Effect of Pressure on Pyrolysis of Milled Wood Lignin and Acid

Subscriber access provided by UNIV OF NEWCASTLE

Article

Effect of Pressure on Pyrolysis of Milled Wood Lignin and Acid Washed Hybrid Poplar Wood M. Brennan Pecha, Evan Terrell, Jorge Montoya, Filip Stankovikj, Linda J. Broadbelt, Farid Chejne, and Manuel Garcia-Perez Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.7b02085 • Publication Date (Web): 20 Jul 2017 Downloaded from http://pubs.acs.org on July 21, 2017

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 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

Industrial & Engineering Chemistry Research

Effect of Pressure on Pyrolysis of Milled Wood Lignin and Acid Washed Hybrid Poplar Wood M. Brennan Pecha1, Evan Terrell1, Jorge Ivan Montoya1,2, Filip Stankovikj1, Linda J. Broadbelt3, Farid Chejne2, Manuel Garcia-Perez1* 1

2

3

Biological Systems Engineering, Washington State University, Pullman, WA 99164, USA

Grupo Tayea, Facultad de Minas Universidad Nacional de Colombia, Medellín, Colombia

Department of Chemical and Biological Engineering, Northwestern University, Evanston, IL 60208, USA

1 ACS Paragon Plus Environment


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

Abstract: Thin films (~115 µm thick) of milled wood lignin from hybrid poplar and acidwashed hybrid poplar were pyrolyzed at 500 °C and ~55 °C/s at five pressures (4, 250, 500, 750, and 1000 mbar) to determine the impact of secondary liquid intermediate reactions on the product distribution. For both milled wood lignin extracted from poplar and acid washed hybrid poplar wood, pressure had a significant effect on the product distribution for thin film pyrolysis between 4 and 1000 mbar. For lignin, lowering the pressure from 1000 mbar to 4 mbar reduced the char yield from 36 % to 23 % and enhanced production of large cluster pyrolytic lignin. However, the pressure did not dramatically impact the gas yield (CO2, CO, methane, H2, ethane, propane, and butane), nor did it significantly impact the release of monomeric phenolic compounds. ICR-MS shows limited changes in the composition of lignin oligomers. The increase in the production of large lignin oligomers observed by UV fluorescence and the reduction of char yield with vacuum confirm the importance of oligomeric combination reactions to form large polyaromatic structures in the liquid intermediate. For hybrid poplar, lowering the pressure from 1000 mbar to 4 mbar decreased the char yield from 19 % to 7 % and enhanced production of heavy sugars (cellobiosan and cellotriosan). ICR-MS results clearly show the importance of dehydration reactions in the liquid intermediate. Lowering the pressure also enhanced production of CO, CO2, and methane due to heterogeneous catalysis by residual alkali and alkaline earth metals in the solid wood matrix. However, it also decreased production of levoglucosan from 10 to 6.1 wt.%. The yields of levoglucosan and cellobiosan obtained for hybrid poplar were higher and lower, respectively, compared with those expected if the pyrolysis products were the result of the additive contribution of hybrid poplar constituents. This result could be explained by the tendency of lignin liquid intermediate to bubble vigorously, contributing in this way to the removal of cellulose oligomers from the liquid intermediate.


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


Keywords: Milled wood lignin, poplar, pyrolysis, effect of pressure, secondary reactions, primary reactions

*Corresponding author:

Manuel Garcia-Perez Associate Professor, Biological Systems Engineering, WSU LJ Smith, Room 205, Pullman, WA, 99164-6120 Phone: 509-335-7758, Fax: 509-335-2722, e-mail: [email protected]



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

Page 4 of 38

Introduction

Greenhouse gas and CO2 emissions are expected to grow in the USA and are on track to cause temperatures to rise and extreme weather 1, 2. One way to mitigate the effects of greenhouse gas emissions is to partially replace petroleum with renewable biomass to make fuels and chemicals 3

. One of the most promising energy crop feedstocks is poplar due to its short growing cycle and

easy harvesting in the United States’ Northwest, Lake States, and Mississippi Delta 4. Pyrolysis, a thermochemical processing technology to convert lignocellulosic feedstocks like poplar into bio-oil, char, and gas, has great potential for producing renewable fuels and chemicals

5-8

. In

particular, a version of the technology known as “fast pyrolysis” is capable of transforming up to 80 wt.% of wood into a highly oxygenated crude oil known as “bio-oil”

9-11

. Although many

studies and models exist for fast pyrolysis, there has yet to be a universal model for the technology due to the multi-scale and multi-component nature of the phenomena that occur simultaneously during heating

12, 13

. One of the key pieces of information missing from most

models is the production of aerosols which leads to collection of heavy tars (oligomeric products) in the oil 14-18.

