ORIGINAL PAPER Catalytic gasification of pyrolytic oil from tire

0 downloads 0 Views 338KB Size Report
Keywords: tire oil, gasification, pyrolysis, dolomite, catalyst .... they are produced, and use the formed process gas for electricity and heat ... On the other hand, all compounds pre- sented in ... 1-Methyl-7-propan-2-ylphenanthrene. 390. Material and methods. Materials .... ensure that also the upper part of the catalyst layer is.
Chemical Papers 67 (12) 1504–1513 (2013) DOI: 10.2478/s11696-013-0371-3

ORIGINAL PAPER

Catalytic gasification of pyrolytic oil from tire pyrolysis process Lukáš Gašparovič, Lukáš Šugár, Ľudovít Jelemenský, Jozef Markoš* Institute of Chemical and Environmental Engineering, Faculty of Chemical and Food Technology, Slovak University of Technology in Bratislava, 812 37, Bratislava, Radlinského 9, Slovakia Received 14 November 2012; Revised 24 January 2013; Accepted 27 January 2013

The present work deals with thermo-catalytic decomposition of pyrolytic oil from the scrap tire pyrolysis process. Such oil can be used as a model tar in an experimental study of tar removal from pyrolysis or gasification process gas. Several experiments under different conditions were carried out in order to determine conditions of the gasification and pyrolysis processes. Influence of the oil to steam ratio, temperature, and of the presence of dolomite catalyst was studied. Addition of water steam has positive effect on the hydrogen content in the outgoing process gas as well as on the conversion of the injected oil. The catalytic gasification experiment in a quasi steady state produced process gas with the composition: 61 mole % of H2 , 6.4 mole % of CO, and 11.7 mole % of CH4 . At temperatures lower than 800 ◦C, the amount of process gas decreased resulting also in a decrease of the oil conversion. A comparison of gasification experiments using fresh calcined dolomite with experiments proceeding with regenerated dolomite was done under the same conditions. There was a decrease in the process gas volumetric flow when regenerated catalyst was used. c 2013 Institute of Chemistry, Slovak Academy of Sciences  Keywords: tire oil, gasification, pyrolysis, dolomite, catalyst

Introduction One of the main interests in the field of energy production is the replacement of standard fossil fuel reserves utilization by new alternative resources connected with the decrease of green house gases emissions, specifically of carbon dioxide. Therefore, large effort has been made in waste conversion to energy research concerning thermal processes. Thermal decomposition processes such as gasification or pyrolysis offer an environmentally attractive method of the decomposition of a wide range of wastes, including scrap tires. There are three main products of the scrap tire pyrolysis process: combustible gas (5–20 mass %), liquid (tar) (40–60 mass %), and solid char (30–40 mass %). Their relative ratio depends mainly on the process conditions (Juma et al., 2007). The amount of condensable vapors, called tars contained in the gaseous phase plays a crucial role in the further processing of process gas. The composition and amount of tars are strongly dependent on the process conditions,

such as temperature, residence time, addition of oxygen or steam water (García et al., 1999; Hosoya et al., 2008; Gilbert et al., 2009; Zhang et al., 2009), etc., and on the origin of the tar sample (Milne et al., 1998; García et al., 1999). Since tars are often regarded as catalyst poisons (Zhang et al., 2004; Schmidt et al., 2011) or cause problems in process gas burning (Schmidt et al., 2011), several ways of tar removal have been proposed (Anis & Zainal, 2011). The basic and simplest way of tar removal from process gas is their condensation followed by a mechanical wet-low temperature washing process (Han & Kim, 2008). A disadvantage of this gas cleaning method is the loss of the energy potential of cleaned gas by its cooling. Moreover, if condensed tars are not directly processed, they have to be stored. Open reservoirs which represent environmental threat are often used. If process gas is needed at high temperature for further processing, the most often used method of tar removal is its thermo-catalytic destruction in a thermo-catalytic reactor. Dolomite is one of the most frequently used catalysts for its good

*Corresponding author, e-mail: [email protected]

