539722
research-article2014
WMR0010.1177/0734242X14539722Waste Management & ResearchUzun and Yaman
Original Article
Thermogravimetric characteristics and kinetics of scrap tyre and Juglans regia shell co-pyrolysis
Waste Management & Research 1–10 © The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0734242X14539722 wmr.sagepub.com
BB Uzun1 and E Yaman2
Abstract The degradation kinetics of Juglans regia shell, scrap tyre and their blends were investigated using a thermogravimetric analysis method. Experiments were performed under dynamic conditions and a nitrogen atmosphere in the range 293 to 973 K at different heating rates. During pyrolysis of J. regia shell three mass loss zones were specified as removal of water, decomposition of hemicelluloses and cellulose, and decomposition of lignin. The degradation curves of scrap tyre showed merely one stage which was due to decomposition of styrene butadiene rubber. The kinetic parameters were calculated using both Arrhenius and Coats–Redfern methods. By adopting the Arrhenius method, the average value of activation energies of J. regia shell, scrap tyre and their 1 : 1 blends were found to be 69.22, 71.48 and 47.03 kJ mol−1, respectively. Additionally, by using the Coats–Redfern method, the average value of activation energies of J. regia shell, scrap tyre and their 1 : 1 blend were determined as 99.85, 78.72 and 63.81 kJ mol−1, respectively. The addition of J. regia shell to scrap tyre caused a reduction in the activation energies. The difference of weight loss was measured to examine interactions between raw materials. The maximum difference between experimental and theoretical mass loss was 5% at about 648 K with a heating rate of 20 K min−1. These results indicated a significant synergistic effect was available during co-pyrolysis of J. regia shell and scrap tyre in the high temperature region. Keywords Juglans regia shell, scrap tyre, co-pyrolysis, kinetic study, thermogravimetric analysis, synergistic effect
Introduction The utilization of human-generated waste such as plastic and tyre is one of the greatest challenges of the twenty-first century. Therefore, their recovery or disposal methods are crucial for sustainable development. The valorization of these wastes contributes to decrease the acceleration of climate change and fossil fuel consumption (Lopez et al., 2009). As known, the number of cars is increasing year by year in developing countries. Due to the increasing number of used tyres, the valorization of this waste material could solve an important environmental problem (Aguado et al., 2005). Approximately 70–85% of scrap tyres are landfilled or stockpiled. Landfilling has been favoured, because of its simplicity. However, a large space is required and the reusable resources are wasted. For example, approximately 2 billion scrap tyres have accumulated in nationwide stockpiles or uncontrolled dumps (Clark et al., 1991; EPA, 1991). The piled tyres provide breeding sites for mosquitoes, which can cause serious diseases (Kim et al., 1995). Landfilling is a potential danger due to the possibility of accidental fires that produce high emissions of hazardous gases. In addition, it requires a considerable amount of space because the volume of the tyres cannot be reduced by compaction (Islam et al., 2009). There are different alternative methods for tyre recycling such as retreading, reclaiming, incineration, grinding, etc. but they also have significant limitations.
A thermal degradation method, pyrolysis, appears to be very appropriate for complex materials, such as tyres that cannot be remoulded. In the pyrolysis process, the organic volatile matter of tyres is decomposed to low molecular weight products, which can be useful as fuels or chemicals. The inorganic components mainly from steel and the nonvolatile carbon black remain as ‘solid residue’, relatively unchanged (Rodrigues et al., 2001). Synthetic rubber materials such as scrap tyres have a high carbon content of about 70 wt.% and a low oxygen content of about 17 wt.% to produce hydrocarbon oils by pyrolysis. Additionally, as the biomass has a low carbon content of 47–51 wt.% and a high oxygen content of 42–46 wt.%, it produces highly oxygenated pyrolysis oils. Co-processing of scrap tyre with biomass could balance the carbon and oxygen amount in the feedstock,
1Anadolu
University, Faculty of Engineering, Department of Chemical Engineering, Eskisehir, Turkey 2Bilecik Seyh Edebali University, Central Research Laboratory, Bilecik, Turkey Corresponding author: BB Uzun, Anadolu University, Faculty of Engineering, Department of Chemical Engineering, Eskisehir, 26470, Turkey. Email: [email protected]
2
Waste Management & Research
Table 1. Properties of J. regia shell. wt.% Proximate analyses Moisture Ash Volatile matter Fixed carbona Component analyses Holocellulose Oil Lignin Extractives Hemicellulose Cellulosea aBy
8.06 0.33 76.38 15.23 46.13 3.29 48.11 3.78 22.18 23.95
difference.
with a strong effect on the properties of degradation products. On the other hand, biomass has lower thermal stability in comparison with scrap tyres and thus it could affect their radical degradation mechanism by promoting the degradation of synthetic macromolecules. In co-pyrolysis, the yields and composition of the products strongly depend on the treatment method, processing conditions, type of biomass and synthetic polymers (Brebu, 2010; Zhou et al., 2006). Juglans regia (walnut) is an important crop that is cultivated throughout the world’s temperate regions for its edible nuts. Worldwide J, regia production was approximately 2.2 million tonnes in 2009. In 2012, J. regia production was about 203 212 tonnes in Turkey (TurkStat, 2014). As the shell of J. regia comprises 67% of the total weight of the fruit, around 136 150 tonnes of J. regia shell was left behind in 2012. Juglans regia shell, which is an annual agricultural waste, is the lignocellulosic material forming the thin endocarp or husk of the walnut fruit. Although, burning agricultural residues causes serious environmental problems this waste material is generally combusted directly in situ for heating purposes. Where there is growing interest in industrial utilization of agricultural residues and wastes, this could mean a second income for farmers. Juglans regia, as a bio-waste, has no economic value or industrial usage in Turkey. In general it is discarded or burned in the stove in the winter (Kar, 2011; Pirayesh et al., 2012). The lignocellulosic biomass is a good precursor for a co-pyrolysis process together with scrap tyre. In order to make better use of the cheap and abundant agricultural waste, it is proposed to use J. regia shell. The kinetic analysis of pyrolysis processes has been the subject of interest for many investigators throughout the modern history of thermal decomposition. The interest is fully justified. On one side, kinetic data are essential for designing any kind of device, in which the thermal decomposition takes place; on the other side, kinetics is intrinsically related with the decomposition mechanisms. The knowledge of the mechanism allows the postulation of kinetic equations and kinetics is the starting point for postulating mechanisms for thermal decomposition (EbrahimiKahrizsangi, 2008). Studies on the kinetics of individual scrap
tyre and J. regia shell pyrolysis have been carried out by means of techniques based on thermal analysis, such as thermogravimetry (TG), differential thermogravimetry (DTG) or differential scanning calorimetry (DSC) (Açıkalın, 2011; Aguado et al., 2005; Chen and Yeh, 1997; Islam et al., 2009; Kim et al., 1995; Lopez et al., 2009; Seidelt et al., 2006). Although there are many studies that can be found on co-pyrolysis of various biomass and plastic wastes (Aboulkas et al., 2008, 2009; Brebu et al., 2010; Encinar and Gonzales, 2008; Jean et al., 2011; Matsuzawa et al., 2001), there is insufficient study on the kinetics of co-pyrolysis of scrap tyres and J. regia shell. Therefore, interactions between scrap tyre and J. regia shell are not well known. The aim of this study was to compare the thermal and kinetic behaviours of individual raw materials with their blends. Special attention was directed to clarification of interactions of scrap tyre with J. regia shell in the pyrolysis process. Arrhenius and Coats– Redfern-based kinetic analysis methods have been applied to identify the pyrolysis reaction model, possibly contributing to the research field of pyrolysis kinetics of biomass–plastics blends.
