energies Article
Catalytic Flash Pyrolysis of Biomass Using Different Types of Zeolite and Online Vapor Fractionation Ali Imran 1, *,† , Eddy A. Bramer 1 , Kulathuiyer Seshan 2 and Gerrit Brem 1 1 2
* †
Laboratory of Thermal Engineering, Faculty of Engineering, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands;
[email protected] (E.A.B.);
[email protected] (G.B.) Catalytic Processes and Materials, Faculty of Science & Technology, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands;
[email protected] Correspondence:
[email protected]; Tel.: +31-53-489-2530; Fax: +31-489-3663 Present address: King Abdullah University of Science & Technology, Thuwal 23955-6900, Saudi Arabia;
[email protected]; Tel.: +966-2808-2221.
Academic Editor: Thomas E. Amidon Received: 4 December 2015; Accepted: 2 February 2016; Published: 11 March 2016
Abstract: Bio-oil produced from conventional flash pyrolysis has poor quality and requires expensive upgrading before it can be used as a transportation fuel. In this work, a high quality bio-oil has been produced using a novel approach where flash pyrolysis, catalysis and fractionation of pyrolysis vapors using two stage condensation are combined in a single process unit. A bench scale unit of 1 kg/h feedstock capacity is used for catalytic pyrolysis in an entrained down-flow reactor system equipped with two-staged condensation of the pyrolysis vapor. Zeolite-based catalysts are investigated to study the effect of varying acidities of faujasite Y zeolites, zeolite structures (ZSM5), different catalyst to biomass ratios and different catalytic pyrolysis temperatures. Low catalyst/biomass ratios did not show any significant improvements in the bio-oil quality, while high catalyst/biomass ratios showed an effective deoxygenation of the bio-oil. The application of zeolites decreased the organic liquid yield due to the increased production of non-condensables, primarily hydrocarbons. The catalytically produced bio-oil was less viscous and zeolites were effective at cracking heavy molecular weight compounds in the bio-oil. Acidic zeolites, H-Y and H-ZSM5, increased the desirable chemical compounds in the bio-oil such as phenols, furans and hydrocarbon, and reduced the undesired compounds such as acids. On the other hand reducing the acidity of zeolites reduced some of the undesired compounds in the bio-oil such as ketones and aldehydes. The performance of H-Y was superior to that of the rest of zeolites studied: bio-oil of high chemical and calorific value was produced with a high organic liquid yield and low oxygen content. H-ZSM5 was a close competitor to H-Y in performance but with a lower yield of bio-oil. Online fractionation of catalytic pyrolysis vapors was employed by controlling the condenser temperature and proved to be a successful process parameter to tailor the desired bio-oil properties. A high calorific value bio-oil having up to 90% organics was produced using two staged condensation of catalytic pyrolysis vapor. Zeolite-based acidic catalysts can be used for selective deoxygenation, and the catalytic bio-oil quality can be further improved with staged vapor condensation. Keywords: catalytic pyrolysis; zeolites; biomass; fractionation
1. Introduction There are a variety of techniques to produce energy and sustainable fuels from biomass, and interest in producing the bio-oil via flash pyrolysis of biomass has been increasing recently. However, the bio-oil obtained via conventional flash pyrolysis is very different from crude oil originating from petroleum sources. Compared to fossil oil, the bio-oil is rich in oxygen-containing molecules, it has Energies 2016, 9, 187; doi:10.3390/en9030187
www.mdpi.com/journal/energies
Energies 2016, 9, 187
2 of 17
a heating value of about half of that of the conventional fuel oil and it is only partly miscible with conventional fuels. Due to these properties many problems arise in the handling and utilization of the bio-oil [1]. Therefore, the bio-oil obtained from conventional flash pyrolysis needs to be upgraded before it can be used as a transportation fuel. For catalytic upgrading of bio-oil, a number of process routes have been explored using various catalysts. Deoxygenation of bio-oil can be carried out by a hydrotreating process with a zeolite-based catalyst. This hydrodeoxygenation (HDO) process is similar to the hydrodesulfurization (HDS) process i.e., it is an essential part of oil refining [2]. Hydro-treating requires large volumes of hydrogen which has a negative impact on the process economics. However, research on catalytic aftertreatment of bio-oil has not shown any promising results. Catalytic aftertreatment processes suffer from fast deactivation of the catalysts caused by coking and poor yield of hydrocarbons due to bypassing of the larger molecules in bio-oils [3]. The oxygen content of bio-oils is usually very high, up to 45 wt% [3,4]. The type of oxygen component depends on the raw material and on the pyrolysis process conditions (temperature, residence time and heating rate) [5–7]. In bio-oil, carboxyl, carbonyl, hydroxyl and methoxy functionalities are typically present [5,7–9]. These oxygenated components are mainly responsible for most of the deleterious properties of the bio-oil such as high viscosity, non-volatility, high acidity and resulting corrosiveness and extreme instability upon storage, lower energy density than the conventional fuel by 50%, immiscibility with fossil fuels, thermal instability and tendency to polymerize under exposure to air [5–8,10,11]. Thus, upgrading of bio-oil means removal of oxygen. As bio-oil contains large molecules derived from lignin, the use of cracking catalysts could selectively break down the lignin derivatives in the bio-oil. Since the lignin derivatives also contain a high proportion of oxygen [12], breaking down these molecules will decrease the oxygen content and increase the hydrocarbons in the bio-oil, with a resulting high heating value of the bio-oil. In order to improve bio-oil quality, several catalysts have been proposed in the literature. In-situ catalytic flash pyrolysis of biomass over zeolite-type catalysts has attracted interest in the past decade because it does not require reducing gases, and the process releases the oxygen as carbon dioxide and water. The catalytically produced bio-oil presents a series of improved properties, such as stability and lower acidity. Lappas et al. [13] have proposed that the bio-oil stability depends on the catalyst type and activity (in addition to pyrolysis process conditions). It is a challenge to develop a single catalyst that can selectively remove problematic oxygenates in bio-oil and till now most of studies have been based on off-theshelf catalysts with little effort dedicated to the design of dedicated catalysts that can tailor the targeted properties of bio-oil. In the present work, zeolite-based catalysts were developed and optimised based on lab scale studies of biomass model components. The criteria for catalyst development is: (i) deoxygenation of oxygenated components in the vapor phase in-situ during the pyrolysis process; (ii) acceleration of primary cracking and (iii) inhibition of secondary cracking. Zeolites are typical fluid catalytic cracking (FCC) catalysts being used in oil refining industry for upgrading low-octane components in the gasoline boiling range, and to isomerize low-octane linear olefins to high-octane branched olefins [14]. Zeolites are inexpensive and have long and rich history of use in FCC units. The key properties of zeolites that influence their acidity, thermal stability, and overall catalytic activity are their structure, Si/Al ratio, particle size, and nature of the exchanged cation. It is well known that some of the inorganic constituents of biomass including alkali metals and alkaline earth metals act as a catalyst during the thermal degradation process that can determine the rate of degradation and yield of char in pyrolysis. To study the effect of the acidity of Y zeolites, sodium, calcium and magnesium cation forms were investigated as these alkaline earth metals are indigenous to a variety of biomasses. Broido et al. proposed that the mechanism of thermal decomposition of the biomass during pyrolysis is affected by the alkali metal cations present that form natural polymer chains from the monomers. Raveendran et al. [15] claimed that it is the cations rather than the anions of alkali metals that are mainly affecting the catalytic biomass pyrolysis. In the present work, Y zeolites with different acidities were prepared by exchanging cations and used for in-situ biomass catalytic pyrolysis to investigate the effect of the acidity and cation forms of the Y zeolite on biomass
Energies 2016, 9, 187
3 of 17
