- Email: [email protected]
Effect of initial moisture content on the yields of oily products from pyrolysis of biomass
J. Anal. Appl. Pyrolysis 71 (2004) 803–815
Effect of initial moisture content on the yields of oily products from pyrolysis of biomass Ayhan Demirbas∗ Department of Chemical Engineering, Selcuk University, Konya, Turkey ; accepted 31 October 2003
Abstract In this work, the effects of initial moisture contents on the yields of total oily products from conventional pyrolysis of spruce wood, hazelnut shell and wheat straw were studied. The yields of total oily products of spruce wood (moisture content: 6.5%), hazelnut shell (moisture content: 6.0%) and wheat straw (moisture content: 7.0%) increase from 8.4, 6.7 and 6.2% to 33.7, 30.8 and 27.4%, respectively, with increasing pyrolysis temperature from 575 to 700 K. The yields of total oily products from spruce wood (moisture content: 60.5%) increase from 17.2 to 39.7% with increasing pyrolysis temperature from 600 to 689 K while the yields of total oily products from spruce wood (moisture content: 0%) increase from 12.6 to 26.7% with increasing pyrolysis temperature from 600 to 689 K in nitrogen medium. The yields of total oily products from hazelnut shell (moisture content: 30.7%) increase from 14.6 to 35.9% with increasing pyrolysis temperature from 600 to 693 K while the yields of total oily products from hazelnut shell (moisture content: 0%) increase from 10.8 to 23.8% with increasing pyrolysis temperature from 600 to 703 K in nitrogen medium. The yields of total oily products from wheat straw (moisture content: 34.7%) increase from 12.1 to 33.6% with increasing pyrolysis temperature from 600 to 693 K while the yields of total oily products from wheat straw (moisture content: 0%) increase from 10.3 to 23.0% with increasing pyrolysis temperature from 600 to 703 K in nitrogen medium. The results indicated that the presence of moisture influenced significantly the thermal degradation degrees of the biomass samples during pyrolysis. © 2003 Elsevier B.V. All rights reserved.
Keywords: Spruce wood; Hazelnut shell; Wheat straw; Moisture content; Pyrolysis
∗
Tel.: +90-462-230-7831; fax: +90-462-248-8508. E-mail address: [email protected] (A. Demirbas).
0165-2370/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.jaap.2003.10.008
804
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
1. Introduction The thermochemical conversion of biomass (pyrolysis, gasification, combustion) is one of the most promising non-nuclear forms of future energy. It is a renewable source of energy and has many advantages from an ecological point of view [1]. The pyrolysis is degradation of biomass by heat in the absence of oxygen which results in the production of charcoal, liquid and gaseous products [2]. Pyrolysis can be used as an independent process for the production of useful fuels and/or chemicals. Biomass feedstocks are chemically and physically heterogeneous, and their components have different reactivities and yield different products. During the pyrolysis process, the temperature inside the particle increases, causing (1) removal of moisture that is present in the biomass particle, (2) the pre-pyrolysis and main pyrolysis reaction takes place as discussed by Babu and Chaurasia [1]. During the pyrolysis process, the pores of the residue are enlarged, and the solid particle merely becomes more porous because the biomass converts into gases. The initial pyrolytic degradation reactions include depolymerization, hydrolysis, oxidation, dehydration, and decarboxylation [3]. The oily liquid fraction consisted of two phases: an aqueous phase containing a wide variety of organo-oxygen compounds of low molecular weight and a non-aqueous tarry phase containing insoluble organics (mainly aromatics) of high molecular weight [4,5]. Tar a viscous black fluid that is a byproduct of the pyrolysis of woody biomass and is used in pitch, varnishes, cements, preservatives, and medicines as disinfectants and antiseptics. Tar from wood is a very complex body, its composition varying with the kind of wood and method of preparation. The chief constituents are pyrocatechol, phenol, guaiacol, cresol, creosol, methyl-creosol, phlorol, toluene, xylene, naphthalene, and other hydrocarbons [2,4,6]. Depending on the operating conditions, the pyrolysis process can be divided into three subclasses: conventional pyrolysis (carbonization), fast pyrolysis and flash pyrolysis. The range of the main operating parameters for pyrolysis processes are given in Table 1 [5]. High yields of liquefied products can be obtained under optimized conditions: very high sample heating and heat transfer rate that requires a finely ground wood type biomass feed, carefully controlled pyrolysis temperature of around 750 K and rapid cooling of pyrolysis vapors to give the oily product. At present, the preferred technology for production of oily products is fast or flash pyrolysis at high temperatures with very short residence times [7]. The aim of this work was the effects of initial moisture contents on the yields of oily products from conventional pyrolysis (slow heating rates, low to intermediate temperatures and long residence times) of spruce wood, hazelnut shell and wheat straw. Table 1 The range of main operating parameters for pyrolysis processes
Pyrolysis temperature (K) Heating rate (K/s) Particle size (mm) Heating time (s)
Conventional pyrolysis
Fast pyrolysis
Flash pyrolysis
550–950 0.1–1 5–50 450–550
850–1250 10–200 <1 0.5–10