Experimental Materials and sample preparation The J. regia shell used in this study was obtained from Bilecik, which is located in the north-west of Turkey. The raw material was dried at room temperature, ground using a rotary cutting mill and sieved. Samples with mesh sizes of 425–600 µm were used in all experiments. Powdered scrap tyre samples were obtained from Orbay Tire Recycling Company (Izmir, Turkey). Ultimate and component analyses were performed for J. regia shell and the properties of raw material are given in Table 1. The Fourier transform infrared (FT-IR) spectrum of J. regia shell which shows indications of various surface functional groups is shown in Figure 1. The wide peak at around 3356 cm−1 is attributed to hydroxyl groups or absorbed water. The bands located at around 2924 and 2856 cm−1 correspond to C–H stretching vibrations in methyl and methylene groups. Carbonyl (C=O) groups appear as a band at 1742 cm−1. The band at 1241 cm−1 and a relatively intense band at about 1036 cm−1 can be assigned to C–O stretching vibrations in alcohols, phenols, or ether or ester groups. The C–H out-of-plane bending vibrations in benzene derivative cause the bands at 894 and 841 cm−1. It could be determined that the J. regia shell include oxygen groups, carbonyl groups, ethers, esters, alcohols, and phenol groups. The powder-form scrap tyre was provided by Orbay Tyre Recycling Company (Izmir, Turkey). The weight fractions of carbon, hydrogen and nitrogen were determined by using an elemental analyser Leco CHN628 Series. Elemental analysis of scrap tyre and J. regia shell is presented in Table 2. Du-Long’s formula (Equation (1)) was used for the calculation of calorific values using elemental compositions. According to the results, calorific values of J. regia shell and scrap tyre were calculated as 16.7 and 28.45 MJ kg−1, respectively (Harker and Bakhurst, 1981).
3
Uzun and Yaman
Figure 1. FT-IR spectrum of J. regia shell. Table 2. Elemental analysis of J. regia shell and scrap tyre. Substance
C (%)
H (%)
N (%)
S (%)
O (%)a
J. regia shell Scrap tyre
47.50 70.02
6.39 6.76
0.46 1.09
– 4.21
47.65 17.92
aBy
difference.
The decomposition rate of a solid is generally given as:
Conversion, x, is a normalized form of weight loss data of the decomposed sample and is defined as follows:
O QGVC = 338.2C + 1442.8 H − + 94.2S 8
( kJ kg ) (1) −1
dx = k (T ) f ( x) (2) dt
x=
wi − wt (3) wi − wf
where C, H, O and S are the mass fractions of carbon, hydrogen, nitrogen, oxygen and sulfur.
where wi is the initial mass of the sample, wt is the actual mass and wf is the mass after pyrolysis. Parameter k(T) is defined as the rate constant of reaction whose temperature dependence is expressed by the Arrhenius equation.
Experimental techniques
Juglans regia shell, scrap tyre and their blends were subjected to thermogravimetric analysis (TGA) under a nitrogen atmosphere. A Setaram Labsysevo TGA system was used to measure and record the mass loss versus temperature of the sample over the course of the pyrolysis reaction. A 10 ± 3 mg sample in an Al2O3 100 µL crucible was used for each run, and the J. regia shell/scrap tyre blends (0 : 1; 1 : 2; 1 : 1; 2 : 1; 1 : 0 mass ratio) were mixed in an agate mortar in order to achieve homogeneity. The experiments were carried out under dynamic conditions. Nitrogen was used as an inert purge gas to displace air in the pyrolysis zone in order to avoid undesirable oxidation of the sample. After the initial purging, the nitrogen flow rate was 20 mL min−1 and the sample was heated from ambient temperature to 973 K at four heating rates: 5, 10, 15, and 20 K min−1.
where Ea is the activation energy (kJ mol−1), T is the absolute temperature (K), R is the universal gas constant (8.314 J K−1 mol−1) and A is the pre-exponential factor (min−1) (Slopiecka et al., 2011). The expression of the function f(x) and its derivative are used for describing the solid-state, first-order reaction; hence many authors restrict the mathematical function f(x) to the following expression (Sait et al., 2012):
− Ea
k (T ) = Ae RT (4)
f ( x ) = (1 − x ) (5) n
For constant heating rate, β = dT / dt ,
dx A − E / RT n = e (1 − x ) (6) dT β a
Arrhenius method Kinetic modelling The decomposition of a solid in a non-isothermal mode can be represented by the following reaction (Aboulkas et al., 2007):
aA ( solid ) → bB ( solid ) + cC ( gas )
In the Arrhenius method the final rate equation given below is obtained by linearization of Equation (6) and making some rearrangements:
A Ea (7) dx ln − nln (1 − x ) = ln β − RT T d
4
Waste Management & Research
Figure 2. TG-DTG curves of J. regia shell.
The plot of ln ( dx / dT ) − n ln (1 − x ) versus (1/T) should give a straight line for the correct value of reaction order n. The activation energy and pre-exponential factors are calculated from slope (−Ea/R) and intercept (ln(A/β)), respectively (Açıkalın, 2011).
Coats–Redfern method The Coats-Redfern method is an integral method and involves the thermal degradation mechanism. Using an asymptotic approximation for the resolution of Equation (6):
1 − (1 − x )1−n AR Ea = ln ln 2 − T (1 − n ) β Ea RT ln (1 − x ) AR Ea ln − = ln − 2 T β Ea RT
( n ≠ 1) (8) ( n = 1) (9)
The above-mentioned Equations (8) and (9) will result in a straight line with slope −Ea/R and an intercept of ln[AR/βEa]. This was done by plotting a graph between the following terms (Sait et al., 2012):
1 − (1 − x )1−n versus 1 / T ln 2 T (1 − n )
( n ≠ 1)
ln (1 − x ) ln − versus 1 / T T 2
( n = 1)
Results and discussion Thermal degradation of J. regia shell The devolatilization stage plays an important role in the conversion processes. Attention should be particularly paid to the light hydrocarbons which are produced at the end of thermal
degradation (Uzun and Sarıoğlu, 2009). The results of TG and DTG curves of the J. regia shell at different heating rates are given in Figure 2. According to the TG curves, the mass loss range can be divided into three steps associated with the variable slope curves. The first step starts at 323–343 K and finishes at 403–423 K with respect to the applied heating rate. This could be due to the vaporization of the moisture as a consequence of physically absorbed water within the sample. The step is specified as a small peak on the left-most side of the DTG curve. This is followed by the main devolatilization step on the TG curve which begins at 481–505 K and finishes at 645–688 K with regard to heating rate. This step indicates decomposition of hemicellulose and cellulose and is referred to as the active pyrolysis zone because the mass loss rate is higher. The passive pyrolysis zone starts above 688 K and the mass loss rate is lower at this step. Subsequently, there is no further significant mass loss. The stages of thermal behaviour are associated with the components of J. regia shell, which mainly consists of hemicellulose, cellulose and lignin as with all other lignocellulosic biomasses. The thermogravimetric behaviour of these components has been studied before and it is well known that hemicellulose, cellulose and lignin accomplish their decomposition within the temperature ranges of 483–598, 583–673, and 433–1173 K, respectively (Uzun and Sarıoğlu, 2009). Based on these temperature intervals, the minor and major reactions observed in the active pyrolysis zone can be attributed to decomposition of hemicellulose and cellulose (Yaman, 2013). In Table 1, the sum of hemicellulose and cellulose is calculated as 46.13% and the average mass loss of J. regia shell at the active pyrolysis zone is about 45%. This case is further evidence that hemicellulose and cellulose decomposition occurs in the active pyrolysis zone. The characteristic temperatures of the active pyrolysis zone of J. regia shell, namely starting temperature (Ti), ending temperature (Tf) and the temperature of maximum mass loss rate that occurred are given in Table 3. All characteristic temperatures increase with increasing heating rate. An increase in the heating
5
Uzun and Yaman Table 3. Properties of active pyrolysis zone at different heating rate. Heating rate Temperature (K min−1) (K) J. regia shell
Scrap tyre
J. regia shell : scrap tyre 1 : 2 mass ratio J. regia shell : scrap tyre 1 : 1 mass ratio J. regia shell : scrap tyre 2 : 1 mass ratio
5 10 15 20 5 10 15 20 5 10 15 20 5 10 15 20 5 10 15 20
Residue (wt.%)
Ti
Tmax
Tf
481 488 503 505 493 501 508 513 443 453 458 462 463 473 483 485 474 473 468 462
621 632 633 638 655 666 674 682 625 614 640 668 622 638 642 645 610 599 624 629
638 643 683 688 763 768 783 833 751 765 776 789 768 773 813 873 751 767 775 780
33.41 27.30 31.87 33.20 47.35 46.82 44.98 48.41 49.11 46.07 47.29 55.02 42.05 38.87 39.67 37.65 47.34 52.58 46.78 58.53
rate tends to delay the thermal degradation process towards higher temperatures, most probably due to the presence of increased thermal lag. Thermal lag is the time required to add or remove heat from the biomass before it reaches the desired temperature. At a given temperature, a higher heating rate implies that the material reaches that temperature in a shorter time. It can be clearly seen in Figure 2 that maximum mass loss rates are also shifted to higher temperatures as the heating rate is increased. This can be explained by the effect of inertia of devolatilization process, as the characteristic time of the process is decreased (Aboulkas et al., 2008; Damartzis et al., 2011; Lapuerta et al., 2004; Uzun and Sarıoğlu, 2009).