pyrolysis and bio-oil properties. H-ZSM5 is a well-known catalyst used for cracking, deoxygenation and synthesis of aromatic hydrocarbons, and it is investigated in current work to study the effect of different zeolite structures on biomass pyrolysis and bio-oil properties. Fractionation of the bio-oil vapors by staged condensation in the downstream phase of the pyrolysis process is an option to tailor the bio-oil composition. Fractionation of the bio-oil can also be achieved for example by distillation, but re-heating of the bio-oil results in the formation of cross-linked solid carbonaceous residues because of the high reactivity of some fractions present in the bio-oil. There is very limited research available on fractionation during condensation of biomass pyrolysis vapors, Westerhof et al. [16] studied the effect of the condenser conditions on the water and organics content in the liquid obtained in the first condenser. Effendi et al. [17] studied the fractional condensation to concentrate phenolic compounds in a targeted fraction of the bio-oil. Oasmaa et al. [18] studied the use of fractional condensation to remove the light fractions from the bio-oil and found an increase the stability the bio-oil when the light fraction was replaced by an alcohol. In the present work, online fractionation of the catalytic pyrolysis vapors was used to improve the overall quality of the bio-oil, including its water content, elemental composition and calorific value. Online fractionation was employed using two condensers operated at different temperatures to produce two different liquids, an organic rich one and a water rich one. A bench scale unit consisting of an entrained flow downer reactor system is designed for in-situ catalytic flash pyrolysis of biomass. The entrained flow downer reactor type is selected because it features easy operation handling, and demonstrates high biomass heating rates while maintaining good control of the reaction conditions, for instance the residence time of reactants and products [19,20]. Experiments were conducted for in-situ upgrading of pyrolysis vapor with premixed feedstocks of catalyst and biomass, using different cation forms of faujasite zeolites, different zeolite structures (MFI zeolites), different reaction temperatures and different catalyst to biomass ratios. Furthermore, fractional condensation results are presented where flash pyrolysis, catalytic upgrading and fractional condensation are integrated in a single pyrolysis system. 2. Materials and Methods A continuous bench scale unit of 1 kg/h feedstock (biomass or/and catalyst) capacity has been used for the catalytic flash pyrolysis of woody biomass. The experimental set-up used was described in detail in an earlier publication [21], and a schematic of the unit is presented in Figure 1. An entrained flow reactor made of a cylindrical quartz tube of 4.2 m length with an internal diameter of 50 mm is used for in-situ catalytic flash pyrolysis to produce bio-oil. The catalyst and biomass are mixed before feeding to the pyrolysis unit. The residence time of biomass and pyrolysis products in the reactor is controlled with nitrogen carrier gas. The vapors, gases and solids leaving the reactor enter tangentially into a cyclone that allows removal of solids particles up to 20 µm. The solids consist of char (ash and unconverted biomass) for non-catalytic experiments, and char plus spent catalyst for in-situ catalytic experiments. Due to the wide particle size distribution of the biomass and the catalyst, fine particles are separated in a hot filter element downstream of the cyclone. The solids recovery system allows complete removal of char and catalyst. Knowing the concentration of catalyst in the biomass and the amount of feedstock consumed for each experiment, the amount of catalyst in the solids can be determined. The condensation unit consists of two double tube heat exchangers that are operated with a coolant mixture of glycol and water at ´5 ˝ C, the bio-oil fractions are collected together from both condensers except for online fractionation experiments. When fractionation of pyrolysis vapors is required, tap water is used in first condenser instead of glycol/water mixture and water flow rate is controlled to set desired temperature of the outlet gas/vapor stream, the second condenser is operated with coolant at ´5 ˝ C to for maximum recovery of liquid. The light and heavy fractions of bio-oil are recovered in the first and second condenser respectively, and collected separately from both condensers. Separation of light fraction from bio oil can improve its quality because the light fraction carries mainly
Energies 2016, 9, 187
4 of 17
the water and oxygenate components. In this two-step condensation scheme, major portion of the condensable Energies 2016, 9, is 187recovered. Other operating conditions and parameters are listed in Table 1.
Figure 1. 2. Pyrolysis Pyrolysis Reactor. Reactor. 1. Schematic presentation of the experimental setup; 1. Carrier Gas (N2). 2. 3. Solids recovery. Liquid collection. collection. 8. 8. Gas Gas to to analysis analysis unit. recovery. 4. Hot particle filter element. element. 5-6-7. Liquid 9. Coolant. Table Table 1. 1. Operating Operating conditions. conditions.
Reactor temperature Reactor temperature Reactor Reactorpressure pressure Feedhopper hoppercapacity capacity Feed Feedstock feed rate Feedstock feed rate Biomass particle size range Biomass particle size range Vapour residence time in reactor Vapour residence time in reactor
400–550 °C
400–550 ˝ C 1 atm 1 atm 4 kg 4 kg 1 kg/h 1 kg/h 150–1000 µm 150–1000 µ m 2–4 s
2–4 s
2.1. Product Analysis 2.1. Product Analysis The non-condensable gases are mainly composed of carbon dioxide, carbon monoxide and C1 –C3 The non-condensable gases are mainly composed of carbon dioxide, carbon monoxide and hydrocarbons. The volumetric flow rate of these gases is measured with a gas flow meter and a sample C1–C3 hydrocarbons. The volumetric flow rate of these gases is measured with a gas flow meter and of this stream is pumped to an on-line gas analysis unit. Infrared analyzers are used to measure CO, a sample of this stream is pumped to an on-line gas analysis unit. Infrared analyzers are used to CO2 and an FID analyzer is used to measure hydrocarbons. The elemental composition of the liquid is measure CO, CO2 and an FID analyzer is used to measure hydrocarbons. The elemental composition analyzed with an elemental analyzer (Flash 2000 Thermo Scientific, Breda, The Netherlands). The water of the liquid is analyzed with an elemental analyzer (Flash 2000 Thermo Scientific, Breda, contents of the oil are quantified by Karl Fisher titration, and a bomb calorimeter is used to determine The Netherlands). The water contents of the oil are quantified by Karl Fisher titration, and a bomb the heating value of the bio-oil. A GC-MS instrument (Agilent, Amstelveen, The Netherlands) is used calorimeter is used to determine the heating value of the bio-oil. A GC-MS instrument (Agilent, to analyze the bio-oil and the NIST8 library is used to identify the components. Amstelveen, The Netherlands) is used to analyze the bio-oil and the NIST8 library is used to identify the components. 2.2. Biomass The biomass used for the experiments consisted of wood fibres commercially available under 2.2. Biomass the trade name LIGNOCEL from J. Rettenmaier & Söhne GmbH (Rosenberg, Germany). The biomass The biomass used for the experiments consisted of wood fibres commercially available under particle size varied from 0.1 to 1 mm. The ultimate and proximate analysis, and the bio-chemical the trade name LIGNOCEL from J. Rettenmaier & Söhne GmbH (Rosenberg, Germany). The biomass composition [16] of the biomass are presented in Table 2. particle size varied from 0.1 to 1 mm. The ultimate and proximate analysis, and the bio-chemical composition [16] of the biomass are presented in Table 2.
Energies 2016, 9, 187
5 of 17
Table 2. Ultimate and proximate analysis, and bio-chemical composition of biomass. Elemental Composition [wt%, dry basis] Biomass
Lignocel
Bio-Chemical Composition [wt%, dry basis]
Proximate Analysis [wet basis]
C
H
N
O1
Moisture [%]
45.0
5.7
0.1
49.2
6.5
Fixed Volatiles Carbon [%] [%] 76.4 1
13.6
Ash [%]
LHV [MJ/kg]
Cellulose
0.5
16.6
35
Hemicellulose Lignin 29
28
By difference.
2.3. Catalysts Solid acid faujasite (Y and MFI ZSM5) zeolites were used to catalyze the deoxygenation reactions. The basic properties of these catalysts are presented in Table 3. The faujasite zeolites were used in different cation forms, H-Y and Na-Y were obtained from Zeolyst International (Conshohocken, PA, USA), Ca-Y and Mg-Y were prepared by Petrobras (Rio de Janeiro, Brazil) by ion exchange of Na-Y zeolite. H-ZSM5 was obtained from Albemarle B.V. (Amsterdam, The Netherlands). All the catalysts were in powder form, with a particle size ranging between 5 µm to 50 µm. Table 3. Properties of zeolite catalysts. Catalyst
SiO2 /Al2 O3 Mole Ratio
Nominal form
Na2 O [wt%]
Surface Area [m2 /g]
Manufacturer
H-ZSM5 H-Y Na-Y Mg-Y Ca-Y
23 5.2 5.1 5.1 5.1
H+ H+ Na+ Mg2+ Ca2+
0.05 0.2 13.0 -
425 660 900 -
Albemarle Zeolyst International Zeolyst International Petrobras Petrobras
3. Results and Discussion Preliminary experiments were conducted to optimize the operating conditions, e.g., the solids residence time and reaction temperature (T = 450–550 ˝ C, 2 s < residence time < 4 s) for a maximum solid conversion and liquid yield. In all these tests the stability of the unit was satisfactory and reproducible experiments could be performed for 1 to 2 h. A reasonably good mass balance of around 90% and liquid yields around 60% were achieved. An optimum liquid yield was obtained at 500 ˝ C reactor temperature and 4 s vapor residence time in the reactor. Maximum biomass conversion was achieved at 4 s vapor residence time, and vapor residence times longer than 4 s did not improve the biomass conversion and had a negative effect on pyrolysis product properties. Longer vapor residence times caused a significant increase in the production of gases and water contents of the liquid due to undesired secondary cracking of the bio-oil vapors. For catalytic experiments, these operating conditions were fixed in order to compare the effect of catalyst yield and characteristics of biomass pyrolysis products. Catalytic experiments were performed with a premixed feedstock of biomass and catalyst. The two step condensation scheme was employed for fractional condensation of bio-oil to separate the organic and aqueous rich phases of bio-oil. 3.1. Effect of the Catalyst to Biomass Ratio For heterogeneous catalytic systems, the contact between catalyst and biomass/vapor is an important issue and poses practical challenges. The contact can be improved by increasing the catalyst to biomass ratio. ZSM5 and Y zeolites were used to study the effect of the catalyst to biomass ratio on the yield and properties of the pyrolysis products. The results are presented separately for both catalyst types in Subsections 3.1.1 and 3.1.2, respectively.