1050–1300 >1000 <0.2 <0.5
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
805
Table 2 Chemical compositions of wet biomass samples
Spruce wood Hazelnut shell Wheat straw
C (wt.%)
H (wt.%)
N (wt.%)
O (wt.%)
Ash (wt.%)
Cl (wt.%)
Moisture (wt.%)
20.3 34.8 27.5
2.4 3.9 3.5
0.1 0.7 0.4
16.5 28.3 22.8
0.04 1.42 9.96
0.11 0.12 0.98
60.5 30.7 34.7
2. Experimental The biomass samples (spruce wood, hazelnut shell and wheat straw) were ground and then sieved a size range of 0.180 and 0.250 mm sieve. The elemental compositions and ash of wet biomass samples are given in Table 2. The heating was carried out from 298 to 825 K. The experimental runs were carried out with dried samples (0% moisture), air dried samples, wet samples and in nitrogen medium. In order to assess the thermal behavior of the samples, moisture losses and reaction water yields were determined in pyrolysis runs. The oil yields from pyrolysis processes were determined for each sample with different moisture content. The pyrolysis experiments were carried out in a stainless steel tube with height 95.1 mm, i.d. 17.0 mm and o.d. 19.0 mm. Its total volume was about 21.6 ml. The ground sample was weight for a run 1.00 g and pyrolyzed in the tube. Heat to tube was supplied from external electrical heater and the power was adjusted to give an appropriate heat up time. The simple thermocouple (NiCr—constantan) was placed directly in the pyrolysis medium. For each run, the heater was started at 298 K and terminated when the desired temperature. The pyrolysis products were collected within three different groups as condensables as oily products with two phases (aqueous phase and tarry materials phase), non-condensable gaseous products and solid residue (charcoal). The oily products were separated into two fractions using benzene by simple liquid–liquid extraction in a separatory funnel. The benzene solubles from oily products were called as total water-insolubles or tarry materials. Fig. 1 shows the simple diagram of pyrolysis.
Fig. 1. Simple diagram of pyrolysis. (1) Stainless steel tube, (2) electrical heater, (3) temperature control monitor, (4) nitrogen pressure control monitor, (5) product exit valve, (6) condenser, (7) oily products collecting vessel, (8) nitrogen tube.
806
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
3. Results and discussion Pyrolysis of biomass is complex functions of the experimental conditions, under which the pyrolysis process proceeds. The most important factors, which affect the yield and composition of the volatile fraction liberated, are: biomass species, chemical and structural composition of biomass, particle size, temperature (i.e. temperature-time history), heating conventional pyrolysis rate, atmosphere, pressure and reactor configuration. Total water produced from pyrolysis (TWPP) was calculated as follow: TWPP = moisture content − water from dehydration and pyrolysis reactions
(1)
Figs. 2–4 show the curves for the TWPPs versus temperature for the biomass samples with the runs of 0.2, 0.5, 1.0 and 5.0 K/s heating rate. The rate of TWPP was the lowest in a 0.2 K/s heating run. The correlation equations for the TWPP versus the temperature of pyrolysis (T) and related correlation coefficients (r) from Figs. 2–4 are: For spruce wood: TWPP = 0.1481T − 59171,
r = 0.9112
(2)
r = 0.9582
(3)
For hazelnut shell: TWPP = 0.0650T − 27.063,
Fig. 2. Curves for total water produced from pyrolysis (TWPP) vs. temperature for spruce wood (moisture content: 60.5%) with different heating rates.
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
807
Fig. 3. Curves for total water produced from pyrolysis (TWPP) vs. temperature for hazelnut shell (moisture content: 30.7%) with different heating rates.
For wheat straw: TWPP = 0.0754T − 30.345,
r = 0.9497
(4)
As can be seen from Eq. (2)–(4), there is highly correlation between the TWPP and the temperature of pyrolysis (T) for all runs. Figs. 5–7 show the curves for total water produced from pyrolysis (TWPP) versus pyrolysis temperature for the biomass samples with different moisture contents. The presence of moisture in biomass influences its behavior during pyrolysis and affects the physical properties and quality of the pyrolysis liquid: qualitative observations show that dry feed material led to the production of very viscous oil, particularly at higher reaction temperatures. The biomass samples were spruce, hazelnut shell and wheat straw; the range of temperatures covered 575–825 K and the range of feed moisture contents from bone dry to 11.5, 10.5 and 12.0% wet basis for the samples of spruce, hazelnut shell and wheat straw, respectively. The heating rate was 5 K/s. The results obtained show that for higher feed moisture contents the maximum liquid yield on a dry feed basis occurs at the range of 690–700 K temperatures. The water yield decreases with increasing pyrolysis temperature and that higher feed moisture content slightly increases reaction water. Figs. 8–10 show the plots for total oil (aqueous phase + tarry materials phase) yields of pyrolyses versus pyrolysis temperature for the biomass samples with different moisture contents. The yields of gaseous products increase with increasing pyrolysis temperature for all runs. In general, the yields of oily products increase with increasing pyrolysis temperature
808
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
Fig. 4. Curves for total water produced from pyrolysis (TWPP) vs. temperature for wheat straw (moisture content: 34.7%) with different heating rates.