Unauthenticated Download Date | 9/24/15 11:23 PM

L. Gašparovič et al./Chemical Papers 67 (12) 1504–1513 (2013)

catalytic properties, price, and availability, but also other types of catalysts such as nickel based, alkali metal based, and some new types of catalysts based on rhodium, ruthenium, palladium, or platinum addition are employed (Sutton et al., 2001; He et al., 2009; Yu et al., 2009). Some authors have studied thermo-catalytic destruction of real tars contained in the process gas after gasification (Gilbert et al., 2009; Zhang et al., 2009) or in a model gas (García et al., 1999). Due to the complexity of tars (oxygenated compounds, phenolics, heterocyclic ethers, poly-aromatic hydrocarbons, etc.) (Milne et al., 1998; Han & Kim, 2008; Li & Suzuki, 2009), some authors used model compounds of tar such as benzene, toluene, or naphthalene instead of real tars in their experiments (Abu El-Rub et al., 2008; Swierczynski et al., 2008; L  amacz et al., 2009; Li et al., 2009, 2010; Yoon et al., 2010). According to these authors, an increase of temperature helps to convert tars as well as to increase the amount and improve the composition of the formed process gas while the amount of hydrogen and carbon monoxide increase and the amount of methane decreases due to thermal cracking. In the work of Yoon et al. (2010), the extra addition of water during the toluene catalytic gasification at 800 ◦C resulted in an increase of hydrogen production while conversion of toluene decomposition reached 100 %. All these studies solved the problem with tars or model compounds removal when they are present in low concentrations in the gaseous phase (high temperature). There is a lack of publications dealing with processing of condensed liquid tars. It might be caused by their nature. Often, a high viscosity and heterogeneous mixture of high boiling point hydrocarbons with a wide distillation curve makes real liquid tars an unfavorable sample for gasification or pyrolysis experiments. One of the few paper outcomes regarding thermal processing of a real tar sample was published by García et al. (1999). Catalytic gasification of three different feeds such as the petroleum distillation residue (PDR), coke oven tar (COT), and coal gasification tar (CGT) was performed in a bench scale. Calcium oxide was used as a catalyst. The gasification process was divided into two steps: thermal decomposition of tar causing gas and coke formation and steam gasification of coke and non decomposed tar through water gas, CO shift, and steam reforming reactions. Temperature of above 750 ◦C and higher feeding rates (steam to tar ratio) led to complete tar conversion since the gasification rate was higher than that of the coke formation. Below this temperature, the gas yield varied due to solid carbon deposition on the catalyst surface. During the pyrolysis experiments, thermal stability of tars decreased in the order: CGT, COT, PDR, which, according to the authors, correlates with higher aromaticity, low contents of alkylated and oxygenated compounds of CGT, as well as with higher content of aliphatic compounds in the inverse order. Gasification

1505

experiments at above 750 ◦C showed that the reactivity of tars decreased in the order: PDR, CGT, COT. However, H2 and CO yields were higher for the less reactive tars showing that gasification of deposited carbon is the predominant factor. Structural characterization of the tars showed higher aromaticity of the less reactive tars thus predicting a higher amount of deposited carbon that can be directly gasified by steam to H2 and CO. The main goal of this work was to study the possibilities of thermo-catalytic decomposition of pyrolysis oil derived from pyrolysis of scraped tires. Thermal decomposition of waste tires is a common and suitable way of their removal although it is more convenient to build small local pyrolysis units for tire decomposition to avoid their collection and transportation for long distances. When the gasification of tire proceeds, except of the desired process gas, some tars and coke are formed as well. Their collection and further processing into further products with higher added value seems not economic if the amount of the produced tars is in the range of hundred tons per year, what is the case of a small country like Slovakia. Because tars are qualified as “dangerous waste”, there are severe legislation restrictions for their transportation and storage. Therefore it is more convenient to decompose tars into process gas directly in the local pyrolytic unit where they are produced, and use the formed process gas for electricity and heat production. Another reason why condensed pyrolytic oil was chosen is the search for a more efficient method of oil utilization than ordinary combustion. Real tar sample is a wide mixture of several hydrocarbons such as aromatic, heterocyclic hydrocarbons, etc., and cannot be simply substituted by one or two model compounds (toluene, naphthalene) as it is often the case in literature. On the other hand, all compounds presented in the tars react differently with the compounds present in the gas (hydrogen, water steam, oxygen, CO2 , CO) and among themselves (alkylation, hydrogenation, dehydrogenation, hydratation, dehydratation, etc.) under pyrolysis conditions. The experiments were focused on setting of such experimental conditions which would ensure the decomposition of all present compounds as a whole. This information can be useful also by researchers dealing with pyrolytic tars originating from the pyrolysis or gasification of different types of biomass, which have similar composition and properties as tars coming from the pyrolysis of waste tires. Therefore, an experimental fixed bed catalytic reactor was designed and its applicability for oil decomposition to process gas (oriented on the production of H2 , CO, and CH4 ) was studied. The aim of the paper was not to determine the exact mechanism of the tar destruction process but to show the possibilities, restrictions, and potential problems connected with pyrolytic oil (tars) processing.

Unauthenticated Download Date | 9/24/15 11:23 PM

1506

L. Gašparovič et al./Chemical Papers 67 (12) 1504–1513 (2013)

Table 1. Elemental composition of the tar sample Component

Na

Ca

Ha

Sa

Ob

HHVc /(MJ kg−1 )

Content w/mass %

0.79

85.83

11.07

0.90

1.41

40.8

a) Determined by elemental analysis, measured on a Vario Macro Cube (Elementar Analysen Systeme, Germany); b) calculated from the difference to 100 %; c) higher heating value determined by a Calorimetric bomb (Fire Testing Technology, UK). Table 2. Major components of the tar detected by GC-MS No.

