Article pubs.acs.org/EF
Stepwise Fast Pyrolysis of Pine Wood Roel J. M. Westerhof,*,† D. Wim. F. Brilman,† Manuel Garcia-Perez,‡ Zhouhong Wang,‡ Stijn R. G. Oudenhoven,† and Sascha R. A. Kersten† †
Thermal Chemical Conversion of Biomass Group, Faculty of Science and Technology, University of Twente, PO Box 217, 7500 AE, Enschede, The Netherlands ‡ Department of Biological Systems Engineering, Washington State University, Pullman, Washington 99164, United States ABSTRACT: This paper reports the yields and product composition obtained from the stepwise pyrolysis of pine wood in a fluidized bed reactor. The first step temperature was varied between 260 °C and 360 °C. After the first step, the solid residue was cooled to ambient temperature and pyrolyzed again at 530 °C. If the first step temperature was below 290 °C, the cumulated yields (the sum of steps 1 and 2) were identical to yields of the single-step experiment at 530 °C. This indicates that the chemical and transfer processes taking place below 290 °C do not lead to chemical and structural changes that affect the outcome of the processes occurring in the temperature range from 290 °C to 530 °C. When the first step temperature was higher than 310 °C the cumulated yields of char, water, light organic compounds, and furans were higher, whereas the cumulated yields of volatiles (gases plus organic liquids) were lower than those obtained if the pyrolysis was conducted in a single step at 530 °C. To explain these observations, a mechanism is proposed in which the main emphasis lies on the competition between routes that lead to char formation and routes that release compounds from the biomass particle. Single compounds and lumped groups could not be concentrated with the stepwise approach. A separation into lighter and heavier oils turned out to be possible.
1. INTRODUCTION Fast pyrolysis is a technology that converts bulky inhomogeneous biomass into a liquid, often called pyrolysis oil or bio-oil. The yield and composition of pyrolysis oil depend on, among other factors, feedstock composition, particle size,1 reactor temperature,2−4 and condensation strategy.5,6 Most fast pyrolysis reactors must operate with (very) small particles to achieve high bio-oil yields. For grinding of the raw biomass, a significant amount of energy is required. A thermal treatment (torrefaction) step could help to reduce the grinding energy significantly7 and improve the economic viability of these technologies. However, it is not known how torrefied biomass will behave during pyrolysis. Furthermore, pyrolysis oil obtained from a one-step fast pyrolysis process has some adverse properties, such as high acidity, high water content, high oxygen content, low heating value, and the inherent complexity related to being a multicomponent mixture.8,9 Biooil performance parameters in a given application are often related to the presence of certain compounds.2,9,10 For example, acidity is related to the presence of acetic acid; viscosity is controlled by the content of water and (lignin) oligomers; combustion performance is dependent mostly on water content and charring compounds; and, last, stability is mainly dependent on (lignin) oligomers, aldehydes, and guaiacols.6,9 Stepwise pyrolysis (torrefaction followed by pyrolysis) could be an interesting process to produce pyrolysis oils with a lower content of acids, concentrated desired compounds and with improved properties. There are few papers available that report on stepwise pyrolysis. Most of them focus on cellulose and other model compounds as feed and on the unraveling of reaction pathways via the stepwise approach. Several groups11−14 have studied the stepwise pyrolysis of cellulose. The evolution of phenols was studied by Murwanashyaka et al.15 via the vacuum stepwise © 2012 American Chemical Society
pyrolysis of a mixture of birch bark (ca. 46%) and birch sapwood (ca. 54%). Stepwise pyrolysis of hybrid poplar and the wood constituents hemicellulose, cellulose, and lignin was studied by Jones et al.,16 using analytical pyrolysis in a batchwise micropyrolyzer. Stepwise pyrolysis of biomass, such as beech, poplar, spruce, and straw, was recently studied by de Wild et al.17 in an auger reactor at temperatures of 250−300 °C (step 1) and 350−400 °C (step 2). Results of these studies are discussed, together with our data, in the “Analysis of Results” section. In the current contribution, we report stepwise pyrolysis experiments of pine wood in a fluidized bed reactor while stretching the