Fig. 5. Curves for total water produced from pyrolysis (TWPP) vs. pyrolysis temperature for spruce wood with different moisture content. Heating rate: 5 K/s.
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
809
Fig. 6. Curves for total water produced from pyrolysis (TWPP) vs. pyrolysis temperature for hazelnut shell with different moisture content. Heating rate: 5 K/s.
Fig. 7. Curves for total water produced from pyrolysis (TWPP) vs. pyrolysis temperature for wheat straw with different moisture content. Heating rate: 5 K/s.
810
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
Fig. 8. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for spruce wood with different moisture content. Heating rate: 5 K/s.
Fig. 9. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for hazelnut shell with different moisture content. Heating rate: 5 K/s.
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
811
Fig. 10. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for wheat straw with different moisture content. Heating rate: 5 K/s.
from 575 to 700 K then it decreases with increasing the temperature. The maximum yield of oil increases with increasing the initial moisture content of the sample. Figs. 11–13 show the plots for total oil (aqueous phase + tarry materials phase) yields from pyrolyses versus pyrolysis temperature for wet samples with maximum initial moistures and dry samples (moisture content: 0%) in nitrogen medium biomass samples. The results indicated that the presence of moisture influenced significantly the thermal degradation rate of biomass pyrolysis. The peak temperatures in pyrolysis were about 689 and 703 K for heating rates of 5 K/min. The oil yields of pyrolysis processes from spruce wood, hazelnut shell and wheat straw were about 14.5, 13.4 and 10.6 higher than those in nitrogen medium, respectively. This may be ascribed to the intermolecular/intramolecular transfer enhanced by the water molecule. The nitrogen medium runs with dry feed biomass material led to the production of very viscous oil, particularly at higher reaction temperatures. As the pyrolysis temperature increased from 575 to 825 K, the yield of char decreased and the yield of gases increased. The trend was not affected by feed moisture content, but higher moisture content feed leads to lower char yields [2,5]. One additional parameter which may also have an effect on char formation is the moisture content of the biomass used. Qualitative observations show that the presence of moisture increased the yield of char from the pyrolyses of spruce wood and hazelnut shell at temperatures between 660 and 730 K. The mechanism of the pyrolytic degradation of structural components of the biomass samples was separately studied [2,4]. The hemicelluloses breaking down first, at temperatures of 470–530 K and cellulose follows in the temperature range 510–620 K, with lignin being the last component to pyrolyse at temperatures of 550–770 K [8].
812
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
Fig. 11. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for wet and dry spruce wood (moisture content: 0%) samples. Heating rate: 5 K/s.
Fig. 12. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for wet and dry hazelnut shell (moisture content: 0%) samples. Heating rate: 5 K/s.
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
813
Fig. 13. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for wet and dry wheat straw (moisture content: 0%) samples. Heating rate: 5 K/s.
The pyrolysis products of lignin, of which guaiacol is that chiefly obtained from coniferous wood, and guaiacol and pyrogallol dimethyl ether show the aromatic nature of lignin from deciduous woods. Lignin gives higher yields of charcoal and tar from wood although lignin has a threefold higher methoxyl content than wood. Cleavage of the aromatic C–O bond in lignin led to the formation of one-oxygen atom products and the cleavage of the methyl C–O bond to form two-oxygen atom products is the first reaction to occur in the thermolysis of 4-alkylguaiiacol at 600–650 K. Cleavage of the side chain C–C bond occurs between the aromatic ring and ␣-carbon atom. Table 3 shows chemical composition of liquid fraction from pyrolyses of the samples of tobacco stalk and yellow pine [9]. Water is formed by dehydration. In the pyrolysis reactions, methanol arises from the breakdown of methyl esters and/or ethers from decomposition of pectin-like plant materials. Methanol also arises from methoxyl groups of uronic acid [10,11]. Acetic acid is formed in the thermal decomposition of all three main components of wood. Acetic acid comes from the elimination of acetyl groups originally linked to the xylose unit. Furfural is formed by dehydration of the xylose unit. Quantitatively, 1-hydroxy-2-propanone and 1-hydroxy-2-butanone present high concentrations in the liquid products. These two alcohols are partly esterified by acetic acid. In conventional slow pyrolysis, these two products are not found in so great a quantity because of their low stability [12]. Degradation of xylan yields eight main products: water, methanol,
814
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
Table 3 Chemical compositions of aqueous phase from liquid fraction of pyrolysis products from the samples of tobacco stalk and yellow pine Compound
Tobacco stalk
Yellow pine
Acetaldehyde Methanol Acetone Methyl acetate Guaiacol 4-Methyl-guaiacol 2-Butanone Acetic acid 1-Hydroxy-2-propanone 1-Hydroxy-2-butanone Furfural Furfurylic alcohol 2,6-Dimethoxyphenol 2-Methyl-2,6-dimethoxyphenol Undetermined
0.34 6.26 0.91 0.51 0.25 0.43 0.32 4.55 8.46 3.82 1.65 1.72 0.56 0.44 69.78
0.76 5.87 0.64 0.38 0.39 0.38 0.25 13.85 10.40 4.25 1.52 1.28 0.80 0.73 58.50
Source: Ref. [9].