Energies 2016, 9, 187
6 of 17
Energies 2016, 9, 187
3.1.1. Effect ofof the 3.1.1. Effect theH-ZSM5 H-ZSM5totoBiomass BiomassRatio Ratio ToTo study thethe effect ofofthe ratio on onthe theyields yieldsand andcharacteristics characteristics pyrolysis study effect theH-ZSM5/biomass H-ZSM5/biomass ratio ofof thethe pyrolysis products, experiments were performed with mixture of 10, 0, 10, 50 wt% of H-ZSM5, products, experiments were performed with mixture of 0, 30, 30, andand 50 wt% of H-ZSM5, andand 100,100, 90, 70 90, 70 and 50 wt% of biomass, respectively. H-ZSM5 and biomass were physically mixed and this and 50 wt% of biomass, respectively. H-ZSM5 and biomass were physically mixed and this premixed premixed feedstock usedcatalytic for in-situ catalytic Figure pyrolysis. Figurethe 2 shows thedistribution product distribution feedstock was used forwas in-situ pyrolysis. 2 shows product (dry basis) (dry and, the gas and boil-oil characteristics. Bio-oilinwas in (no onefractionation) condenser (no and, thebasis) gas and boil-oil characteristics. Bio-oil was collected onecollected condenser and fractionation) and its water content was determined by Karl Fisher titration, and the organic and its water content was determined by Karl Fisher titration, and the organic and water contents/fractions contents/fractions the bio-oilfor areapresented separately for a better comparison. of water the bio-oil are presentedofseparately better comparison. 50
Organic Fraction Water Gas Char
Weight %
40 30 20 10 0 0
10 Catalyst wt%
30
50
(a) 50
CO2
CO
Hydrocarbons
Weight %
40 30 20 10 0 0
10 Catalyst wt%
30
50
(b) 60 C [wt%]
50
O [wt%]
40 30
Water [wt%]
20
HHV [MJ/Kg]
10
H [wt%]
0 0
10
30 50 Catalyst wt % (c)
Figure (a)Effect Effect of of the the H-ZSM5/biomass H-ZSM5/biomass ratio onon pyrolysis products yields. (b) Effect of theofH-the Figure 2. 2.(a) ratio pyrolysis products yields; (b) Effect ZSM5/biomass ratio on the gas composition. (c) Effect of the H-ZSM5/biomass ratio onon thethe H-ZSM5/biomass ratio on the gas composition; (c) Effect of the H-ZSM5/biomass ratio characteristics of the bio-oil. characteristics of the bio-oil.
6
Energies Energies2016, 2016,9,9,187 187
7 of 17
The catalyst/biomass ratio had a significant effect on the yields of the different pyrolysis The catalyst/biomass ratio had a significant effect on the yields of the different pyrolysis products products (see Figure 2a). A higher catalyst/biomass ratio led to an increase in the gas production (see Figure 2a). A higher catalyst/biomass ratio led to an increase in the gas production while while the organic liquid fraction significantly reduced. Using 50% H-ZSM5, the organic liquid the organic liquid fraction significantly reduced. Using 50% H-ZSM5, the organic liquid fraction fraction reduced from 47.2% to 14.9% and the gas fraction was increased from 11.6% to 46.8% reduced from 47.2% to 14.9% and the gas fraction was increased from 11.6% to 46.8% compared to the compared to the non-catalytic experiment. non-catalytic experiment. The catalyst/biomass ratio had no significant effect on the char content. The gas composition is The catalyst/biomass ratio had no significant effect on the char content. The gas composition presented in Figure 2b as relative selectivity to CO, CO2 and hydrocarbons. H-ZSM5 had a significant is presented in Figure 2b as relative selectivity to CO, CO2 and hydrocarbons. H-ZSM5 had a effect on the gas composition compared to the non-catalytic situation, but changing the significant effect on the gas composition compared to the non-catalytic situation, but changing the catalyst/biomass ratio did not alter the gas composition to a great extent. A biomass mixture with catalyst/biomass ratio did not alter the gas composition to a great extent. A biomass mixture with 50% H-ZSM5 produced a gas with 32.5% hydrocarbons and a relative concentration of CO2 and CO 50% H-ZSM5 produced a gas with 32.5% hydrocarbons and a relative concentration of CO2 and CO of 25% and 42.5%. A high catalyst/biomass ratio enhanced the secondary cracking of the oil vapors of 25% and 42.5%. A high catalyst/biomass ratio enhanced the secondary cracking of the oil vapors and caused formation of hydrocarbon gases. Increasing the H-ZSM5 concentration did not result in and caused formation of hydrocarbon gases. Increasing the H-ZSM5 concentration did not result in deoxygenation of the bio-oil vapor via the preferred route of decarboxylation, but a significant deoxygenation of the bio-oil vapor via the preferred route of decarboxylation, but a significant amount amount of oxygen is released via decarbonization as CO gas. During catalytic de-oxygenation of bioof oxygen is released via decarbonization as CO gas. During catalytic de-oxygenation of bio-oil vapors, oil vapors, the oxygen may be given off by dehydration, decarbonization and decarboxylation the oxygen may be given off by dehydration, decarbonization and decarboxylation leading to the leading to the formation of water, CO and CO2, respectively. In order to assess the best mode of deformation of water, CO and CO2 , respectively. In order to assess the best mode of de-oxygenation of oxygenation of the vapors, basic calorific value calculations were done based on the elemental the vapors, basic calorific value calculations were done based on the elemental formula (C6 H8 O4 ) for formula (C6H8O4) for bio-oil (see Figure 3). Calculation shows that 75% de-oxygenation via debio-oil (see Figure 3). Calculation shows that 75% de-oxygenation via de-carboxylation will improve carboxylation will improve the heating value of the bio-oil from 19.4 MJ/Kg to 34.1 MJ/Kg. It is evident the heating value of the bio-oil from 19.4 MJ/Kg to 34.1 MJ/Kg. It is evident from these calculations from these calculations that the decarboxylation is the preferred route for deoxygenation in that the decarboxylation is the preferred route for deoxygenation in comparison to dehydration. comparison to dehydration. Decarboxylation removes oxygen with minimal carbon loss, retains the Decarboxylation removes oxygen with minimal carbon loss, retains the hydrogen in bio-oil and hydrogen in bio-oil and thereby increases heating value, decreases the aromatic compounds, thereby increases heating value, decreases the aromatic compounds, minimizes the water content of minimizes the water content of the bio-oil. Hence, in order to the quality of the bio-oil, selective the bio-oil. Hence, in order to the quality of the bio-oil, selective scission of bonds should follow the scission of bonds should follow the order of C-C > C-O > C-H. order of C-C > C-O > C-H.
35 33 HHV [MJ/kg]
31
CO2
29
CO
27
H2O
25 23 21 19 0
25
50
75
Deoxygenation [%] Figure 3. Effectiveness of different de-oxygenation routes to improve the heating value of the bio-oil. Figure 3. Effectiveness of different de-oxygenation routes to improve the heating value of the bio-oil.
AAdetailed detailedanalysis analysisofofbio-oil bio-oilisispresented presentedininFigure Figure2c. 2c.The Theelemental elementalcomposition compositionand andhigher higher heating value of bio-oil is given on a dry basis. H-ZSM5 caused an increase in the carbon and heating value of bio-oil is given on a dry basis. H-ZSM5 caused an increase in the carbon andaa decrease decreaseininthe theoxygen oxygencontent, content,while whilethe thehydrogen hydrogencontent contentremained remainedthe thesame. same.At At50% 50%H-ZSM5 H-ZSM5 concentration, concentration,the theoxygen oxygencontent contentofofbio-oil bio-oilisisreduced reducedfrom from45.4% 45.4%toto41.6% 41.6%and andthe thehigher higherheating heating value increased from 17.8 to 20.7 MJ/kg. Although this is a minor difference, a considerable difference value increased from 17.8 to 20.7 MJ/kg. Although this is a minor difference, a considerable difference ininphysical of of thethe bio-oil could be noted. TheThe thermal pyrolysis oil was viscous and physicalappearance appearance bio-oil could be noted. thermal pyrolysis oil highly was highly viscous heavy tar was in thein condenser, whilewhile the bio-oil obtained via catalytic pyrolysis was less and heavy tarobserved was observed the condenser, the bio-oil obtained via catalytic pyrolysis was viscous and no heavy fraction was observed in the condensers. Obviously the heavy fraction of theof less viscous and no heavy fraction was observed in the condensers. Obviously the heavy fraction bio-oil was cracked to lighter fractions and gaseous components at high concentrations. the bio-oil was cracked to lighter fractions and gaseous components at catalyst high catalyst concentrations.
