Kinetic Modeling of Cellulose Fractional Pyrolysis

developed by Broido and his colleagues. 15-17 involved a competitive .... material and derived kinetic parameters were used to develop a flexible empi...
1 downloads 19 Views 3MB Size
Subscriber access provided by MT ROYAL COLLEGE

Article

Kinetic Modeling of Cellulose Fractional Pyrolysis Hayat Bennadji, Lavrent Khachatryan, and Slawomir M Lomnicki Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b03078 • Publication Date (Web): 22 Jan 2018 Downloaded from http://pubs.acs.org on January 22, 2018

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.

Energy & Fuels 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 38 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

Energy & Fuels

Kinetic Modeling of Cellulose Fractional Pyrolysis

Hayat Bennadji1,2, Lavrent Khachatryan2, Slawomir M. Lomnicki1*

1

Department of Environmental Sciences, Louisiana State University, Baton Rouge, LA 70803, USA

2

Department of Chemistry, Louisiana State University, Baton Rouge, LA 70803, USA

Abstract The kinetics of cellulose fractional pyrolysis was studied for the first time in the temperature range of 200-900 ºC, with 25 ºC increment under nitrogen atmosphere. A detailed analysis of the major and minor pyrolysis products was performed using a System for Thermal Diagnostic Studies (STDS) and FTIR techniques. A semi-global kinetic model was proposed, with products grouped into kinetic lumps, based on their formation profile similarity. Kinetic parameters (preexponential factor A, and activation energy Ea) for formation of major products grouped into heavy volatiles 1 lump (levoglucosan and anhydrosugars) and light volatiles 2 lump (furans and carbonyls) were obtained based on the performed experimental studies. The final model accurately predicts not only the weight loss, the temperature-distribution of major lumped products, the total yields of tar and gases from the fractional pyrolysis of cellulose, but also shows a good performance toward literature data for experimental studies of others.

Key words: Cellulose, Kinetic modeling, Lumped, Isothermal pyrolysis, fractional pyrolysis.

1 ACS Paragon Plus Environment

Energy & Fuels 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. Introduction Pyrolysis is a fundamental thermochemical conversion reaction that can be used to transform biomass directly into synthesis gas, liquid fuels, and chemicals and commonly accompanies combustion and gasification processes. A good understanding of pyrolysis kinetics is important for the assessment of the feasibility, design, and scaling of industrial biomass conversion applications

1, 2

. Primary components of biomass are cellulose, hemicellulose, and lignin along

with minor amounts of water, mineral, and extractives such as terpenes, fatty acids, oils, and resins3. Cellulose is the most abundant organic compound in plants, comprising up to ~50 wt% of dry biomass 4, 5 and it is one of the major compounds of tobacco representing ~25 wt% of dry tobacco plant and ~36 wt% of a tobacco leaf and it is the main ingredient of a cigarette tobacco rod 6. The formation of the large variety of primary and secondary pyrolysis products and structural changes in cellulose-derived chars during pyrolysis has been extensively studied by many authors 7-9. The pathway of thermal decomposition of cellulose and resulting products formation are influenced by temperature 10, type of atmosphere 11, presence of impurities such as minerals 12

, sample size and texture 10 and crystallinity 13, 14. Despite extensive studies, the mechanism and primary reaction pathway of cellulose

pyrolysis are still in debate

7, 9

. The predictive mechanism for cellulose pyrolysis kinetics

developed by Broido and his colleagues

15-17

involved a competitive, multi-step reaction

sequence, where a stable form of cellulose is converted to a reactive cellulose intermediate (labeled “active cellulose”) which further degrades thermally by two parallel routes, forming either volatiles with no char, or proceeding via a sequence of reactions to form chars and different set of volatiles. The later Broido-Shafizadeh (BS) model 18 was developed based on the cellulose pyrolysis experiments performed under vacuum conditions to prevent secondary 2 ACS Paragon Plus Environment

Page 2 of 38

Page 3 of 38 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

Energy & Fuels

reactions and to improve the recovery of primary products and did not include the secondary reactions. Further studies updated the BS model by considering the formation of an intermediate active cellulose 7, 9. The identification and quantification of active cellulose intermediate remains a controversial and challenging issue, however, rapid heating and cooling of parent material indicated morphological changes to solid cellulose indicative of a liquid intermediate 19. Several kinetic models of cellulose decomposition include such “active cellulose” intermediate formation 7, 9, 18, 20-24

.

