chemical feedstock by a variety of thermochemical processes. Each of these processes involves two highly significant and inter-related general aspects: first, material and energy transformations that could be explored and understood through the discipline of chemistry; second, material transport and heat transfer that could be investigated and designed through the disciplines of process engineering. Variations of the reaction conditions and the processing design, which are closely related, are often investigated in a sporadic or empirical manner in order to define the optimum conditions that provide high yields of operational efficiency.
Article
Seeking to systemalically identify and organize the physical and chemical properties of biomass particles for combustion in furnaces, an attempt is made here to theoretically describe the temporal change in mass of a burning particle of an organic solid. The development involves a series of simple models to highlight the mechanisms involved, the physical and chemical properties of concern and the expected results. Although many simplifying assumptions are made, the properties identified, the parameters which evolved and the final functional form of the results are expected to becorrect.
Article
Single, thermally thick particles of lodgepole pinewood were pyrolyzed under well-defined conditions of industrial importance. Particle thickness, heating level, moisture content, density, and grain axis relative to one-dimensional heating were varied using a Box-Behnken experimental design. Gross product fractions, as well as components therein, were measured and the batch yields were correlated with second-order polynominals. The empirical equations correlating the batch yields, together with their prediction uncertainties, are presented and are suitable for use in simulations of wood combustion and thermal conversion. Comparison of large particle pyrolysis product distributions to other studies of small-particle pyrolysis yields shows the trends with particle size to be consistent. Tar yield minima depend on both particle size and heating rate. Gas yield is dependent on both particle size and heating intensity. Because some process controllables were found to alter product yields from large particles in a multiplicative way, rather than an additive way, suggestions for future experiments are made.
Article
A lumped-parameter kinetic model is applied to simulate the pyrolysis of lignocellulosic particles, exposed to a high temperature environment. Physical processes account for radiative, conductive and convective heat transport, diffusion and convection of volatile species and pressure and velocity variations across a two-dimensional (2-D) , anisotropic, variable property medium. The dynamics of particle degradation are found to be strongly affected by the grain structure of the solid. A comparison is made between the total heat transferred to the virgin solid (conduction minus convection) along and across the grain. Notwithstanding the lower thermal conductivities, because of the concomitant slower convective transport (lower gas permeabilities) , the largest contribution is that across the solid grain. The role played by convective heat transport is successively less important as the particle size is increased. Finally, the 2-D and the widely applied one-dimensional (1-D) predictions are compared.
Article
The pyrolysis regimes of thermoplastic polymers (polyethylene) are examined through a mathematical model, including transport phenomena and chemical reactions. Depolymerization and melting are followed by devolatilization. Surface regression, property variation, heat convection and conduction are properly taken into account. Simulations are carried out for external heating conditions corresponding to fixed-bed reactors, hot-plate contact and fire-level radiation exposures. Only very slow external heat transfer rates, associated with low reactivity of the fuel, give rise to a pure kinetic control. For conditions of interest in fixed-bed reactors and radiative heating, a transition from a thermally thin to a thermally thick regime is observed, as the sample size is increased. Two regimes can be established during hot-plate contact, namely an ablative regime or again a thermally thick regime, depending on the relative importance between internal heat transfer and chemical reaction kinetics.
Chapter
The effect of wood preservatives on the amount and composition of pyrolysis products were investigated over a wide range of temperatures (300 to 700 °C). The experiments were carried out in a modified bench scale unit according to the Waterloo-Process. Of particular interest was the distribution of heavy metals in the different pyrolysis products. Under strict German environmental laws, the combustion of wood waste in incinerators is restricted to those with gas cleaning systems which comply to the 17th Ordinance on the Implementation of the Federal Imission Control Act (17. BImSchV), because of the various wood preservatives, insecticides or coatings present in the wood. In this situation it is essential to find new and cheaper ways for wood waste management. After evaluating a series of experiments flash pyrolysis seems to be a promising approach of wood waste managing and the production of liquid fuel. Treated wood does not have a negative effect on the pyrolysis products. The amount of liquid remains high and the heavy metals from the preservatives accumulate in the char, not in the liquid. Furthermore, boron has a significant effect on the amount of levoglu-cosan, hydroxyacetone and hydroxyacetaldehyde. The levoglucosan yield in the pyrolysis liquid from treated wood increased whilst the hydroxyacetone and the hydroxyacetaldehyde yields decreased. The presence of creosote during the pyrolysis did not affect the products from the wood and the creosote did not change either.