Energy balance for the non-catalytic pyrolysis of biomass is presented in Figure 4. Energy yields for each product were calculated from the mass yields (see Figure 2, non-catalytic pyrolysis 7
Energies 2016, 9, 187
8 of 17
Energies 2016, 9, 187
Energy balance Energies 2016, 9, 187 for the non-catalytic pyrolysis of biomass is presented in Figure 4. Energy yields experiment) and the gross energy contents, or higher heating values (HHV) of the corresponding for each product were calculated from the mass yields (see Figure 2, non-catalytic pyrolysis experiment) products. HHVand for liquid andenergy char was measured, and calculated for the gas stream. A good energy gross contents, or values higher heating of values (HHV) of the corresponding andexperiment) the gross energythe contents, or higher heating (HHV) the corresponding products. HHV balance closure was liquid achieved. The was optimum typeand of catalytic pyrolysis should have an and char measured, for A thegood gas reactor stream. good energy for products. liquid andHHV charfor was measured, and calculated for thecalculated gas stream. energyA balance closure integrated heat recovery system from combustible gases, and char and can increase the overall closure was achieved. optimum type reactor of catalytic pyrolysis should an wasbalance achieved. The optimum type ofThe catalytic pyrolysis should have anreactor integrated heathave recovery process efficiency and economics. integrated heat recovery system from combustible gases, and char and can increase the overall system from combustible gases, and char and can increase the overall process efficiency and economics. process efficiency and economics.
Closure Closure Liquid Liquid Char Char Gas Gas 0
20 20
0
40 60 40 60 % of energy in feed % of energy in feed
80
100
80
100
Figure Figure 4. 4. Energy Energy balances balances as as % % of of energy energyin infeed feedfor fornon-cataltyic non-cataltyicpyrolysis pyrolysisof ofbiomass. biomass. Figure 4. Energy balances as % of energy in feed for non-cataltyic pyrolysis of biomass.
3.1.2. Effect of Na-Y Na-Y toto Biomass Ratio 3.1.2. Effect of to Biomass Ratio 3.1.2. Effect of Na-Y Biomass Ratio
Weight %
Weight %
Tostudy study thethe effect ofof the faujasite zeolite/biomass productyields yields(liquid, (liquid,gas gas and char) To the effect of the faujasite zeolite/biomass product yields (liquid, gas and char) To study effect the faujasite zeolite/biomass ratio ratio on product and char) andand onthe the liquid product characteristics, experiments were performed using 0, 10, 30, and 50 wt% of and on liquid product characteristics, experiments were performed using 0, 10, 30, and 50 wt% on the liquid product characteristics, experiments were performed using 0, 10, 30, and 50 wt% ofof Na-Y zeolite. The product yields are presented in Figure and the theresults resultsshow showthat thatan anincreasing increasing Na-Y zeolite. The product yields results show that an increasing Na-Y zeolite. The product yieldsare arepresented presentedin inFigure Figure 5a 5a and and Na-Y/biomass ratio had yields. The trends are similar those Na-Y/biomass ratio hadaaasignificant significanteffect on the product trends are similar toto those of of Na-Y/biomass ratio had significant effecton onthe theproduct productyields. yields.The The trends are similar to those H-ZSM5. A higher Na-Y/biomass ratio resulted in more formation of gas and a lower organic fraction H-ZSM5. A higher Na-Y/biomass ratio resulted in more formation of gas and a lower organic fraction of H-ZSM5. A higher Na-Y/biomass ratio resulted in more formation of gas and a lower organic yield. For 50% Na-Y, theorganic organic liquid fraction reduced from 47.2% to 14.1% and the gas yield. For 50% Na-Y, the liquid fraction reduced from 47.2% toto 14.1% and the gasyield yield fraction yield. For 50% Na-Y, the organic liquid fraction reduced from 47.2% 14.1% and the yield increased from 11.6% 49.7% andthere therewas wasno noconsiderable considerable change change toto increased from 11.6% toto 49.7% and there was no considerable changein inthe thechar charyield yieldcompared compared to increased from 11.6% to 49.7% and in the char yield compared non-catalytic pyrolysis. The gas composition is presented in Figure 5b. For 50% Na-Y, the non-catalytic pyrolysis. The gas composition is presented in Figure 5b. For 50% Na-Y, the non-catalytic pyrolysis. The gas composition is presented in Figure 5b. For 50% Na-Y, the hydrocarbons hydrocarbons concentration thegas gas increased from 16.1 44.7% the 2 and CO hydrocarbons ininthe from 16.1 % to 44.7% while theCO CO 2 and CO concentration inconcentration the gas increased from 16.1increased % to 44.7% while the% COto COwhile concentration decreased 2 and concentration decreased from 40.4% and 43.5% to 24.4% and 31% respectively compared to the nonconcentration decreased 40.4% to 24.4%compared and 31% respectively comparedexperiment. to the nonfrom 40.4% and 43.5% tofrom 24.4% andand 31%43.5% respectively to the non-catalytic catalytic experiment. A higher Na-Y/biomass ratio favored the secondary cracking of vapors and the experiment. A higher Na-Y/biomass ratio favored the ofisvapors Acatalytic higher Na-Y/biomass ratio favored the secondary cracking ofsecondary vapors andcracking the effect largerand thanthe it effect is larger than it was in case of H-ZSM5. The bio-oil analysis is presented in Figure 5c. For 50% effect larger than it was inbio-oil case ofanalysis H-ZSM5. bio-oilin analysis Figure 5c.biomass For 50% was iniscase of H-ZSM5. The is The presented Figureis5c.presented For 50% in Na-Y in the Na-Y in the biomass mixture, the oxygen content of the bio-oil is reduced from 45.4% to 41.1% and Na-Y in the the oxygen biomasscontent mixture, thebio-oil oxygenis content the 45.4% bio-oiltois41.1% reduced 45.4%heating to 41.1% and mixture, of the reducedof from andfrom the higher value the higher heating value of the bio-oil increased from 17.8 to 21.6 MJ/kg compared to the thethehigher bio-oil increased from 17.8non-catalytic to 21.6 MJ/kg compared to the of bio-oil heating increasedvalue from of 17.8the to 21.6 MJ/kg compared to the situation. non-catalytic situation. non-catalytic situation. 60 60 Organic Fraction 50 Organic Water Fraction 50 Water Gas 40 Gas 40 Char 30 Char 30 20 2010 10 0 0
0
0
10 Catalyst 10wt%
Catalyst wt% (a) (a)
Figure 5. Cont.
8
8
30
30
50
50
Energies 2016, 9, 187 Energies 2016, 9, 187
9 of 17
50
CO2
CO
Hydrocarbons
Weight %
40 30 20 10 0 0
10
30
50
Catalyst wt% (b) 70 60
C [wt%]
50
O [wt%]
40 30
Water [wt%]
20
HHV [MJ/Kg]
10
H [wt%]
0 0
10
30
50
Catalyst wt% (c) Figure Effectof ofthetheNa-Y/biomass Na-Y/biomass ratio ratio on on pyrolysis pyrolysis products products yields. Figure 5. 5. (a)(a) Effect yields; (b) (b) Effect Effectofofthe the Na-Y/biomass ratio on the gas composition. (c) Effect of the Na-Y/biomass ratio on the characteristics Na-Y/biomass ratio on the gas composition; (c) Effect of the Na-Y/biomass ratio on the characteristics of the bio-oil. of the bio-oil.
The experimental results show that an increasing catalyst/biomass ratio enhanced the formation results show that anyield. increasing catalyst/biomass enhanced ofThe gasexperimental and lowered the organic fraction The higher gas yields in ratio the presence of the the formation H-ZSM5 of gas and lowered the organic fraction yield. The higher gas yields in the presence of the H-ZSM5 and and Na-Y catalysts are due to the secondary cracking and deoxygenation reactions of the product Na-Y catalysts are due to the secondary cracking and deoxygenation reactions of the product vapors. vapors. Oxygen is removed as CO and CO2 and hydrocarbon gases are produced as a result of the Oxygen is removed COlower and CO hydrocarbon produced as a result the cracking 2 andliquid cracking reactions.asThe organic yield duegases to gasare formation is more or lessof compensated reactions. organic liquid yield due to gas is more less compensated by the The betterlower quality. The thus produced could be formation used for e.g., local or electricity productionby in the a better gasquality. engine. The thus produced could be used for e.g., local electricity production in a gas engine.