Thermal degradation of scrap tyre Tyres are made up of more than 100 different substances: rubber, fillers such as carbon black or silica gel, steel, sulfur, zinc oxide and many other additives such as plasticizers. The most common rubbers used for tyres are natural rubber (NR), styrene-butadiene rubber (SBR) and butadiene rubber (BR) or their blends. These components decompose at different temperature ranges and many researchers have emphasized that the thermal degradation behaviour of scrap tyres provide information about the type of rubber and its contents (Islam et al., 2009; Leung and Wang 1998; Seidelt et al., 2006). Previous researchers have reported that oil, plasticizer and other organic additives lose their weight within the temperature range of 423–623 K (Leung and Wang 1998). Devolatilization of tyre rubber and its components (SBR, NR and BR) at heating rates of 5 and 80 K min−1 was studied by
Williams and Besler (1995). During thermal decomposition, NR provides a sharp peak at 648 K, SBR provides a comparatively round peak at around 723 K, whereas BR generates two peaks at 673 and 748 K for a heating rate of 5 K min−1. When the scrap tyre sample consists of three major components, NR, SBR and BR components, a sharp peak at around 643 K, a round peak in the range of 673–733 K and another sharp peak at around 733 K must appear in the DTG curve. Figure 3 shows the thermograms of scrap tyre samples at various heating rates of 5, 10, 15 and 20 K min−1. As it can be seen, the first weight loss about 1% occurred at a range of 323–403 K owing to the vaporization of the moisture absorbed by the scrap tyre sample. This step is identified as a small peak on the DTG curve. From the TG data, it can be seen that thermal decomposition of scrap tyre started at about 493 K, following which a minor and a major loss of weight during the main devolatilization, and pyrolysis was essentially complete near 833 K. Above this temperature, there was no further remarkable mass loss. This decomposition profile shows that the sample consisted of SBR. Table 3 shows that there is a shift in the TG and DTG curves to higher temperatures as the heating rate increases. As previously mentioned, the shift to higher temperatures of thermal degradation profile is seen due to the presence of increased thermal lag. Owing to the addition of carbon black during tyre manufacture, the scrap tyre sample gives high char yields between 44.98 and48.41 wt%.
Thermal degradation of J. regia shellscrap tyre blends Juglans regia shell was blended with scrap tyre in various weight proportions of 0 : 1, 1 : 2, 1 : 1, 2 : 1, 1 : 0 to evaluate the effect of blending ratios on co-pyrolysis. The TG-DTG curves of these samples are given in Figure 4 and their parameters are listed in Table 3. The characteristic temperatures of the blends differ from the characteristic temperatures of their individual components. A shift of Tmax to higher values can be observed as the ratio of scrap tyre in the blends increased. This implies that the addition of scrap tyre to biomass increases the thermal stability of J. regia shell. To investigate whether interactions existed between the J regia shell and scrap tyre, difference of weight loss ∆w = wblend − (( x1w1 + x2 w2 )) is defined, where wblend is the weight loss of blend, xi is the mass fraction of each material and wi is the weight loss of each material in the same operational conditions. In Figure 5, Δw describes the extent of the synergistic effect which is calculated for different heating rates of 5, 10, 15 and 20 K min−1. For all blends, the co-components showed a different behaviour from that of the pure materials. It was found that for the J. regia shell and scrap tyre blends Δw ±2% at temperatures lower than 550 K, indicating poor interactions between J. regia shell and scrap tyre. Above 550 K, the value of Δw declined first and then increased up to 630–650 K. A significant interaction was observed after about 630 K, which is the starting temperature of
6
Waste Management & Research
Figure 3. TG-DTG curves of scrap tyre.
Figure 4. TG/DTG curves of J. regia shell/scrap tyre blends with a heating rate of 10 K min−1.
Figure 5. Variation of Δw for J. regia shell-scrap tyre blends (1 : 1 mass ratio) at different heating rates.
7
Uzun and Yaman Table 4. Kinetic parameters for pyrolysis of J. regia shell, scrap tyre and their blends from Arrhenius method. Sample
Heating rate
Kinetic parameters
(K min−1)
Ea (kJ mol−1)
log A (min−1)
R2
J. regia shell
5 10 15 20
69.11 79.73 60.99 67.03
5.27 6.41 4.69 5.34
0.976 0.955 0.984 0.924
Scrap tyre
5 10 15 20
68.59 73.54 76.56 67.21
4.42 5.04 5.37 4.67
0.979 0.980 0.981 0.966
J. regia shell : scrap tyre 1 : 2 mass ratio
5 10 15 20
60.90 64.90 64.02 62.04
7.40 3.58 2.06 1.73
0.981 0.876 0.973 0.954
J. regia shell : scrap tyre 1 : 1 mass ratio
5 10 15 20
50.40 53.98 44.39 39.33
3.13 3.64 5.34 1.44
0.968 0.982 0.973 0.888
Sample
J. regia shell : scrap tyre 2 : 1 mass ratio
5 10 15 20
29.30 45.43 71.92 60.94
3.53 2.31 2.20 1.64
0.744 0.829 0.963 0.955
SBR decomposition. These experimental results indicate that the maximum synergistic effect during co-pyrolysis at 20 K min−1 was about 5% at 669 K. The positivity of the values and their extent imply the interactions between J. regia shell and scrap tyre co-pyrolysis. Thus, at a higher rate a significant synergistic effect could be achieved. Above 750 K, the devolatilization process of the blend was complete and hence Δw became stable at this stage.
Determination of kinetic parameters Thermal decomposition of carbonaceous materials involves a large number of parallel and series reactions, thus the analysis of pyrolysis kinetics presents difficulties. Although thermogravimetry provides general information about overall reaction kinetics, rather than individual reactions, it can be used as a useful tool for providing a comparison of the kinetic data of various reaction parameters such as temperature and heating rate (Aboulkas et al., 2008). The Arrhenius and Coats–Redfern methods were adopted for the study of thermal decomposition of the J. regia shell, scrap tyre and their blends. In Arrhenius method, the order of reaction of J. regia shell, scrap tyre and their blends pyrolysis is assumed to be 1. The kinetic parameters such as activation energy (Ea) and pre-exponential factor (A) are given in Table 4. The average value of activation energies of J. regia shell, scrap tyre and their 1 : 1 blends were found to be 69.22, 71.48 and 47.03 kJ mol−1, respectively. In the Coats–Redfern method, the reaction order of
Figure 6. R2 –n curves for J. regia shell pyrolysis obtained by the Coats–Redfern method. Table 5. Kinetic parameters for pyrolysis of J. regia shell, scrap tyre and their blends from the Coats–Redfern method. Kinetic parameters
Heating Rate (K min−1)
Ea (kJ mol−1)
log A (min−1)
R2
J. regia shell
5 10 15 20
102.31 109.83 97.51 89.78
7.86 9.29 8.03 7.91
0.962 0.955 0.963 0.957
Scrap tyre
5 10 15 20
69.71 83.44 86.11 75.63
4.45 5.75 6.06 5.68
0.989 0.978 0.982 0.977
J. regia shell : scrap tyre 1 : 2 mass ratio
5 10 15 20
70.96 78.92 79.54 88.08
4.73 5.68 5.75 6.78
0.957 0.900 0.904 0.906
J. regia shell : scrap tyre 1 : 1 mass ratio
5 10 15 20
68.42 71.82 61.29 53.72
4.51 5.00 4.18 3.44
0.928 0.945 0.940 0.924
J. regia shell : scrap tyre 2 : 1 mass ratio
5 10 15 20
33.87 67.26 71.92 73.12
1.44 4.74 5.20 5.32
0.929 0.858 0.936 0.939
J. regia shell pyrolysis was found to be 1.5, which gave the maximum correlation factor (Figure 6). The reaction order of scrap tyre and J. regia shell-scrap tyre blend pyrolysis was assumed to be 1, as there was no significant change in correlation factor. The kinetic parameters based on the Coats–Redfern method are given in Table 5 with the activation energy values of J. regia shell, scrap tyre and their 1 : 1 blends being 99.86, 78.72 and 63.81 kJ mol−1, respectively. Comparing results of blends with those of the thermal decomposition of the single materials, it is noteworthy that the
8
Waste Management & Research
Figure 7. Comparison of experimental and model values of x plotted against T from the Arrhenius method (10 K min−1): (a) J. regia shell; (b) scrap tyre; (c) J. regia shell : scrap tyre 1 : 2 mass ratio; (d) J. regia shell : scrap tyre 1 : 1 mass ratio; (e) J. regia shell : scrap tyre 2 : 1 mass ratio.
activation energy of scrap tyre in the blend was lower than the activation energy of scrap tyre alone. The decreases in the activation energies are caused by the probable influence of J. regia shell degradation products on the scrap tyre degradation process. This behaviour was also considered for blends of biomass derivatives with other common polymers, with changes in the mechanism of thermal degradation (Aboulkas et al., 2009). The results for xmodel, calculated using the predicted kinetic parameters by the Arrhenius and Coats–Redfern method have been compared with the experimental results and are shown in Figures 7 and 8 for the heating rate of 10 K min−1. It can be seen that all the calculated curves almost overlap with the experimental ones. In general, experimental values of x were lower than the model values. For J. regia shell pyrolysis small differences were observed at low temperatures and the differences increased as temperature increased.