Article
A mathematical model is presented for the combustion of small wooden dowels heated externally. The heat transfer is described by the Fourier conduction equation, including a heat-source term, with both convection and radiation at the surface. The decomposition is assumed to follow a first-order Arrhenius equation. Both equations are solved simultaneously, for successive shells of the dowel. After one-half or more of a particular shell is consumed, the energy of activation, velocity constant, and heat of decomposition for that shell are all increased. These increases match a presumed change in the degradation process from a rapid evolution of gaseous products to a slower degradation of the remaining charcoalike substance.Experimental results are presented showing center-temperature, weight-loss, and internal-pressure histories for 3/8-inch dowels, which were inserted into a preheated air furnace whose temperature was held constant in the range of 300° to 650°C. The center temperatures are fairly well predicted except for peak values of the exothermic contribution. The experimental and theoretical weight-loss histories agree quite well for all furnace temperatures. The pressure at the center rises to several psig and then drops suddenly to zero before the center temperature reaches the furnace temperature. While the pressure is dropping, serious structural failures, such as longitudinal channeling and surface cracking, occur.
Article
Radiative pyrolysis of thermally thick beech wood has been investigated through a comparison between dry and moist [11% dry basis (db)] particles, for heat fluxes in the range 27.5−80 kW/m2. The initial moisture content has also been varied from 0 to 50% (db) for two radiative fluxes, 27.5 and 49 kW/m2, corresponding to slow and fast external heat-transfer rates, as steady surface temperatures are about 625 and 800 K, respectively. For very slow heating, moisture evaporation precedes wood pyrolysis. As the external heating conditions are made more severe and/or the initial moisture content is increased, the two processes take place simultaneously, associated with the propagation of separate fronts along the particle radius. Spatial gradients also increase, while apparent weight loss kinetics from a single-peak rate turn into a two-peak rate. The conversion times increase almost linearly with the initial moisture content, but differences in primary product (char, gas, and liquids) yields and gas composition are negligible.
Article
The pyrolysis characteristics of agricultural residues (wheat straw, olive husks, grape residues, and rice husks) and wood chips have been investigated on a bench scale. The experimental system establishes the conditions encountered by a thin (4 × 10-2 m diameter) packed bed of biomass particles suddenly exposed in a high-temperature environment, simulated by a radiant furnace. Product yields (gases, liquids, and char) and gas composition, measured for surface bed temperatures in the range 650−1000 K, reproduce trends already observed for wood. However, differences are quantitatively large. Pyrolysis of agricultural residues is always associated with much higher solid yields (up to a factor of 2) and lower liquid yields. Differences are lower for the total gas, and approximate relationships exist among the ratios of the main gas species yields, indicating comparable activation energies for the corresponding apparent kinetics of formation. However, while the ratios are about the same for wood chips, rice husks, and straw, much lower values are shown by olive and grape residues. Large differences have also been found in the average values of the specific devolatilization rates. The fastest (up to factors of about 1.5 with respect to wood) have been observed for wheat straw and the slowest (up to factors of 2) for grape residues.