temperature range of the first step to 360 °C, as compared to that reported by De Wild and co-workers.17 By taking this wider range, it is anticipated that more information will be obtained about the reaction mechanisms, in particular, char formation pathways. It is discussed how the observed pyrolysis behavior in the stepwise concept can be understood by combining the data with existing theoretical approaches and modifications of those. In addition to having practical implications, the stepwise pyrolysis experiments herein reported were designed to provide information on how important certain processes (reactions and transfer processes) are at certain temperatures within the range from ambient temperature to a typical fast pyrolysis temperature (in our case, 530 °C). Thus, the main goals of this paper are to identify the potential of stepwise pyrolysis for pyrolysis oil improvement, to maximize the production of targeted compounds, and to gain insight into the pyrolysis mechanisms. Received: August 8, 2012 Revised: November 6, 2012 Published: November 6, 2012 7263
dx.doi.org/10.1021/ef301319t | Energy Fuels 2012, 26, 7263−7273
Energy & Fuels
Article
At the end of the first pyrolysis step experiment (260−360 °C), the partial converted biomass (char) left in the reactor was quickly cooled to room temperature by feeding large volumes of cold sand into the fluidized bed reactor in order to freeze further pyrolysis reactions. An important part of the setup in controlling the pyrolysis time of the particles was the water-cooled overflow vessel for char collection, which was kept at 20 °C during the experiments. In this way, immediate cooling of the char after leaving the reactor was achieved. The solid residue (char/partially converted biomass) of the first step was obtained in a mixture with sand, which was stored in a sealed vessel to prevent moisture uptake. Prior to the second pyrolysis step, the solid residue was effectively separated from the majority of the sand by sieving. Five samples of this fraction, containing almost all char/unconverted biomass, were combusted to determine the minerals (ash including sand) content with good reproducibility. This gave the actual solid residue yield of the first step and enabled the calculation of the actual char/partially converted biomass concentration in the feed of the second step. More than 95 wt % of the char obtained in step one was fed into the second step. The char could be fed without size reduction. At the end of every experiment (the 530 °C test), the sand/ char mixture is recovered again and the cumulative char yield was calculated by weight difference (total mass of solids in the system − mass of sand initially fed). The liquids collected in condenser 1, condenser 2, and the intensive cooler were combined and this mixture was used to determine the yield for further calculations and analysis. These liquids are called oil. Organics (yield) were calculated by subtracting the water content from the oils. Produced water was calculated as water in the oils minus the water in the feed. The gas yield was determined by the weight difference of gas flow measured after the condensers and the fed fluidization gas (N2). Product yields of the first and second pyrolysis step were calculated on a dry-ash-free (daf) biomass (fed during the first step) basis. 2.2. Pyrolysis Oil Analysis. The analytical methods used are shown in Table 2. The water content of the pyrolysis oil was quantified by the Karl Fisher titration method. The cold water precipitation method described by Garcia-Perez et al.19 was used to determine the water-insoluble oligomers. The water-insoluble compounds were further extracted by CH2Cl2. The CH2Cl2 soluble compounds are mostly low-molecular-mass, water-insoluble (LMM) compounds and the water insolubles and ether insolubles were ascribed to the highmolecular-mass, water-insoluble (HMM) compounds. The sugar content of the oils was quantified by ion exchange chromatography (IEC) after hydrolysis with sulfuric acid. After neutralization of the solution, the oil was analyzed using a Dionex ICS-3000 system equipped with an AS 50 autosampler, GP50 gradient pump, and ED50 electrochemical detector. The separation of sugars was performed with a Carbopac PA20 column. The mobile phase was an aqueous NaOH solution at a flow rate of 0.50 mL/min. Xylose/ mannose showed a peak overlap and, therefore, could not be analyzed
