Detailed Chemical Kinetic Modeling of Vapor-Phase Reactions of

Subscriber access provided by UNIV OF MISSISSIPPI

Article

Detailed Chemical Kinetic Modeling of Vapor-Phase Reactions of Volatiles Derived from the Fast Pyrolysis of Lignin Hua-Mei Yang, Srinivas Appari, Shinji Kudo, Jun-ichiro Hayashi, and Koyo Norinaga Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.5b01289 • Publication Date (Web): 18 Jun 2015 Downloaded from http://pubs.acs.org on July 1, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Industrial & Engineering Chemistry Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Detailed Chemical Kinetic Modeling of Vapor-Phase Reactions of Volatiles Derived from Fast Pyrolysis of Lignin Hua-Mei Yang,† Srinivas Appari‡ Shinji Kudo,‡ Jun-ichiro Hayashi‡,§ and Koyo Norinaga*,‡ ‡ Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Kasuga, 816-8580, Japan. ‡ Institute for Materials Chemistry and Engineering, Kyushu University, Kasuga, 816-8580, Japan. §

Research and Education Centre of Carbon Resources, Kyushu University, Kasuga, 816-8580, Japan. ABSTRACT The vapor-phase reactions of nascent volatiles derived from the fast pyrolysis of lignin were investigated both experimentally and numerically. Lignin residue after enzymatic hydrolysis was pyrolyzed in a two-stage tubular reactor at 773–1223 K. The nascent volatiles formed in the first stage underwent vapor-phase reactions in situ in the second stage. A detailed chemical kinetic model that consists of more than 500 species and 8000 elementary reactions was used to simulate the vapor-phase reactions of volatiles derived from fast pyrolysis of lignin. The contribution of tar in the vapor-phase reactions was considered in global reactions, which improves the predictive capability of the kinetic model. The experimental data and numerical predictions of 31 products were compared and analyzed to understand the mechanism of vapor-phase reactions of lignin. The model predictions were in good agreements with the experimental observations. Reaction pathway analysis for the formation of aromatic hydrocarbons is performed. Cyclopentadienyl radicals produced from phenol decomposition were suggested as important intermediates in the formation of aromatic hydrocarbons during lignin pyrolysis. KEYWORDS: lignin; fast pyrolysis; elementary reactions, mechanism. 1. INTRODUCTION Owing to its renewability, carbon neutrality, and low sulfur content, biomass has the potential to help relieve the energy and environmental crisis.1-4 Biomass can be converted into heat, fuels, and chemicals through pyrolysis, gasification, liquefaction, and combustion.1, 5 Pyrolysis is always involved in biomass thermal conversion processes as either the entire process or an important step to cause fragmentation of the biomass structure.6, 7 Thus, the pyrolysis of biomass has been an attractive research subject in recent years. The effects of pressure, temperature, heating rate, heat and mass transfer, and reactor configuration on the product distributions, product upgrading, and the chemistry and kinetics have been thoroughly reviewed.2, 5, 8-11 During pyrolysis, typically 75–90 wt% of biomass is converted into volatiles at temperatures higher than 773 K. Volatiles escaped from biomass particles, and continue to crack, combine, or condense in vapor phase. This process is named as vapor-phase reactions of volatiles which is unavoidable during biomass pyrolysis.6, 7, 12 This high conversion into volatile products suggests that vapor-phase reactions have an important role during biomass pyrolysis,

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 27

influencing the formation of the final products. However, there are very limited studies on the chemistry and kinetics of the vapor-phase reactions of biomass. The effect of vapor-phase reactions on the final product distribution is still not well understood. A wide variety of volatiles are generated from the pyrolysis of biomass, indicating that a large number of reaction pathways exist in the vapor-phase reactions of volatiles derived from biomass pyrolysis. The complexity of biomass pyrolysis has been a barrier to understanding the formation mechanisms of target products. Studying the pyrolysis behaviors of the individual components of biomass, such as cellulose, hemicellulose, and lignin, would simplify the volatiles obtained and the available reaction pathways. Cellulose, which is the most abundant component in biomass, has a polymeric structure, and its pyrolysis behavior has been investigated widely.7, 13-18 We have previously investigated the mechanism of the vapor-phase reactions of the volatiles generated from cellulose pyrolysis experimentally and numerically with a detailed chemical kinetic model (DCKM).7, 18 Light hydrocarbons derived from cellulose, such as propylene, butane, cyclopentadiene, and propadiene, were proposed to be important precursors in the formation of aromatic hydrocarbons during cellulose pyrolysis. Compared with cellulose, lignin has a complex structure formed by the polymerization of three phenyl propane monomers, i.e., guaiacyl, syringyl, and p-hydroxyphenyl units.19 Research efforts have been focused on understanding the primary pyrolysis of lignin, including the cleavage of the linkages between the aromatic units and the formation of phenols by free radical reactions, concerted reactions, and rearrangement reactions.20-26 However, only a few studies have focused on the vapor-phase reactions of volatiles derived from lignin pyrolysis. Zhou et al. investigated the effect of vapor-phase reactions on the yield and composition of lignin oligomers obtained from the fast pyrolysis of pine wood and organosolv lignin;12, 27 however, they did not investigate the kinetic mechanism. Jegers et al.6 and Caballero et al.28 investigated the product distribution obtained after the secondary thermal decomposition of lignin and checked the kinetic with a lumped kinetic model. However, lumped kinetic models are established based on global product categories (char, tar or bio-oil, and gases), which limits the validity of this model to provide a full description of the formation mechanisms for specific products. As mentioned above, a DCKM successfully reproduced the vapor-phase reactions of the volatiles derived from cellulose pyrolysis.7, 18 This is a powerful tool for understanding the interactions between molecules and reactive intermediates at the molecular level. Ranzi et al.29 employed a detailed kinetic model for the pyrolysis and oxidation of hydrocarbon fuels to simulate the main kinetic features of biomass pyrolysis. Some detailed kinetic models were also developed to simulate the pyrolysis of phenols.30-32 To date, only Faravelli et al.33 proposed a DCKM for the pyrolysis of lignin. This model, which involves about 100 molecular and radical species with 500 elementary and lumped reactions, was used to predict the degradation rates of lignin and product compositions. The DCKM mainly considered the breaking of linkages between aromatic units, but neglected the vapor-phase reactions of