Article
Results from a numerical model for endothermic biomass pyrolysis, which includes both high activation energy kinetics and heat transfer across a boundary layer to the reacting solid particle, are presented. The model accounts for conventional thermocouple thermal lag and unconventional thermal lag due to heat demand by the chemical reaction (which is governed by Arrhenius kinetics). Biomass fusion, first identified quantitatively by Lédé and Villermaux, is shown to be a manifestation of severe thermal lag that results from the chemical reaction heat demand. Over the wide range of conditions studied, the true substrate temperature remains almost constant during pyrolysis, as is the case with compounds undergoing fusion or sublimation at constant pressure. A simple algebraic model, whose derivation presupposes the idea that biomass pyrolysis mimics the melting of a block of ice, accurately predicts the maximum value of thermal lag during pyrolysis. Unidentified thermal lag in TGA experiments lowers the values of the apparent activation energy and frequency factor associated with the experimental data but approximately retains the true value of their ratio. Thus, the widely varying values of kinetic parameters for cellulose pyrolysis reported in the literature may be a result of differing thermal lag characteristics of the experiments.
Article
The coupled effects of particle size and external heating conditions (reactor heating rate and final temperature) on cellulose pyrolysis are investigated by means of a computer model accounting for all main transport phenomena, variable thermophysical properties and primary, and secondary reaction processes. The dynamics of particle conversion are predicted, and final product distributions are favorably compared with experimental measurements. A map is constructed, in terms of particle size as a function of the reactor temperature, to identify the transition from a kinetically controlled conversion to a heat transfer controlled conversion (thermally thin and thermally thick regimes) and from flash to slow-conventional pyrolysis. Conditions for maximizing oil, gas, or char yields are also discussed.
Article
The release of volatiles of pine wood particles was analyzed by means of continuous measurements of the CO2 and O2 concentrations obtained after the complete combustion of the volatiles and from flame extinction times. A relatively simple mathematical model was used to predict the O2 consumed, as a function of time, during the devolatilization and further volatiles combustion of wood chips in fluidized beds. In this model, the drying and pyrolysis of wood particles is considered as a coupled process controlled by the kinetics of devolatilization as well as heat transfer to and through the particles. A distributed activation energy model is adopted for the kinetics of devolatilization of the wood particles. The wood chips are characterized by an equivalent particle diameter and a shape factor and transformed into spherical shape. The model was in satisfactory agreement with experimental data over a wide range of operating conditions, particle sizes and shapes, and particle moisture contents of practical importance for wood combustion in fluidized beds.
Article
Weight loss curves of thin layers (150 μm) of beech wood powder, measured for heating rates of 1000 K/min and final temperatures in the range 573−708 K, show final char yields of 37−11%. The process is kinetically controlled and, for the most part, isothermal. A one-step global reaction, with E = 141.2 ± 15.8 kJ/mol and ln A = 22.2 ± 2.9 s-1, is a degradation mechanism capable of capturing the main features of the process. The thermogravimetric curves also allow the formation rate constants to be estimated for char and total volatiles (activation energies of 111.7 ± 14.3 and 148.6 ± 17.4 kJ/mol, respectively) and, once integrated byproduct distribution, those for liquids and gases (activation energies of 148 ± 17.2 and 152.7 ± 18.2 kJ/mol, respectively). A comparison is provided with pyrolysis mechanisms available in the literature.
Article
Cellulose and maple sawdust have been pyrolyzed by different workers in two different reactors (a fluid bed and a transport reactor) in separate laboratories. The Avicel cellulose sample used by both groups was from the same batch, while the maple was different samples of the same species. Fast pyrolysis product yields were compared at a constant vapor residence time of 500 ms over a temperature range of 450-900°C and were found to be in very good agreement. It is proposed that if particle heat-up time to 500°C, for any reactor, is significantly less than particle residence time, or if particle weight loss is less than 10% before the particle temperature reaches 450°C, then the temperature of the reactor will be the only variable determining the yields of char, oil, and gases for a given feed material and a given gas residence time. The implications of the results in terms of product yields and possible pyrolysis mechanisms are discussed. The oil yield as temperature increases can be described adequately by a simple kinetic model.