Conclusion TG and DTG curves provide valuable information on pyrolysis mechanism and kinetics of heterogeneous mixtures like scrap tyre. This is crucial for industrial scale application. For this purpose non-isothermal thermogravimetric analysis of J. regia shell, scrap tyre and their blends (weight proportions of 1 : 2; 1 : 1; 2 : 1) have been carried out in the temperature range of 323– 900 K at the heating rate of 5, 10, 15 and 20 K min−1 under a nitrogen flow of 20 mL min−1. Among pyrolysis of J. regia shell, three mass loss zones were specified as removal of water, decomposition of hemicelluloses and cellulose and decomposition of lignin. Most of the mass loss (about 45%) occurred in the second zone, which is referred to as the active pyrolysis zone, In the course of thermal degradation of scrap tyre, mass loss was detected due to decomposition of plasticizers and styrene-butadiene rubber. It was found that the
9
Uzun and Yaman
Figure 8. Comparison of experimental and model values of x plotted against T from Coats–Redfern method (10 K min−1): (a) J. regia shell; (b) scrap tyre; (c) J. regia shell : scrap tyre 1 : 2 mass ratio: (d) J. regia shell : scrap tyre 1 : 1 mass ratio; (e) J. regia shell : scrap tyre 2 : 1 mass ratio.
heating rate had a significant role in the pyrolysis process. The weight loss region was shifted to a higher temperature range for all the materials. The overall rate equations for various heating rates were determined by the Arrhenius and Coats–Redfern equations, by calculating the kinetic parameters such as activation energy and pre-exponential factors. The results are useful for the design of pyrolysis units. As a conclusion, the results obtained in this study showed that co-pyrolysis of J. regia shell and scrap tyre can be an environment-friendly process for the transformation of waste materials into valuable products such as chemicals and fuels.
Declaration of conflicting interests The authors declare that there is no conflict of interest.
Funding The authors are grateful to ‘Anadolu University Scientific Research Projects Council’ for the financial support of this work through the project 1205F104.
References Aboulkas A, El Harfi K, El Bouadili A, et al. (2007) Pyrolysis kinetics of polypropylene. Journal of Thermal Analysis and Calorimetry 89: 203–209. Aboulkas A, El Harfi K, Nadifiyine M, et al. (2008) Thermogravimetric characteristics and kinetic of co-pyrolysis of olive residue with high density polyethylene. Journal of Thermal Analysis and Calorimetry 91: 737–743. Aboulkas A, El Harfi K, El Bouadili A, et al. (2009) Pyrolysis kinetics of olive residue/plastic mixtures by non-isothermal thermogravimetry. Fuel Processing Technology 90: 722–728. Açıkalın K (2011) Thermogravimetric analysis of walnut shell as pyrolysis feedstock. Journal of Thermal Analysis and Calorimetry 105: 145–150.
10 Aguado R, Olazar M, Velez D, et al. (2005) Kinetics of scrap tyre pyrolysis under fast heating conditions. Journal of Analytical and Applied Pyrolysis 73: 290–298. Brebu M, Ucar S, Vasile C, et al. (2010) Co-pyrolysis of pine cone with synthetic polymers. Fuel 89: 1911–1918. Chen KS and Yeh RZ (1997) Kinetics of thermal decomposition of styrenebutadiene rubber at low heating rates in nitrogen and oxygen. Combustion and Flame 108: 408–418. Clark C and Meardon Kand Russell D (1991) Burning Tires for Fuel and Tire Pyrolysis. Research Triangle Park, NC, USA: Environmental Protection Agency. Damartzis Th, Vamvuka D, Sfakiotakis S, et al. (2011) Thermal degradation studies and kinetic modelling of cardoon (Cynara cardunculus) pyrolysis using thermogravimetric analysis (TGA). Bioresources Technology 102: 6230–6238. Ebrahimi-Kahrizsangi R (2008) Evaluation of reliability of Coats–Redfern method for kinetic analysis of non-isothermal TGA. Transactions of Nonferrous Metals Society of China 18: 217–221. Encinar JM and Gonzales JF (2008) Pyrolysis of synthetic polymers and plastic wastes: Kinetic study. Fuel Processing Technology 89: 678–686. Environmental Protection Agency (EPA) (1991) Burning Tires for Fuel and Tire Pyrolysis: Air Implications. EPA-450/3-91-024. NC, USA: Environmental Protection Agency. Harker JH and Backhurst JR (1981) Fuel and Energy. London, UK: Academic Press Limited. Islam MR, Haniu H and Fardoush J (2009) Pyrolysis kinetics behavior of solid tire wastes available in Bangladesh. Waste Management 29: 668–677. Jeon MJ, Choi SJ, Yoo KS, Ryu C, Park, SH, Lee JM, Jeon JK, Park YK, Kim S (2011) Copyrolysis of block poypropylene with waste wood chip. Korean Journal of Chemical Engineering 28(2): 497–501. Kar Y (2011) Co-pyrolysis of walnut shell and tar sand in a fixed-bed reactor. Bioresource Technology 102: 9800–9805. Kim SD, Park JK and Chun HD (1995) Pyrolysis kinetics of scrap tire rubbers I: Using DTG and TGA. Journal of Environmental Engineering 121: 507–514.
Waste Management & Research Lapuerta M, Hernandez JJ and Rodriguez J (2004) Kinetics of devolatilization of forescrap tyre wastes from thermogravimetric analysis. Biomass and Bioenergy 27: 385–391. Leung DYC and Wang CL (1998) Kinetic study of scrap tyre pyrolysis and combustion. Journal of Analytical and Applied Pyrolysis 45: 153–169. Lopez G, Aguado R, Olazar M, et al. (2009) Kinetics of scrap tyre pyrolysis under vacuum conditions. Waste Management 29: 2649–2655. Matsuzawa Y, Ayabe M and Nishino J (2001) Acceleration of cellulose co-pyrolysis with polymer. Polymer Degradation and Stability 71: 453–444. Pirayesh H, Khazaeian A and Tabarsa T (2012) The potential for using walnut (Juglans regia L.) shell as a raw material for wood-based particleboard manufacturing. Composites: Part B 43: 3276–3280 Rodriguez IM, Laresgoiti MF, Cabrero MA, et al. (2001) Pyrolysis of scrap tyres. Fuel Processing Technology 72: 9–22. Sait HH, Hussain A, Salema AA, et al. (2012) Pyrolysis and combustion kinetics of date palm biomass using thermogravimetric analysis. Bioresource Technology 118: 382–389. Seidelt S, Müller-Hagedorn M and Bockhorn H (2006) Description of tire pyrolysis by thermal degradation behaviour of main components. Journal of Analytical and Applied Pyrolysis 75: 11–18. Slopiecka K, Bartocci P and Fantozzi F (2011) Thermogravimetric analysis and kinetic study of poplar wood pyrolysis. Applied Energy 97: 491–497. TurkStat (Turkish Statistical Institute) 2014 Available at http://www.tuik. gov.tr/PreTablo.do?alt_id=1001) (accessed 20 March 2014). Uzun BB and Sarioğlu N (2009) Rapid and catalytic pyrolysis of corn stalks. Fuel Processing Technology 90: 705–716. Williams PT and Besler S (1995) Pyrolysis-thermogravimetric analysis of tyres and tyre components. Fuel 74: 1277–1283. Yaman E (2013) Investıgatıon of thermal degradation kinetics of walnut shell with waste plastics via thermogravimetric analysis. MSc. Thesis, Anadolu University, Institute of Applied Science, pp. 37–39. Zhou L, Wang Y, Huang Q, et al. (2006) Thermogravimetric characteristics and kinetics of plastic and biomass blends co-pyrolysis. Fuel Processing Technology 87: 963–969.