2

Page 3 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


non-aromatic compounds, such as butadiene, cyclopentadiene, and acetylene, which are important products obtained from the pyrolysis of lignin model compounds.34-41 Therefore, a more detailed model is necessary to reproduce the vapor-phase reactions of volatiles derived from lignin pyrolysis. In this work, we employed a DCKM to simulate the vapor-phase reactions of volatiles derived from lignin pyrolysis. First, lignin was pyrolyzed in a two-stage tubular reactor with temperatures ranging from 773 to 1223 K and residence times of volatiles ranging from 0.1 s to 3.6 s. Then, the vapor-phase reactions of volatiles derived from lignin pyrolysis was simulated with a DCKM coupled with batch reactor simulations and validated with experimental observations. Finally, a reaction pathway analysis was proposed to understand the formation of aromatic hydrocarbons during lignin pyrolysis. 2. EXPERIMENTAL SECTION The lignin sample used in this study was prepared by enzymatic hydrolysis of empty fruit bunches and is hereafter referred to as EHL. EHL has a structure similar to that of lignin in the original biomass because it was exposed to less severe chemical reactions.42 The preparation process has been previously reported,43 and is briefly described here. Hemicellulose was first removed by hydrothermal treatment. After filtration and washing with water, the pretreated solid was enzymatically hydrolyzed to obtain lignin residue. Filtration and washing of the lignin residue yielded the lignin used in this study. The content of lignin in EHL was 96 wt%; the elemental composition and chemical structure characterized by 13C NMR analysis of EHL have been previously reported.43 The sample used in this study had particle sizes of 75–150 µm and was dried under vacuum overnight before the use. EHL (1 mg) was pyrolyzed in a two-stage tubular reactor (TS-TR) connected to a gas chromatograph (GC), which has been previously described in detail.7, 18 TS-TR was made of quartz and consists of two zones divided by a quartz wool filter. First zone was for the primary pyrolysis, producing char and volatiles. Secondary zone was for the vapor-phase reactions of volatiles. TS-TR was heated by an electric furnace. 1 mg EHL was wrapped by a 10 mm * 10 mm sheet of stainless (SUS316) wire mesh with mesh opening of 45 µm. The prepared sample was fixed at the upper part of the TS-TR where the maximum temperature was lower than 343 K during the experiment. EHL was not converted at temperature lower than 343 K when waiting for the stabilization of the furnace temperature and the GC baseline. The sample was dropped into the bottom of first isothermal zone (the primary pyrolysis zone) after the furnace temperature and the GC baseline was stable. Fast pyrolysis of EHL in the primary pyrolysis zone of the TS-TR generated nascent volatiles which were involved in the vapor-phase reactions in the secondary isothermal zone. Volatiles derived from EHL primary pyrolysis were flown to secondary isothermal zone with carrier gas at 241 kPa. Through adjusting the length of TS-TR in furnace, the residence time of vapor-phase reactions of volatiles could be varied from 0.1 s to 3.6 s. Hereafter, tr refers to the residence time of vapor-phase reactions in TS-TR. Pyrolysis temperature was changed from 773 K to 1223 K with tr = 0.1 – 3.6 s. Passed the secondary isothermal zone, volatiles were carried with carrier gas to the GC column and further identified by the


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 27

GC detector. The composition of the volatiles was measured directly by two GCs equipped with TCD and/or FID. To detect the products in detail, four different columns were employed. The chromatographic conditions are listed in Table 1. In this study, fast volatilization could be realized in the first isothermal zone, and well-resolved chromatograms were obtained at entire range of the pyrolysis temperature without suffering peak broadening. The typical gas chromatograms along with peak assignments can be found in the supporting information (Figure S1-S5). When standard compounds for the target products were available, the quantification of the products was conducted using an external calibration method to estimate the amount of analyte in a separately analyzed sample. When standards for the products were unavailable, the effective carbon number concept was employed to estimate the amount of analyte.44, 45 The detailed procedure has been described elsewhere.18 The char yield was determined by measuring its mass, and was defined as the ratio of char mass to the mass of feed sample. In this paper, tar was defined as a group of phenols (except phenol, cresols, and catechol) and the undetectable products that could not be introduced into GC detector and quantified by the current GC. Tar yield was defined as the sum of yield of phenols (except phenol, cresols, and catechol) and the yield of undetectable products. The yield of undetectable products was calculated as followed: Yield of undetectable products = 100% – char yield – sum of all the detected products yield. 3. DETAILED CHEMICAL KINETIC MODELING In the simulation, it was assumed that EHL produced volatiles spontaneously and instantaneously under the rapid heating rate at first stage. The volatile cloud was considered to be a batch of multicomponent mixture and underwent vapor-phase reactions at a fixed residence time in the second stage of the reactor. The residence time for the vapor-phase reactions of the molecules contained in the volatiles was thus determined based on the volume of the second reactor and flow rate of the carrier gas. It was possible to consider this situation as a problem to solve time dependent change in the composition of the multicomponent gas mixtures at isothermal condition. Thus we chose a batch reactor model and used the DETCHEMBATCH code to run the DCKM. The DETCHEMBATCH code is designed to simulate the time-dependent homogeneous gas-phase reactions in a batch reactor.7, 18, 46 Reaction rate parameters and thermodynamic polynomials for all participating species, and mass fraction of feeding species were prepared as the input files for the numerical simulation. Numerical simulations were performed at isobaric and isothermal conditions. The DCKM consists of 548 species with 8159 elementary step-like reactions without any modifications in the kinetic parameters. To offer a comprehensive kinetic model for the biomass pyrolysis, the reaction mechanism of the DCKM was kept same with our previous studies on cellulose pyrolysis7, 18 and coffee extraction residue47. The DCKM is still not sufficient to describe the complex phenomena of solid fuel conversion. For example, the knowledge on the detailed decomposition pathways of phenols (such as guaiacol, syringol, catechol, and their derivatives) is still limited, and elementary reactions of these phenols are not involved in the DCKM. So the addition of global reaction (a non-elementary step reaction) is very helpful to explain the contribution of tar (a sum of phenols and undetectable products)


4

Page 5 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


in the vapor-phase reactions of lignin pyrolysis. In this study, global reactions of tar decomposition were developed in the temperature range of 923–1223 K and coupled with the DCKM. The global reactions were developed with the similar method in our previous paper.48 The global reactions were assumed as first order reaction. Stoichiometric coefficient of each product in the global reaction was determined by the mole difference between experimental observations and the simulation results without global reactions at tr = 3.6 s. Soot was assumed in the global reaction as the final product which was used to meet the elemental balance of the global reaction. The rate coefficient (k) is a fitting parameter, and estimated empirically to minimize the gaps between the predicted and experimental results of the major species composition. More detailed information on the development of the global reactions is given in sections 4.2 and 4.3. 4. RESULTS AND DISCUSSION 4.1 Molecular Composition of the Nascent Volatiles Derived from EHL Pyrolysis The molecular composition of the nascent volatiles derived from EHL pyrolysis was used as the boundary condition for the numerical simulation of the vapor-phase reactions using the DCKM. It was difficult to determine the molecular composition of the products formed during the primary pyrolysis of EHL without the effect of vapor-phase reactions. Here, we assumed the molecular compositions of the volatiles derived from EHL pyrolysis at tr = 0.1 s as the nascent volatiles. The char yield and the compositions of nascent volatiles (obtained at 0.1 s) at 773, 923, 1023, 1123, and 1223 K are provided in the Supporting information (Table S1). The yield of char was reduced from 34 to 24 wt% with increasing temperature. In total, 61 products were quantified by GC. 45 products were grouped in “light products”. The total mass fraction of the light products varied from 22 to 68 wt% of the EHL samples fed at 773–1223 K. The detailed analysis of light products will be discussed in section 4.3. As described above, there is a lack of decomposition mechanism of phenols in the DCKM, hence, 16 phenols as well as undetectable products were grouped as “tar”. Yield of phenols in tar reduces with temperature increasing, which was consistent with the results of Patwardhan et al.49 and Jiang et al.26 There are almost no methoxyphenols in tar at temperature higher than 1023 K. The total mass fraction of undetectable products varied with temperature and was highest at 773 K and lowest at 1223 K. More complex phenols, oligomers and polycyclic aromatic hydrocarbons that were undetectable using the current GC conditions could account for the majority of missing mass of undetectable products. Based on the elemental balance between the EHL feed and the products (including char and the light products), the elemental composition of tar at 0.1 s was estimated at 773–1223 K, as shown in Table 2. The ratio of carbon and hydrogen (C:H) of tar first goes down and then up again. At low temperature, oligomers, phenol, guaiacol, syringol, and their derivatives were released from lignin after breakages of weak bonds such as β-O-4 and α-O-4, and C-C bonds such as Cβ-β, Cβ−5 and C5−5 in EHL.26, 27 With temperature increasing, oligomers and phenols were decomposed to small fragments and light products such as light hydrocarbons (LHs). It makes the H content of the light products increases and the H content of tar declines, leading to the C:H ratio of tar going down. At