Article
Recent advances in experimental methods and computer modeling have shed new light on the kinetics of cellulose pyrolysis. The rich slate of reaction products that evolve when cellulose is heated implies that the pyrolysis chemistry is exceedingly complex. Nevertheless, a simple, first order, high activation energy (ca. 238 kJ/mol) model accurately describes the pyrolytic decomposition of an extraordinary variety of cellulosic substrates. Secondary vapor-solid interactions are the main source of char formed during cellulose pyrolysis. When a whole biomass substrate is pretreated to remove mineral matter, the pyrolysis kinetics of its cellulose component are very similar to those of pure cellulose. Future work should focus on the effects of mineral matter on pyrolysis, and the secondary vapor-solid reactions which govern char formation.
Article
A pyrolysis unit was developed to study effects of temperature, heating rate, wood particle size, moisture, gaseous environment, and catalyst impregnation on the wood pyrolysis. Effects of these parameters on char, oil, and gas yields and composition are presented and interpreted to assist the industrial use of the wood-pyrolysis technology.
Article
Experimental and modeling work on pyrolysis of wood under regimes controlled by heat and mass transfer are presented. In a single-particle, bell-shaped Pyrex reactor, one face of a uniform and well-characterized cylinder (D = 20 mm, L = 30 mm) prepared from Norwegian spruce has been one-dimensionally heated by using a Xenon-arc lamp as a radiant heat source. The effect of applied heat flux on the product yield distributions (char, tar, and gas yield) and converted fraction have been investigated. The experiments show that heat flux alters the pyrolysis products as well as the intraparticle temperatures to a great extent. A comprehensive mathematical model that can simulate pyrolysis of wood is presented. The thermal degradation of wood involves the interaction in a porous media of heat, mass, and momentum transfer with chemical reactions. Heat is transported by conduction, convection, and radiation, and mass transfer is driven by pressure and concentration gradients. The modeling of these processes involves the simultaneous solution of the conservation equations for mass and energy together with Darcy's law for velocity and kinetic expressions describing the rate of reaction. By using three parallel competitive reactions to account for primary production of gas, tar, and char, and a consecutive reaction for the secondary cracking of tar, the predicted intraparticle temperature profiles, ultimate product yield distributions, and converted fraction agreed well with the experimental results.
Article
Biomass pyrolysis oils were produced from stored biomass feedstocks by rapid pyrolysis in a fluidized bed reactor. The feedstocks used for these studies were switchgrass, corn stover, and hybrid poplar. The woody and herbaceous feedstocks were stored in chip piles and bales, respectively, unprotected in an open field for 6 months. At the end of the storage period, biomass samples were taken from the interior of bales and the centers of chip piles for pyrolysis studies. The materials were ground to pass -20/+80 mesh and dried to less than 10% moisture content before pyrolyzing in the fluidized bed reactor. Pyrolysis was conducted at 500 degrees C and with less than 0.4 s apparent vapor residence time. Total liquid yields were as high as 66% for the hybrid poplar and as low as 58% for the corn stover. Moisture content of the oils was between 10 and 13%. Gas and char/ash yields were 10-15% and 12-22%, respectively. The char/ash yields were feedstock dependent, but storage influence was significant for only the corn stover feedstock. Gas and liquid yields were not influenced by storage time. The oils were highly oxygenated and had higher heating values (HHV) of 23-24 MJ/kg that decreased slightly with storage time for all the feedstocks except the switchgrass. The oils, as currently produced, are high in ash and alkali metals. Ultimately, they may be upgraded and used as boiler and turbine fuels.