3.2. 3.2. Effect of the Catalytic Pyrolysis Reaction Effect of the Catalytic Pyrolysis ReactionTemperature Temperature To investigate thethe effect of of the flash pyrolysis pyrolysisprocess, process, To investigate effect thereaction reactiontemperature temperature on on the the catalytic catalytic flash experiments were conducted C and C. The product product yields yieldsand and experiments were conductedwith with50% 50%H-ZSM5 H-ZSM5atat550 550 ˝°C and 500 500 ˝°C. composition of the gaseous products are compared in Table 4 while the characteristics of the bio-oil composition of the gaseous products are compared in Table 4 while the characteristics of the bio-oil presented in Table 5, respectively. Increasingthe thereaction reaction temperature temperature caused are are presented in Table 5, respectively. Increasing causedaadecrease decreaseininthe the organic liquid yield and increaseininthe thegas gasyield. yield.ItIt isis evident evident from from the organic liquid yield and anan increase the gas gas composition compositionthat thata a higher reaction temperature enhancedthe thesecondary secondary cracking cracking of of the the bio-oil higher reaction temperature enhanced bio-oil vapors vaporsresulting resultinginina a lower yield of organic liquid. A lower char yield at a higher temperature is expected because lower yield of organic liquid. A lower char yield at a higher temperature is expected becauseelevated elevated reaction temperatures arefavorable favorablefor for aa better better conversion components in the biomass. A reaction temperatures are conversionofoflignin lignin components in the biomass. higher reaction temperature resulted in a significant increase in the higher heating value of the bioA higher reaction temperature resulted in a significant increase in the higher heating value of the oil (22.6 MJ/kg) because of a higher deoxygenation achieved, but this improvement is at the costs of bio-oil (22.6 MJ/kg) because of a higher deoxygenation achieved, but this improvement is at the costs a lower yield of organic liquid. of a lower yield of organic liquid. 9
Energies 2016, 9, 187 Energies 2016, 9, 187
10 of 17
Table 4. Effect of the pyrolysis reaction temperature on different pyrolysis products yields and gas Table 4. Effect of the pyrolysis reaction temperature on different pyrolysis products yields and composition. gas composition.
50% H-ZSM5 Organic Fraction 50% H-ZSM5 Organic Fraction 500 °C 14.9 ˝ 500 C 14.9 550 °C 9.9 550 ˝ C
9.9
Water 18.2 18.2 18.0
Gases 46.7 46.7 62.7
Char 16.3 16.3 12.0
CxHy 32.5 32.5 45.2
CO2 25 25 17.3
Water
Gases
Char
Cx Hy
CO2
18.0
62.7
12.0
45.2
17.3
CO 42.5 42.5 37.5 CO
37.5
Table 5. Effect of the pyrolysis reaction temperature on the characteristics of the liquid 1. Table 5. Effect of the pyrolysis reaction temperature on the characteristics of the liquid 1 .
50% H-ZSM5
50% H-ZSM5
500500 °C ˝ C 550550 °C ˝ C
C H O Water HHV [MJ/Kg] C [wt%][wt%] H [wt%] [wt%] O [wt%] [wt%] Water [wt%][wt%] HHV [MJ/Kg] 47.2 4.9 46.8 55.0 20.7 47.2 4.9 46.8 55.0 20.7 4.9 42.5 42.5 64.5 64.5 22.6 50.7 50.7 4.9 22.6
1 The elemental Note: The elemental composition and higher heating value ofofthe presentedonon a dry basis. Note: composition and higher heating value thebio-oil bio-oil are are presented a dry basis.
3.3. Effect of of Zeolite Structures 3.3. Effect Zeolite Structures
Weight %
ToToinvestigate investigatethe theeffect effectofofdifferent differentzeolite zeolitestructures structuressuch suchasasfaujasite faujasite(H-Y) (H-Y)and andMFI MFI (Pentasil/H-ZSM5), experiments were performed with H-Y zeolite and thethe results were compared (Pentasil/H-ZSM5), experiments were performed with H-Y zeolite and results were compared with those with experiments to with those withH-ZSM5. H-ZSM5.For Forcomparison comparison50 50wt% wt%of ofcatalyst catalystwas was used used for for this series of experiments to observe observe the clear effect of the zeolite structure. The product yields (dry basis), gas composition and the zeolite structure. The product yields (dry basis), gas composition and liquid characteristics forfor 50% H-Y are compared with the results obtained forfor H-ZSM5 (Figure liquid characteristics 50% H-Y are compared with the results obtained H-ZSM5 (Figure6a–c). 6a–c). It It can be be seen thatthat H-YH-Y produced much less less gas resulting in a higher yieldyield of organic liquid.liquid. Hcan seen produced much gas resulting in a higher of organic Y H-Y produced less hydrocarbon gas and CO and the ispreferred route route for de-for produced less hydrocarbon gas more and more CO CO and2 gas; CO2 this gas;isthis the preferred oxygenation of theofbio-oil vapors to keep maximum hydrogen andand carbon in the bio-oil. It is de-oxygenation the bio-oil vapors to keep maximum hydrogen carbon in the bio-oil. It evident is evident that H-Y is is less prone toto secondary cracking ofof the oiloil vapors. that H-Y less prone secondary cracking the vapors. A Amajor differencebetween between both catalysts structures becomes clear from characteristics. the liquid major difference both catalysts structures becomes clear from the liquid characteristics. H-Y was more effective for the deoxygenation of the bio-oil vapors and lowered the H-Y was more effective for the deoxygenation of the bio-oil vapors and lowered the oxygen content oxygen content of the bio-oil to 41.7, and increased the carbon and hydrogen content of the bio-oil of the bio-oil to 41.7, and increased the carbon and hydrogen content of the bio-oil to 56.2 to and 56.2 5.2respectively. wt%, respectively. the value heating value increased 17.8MJ/kg to 22.9 MJ/kg 5.2and wt%, For 50%For H-Y50% the H-Y heating increased from 17.8from to 22.9 compared compared to non-catalytic experiment. Thisheating higher value heating value can be attributed to the changed to non-catalytic experiment. This higher can be attributed to the changed elemental elemental composition of theH-Y bio-oil. has aperformance superior performance compared H-ZSM5 composition of the bio-oil. has aH-Y superior compared to H-ZSM5 to with respectwith to the respect torecovery the energy recovery in the bio-oil. energy in the bio-oil. Organic Fraction Water Gases Char
50 45 40 35 30 25 20 15 10 5 0 Thermal
H-ZSM5
H-Y
Na-Y
(a) Figure 6. Cont.
10
Mg-Y
Ca-Y
Energies 2016, 9, 187 Energies 2016, 9, 187
11 of 17
70 Thermal H-ZSM5 H-Y Na-Y Mg-Y Ca-Y
60 50 40 30 20 10 0 C [wt%]
H [wt%]
O [wt%]
Water [wt%]
HHV [MJ/Kg]
(b) 60 CO2
Weight %
50
CO
Hydrocarbons
40 30 20 10 0 Thermal H-ZSM5
H-Y
Na-Y
Mg-Y
Ca-Y
(c) Figure Figure 6. 6. (a) (a) Effect Effect of of zeolite zeolitestructure structureand andvarying varyingacidities aciditiesof ofYYzeolite zeoliteon onpyrolysis pyrolysisproducts productsyields. yields; (b) (b) Effect Effect of of zeolite zeolite structure structure and and varying varying acidities acidities of of YY zeolite zeolite on on the the characteristics characteristics of of the the bio-oil. bio-oil; (c) (c) Effect Effect of zeolite structure and varying acidities of Y zeolite on the gas composition.