2. EXPERIMENTAL METHODS 2.1. Stepwise Pyrolysis Tests. Experiments were performed in a 1 kg/h fluidized bed (sand) pyrolysis plant. Details of the experimental setup can be found elsewhere.2,5 The fluidized bed reactor was equipped with an overflow to remove the majority of char from the bed. A knockout vessel and two cyclones collected the remaining char fines entrained from the reactor. The pyrolysis vapors were condensed in two identical wall-cooled counter-current spray columns and one intensive cooler was installed to condense the very light organic compounds. Experimental conditions are listed in Table 1. Pine wood
Table 1. Experimental Conditions operating condition
value(s)
biomass (pine) feeding rate biomass particle size biomass residence time steps 1 and 2 sand particle size U/Umf residence time vapors, hot part Treactor, step 1 Treactor, step 2 temp, condenser 1 temp, condenser 2 temp, intensive cooler experimental run time
1 kg/h ∼1 mm 25 min 220 μm 2.5−3 1.6−1.9 s 260, 290, 310, 330, 360, 530 °C 530 °C 20 °C 20 °C 0 °C 120 min
of ca. 1 mm and 8 wt % moisture was used as feedstock (lignocel 9 purchased from Rettenmaier & Sohne, Germany). The reactor temperature was measured at four different places in the reactor, namely, at the bottom and in the reactor top and two times in the reactor center. The temperature differences between these locations was not more than 5 °C. By controlling the biomass feed rate and sand feed rate, the (average) particle residence time in the well-mixed fluidized bed was set to ca. 25 min. This time is selected to ensure that, even at lowest temperatures, a significant (measurable) conversion is obtained. Simulation results obtained from a single particle model described by Kersten et al.,18 assuming that the heat of reaction is negligible, show that pyrolysis at the pretreatment conditions (first step) proceeds with hardly any temperature gradients inside the particle and that the heating rate is much faster than the reaction rate. Hence, we can say that we have been studying the reactions at the pretreatment temperatures prevailing in the fluidized bed. At 530 °C, which is the fluid bed temperature, the heating time of the particles is 300 °C). The reactions running in the range of 310−360 °C have a profound effect on char formation in the stepwise approach; more cumulative char and water is produced at the expense of organic volatiles. The presence of reactive (liquid) intermediates (shown by water/acetone extraction of the char/ unconverted biomass) at these temperatures is expected to have an important impact on the outcome of pyrolysis reactions. A mechanism is proposed that includes several stages of the reacting solid and in which the main emphasis lies on the competition between routes that lead to char formation and routes that release compounds from the biomass particle. From an engineering and cost viewpoint, it can be interesting to pyrolyze the biomass at a low temperature, followed by grinding and pyrolysis in a second step at higher temperatures. This will be a subject for further research.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Dr. Garcia-Perez and Mr. Zhouhong Wang are very grateful to the U.S. National Science Foundation (No. CBET-0966419), the Sun-Grant Initiative (Interagency Agreement No. T0013GA), and the Washington State Agricultural Research Center for their financial assistance.
■
REFERENCES
(1) Westerhof, R. J. M. K., S.R.A.; Nygard, H. S.; van Swaaij, W. P. M.; Brilman, D. W. F. Effect of particle geometry and microstructure on fast pyrolysis of beech wood. Energy Fuels 2012, 26 (4), 2274− 2280. (2) Westerhof, R. J. M.; Brilman, D. W. F.; van Swaaij, W. P. M.; Kersten, S. R. A. Effect of Temperature in Fluidized Bed Fast Pyrolysis of Biomass: Oil Quality Assessment in Test Units. Ind. Eng. Chem. Res. 2009, 49 (3), 1160−1168. (3) Garcia-Perez, M.; Wang, X. S.; Shen, J.; Rhodes, M. J.; Tian, F.; Lee, W.-J.; Wu, H.; Li, C.-Z. Fast Pyrolysis of Oil Mallee Woody 7272
dx.doi.org/10.1021/ef301319t | Energy Fuels 2012, 26, 7263−7273
Energy & Fuels
Article