Article
A continuous atmospheric pressure flash pyrolysis process for the production of organic liquids from cellulosic biomass has been demonstrated at a scale of 1–3 kg/hr of dry feed. Organic liquid yields as high as 65–70% of the dry feed can be obtained from hardwood waste material, and 45–50% from wheat straw. The fluidized sand bed pyrolysis reactor operates on a unique principle so that char does not accumulate in the bed and treatment of the sand is not necessary. The product gas, about 15% of the yield, has a medium heating value. The liquid product is an acidic fluid, which pours easily and appears to be stable. A preliminary economic analysis suggests that if the pyrolysis oil can be used directly as a fuel, its production cost from wood waste is probably competitive with conventional fuel oil at the present time. On a fait la démonstration d'un procédé continu de pyrolyse éclair à l'échelle de l à 3 kg/h d'alimentation sèche, à la pression atmosphérique, pour la production de liquides organiques à partir d'une biomasse cellulosique. On peut obtenir des rendements en liquide atteignant 65 à 70% de l'alimentation sèche à partir de déchets de bois dur et de 45 à 50% à partir de paille de blé. Le réacteur de pyrolyse, à lit de sable fluidisé, fonctionne sur un principe unique, de sorte que le charbon ne s'accumule pas dans le lit et qu'il n'est pas nécessaire de traiter le sable. Le produit gazeux, qui correspond à environ 15% de rendement, a une valeur calorifique moyenne. Le produit liquide est un fluide acide qui se déverse facilement et semble assez stable. Une analyse économique préliminaire indique que, si l'on peut employer directement l'huile de pyrolyse comme combustible, son coǔt de production à partir de déchets de bois est probablement compétitif actuellement avec celui de l'huile combustible classique.
Article
The devolatilization time of mm-size coal particles affects the in-bed combustion efficiency of volatiles in fluidized bed combustors and has been studied by numerous workers. Conflicting effects of the various parameters on devolatilization time have been reported, when different rigs, measurement methods and conditions were used. Therefore, this work aimed to study the effects of experimental parameters using the same rig, procedure and measurement method to eliminate these differences. Four coals, ranging from lignite to bituminous, from the Penn State coal sample bank were studied in a thermogravimetric analysis (TGA) apparatus. Higher heating rates, closer to those in a fluidized bed combustor, were obtained by inserting the coal particles directly into the already hot furnace, rather than using the relatively slow heating rate programmed into the apparatus. The devolatilization time was measured from the weight loss curve for individual coal particles in the 1–5 mm size range. The effect of particle size was correlated by equations of the form, tv = Adp1.5, consistent with a heat transfer-controlled regime. No effects of coal type or gas flow rate on the devolatilization time were found. The effects of atmosphere (inert or air) and devolatilization time definition (90 or 95% weight loss) were expressed as ratios of the values of the correlation parameter, A, and the effect of temperature was presented as an Arrhenius-type plot for A. The decreases in devolatilization time with increasing temperature or oxygen concentration again reflect the controlling nature of the heat transfer process. No fragmentation of the coal was observed in this work, but the effect of primary fragmentation was deduced by comparison with literature data. It is suggested that an effect of coal type on devolatilization time is only observed when different degrees of primary fragmentation occur for different coals, and this may explain the differing observations for coal type effect in the literature.
Article
A transient unimodel for reacting pellets is considered with various modes of heat and mass transfer and structural changes for wood pyrolysis. Pellet breakup was found to be possible from strength calculations. This leads to an increase in the number of pellets and a decrease in the resistance to heat and mass transfer. The pressure and temperature buildup within 2.7 mm thick pellets was measured for wood pyrolysis/combustion experimentally. The bimodal wood pyrolysis was analyzed, and the rate constant and activation energy were found. Pellet breakup may also be used as a transient catalytic process where the catalysts become smaller as they break up in time.
Article
Devolatilization and char burning were studied in an electrically heated bench-scale fluidized-bed reactor at 750 to 900°C bed temperature, gas oxygen mole fractions ranging from zero to 0.21, superficial gas velocities from 0.3 to 0.7 m/s and coal particle diameters 5 to 35 mm. The coals investigated include lignite, bituminous and anthracite. The coal devolatilization and char burning times, H/C ratio histories, and particle fragmentation were measured. Statistical correlations with the operating variables were developed for the devolatilization time. A mathematical model is given for the combustion of char. Most predictions of the model agree quite well with the experimental results. On a étudié la dévolatilisation et la combustion de produits de carbonisation dans un réacteur à lit fluidisé chauffé à l'électricité à l'échelle expérimental pour des températures de 750 à 900°C, des fractions molaires de l'oxygène gazeux de zéro à 0,21, des vitesses superficielles de gaz de 0,3 à 0,7 m/s et des diamètres de particules de charbon de 5 à 35 mm. Les charbons étudiés comprennent la lignite, les bitumineux et l'anthracite. On a mesuré la dévolatilisation du charbon et les temps de combustion des produits de carbonisation, l'histoire des rapports H/C ainsi que la fragmentation des particules. Des corrélations statistiques du temps de dévolatilisation avec les variables de fonctionnement ont été mises au point. On présente un modèle mathématique pour la combustion des produits de carbonisation. La plupart des prédictions du modéle concordent relativement bien avec les résultats expérimentaux.