Formula

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

C6 H 6 C7 H 8 C8 H10 C8 H10 C8 H 8 C10 H16 C9 H12 C10 H18 C9 H10 C9 H12 C10 H14 C10 H16 C10 H12 C10 H12 C10 H8 C11 H12 C12 H14 C7 H5 NS C11 H10 C10 H8 C12 H15 N C18 H24 N2 C12 H12 C13 H14 C14 H16 C18 H18

Molar mass 78.1 92.1 106.2 106.2 104.2 136.2 120.2 138.2 118.2 120.2 134.2 136.2 132.2 132.2 128.2 144.2 158.2 135.2 142.2 128.2 173.1 268.4 156.2 170.3 184.3 234.3

Possible compounds Benzene Toluene Ethylbenzene p-Xylene Styrene 1,4-Dimethyl-4-vinylcyclohexane 1-Ethyl-2-methylbenzene 3-Methyl-6-(propan-2-yl)cyclohexene Isopropenylbenzene 1,2,4-Trimethylbenzene 1-Isopropyl-2-methylbenzene 1-Methyl-4-(1-methylethenyl)cyclohexene 1-Methyl-4-(prop-1-en-2-yl)benzene 1-Ethenyl-3-ethylbenzene Naphthalene 2,3-Dimethyl-1H-indene 1,4-Diisopropenylbenzene 1,3-Benzothiazole 2-Methylnaphthalene Azulene (bicyclo[5.3.0]decapentaene) 2,4,4-Trimethyl-3,4-dihydroquinoline 1,4-Benzenediamine, N-(1,3-dimethylbutyl)-N  -phenyl2,6-Dimethylnaphthalene 1,6,7-Trimethylnaphthalene, 7-Ethyl-1,4-dimethylazulene 1-Methyl-7-propan-2-ylphenanthrene

Material and methods Materials Oil used in the present work represents the distillated heavy fraction of pyrolysis oil originating from a scrap tire pyrolysis unit. The sample was a deep dark brown liquid. Pyrolysis oil characterization by elemental analysis and high heating value (HHV) is shown in Table 1. Qualitative analysis of oil was done on a GC-MS equipment (Agilent Technologies 7890A GC System, 5975C inert XL MSD with a Triple-Axis Detector, USA) and is summarized in Table 2 (results were obtained considering the comparison of real mass spectra with the NIST 0.8 library). Separation was made on a 30 m × 250 µm × 0.25 µm i.d. fused silica capillary column J&W HP-5MS using a 7890A gas chromatograph Agilent Technologies. The GC oven temperature was held at 40 ◦C for 5 min, increased by 5 ◦C min−1 to 100 ◦C, and then programmed to 300 ◦C at 20 ◦C min−1 . The injector temperature was 260 ◦C in

BP/ ◦C 80.1 110.6 136 138.4 146 160.8 165.2 165.2 169 171 177 178 182 191.5 217 225 228 230 241 242 252 260 267 285 299.1 390

the split mode (split ratio of 70 : 1). He, 3 mL min−1 , was used as the carrier gas. The end of the column was directly inserted into the ion source of the Agilent Technologies model 5975C series mass selective detector (MSD) operated in the electron impact ionization mode. Qualitative analysis of the pyrolysis oil composition mentioned in Table 2 has just an informative character showing the complexity of tar composition. Such a wide mixture of several hydrocarbons such as aromatic, heterocyclic hydrocarbons, etc., cannot be simply substituted by one or two model compounds. Moreover, direct utilization of tar without any treatment is problematic due to its physical character and the separation of individual compounds is pretty costly. Therefore, the consequent thermal decomposition of the tars to the process gas in a small pyrolytic unit seems to be an effective way of their disposal. According to literature reviews, the cheapest catalyst with sufficient activity for tar decomposition is calcined dolomite (Sutton et al., 2001; Han & Kim, 2008). Therefore, our experiments were carried out in

Unauthenticated Download Date | 9/24/15 11:23 PM

L. Gašparovič et al./Chemical Papers 67 (12) 1504–1513 (2013)

1507

Table 3. Specific surface area of used catalyst

Material

Raw dolomite Calcined dolomite

SORPTOMATIC

POROSIMETER

Surface area m2 g−1

Surface area Porosity m2 g−1 %

4.30 10.49

6.78 11.67

22.6 28.1

the presence of calcined dolomite as the catalyst. Raw dolomite ore, a mixture of CaCO3 and MgCO3 , was taken from a stone pit in Malé Kršteňany, Slovakia. The desired fraction between 7 mm and 4 mm was used. Since calcined dolomite (composed of CaO and MgO) has better catalytic properties, each batch of the catalyst was calcined at 1000 ◦C before the experiment for four hours while purged with air as a carrier gas until the concentration of CO2 in the outgoing gas was practically zero. The amount of evolved CO2 during the calcination determined by thermogravimetric analysis is 48 % of the original mass of dolomite. Specific surface area and porosity were estimated from the adsorption isotherm of nitrogen at the temperature of liquid nitrogen by a Sorptomatic 1900 and by mercury porosimetry using a Porosimeter P2000 (both Carlo Erba, Milan, Italy) and its values are presented in Table 3. From data reported in Table 3 it is evident that calcination had a positive effect on the surface area and the catalyst porosity increase. Surface area of calcined dolomite is in accordance with 10 m2 g−1 reported by García et al. (1999) (estimated by the same adsorption method). Dolomite which comes from the only source of crushed dolomite in Slovakia was used. If a cheap catalyst can be used, the use of a catalyst from a domestic source and finding optimal conditions for its use for tar decomposition into process gas is the best choice. Therefore, the presence of different metal traces (like iron, nickel etc.) and their influence on the qualitative tar decomposition was not studied in this paper. Experimental equipment and products analysis A vertical tubular reactor made of Cr/Ni stainless steel placed inside an electric furnace (CLASIC CZ, Czech Republic) was the main part of the whole experimental equipment. Its inner diameter was 32 mm and the total length was 450 mm. The layer of dolomite inside the reactor was placed 50 mm under the top of the reactor, due to the furnace temperature gradient, to ensure that also the upper part of the catalyst layer is placed in the area with the desired temperature. The mass of dolomite used in the experiments was 420 g (before calcination). Three inlets providing the feed of oil, water, and gas (air, pure oxygen, or pure nitrogen) were located at the top of the reactor. Oil and water