(25) Molton, P. M.; Demmitt, T. F. Reaction mechanisms in cellulose pyrolysisA literature review; Battelle−Pacific Northwest Laboratories: Richland, WA, 1977. (26) Mok, W. S. L.; Antal, M. J. Effects of pressure on biomass pyrolysis. II. Heats of reaction of cellulose pyrolysis. Thermochim. Acta 1983, 68 (2−3), 165−186. (27) Lin, Y.-C.; Cho, J.; Tompsett, G. A.; Westmoreland, P. R.; Huber, G. W., Kinetics and Mechanism of Cellulose Pyrolysis. J. Phys. Chem. C 2009, 113 (46), 20097−20107. (28) Piskorz, J.; Majerski, P.; Radlein, D.; Vladars-Usas, A.; Scott, D. S. Flash pyrolysis of cellulose for production of anhydro-oligomers. J. Anal. Appl. Pyrolysis 2000, 56 (2), 145−166. (29) Hoekstra, E.; Westerhof, R. J. M.; Windt, M.; Brilman, W.;Van Swaaij, W. P. M.; Kersten,S. R. A.; Hogendoorn, K. J. A. Heterogeneous and homogeneous reactions of pyrolysis vapors from pine wood. AIChE J. 2011, 58 (9), 2830−2842 (DOI: 10.1002/ aic.12799). (30) Sharma, R. K.; Wooten, J. B.; Baliga, V. L.; Lin, X.; Chan, W. G.; Hajaligol, M. R. Characterization of chars from pyrolysis of lignin. Fuel 2004, 83 (11−12), 1469−1482. (31) Scheirs, J.; Camino, G.; Tumiatti, W. Overview of water evolution during the thermal degradation of cellulose. Eur. Polym. J. 2001, 37 (5), 933−942. (32) Zhang, J.; Luo, J.; Tong, D.; Zhu, L.; Dong, L.; Hu, C. The dependence of pyrolysis behavior on the crystal state of cellulose. Carbohydr. Polym. 2010, 79 (1), 164−169. (33) Oja, V.; Suuberg, E. M. Vapor Pressures and Enthalpies of Sublimation of d-Glucose, d-Xylose, Cellobiose, and Levoglucosan. J. Chem. Eng. Data 1998, 44 (1), 26−29. (34) Mamleev, V.; Bourbigot, S.; Le Bras, M.; Yvon, J. The facts and hypotheses relating to the phenomenological model of cellulose pyrolysis: Interdependence of the steps. J. Anal. Appl. Pyrolysis 2009, 84 (1), 1−17. (35) Lede, J. Diebold, J. P.; Peacocke, G. V. C.; Piskorz, J. Fast Pyrolysis of Biomass: A Handbook. CPL Press: Newbury, U.K., 1999. (36) Nowakowski, D. J.; Bridgwater, A. V.; Elliott, D. C.; Meier, D.; de Wild, P. Lignin fast pyrolysis: Results from an international collaboration. J. Anal. Appl. Pyrolysis 2010, 88 (1), 53−72. (37) Haas, T. J.; Nimlos, M. R.; Donohoe, B. S. Real-Time and Postreaction Microscopic Structural Analysis of Biomass Undergoing Pyrolysis. Energy Fuels 2009, 23 (7), 3810−3817. (38) Shafizadeh, F.; Chin, P. P. S. Thermal deterioration of wood. In Wood Technology: Chemical Aspects; ACS Symposium Series 43; American Chemical Society: Washington, DC, 1977; pp 57−81. (39) Chaiwat, W.; Hasegawa, I.; Tani, T.; Sunagawa, K.; Mae, K. Analysis of Cross-Linking Behavior during Pyrolysis of Cellulose for Elucidating Reaction Pathway. Energy Fuels 2009, 23 (12), 5765− 5772. (40) Antal, M. J. Biomass pyrolysis: A review of the literature, Part I, Carbohydrate pyrolysis. Adv. Solar Energy 1982, 1, 61−111. (41) Oasmaa, A.; Kuoppala, E. Solvent Fractionation Method with Brix for Rapid Characterization of Wood Fast Pyrolysis Liquids. Energy Fuels 2008, 22 (6), 4245−4248. (42) Antal Jr, M. J., Biomass Pyrolysis: A Review of the Literature, Part 2Lignocellulose Pyrolysis. Plenum Press: 1985; Vol. 2, p 175−255. (43) Bridgwater, A. V.; Meier, D.; Radlein, D. An overview of fast pyrolysis of biomass. Org. Geochem. 1999, 30 (12), 1479−1493. (44) Sunagawa, K.;Hasegawa, I.; Mae, K., A new kinetic model for cellulose pyrolysis based on changes of its macromolecule structure. In Science in Thermal and Chemical Biomass Conversion, Bridgwater, A. V., Boocock, D. G. B., Eds.; CPL Press: Newbury, U.K., 2006. (45) Lédé, J. Cellulose pyrolysis kinetics: An historical review on the existence and role of intermediate active cellulose. J. Anal. Appl. Pyrolysis 2012, 94 (0), 17−32. (46) Shafizadeh, F. J. Appl. Polym. Sci. Symp. 1975, 28, 153. (47) Yu, Y.; Wu, H. Characteristics and Precipitation of Glucose Oligomers in the Fresh Liquid Products Obtained from the Hydrolysis of Cellulose in Hot-Compressed Water. Ind. Eng. Chem. Res. 2009, 48 (23), 10682−10690.
(48) Koufopanos, C. A.; Maschio, G. Can. J. Chem. Eng. 1989, 67 (1), 75−84. (49) Di Blasi., C. Comparison of semi-global mechanisms for primary pyrolysis of lignocellulosic fuels. J. Anal. Appl. Pyrolysis 1998, 47 (1), 43−64. (50) Prins, W.; van Swaaij, W. P. M. Heat Transfer to Immersed Spheres Fixed or Freely Moving in a Gas-Fluidized Bed; Van Swaaij, W. P. M., Ed.; Hemisphere Publishing Corporation: Washington, DC, 1986; pp 317−331. (51) Wagenaar, B. M.; Prins, W.; van Swaaij, W. P. M. Flash pyrolysis kinetics of pine wood. Fuel Processing Technology 1993, 36, 291−298. (52) Chan, W.-C. R.; Kelbon, M.; Krieger, B. B. Modelling and experimental verification of physical and chemical processes during pyrolysis of a large biomass particle. Fuel 1985, 64 (11), 1505−1513. (53) Wang, X.; Kersten, S. R. A.; Prins, W.; van Swaaij, W. P. M. Biomass Pyrolysis in a Fluidized Bed Reactor. Part 2: Experimental Validation of Model Results. Ind. Eng. Chem. Res. 2005, 44 (23), 8786−8795.
7273
dx.doi.org/10.1021/ef301319t | Energy Fuels 2012, 26, 7263−7273