formic, acetic, and propionic acids, 1-hydroxy-2-propanone, 1-hydroxy-2-butanone and 2-furfuraldeyde.
4. Conclusion In this work, the effects of initial moisture contents on the yields of oily products from conventional pyrolysis (slow heating rates, low to intermediate temperatures and long residence times) of spruce wood, hazelnut shell and wheat straw have been investigated. The liquid fraction of the pyrolysis products consists of two phases: an aqueous phase containing a wide variety of organo-oxygen compounds of low molecular weight and a non-aqueous phase containing water-insoluble organics (mainly aromatics) of high molecular weight. This phase is called as tarry materials or tar and is the product of greatest interest. The ratios of acetic acid, methanol, and acetone of aqueous phase were higher than those of non-aqueous phase. If the purpose is to maximize the yield of liquid products resulting from biomass pyrolysis, a low temperature, high heating rate, short gas residence time process would be required. For a high char production, a low temperature, low heating rate process would be chosen. If the purpose were to maximize the yield of fuel gas resulting from pyrolysis, a high temperature, low heating rate, long gas residence time process would be preferred. In general, the yields of oily products increase with increasing pyrolysis temperature from 575 to 700 K then it decreases with increasing the temperature. The yield of total oil increases with increasing the initial moisture content of the sample. The yields of total oily products of spruce wood (moisture content: 6.5%), hazelnut shell (moisture content: 6.0%) and wheat straw (moisture content: 7.0%) increase from 8.4, 6.7
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
815
and 6.2% to 33.7, 30.8 and 27.4%, respectively, with increasing pyrolysis temperature from 575 to 700 K. The yields of total oily products from spruce wood (moisture content: 60.5%) increase from 17.2 to 39.7% with increasing pyrolysis temperature from 600 to 689 K while the yields of oily products from spruce wood (moisture content: 0%) increase from 12.6 to 26.7% with increasing pyrolysis temperature from 600 to 689 K in nitrogen medium. Moisture percentage of the biomass species varied from 41 to 70% [13,14]. The presence of water in wood influences its behavior during pyrolysis and affects the physical properties and quality of the pyrolysis liquid. The results obtained show that for higher initial moisture contents the maximum liquid yield on a dry feed basis occurs at lower pyrolysis temperatures between 689 and 703 K. Qualitative observations show that dry feed material led to the production of very viscous oil, particularly at higher reaction temperatures. The highest liquid yield from spruce wood pyrolysis, 39.7% dry feed basis, was found for initial moisture content 60.5% and reactor temperature 689 K, condition that also gave the lowest viscosity.
References [1] B.V. Babu, A.S. Chaurasia, Energy Convers. Mgmt. 44 (2003) 2251–2275. [2] A. Demirbas, Energy Edu. Sci. Technol. 6 (2000) 19–41. [3] R. Zanzi, Pyrolysis of Biomass, Dissertation, Royal Institute of Technology, Department of Chemical Engineering and Technology, Stockholm, 2001. [4] A. Demirbas, Energy Edu. Sci. Technol. 5 (2000) 21–45. [5] G. Maschio, C. Koufopanos, A. Lucchesi, Bioresource Technol. 42 (1992) 219–231. [6] A. Demirbas, Energy Convers. Mgmt. 41 (2000) 633–646. [7] D.C. Elliot, D. Beckman, A.V. Bridgewater, J.P. Diebold, S.B. Gevert, Y. Solantausta, Energy Fuels 5 (1991) 399–410. [8] A. Demirbas, M.M. Küçük, Cellulose Chem. Technol. 27 (1993) 679–686. [9] A. Demirbas, Energy Sources 24 (2002) 337–345. [10] D. Gullu, Energy Sources 25 (2003) 753–765. [11] A. Demirbas, D. Gullu, Energy Edu. Sci. Technol. 1 (1998) 111–115. [12] O. Beaumont, Wood Fiber Sci. 17 (1985) 228–239. [13] A. Demirbas, Energy Sources 25 (2003) 309. [14] R. Kataki, D. Konwer, Biomass Bioenergy 20 (2001) 17.