Fig. 1. Flow diagram of the experimental equipment: 1. water reservoir, 2. water piston pump, 3. oil reservoir, 4. oil piston pump, 5. air, N2 , or O2 pressure vessel, 6. mass flow meter, 7. reactor, 8. reactor oven, 9. oven controller, 10. pressure indicator, 11. cooling and cleaning gas system, 12. GC carrier gas – Ar, 13. gas chromatograph, 14. flow meter, I. outgoing process gas, II. cleaned and cooled gas, II’. branch to gas chromatograph, III. waste gas from chromatograph, IV. exhaust gas.

were fed to the reactor in the liquid phase by piston pumps, preheated and partially evaporated in the tubing and totally evaporated in the reactor. Gases such as air, pure oxygen, or nitrogen were fed (if needed) from an air compressor or pressure vessel through a mass flow meter. To prevent reactor intersection due to coke formation inside the reactor, the top of the reactor was equipped with a pressure indicator. The bottom of the reactor was equipped with just one outlet for the gaseous phase. Outgoing gas from the bottom of the reactor passed through the gas cooling and purification system providing condensation of residual tars and separation of solids and aerosols. On the whole, there are five separatory cooling traps equipped with cotton wool filters. The first one was cooled by tap water (10–12 ◦C); the other two were put into an ice bath (around 0 ◦C) and the last two are placed inside a glycol bath with the temperature of –25 ◦C. The volume of the outgoing gas was measured by a volumetric flow meter and led to the gas exhaust. A small part of the gas was pumped through a bypass loop to the gas chromatograph. Laboratory equipment is described on a detailed scheme in Fig. 1. The outgoing gas mixture from the reactor, after passing through the cooling and cleaning systems, was analyzed using a gas chromatograph GSM MicroBox III (SLS Micro Technology, Germany). The following chromatographic conditions were set: separation column Carbosphere able to separate H2 , N2 , CO, CH4 ,

Unauthenticated Download Date | 9/24/15 11:23 PM

1508

L. Gašparovič et al./Chemical Papers 67 (12) 1504–1513 (2013)

Table 4. List of experimental runs Reactor temperature

Oil

Steam

◦C

g min−1

g min−1

800 800 800 800 700 600 800 800 800

0.45 0.46 0.46 0.46 0.46 0.45 0.46 0.42 0.44

0.50 0.10 0 0.50 0.5 0.5 0.50 0.50 0.49

Run No.

E1. E2. E3. E4. E5. E6. E7. E8. E9.

Catalytic gasification Catalytic gasification Catalytic pyrolysis Non-catalytic gasification Catalytic gasification Catalytic gasification Catalytic gasification Catalytic gasification Catalytic gasification

Oil/Steam ratio

Reactor packing

0.90 4.60 – 0.92 0.92 0.90 0.92 0.84 0.90

dolomite dolomite dolomite Al2 Oa 3 dolomite dolomite dolomite regenerated dolomite regenerated dolomite

a) Aluminium oxide structured pellets (originally intended for the filtration of molten metal alloys, e.g. characterized by high bed void fraction and high thermo-mechanic-chemical stability with very low specific surface area and practically no catalytic effect).

N2 O, CO2 , C2 H2 , C2 H4 , and C2 H6 ; column length of 65 cm; carrier gas: He; flow: 500 µL min−1 ; analysis time: 70 s; start temperature: 50 ◦C; stop temperature: 240 ◦C; heating rate: 5 ◦C s−1 ; heat delay time: 10 s; detector: thermal conductivity detector (TCD); retention times: H2 (5 s), N2 (11.9 s), CO (15.2 s), CH4 (26 s), CO2 (37.6 s). Quantification of process gas composition was estimated by direct multipoint calibration of the gas chromatograph.