Most of the cellulose experimental and kinetics studies centers on the direct conventional pyrolysis; while the studies of fractional pyrolysis (called also stepwise or distillation pyrolysis) remains scarce. In direct conventional pyrolysis, a new sample is ramp-heated followed by isothermal heating stage at prescribed pyrolysis temperature, and each pyrolysis temperature studies starts from a fresh cellulose material and products includes all species formed during the ramp heating and isothermal stage. In fractional pyrolysis, the same sample is pyrolyzed at consecutive incremental temperatures for the same or different time in order to study special fractions of the sample (see more nomenclature in Uden, 1995) 25. Pyrolysis oil (bio-oil) produced by direct pyrolysis of biomass is acidic and cannot be directly used in most applications (chemical and fuel production) as it contains many compounds with various reactive functionalities such as furans, phenols, dehydrated carbohydrates, esters, alcohols, aldehydes, ketones, and acids

26

. High concentration of oxygenates results in product

instability during storage, transport, and especially during heating process

27

. Various costly

methods were proposed to treat bio-oil before it can be accepted commercially

28

. Fractional

pyrolysis offers economic strategy to produce valuable oil with fractionation of formed products during decomposition of biopolymers in biomass 27, 29-32. Hammer et al, (2015) 31 reported a two-

3 ACS Paragon Plus Environment

Energy & Fuels 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 38

step pyrolysis process; in the first step, the biomass was pyrolyzed between 300 and 350 °C, targeting the hemicellulose and cellulose components to produce sugar-rich oil fraction whereas in the second step, the lignin is pyrolyzed at 500 °C to generate phenol-rich oil fraction. Fractional pyrolysis represents accurately cellulose behavior within the cigarette smoking, mimicking well progressing thermal conditions in different zones

33-35

. To our

knowledge few literature data were reported on the fractional pyrolysis of biomass and its constituent

29, 30, 32, 33, 36-39

and focus mainly on understanding the effect of substrates thermal

pretreatment on the yield and properties of the solid products 16, 36, 40-42. Present study assessed the pyrolysis characteristics of crystalline cellulose under isothermal, fractional pyrolysis conditions, in 25 °C temperature increments within the temperature range of 200-900 °C. In contrast to published studies of fractional pyrolysis with isothermal periods from several minutes to a day

16, 36, 40-43

, our experimental data represent a 3 minutes isothermal

periods, with no pre-heating thermal ramp; which represent well gradual heating of biopolymer material and derived kinetic parameters were used to develop a flexible empirical kinetic model This model can be applied in engineering applications, as a predictive tool to optimize yields of pyrolysis products, assist in reactor design and to understand the process of cellulose partial decomposition during puffing a cigarette.

2. Experimental Section 2.1. Materials The cellulose used in this study was microcrystalline cellulose (Avicel PH-105), purchased from FMC BioPolymer with 20 µm nominal particle size. We have determined approx.6 wt% moisture in Avicel cellulose using ASTM E871 analysis method

44

. The ash content was not

more than 0.05%. Ultra-high purity Nitrogen (Grade 5.0) and CO and CO2 standard mixture

4 ACS Paragon Plus Environment

Page 5 of 38 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

Energy & Fuels

were supplied from Airgas. The organic compounds and solvents (grade, ≥99%) were purchased from Sigma-Aldrich and used as received. 2.2. System for Thermal Diagnostic Studies: STDS The experimental technique applied in this study for determining cellulose fractional pyrolysis kinetics is unique and complement the numerous studies reported in the literature. The isothermal fractional pyrolysis was performed using System for Thermal Diagnostic Studies (STDS)

33, 35

. The STDS was designed for in-line analysis of thermal reaction products using

gas-chromatography/mass spectroscopy system at each temperature and to identify evolved chemical compounds on the basis of their different retention times and mass spectra