Effect of initial moisture content on the yields of oily products from pyrolysis of biomass Ayhan Demirbas∗ Department of Chemical Engineering, Selcuk University, Konya, Turkey ; accepted 31 October 2003
Abstract In this work, the effects of initial moisture contents on the yields of total oily products from conventional pyrolysis of spruce wood, hazelnut shell and wheat straw were studied. The yields of total oily products of spruce wood (moisture content: 6.5%), hazelnut shell (moisture content: 6.0%) and wheat straw (moisture content: 7.0%) increase from 8.4, 6.7 and 6.2% to 33.7, 30.8 and 27.4%, respectively, with increasing pyrolysis temperature from 575 to 700 K. The yields of total oily products from spruce wood (moisture content: 60.5%) increase from 17.2 to 39.7% with increasing pyrolysis temperature from 600 to 689 K while the yields of total oily products from spruce wood (moisture content: 0%) increase from 12.6 to 26.7% with increasing pyrolysis temperature from 600 to 689 K in nitrogen medium. The yields of total oily products from hazelnut shell (moisture content: 30.7%) increase from 14.6 to 35.9% with increasing pyrolysis temperature from 600 to 693 K while the yields of total oily products from hazelnut shell (moisture content: 0%) increase from 10.8 to 23.8% with increasing pyrolysis temperature from 600 to 703 K in nitrogen medium. The yields of total oily products from wheat straw (moisture content: 34.7%) increase from 12.1 to 33.6% with increasing pyrolysis temperature from 600 to 693 K while the yields of total oily products from wheat straw (moisture content: 0%) increase from 10.3 to 23.0% with increasing pyrolysis temperature from 600 to 703 K in nitrogen medium. The results indicated that the presence of moisture influenced significantly the thermal degradation degrees of the biomass samples during pyrolysis. © 2003 Elsevier B.V. All rights reserved.
Keywords: Spruce wood; Hazelnut shell; Wheat straw; Moisture content; Pyrolysis
∗
Tel.: +90-462-230-7831; fax: +90-462-248-8508. E-mail address: [email protected] (A. Demirbas).
0165-2370/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.jaap.2003.10.008
804
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
1. Introduction The thermochemical conversion of biomass (pyrolysis, gasification, combustion) is one of the most promising non-nuclear forms of future energy. It is a renewable source of energy and has many advantages from an ecological point of view [1]. The pyrolysis is degradation of biomass by heat in the absence of oxygen which results in the production of charcoal, liquid and gaseous products [2]. Pyrolysis can be used as an independent process for the production of useful fuels and/or chemicals. Biomass feedstocks are chemically and physically heterogeneous, and their components have different reactivities and yield different products. During the pyrolysis process, the temperature inside the particle increases, causing (1) removal of moisture that is present in the biomass particle, (2) the pre-pyrolysis and main pyrolysis reaction takes place as discussed by Babu and Chaurasia [1]. During the pyrolysis process, the pores of the residue are enlarged, and the solid particle merely becomes more porous because the biomass converts into gases. The initial pyrolytic degradation reactions include depolymerization, hydrolysis, oxidation, dehydration, and decarboxylation [3]. The oily liquid fraction consisted of two phases: an aqueous phase containing a wide variety of organo-oxygen compounds of low molecular weight and a non-aqueous tarry phase containing insoluble organics (mainly aromatics) of high molecular weight [4,5]. Tar a viscous black fluid that is a byproduct of the pyrolysis of woody biomass and is used in pitch, varnishes, cements, preservatives, and medicines as disinfectants and antiseptics. Tar from wood is a very complex body, its composition varying with the kind of wood and method of preparation. The chief constituents are pyrocatechol, phenol, guaiacol, cresol, creosol, methyl-creosol, phlorol, toluene, xylene, naphthalene, and other hydrocarbons [2,4,6]. Depending on the operating conditions, the pyrolysis process can be divided into three subclasses: conventional pyrolysis (carbonization), fast pyrolysis and flash pyrolysis. The range of the main operating parameters for pyrolysis processes are given in Table 1 [5]. High yields of liquefied products can be obtained under optimized conditions: very high sample heating and heat transfer rate that requires a finely ground wood type biomass feed, carefully controlled pyrolysis temperature of around 750 K and rapid cooling of pyrolysis vapors to give the oily product. At present, the preferred technology for production of oily products is fast or flash pyrolysis at high temperatures with very short residence times [7]. The aim of this work was the effects of initial moisture contents on the yields of oily products from conventional pyrolysis (slow heating rates, low to intermediate temperatures and long residence times) of spruce wood, hazelnut shell and wheat straw. Table 1 The range of main operating parameters for pyrolysis processes
Pyrolysis temperature (K) Heating rate (K/s) Particle size (mm) Heating time (s)
Conventional pyrolysis
Fast pyrolysis
Flash pyrolysis
550–950 0.1–1 5–50 450–550
850–1250 10–200 <1 0.5–10
1050–1300 >1000 <0.2 <0.5
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
805