Results and discussion Several experiments were carried out under different reaction conditions (temperature, usage of steam, catalyst) in order to obtain more detailed information about the gasification and pyrolysis processes. Also, the influence of the utilization of regenerated dolomite on the gasification process was analyzed. Regeneration of fouled dolomite was done by the oxidation of coke at high temperatures (800 ◦C) in a stream of air. All experiments were carried out with an injection of 90 mL min−1 of air. The amount of fed oxygen represents approximately 1 % of the total stoichiometric amount necessary for total combustion of the fed oil. Conditions in each experimental run are presented in Table 4. Influence of water and catalyst Four different experiments were carried out under different process conditions in order to obtain reasonable information about the influence of water and catalyst on experiments E1–E4, reported in Table 4. Time profile of every process gas component in each experimental run and the total volumetric flow of the process gas in each experiment are shown in Figs. 2 and 3. From the hydrogen production point of view it can be expressly seen that if more water is in contact with carbohydrate vapors in the presence of the catalyst, the yield of the produced hydrogen increases. The influence of used dolomite on the production of hydro-

gen is more important than that of the addition of water, as can be seen when the first catalytic gasification (E1) and non-catalytic gasification (E4) are compared. The difference in the produced amount of hydrogen in these two experiments is higher than 30 mole %. Moreover, from Fig. 3 it is evident that the volumetric flow of the produced gas increased twice when the catalyst was used compared with the noncatalytic process. Feeding of water has two reasons: i) the presence of water steam increases the hydrogen production through the hydrocarbons steam reforming reaction Eq. (1) Cn Hm + nH2 O → nCO + (n + m/2) H2

(1)

ii) addition of water slows down the rate of the catalyst deactivation by coking according to the steam gasification reaction Eq. (2) C + H2 O → CO + H2

(2)

Again, when experiment E1 is compared with experiments E2 and E3 it can be concluded that in the absence of water or its insufficient feed, the rate of catalyst deactivation is higher and consequently, the production of hydrogen decreases. The production of carbon monoxide is also dependent on the amount of added water. Due to the high temperature inside the reactor, the reaction of hydrocarbons steam reforming Eq. (1), which is responsible for hydrogen and carbon monoxide production, can take place. On the other hand, the excess steam can react with produced carbon monoxide due to the water gas shift reaction while carbon dioxide and hydrogen are produced by reaction in Eq. (3) as it can be seen in the CO2 production during E1. CO + H2 O → CO2 + H2

(3)

The water gas shift reaction is strongly influenced by the presence of a catalyst. Experiments E1 (with catalyst) and E4 (without catalyst) were performed

Unauthenticated Download Date | 9/24/15 11:23 PM

L. Gašparovič et al./Chemical Papers 67 (12) 1504–1513 (2013)

1509

Fig. 2. Comparison of process gas composition under different experimental conditions: E1 (squares), E2 (circles), E3 (triangles), E4 (diamonds).

flow of the process gas rich in hydrogen and carbon monoxide. The decrease of hydrogen concentration in the process gas indicates that the efficiency of the steam reforming reaction of methane (Eq. (4)) decreases as the catalyst is deactivated, which is also confirmed by the slight increase of the methane concentration. CH4 + H2 O → CO + 3H2

Fig. 3. Comparison of volumetric flow of produced process gas under different experimental conditions: E1 (squares), E2 (circles), E3 (triangles), E4 (diamonds).

with the same oil and water loading of the reactor. The time profiles of process gas composition and its overall volumetric flow are depicted in Figs. 2 and 3. From the figures it is evident that the experiment with a catalyst provided more than twice higher volumetric

(4)

Regarding the stoichiometry of the methane reforming reaction, a small decrease in the methane consumption resulted in a more visible increase of the hydrogen production. From Figs. 2 and 3 it is evident that the volumetric flow and composition of the produced process gas are varying with time. It is caused by the catalyst fouling by coke formed on its surface. Therefore, the reactor is working in a non steady state. However, an observable period of time is applied (time period between 50 min and 200–250 min of experimental time), during which the process can be assumed to proceed in a quasi steady state, and the results of individual experiments can be compared. In Table 5, such a comparison for the 100th minute of the experimental time

Unauthenticated Download Date | 9/24/15 11:23 PM

1510

L. Gašparovič et al./Chemical Papers 67 (12) 1504–1513 (2013)

Table 5. Results of experiments E1–E4 during the quasi steady state Volumetric flow Component content/mole %

L min−1

Run

E1 E2 E3 E4

∆c H

Collected tara

Gas yieldb

H2

CO

CH4

CO2

N2

Exp.

Calc.

MJ kg−1

%

L g−1

61 49 50 24.7

6.4 3.5 3.3 4.4

11.7 22.5 27.2 24

12.7 2.5 0 3.1

8.8 13.6 15.7 21.9

0.76 0.46 0.44 0.31

0.82 0.48 0.46 0.33

20.6 32.1 34.8 27.1

14 27 17 36

1.62 0.91 0.87 0.69

a) Ratio between the amount of collected tar in separatory traps and the total amount of fed oil; b) ratio between the amount of total produced process gas (amount of nitrogen was subtracted) to the difference of the fed oil and collected tar; ∆c H – heat of combustion calculated according to Lide (1993). Table 6. Results of experiments E1, E5, and E6 during the quasi steady state Volumetric flow T Run

E1 E5 E6

Component content/mole %

L min−1

Collected tara

Gas yieldb

◦C

H2

CO

CH4

CO2

N2

Exp.

