Pyrolysis kinetic behaviour and thermodynamic analysis of waste wind turbine blades (carbon fibres/unsaturated polyester resin)

ABSTRACT This research aims to study the pyrolysis characteristics and thermodynamic analysis of waste wind turbine blades (WTBs: composed of carbon fiber/unsaturated polyester resin) using a thermogravimetric analyzer (TG). The distribution of pyrolysis vapor products and their chemical compositions were analyzed. Also, pyrolysis experiments were carried out at various heating conditions to examine its effect on the vapor composition, and their pyrolysis kinetics was simulated using several modeling routes. In addition, all degradation regions of WTBs were mathematically simulated. The results showed that WTBs are rich in carbon element (53–88 wt.%) and volatile matter (26–30 wt.%), whereas TG measurements revealed that WTBs can fully degrade up to 500°C with a mass loss of 31%. Carbonyl (C=O) was the major functional group in the released vapor with a constant intensity even when the heating rate changed, while styrene was the major Gas chromatography-mass spectrometry compound with a significant abundance of 87% (at 30°C/min) versus 84% (at 5°C/min). Finally, kinetic investigations manifested that WTBs with a specific configuration have lower average activation energies (Ea) estimated at 172 kJ/mol (KAS), 220 kJ/mol (FWO), 194 kJ/mol (Friedman), and 159 kJ/mol (Vyazovkin and Cai), and FWO provides the highest R2 of 0.99. This decrease in Ea and the reaction complexity is caused by high conductivity of carbon fibers which helps in increasing the heat transfer and speeding up the reaction. The average enthalpy and Gibbs free energy were estimated in the ranges of 166–215 kJ/mol and 114–226 kJ/mol, respectively. Based on that, resin and carbon fibers can be recovered from WTBs using pyrolysis treatment.