research-article2014
WMR0010.1177/0734242X14539722Waste Management & ResearchUzun and Yaman
Original Article
Thermogravimetric characteristics and kinetics of scrap tyre and Juglans regia shell co-pyrolysis
Waste Management & Research 1–10 © The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0734242X14539722 wmr.sagepub.com
BB Uzun1 and E Yaman2
Abstract The degradation kinetics of Juglans regia shell, scrap tyre and their blends were investigated using a thermogravimetric analysis method. Experiments were performed under dynamic conditions and a nitrogen atmosphere in the range 293 to 973 K at different heating rates. During pyrolysis of J. regia shell three mass loss zones were specified as removal of water, decomposition of hemicelluloses and cellulose, and decomposition of lignin. The degradation curves of scrap tyre showed merely one stage which was due to decomposition of styrene butadiene rubber. The kinetic parameters were calculated using both Arrhenius and Coats–Redfern methods. By adopting the Arrhenius method, the average value of activation energies of J. regia shell, scrap tyre and their 1 : 1 blends were found to be 69.22, 71.48 and 47.03 kJ mol−1, respectively. Additionally, by using the Coats–Redfern method, the average value of activation energies of J. regia shell, scrap tyre and their 1 : 1 blend were determined as 99.85, 78.72 and 63.81 kJ mol−1, respectively. The addition of J. regia shell to scrap tyre caused a reduction in the activation energies. The difference of weight loss was measured to examine interactions between raw materials. The maximum difference between experimental and theoretical mass loss was 5% at about 648 K with a heating rate of 20 K min−1. These results indicated a significant synergistic effect was available during co-pyrolysis of J. regia shell and scrap tyre in the high temperature region. Keywords Juglans regia shell, scrap tyre, co-pyrolysis, kinetic study, thermogravimetric analysis, synergistic effect
Introduction The utilization of human-generated waste such as plastic and tyre is one of the greatest challenges of the twenty-first century. Therefore, their recovery or disposal methods are crucial for sustainable development. The valorization of these wastes contributes to decrease the acceleration of climate change and fossil fuel consumption (Lopez et al., 2009). As known, the number of cars is increasing year by year in developing countries. Due to the increasing number of used tyres, the valorization of this waste material could solve an important environmental problem (Aguado et al., 2005). Approximately 70–85% of scrap tyres are landfilled or stockpiled. Landfilling has been favoured, because of its simplicity. However, a large space is required and the reusable resources are wasted. For example, approximately 2 billion scrap tyres have accumulated in nationwide stockpiles or uncontrolled dumps (Clark et al., 1991; EPA, 1991). The piled tyres provide breeding sites for mosquitoes, which can cause serious diseases (Kim et al., 1995). Landfilling is a potential danger due to the possibility of accidental fires that produce high emissions of hazardous gases. In addition, it requires a considerable amount of space because the volume of the tyres cannot be reduced by compaction (Islam et al., 2009). There are different alternative methods for tyre recycling such as retreading, reclaiming, incineration, grinding, etc. but they also have significant limitations.
A thermal degradation method, pyrolysis, appears to be very appropriate for complex materials, such as tyres that cannot be remoulded. In the pyrolysis process, the organic volatile matter of tyres is decomposed to low molecular weight products, which can be useful as fuels or chemicals. The inorganic components mainly from steel and the nonvolatile carbon black remain as ‘solid residue’, relatively unchanged (Rodrigues et al., 2001). Synthetic rubber materials such as scrap tyres have a high carbon content of about 70 wt.% and a low oxygen content of about 17 wt.% to produce hydrocarbon oils by pyrolysis. Additionally, as the biomass has a low carbon content of 47–51 wt.% and a high oxygen content of 42–46 wt.%, it produces highly oxygenated pyrolysis oils. Co-processing of scrap tyre with biomass could balance the carbon and oxygen amount in the feedstock,
1Anadolu
University, Faculty of Engineering, Department of Chemical Engineering, Eskisehir, Turkey 2Bilecik Seyh Edebali University, Central Research Laboratory, Bilecik, Turkey Corresponding author: BB Uzun, Anadolu University, Faculty of Engineering, Department of Chemical Engineering, Eskisehir, 26470, Turkey. Email: [email protected]
2
Waste Management & Research
Table 1. Properties of J. regia shell. wt.% Proximate analyses Moisture Ash Volatile matter Fixed carbona Component analyses Holocellulose Oil Lignin Extractives Hemicellulose Cellulosea aBy
8.06 0.33 76.38 15.23 46.13 3.29 48.11 3.78 22.18 23.95
difference.
with a strong effect on the properties of degradation products. On the other hand, biomass has lower thermal stability in comparison with scrap tyres and thus it could affect their radical degradation mechanism by promoting the degradation of synthetic macromolecules. In co-pyrolysis, the yields and composition of the products strongly depend on the treatment method, processing conditions, type of biomass and synthetic polymers (Brebu, 2010; Zhou et al., 2006). Juglans regia (walnut) is an important crop that is cultivated throughout the world’s temperate regions for its edible nuts. Worldwide J, regia production was approximately 2.2 million tonnes in 2009. In 2012, J. regia production was about 203 212 tonnes in Turkey (TurkStat, 2014). As the shell of J. regia comprises 67% of the total weight of the fruit, around 136 150 tonnes of J. regia shell was left behind in 2012. Juglans regia shell, which is an annual agricultural waste, is the lignocellulosic material forming the thin endocarp or husk of the walnut fruit. Although, burning agricultural residues causes serious environmental problems this waste material is generally combusted directly in situ for heating purposes. Where there is growing interest in industrial utilization of agricultural residues and wastes, this could mean a second income for farmers. Juglans regia, as a bio-waste, has no economic value or industrial usage in Turkey. In general it is discarded or burned in the stove in the winter (Kar, 2011; Pirayesh et al., 2012). The lignocellulosic biomass is a good precursor for a co-pyrolysis process together with scrap tyre. In order to make better use of the cheap and abundant agricultural waste, it is proposed to use J. regia shell. The kinetic analysis of pyrolysis processes has been the subject of interest for many investigators throughout the modern history of thermal decomposition. The interest is fully justified. On one side, kinetic data are essential for designing any kind of device, in which the thermal decomposition takes place; on the other side, kinetics is intrinsically related with the decomposition mechanisms. The knowledge of the mechanism allows the postulation of kinetic equations and kinetics is the starting point for postulating mechanisms for thermal decomposition (EbrahimiKahrizsangi, 2008). Studies on the kinetics of individual scrap
tyre and J. regia shell pyrolysis have been carried out by means of techniques based on thermal analysis, such as thermogravimetry (TG), differential thermogravimetry (DTG) or differential scanning calorimetry (DSC) (Açıkalın, 2011; Aguado et al., 2005; Chen and Yeh, 1997; Islam et al., 2009; Kim et al., 1995; Lopez et al., 2009; Seidelt et al., 2006). Although there are many studies that can be found on co-pyrolysis of various biomass and plastic wastes (Aboulkas et al., 2008, 2009; Brebu et al., 2010; Encinar and Gonzales, 2008; Jean et al., 2011; Matsuzawa et al., 2001), there is insufficient study on the kinetics of co-pyrolysis of scrap tyres and J. regia shell. Therefore, interactions between scrap tyre and J. regia shell are not well known. The aim of this study was to compare the thermal and kinetic behaviours of individual raw materials with their blends. Special attention was directed to clarification of interactions of scrap tyre with J. regia shell in the pyrolysis process. Arrhenius and Coats– Redfern-based kinetic analysis methods have been applied to identify the pyrolysis reaction model, possibly contributing to the research field of pyrolysis kinetics of biomass–plastics blends.