Calc.

%

L g−1

800 700 600

61 56.4 44

6.4 0.8 0.2

11.7 9.7 8.8

12.7 0 0

8.8 16.1 38.7

0.76 0.44 0.19

0.82 0.46 0.18

14 42 43

1.62 1.24 0.35

a) Ratio between the amount of collected tar in separatory traps and the total amount of fed oil; b) ratio between the amount of total produced process gas (amount of nitrogen was subtracted) to the difference of the fed oil and collected tar.

is reported. Heating values of the produced gas were calculated according to Lide (1993) and they are lower due to the absence of the C2 fraction which was not measured. Assuming an ideal gas sample and considering the known amount of injected nitrogen and the content of nitrogen in the process gas, the total volumetric flow of the outgoing process gas was calculated. A comparison of the calculated volumetric flow with the experimental flow is shown in Table 5. The amount of collected tars indicates a conversion of oil, where the lower value of separated tars represents higher conversion. After the catalytic pyrolysis, significantly higher deactivation of the used dolomite (visibly thicker coke and non-reacted matter layer on the catalyst surface and inside the catalyst) was observed than after the catalytic steam gasification. Therefore it can be assumed that although the amount of collected tars in the process of pyrolysis reported in Table 5 is comparable with the value obtained from the first experiment, the total conversion is definitely lower due to the more significant coking inside the reactor during the pyrolysis. Also the yield of process gas in dependence on the fed oil was considered, see Table 5. Influence of temperature

on the temperature influence on the catalytic gasification of tars was carried out. Experimental conditions of three experiments, E1, E5, and E6, performed at different temperatures are presented in Table 4. Results of the measurements from the quasi steady state in the 100th minute of the process are reported in Table 6. A decrease of the catalytic reactor temperature leads to the decrease of the amount of formed combustible gases. Practically, the presence of carbon dioxide and low concentration of carbon monoxide in the process gas from experiments E5 and E6 result from low temperature in the reactor which is not favorable for the water gas shift reaction or the steam gasification reactions (Eqs. (1) and (2)). These reactions start to take place at temperatures higher than 800 ◦C. Lower temperature has not only negative influence on the gas composition but also on the overall volumetric gas flow produced. Higher temperature has also a positive effect on the amount of destructed tars, which can be seen from the comparison of the amounts of tar collected in separatory traps. At higher reaction temperatures (750 ◦C), the reaction between the formed coke and carbon dioxide Eq. (5) takes place in the reactor (Žajdlík et al., 2000) simultaneously with reaction Eq. (2) which has a positive influence on the catalyst operational life.

Temperature is a very important parameter influencing the decomposition of tars. Therefore, a study

Unauthenticated Download Date | 9/24/15 11:23 PM

C + CO2 → 2CO

(5)

L. Gašparovič et al./Chemical Papers 67 (12) 1504–1513 (2013)

1511

Fig. 4. Comparison of process gas composition: E1 (squares), E7 (circles), E8 (triangles), E9 (diamonds).

Catalyst regeneration Four steam gasification catalytic experiments were carried out under similar conditions in order to verify the reproducibility of measurements and to compare catalytic impact of fresh and regenerated catalyst. Regeneration of coked catalyst after the experiment was done by air at high temperature (800 ◦C). All impurities on the catalyst surface were supposed to be combusted in the presence of oxygen. According to Table 4, experiments E1 and E7 were performed with fresh calcined dolomite. Experiments E8 and E9 were carried out with regenerated dolomite from the experimental run E7 (catalyst after the first regeneration was used in E8 and that after the second regeneration in E9). Gas composition and volumetric flows of the gas phase are reported in Figs. 4 and 5, respectively. Results of all compared experimental runs show similar trends for each produced compound. The higher production of hydrogen in E7 and E8 can be a consequence of a higher number of catalytic active centers. A slightly steeper decrease of the hydrogen production in E9 can be caused by permanent deactivation of some catalytic active centers due to the re-use of the catalyst. This leads to the assumption

Fig. 5. Comparison of volumetric flow of produced process gas: E1 (squares), E7 (circles), E8 (triangles), E9 (diamonds).

that it is possible to use regenerated dolomite only for a limited number of gasification cycles. The production of carbon monoxide of around 5–6 mole % seems to be almost the same in all experiments as well as the production of methane of 11–13 mole %. Volumetric flow of the produced gas in experiments E1 and E7

Unauthenticated Download Date | 9/24/15 11:23 PM

1512

L. Gašparovič et al./Chemical Papers 67 (12) 1504–1513 (2013)

Table 7. Results of experiments E1, E7, E8, and E9 during the quasi steady state in the 150th minute Volumetric flow Component content/mole %

L min−1

Run

E1. E7. E8. E9.

fresh 1. reg. 2. reg.

∆c H

Collected tara

H2

CO

CH4

CO2

N2

Exp.