Introduction
Wind energy is one of the most promising renewable energy sources, the demand of which has increased in recent times due to its sustainability, high production capacity, cleanliness, and safety.However, its limited operational life (less than 25 years) is a major challenge for environment and waste management sectors, especially since the cumulative of its waste of wind turbine blades (WTBs) is expected to reach 43 million tons globally by 2050 (Lichtenegger et al. 2020).Therefore, good management of WTBs is highly required to close the loop of wind energy's circular economy and to help turn it into a completely green approach.These WTBs are a type of fiber-reinforced polymer composite with sandwich construction (Beauson et al. 2022).In addition, WTBs have superior thermochemical stability because of strong bonding between polymer resin and fibers through chemical and frictional bonds that are resistant to degradation with difficult recycling and fiber extraction (Subadra et al. 2020(Subadra et al. , 2020)).The polymer resin in WTBs can be of a thermoplastic or thermosetting cross-linked type, and manufacturers choose between these types based on several factors, where cost and mechanical performance are among the most important ones (Striūgas et al. 2021;Yousef et al. 2021).Meanwhile, glass fibers and carbon fibers are the most common types of fibers that can be found in WTBs (Subadra, Griskevicius, and Yousef 2020); however, carbon fibers have higher mechanical properties but are more expensive, and their recovery or reuse contributes significantly to sustainability of this sector (Upadhyayula et al. 2022).Recently, several conventional and advanced techniques have been employed to extract fibers from the blade matrix and to reuse them in some ancillary fields.These recycling techniques can be limited to mechanical processes, chemical solvolysis, and thermal conversion treatment (Leon 2023).In fact, mechanical processes (e.g., milling, crushing, shredding, etc.) cannot be categorized as main processing but are closer to easy pretreatment processing that is used to cut WTBs into small pieces and then grind them into resin-rich fibers called "wet fibers" (Yazdanbakhsh et al. 2018).Due to random distribution of resins on the fibers even after mechanical sieving, these wet fibers have inhomogeneous sub-applications and properties (Xu et al. 2022).Therefore, chemical solvolysis has been used to decompose cross-linked resins from the surface of wet fibers using several solvents such as Dimethyl sulfoxide, Dimethylacetamide, acids, subcritical water, etc., by overcoming Van der Waals' bonds (Tatariants et al. 2018;Yousef et al. 2020).Although this approach helped to recover fibers from wet fibers matrix with high purity and minor effects on their mechanical behavior, however, resin extraction and solvent regeneration are the main limitations for their further industrial application, especially since they must take place almost continuously (Martinez-Marquez et al. 2022).In addition to the toxicity of solvents, emissions, and pollution, secondary residues (secondary contamination) resulting from handling of fluids used in chemical solvolysis have to be disposed of safely as well (Khalid et al. 2023).
All of these limitations can be avoided by utilizing a thermal conversion scenario that can be used for thermal depolymerization resins (Ming-Xin et al. 2023;Xiong et al. 2022).Pyrolysis and gasification are the common terms for thermal processing with high techno-economic performance (Lekavičius, Striūgas, and Striūgas 2023;Stasiulaitiene, Zakarauskas, and Striūgas 2023).However, the pyrolysis process needs lower temperature (550°C) to decompose resins into liquid and gaseous products in the absence of air.The gasification technique that needs higher decomposition temperature of 1000°C and high pressure of 45 MPa to convert polymer resin into a gaseous product called syngas, which can affect the properties of the fibers (Chen et al. 2022).Typically, a gaseous product is used as a heating source for a pyrolysis reactor on a commercial scale, while oil can be used as a fuel product or chemically separated into its original compounds (Striūgas et al. 2021;Yousef et al. 2022).At the same time, fibers can be extracted without deteriorating or affecting their mechanical properties, which makes pyrolysis process one of the most promising emerging technology for fibers and energy recovery from WTBs (Yang, Kim, and Lee 2022).Accordingly, recent studies have accelerated the investigation of WTB pyrolysis based on three research directions.In the first direction, the researchers were centered on study of the influence of WTBs composition (commercial and laboratory prepared) on thermal decomposition process [24] and compounds formulated in laboratory using thermogravimetry (TG) (Lichao et al. 2023) and Gas chromatography-mass spectrometry (GC/MS) (Kiminaitė et al. 2022;Striūgas, Eimontas, and Abdelnaby 2021).The results announced that resins can decompose further into 60% phenol and acetic acid compounds (Kiminaitė et al. 2022).Also, the type and abundance of GC/MS compounds can change by other micro and nano additives or inclusion of catalysts into the reaction (Striūgas et al. 2023).In the second direction, the researches were focused more on studying pyrolysis kinetics behavior of WTBs with different compositions (Lichao et al. 2023) using various kinetics models (Chen et al. 2023) to determine their activation energy (Ea) and reaction mechanism (Abdelnaby, Eimontas, and Striūgas 2022;Yousef et al. 2022).The results showed that these additives and catalysts can reduce Ea and the reaction complexity (Amin Mousavi Khorasani, Sahebian, and Zabett 2020).In the last direction, the optimal conditions of pyrolysis (temperature, heating rate, reactive gaseous agents like H 2 O and CO 2 , and pyrolysis time) that can produce more volatile compouds and reclaim of fibers were studied.The results in this respect showed that resin matrix is easy to crack at above 500°C with higher polymerization of resins and formulation of more aromatic chemicals with high abundance, especially at higher heating rate regardless of the chemical composition of resins (Ming-Xin et al. 2023).Despite of all the interesting findings in the specific directions, these studies were focused on epoxy resins, thermoplastic polyurethanes, and poly(methyl methacrylate)-PMMA, while polyester thermoset resin is still absent from the researches, although it is considered to be more cost-effective option for production of blades (Amin Mousavi Khorasani, Sahebian, and Zabett 2020).This has created a huge gap in understanding of pyrolysis of WTBs composed of carbon fibers/polyester, and this is critical in the fields of commercial WTBs recycling.This research aims to fill this gap and study the fundamental pyrolysis properties and thermodynamic analysis of commercial WTB composites composed of carbon fibers/unsaturated polyester resin composites and their kinetic behavior.Kinetic parameters were investigated using various modeling methods.This proposed research can provide the right path fo the pyrolysis disposal mechanism of WTBs and the recovery of resin and carbon fibers from them and their basic data for future reactor construction and operation.

Feedstock preparation
The WTBs in the form of small panels with carbon fiber-reinforced polyester resin composites were supplied by European Energy, Vilnius, Lithuania.In order to improve heat transfer during the treatment process of WTBs (Striūgas et al. 2022), the supplied panels were crushed and milled, then exposed to sieving process to prepare uniform powder for the experiments.The physicochemical behavior of the milled WTBs, including proximate and ultimate properties, were analyzed by Perkin Elmer 2400 CHN apparatus according to E1756-01, E872-82, and E1755-01 standard procedures three times for each sample, followed by calculation of average values (Abdelnaby, Eimontas, and Striūgas 2022b;Striūgas et al. 2021).