Article
The present work provides a rationally-based model to describe the pyrolysis of a single solid particle of biomass. As the phenomena governing the pyrolysis of a biomass particle are both chemical (primary and secondary reactions) and physical (mainly heat transfer phenomena), the presented model couples heat transport with chemical kinetics. The thermal properties included in the model are considered to be linear functions of temperature and conversion, and have been estimated from literature data or by fitting the model with experimental data. The heat of reaction has been found to be represented by two values: one endothermic, which prevails at low conversions and the other exothermic, which prevails at high conversions. Pyrolysis phenomena have been simulated by a scheme consisting of two parallel reactions and a third reaction for the secondary interactions between charcoal and volatiles. The model predictions are in agreement with experimental data regarding temperature and mass-loss histories of biomass particles over a wide range of pyrolysis conditions.
Article
A mathematical model for intra-particle transport phenomena and chemical reactions is coupled with an external heat transfer model, taking into account fluid-bed hydrodynamics, to predict the fast pyrolysis characteristics of cellulosic fuels. Good agreement is obtained between predicted and measured product yields as functions of the reactor temperature. For practical applications aimed at liquid fuel production, particle size and external temperatures greatly affect the average particle heating rate (values roughly comprised between 300 and ), whereas the actual degradation temperature vary in a narrow range (600–). Consequently, variations in the conversion time are significantly larger than in product distribution and yields. Finally, comparisons are made with the Ranz–Marshall correlation and the limit case of infinitely fast external heat transfer rates.
Article
Over the past two decades a great deal of experimental work as been carried out on the development of fast pyrolysis processes, particularly for biomass and for lignocellulosic waste materials. High yields of an organic liquid product (50–70%) are typical of atmospheric pressure short contact time pyrolysis of such feedstocks. This liquid product has been shown to be usable both as an alternative liquid fuel, and as a chemical feedstock because of its content of significant concentrations of potentially useful organic chemicals. The characteristics of the more important fast pyrolysis processes are reviewed, and the advantages and problems existing with present pyrolysis reactors are discussed, with the emphasis on bubbling fluidized bed systems. Experience with the process has led us to a somewhat different view of the optimal conditions for fast pyrolysis, and has resulted in the recent development of a new fluid bed process—the RTI Process. Characteristics of the RTI process are described and its advantages over existing fast pyrolysis technologies summarized.
Article
Times required for devolatilisation of large coal particles in a fluidised-bed operating at 750, 850 and 950°C and in gas environments simulating pyrolysis, combustion and gasification conditions were investigated. The effect of coal type and coal moisture content was also investigated at a bed temperature of 850°C under fluidised-bed combustion conditions. The devolatilisation time was defined as the time taken from immersion into the bed until the centre temperature of the coal particle equalled the bed temperature. Six coals of varying rank in the size range from 6 to 17 mm, were used in this present study. The devolatilisation time was found to be influenced by the bed temperature, gas environment, coal moisture content and coal type which is in agreement with data reported in the literature. The results were correlated with the classic empirical particle diameter power law relation (tv=Adpn) where it was observed that the devolatilisation time was directly proportional to the particle diameter. A new definition to distinguish between heat transfer and chemical-kinetically controlled regimes during devolatilisation validates present method and explains experimental observations relating to the influence of bed temperature and gas environment.