Experimental Materials and sample preparation The J. regia shell used in this study was obtained from Bilecik, which is located in the north-west of Turkey. The raw material was dried at room temperature, ground using a rotary cutting mill and sieved. Samples with mesh sizes of 425–600 µm were used in all experiments. Powdered scrap tyre samples were obtained from Orbay Tire Recycling Company (Izmir, Turkey). Ultimate and component analyses were performed for J. regia shell and the properties of raw material are given in Table 1. The Fourier transform infrared (FT-IR) spectrum of J. regia shell which shows indications of various surface functional groups is shown in Figure 1. The wide peak at around 3356 cm−1 is attributed to hydroxyl groups or absorbed water. The bands located at around 2924 and 2856 cm−1 correspond to C–H stretching vibrations in methyl and methylene groups. Carbonyl (C=O) groups appear as a band at 1742 cm−1. The band at 1241 cm−1 and a relatively intense band at about 1036 cm−1 can be assigned to C–O stretching vibrations in alcohols, phenols, or ether or ester groups. The C–H out-of-plane bending vibrations in benzene derivative cause the bands at 894 and 841 cm−1. It could be determined that the J. regia shell include oxygen groups, carbonyl groups, ethers, esters, alcohols, and phenol groups. The powder-form scrap tyre was provided by Orbay Tyre Recycling Company (Izmir, Turkey). The weight fractions of carbon, hydrogen and nitrogen were determined by using an elemental analyser Leco CHN628 Series. Elemental analysis of scrap tyre and J. regia shell is presented in Table 2. Du-Long’s formula (Equation (1)) was used for the calculation of calorific values using elemental compositions. According to the results, calorific values of J. regia shell and scrap tyre were calculated as 16.7 and 28.45 MJ kg−1, respectively (Harker and Bakhurst, 1981).
3
Uzun and Yaman
Figure 1. FT-IR spectrum of J. regia shell. Table 2. Elemental analysis of J. regia shell and scrap tyre. Substance
C (%)
H (%)
N (%)
S (%)
O (%)a
J. regia shell Scrap tyre
47.50 70.02
6.39 6.76
0.46 1.09
– 4.21
47.65 17.92
aBy
difference.
The decomposition rate of a solid is generally given as:
Conversion, x, is a normalized form of weight loss data of the decomposed sample and is defined as follows:
O QGVC = 338.2C + 1442.8 H − + 94.2S 8
( kJ kg ) (1) −1
dx = k (T ) f ( x) (2) dt
x=
wi − wt (3) wi − wf
where C, H, O and S are the mass fractions of carbon, hydrogen, nitrogen, oxygen and sulfur.
where wi is the initial mass of the sample, wt is the actual mass and wf is the mass after pyrolysis. Parameter k(T) is defined as the rate constant of reaction whose temperature dependence is expressed by the Arrhenius equation.
Experimental techniques
Juglans regia shell, scrap tyre and their blends were subjected to thermogravimetric analysis (TGA) under a nitrogen atmosphere. A Setaram Labsysevo TGA system was used to measure and record the mass loss versus temperature of the sample over the course of the pyrolysis reaction. A 10 ± 3 mg sample in an Al2O3 100 µL crucible was used for each run, and the J. regia shell/scrap tyre blends (0 : 1; 1 : 2; 1 : 1; 2 : 1; 1 : 0 mass ratio) were mixed in an agate mortar in order to achieve homogeneity. The experiments were carried out under dynamic conditions. Nitrogen was used as an inert purge gas to displace air in the pyrolysis zone in order to avoid undesirable oxidation of the sample. After the initial purging, the nitrogen flow rate was 20 mL min−1 and the sample was heated from ambient temperature to 973 K at four heating rates: 5, 10, 15, and 20 K min−1.
where Ea is the activation energy (kJ mol−1), T is the absolute temperature (K), R is the universal gas constant (8.314 J K−1 mol−1) and A is the pre-exponential factor (min−1) (Slopiecka et al., 2011). The expression of the function f(x) and its derivative are used for describing the solid-state, first-order reaction; hence many authors restrict the mathematical function f(x) to the following expression (Sait et al., 2012):
− Ea
k (T ) = Ae RT (4)
f ( x ) = (1 − x ) (5) n
For constant heating rate, β = dT / dt ,
dx A − E / RT n = e (1 − x ) (6) dT β a
Arrhenius method Kinetic modelling The decomposition of a solid in a non-isothermal mode can be represented by the following reaction (Aboulkas et al., 2007):
aA ( solid ) → bB ( solid ) + cC ( gas )
In the Arrhenius method the final rate equation given below is obtained by linearization of Equation (6) and making some rearrangements:
A Ea (7) dx ln − nln (1 − x ) = ln β − RT T d
4
Waste Management & Research
Figure 2. TG-DTG curves of J. regia shell.
The plot of ln ( dx / dT ) − n ln (1 − x ) versus (1/T) should give a straight line for the correct value of reaction order n. The activation energy and pre-exponential factors are calculated from slope (−Ea/R) and intercept (ln(A/β)), respectively (Açıkalın, 2011).
Coats–Redfern method The Coats-Redfern method is an integral method and involves the thermal degradation mechanism. Using an asymptotic approximation for the resolution of Equation (6):
1 − (1 − x )1−n AR Ea = ln ln 2 − T (1 − n ) β Ea RT ln (1 − x ) AR Ea ln − = ln − 2 T β Ea RT
( n ≠ 1) (8) ( n = 1) (9)
The above-mentioned Equations (8) and (9) will result in a straight line with slope −Ea/R and an intercept of ln[AR/βEa]. This was done by plotting a graph between the following terms (Sait et al., 2012):
1 − (1 − x )1−n versus 1 / T ln 2 T (1 − n )
( n ≠ 1)
ln (1 − x ) ln − versus 1 / T T 2
( n = 1)
Results and discussion Thermal degradation of J. regia shell The devolatilization stage plays an important role in the conversion processes. Attention should be particularly paid to the light hydrocarbons which are produced at the end of thermal
degradation (Uzun and Sarıoğlu, 2009). The results of TG and DTG curves of the J. regia shell at different heating rates are given in Figure 2. According to the TG curves, the mass loss range can be divided into three steps associated with the variable slope curves. The first step starts at 323–343 K and finishes at 403–423 K with respect to the applied heating rate. This could be due to the vaporization of the moisture as a consequence of physically absorbed water within the sample. The step is specified as a small peak on the left-most side of the DTG curve. This is followed by the main devolatilization step on the TG curve which begins at 481–505 K and finishes at 645–688 K with regard to heating rate. This step indicates decomposition of hemicellulose and cellulose and is referred to as the active pyrolysis zone because the mass loss rate is higher. The passive pyrolysis zone starts above 688 K and the mass loss rate is lower at this step. Subsequently, there is no further significant mass loss. The stages of thermal behaviour are associated with the components of J. regia shell, which mainly consists of hemicellulose, cellulose and lignin as with all other lignocellulosic biomasses. The thermogravimetric behaviour of these components has been studied before and it is well known that hemicellulose, cellulose and lignin accomplish their decomposition within the temperature ranges of 483–598, 583–673, and 433–1173 K, respectively (Uzun and Sarıoğlu, 2009). Based on these temperature intervals, the minor and major reactions observed in the active pyrolysis zone can be attributed to decomposition of hemicellulose and cellulose (Yaman, 2013). In Table 1, the sum of hemicellulose and cellulose is calculated as 46.13% and the average mass loss of J. regia shell at the active pyrolysis zone is about 45%. This case is further evidence that hemicellulose and cellulose decomposition occurs in the active pyrolysis zone. The characteristic temperatures of the active pyrolysis zone of J. regia shell, namely starting temperature (Ti), ending temperature (Tf) and the temperature of maximum mass loss rate that occurred are given in Table 3. All characteristic temperatures increase with increasing heating rate. An increase in the heating
5
Uzun and Yaman Table 3. Properties of active pyrolysis zone at different heating rate. Heating rate Temperature (K min−1) (K) J. regia shell
Scrap tyre
J. regia shell : scrap tyre 1 : 2 mass ratio J. regia shell : scrap tyre 1 : 1 mass ratio J. regia shell : scrap tyre 2 : 1 mass ratio
5 10 15 20 5 10 15 20 5 10 15 20 5 10 15 20 5 10 15 20
Residue (wt.%)
Ti
Tmax
Tf
481 488 503 505 493 501 508 513 443 453 458 462 463 473 483 485 474 473 468 462
621 632 633 638 655 666 674 682 625 614 640 668 622 638 642 645 610 599 624 629
638 643 683 688 763 768 783 833 751 765 776 789 768 773 813 873 751 767 775 780
33.41 27.30 31.87 33.20 47.35 46.82 44.98 48.41 49.11 46.07 47.29 55.02 42.05 38.87 39.67 37.65 47.34 52.58 46.78 58.53
rate tends to delay the thermal degradation process towards higher temperatures, most probably due to the presence of increased thermal lag. Thermal lag is the time required to add or remove heat from the biomass before it reaches the desired temperature. At a given temperature, a higher heating rate implies that the material reaches that temperature in a shorter time. It can be clearly seen in Figure 2 that maximum mass loss rates are also shifted to higher temperatures as the heating rate is increased. This can be explained by the effect of inertia of devolatilization process, as the characteristic time of the process is decreased (Aboulkas et al., 2008; Damartzis et al., 2011; Lapuerta et al., 2004; Uzun and Sarıoğlu, 2009).