Table 2 Chemical compositions of wet biomass samples
Spruce wood Hazelnut shell Wheat straw
C (wt.%)
H (wt.%)
N (wt.%)
O (wt.%)
Ash (wt.%)
Cl (wt.%)
Moisture (wt.%)
20.3 34.8 27.5
2.4 3.9 3.5
0.1 0.7 0.4
16.5 28.3 22.8
0.04 1.42 9.96
0.11 0.12 0.98
60.5 30.7 34.7
2. Experimental The biomass samples (spruce wood, hazelnut shell and wheat straw) were ground and then sieved a size range of 0.180 and 0.250 mm sieve. The elemental compositions and ash of wet biomass samples are given in Table 2. The heating was carried out from 298 to 825 K. The experimental runs were carried out with dried samples (0% moisture), air dried samples, wet samples and in nitrogen medium. In order to assess the thermal behavior of the samples, moisture losses and reaction water yields were determined in pyrolysis runs. The oil yields from pyrolysis processes were determined for each sample with different moisture content. The pyrolysis experiments were carried out in a stainless steel tube with height 95.1 mm, i.d. 17.0 mm and o.d. 19.0 mm. Its total volume was about 21.6 ml. The ground sample was weight for a run 1.00 g and pyrolyzed in the tube. Heat to tube was supplied from external electrical heater and the power was adjusted to give an appropriate heat up time. The simple thermocouple (NiCr—constantan) was placed directly in the pyrolysis medium. For each run, the heater was started at 298 K and terminated when the desired temperature. The pyrolysis products were collected within three different groups as condensables as oily products with two phases (aqueous phase and tarry materials phase), non-condensable gaseous products and solid residue (charcoal). The oily products were separated into two fractions using benzene by simple liquid–liquid extraction in a separatory funnel. The benzene solubles from oily products were called as total water-insolubles or tarry materials. Fig. 1 shows the simple diagram of pyrolysis.
Fig. 1. Simple diagram of pyrolysis. (1) Stainless steel tube, (2) electrical heater, (3) temperature control monitor, (4) nitrogen pressure control monitor, (5) product exit valve, (6) condenser, (7) oily products collecting vessel, (8) nitrogen tube.
806
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
3. Results and discussion Pyrolysis of biomass is complex functions of the experimental conditions, under which the pyrolysis process proceeds. The most important factors, which affect the yield and composition of the volatile fraction liberated, are: biomass species, chemical and structural composition of biomass, particle size, temperature (i.e. temperature-time history), heating conventional pyrolysis rate, atmosphere, pressure and reactor configuration. Total water produced from pyrolysis (TWPP) was calculated as follow: TWPP = moisture content − water from dehydration and pyrolysis reactions
(1)
Figs. 2–4 show the curves for the TWPPs versus temperature for the biomass samples with the runs of 0.2, 0.5, 1.0 and 5.0 K/s heating rate. The rate of TWPP was the lowest in a 0.2 K/s heating run. The correlation equations for the TWPP versus the temperature of pyrolysis (T) and related correlation coefficients (r) from Figs. 2–4 are: For spruce wood: TWPP = 0.1481T − 59171,
r = 0.9112
(2)
r = 0.9582
(3)
For hazelnut shell: TWPP = 0.0650T − 27.063,
Fig. 2. Curves for total water produced from pyrolysis (TWPP) vs. temperature for spruce wood (moisture content: 60.5%) with different heating rates.
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
807
Fig. 3. Curves for total water produced from pyrolysis (TWPP) vs. temperature for hazelnut shell (moisture content: 30.7%) with different heating rates.
For wheat straw: TWPP = 0.0754T − 30.345,
r = 0.9497
(4)
As can be seen from Eq. (2)–(4), there is highly correlation between the TWPP and the temperature of pyrolysis (T) for all runs. Figs. 5–7 show the curves for total water produced from pyrolysis (TWPP) versus pyrolysis temperature for the biomass samples with different moisture contents. The presence of moisture in biomass influences its behavior during pyrolysis and affects the physical properties and quality of the pyrolysis liquid: qualitative observations show that dry feed material led to the production of very viscous oil, particularly at higher reaction temperatures. The biomass samples were spruce, hazelnut shell and wheat straw; the range of temperatures covered 575–825 K and the range of feed moisture contents from bone dry to 11.5, 10.5 and 12.0% wet basis for the samples of spruce, hazelnut shell and wheat straw, respectively. The heating rate was 5 K/s. The results obtained show that for higher feed moisture contents the maximum liquid yield on a dry feed basis occurs at the range of 690–700 K temperatures. The water yield decreases with increasing pyrolysis temperature and that higher feed moisture content slightly increases reaction water. Figs. 8–10 show the plots for total oil (aqueous phase + tarry materials phase) yields of pyrolyses versus pyrolysis temperature for the biomass samples with different moisture contents. The yields of gaseous products increase with increasing pyrolysis temperature for all runs. In general, the yields of oily products increase with increasing pyrolysis temperature
808
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
Fig. 4. Curves for total water produced from pyrolysis (TWPP) vs. temperature for wheat straw (moisture content: 34.7%) with different heating rates.