Lignin, one of the three major components of lignocellulosic materials, makes up ~15-20% of grasses and straws, ~30% of softwoods like pine, and 25% of poplar

19, 20

. Pyrolysis of lignin

alone is not commonly performed on the large scale for production of oil due to its propensity to agglomerate and form char. Lignin-rich materials also tend to form a liquid intermediate responsible for the defluidization of sand bed reactors 21, 22. However, studying the fundamentals of neat lignin pyrolysis can provide critical insights into pyrolysis of lignocellulosic feedstocks.


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


The heating of lignin has been studied by the pulp and paper industry at lower temperatures to determine when it softens 23. Many early researchers used Kraft lignin from pulp and paper mills to develop lumped kinetic schemes in slow-heating systems

24

. However, some performed fast

pyrolysis studies with lignin 25, 26. More recently, lignin pyrolysis studies have had a renaissance due to the desire to process lignin from enzymatic hydrolysis processes, as reviewed by Pandey et al. 27. Advanced modeling work has been done to understand the chemistry of bond cleavages in the complex structure of lignin 28-32. Experimental studies have been performed on lignin and model compounds 33-36. Notably, Nunn et al.25 performed thin film (pressed powder) pyrolysis of milled wood lignin (dry, ash-free) in a wire mesh reactor and reported yields of 34.4 % tar and 50% char at close to 500 °C.

In lignin pyrolysis, one of the main challenges in studying the decomposition chemistry of lignin is distinguishing primary from secondary reactions; secondary reactions occur both in the liquid intermediate phase as well as in the gas phase. There are two competing secondary gas phase reaction classes: fragmentation of oligomers into smaller units and cross-linking of monomers to form polyaromatic rings by Yang et al.

38

37

. These gas phase secondary reactions were formalized into a model

. Following the secondary gas phase reaction study, Zhou et al.39 studied the

competition between primary and secondary reactions in the liquid intermediate phase; results from the visualization portion of this study found that there are three main reaction steps: 1) formation of liquid, 2) foaming of this phase due to gas formation, and 3) evaporation. Furthermore, by minimizing gas phase reactions, no lignin monomers were detected in significant quantities. More recently, Montoya et al.17,

40

studied the bubbling and ejection of

aerosol droplets from lignin pyrolysis. These studies showed that ejection of aerosols in the form



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 38

of lignin oligomers can be on the order of 20% from Organosolv lignin during fast pyrolysis (~100 °C/s). However, despite many studies at atmospheric pressure, there are very few reported studies to isolate and study the primary reactions in lignin pyrolysis using vacuum. One recent study by Montoya et al.41 highlighted the effect of pressure and heating rate on aerosol release during pyrolysis at various pressures, finding that lower pressures and higher heating rates enhanced droplet ejection. Pakdel et al.42 performed pyrolysis of lignin in a packed bed reactor under vacuum (2 kPa) and reported yields of 42.7% oil, 10.0 % gas, 38.0% char, and 9.3 % water (dry, ash-free basis) with nearly 40 % of the oil being monomers and dimers. It should be noted that the reactor used by Pakdel and coworkers would have significant gas phase secondary reactions as well as slow heating of the sample due to the large sample size 43.

Fast pyrolysis of poplar at atmospheric pressure commonly provides tar yields of 65%

44

.

Vacuum pyrolysis of poplar at 315 °C gave yields of 52.8% oil, 27.7 % char, 12.3 % water, and 7.2 % gas 45. In another study, Roy et al.46 compared pyrolysis yields under vacuum (10 kPa) and close to atmospheric pressure (80 kPa) at 450 °C and reported yields of 50.0 and 39.7 wt.% oil, respectively. The effect of vacuum on the pyrolysis of poplar and other lignocellulosic materials is still poorly known.