MJ kg−1

%

60.5 60.6 61.5 54.5

6.0 5.2 5.6 5.7

12.0 13 11.7 11.9

11.8 9.2 9.2 9.8

8.4 8.8 8.8 9.2

0.74 0.71 0.68 0.66

22.6 25.0 25.2 26.4

14 11 18 18

a) Ratio between the amount of collected tar in separatory traps and the total amount of fed oil; ∆c H – heat of combustion calculated according to Lide (1993).

was approximately the same. However, a decrease in the process gas volumetric flow can be observed when regenerated catalyst was used. Therefore it is necessary to find an optimal number of regenerations. Gas composition and volumetric flows in the 150th minute (as E9 was still not stable in the 100th min) expected to be in the range of the quasi steady state are reported in Table 7. From the data reported in Table 7, also the decrease of apparent conversion indicates that a decrease of the catalytic activity of dolomite occurred while the same catalyst was repeatedly utilized in the gasification process. Although the apparent conversion in experiments E8 and E9 is the same, it has to be kept in mind that the reaction time was 10 h in E8 while it was only 6 h in E9. It is evident that longer time of catalyst loading decreases its activity. Therefore it is highly probable that if both these experiments proceeded for the same time, the apparent conversion in case of E8 would be slightly higher.

Conclusions A concept of pyrolytic oil (from thermal decomposition of scrap tires) thermal degradation has been proposed and developed based on catalytic gasification in a fixed bed catalytic reactor. There are several reasons why the pyrolytic oil from thermal pyrolysis of scrap tires was chosen (see also our notes in the introduction part): i) in pyrolytic units of scrap tires, this oil has practically no utilization because of its complex composition and physico–chemical properties; ii) further processing requires its collection from several pyrolytic units (Briens et al., 2008) increasing the costs of the whole process and therefore the oil is only combusted; iii) its conversion into process gas can increase the efficiency of the pyrolysis process; iv) pyrolytic oil represents the “model” mixture of tars produced during pyrolysis/gasification of different polymeric materials: plastics, biomass, municipal waste etc., so the results are applicable also for these materials.

A laboratory fixed bed catalytic reactor for pyrolysis and/or gasification of liquid tars (pyrolytic oil) was built. Calcined dolomite as a very cheap and sufficiently active catalyst was used. The influence of the presence of a catalyst, steam, and of temperature was studied to determine conditions maximizing the production of combustible permanent gases (hydrogen, carbon monoxide, methane) in process gas. Because of strong fouling of the catalyst, its regeneration and re-use was also tested. Simultaneous feeding of water steam showed positive effect on the hydrogen production as well as on the prevention of catalyst coking and deactivation. Positive effect of the presence of a catalyst on the production of hydrogen and carbon monoxide was also observed. Almost no carbon monoxide or carbon dioxide was produced in the gasification experiments at temperatures lower than 800 ◦C, indicating higher carbonization of the pyrolytic oil on the catalyst surface inside the reactor. Also, a rapid flow decrease of the process gas production with the decrease of the reactor temperature was observed. Dolomite as a catalyst seems to have suitable properties regarding the reproducibility of measurements. Moreover, it is an easily available material with low price. In further research, more attention should be paid to the determination of the life cycle of dolomite regeneration. Also, it is necessary to test the catalyst loading, to determine a suitable ratio of the feed to catalyst in order to reach total conversion of tars and to propose the regeneration conditions. From the catalyst regeneration point of view, optimization of the regeneration process can ensure energy savings, since regenerated catalyst can be overheated and thus the accumulated energy can be used in further process of tar decomposition. Acknowledgements. This work was supported by the Slovak Research and Development Agency under the contract LPP0230-07. This contribution is result of the project implementation: Finalizing of the National Centre for Research and Application of Renewable Energy Sources, ITMS 26240120016, supported by the Research and Development Operational Programme funded by the ERDF. Authors would like to thank Mr. Aleš Ház for GC-MS measurements (Faculty of Chemical and Food Technology, Slovak University of Technology in Bratislava).