H-Y produced an organic liquid with a higher yield and higher calorific value, and with lower H-Y produced an organic liquid with a higher yield and higher calorific value, and with lower water content. A detailed comparison of the two zeolite structures is carried out in Section 3.4 based water content. A detailed comparison of the two zeolite structures is carried out in Section 3.4 based on the chemical composition of the produced bio-oils. on the chemical composition of the produced bio-oils. 3.4. Zeolite 3.4. Effect Effect of of Varying Varying Acidities Acidities of of Faujasite Faujasite Zeolite The natureofofthe the (exchanged) cation is aproperty key property of faujasite zeolite catalysts and it The nature (exchanged) cation is a key of faujasite zeolite catalysts and it influences influences the acidity of zeolite and its catalytic activity. To investigate the effect of varying acidities the acidity of zeolite and its catalytic activity. To investigate the effect of varying acidities of faujasite of faujasite zeolite on catalytic biomassexperiments pyrolysis, experiments were with performed withCa-Y, Mg-Yand anda zeolite on catalytic biomass pyrolysis, were performed Mg-Y and Ca-Y, and a was comparison wasH-Y, made withand H-Y, Na-Y and non-catalytic all these comparison made with Na-Y non-catalytic experiments. experiments. For all these For experiments experiments 50 wt% of catalyst was used in the biomass mixture. The acidity of Y zeolite is 50 wt% of catalyst was used in the biomass mixture. The acidity of Y zeolite is changed in thechanged order of in the>order H-Y > > Na-Y > Mg-Y > Ca-Y.yields The product yields basis), gasand composition and liquid H-Y Na-Yof > Mg-Y Ca-Y. The product (dry basis), gas(dry composition liquid characteristics characteristics areFigure presented are presented in 6c. in Figure 6c. Compared to the non-catalytic Compared to the non-catalytic experiment, experiment, all all cation cation forms forms of of faujasite faujasite zeolites zeolites produced produced less less organic liquid and and more moregas. gas.Among Amongthe theallallfaujasite faujasite zeolites, H-Y produced lowest amount of organic liquid zeolites, H-Y produced thethe lowest amount of gas gas and highest amount of organic liquid. Reducing the acidity of Y zeolite resulted in more CO and and highest amount of organic liquid. Reducing the acidity of Y zeolite resulted in more CO and CO2 CO 2 formation and fewer hydrocarbons in the gas, except for H-Y. H-Y produced the highest amount formation and fewer hydrocarbons in the gas, except for H-Y. H-Y produced the highest amount of CO of CO and lowest amount of hydrocarbon while the production COsimilar 2 was quite similar to and lowest amount of hydrocarbon gas while gas the production of CO2 was of quite to other zeolites. otherThe zeolites. liquid product is not very different in terms of elemental composition. The oxygen content Thedecreased liquid product not very acidity different terms of elemental composition. The oxygen content slightly with aisreducing of in Y zeolite while the carbon content increased in all cases. slightly decreased with a reducing acidity of Y zeolite while the carbon content increased in all cases. Acidic zeolite, H-Y, showed a superior performance compared to the other zeolites in terms of keeping Acidic zeolite, H-Y, showed a superior performance the other in terms of the energy in the liquid product. It produced the highestcompared amount ofto organic liquidzeolites with almost similar keeping the energy in the liquid product. It produced the highest amount of organic liquid with 11
Energies 2016, 9, 187 Energies 2016, 9, 187
12 of 17
almost similar elemental composition, slightlyvalue lowerand calorific lower compared water formation elemental composition, slightly lower calorific lower value water and formation to the compared to the other zeolites. other zeolites. 3.5. 3.5.Chemical ChemicalCharacterization CharacterizationofofLiquid Liquid The Thebio-oil bio-oilobtained obtainedfrom fromthe thecatalytic catalyticand andnon-catalytic non-catalyticpyrolysis pyrolysisexperiments experimentswas wasdiluted dilutedin in acetone and analyzed in a GC–MS. A wide range of organic compounds are found in the bio-oil. acetone and analyzed GC–MS. A wide range of organic compounds are found in the bio-oil. In In Figure chemical composition of the bio-oil described the compounds Figure 7, 7, thethe chemical composition of the bio-oil has has beenbeen described and and the compounds havehave been been classified according to their chemical nature hydrocarbons,phenols, phenols, furans, furans, acids, alcohols, classified according to their chemical nature asashydrocarbons, alcohols, aldehydes, aldehydes,ketones, ketones,sugars sugarsand andaromatics. aromatics.
35
TIC area %
30
Thermal H-ZSM5
25
H-Y
20
Na-Y
15
Mg-Y
10
Ca-Y
5 0
Figure 7. Chemical composition of the bio-oil obtained by non-catalytic and catalytic pyrolysis Figure 7. Chemical composition of the bio-oil obtained by non-catalytic and catalytic pyrolysis of biomass. of biomass.
Hydrocarbons are assumed to be the desirable fractions since they are chemicals of high Hydrocarbons are assumed to be the desirable fractions since they are chemicals of high commercial value. In addition to the hydrocarbons, compounds like phenol and its alkylated commercial value. In addition to the hydrocarbons, compounds like phenol and its alkylated derivatives also have a high value, especially for the resin or adhesive industry and they could derivatives also have a high value, especially for the resin or adhesive industry and they could make make the pyrolysis process economically more feasible. Furans are stable compounds with a high the pyrolysis process economically more feasible. Furans are stable compounds with a high energy energy value and fuel compatible. While the oxygen-containing compounds, such as acids and value and fuel compatible. While the oxygen-containing compounds, such as acids and carbonyls, carbonyls, are considered as undesirable fractions. The reduction of the acids is important factor to are considered as undesirable fractions. The reduction of the acids is important factor to improve the improve the corrosive nature of the bio-oil and the reduction of carbonyls is important for its stability. corrosive nature of the bio-oil and the reduction of carbonyls is important for its stability. These two These two factors are critical with respect to handling, storage and fuel applications of the bio-oil. factors are critical with respect to handling, storage and fuel applications of the bio-oil. A catalyst is A catalyst is expected to produce more desirable and less undesirable compounds in the bio-oil. expected to produce more desirable and less undesirable compounds in the bio-oil. Acids, phenols, ketones and furans were the dominant groups of bio-oil and acetic acid was Acids, phenols, ketones and furans were the dominant groups of bio-oil and acetic acid was the the single most dominant and undesired compound in the bio-oil. In general it can be said that single most dominant and undesired compound in the bio-oil. In general it can be said that increasing increasing the acidity of the zeolite catalyst lowered the acid yields except for Mg-Y. H-ZSM5 has the acidity of the zeolite catalyst lowered the acid yields except for Mg-Y. H-ZSM5 has the highest the highest acidity among the investigated zeolites and produced the lowest amount of acids. The acidity among the investigated zeolites and produced the lowest amount of acids. The acid yield was acid yield was reduced to half of what was obtained during non-catalytic pyrolysis. Increasing the reduced to half of what was obtained during non-catalytic pyrolysis. Increasing the acidity of zeolites acidity of zeolites enhanced the formation of ketones except for H-ZSM5. H-Y produced the highest enhanced the formation of ketones except for H-ZSM5. H-Y produced the highest amount of ketones amount of ketones and Ca-Y formed the lowest amount of ketones. All zeolites reduced the amount and Ca-Y formed the lowest amount of ketones. All zeolites reduced the amount of aldehydes in the of aldehydes in the bio-oil, and the effect was more pronounced for lower acidities of the zeolites, bio-oil, and the effect was more pronounced for lower acidities of the zeolites, except for H-Y that except for H-Y that produced the least amount of aldehydes. Increasing the acidity of the zeolites produced the least amount of aldehydes. Increasing the acidity of the zeolites favored the production favored the production of furans except for H-ZSM5. Both H-ZSM5 and Ca-Y produced the lowest of furans except for H-ZSM5. Both H-ZSM5 and Ca-Y produced the lowest amount of furans. amount of furans. Reducing the acidity of zeolites favored the formation of valuable alcohols, and Reducing the acidity of zeolites favored the formation of valuable alcohols, and Ca-Y produced the Ca-Y produced the highest amount of alcohols. Compared to the non-catalytic runs, the application of highest amount of alcohols. Compared to the non-catalytic runs, the application of zeolite catalysts zeolite catalysts increased the production of phenols, a valuable commodity for resins and adhesive increased the production of phenols, a valuable commodity for resins and adhesive industry. In general, increasing the acidity of zeolites favored the phenols production except for H-Y. Zeolites did
12
Energies 2016, 9, 187
13 of 17
Energies 2016, 9, 187
industry. In general, increasing the acidity of zeolites favored the phenols production except for H-Y. Zeolites favor the production of hydrocarbons i.e., the important for fuel application. not favordid thenot production of hydrocarbons i.e., the important compoundcompound for fuel application. Only HOnly H-Ywas zeolite able tocompletely almost completely convert the sugar components. Y zeolite ablewas to almost convert the sugar components. To evaluate the overall performance of the zeolite catalysts and the effect of their acid nature on the chemical composition of the bio-oil, the chemical compounds are grouped in so called undesired desired/valuablecompounds compounds(see (seeFigure Figure 8). 8). Acids, Acids, ketones, ketones, aldehydes aldehydes and and sugars sugars are are grouped and desired/valuable as undesired compounds, while hydrocarbons, phenols, phenols, furans furans and and alcohols alcohols are are grouped as desired Compared to the non-catalytic (thermal) experiments, experiments, both H-ZSM5 and and valuable compounds. Compared H-Y were successful in enhancing the yields of the desired components and reducing the undesired components in Na-Y alsoalso produced a high of desired compound but also enhanced inthe thebio-oil. bio-oil. Na-Y produced a amount high amount of desired compound but also the formation of undesired compounds. ReducingReducing the acidity Y-zeolite had a negative on enhanced the formation of undesired compounds. theofacidity of Y-zeolite had a effect negative the bio-oil composition and produced less desired compounds and more undesired compounds. effect on the bio-oil composition and produced less desired compounds and more undesired SummarizingSummarizing it can be stated thatbe H-Y produced a bio-oil with both a high chemical and calorific compounds. it can stated that H-Y produced a bio-oil with both a high chemicalvalue and in combination with a high organic yield. H-Y has a high potential for athe production a high calorific value in combination withliquid a high organic liquid yield. H-Y has high potentialoffor the quality fuel via catalytic flash pyrolysis of lignocellulosic biomass. H-ZSM5 is a very close competitor production of a high quality fuel via catalytic flash pyrolysis of lignocellulosic biomass. H-ZSM5 is a of H-Y butcompetitor it results inof a lower yield of organic liquid.yield of organic liquid. very close H-Y but it results in a lower
90 80
TIC Area %
70
Undesired Desired/Valuable
60 50 40 30 20 10 0 Thermal
H-ZSM5
H-Y
Na-Y
Mg-Y
Ca-Y
Figure 8. Effect of catalyst on the desired and undesired compounds of the bio-oil. Figure 8. Effect of catalyst on the desired and undesired compounds of the bio-oil.