Thus, the goal of the work described in this manuscript is to study the effect of pressure on the vacuum pyrolysis of lignin and hybrid poplar as a way to better understand the nature of secondary reactions in the liquid intermediate on the final outcome of pyrolysis reactions. Vacuum pyrolysis does not affect the reactions in the solid phase, but it could influence the rate at which the pyrolysis primary products are removed either by evaporation or thermal ejection.


Page 7 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


2

2.1

Materials and Methods

Feedstock and lignin preparation

Hybrid poplar (heartwood and sapwood) was donated by the Boise Cascade Corporation in Pasco, WA. Untreated, unmilled wood contained 1 % ash, 3.8 % extractives, 16.3 % acid insoluble lignin, 3.2 % acid soluble lignin, 0.3 % arabinose, 0.4 % galactose, 20.8 % mannose, and 43.1 % glucose, as reported by Zhou et al.47 following the NREL compositional analysis method 48, 49. The poplar was milled at 580 RPM with a 5 min run 1 min rest cycle for 10 hr until particles could be sieved through a 150 US Tyler mesh (105 µm) (See Figure 1). This wood flour was acid washed with 0.1 M nitric acid for 15 min in a shaker then washed with E-pure water three times and centrifuged to remove residual acids and alkali and alkaline earth metals (AAEMs). Acid washed wood was vacuum-dried for more than 48 hr at room temperature and stored in a desiccator.

Proximate analysis of the wood before and after acid washing was performed to quantify moisture and ash content. Experiments were done in a Mettler-Toledo TGA/SDTA 851e thermogravimetric analyzer with the following heating schedule: 25-120 oC at 50 oC/min (N2), Hold 120 oC for 3 min (N2), 120-950 oC at 100 oC/min (N2), Hold 950 oC for 5 min (N2), 950450 oC (N2) at -100 oC/min (N2), 450-600 oC at 100 oC/min (O2), Hold 600 oC 5 min (O2). This method and calculations were based on the protocol described elsewhere.50. ICP metal content analysis was performed using an Agilent 7500cx series with a method described elsewhere in in detail51.



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

Results of the proximate analysis and metal analysis are shown in Table 1. The detectable metal content was reduced from 7.59 to 1.84 mg/g by the acid wash with the residual 20 mg/g being silicate contamination from the ball milling, as calculated by difference. Notably, Na, K, Ca and Mg contents were reduced by 50 %, 95%, 86%, and 80%, respectively. These metals are known to catalyze gas formation and fragmentation reactions during pyrolysis 51.

Figure 1. Non-milled poplar slivers, ball-milled poplar, and milled wood lignin.


Page 8 of 38

Page 9 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


Table 1. Proximate and metal analysis of ball milled poplar and acid washed, ball milled poplar. AF = ash free; errors represent standard deviation; proximate was performed in triplicate, metal analysis in duplicate. Silicates = ash - Total metal by ICP. Parameter

Unit

Ball milled (BM) poplar

Acid washed BM poplar

Moisture Volatile matter (dry) Fixed carbon (dry, AF) Ash (dry) Na Mg K Ca Fe Al Zn Cu Ba Pb Cr Mo Mn Ni Total metal by ICP Silicates by difference

wt.% wt.% wt.% wt.% mg/g mg/g mg/g mg/g mg/g ug/g ug/g ug/g ug/g ug/g ug/g ug/g ug/g ug/g mg/g mg/g

5.89 +/-1.27 80.39 +/- 0.67 16.90 +/- 0.11 2.72 +/- 0.64 0.48 +/- 0.01 0.50 +/- 0.01 3.11 +/- 0.01 2.33 +/ 0.01 0.42 +/- 0.01 654.0 +/- 3.51 12.86 +/- 4.28 38.40 +/- 1.43 26.03 +/- 2.47 11.97 +/- 3.45 1.19 +/- 0.01 0.64 +/- 0.31 11.74 +/- 0.03 0.63 +/- 0.05 7.59 +/- 0.07 19.59 +/- 6.47

1.92 +/- 0.59 83.92 +/- 0.61 13.81 +/- 0.14 2.27 +/- 0.54 0.25 +/- 0.02 0.10 +/- 0.01 0.18 +/- 0.00 0.32 +/- 0.03 0.36 +/- 0.00 591.6 +/- 3.09 10.18 +/- 0.61 7.97 +/- 0.28 3.86 +/- 0.01 1.21 +/- 0.02 0.35 +/- 0.08 5.06 +/- 0.28 0.42 +/- 0.05 1.84 +/- 0.07 20.82 +/- 5.47