Thermal degradation of scrap tyre Tyres are made up of more than 100 different substances: rubber, fillers such as carbon black or silica gel, steel, sulfur, zinc oxide and many other additives such as plasticizers. The most common rubbers used for tyres are natural rubber (NR), styrene-butadiene rubber (SBR) and butadiene rubber (BR) or their blends. These components decompose at different temperature ranges and many researchers have emphasized that the thermal degradation behaviour of scrap tyres provide information about the type of rubber and its contents (Islam et al., 2009; Leung and Wang 1998; Seidelt et al., 2006). Previous researchers have reported that oil, plasticizer and other organic additives lose their weight within the temperature range of 423–623 K (Leung and Wang 1998). Devolatilization of tyre rubber and its components (SBR, NR and BR) at heating rates of 5 and 80 K min−1 was studied by
Williams and Besler (1995). During thermal decomposition, NR provides a sharp peak at 648 K, SBR provides a comparatively round peak at around 723 K, whereas BR generates two peaks at 673 and 748 K for a heating rate of 5 K min−1. When the scrap tyre sample consists of three major components, NR, SBR and BR components, a sharp peak at around 643 K, a round peak in the range of 673–733 K and another sharp peak at around 733 K must appear in the DTG curve. Figure 3 shows the thermograms of scrap tyre samples at various heating rates of 5, 10, 15 and 20 K min−1. As it can be seen, the first weight loss about 1% occurred at a range of 323–403 K owing to the vaporization of the moisture absorbed by the scrap tyre sample. This step is identified as a small peak on the DTG curve. From the TG data, it can be seen that thermal decomposition of scrap tyre started at about 493 K, following which a minor and a major loss of weight during the main devolatilization, and pyrolysis was essentially complete near 833 K. Above this temperature, there was no further remarkable mass loss. This decomposition profile shows that the sample consisted of SBR. Table 3 shows that there is a shift in the TG and DTG curves to higher temperatures as the heating rate increases. As previously mentioned, the shift to higher temperatures of thermal degradation profile is seen due to the presence of increased thermal lag. Owing to the addition of carbon black during tyre manufacture, the scrap tyre sample gives high char yields between 44.98 and48.41 wt%.
Thermal degradation of J. regia shellscrap tyre blends Juglans regia shell was blended with scrap tyre in various weight proportions of 0 : 1, 1 : 2, 1 : 1, 2 : 1, 1 : 0 to evaluate the effect of blending ratios on co-pyrolysis. The TG-DTG curves of these samples are given in Figure 4 and their parameters are listed in Table 3. The characteristic temperatures of the blends differ from the characteristic temperatures of their individual components. A shift of Tmax to higher values can be observed as the ratio of scrap tyre in the blends increased. This implies that the addition of scrap tyre to biomass increases the thermal stability of J. regia shell. To investigate whether interactions existed between the J regia shell and scrap tyre, difference of weight loss ∆w = wblend − (( x1w1 + x2 w2 )) is defined, where wblend is the weight loss of blend, xi is the mass fraction of each material and wi is the weight loss of each material in the same operational conditions. In Figure 5, Δw describes the extent of the synergistic effect which is calculated for different heating rates of 5, 10, 15 and 20 K min−1. For all blends, the co-components showed a different behaviour from that of the pure materials. It was found that for the J. regia shell and scrap tyre blends Δw ±2% at temperatures lower than 550 K, indicating poor interactions between J. regia shell and scrap tyre. Above 550 K, the value of Δw declined first and then increased up to 630–650 K. A significant interaction was observed after about 630 K, which is the starting temperature of
6
Waste Management & Research
Figure 3. TG-DTG curves of scrap tyre.
Figure 4. TG/DTG curves of J. regia shell/scrap tyre blends with a heating rate of 10 K min−1.
Figure 5. Variation of Δw for J. regia shell-scrap tyre blends (1 : 1 mass ratio) at different heating rates.
7
Uzun and Yaman Table 4. Kinetic parameters for pyrolysis of J. regia shell, scrap tyre and their blends from Arrhenius method. Sample
Heating rate
Kinetic parameters
(K min−1)
Ea (kJ mol−1)
log A (min−1)
R2
J. regia shell
5 10 15 20
69.11 79.73 60.99 67.03
5.27 6.41 4.69 5.34
0.976 0.955 0.984 0.924
Scrap tyre
5 10 15 20
68.59 73.54 76.56 67.21
4.42 5.04 5.37 4.67
0.979 0.980 0.981 0.966
J. regia shell : scrap tyre 1 : 2 mass ratio
5 10 15 20
60.90 64.90 64.02 62.04
7.40 3.58 2.06 1.73
0.981 0.876 0.973 0.954
J. regia shell : scrap tyre 1 : 1 mass ratio
5 10 15 20
50.40 53.98 44.39 39.33
3.13 3.64 5.34 1.44
0.968 0.982 0.973 0.888
Sample
J. regia shell : scrap tyre 2 : 1 mass ratio
5 10 15 20
29.30 45.43 71.92 60.94
3.53 2.31 2.20 1.64
0.744 0.829 0.963 0.955
SBR decomposition. These experimental results indicate that the maximum synergistic effect during co-pyrolysis at 20 K min−1 was about 5% at 669 K. The positivity of the values and their extent imply the interactions between J. regia shell and scrap tyre co-pyrolysis. Thus, at a higher rate a significant synergistic effect could be achieved. Above 750 K, the devolatilization process of the blend was complete and hence Δw became stable at this stage.
Determination of kinetic parameters Thermal decomposition of carbonaceous materials involves a large number of parallel and series reactions, thus the analysis of pyrolysis kinetics presents difficulties. Although thermogravimetry provides general information about overall reaction kinetics, rather than individual reactions, it can be used as a useful tool for providing a comparison of the kinetic data of various reaction parameters such as temperature and heating rate (Aboulkas et al., 2008). The Arrhenius and Coats–Redfern methods were adopted for the study of thermal decomposition of the J. regia shell, scrap tyre and their blends. In Arrhenius method, the order of reaction of J. regia shell, scrap tyre and their blends pyrolysis is assumed to be 1. The kinetic parameters such as activation energy (Ea) and pre-exponential factor (A) are given in Table 4. The average value of activation energies of J. regia shell, scrap tyre and their 1 : 1 blends were found to be 69.22, 71.48 and 47.03 kJ mol−1, respectively. In the Coats–Redfern method, the reaction order of
Figure 6. R2 –n curves for J. regia shell pyrolysis obtained by the Coats–Redfern method. Table 5. Kinetic parameters for pyrolysis of J. regia shell, scrap tyre and their blends from the Coats–Redfern method. Kinetic parameters
Heating Rate (K min−1)
Ea (kJ mol−1)
log A (min−1)
R2
J. regia shell
5 10 15 20
102.31 109.83 97.51 89.78
7.86 9.29 8.03 7.91
0.962 0.955 0.963 0.957
Scrap tyre
5 10 15 20
69.71 83.44 86.11 75.63
4.45 5.75 6.06 5.68
0.989 0.978 0.982 0.977
J. regia shell : scrap tyre 1 : 2 mass ratio
5 10 15 20
70.96 78.92 79.54 88.08
4.73 5.68 5.75 6.78
0.957 0.900 0.904 0.906
J. regia shell : scrap tyre 1 : 1 mass ratio
5 10 15 20
68.42 71.82 61.29 53.72
4.51 5.00 4.18 3.44
0.928 0.945 0.940 0.924
J. regia shell : scrap tyre 2 : 1 mass ratio
5 10 15 20
33.87 67.26 71.92 73.12
1.44 4.74 5.20 5.32
0.929 0.858 0.936 0.939
J. regia shell pyrolysis was found to be 1.5, which gave the maximum correlation factor (Figure 6). The reaction order of scrap tyre and J. regia shell-scrap tyre blend pyrolysis was assumed to be 1, as there was no significant change in correlation factor. The kinetic parameters based on the Coats–Redfern method are given in Table 5 with the activation energy values of J. regia shell, scrap tyre and their 1 : 1 blends being 99.86, 78.72 and 63.81 kJ mol−1, respectively. Comparing results of blends with those of the thermal decomposition of the single materials, it is noteworthy that the
8
Waste Management & Research
Figure 7. Comparison of experimental and model values of x plotted against T from the Arrhenius method (10 K min−1): (a) J. regia shell; (b) scrap tyre; (c) J. regia shell : scrap tyre 1 : 2 mass ratio; (d) J. regia shell : scrap tyre 1 : 1 mass ratio; (e) J. regia shell : scrap tyre 2 : 1 mass ratio.
activation energy of scrap tyre in the blend was lower than the activation energy of scrap tyre alone. The decreases in the activation energies are caused by the probable influence of J. regia shell degradation products on the scrap tyre degradation process. This behaviour was also considered for blends of biomass derivatives with other common polymers, with changes in the mechanism of thermal degradation (Aboulkas et al., 2009). The results for xmodel, calculated using the predicted kinetic parameters by the Arrhenius and Coats–Redfern method have been compared with the experimental results and are shown in Figures 7 and 8 for the heating rate of 10 K min−1. It can be seen that all the calculated curves almost overlap with the experimental ones. In general, experimental values of x were lower than the model values. For J. regia shell pyrolysis small differences were observed at low temperatures and the differences increased as temperature increased.