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 27

the same time, the combinations of light products were promoted with temperature increasing, leading to the formation of heavy-weight molecules which could not be detected by the GC. This could be proved by obvious secondary tar/soot attached in the reactor wall of secondary isothermal zone above 1023 K in our experiment. The formation of secondary tar/soot caused the H content of the light products declines and the H content of tar increased, leading to the C:H ratio of tar going up at 1223 K as shown in Table 2. 4.2 Development of Global Reaction The DCKM used in this study has been validated by our group to predict the vapor-phase reactions of volatiles derived from cellulose pyrolysis.7, 18 In this system, the missing products (tar) in the nascent volatiles were assumed as levogulcosan, which is a major product in cellulose pyrolysis. The model was shown to have an acceptable capability for predicting the concentration profiles of the major products, such as hydrogen, carbon monoxide, carbon dioxide, and methane, in the vapor-phase reactions of cellulose at 923, 1023, and 1123 K. Unlike the simple structure of cellulose, lignin has a more complex polymeric structure, and its pyrolysis products consist of a wide variety of phenols. From 44 to 8 wt% of tar (Table 2) generated from primary pyrolysis of EHL are still not involved in the species used in the DCKM. Since the DCKM is still under development, there is a lack of detailed decomposition mechanism of phenols, 16 phenols detected in this study were considered as a part of tar in the simulation in this study. Thimthong et al. added a global reaction to understand the significance of the missing products on the product distribution obtained for the partial oxidation of nascent volatiles derived from fast pyrolysis of woody biomass, and good agreement was achieved between the experimental observation and predicted results.48 Here, we employed a similar approach to reproduce the vapor-phase reactions of volatiles derived from EHL pyrolysis with the DCKM added a global reaction for the contribution of tar in the vapor-phase reactions. Model predictions without a global reaction were conducted at 923 K by considering the tar and char as inert substances, such as nitrogen. These simulation results at tr = 0.1–3.6 s are illustrated as dashed lines in Figure 1. The numerical predictions without a global reaction underestimated the yields of most of the products, such as carbon monoxide (CO), carbon dioxide (CO2), methane (CH4), ethene (C2H4), propene (C3H6), 1,3-butadiene (C4H6), cyclopentadiene (C5H6), benzene (C6H6), and phenol (C6H5OH). This under-estimation was caused by the absence of the tar at 923 K.26 By considering these compounds generated from tar, a global reaction scheme was estimated as listed in Table 3. The stoichiometric coefficients of each species in the global reaction were estimated by the mole difference at tr = 3.6 s between the experimental value and predicted value without the global reaction. The k of the global reactions at 923 K was estimated as 0.25 s-1 empirically by fitting the experimental observations and predicted results. The model predictions with the global reaction are also illustrated in Figure 1, as shown by the solid lines. Good agreement between the model predictions and experimental data was obtained at 923 K for all the major products illustrated in Figure 1. The global reaction was also tested at 1023 K. Unlike at 923 K, the predictions for the C2–C3


6

Page 7 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


hydrocarbons agreed well with the experimental observations, indicating that most of the substituents of the lignin units were broken quickly in the primary pyrolysis at 1023 K. The yields of CO, CO2, H2O, H2, CH4, C6H6, and C6H5OH were underestimated at 1023 K with tr ranging from 0.1s to 3.6 s by the model without a global reaction, as illustrated by the dashed lines in Figure 2. With the global reaction listed in Table 3, the model predictions (Figure 2, solid lines) were in good agreements with the experimental observations (Figure 2, points) at 1023 K. 4.3 Numerical Simulation of the Vapor-Phase Reactions as a Function of Temperature Numerical simulations of the vapor-phase reactions were conducted at 773–1223 K. The molecular compositions of volatiles (at 0.1 s) derived from the fast pyrolysis of EHL (Table S1) were served as the species boundary. At 773 K, the vapor-phase reactions were not so drastic, and the products yield could be predicted well without the addition of global reaction. Similar to the cases at 923 and 1023 K, the global reaction (Table 3) was considered to improve the capability of DCKM at 1123 and 1223 K. As shown in Table 3, k of global reactions declines with temperature increasing from 923 K to 1123 K , and increases above 1123 K. This would be caused by the different compositions of tar at different temperature. At lower temperature, the compositions of tar were mainly phenols and oligomers. k of global reactions decreases with increasing temperature up to 1123 K likely because more stable phenols or unreactive products are included in “tar” with temperature increasing from 923 K to 1123 K.12, 26, 27 Above 1123 K, the temperature effect on k would be larger than the effect of tar reactivity, leading to the increase of k. As shown in Table 3, more H2 was generated from tar decomposition based on the global reaction in order to predict the H2 yield at higher temperature. The elemental compositions of soot were also shown in Table 3. The C:H ratio of soot in the global reaction goes down with temperature increasing, which was mainly caused by the condensation of carbon structure with the release of H2. The yield profiles of inorganic compounds (IGs), light hydrocarbons (LHs), light oxygenated compounds (LOCs), phenols, and aromatic hydrocarbons (AHs) are summarized in Figures 3–7. The lines and symbols represent the predictions and the experimental data, respectively, in the temperature range of 773–1223 K with tr = 3.6 s. Comparisons of 31 species showed that good agreement was observed between the model predictions and experimental observations of not only the major products but also the minor compounds. In other words, the DCKM along with global reactions at different temperatures had a high capability of reproducing the vapor-phase reactions of volatiles derived from the fast pyrolysis of lignin. IGs, including CO, CO2, H2O and H2, are main products above 773 K. Figure 3 showed the experimental and predicted yield profiles of IGs obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s. The predicted results agree well with the experimental observations for the IGs. The yield of IGs (excluding H2O) increases as temperature increases. The yield of CO is much lower than CO2 and H2O below 923 K. Formation of CO was significantly enhanced above 923 K, and its yield increases sharply, much more than that of CO2 and H2O above 1023 K. Deoxygenation of light oxygen compounds in