The result in this study with with zeolitezeolite catalysts in an entrained flow reactor are reactor comparable The resultobtained obtained in this study catalysts in an entrained flow are to other studies conducted at bench scale at units. Aho et al.units. [22] performed pyrolysiscatalytic of pine comparable to other studies conducted bench scale Aho et al.catalytic [22] performed wood in a of fluidized bed reactor to investigate the effect of differentthe structures of different acidic zeolite catalysts. pyrolysis pine wood in a fluidized bed reactor to investigate effect of structures of H-Y and H-ZSM5 were tested as catalysts thetested chemical composition bio-oil was found to be acidic zeolite catalysts. H-Y and H-ZSM5and were as catalysts and of thethe chemical composition of dependent on the structure acidic zeolite catalysts. H-Yofzeolite H-ZSM5 catalyst the bio-oil was found to be of dependent on the structure acidic and zeolite catalysts. H-Ysignificantly zeolite and reduced the organic liquid yield. The formation of acids and alcohols over H-ZSM5 was lower while H-ZSM5 catalyst significantly reduced the organic liquid yield. The formation of acids and alcohols the formation of ketones was the higher than over the H-Y Lappas etthe al. H-Y [23] zeolite. used FCC and over H-ZSM5 was lower while formation of ketones waszeolite. higher than over Lappas H-ZSM5 for catalytic flash pyrolysis of biomass in a circulating fluid bed reactor with continuous et al. [23] used FCC and H-ZSM5 for catalytic flash pyrolysis of biomass in a circulating fluid bed solids regeneration. Thesolids effect regeneration. of specific operating variables including the type of catalyst and the the reactor with continuous The effect of specific operating variables including catalyst/biomass ratio oncatalyst/biomass the final liquid product andliquid yield was studied. FCC andyield H-ZSM5 type of catalyst and the ratio onquality the final product quality and was catalysts produce high gas yields, with dominating oxygenated gaseos compounds as CO and CO. 2 gaseos studied. FCC and H-ZSM5 catalysts produce high gas yields, with dominating oxygenated A high catalyst to biomass ratio enhanced thetoproduction of additional and resulted in a lower compounds as CO 2 and CO. A high catalyst biomass ratio enhancedwater the production of additional yield of organic liquid obtained with less oxygenated compounds but no improvement in the calorific water and resulted in a lower yield of organic liquid obtained with less oxygenated compounds but value. H-ZSM5 produced more carbonyls while the production of acids, HCs, phenols and heavy no improvement in the calorific value. H-ZSM5 produced more carbonyls while the production of oxygenates not significantly changed. Atutxa et al. [24] also reported thatAtutxa bio-oiletfrom catalytic acids, HCs, was phenols and heavy oxygenates was not significantly changed. al. [24] also pyrolysisthat with H-ZSM5 less oxygenated than oil from pyrolysis. Atutxa et al. used reported bio-oil fromwas catalytic pyrolysis with H-ZSM5 wasthermal less oxygenated than oil from thermal ˝ C and found a conical spouted bed reactor to investigate catalytic pyrolysis of pine wood at 400 pyrolysis. Atutxa et al. used a conical spouted bed reactor to investigate catalytic pyrolysis of pine a substantial in gaseous products (mainly CO2 and CO) and water, a CO) corresponding wood at 400 °Cincrease and found a substantial increase in gaseous products (mainly COand 2 and and water,
and a corresponding decrease in the organic liquid and char yield compared to thermal pyrolysis. The obtained liquid product was less corrosive and more stable than thermal bio-oil. 13
Energies 2016, 9, 187
14 of 17
decrease in the organic liquid and char yield compared to thermal pyrolysis. The obtained liquid product was less corrosive and more stable than thermal bio-oil. It can be concluded that in general zeolites reduce oxygenates compounds in the bio-oil via various routes of deoxygenation e.g., dehydration, decarbonylation and decarboxylation. However, main drawback of the zeolites for catalytic pyrolysis is a significant decrease in bio-oil organics yield due to the increase in the production of gases, water and coke. Organics yield in the bio-oil can be increased if deep cracking and formation of hydrocarbon gas can be avoided, it may be controlled by manipulating the strength and concentration of acid sites in zeolites. 3.6. Online Fractionation via Staged Condensation of Pyrolysis Vapors Online fractionation via two staged condensation of pyrolysis vapors was employed to investigate the possibilities to improve the composition and thus the quality of the bio-oil. The pyrolysis vapors leaving the pyrolysis reactor were condensed via two condensers at different temperatures. The first condenser was used to condense the heavy and middle fractions of the bio-oil and it was operated at an outlet temperature of 55 ˝ C. The second condenser was used to condense the water and light oxygenate fractions of the bio-oil and operated with a coolant (glycol/water) temperature of ´5 ˝ C resulting in an outlet gas temperature of 6–8 ˝ C. The fractionation experiments were performed with 50% Na-Y and 50% H-ZSM5 and the liquid characteristics from the first condenser are compared with the results of the experiments without fractionation in Table 6. Table 6. Liquid Characteristics for experiments with and without fractionation, dry basis. Exp. #
Description
C [wt%]
H [wt%]
O [wt%]
Water [wt%]
HHV [MJ/Kg]
1 2 3 4
Na-Y (no Fractionation) Na-Y-1st Condenser H-ZSM5 (no Fractionation) H-ZSM5-1st Condenser
52.9 53.6 51.6 55.4
4.9 6.5 5.9 6.2
41.1 39.9 41.6 37.9
63.0 31.0 55.0 11.0
21.6 23.1 20.7 23.0
Experiments 1 and 3 were performed with both condensers at the same low temperature and the liquid was collected together (no fractionation), while experiments 2 and 4 were performed with two staged condensation. For the fractionation experiments, the liquid obtained in the second condenser was mainly water (the water content was more than 90 wt%) and it was not possible to perform a reliable and accurate elemental analysis of the liquid because of this high water content. The liquid obtained from the first condenser was rich in organic compounds and its characteristics are presented here in Table 6 being the fraction of interest. A significant difference in the composition of the bio-oil is seen in case of fractionation. In case of Na-Y and fractionation, the water content goes down to 31 wt%, the hydrogen content increased and the higher heating value of the bio-oil increased to 23.1 MJ/kg. In case of H-ZSM5, the fractionation experiment was more effective and resulted in a bio-oil with a water content of 11 wt%. The oxygen content of the bio-oil decreased to 37.9 wt% while both the hydrogen and carbon content increased. The higher heating value of the bio-oil increased to 23 MJ/kg, this effect can be attributed to the concentration of lignin derivatives in the bio-oil that are mainly condensed in the second condenser, the lignin derivatives contains a relative low oxygen compared to the light organic compounds. The GC/MS analysis of the water-rich liquid collected from the second condenser (see Table 7) revealed that acetic acid and furfural were the dominant components in the liquid, contributing up to 60% of the total organic fraction of the liquid collected in the second condenser. The remaining 40% of the liquid mainly consisted of ketone and phenol groups. In general, online fractionation was effective to concentrate the high calorific value compounds of the bio-oil in the first condenser and the water and oxygenated compounds in the second condenser. The online fractionation could be cheap downstream approach to improve the quality of the bio-oil. Operating the
Energies 2016, 9, 187
15 of 17
first condenser at higher temperatures generates a bio-oil with up to 90 wt% organics, which could be promising fraction for gasoline production via e.g., FCC, hydrotreatment. The decrease in the oxygen content and the increase in the carbon and hydrogen content of the bio-oil could help to reduce the amount of hydrogen needed in case of e.g., hydrotreatment of this bio-oil. The water-rich fraction of the pyrolysis liquid, obtained in the second condenser, contains up to 10 wt% light organics and could be an interesting raw stock for further extraction applications or for supercritical water gasification to produce hydrogen for hydro-treatment of the bio-oil [22,23] Table 7. Component concentration of the water-rich fraction from the second condenser.