Fig. 5. Curves for total water produced from pyrolysis (TWPP) vs. pyrolysis temperature for spruce wood with different moisture content. Heating rate: 5 K/s.
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
809
Fig. 6. Curves for total water produced from pyrolysis (TWPP) vs. pyrolysis temperature for hazelnut shell with different moisture content. Heating rate: 5 K/s.
Fig. 7. Curves for total water produced from pyrolysis (TWPP) vs. pyrolysis temperature for wheat straw with different moisture content. Heating rate: 5 K/s.
810
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
Fig. 8. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for spruce wood with different moisture content. Heating rate: 5 K/s.
Fig. 9. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for hazelnut shell with different moisture content. Heating rate: 5 K/s.
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
811
Fig. 10. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for wheat straw with different moisture content. Heating rate: 5 K/s.
from 575 to 700 K then it decreases with increasing the temperature. The maximum yield of oil increases with increasing the initial moisture content of the sample. Figs. 11–13 show the plots for total oil (aqueous phase + tarry materials phase) yields from pyrolyses versus pyrolysis temperature for wet samples with maximum initial moistures and dry samples (moisture content: 0%) in nitrogen medium biomass samples. The results indicated that the presence of moisture influenced significantly the thermal degradation rate of biomass pyrolysis. The peak temperatures in pyrolysis were about 689 and 703 K for heating rates of 5 K/min. The oil yields of pyrolysis processes from spruce wood, hazelnut shell and wheat straw were about 14.5, 13.4 and 10.6 higher than those in nitrogen medium, respectively. This may be ascribed to the intermolecular/intramolecular transfer enhanced by the water molecule. The nitrogen medium runs with dry feed biomass material led to the production of very viscous oil, particularly at higher reaction temperatures. As the pyrolysis temperature increased from 575 to 825 K, the yield of char decreased and the yield of gases increased. The trend was not affected by feed moisture content, but higher moisture content feed leads to lower char yields [2,5]. One additional parameter which may also have an effect on char formation is the moisture content of the biomass used. Qualitative observations show that the presence of moisture increased the yield of char from the pyrolyses of spruce wood and hazelnut shell at temperatures between 660 and 730 K. The mechanism of the pyrolytic degradation of structural components of the biomass samples was separately studied [2,4]. The hemicelluloses breaking down first, at temperatures of 470–530 K and cellulose follows in the temperature range 510–620 K, with lignin being the last component to pyrolyse at temperatures of 550–770 K [8].
812
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
Fig. 11. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for wet and dry spruce wood (moisture content: 0%) samples. Heating rate: 5 K/s.
Fig. 12. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for wet and dry hazelnut shell (moisture content: 0%) samples. Heating rate: 5 K/s.
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
813
Fig. 13. Plots for total oil yield of pyrolysis vs. pyrolysis temperature for wet and dry wheat straw (moisture content: 0%) samples. Heating rate: 5 K/s.
The pyrolysis products of lignin, of which guaiacol is that chiefly obtained from coniferous wood, and guaiacol and pyrogallol dimethyl ether show the aromatic nature of lignin from deciduous woods. Lignin gives higher yields of charcoal and tar from wood although lignin has a threefold higher methoxyl content than wood. Cleavage of the aromatic C–O bond in lignin led to the formation of one-oxygen atom products and the cleavage of the methyl C–O bond to form two-oxygen atom products is the first reaction to occur in the thermolysis of 4-alkylguaiiacol at 600–650 K. Cleavage of the side chain C–C bond occurs between the aromatic ring and ␣-carbon atom. Table 3 shows chemical composition of liquid fraction from pyrolyses of the samples of tobacco stalk and yellow pine [9]. Water is formed by dehydration. In the pyrolysis reactions, methanol arises from the breakdown of methyl esters and/or ethers from decomposition of pectin-like plant materials. Methanol also arises from methoxyl groups of uronic acid [10,11]. Acetic acid is formed in the thermal decomposition of all three main components of wood. Acetic acid comes from the elimination of acetyl groups originally linked to the xylose unit. Furfural is formed by dehydration of the xylose unit. Quantitatively, 1-hydroxy-2-propanone and 1-hydroxy-2-butanone present high concentrations in the liquid products. These two alcohols are partly esterified by acetic acid. In conventional slow pyrolysis, these two products are not found in so great a quantity because of their low stability [12]. Degradation of xylan yields eight main products: water, methanol,
814
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
Table 3 Chemical compositions of aqueous phase from liquid fraction of pyrolysis products from the samples of tobacco stalk and yellow pine Compound
Tobacco stalk
Yellow pine
Acetaldehyde Methanol Acetone Methyl acetate Guaiacol 4-Methyl-guaiacol 2-Butanone Acetic acid 1-Hydroxy-2-propanone 1-Hydroxy-2-butanone Furfural Furfurylic alcohol 2,6-Dimethoxyphenol 2-Methyl-2,6-dimethoxyphenol Undetermined
0.34 6.26 0.91 0.51 0.25 0.43 0.32 4.55 8.46 3.82 1.65 1.72 0.56 0.44 69.78
0.76 5.87 0.64 0.38 0.39 0.38 0.25 13.85 10.40 4.25 1.52 1.28 0.80 0.73 58.50
Source: Ref. [9].