Conclusion TG and DTG curves provide valuable information on pyrolysis mechanism and kinetics of heterogeneous mixtures like scrap tyre. This is crucial for industrial scale application. For this purpose non-isothermal thermogravimetric analysis of J. regia shell, scrap tyre and their blends (weight proportions of 1 : 2; 1 : 1; 2 : 1) have been carried out in the temperature range of 323– 900 K at the heating rate of 5, 10, 15 and 20 K min−1 under a nitrogen flow of 20 mL min−1. Among pyrolysis of J. regia shell, three mass loss zones were specified as removal of water, decomposition of hemicelluloses and cellulose and decomposition of lignin. Most of the mass loss (about 45%) occurred in the second zone, which is referred to as the active pyrolysis zone, In the course of thermal degradation of scrap tyre, mass loss was detected due to decomposition of plasticizers and styrene-butadiene rubber. It was found that the
9
Uzun and Yaman
Figure 8. Comparison of experimental and model values of x plotted against T from Coats–Redfern method (10 K min−1): (a) J. regia shell; (b) scrap tyre; (c) J. regia shell : scrap tyre 1 : 2 mass ratio: (d) J. regia shell : scrap tyre 1 : 1 mass ratio; (e) J. regia shell : scrap tyre 2 : 1 mass ratio.
heating rate had a significant role in the pyrolysis process. The weight loss region was shifted to a higher temperature range for all the materials. The overall rate equations for various heating rates were determined by the Arrhenius and Coats–Redfern equations, by calculating the kinetic parameters such as activation energy and pre-exponential factors. The results are useful for the design of pyrolysis units. As a conclusion, the results obtained in this study showed that co-pyrolysis of J. regia shell and scrap tyre can be an environment-friendly process for the transformation of waste materials into valuable products such as chemicals and fuels.
Declaration of conflicting interests The authors declare that there is no conflict of interest.
Funding The authors are grateful to ‘Anadolu University Scientific Research Projects Council’ for the financial support of this work through the project 1205F104.
References Aboulkas A, El Harfi K, El Bouadili A, et al. (2007) Pyrolysis kinetics of polypropylene. Journal of Thermal Analysis and Calorimetry 89: 203–209. Aboulkas A, El Harfi K, Nadifiyine M, et al. (2008) Thermogravimetric characteristics and kinetic of co-pyrolysis of olive residue with high density polyethylene. Journal of Thermal Analysis and Calorimetry 91: 737–743. Aboulkas A, El Harfi K, El Bouadili A, et al. (2009) Pyrolysis kinetics of olive residue/plastic mixtures by non-isothermal thermogravimetry. Fuel Processing Technology 90: 722–728. Açıkalın K (2011) Thermogravimetric analysis of walnut shell as pyrolysis feedstock. Journal of Thermal Analysis and Calorimetry 105: 145–150.
10 Aguado R, Olazar M, Velez D, et al. (2005) Kinetics of scrap tyre pyrolysis under fast heating conditions. Journal of Analytical and Applied Pyrolysis 73: 290–298. Brebu M, Ucar S, Vasile C, et al. (2010) Co-pyrolysis of pine cone with synthetic polymers. Fuel 89: 1911–1918. Chen KS and Yeh RZ (1997) Kinetics of thermal decomposition of styrenebutadiene rubber at low heating rates in nitrogen and oxygen. Combustion and Flame 108: 408–418. Clark C and Meardon Kand Russell D (1991) Burning Tires for Fuel and Tire Pyrolysis. Research Triangle Park, NC, USA: Environmental Protection Agency. Damartzis Th, Vamvuka D, Sfakiotakis S, et al. (2011) Thermal degradation studies and kinetic modelling of cardoon (Cynara cardunculus) pyrolysis using thermogravimetric analysis (TGA). Bioresources Technology 102: 6230–6238. Ebrahimi-Kahrizsangi R (2008) Evaluation of reliability of Coats–Redfern method for kinetic analysis of non-isothermal TGA. Transactions of Nonferrous Metals Society of China 18: 217–221. Encinar JM and Gonzales JF (2008) Pyrolysis of synthetic polymers and plastic wastes: Kinetic study. Fuel Processing Technology 89: 678–686. Environmental Protection Agency (EPA) (1991) Burning Tires for Fuel and Tire Pyrolysis: Air Implications. EPA-450/3-91-024. NC, USA: Environmental Protection Agency. Harker JH and Backhurst JR (1981) Fuel and Energy. London, UK: Academic Press Limited. Islam MR, Haniu H and Fardoush J (2009) Pyrolysis kinetics behavior of solid tire wastes available in Bangladesh. Waste Management 29: 668–677. Jeon MJ, Choi SJ, Yoo KS, Ryu C, Park, SH, Lee JM, Jeon JK, Park YK, Kim S (2011) Copyrolysis of block poypropylene with waste wood chip. Korean Journal of Chemical Engineering 28(2): 497–501. Kar Y (2011) Co-pyrolysis of walnut shell and tar sand in a fixed-bed reactor. Bioresource Technology 102: 9800–9805. Kim SD, Park JK and Chun HD (1995) Pyrolysis kinetics of scrap tire rubbers I: Using DTG and TGA. Journal of Environmental Engineering 121: 507–514.
Waste Management & Research Lapuerta M, Hernandez JJ and Rodriguez J (2004) Kinetics of devolatilization of forescrap tyre wastes from thermogravimetric analysis. Biomass and Bioenergy 27: 385–391. Leung DYC and Wang CL (1998) Kinetic study of scrap tyre pyrolysis and combustion. Journal of Analytical and Applied Pyrolysis 45: 153–169. Lopez G, Aguado R, Olazar M, et al. (2009) Kinetics of scrap tyre pyrolysis under vacuum conditions. Waste Management 29: 2649–2655. Matsuzawa Y, Ayabe M and Nishino J (2001) Acceleration of cellulose co-pyrolysis with polymer. Polymer Degradation and Stability 71: 453–444. Pirayesh H, Khazaeian A and Tabarsa T (2012) The potential for using walnut (Juglans regia L.) shell as a raw material for wood-based particleboard manufacturing. Composites: Part B 43: 3276–3280 Rodriguez IM, Laresgoiti MF, Cabrero MA, et al. (2001) Pyrolysis of scrap tyres. Fuel Processing Technology 72: 9–22. Sait HH, Hussain A, Salema AA, et al. (2012) Pyrolysis and combustion kinetics of date palm biomass using thermogravimetric analysis. Bioresource Technology 118: 382–389. Seidelt S, Müller-Hagedorn M and Bockhorn H (2006) Description of tire pyrolysis by thermal degradation behaviour of main components. Journal of Analytical and Applied Pyrolysis 75: 11–18. Slopiecka K, Bartocci P and Fantozzi F (2011) Thermogravimetric analysis and kinetic study of poplar wood pyrolysis. Applied Energy 97: 491–497. TurkStat (Turkish Statistical Institute) 2014 Available at http://www.tuik. gov.tr/PreTablo.do?alt_id=1001) (accessed 20 March 2014). Uzun BB and Sarioğlu N (2009) Rapid and catalytic pyrolysis of corn stalks. Fuel Processing Technology 90: 705–716. Williams PT and Besler S (1995) Pyrolysis-thermogravimetric analysis of tyres and tyre components. Fuel 74: 1277–1283. Yaman E (2013) Investıgatıon of thermal degradation kinetics of walnut shell with waste plastics via thermogravimetric analysis. MSc. Thesis, Anadolu University, Institute of Applied Science, pp. 37–39. Zhou L, Wang Y, Huang Q, et al. (2006) Thermogravimetric characteristics and kinetics of plastic and biomass blends co-pyrolysis. Fuel Processing Technology 87: 963–969.