Chemical Compound 2-Butenal Acetic acid 2-Isopropoxyethylamine Butanedial 2-Cyclopenten-1-one Furfural 1-(Acetyloxy)-2-propanone 2-Methyl-2-cyclopenten-1-one (E,E))- 6,10-dimethyl-5,9-dodecadien-2-one 4(3H)-Pyrimidinone 5-Methyl-2-furancarboxaldehyde 3-Methyl-2-cyclopenten-1-one Butyrolactone 2(3H)-Furanone 2H-Pyran-2-one 3-Methyl-1,2-cyclopentanedione Phenol 2-Methoxyphenol 2-Methylphenol 4-Methylphenol 2-Methoxy-5-methylphenol 2,4-Dimethylphenol 2,6-Dimethoxyphenol
TIC Area % H-ZSM-5
Na-Y
3.6 28.1 1.7 — 3.5 25.5 — 2.5 — — 1.9 2.3 4.1 3.7 1.6 3.0 3.2 4.1 2.1 1.6 1.8 1.6 4.0
2.8 37.1 1.3 4.9 2.7 19.3 1.8 — 2.0 1.6 1.5 1.8 3.1 2.8 1.2 2.2 2.4 3.1 1.6 1.2 2.3 — 3.0
4. Conclusions The bench scale unit can be operated successfully for in situ catalytic flash pyrolysis of biomass. The unit demonstrated good operating stability and flexibility for varying operating conditions. The bio-oil recovery system provides the possibility for fractionation of different bio-oil fractions and adds to the degrees of freedom available for tailoring bio-oil properties. Zeolite-based acidic catalysts can perform deoxygenation, at the cost of reduced organic liquid yield because of secondary cracking reaction of vapors leading to formation of gases. Nevertheless, the gas or char generated during the catalytic pyrolysis can supply a part of the heat required for the pyrolysis step in an integrated process or it can be used for biomass drying. In general, increasing the zeolite acidity favored the formation of desirable compounds such as phenols and furans and reduced the most deleterious components in the bio-oil such as acids. However some of the undesired components in the bio-oil such as ketones, and aldehydes increased with increasing zeolite acidity. Online fractionation via staged condensation was a successful technique to produce high calorific value bio-oil with very low water content of 10 wt%. It can be concluded that fractionation is a promising cheap downstream approach to control the bio-oil properties. The integrated process of catalytic pyrolysis and in-situ fractionation can be used for the production of tailor made biofuels and/or biochemicals. The good quality bio-oil produced by integrated catalytic pyrolysis of biomass
Energies 2016, 9, 187
16 of 17
may be used as a co-feedstock for conventional refineries and existing infrastructure can be used for the production of sustainable transportation fuels. Acknowledgments: This work was funded by the Dutch Technology Foundation STW and carried out at the University of Twente, The Netherlands. Author Contributions: Ali Imran, Eddy A. Bramer, Kulathuiyer Seshan and Gerrit Brem conceived and designed the experiments. Ali Imran conducted the experiments, analysis of the data, and writing of the manuscript. Eddy Bramer, Kulathuiyer Seshan and Gerrit Brem revised the manuscript. Conflicts of Interest: The authors declare no conflict of interest.
References 1. 2. 3. 4. 5. 6. 7. 8.
9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
Nokkosmäki, M.I.; Kuoppala, E.T.; Leppämäki, E.A.; Krause, A.O.I. Catalytic conversion of biomass pyrolysis vapours with zinc oxide. J. Anal. Appl. Pyrolysis 2000, 55, 119–131. [CrossRef] Mohebali, G.; Ball, A.S. Biocatalytic desulfurization (BDS) of petrodiesel fuels. Microbiology 2008, 154, 2169–2183. [CrossRef] [PubMed] Oasmaa, A.; Czernik, S. Fuel Oil Quality of Biomass Pyrolysis Oils—State of the Art for the End Users. Energy Fuels 1999, 13, 914–921. [CrossRef] Scholze, B.; Meier, D. Characterization of the water-insoluble fraction from pyrolysis oil (pyrolytic lignin). Part I. PY–GC/MS, FTIR, and functional groups. J. Anal. Appl. Pyrolysis 2001, 60, 41–54. [CrossRef] Sharma, R.K.; Bakhshi, N.N. Catalytic upgrading of biomass-derived oils to transportation fuels and chemicals. Can. J. Chem. Eng. 1991, 69, 1071–1081. [CrossRef] Bridgwater, A.V. Catalysis in thermal biomass conversion. Appl. Catal. A: Gen. 1994, 116, 5–47. [CrossRef] Maggi, R.; Delmon, B. Characterization and upgrading of bio-oils produced by rapid thermal processing. Biomass Bioenergy 1994, 7, 245–249. [CrossRef] Maggi, R.; Delmon, B. A Review of Catalytic Hydrotreating Processes for the Upgrading of Liquids Produced by Flash Pyrolysis. In Studies in Surface Science and Catalysis; Froment, G.F., Delmon, B., Grange, P., Eds.; Elsevier: Amsterdam, The Netherlands, 1997; Volume 106, pp. 99–113. Piskorz, J.; Majerski, P.; Radlein, D.; Scott, D.S. Conversion of lignins to hydrocarbon fuels. Energy Fuels 1989, 3, 723–726. [CrossRef] Bridgwater, A.V. Review of fast pyrolysis of biomass and product upgrading. Biomass Bioenergy 2012, 38, 68–94. [CrossRef] Bridgwater, A.V.; Cottam, M.L. Opportunities for biomass pyrolysis liquids production and upgrading. Energy Fuels 1992, 6, 113–120. [CrossRef] Pattiya, A.; Titiloye, J.O.; Bridgwater, A.V. Fast pyrolysis of cassava rhizome in the presence of catalysts. J. Anal. Appl. Pyrolysis 2008, 81, 72–79. [CrossRef] Lappas, A.A.; Samolada, M.C.; Iatridis, D.K.; Voutetakis, S.S.; Vasalos, I.A. Biomass pyrolysis in a circulating fluid bed reactor for the production of fuels and chemicals. Fuel 2002, 81, 2087–2095. [CrossRef] Sadeghbeigi, R. Chapter 1—Process Description. In Fluid Catalytic Cracking Handbook, 2nd ed.; Sadeghbeigi, R., Ed.; Gulf Professional Publishing: Houston, TX, USA, 2000; pp. 1–39. Raveendran, K.; Ganesh, A.; Khilar, K.C. Influence of mineral matter on biomass pyrolysis characteristics. Fuel 1995, 74, 1812–1822. [CrossRef] Westerhof, R.J.M.; Kuipers, N.J.M.; Kersten, S.R.A.; van Swaaij, W.P.M. Controlling the Water Content of Biomass Fast Pyrolysis Oil. Ind. Eng. Chem. Res. 2007, 46, 9238–9247. [CrossRef] Effendi, A.; Gerhauser, H.; Bridgwater, A.V. Production of renewable phenolic resins by thermochemical conversion of biomass: A review. Renew. Sustain. Energy Rev. 2008, 12, 2092–2116. [CrossRef] Oasmaa, A.; Sipilä, K.; Solantausta, Y.; Kuoppala, E. Quality Improvement of Pyrolysis Liquid: Effect of Light Volatiles on the Stability of Pyrolysis Liquids. Energy Fuels 2005, 19, 2556–2561. [CrossRef] Biagini, E.; Cioni, M.; Tognotti, L. Development and characterization of a lab-scale entrained flow reactor for testing biomass fuels. Fuel 2005, 84, 1524–1534. [CrossRef] Brown, A.L.; Dayton, D.C.; Nimlos, M.R.; Daily, J.W. Design and Characterization of an Entrained Flow Reactor for the Study of Biomass Pyrolysis Chemistry at High Heating Rates. Energy Fuels 2001, 15, 1276–1285. [CrossRef]
Energies 2016, 9, 187
21.
22. 23. 24.
17 of 17
Imran, A.; Bramer, E.A.; Seshan, K.; Brem, G. High quality bio-oil from catalytic flash pyrolysis of lignocellulosic biomass over alumina-supported sodium carbonate. Fuel Process. Technol. 2014, 127, 72–79. [CrossRef] Demirbas, A. Hydrogen Production from Biomass via Supercritical Water Gasification. Energy Sources Part A Recovery Util. Environ. Effects 2010, 32, 1342–1354. [CrossRef] Guo, Y.; Wang, S.Z.; Xu, D.H.; Gong, Y.M.; Ma, H.H.; Tang, X.Y. Review of catalytic supercritical water gasification for hydrogen production from biomass. Renew. Sustain. Energy Rev. 2010, 14, 334–343. [CrossRef] Atutxa, A.; Aguado, R.; Gayubo, A.G.; Olazar, M.; Bilbao, J. Kinetic Description of the Catalytic Pyrolysis of Biomass in a Conical Spouted Bed Reactor. Energy Fuels 2005, 19, 765–774. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