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 27

vapor-phase reactions is one pathway for the formation of CO. Pyrolysis of lignin model compounds32, 34-37, 50 suggested that the decomposition of phenols also contributed to the formation of CO which will be proved by the reaction pathway analysis in section 4.4. With temperature increasing, the yield of CO2 slightly increases, while the yield of H2O reduces with the temperature increasing. CO2 was mainly eliminated from the primary pyrolysis of EHL and derived from the breakages of substituents on aromatic rings at lower temperature.23 H2O was formed by the combination of H and OH. OH was mainly generated from the phenolic hydroxyl group during lignin pyrolysis. With temperature increasing, water-gas shift reaction (CO + H2O → H2 + CO2) could account for the H2O consumption and CO2 formation according to the reaction pathway analysis of DCKM. In the current temperature range, the equilibrium constant of the shift reaction ranges from 0.7 to 5.00. However the value of P(H2)*P(CO2)/(P(CO)*P(H2O) that can be obtained based on the experimental results ranges from 0.3 and 1.0, indicating the system is far from the equilibrium. The enhanced formation of CO at higher temperature can promote the rate of water consumption according to the shift reaction at conditions far from the chemical equilibrium. This also supports that the shift reaction is the major reaction for water consumption. Predictions of C1-C5 LHs obtained from EHL pyrolysis with DCKM are in good agreement with experiment observations at 773−1223 K with tr = 3.6 s as shown in Figure 4. C1-C4 alkanes and C2-C3 alkenes have been widely investigated in the fast pyrolysis of lignin. CH4 is the most abundant LH and its yield increases remarkably with increasing temperature. Main sources for CH4 formation are the breakage of oxygen-carbon weak bond in methoxyl phenols and the cracking of small fragments. Each C2-C3 LHs, excluding acetylene (C2H2) showed its maximum yield at 923 or 1023 K. The breakage of substituents in lignin units at lower temperature and the cracking of C4–C6 LHs at higher temperature could contribute to the formation of C2-C3 LHs. Above 1023 K, the combination of these products was enhanced, causing decrease in their yields with temperature increasing. Different from other LHs, the yield of C2H2 increases with temperature increasing, which is consistent with the results obtained with the pyrolysis of lignin model compounds. 31, 35, 36 Unlike the C1–C4 alkanes and C2–C3 alkenes previously observed to be formed during the fast pyrolysis of lignin,6, 51-53 several LHs were newly detected and investigated during the fast pyrolysis of EHL in this study, including propadiene, 1-butene-3-yne, 1,3-butadiene, cyclopentadiene, 1,4-pentadiene, and 1,2-pentadiene. Propadiene, 1,3-butadiene, and cyclopentadiene have been reported to be important intermediates in the formation of AHs from lignin model compounds, such as phenol, eugenol, and catechol.34-39 Phenol is an important precursor of C5 LHs, such as cyclopentadiene, whereas methoxy- or hydroxyphenol favor the formation of C4 LHs, such as 1,3-butadiene.34, 35, 40, 41 As shown in Figure 4, the LHs as a function of temperature are predicted well by the model in the temperature range of 773–1223 K. Six LOCs generated from EHL pyrolysis were also used to validate the model at 773−1223 K with tr = 3.6 s as shown in Figure 5. The numerical predictions are in good agreement with the experimental


8

Page 9 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


observations. Methanol, acetic acid, and acetaldehyde are three main oxygen compounds at different temperatures. LOCs are mainly formed from the primary pyrolysis of EHL. After breaking the oxygen-carbon bonds (such as β-O-4 bond) between lignin units, it is easy to break substituents from benzene ring to form LOCs.20 The yields of LOCs reduce obviously above 923 K. LOCs are main feedstock for the vapor-phase reactions during lignin pyrolysis, and condensed into secondary tar or decomposed into IGs and LHs through deoxygenation after 923 K. Phenols are one of main chemical groups produced from lignin pyrolysis. Considering the model, the prediction yield profile of phenol, catechol, and cresols were compared with experimental observation in Figure 6. The predicted yields of phenol are in good agreement with experimental observations. The yield of catechol was over-estimated which would be caused by limited reactions of the decomposition of catechol in DCKM. In contrast, the yield of cresols was under-estimated at 923 K. As mentioned above at low temperature, the composition of tar would be substituent phenols, which can contribute to the formation of cresols. In this study, cresols were not considered in the global reaction of tar, causing the under-estimation of cresols. As mentioned above, 16 phenols were considered as tar since there is a lack of detailed mechanism of phenols in the DCKM. Now, our group is investigating the pyrolysis mechanisms of catechol, guaiacol and their derivatives, with an expectation to well predict the phenols yield during lignin pyrolysis. In this paper, the yields of the 16 phenols were not considered in detail. Figure 7 illustrates the experimental and predicted yield profiles of major AHs obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s. Good agreements were obtained between the model prediction and experimental observation of benzene, toluene, styrene, ethylbenzene, and naphthalene. The yield of indene was obviously over-estimated as shown in Figure 7, which would be accounted for by the uncertainty of the rate parameters of reactions for the formation of indene. With the increase in temperature, all the AHs yields shown in Figure 7 reduced, which was caused by the formation of secondary char or soot at high temperature. It could be proved by our experimental observation that there was obvious tar or soot attached on the reactor wall of secondary isothermal zone above 1023 K. 4.4. Reaction Pathway Analysis AHs are the major constituents of the tar obtained from the thermochemical conversion of biomass. The presence of AHs in fuels could cause engine problems and environmental pollution. The good agreement between the simulation results and experimental observation encouraged us to analyze the reaction pathways based on the DCKM. Figure 8 shows the reaction pathway analysis for the important reactions leading to the formation of AHs in EHL pyrolysis at 1023 K. The arrows represent the formation of a species, and the arrow thickness indicates the contribution of a reaction to the formation of a specific species. The reaction flux diagram in Figure 8 shows that the formation of cyclopentadienyl radical (C5H5) was an important bridge for the formation of aromatic hydrocarbons from lignin pyrolysis, which will be described in detail as the following part. The vapor-phase reactions of volatiles generated from cellulose have been studied in our previous