Thermogravimetric analysis
A STA449 F3 TG analyzer was utilized to perform thermal decomposition measurements on the prepared WTBs.Approximately, 10 mg of WTBs were filled into a crucible, then put into a furnace, and the reaction temperature was raised from 24°C to 900°C with an inert gas (nitrogen) flow rate at 60 ml/min.The temperature was increased at different heating rates of 5, 10, 15, 20, 25, and 30°C/min.The mass loss obtained from TG analysis and its loss rate obtained from Differential thermogravimetric analysis (DTG) data were recorded, followed by numerical differentiation.Their curves were fitted and pyrolysis kinetics of WTBs was described.Also, based on the formulas Eqs.(1, 2) in Table 1 and the data derived from TG and DTG, devolatilization index and the heat-resistance index (THRI) were studied (Striūgas et al. 2022;Yousef et al. 2022).

Chemical analysis of pyrolysis gaseous products
A TG and Fourier-transform infrared spectroscopy (FTIR) were applied to perform the TG-FTIR coupling experiments and to define the structural units of gaseous products generated at the heating rates specified above.In addition, the different substances of these gaseous products were generated from primary reactions of WTBs in the main decomposition region, and the effect of heating conditions on abundance of their chemical compounds was detected via GC/MS.The analysis of the captured gaseous fraction using Micro-Autoinjector™ was characterized based on the following column settings: Argon inert gas ≥ 99.99%, applied pressure = 20 psi, temperature = 100°C, and time = 120 s.The chromatographic conditions were adjusted in this way: injector temperature = 95°C, measurement range is 30-600 m/s, frequency is 50 Hz, TCD temperature is 70 ± 1°C, pumping time = 20 s, and injection times = 30 ms (Mohamed et al. 2023;Yousef et al. 2021).

Pyrolysis kinetics of WTBs
The previous studies showed that the TG process of WTBs remains a heterogeneous interaction under non-isothermal conditions even when WTBs composition changes (Lichao et al. 2023;Striūgas et al. 2023).In addition, these studies agreed that thermal degradation of WTBs is a complex process which can be identified by studying their pyrolysis kinetics and Ea (Praspaliauskas et al. 2020).In fact, there are many kinetic models that can be used for this purpose, and the quality and accuracy of most of them are usually controversial and choosing the best model is challenging.These models can be divided into two main approaches: estimation of Ea for the whole degradation process and estimation of Ea for each conversion rate (α) (Mohamed et al. 2022).Kissinger is the most common model used to determine Ea for the whole process due to its simplicity; its formulation is shown in Eq. (3) (Eimontas et al. 2021).Meanwhile, Kissinger Akahira Sunose (KAS), Friedman, Flynn Wall Ozawa (FWO), Vyazovkin, and Cai are the most common linear and nonlinear models used to determine Ea at each α, and their formulation are illustrated in Eq. (4-8) (Abdelnaby et al. 2021).In the present research, Ea of WTBs was determined using all these models, then choosing the most suitable model based on the highest value of coefficient of determination (R 2 ).In addition, pre-exponential factors (A) were calculated at each α by estimating the the slope of ln(β/T 2 )-1/T (KAS), ln(β)-1/T (FWO), and the corresponding DTG values to the conversion rate −1/T (Friedman) fitting curves.In addition, the experimental data obtained from the TGA-DTG measurements of WTBs with different heating rates were mathematically predicted using the distributed activation energy (12) Y-axis of Cai plot (13) Dev.% Dev:% ¼ 100 ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi Tm method (DAEM) and the independent parallel reactions (IPR), and their formulations are shown in Eqs.(9, 10) in Table 1.The definitions of these parameters are illustrated in Supplementary materials section (Table S1).

FTIR analysis of WTBs components
Figure 1 shows FTIR spectra of resin polymer and fibers' components of the supplied WTBs.The analysis was performed on two samples collected from different places.The FTIR spectra of polymer resin components (Figure 1a) show similar features with only one strong functional group at 890 cm −1 corresponding to bending modes of SiO 2 .Other few peaks were noticed at 2918 cm −1 (C-H stretching), 2366 cm −1, and 1700 cm −1 (O-H bonds stretching), 1970 cm −1 (O-H stretching), and 700 cm −1 (aromatic bonds C-H stretching).These spectra and functional groups showed that the analyzed organic material was unsaturated polyester polymer (resin) mixed with gelcoat that was used as a coating layer to protect the surface of wind blades from erosion and wear in harsh climates (Khan et al. 2013;Kjaerside Storm 2013;Lakshmi et al. 2013).Meanwhile, the FTIR spectra of the fibers' components (Figure 1b) manifested many functional groups, and the significant one was at 700 cm −1 (C-H bonds stretching) because the surface of short fibers was contaminated with resin debris.Besides, some other vibrations in the ranges of 1948-2155 cm −1 and 1447-1507 cm −1 corresponding to C=O bond and carbonyl group stretching, respectively, were detected, which is typical to the main spectrum of carbon fibers (Subadra et al. 2022;Tiwari, Bijwe, and Panier 2011).