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

Preparation of milled wood lignin was performed following the protocol described by Smith who used a modified Bjorkman method

53

Page 10 of 38

52

. Briefly, poplar extracts were removed using 9:1

acetone:water followed by 2:1 ethanol:benzene for 8 hours each in a Soxhlet extractor, then dried at 80 °C overnight. This wood was then milled for 50 hours at 600 RPM with a 15 min on/off cycle. The wood flour was then extracted with a 96 % dioxane solution and centrifuged thrice. Dioxane was removed with a rotary evaporator. This crude lignin was then dissolved in 9:1 acetic acid:water to hydrolyze any carbohydrate residue, then precipitated into cold water and centrifuged. The solids were dissolved in dichloroethane:ethanol 2:1 and precipitated into diethyl ether and washed thrice with ether. The lignin was stored in a brown glass bottle at 0 °C to minimize degradation.

2.2

Thin film pyrolysis at various pressures

Sample preparation for the milled wood lignin and poplar was done by creating thin films on the inside of quartz tubes. A slurry was made with water for the poplar and methanol for the lignin and pushed inside the quartz tubes (CDS Analytical item # 10A1-3015) which were cut to 1.75 cm. The slurry was made to create a film on the inside of the quartz tube after drying. The film sticks to the walls due to surface tension after drying. Water was used for poplar because it suspended the poplar well and methanol for lignin because the lignin dissolves and is suspended in the organic solvent, but not in water due to its polarity. The paste was rolled into a film coating the inside of the tubes, and the outside surfaces of the tubes were cleaned off. The quartz sample tubes were dried overnight at 108 °C, and the top and bottom 2 mm of cellulose were scraped off each end to keep the sample in the isothermal region (+/- 13 °C). Tubes with the


Page 11 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


samples were then weighed. Film thicknesses were measured using photographs taken from above; thus thicknesses are likely overestimates of the overall thickness, representing a conservative value. The thicknesses of the films were 124 +/- 75 µm and 114 +/- 45 µm for lignin and poplar, respectively, as illustrated in Figure 2. Sample masses for poplar ranged from 1.15 to 5.5 mg and for lignin from 2.4 to 7.8 mg.

Figure 2. Lignin and poplar films viewed from above.

The reactor used for this research was described in detail by Pecha et al.54, and the calibration and operation were described in Pecha et al.55. An illustration can be seen in Figure 3. Briefly, a modified pyroprobe was used to pyrolyze the lignin and wood samples (separately) at five pressures: 4, 250, 500, 750, and 1000 mbar (absolute pressures). A system blank was made using 11 ACS Paragon Plus Environment


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 38

the same experimental protocol less a pyrolyzed sample. Gas samples were taken from the reactor and analyzed by GC for light gases, and char yields were determined by mass, as described in Pecha et al.55. Walls of the reactor were washed with methanol and (for poplar) water to analyze product yields. For both the lignin and poplar experiments, the methanol wash was analyzed by GC/MS as described previously in

55

. For lignin, the methanol wash was also

analyzed by UV fluorescence. For poplar, the water wash was analyzed by HPLC to calculate sugar yields, as described in 55.

Figure 3. Illustration of modified pyroprobe reactor setup used in this study.


Page 13 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


2.3

UV fluorescence analysis

For lignin, condensed samples on the surface of the reactor were analyzed by UV fluorescence to characterize

pyrolytic

lignin.

For

these

experiments,

a

Shimadzu

RF-5301PC

spectrofluorophotometer was used. Samples were diluted 100 fold in methanol, and the instrument parameters were the following: Excitation λ 250-500 nm, emission λ 265-515 nm with slit 5 nm, sensitivity high, scan speed fast, response time 0.5s, sample interval 1nm. Data was exported into Matlab, and a background was subtracted using data from a system blank. Peaks were deconvoluted using the lsqnonlin function with four Gaussian curves centered at 310, 328, 355, and 410 nm. Areas were taken and normalized by wash solvent mass and starting lignin mass for comparison relative to lignin. Pyrolytic lignin samples from a fluidized bed poplar pyrolysis oil sample were also run for comparison. These samples were generously provided by Jack Ferrell (at the National Renewable Energy Laboratory) and were obtained by the well accepted cold water precipitation method described in the literature15.