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 27

study, and the reaction pathways leading to benzene and naphthalene were discussed.7, 18 Cellulose is originally free from aromatic structures, while EHL is rich in aromatic structure. But it was noted that the reaction pathway of naphthalene formation from cellulose pyrolysis and lignin pyrolysis was similar. Combination of C5H5 is a dominant route to naphthalene in both cellulose pyrolysis and lignin pyrolysis. Different from the formation of C5H5 through combination of acetylene with propyne or allyl radical in cellulose pyrolysis, C5H5 were mainly generated from the decomposition of phenols (phenol and cresol shown in Figure 8) in lignin pyrolysis. Toluene was mainly formed from the combination of benzyl radical (C7H7) with H as shown in Figure 8. The formation of indene is also shown in Figure 8, and indene was mainly formed from C7H7 by the combination with C2H2. In lignin pyrolysis, C7H7 was mainly formed from the combination of C5H5 and C2H2. The combination of propargyl radical (C3H3) and vinylacetylene also contributed to the formation of C7H7 in lignin pyrolysis. Based on the investigation of pyrolysis of phenolic model compounds, 1,3-butadiene (a precursor of vinylacetylene as shown in Figure 8) were mainly generated from methoxy- or hydroxyphenols.34, 35, 40, 41 Since there is a lack of detailed decomposition mechanism of phenols in the DCKM, the formation of 1,3-butadiene from phenols is not shown in Figure 8. Different from the lignin pyrolysis, the combination of C3H3 with vinylacetylene was the main pathway for the formation of C7H7 in cellulose pyrolysis. Vinylacetylene was mainly formed from the combination of C2H2 or C2H4 during cellulose pyrolysis. Benzene was mainly formed from two types of reactions in lignin pyrolysis. The first type was H substitution reactions of phenol and toluene. The second type involved the combination of small hydrocarbons, including the combination of propadiene or propyne with C3H3 and the combination of C5H6 with C2H4 with the concomitant formation of a methyl radical. C5H6 was generated from the combination of C5H5 with H, and the combination of C2H2 with C3H3, and C5H6 was also directly formed from the decomposition of phenol. Since cellulose is free from aromatic structure, the formation of benzene was mainly formed from the combination of propadiene or propyne with C3H3 and H substitution reactions of toluene which was formed as mentioned above. Different composition of volatiles derived from lignin pyrolysis and cellulose pyrolysis caused different reaction pathway in the vapor-phase reactions of lignin pyrolysis and cellulose pyrolysis. And close agreements for most of the compounds between prediction and experiment observations were obtained in both cellulose pyrolysis and lignin pyrolysis. In another word, the DCKM is competent to deal with the reactions of multi-component mixtures. It is expectable in the future work to extend DCKM to predict the vapor-phase reactions of biomass pyrolysis which always generates complex volatiles. 5. CONCLUSIONS Vapor-phase reactions of volatiles decide the final products distribution during the pyrolysis of biomass or biomass materials. To investigate the vapor-phase reactions of volatiles derived from lignin pyrolysis, EHL prepared by enzymatic hydrolysis was pyrolyzed in a TS-TR connected to GC. A


10

Page 11 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


DCKM was used to predict the vapor-phase reactions of nascent volatiles generated from EHL pyrolysis in the temperature range from 773 to 1223 K with tr varying from 0.1 to 3.6 s. Because from 43.5 to 8.2 wt% tar were obtained at different temperatures, global reactions were developed and used with the DCKM to improve the model prediction capability. The model was evaluated through the comparison of the numerically predicted results with experimentally measured yields for 31 compounds, and the numerical results were found to be in good agreement with the experimental results. It strongly encourages our DCKM approach in understanding lignin vapor-phase reaction chemistry at mechanistic level. A reaction pathway analysis based on the DCKM was proposed for a deeper understanding of the formation of aromatic compounds during lignin pyrolysis. C5H5 was an important precursor of aromatic hydrocarbons such as toluene, naphthalene, and indene formed from lignin pyrolysis. ASSOCIATED CONTENT S Supporting Information ○ GC chromatograms and peak assignment with the different columns, and molecular compositions of volatiles (at 0.1 s) derived from the fast pyrolysis of EHL. This information is available free of charge via the Internet at http://pubs.acs.org/.

AUTHOR INFORMATION

Corresponding Author * E-mail: [email protected]. Tel: +81 92 583 7794. ACKNOWLEDGEMENT This study was financially supported in part by a Grant in-Aid for Young Scientist (A) (Grant Number 23686112) and MOST-JST, Strategic International Collaborative Research Program, SICORP. H.-M. Y. is grateful to the China Scholarship Council (File Number: 201206420006). REFERENCES (1) Zhang, L. H.; Xu, C. B.; Champagne, P. Overview of recent advances in thermo-chemical conversion of biomass. Energy Convers. Manage. 2010, 51, 969-982. (2) White, J. E.; Catallo, W. J.; Legendre, B. L. Biomass pyrolysis kinetics: A comparative critical review with relevant agricultural residue case studies. J. Anal. Appl. Pyrolysis 2011, 91, 1-33. (3) Pandey, M. P.; Kim, C. S. Lignin Depolymerization and Conversion: A Review of Thermochemical Methods. Chem. Eng. Technol. 2011, 34, 29-41. (4) Mohan, D.; Pittman, C. U.; Steele, P. H. Pyrolysis of wood/biomass for bio-oil: A critical review. Energy Fuels 2006, 20, 848-889. (5) Bridgwater, A. V. Review of fast pyrolysis of biomass and product upgrading. Biomass Bioenergy 2012, 38, 68-94. ACS Paragon Plus Environment