Ultimate and proximate properties
The ultimate and proximate properties of the resin and carbon fibers components of the supplied WTBs are shown in Table 2.The ultimate analysis released that the fibers' component is rich in carbon content (C: 88 wt.%), which serves as an evidence that the fibers used in production of the matrix was carbon fibers.In addition, 1.88 wt.% of hydrogen (H) and 29.88 wt.% of volatile matter contents were observed due to contamination of the sample with some organic compounds resulting from resin component (Gonçalves et al. 2017).Meanwhile, fixed carbon (F.C) was the predominant content due to high proportion of fibers in the tested sample.At the same time, a complete absence of sulfur (S) element in both components was noticed, but nitrogen (N: 2.985 wt.%) was present only in the fibers sample, hence indicating the potential for NOx emissions during the conversion process, which can be easily managed and controlled using a tight seal system to prevent any leakage during processing on a large industrial scale (Mohamed et al. 2021;Zakarauskas et al. 2022).Finally, these organic compounds in the resin can decompose thermally using the pyrolysis process and the ash fraction in carbon fibers which has a negative effect on the pyrolysis process (Yousef et al. 2021), while the fibers can remain as a solid residue, as it will be explained later.

Thermogravimetric analysis
Figure 2 shows the effect of heating rates on the TG and DTG profiles of the supplied WTBs samples, while all the pyrolysis characteristics are presented in Table 3.As shown in TG profiles (Figure 2a), the samples manifested a very high resistance to degradation up to 200°C without almost any mass loss.This resistance decreased gradually with slight mass loss of <2 wt.% up to 380°C due to evaporation of  the remaining chemicals (e.g, solvents, curing, hardener, etc.) during the manufacturing process.Afterward, this resistance collapsed significantly till 500°C with mass loss of 31 wt% due to degradation of resin and coat layer's sub-components, which can decay together in parallel reactions (Manfredi et al. 2006).Finally, another stability was observed in the last phase of reaction due to formation of soild residue fraction (including ash and fibers which usually constitute a high proportion in the matrix) without any further weight loss (Subadra et al. 2021).These results agree with DTG (Figure 2b) results that show only one significant degeneration peak in the ranges of 370-500°C (main degeneration zone) even with changing heating conditions.However, the location of the degraded peaks shifted since the melting temperature raised from 411°C to 448°C with an increase in the heating rate as a result of production of more heat flux supporting heat transfer from the surroundings of the dispersion particles and their inner particles until complete degration occured (Abdelnaby, Eimontas, and Striūgas 2022a;Ni et al. 2022).

TG-FTIR analysis of the pyrolytic compounds
Figures 3 and 4 display 2D and 3D FTIR spectroscopy of the released pyrolytic compounds from WTBs under different rates.The 2D FTIR spectrum (Figure 3) showed that all gaseous compounds generated under the specified heating rates had identical functional groups, in particular at 3029 cm −1 , 2341 cm −1 , 1741 cm −1 , and 1000-1136 cm −1 corresponding to CH 2 asymmetric stretching vibrations, carbonyl group stretching, and aromatic hydrocarbons, respectively (Selvaraj et al. 2020;Wu et al. 2022).As shown, the generated gaseous product was very rich in aromatic hydrocarbons, but their intensity and that of the other groups did not change with increasing heating rate.Whereas, the 3D FTIR Figure 4(4) spectrum showed a clear spectra without noisy vibrations, especially at smaller and larger heating rates.