33

. The

STDS with modified reaction chamber is schematically depicted in Figure 1. It consists of a vertical movable sample holder with a basket at the end, a quartz pyrolysis reactor 7 mm ID x 110 mm long, with 30 mm central section residing in the isothermal zone of the furnace (Figure S1 in the supporting information). The movable sample basket can be lifted up outside of the heated section of the reactor to an ambient temperature zone, for a rapid quenching of the reactions. The reactor was heated to 200-900 ºC by furnace and transfer lines were housed in the heated chamber (300 ºC) to ensure efficient transport of the reaction products to GC-MS. The diameter of transfer lines is 0.5 mm to ensure minimal residence time and no reaction during transport. All transport lines beyond the heated chamber were also heated to 300 °C (dashed square, Figure 1). The furnace temperature was regulated by three thermocouples at three locations of the quartz tubular reactor during pyrolysis. In a typical experiment, about 25±2 mg of cellulose was loaded in a deactivated sample basket (stainless steel or quartz). Two set of experiments were performed: the first pyrolysis set was conducted using a deactivated stainless steel basket at temperature range of 200-500 °C, and 5 ACS Paragon Plus Environment

Energy & Fuels 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

second one - quartz basket at temperature range of 200-900 °C to ensure lack of catalytic effects of basket material on the reaction. Each time, a cellulose sample in the basket was purged by ultra-high purity N2 for 3 minutes at room temperature in cold zone (see Figure 1) and then inserted quickly into the isothermal zone of the preheated reactor at desired temperature and pyrolysed for 3 minutes. A laminar continuous flow of N2 (Reynolds number Re300 - 350 °C 56). These results suggested that the relative evaporation/condensation efficiency of LVG is the main factor that determines the selectivity for volatiles/char in cellulose pyrolysis

54, 59

. Hosoya et al. (2008)

56

compared the reactivity of molten and gas phase LVG at

400 °C and found gaseous LVG produced selectively CO and CO2, in contrast with the formation of char and other low-molecular-weight products from the molten LVG. Shin et al. (2001)

52

showed the polymerization and other molten-phase reactions of LVG to occur at

temperatures above 230 °C, i.e., prior to vaporization. Fukutome et al.(2015, 2017) 63, 64 reported that the gas-phase reactions of LVG contribute to the formation of the gases. On the contrary, molten-phase reactions result in tar and char formation. 3.2. Insignificance of secondary homogeneous reactions The extent of secondary gas-phase reactions is a function of both temperature and residence time. Gas-phase reactions kinetics of the tars and gaseous species produced by 13 ACS Paragon Plus Environment

Energy & Fuels 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

pyrolysis reactions were proposed by many researchers

Page 14 of 38

66, 72-75

. The decomposition of tars

includes an large number of reactions as shown by the mechanism developed by Ranzi et al. (2008)

76

and Calonaci et al.(2010)

77

; For the purpose of model development and ensure no

secondary reactions occur in our experiments, a simple first order overall kinetic model was considered for to evaluate potential of secondary gas-phase reactions of primary pyrolysis products. The lifetime (1/e) of pyrolysis products at our experimental conditions was calculated by applying the experimentally derived rates constants for gas-phase pyrolysis of celluloseevolved volatile matter reported by Antal (1983)

78

and Graham et al. (1994)

79

. The average

calculated lifetime for primary tars gas-phase cracking was found to be 2 s at temperature of 600 ºC (cf. Table S2 in supporting information), which is 10 times higher than the residence time of pyrolysis vapor in the present work (0.2 s). Fukutome et al (2015, 2017)

63, 64

studied the gas-

phase pyrolysis of levoglucosan (LVG) at residence times of 0.8-1.4 s. They found that, for temperature of 500 °C and higher, the gas-phase pyrolysis of LVG generated only noncondensable gases and fragmentation products (C1 and C2 aldehydes/ketones) whereas, the pyrolysis products of molten LVG (furans and char) were not detected under these conditions. In our experiments, majority of primary products were released at temperature less than 400 °C, a temperature not sufficient for the secondary reactions to occur. Consequently, at the temperatures and residence time reported here, the gas-phase reactions outside of the solid matrix are neglected and not considered in the present kinetic model. 3.3. Estimation of kinetic parameters Over 100 species were formed from cellulose pyrolysis. However, the large number of products can be included within the manageable sets of kinetic “lumps”, each lump containing products with sufficiently similar characteristics. A suitable kinetic approach assumes the solid material decomposes directly to each lump by a single independent reaction and that the kinetics 14 ACS Paragon Plus Environment