11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 27

(6) Jegers, H. E.; Klein, M. T. Primary and Secondary Lignin Pyrolysis Reaction Pathways. Ind. Eng. Chem. Process Des. Dev. 1985, 24, 173-183. (7) Norinaga, K.; Shoji, T.; Kudo, S.; Hayashi, J.-i. Detailed chemical kinetic modelling of vapour-phase cracking of multi-component molecular mixtures derived from the fast pyrolysis of cellulose. Fuel 2013, 103, 141-150. (8) Bridgwater, A. V.; Meier, D.; Radlein, D. An overview of fast pyrolysis of biomass. Org. Geochem. 1999, 30, 1479-1493. (9) Ranzi, E.; Corbetta, M.; Manenti, F.; Pierucci, S. Kinetic modeling of the thermal degradation and combustion of biomass. Chem. Eng. Sci. 2014, 110, 2-12. (10) Hastaoglu, M. A.; Al-Khalid, T. T. Treatment of wood pyrolysis data via a multiple gas-solid reaction model. Ind. Eng. Chem. Res. 2001, 40, 1845-1851. (11) Kersten, S. R. A.; Wang, X. Q.; Prins, W.; van Swaaij, W. P. M. Biomass pyrolysis in a fluidized bed reactor. Part 1: Literature review and model simulations. Ind. Eng. Chem. Res. 2005, 44, 8773-8785. (12) Zhou, S.; Garcia-Perez, M.; Pecha, B.; McDonald, A. G.; Kersten, S. R. A.; Westerhof, R. J. M. Secondary Vapor Phase Reactions of Lignin-Derived Oligomers Obtained by Fast Pyrolysis of Pine Wood. Energy Fuels 2013, 27, 1428-1438. (13) Hosoya, T.; Kawamoto, H.; Saka, S. Different pyrolytic pathways of levoglucosan in vapor-and liquid/solid-phases. J. Anal. Appl. Pyrolysis 2008, 83, 64-70. (14) Patwardhan, P. R.; Dalluge, D. L.; Shanks, B. H.; Brown, R. C. Distinguishing primary and secondary reactions of cellulose pyrolysis. Bioresour. Technol. 2011, 102, 5265-5269. (15) Kawamoto, H.; Yamamoto, D.; Saka, S. Influence of neutral inorganic chlorides on primary and secondary char formation from cellulose. J. Wood Sci. 2008, 54, 242-246. (16) Kawamoto, H.; Morisaki, H.; Saka, S. Secondary decomposition of levoglucosan in pyrolytic production from cellulosic biomass. J. Anal. Appl. Pyrolysis 2009, 85, 247-251. (17) Ronsse, F.; Bai, X. L.; Prins, W.; Brown, R. C. Secondary Reactions of Levoglucosan and Char in the Fast Pyrolysis of Cellulose. Environ. Prog. Sustainable Energy 2012, 31, 256-260. (18) Norinaga, K.; Yang, H. M.; Tanaka, R.; Appari, S.; Iwanaga, K.; Takashima, Y.; Kudo, S.; Shoji, T.; Hayashi, J.-i. A mechanistic study on the reaction pathways leading to benzene and naphthalene in cellulose vapor phase cracking. Biomass Bioenergy 2014, 69, 144-154. (19) Klein, M. T.; Virk, P. S. Modeling of lignin thermolysis. Energy Fuels 2008, 22, 2175-2182. (20) Hu, J.; Shen, D. K.; Xiao, R.; Wu, S. L.; Zhang, H. Y. Free-Radical Analysis on Thermochemical Transformation of Lignin to Phenolic Compounds. Energy Fuels 2013, 27, 285-293. (21) Chu, S.; Subrahmanyam, A. V.; Huber, G. W. The pyrolysis chemistry of a beta-O-4 type oligomeric lignin model compound. Green Chem. 2013, 15, 125-136. (22) Jarvis, M. W.; Daily, J. W.; Carstensen, H. H.; Dean, A. M.; Sharma, S.; Dayton, D. C.; Robichaud, D. J.; Nimlos, M. R. Direct Detection of Products from the Pyrolysis of 2-Phenethyl Phenyl Ether. J.


12

Page 13 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Phys. Chem. A 2011, 115, 428-438. (23) Shen, D. K.; Gu, S.; Luo, K. H.; Wang, S. R.; Fang, M. X. The pyrolytic degradation of wood-derived lignin from pulping process. Bioresour. Technol. 2010, 101, 6136-6146. (24) Garcia-Perez, M.; Wang, S.; Shen, J.; Rhodes, M.; Lee, W. J.; Li, C. Z. Effects of temperature on the formation of lignin-derived oligomers during the fast pyrolysis of Mallee woody biomass. Energy Fuels 2008, 22, 2022-2032. (25) Britt, P. F.; Kidder, M. K.; Buchanan, A. C. Oxygen substituent effects in the pyrolysis of phenethyl phenyl ethers. Energy Fuels 2007, 21, 3102-3108. (26) Jiang, G. Z.; Nowakowski, D. J.; Bridgwater, A. V. Effect of the Temperature on the Composition of Lignin Pyrolysis Products. Energy Fuels 2010, 24, 4470-4475. (27) Zhou, S.; Garcia-Perez, M.; Pecha, B.; Kersten, S. R. A.; McDonald, A. G.; Westerhof, R. J. M. Effect of the Fast Pyrolysis Temperature on the Primary and Secondary Products of Lignin. Energy Fuels 2013, 27, 5867-5877. (28) Caballero, J. A.; Font, R.; Marcilla, A. Kinetic study of the secondary thermal decomposition of Kraft lignin. J. Anal. Appl. Pyrolysis 1996, 38, 131-152. (29) Ranzi, E.; Cuoci, A.; Faravelli, T.; Frassoldati, A.; Migliavacca, G.; Pierucci, S.; Sommariva, S. Chemical Kinetics of Biomass Pyrolysis. Energy Fuels 2008, 22, 4292-4300. (30) Horn, C.; Roy, K.; Frank, P.; Just, T. Shock-tube study on the high-temperature pyrolysis of phenol. Twenty-Seventh Symposium (International) on Combustion, 1998, 1-2, 321-328. (31) Scheer, A. M.; Mukarakate, C.; Robichaud, D. J.; Ellison, G. B.; Nimlos, M. R. Radical Chemistry in the Thermal Decomposition of Anisole and Deuterated Anisoles: An Investigation of Aromatic Growth. J. Phys. Chem. A 2010, 114, 9043-9056. (32) Nowakowska, M.; Herbinet, O.; Dufour, A.; Glaude, P. A. Detailed kinetic study of anisole pyrolysis and oxidation to understand tar formation during biomass combustion and gasification. Combust. Flame 2014, 161, 1474-1488. (33) Faravelli, T.; Frassoldati, A.; Migliavacca, G.; Ranzi, E. Detailed kinetic modeling of the thermal degradation of lignins. Biomass Bioenergy 2010, 34, 290-301. (34) Xu, Z. F.; Lin, M. C. Ab initio kinetics for the unimolecular reaction C6H5OH > CO+C5H6. J. Phys. Chem. A 2006, 110, 1672-1677. (35) Ledesma, E. B.; Marsh, N. D.; Sandrowitz, A. K.; Wornat, M. J. An experimental study on the thermal decomposition of catechol. Proc. Combust. Inst. 2002, 29, 2299-2306. (36) Ledesma, E. B.; Hoang, J. N.; Nguyen, Q.; Hernandez, V.; Nguyen, M. P.; Batamo, S.; Fortune, C. K. Unimolecular Decomposition Pathway for the Vapor-Phase Cracking of Eugenol, A Biomass Tar Compound. Energy Fuels 2013, 27, 6839-6846. (37) Robichaud, D. J.; Scheer, A. M.; Mukarakate, C.; Ormond, T. K.; Buckingham, G. T.; Ellison, G. B.; Nimlos, M. R. Unimolecular thermal decomposition of dimethoxybenzenes. J. Phys. Chem. 2014, 140, 234-302.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 27