Unauthenticated Download Date | 9/24/15 11:23 PM

L. Gašparovič et al./Chemical Papers 67 (12) 1504–1513 (2013)

References Abu El-Rub, Z., Bramer, E. A., & Brem, G. (2008). Experimental comparison of biomass chars with other catalysts for tar reduction. Fuel, 87, 2243–2252. DOI: 10.1016/j.fuel.2008.01. 004. Anis, S., & Zainal, Z. A. (2011). Tar reduction in biomass producer gas via mechanical, catalytic and thermal methods: A review. Renewable and Sustainable Energy Reviews, 15, 2355–2377. DOI: 10.1016/j.rser.2011.02.018. Briens, C., Piskorz, J., & Berruti, F. (2008). Biomass valorization for fuel and chemicals production — A review. International Journal of Chemical Reactor Engineering, 6, 1674. DOI: 10.2202/1542-6580.1674. García, X. A., Alarcón, N. A., & Gordon, A. L. (1999). Steam gasification of tars using a CaO catalyst. Fuel Processing Technology, 58, 83–102. DOI: 10.1016/s0378-3820(98)000873. Gilbert, P., Ryu, C., Sharifi, V., & Swithenbank, J. (2009). Tar reduction in pyrolysis vapours from biomass over a hot char bed. Bioresource Technology, 100, 6045–6051. DOI: 10.1016/j.biortech.2009.06.041. Han, J., & Kim, H. (2008). The reduction and control technology of tar during biomass gasification/pyrolysis: An overview. Renewable and Sustainable Energy Reviews, 12, 397–416. DOI: 10.1016/j.rser.2006.07.015. He, M. Y., Hu, Z. Q., Xiao, B., Li, J. F., Guo, X. J., Luo, S. Y., Yang, F., Feng, Y., Yang, G. J., & Liu, S. M. (2009). Hydrogen-rich gas from catalytic steam gasification of municipal solid waste (MSW): Influence of catalyst and temperature on yield and product composition. International Journal of Hydrogen Energy, 34, 195–203. DOI: 10.1016/j.ijhydene.2008.09.070. Hosoya, T., Kawamoto, H., & Saka, S. (2008). Pyrolysis gasification reactivities of primary tar and char fractions from cellulose and lignin as studied with a closed ampoule reactor. Journal of Analytical and Applied Pyrolysis, 83, 71–77. DOI: 10.1016/j.jaap.2008.06.002. Juma, M., Koreňová, Z., Markoš, J., Jelemenský, Ľ., & Bafrnec, M. (2007). Experimental study of pyrolysis and combustion of scrap tire. Polymers for Advanced Technologies, 18, 144– 148. DOI: 10.1002/pat.811. L  amacz, A., Krzto´ n, A., Musi, A., & Da Costa, P. (2009). Reforming of model gasification tar compounds. Catalysis Letters, 128, 40–48. DOI: 10.1007/s10562-008-9712-1. Li, C. S., Hirabayashi, D., & Suzuki, K. (2009). A crucial role 2− of O− 2 and O2 on mayenite structure for biomass tar steam reforming over Ni/Ca12 Al14 O33 . Applied Catalysis B: Environmental, 88, 351–360. DOI: 10.1016/j.apcatb.2008.11.004.

1513

Li, C. S., & Suzuki, K. (2009). Tar property, analysis, reforming mechanism and model for biomass gasification—An overview. Renewable and Sustainable Energy Reviews, 13, 594–604. DOI: 10.1016/j.rser.2008.01.009. Li, C. S., Hirabayashi, D., & Suzuki, K. (2010). Steam reforming of biomass tar producing H2 -rich gases over Ni/MgOx /CaO1−x catalyst. Bioresource Technology, 101, S97–S100. DOI: 10.1016/j.biortech.2009.03.043. Lide, D. R. (Ed.) (1993). Handbook of chemistry and physics (74th ed.). Boca Raton, FL, USA: CRC Press. Milne, T. A., Evans, R. J., & Abatzaglou, N. (1998). Biomass gasifier “tars”: Their nature, formation and conversion. Golden, CO, USA: National Renewable Energy Laboratory. Schmidt, S., Giesa, S., Drochner, A., & Vogel, H. (2011). Catalytic tar removal from bio syngas—Catalyst development and kinetic studies. Catalysis Today, 175, 442–449. DOI: 10.1016/j.cattod.2011.04.052. Sutton, D., Kelleher, B., & Ross, J. R. H. (2001). Review of literature on catalysts for biomass gasification. Fuel Processing Technology, 73, 155–173. DOI: 10.1016/s03783820(01)00208-9. Swierczynski, D., Courson, C., & Kiennemann, A. (2008). Study of steam reforming of toluene used as model compound of tar produced by biomass gasification. Chemical Engineering and Processing: Process Intensification, 47, 508–513. DOI: 10.1016/j.cep.2007.01.012. Yoon, S. J., Choi, Y. C., & Lee, J. G. (2010). Hydrogen production from biomass tar by catalytic steam reforming. Energy Conversion and Management, 51, 42–47. DOI: 10.1016/j.enconman.2009.08.017. Yu, Q. Z., Brage, C., Nordgreen, T., & Sj¨ ostr¨ om, K. (2009). Effects of Chinese dolomites on tar cracking in gasification of birch. Fuel, 88, 1922–1926. DOI: 10.1016/j.fuel.2009.04.020. Zhang, R. Q., Brown, R. C., Suby, A., & Cummer, K. (2004). Catalytic destruction of tar in biomass derived producer gas. Energy Conversion and Management, 45, 995–1014. DOI: 10.1016/j.enconman.2003.08.016. Zhang, Y., Kajitani, S., Ashizawa, M., & Oki, Y. (2009). Tar destruction and coke formation during rapid pyrolysis and gasification of biomass in a drop-tube furnace. Fuel, 89, 302– 309. DOI: 10.1016/j.fuel.2009.08.045. Žajdlík, R., Markoš, J., Jelemenský, Ľ., & Remiarová, B. (2000). Single coal char particle combustion in the carbon dioxide atmosphere. Chemical Papers, 54, 467–472.

Unauthenticated Download Date | 9/24/15 11:23 PM