Page 15 of 38 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

Energy & Fuels

of the decomposition can be represented by a unimolecular pseudo first-order reaction. Such approach has been widely used to represent the kinetics of the thermal decomposition of various solid feedstocks 47, 79, 80. In current study, 3 lump groups were considered for the empirical model (see Figure 5): •

water (H2O) lump resulting from char formation consists of water and carbon oxides (CO and CO2);



volatiles 1 lump includes heavy volatiles namely levoglucosan, anhydrosugars, and derivatives; and



volatiles 2 lump is composed of light volatiles such as furans and carbonyls (aldehydes, ketones, acids, and others).

The experimental data of cellulose fractional pyrolysis were used to estimate kinetic parameters of the two lumps volatiles 1 and volatiles 2, vide infra Table 3. The following assumptions have been made in the model: •

First-order irreversible cellulose pyrolysis kinetics;



The rate constant k is defined by the Arrhenius expression (eq. 1);



The reactor operates under isothermal condition;



The cellulose particles are kinetically controlled - no temperature gradient along the bed and the cellulose particles are in thermal equilibrium at each temperature.

Based on these assumptions, the rate of formation of any pyrolysis product (or lump) can be written as:  =



(eq. 2)



Where, Y is yield of product, wt% of cellulose at time t (s), Rate equation is also expressed as:  = ( − )

(eq. 3)

15 ACS Paragon Plus Environment

Energy & Fuels 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 38

Where, k is the Arrhenius rate constant (s-1), Y0 is the maximum yield, wt% of cellulose at some extended residence time (t∞), Y is yield of product, wt% of cellulose at time t (s), and n is the order of reaction. The 2 equations (eq. 2 and eq.3) are equivalent:

 

= ( − )

(eq. 4)

The rate constant k is defined by the Arrhenius equation (eq.1). Substituting the Arrhenius equation (eq. 1) into (eq. 4):

 

=  −   ( − )

(eq. 5)

For first-order reaction (as many solid pyrolysis reactions are assume to be), n=1. This can be substituted in equation and after rearrangement:



( )

=   =  −





 

(eq. 6)

Integrating both sides of (eq. 6) and after rearrangement, the resulting kinetic equation of lumps is expressed as follows:  =  1 −  − −





 !

(eq. 7)

The experimental data for volatiles 1 and volatiles 2 lumps formation were employed in a non-linear least-square method implemented in the Matlab software (version R2014a) in order to estimate the kinetic parameters (A and Ea) represented by equation (7)

47, 81

. An optimization

procedure was performed to minimize the error between the experimental and computed values for A and Ea by using the lsqcurvefit subroutine. Figure 6a-b represents the best-fit of the kinetic parameters (A, Ea) to the experimental data for volatiles 1 and volatiles 2 lumps formation, with respectively 0.98 and 0.96 correlation coefficients. The kinetic parameters A and Ea, derived from experimental results (rxns # 3, 4) and collected from literatures (rxns # 1, 2, 5) are summarized in Table 3. All the reactions in current studies proceed in a kinetically controlled regime as evidenced by apparent activation energies (cf. Table 3) - typically the reactions limited by mass or heat transfer, show Ea to be below 50 kJ/mol

79, 82

16 ACS Paragon Plus Environment

. Note that cellulose pyrolysis

Page 17 of 38 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

Energy & Fuels

reaction is in kinetically controlled regime since the rates of heat and mass transfer are very fast due to the small size of the cellulose particles (< 20 µm)

10, 83, 84

. Indeed, the dimensionless

pyrolysis (Py) and Biot (Bi) numbers calculated are Py >> 0.1 (in our case Py between 4 and 2x106 are obtained which indicates that the pyrolysis process is under kinetic control regime; Py is a ratio of the sample heat conductivity to the rate of the reaction) and Bi