(38) Britt, P. F.; Buchanan, A. C.; Cooney, M. J.; Martineau, D. R. Flash vacuum pyrolysis of methoxy-substituted lignin model compounds. J. Org. Chem. 2000, 65, 1376-1389. (39) Thomas, S.; Ledesma, E. B.; Wornat, M. J. The effects of oxygen on the yields of the thermal decomposition products of catechol under pyrolysis and fuel-rich oxidation conditions. Fuel 2007, 86, 2581-2595. (40) Khachatryan, L.; Adounkpe, J.; Maskos, Z.; Dellinger, B. Formation of cyclopentadienyl radical from the gas-phase pyrolysis of hydroquinone, catechol, and phenol. Environ. Sci. Technol. 2006, 40, 5071-5076. (41) Scheer, A. M.; Mukarakate, C.; Robichaud, D. J.; Nimlos, M. R.; Ellison, G. B. Thermal Decomposition Mechanisms of the Methoxyphenols: Formation of Phenol, Cyclopentadienone, Vinylacetylene, and Acetylene. J. Phys. Chem. A 2011, 115, 13381-13389. (42) Cho, J. M.; Chu, S.; Dauenhauer, P. J.; Huber, G. W. Kinetics and reaction chemistry for slow pyrolysis of enzymatic hydrolysis lignin and organosolv extracted lignin derived from maplewood. Green Chem. 2012, 14, 428-439. (43) Yang, H.-M.; Appri, S.; Kudo, S.; Hayashi, J.-i.; Kumagai, S.; Norinaga, K. Chemical structures and primary pyrolysis characteristics of lignins obtained from different preparation methods. J. Japan Institute Energy 2014, 93, 986-994. (44) Huang, Y. R.; Ou, Q. Y.; Yu, W. L. The Effect of Compound Structure on the Elemental Responses in Gas Chromatography-Microwave Induced Plasma Atomic Emission-Spectrometry. J. Chromatogr. Sci. 1990, 28, 584-588. (45) Scanion, J. T.; Willis, D. E. Calculation of Flame Ionization Detector Relative Response Factors Using the Effective Carbon Number Concept. J. Chromatogr. Sci. 1985, 23, 333-340. (46) Deutschmann O.; Tischer S.; Kleditzsch S.; Janardhanan V.; Correa C.; Chatterjee D, DETCHEM Software package, 2.5 ed., [Cited 2014 July 4]. Available from: http://www.detchem.com, Karlsruhe 2014. (47) Shoji, T.; Norinaga, K.; Masek, O. Numerical Simulation of Secondary Gas Phase Reactions of Coffee Grounds with a Detailed Chemical Kinetic Model. J. Japan Institute Energy 2010, 89, 955-961. (48) Thimthong, N.; Appari, S.; Tanaka, R.; Iwanaga, K.; Kudo, S.; Hayashi, J.-i.; Norinaga, K. Kinetic modeling of non-catalytic partial oxidation of nascent volatiles derived from fast pyrolysis of woody biomass with detailed chemistry. Fuel Process. Technol. 2015, 134, 159-167. (49) Patwardhan, P. R.; Brown, R. C.; Shanks, B. H. Understanding the Fast Pyrolysis of Lignin. Chemsuschem 2011, 4, 1629-1636. (50) Altarawneh, M.; Dlugogorski, B. Z.; Kennedy, E. M.; Mackie, J. C. Theoretical Study of Unimolecular Decomposition of Catechol. J. Phys. Chem. A 2010, 114, 1060-1067. (51) Nunn, T. R.; Howard, J. B.; Longwell, J. P.; Peters, W. A. Product Compositions and Kinetics in the Rapid Pyrolysis of Milled Wood Lignin. Ind. Eng. Chem. Process Des. Dev. 1985, 24, 844-852. (52) Caballero, J. A.; Font, R.; Marcilla, A. Pyrolysis of Kraft lignin: Yields and correlations. J. Anal.


14

Page 15 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Appl. Pyrolysis 1997, 39, 161-183. (53) Mullen, C. A.; Boateng, A. A. Catalytic pyrolysis-GC/MS of lignin from several sources. Fuel Process. Technol. 2010, 91, 1446-1458.


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 27

List of Tables Table 1. Gas Chromatography configuration and instrument conditions. Table 2. Elemental Composition of the Quantified Products, Char, and Tar from the Fast Pyrolysis of EHL at 0.1 s. Table 3. Global Reactions at Different Temperatures. List of Figures Figure 1. Comparison between the model predictions with (solid line) and without (dashed line) a global reaction and the experimental observations (circles) for various products at 923 K. Figure 2. Comparison between the model predictions with (solid line) and without (dashed line) a global reaction and the experimental observations (diamonds) for various products at 1023 K. Figure 3. Yield profiles of inorganic compounds obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (squares: experimental data, solid lines: model predictions). Figure 4. Yield profiles of C1−C5 light hydrocarbons obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (circles: experimental data, solid lines: model predictions). Figure 5. Yield profiles of light oxygenated compounds obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (diamonds: experimental data, solid lines: model predictions). Figure 6. Yield profiles of simple phenols obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (squares: experimental data, the solid lines: model predictions). Figure 7. Yield profiles of aromatic hydrocarbons obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (triangles: experimental data, solid lines: model predictions). Figure 8. Reaction flux diagram for the formation of aromatic hydrocarbons (benzene, toluene, naphthalene, and indene) at 1023 K.


16

Page 17 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 1. Gas Chromatography configuration and instrument conditions. Instrument Separation column Shimadzu GC-2014 Gaskuropack 54 Length: 4 m (Packed column, GL Sciences)

VZN-1 Length: 4 m (Packed column, Alltech) Shimadzu GC-2010 PoraBOND Length: 25 m (0.25 mm i.d.) (Capillary column, Varian) TC-1701 Length: 60 m (0.25 mm i.d.) (Capillary column, GL Sciences)

Detector Conditions TCD/FID Injector; 473 K Detector; 493 K TCD current; 110 mA Temperature program; 10 min hold at 313 K, then heated at 5 K min-1 to 473 K, and 30 min hold at 473 K. TCD Injector; 333 K Detector; 333 K TCD current; 110 mA Temperature program; isothermal at 313 K. FID Injector; 618 K Detector; 573 K Temperature program; 10 min hold at 313 K, then heated at 5 K min-1 to 573 K, and 30 min hold at 573 K. FID Injector; 618 K Detector; 573 K Temperature program; 10 min hold at 313 K, then heated at 5 K min-1 to 573 K, and 30 min hold at 573 K.


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 27

Table 2. Elemental Composition (wt%-dry-EHL) of the Quantified Products, Char, and Tar from the Fast Pyrolysis of EHL at 0.1 s. 923 1023 1123 1223 Temperature/K 773 Feed (EHL) 100.00 100.00 100.00 100.00 100.00 C 63.50 63.50 63.50 63.50 63.50 H 5.93 5.93 5.93 5.93 5.93 O 30.57 30.57 30.57 30.57 30.57 Light products 22.19 34.00 48.98 63.07 68.28 C 6.66 14.33 23.63 31.86 32.89 H 1.94 2.95 3.78 4.59 4.66 O 13.59 16.68 21.58 26.64 30.59 Char 34.29 30.30 29.00 25.51 23.53 C 27.00 24.59 24.83 22.28 23.53 H 0.48 0.50 0.32 0.07 0.00 O 6.82 5.21 3.84 3.16 0.00 Tar 43.52 35.70 22.02 11.42 8.19 C 29.84 24.58 15.03 9.35 7.08 H 3.52 2.48 1.83 1.27 1.27 O 10.16 8.67 5.14 0.76 0.00 Formula C100H142O26 C100H121O26 C100H146O26 C100H163O6 C100H215


18

Page 19 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 3. Global Reactions at Different Temperatures. Temperature / K Global reaction 923