formic, acetic, and propionic acids, 1-hydroxy-2-propanone, 1-hydroxy-2-butanone and 2-furfuraldeyde.
4. Conclusion In this work, the effects of initial moisture contents on the yields of oily products from conventional pyrolysis (slow heating rates, low to intermediate temperatures and long residence times) of spruce wood, hazelnut shell and wheat straw have been investigated. The liquid fraction of the pyrolysis products consists of two phases: an aqueous phase containing a wide variety of organo-oxygen compounds of low molecular weight and a non-aqueous phase containing water-insoluble organics (mainly aromatics) of high molecular weight. This phase is called as tarry materials or tar and is the product of greatest interest. The ratios of acetic acid, methanol, and acetone of aqueous phase were higher than those of non-aqueous phase. If the purpose is to maximize the yield of liquid products resulting from biomass pyrolysis, a low temperature, high heating rate, short gas residence time process would be required. For a high char production, a low temperature, low heating rate process would be chosen. If the purpose were to maximize the yield of fuel gas resulting from pyrolysis, a high temperature, low heating rate, long gas residence time process would be preferred. In general, the yields of oily products increase with increasing pyrolysis temperature from 575 to 700 K then it decreases with increasing the temperature. The yield of total oil increases with increasing the initial moisture content of the sample. The yields of total oily products of spruce wood (moisture content: 6.5%), hazelnut shell (moisture content: 6.0%) and wheat straw (moisture content: 7.0%) increase from 8.4, 6.7
A. Demirbas / J. Anal. Appl. Pyrolysis 71 (2004) 803–815
815
and 6.2% to 33.7, 30.8 and 27.4%, respectively, with increasing pyrolysis temperature from 575 to 700 K. The yields of total oily products from spruce wood (moisture content: 60.5%) increase from 17.2 to 39.7% with increasing pyrolysis temperature from 600 to 689 K while the yields of oily products from spruce wood (moisture content: 0%) increase from 12.6 to 26.7% with increasing pyrolysis temperature from 600 to 689 K in nitrogen medium. Moisture percentage of the biomass species varied from 41 to 70% [13,14]. The presence of water in wood influences its behavior during pyrolysis and affects the physical properties and quality of the pyrolysis liquid. The results obtained show that for higher initial moisture contents the maximum liquid yield on a dry feed basis occurs at lower pyrolysis temperatures between 689 and 703 K. Qualitative observations show that dry feed material led to the production of very viscous oil, particularly at higher reaction temperatures. The highest liquid yield from spruce wood pyrolysis, 39.7% dry feed basis, was found for initial moisture content 60.5% and reactor temperature 689 K, condition that also gave the lowest viscosity.
References [1] B.V. Babu, A.S. Chaurasia, Energy Convers. Mgmt. 44 (2003) 2251–2275. [2] A. Demirbas, Energy Edu. Sci. Technol. 6 (2000) 19–41. [3] R. Zanzi, Pyrolysis of Biomass, Dissertation, Royal Institute of Technology, Department of Chemical Engineering and Technology, Stockholm, 2001. [4] A. Demirbas, Energy Edu. Sci. Technol. 5 (2000) 21–45. [5] G. Maschio, C. Koufopanos, A. Lucchesi, Bioresource Technol. 42 (1992) 219–231. [6] A. Demirbas, Energy Convers. Mgmt. 41 (2000) 633–646. [7] D.C. Elliot, D. Beckman, A.V. Bridgewater, J.P. Diebold, S.B. Gevert, Y. Solantausta, Energy Fuels 5 (1991) 399–410. [8] A. Demirbas, M.M. Küçük, Cellulose Chem. Technol. 27 (1993) 679–686. [9] A. Demirbas, Energy Sources 24 (2002) 337–345. [10] D. Gullu, Energy Sources 25 (2003) 753–765. [11] A. Demirbas, D. Gullu, Energy Edu. Sci. Technol. 1 (1998) 111–115. [12] O. Beaumont, Wood Fiber Sci. 17 (1985) 228–239. [13] A. Demirbas, Energy Sources 25 (2003) 309. [14] R. Kataki, D. Konwer, Biomass Bioenergy 20 (2001) 17.