Article
Biomass fast pyrolysis is of rapidly growing interest in Europe as it is perceived to offer significant logistical and hence economic advantages over other thermal conversion processes. This is because the liquid product can be stored until required or readily transported to where it can be most effectively utilised. The objective of this paper is to review the design considerations faced by the developers of fast pyrolysis, upgrading and utilisation processes in order to successfully implement the technologies. Aspects of design of a fast pyrolysis system include feed drying; particle size; pretreatment; reactor configuration; heat supply; heat transfer; heating rates; reaction temperature; vapour residence time; secondary cracking; char separation; ash separation; liquids collection. Each of these aspects is reviewed and discussed.
Article
The thermal decomposition of wood has up to now been represented by an overall first order reaction with a definite heat of reaction.1 There is now sufficient experimental evidence to show that such a treatment is too simple, since values obtained for activation energy and heat of reaction vary with experimental conditions.2,3The present paper describes a series of experiments in which cylinders of wood were decomposed under controlled heating conditions in an atmosphere of nitrogen in a furnace. During each experiment, the specimen was weighed continuously and its temperature was measured at several points. Data were analyzed in terms of the above theory to examine its validity and shortcomings.
Article
Biomass in the form of mixed wood waste was pyrolysed in a fluidized bed reactor at 400, 450, 500 and 550°C. The char, liquid and gas products were analysed to determine their elemental composition and calorific value. In particular, the liquid products were analysed in detail to determine the concentration of environmentally hazardous polycyclic aromatic hydrocarbons (PAH) and potentially high-value oxygenated aromatic compounds in relation to the process conditions. The gases evolved were CO2, CO and C1C4 hydrocarbons. The liquids were homogeneous, of low viscosity and highly oxygenated. The molecular weight range of the liquids was 50–1300 u. Chemical fractionation of the liquids showed that only low quantities of hydrocarbons were present and the oxygenated and polar fractions were dominant. PAH up to MW 252 were present in the liquids; some of the PAH identified have been shown to be carcinogenic and/or mutagenic. The concentration of PAH in the liquids increased with pyrolysis temperature, but even at the maximum pyrolysis temperature of 550°C the total concentration was < 120 ppmw. The liquids contained significant quantities of phenolic compounds and the yield of phenol and its alkylated derivatives was highest at 500 and 550°C. Some of the oxygenated compounds identified are of high value.
Article
Thick wood cylinders have been pyrolyzed with applied radiation intensities in the range 28–80 kW/m2, to investigate the role of wood variety on the degradation characteristics (temperature and weight loss dynamics), product (char, gas, and liquid) yields, and gas composition. Two hardwoods (beech, chestnut) and three softwoods (Douglas fir, redwood, and pine) have been examined. Apart from the higher minimum heat flux needed for softwood pyrolysis, all the varieties present the same qualitative behavior, and the process dynamics tend to become the same for applied heat fluxes above 40 kW/m2, when internal heat transfer is the controlling mechanism. However, quantitative differences remain large in terms of pyrolysis temperature (maximum values of 600–650 K), product yields (minimum char yields of 21–33%, maximum liquid yields of 47–57%), and average devolatilization rate, as a consequence of variations in the chemical composition.
Article
New theory is developed to define the parameters controlling the pyrolysis rate of single particles. It is shown that the relative importance of internal and external heat transfer and of the intrinsic (first order) pyrolysis kinetics can be determined from the Biot number (hR/K) and, depending on the value of the Biot number, one of two Pyrolysis numbers defined by Py = (K/kϱcpR2) or Py′= (h/kϱcpR). On the basis of these groups four regions are defined, and appropriate and simple models of the kinetics of primary pyrolysis outlined. The models are tested against measurements of decomposition and temperature distributions in pyrolysing wood cylinders with diameters in the range 0.6–2.2 cm and at temperatures from 380 to 500°C. Good agreement with theory is found and it is concluded that, under these conditions, internal convection is unimportant and that with suitably large or small values of Bi, Py and Py′ very simple models are adequate. Some implications for reactor design are briefly discussed.