1023

Tar (C100H121O26) → 5.2 CO + 2 CO2 + 2.8 H2O + 6.1 H2 + 3 CH4 + 2 C2H4 + C3H6 + 1.5 C4H613 + 0.4 C5H6 + 0.8 C6H6 + C6H5OH + 2 soot (C32H25O6) (k=0.25 s-1) Tar (C100H146O26) → 11 CO + 2 CO2 + 2.5 H2O + 41 H2 + 4 CH4 + C6H6 + 4.5 C6H5OH + 0.5 C8H8 + soot (C46H6O4) (k =0.21 s-1)

1123

Tar (C100H163O6) → 75 H2 + 2 C6H5OH + soot (C88HO4) (k =0.19 s-1)

1223

Tar (C100H215) → 107 H2 + soot (C100H) (k =0.23 s-1)


19


Figure 1. Comparison between the model predictions with (solid line) and without (dashed line) a global reaction and the experimental observations (circles) for various products at 923 K. 10.0 8.0 6.0 4.0 2.0 0.0

Yield, wt%-dry-EHL

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 27

8.0

CO

0 6.0

1

2

3

6.0

0.3

4.0

0.2

2.0

0.1

0.0

0.0 0

4 3.0

CH4

0.4

CO2

1

2

3

4

0 1.5

Ethene

4.0

2.0

1.0

2.0

1.0

0.5

0.0

0.0 0

0.4

1

2

3

4

1

2

3

10.0 H2O 8.0 6.0 4.0 2.0 0.0 4 0 1 1.5 1.2 0.9 0.6 0.3 0.0

Propene

0.0 0

Cyclopentadiene 1.5

0.3

H2

1

2

3

4

0 4.0

Benzene

1

2

3

4

2

3

4

2

3

4

1,3-Butadiene

0

1

2

3

4

Phenol

3.0

1.0

0.2

2.0 0.5

0.1 0.0

1.0

0.0 0

1

2

3

4

0.0 0

1

2

3

4

0 tr, s

1


20

Page 21 of 27

Figure 2. Comparison between the model predictions with (solid line) and without (dashed line) a global reaction and the experimental observations (diamonds) for various products at 1023 K. 20.0

CO

15.0 10.0 5.0 0.0 0 Yield, wt%-dry-EHL

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1.5

1

2

3

10.0 CO2 8.0 6.0 4.0 2.0 0.0 4 0 1 8.0

H2

1.0 0.5

2

3

10.0 H O 2 8.0 6.0 4.0 2.0 0.0 4 0 1 4.0

Methane

6.0

3.0

4.0

2.0

2.0

1.0

0.0

0.0 0 4.0

1

2

3

1.0 0.8 0.6 0.4 0.2 0.0

Benzene

3.0 2.0 1.0 0.0 0

1

2

3

4

3

4

2

3

4

Phenol

0.0 0

4

2

1

2

3

4

2 3 tr, s

4

0

1

Styrene

0

1


21


Figure 3. Yield profiles of inorganic compounds obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (squares: experimental data, solid lines: model predictions). 40.0

CO

30.0 Yield, wt%-dry-EHL

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 27

3.0

H2

2.0

20.0 1.0

10.0 0.0

0.0 700 900 1100 1300 700 900 1100 1300

10.0 8.0 6.0 4.0 2.0 0.0

10.0 H2O 8.0 6.0 4.0 2.0 0.0 700 900 1100 1300 700 900 1100 1300 Temperature, K CO2


22

Page 23 of 27

Figure 4. Yield profiles of C1−C5 light hydrocarbons obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (circles: experimental data, solid lines: model predictions). 10.0 8.0 6.0 4.0 2.0 0.0 Yield, wt%-dry-EHL

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1.2

Methane

Propene

0.4

1.2 0.8 0.4 0.0

Ethane

4.0

0.6

3.0

0.4

2.0

0.2

1.0

0.0 0.0 700 900 1100 1300 700 900 1100 1300

0.8

0.0

0.8

0.20

1-Propyne

0.04

0.15

0.03

0.10

0.02

0.05

0.01

0.00 0.00 700 900 1100 1300 700 900 1100 1300 1,3-Butadiene

0.16

iso-Butane

0.4

0.12

0.3

0.08

0.2

0.04

0.1

1.0 Ethyne 0.8 0.6 0.4 0.2 0.0 700 900 1100 1300 700 900 1100 1300 Ethene

1.0 Butene 0.8 0.6 0.4 0.2 0.0 700 900 1100 1300 700 900 1100 1300 1,4-Pentadiene Cyclopentadiene 0.3 Propadiene

0.2 0.1

0.00 0.0 0.0 700 900 1100 1300 700 900 1100 1300 700 900 1100 1300 700 900 1100 1300 Temperature, K


23


Figure 5. Yield profiles of light oxygenated compounds obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (diamonds: experimental data, solid lines: model predictions). 3.0

Methanol

2.0 Yield, wt%-dry-EHL

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 27

1.0 0.0 0.4

2.0

0.16

1.5

0.12

1.0

0.08

0.5

0.04

Acetone

0.0 0.00 700 900 1100 1300 700 900 1100 1300 700 900 1100 1300 Acetaldehyde

0.08

0.3

0.06

0.2

0.04

0.1

0.02

0.0

Acetic acid

Acrolein

0.12

Ethanol

0.08 0.04

0.00 0.00 700 900 1100 1300 700 900 1100 1300 700 900 1100 1300 Temperature, K


24

Page 25 of 27

Figure 6. Yield profiles of simple phenols obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (squares: experimental data, the solid lines: model predictions). Yield, wt%-dry-EHL

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


4.0 3.0 2.0 1.0 0.0

0.6 1.0 Catechol Cresols 0.5 0.8 0.4 0.6 0.3 0.4 0.2 0.2 0.1 0.0 0.0 700 900 1100 1300 700 900 1100 1300 700 900 1100 1300 Temperature, K Phenol


25


Figure 7. Yield profiles of aromatic hydrocarbons obtained from EHL pyrolysis at 773−1223 K with tr = 3.6 s (triangles: experimental data, solid lines: model predictions). 1.0 Styrene 0.8 4.0 1.0 0.6 0.4 2.0 0.5 0.2 0.0 0.0 0.0 700 900 1100 1300 700 900 1100 1300 700 900 1100 1300 6.0

Yield, wt%-dry-EHL

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 27

0.10 0.08 0.06 0.04 0.02 0.00

Benzene

1.5

Toluene

2.5 1.5 Naphthalene Indene 2.0 1.0 1.5 1.0 0.5 0.5 0.0 0.0 700 900 1100 1300 700 900 1100 1300 700 900 1100 1300 Temperature, K Ethylbenzene


26

Page 27 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8. Reaction flux diagram for the formation of aromatic hydrocarbons (benzene, toluene, naphthalene, and indene) at 1023 K. CH3

CH2

+

OH

-CO CH O

O

+

+ -H

or

-2H

CH

+ -CO ~20%

CH3 HO

21~40% 41~60% 61~80% 81~100%


27

Detailed Chemical Kinetic Modeling of Vapor-Phase Reactions of

Recommend Documents