2.4

FT-ICR-MS

High resolution mass spectrometry was performed on the lignin pyrolyzate samples dissolved in methanol. The method used was described elsewhere 55. Briefly, experiments were carried out on a 9.4 T Bruker Solarix electrospray ionization Fourier transform ion cyclotron resonance mass spectrometer, which has enough resolving power to calculate the ultimate mass of each ion it detects.52 One sample at each pressure and a control (blank) were diluted 3:1 in 150 µm ammonium acetate and 0.15% acetic acid which act as cationization agents, loosely based on the



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 38

method proposed by Xu et al.53 for electrospray ionization of sugars. The data analysis showed that all carbohydrates were sodium adducts due to the presence of sodium in the methanol wash solution which was dried in a sodium alumino-silicate zeolite molecular sieve (0.3 nm rods, CAS# 1318-02-1, EMD MilliPore Corporation). Instrument parameters are shown in Table S1. The samples were run with time of flight (TOF) 0.5 ms. Data analysis was performed in the software Composer by Sierra Analytics (Modesto, CA, USA) to assign an atomic formula to each ion. Only sodium adducts were considered. Table S2 shows the parameters used in the data analysis of the lignin and poplar samples. Using less stringent criteria for the lignin and poplar samples allowed for unique assignment of more of the peaks (1.2 vs. 0.8 ppm) compared to the previous study with cellulose (Pecha et al.55). The molecular peaks present in the blanks were considered background, and they were excluded from the sample analysis. The peaks analyzed had minimum abundance >1 %.


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


3

3.1

Results and Discussion

Milled wood lignin pyrolysis products

3.1.1 Lignin temperature excursions

The system was calibrated such that the final temperature of the lignin was 500 °C and the maximum heating rates were the same. For all the pressures (4.3 mbar to 1000 mbar) the average heating rate was 53 °C/s with standard deviation 4.9 °C/s, and the final hold temperature was 498 °C with standard deviation 5.1 °C. The lowest pressure run heats more slowly than the higher pressure runs between 300 and 500 °C. The heating rate was the maximum achievable by this system at the lowest pressure used in these experiments. It is the measured heating rate, not the programmed heating rate

54

. This slowed heating could be caused by poor heat transfer under

vacuum, or a lack of exothermic char forming reactions

54

. Benitez-Guerrero et al.56 found that

alkali lignin has four main thermal peaks when heated slowly in a TGA: one endothermic peak centered at 90 °C, an exothermic peak at 292 °C, an endothermic peak at 312 °C, an exothermic peak at 353 °C, and another exothermic peak at 478 °C. It is likely that these broad exothermic peaks contribute to maintaining the heating rate at higher pressures. However, as we will see in the next section, the char formation is lowest under vacuum (mostly due to the removal of char precursor compounds) and thus exothermic char forming reactions occur to a lesser extent.



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

3.1.2 Lignin char formation

Lignin is well known to form high quantities of char (> 40 %) when pyrolyzed alone 25, 57. Even under vacuum with slow heating rates, char yields of 38 % have been reported 42. In this study, the yield of char increases from 23 wt. % at 4.3 mbar to 36 wt. % at 740 mbar and remains at 36 wt. % at 990 mbar (see Figure 4). Lower pressures enhance evaporation and thermal ejection of compounds from the liquid intermediate phase

41

. At lower pressures, we hypothesize,

evaporation likely plays a more important role. At higher pressures, aerosol ejection likely plays a more important role and ejected droplet sizes are more or less random, sometimes large. If these compounds remain in the liquid intermediate phase, they will undergo secondary reactions to form gas and char. As shown in Figure 4, the gas forming reactions are so strong that they foam the liquid to fill the whole tube at atmospheric pressure, forming a porous char; under vacuum, the char does not bridge across the tube.


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


Figure 4. Lignin char formation yield as a function of pressure. Error bars represent standard error.

3.1.3 Gaseous products from lignin pyrolysis

The yields of CO2, CO, methane, hydrogen, ethane, propane, and butane remain relatively constant across the pressures studied herein (see Figure 5). The average CO2 yield is 1.5 wt. %, CO 0.56 wt. %, H2 0.023 wt.%, methane 0.29 wt. %, and ethane, propane, and butane

Effect of Pressure on Pyrolysis of Milled Wood Lignin and Acid

Recommend Documents