Article
In fluidized-bed combustion, relatively few larger and lighter coal particles are fluidized along with more dense and smaller sulphur-sorbent bed particles. Predictions for drying, ignition and devolatilization, as well as particle temperatures during char combustion, require knowledge of the external heat transfer coefficients. Several studies have dealt with mass transfer to freely moving ‘active’ particles. Heat transfer to fixed immersed objects in fluidized beds has also received wide research attention. Very few studies have addressed heat transfer to freely moving ‘active’ spheres. This paper proposes a mechanistic model for heat transfer with the effect of ‘active’ particle motion taken into account. The average heat transfer coefficients calculated from the model compare well with experimental data in the literature. The model predictions are also compared with available empirical correlations. The implications of the model predictions for modelling fluidized-bed combustion are also discussed.
Article
A coupled transport and reaction model is formulated to investigate the effects of various parameters on biomass pyrolysis. The model takes into account formation of chars, tars and gases through mechanisms including both primary reactions of the virgin biomass degradation and secondary reactions of the primary tar. All main transport phenomena, unsteadiness of the gas/vapor-phase processes, variation of the reacting medium properties and particle shrinkage are also described. Numerical simulation of the problem of a wooden particle, subjected to an assigned external radiation, is used to analyze time and space evolution of the main variables and product distribution as the shrinkage parameters and the intensity of the heat flux are varied. The effects of the orientation of the anisotropic wood grain relative to the one-dimensional heat flux are also investigated.
Article
A detailed single-particle model, including a description of transport phenomena and a global reaction mechanism, is coupled with a plug-flow assumption for extraparticle processes of tar cracking, in order to predict the fast pyrolysis of wood in fluid-bed reactors for liquid-fuel production. Good agreement is obtained between predictions and measurements of product yields (liquids, char, and gases) as functions of temperature. Particle dynamics are very affected by the convective transport of volatile products. The average heating rates are on the order of 450-455 K/s, whereas reaction temperatures vary between 770 and 640 K (particle sizes of 0.1-6 mm and a reactor temperature of 800 K). The effects of several factors, such as size, shape, and shrinkage of wood particles, and external heat-transfer conditions are also examined.
Article
The thermal decomposition of relatively large particles of pine wood has been studied. By use of spherical particles of different sizes (2-5.6 cm in diameter), the experimental solid conversion values and the temperature at different points of the solid are analyzed for different heating rates of the system. The experimental results have been compared with those calculated with a simple model, which involves the solution of the heat and mass balances and the kinetic equations, using experimental values as boundary conditions. A good agreement has been obtained for particles up to 4 cm in diameter. For larger particles the mass transfer resistance inside the solid and the existence of secondary reactions may acquire an appreciable importance.
Pellet Breakup Due to Pressure Generated during Wood Pyrolysis 2-10 mm and T r ) 807 K) or 10-14 (T r ) 646-1107 K and d ) 4 mm) That is, internal heat transfer is
  • M Q Syed
Syed, M. Q. Pellet Breakup Due to Pressure Generated during Wood Pyrolysis. Ind. Eng. Chem. Res. 2000, 39, 3255-3263. ) 2-10 mm and T r ) 807 K) or 10-14 (T r ) 646-1107 K and d ) 4 mm). That is, internal heat transfer is (37) Di Blasi, C.; Signorelli, G.; Di Russo, C.; Rea, G. Product Distribution from Pyrolysis of Wood and Agricultural Residues. Ind. Eng. Chem. Res. 1999, 38, 2216-224.
A Theoretical and Experimental Study of the Thermal Degradation of Biomass28) Di Blasi, C. Physicochemical Processes Occurring inside a Degrading Two-Dimensional Anisotropic Porous Medium
  • M G Gronli
Gronli, M. G. A Theoretical and Experimental Study of the Thermal Degradation of Biomass. Ph.D. Thesis, NTNU, Trondheim, Norway, 1996. (28) Di Blasi, C. Physicochemical Processes Occurring inside a Degrading Two-Dimensional Anisotropic Porous Medium. Int. J. Heat Mass Transfer 1998, 41, 4139-4150.