GC/MS analysis of of the pyrolytic compounds
Since the FTIR results showed similar functional groups for gaseous products of WTBs with almost the same intensity for all conditions, GC/MS measurements was perfomed on vapor generated at smaller and larger heating rates to check their effect on chemical composition and their abundances in terms of peak area (%).The GC/MS spectra and abundances of gaseous compounds at 5 and 30°C/min are shown in Figure 5(5) and Table 4. Styrene was the key compound of gaseous compounds with abundance estimated at 84.33% (5°C/min) and 86.88% (30°C/min).In addition, other compounds with much smaller abundance (<3.62%) such as Toluene, 2-Propenoic acid, 2-methyl-, 2-Propenal, etc. were observed.As shown, the heating rate did not have a significant impact on the concentration of styrene, and these results match with the FTIR results.Besides, it was confirmed that resin used in the supplied WTBs was an unsaturated polyester polymer, where styrene was used in such resin to improve its thermal stability and mechanical properties (Cousinet et al. 2015;Sanchez, Zavaglia, and Felisberti 2000).Finally, the recovered styrene can be used in the production of resin and adhesive materials or in other fields, such as rubber manufacturing, packaging, gaskets materials, conveyor belts, tires production, and foam synthesizers (Dahal et al. 2023).

Mechanism of carbon fibres/polyester WTBs pyrolysis
As shown in the above analysis, the supplied WTBs were composed of different components, in particular carbon fibers, unsaturated polyester resin, and gelcoat coat layer.Also, these components consisted of different organic elements, such as modified silicone oil, vinyl monomer, low-molecularweight unsaturated polyester, graphite, and other additives (Khan et al. 2013;Kjaerside Storm 2013;Lakshmi et al. 2013;Sun et al. 2020).All these elements make the thermal degradation mechanism of WTBs a little bit complex.However, carbon fibers can be set aside since it has a high melting point, and we could focus more on decomposition of organic compounds (resin and gelcoat coat) based on the features and characteristics received from TG analysis.The pyrolysis mechanism of these organic components was studied at 30°C/min, where it revealed the highest quantity of styrene (87%), and the proposed mechanism is shown in Figure 6(6).As shown in the mechanism diagram, the blade exhibited resistance as high as 200°C (<0.2 wt.%) with a very small drop due to moisture evaporation.A further slight decrease was observed up to 310°C (<0.8 wt.%) due to evaporation of chemical residues in the coating and resin components (Kjaerside Storm 2013).Afterward, three successive dismantling stages were observed to be distributed in such a manner.The first stage resulted from the dissociation of resin from fibers (<1 wt.% up to 310°C) by breaking the chemical and mechanical bonds between them and dividing into undegradable short fibers and degradable resin debris (Tatariants, Bendikiene, and Denafas 2017).In the second stage, the molecules of resin debris dismantled into smaller molecules (<1 wt.% up to 380°C) after having broken down their polymer chains and having overcome their Waals' bonds under the applied temperature (Tichonovas et al. 2020).In the last stage, the resin particles with smaller molecules were completely decomposed into volatile compounds, such as water, light, and heavy hydrocarbons components, carbon dioxide, etc. (< 31 wt.% up to 500°C).It is worth mentioning that the coat layer was dismantled and decomposed in parallel with resin component or simultaneously by the same degeneration mechanism, and some of their compounds were consumed in the reaction as catalysts to increase the amount of hydrocarbons (Tuckute et al. 2019).Finally, after decomposition of organic components of WTBs, devolatilization took place with higher thermal stability until the end of the reaction due to solid carbonaceous formation, including short carbon fibers and carbon black powder (Kalpokaitė-Dičkuvienė et al. 2021).It is clear that carbon fibers/polyester WTBs have a pyrolysis mechanism that is almost the same as that of glass fiber reinforced epoxy resin composites (Yousef et al. 2022), but with different generated GC compounds.

Activation energy for the entire conversion process
Kissinger kinetic modeling route using Eq. ( 3) was applied to determine Ea for the entire conversion process of WTBs by fitting 1000/Tm and ln(β/Tm 2 ) linear relationship (Figure 7a), then generating its slope value which can be formulated as -Ea/8.31(kJ/mol) (Gajera et al. 2023).The model showed that Ea of WTBs was estimated at 173.5 kJ/mol.Compared to other WTB compositions (such as fiberglass and epoxy resin (Kiminaitė et al. 2022(Kiminaitė et al. , 2022;;Striūgas, Eimontas, and Abdelnaby 2021)), WTB carbon fibers are characterized by low Ea and reaction complexity due to high conductivity of carbon fibers which helps to use entire pyrolysis heating for decomposition of raw materials without any losses (Ye et al. 2022).In addiition, this heat can be used to accelerate digestibility and to convert it into hydrocarbon products (Thuengtung et al. 2023).

Linear modeling methods.
In this approach, Ea of pyrolysis of WTBs at each conversion region in the ranges of 10% to 90% of reaction was determined using three different linear kinetic models, in particular KAS (Eq.( 4)), FWO (Eq.( 5)), and Friedman (Eq.( 6)).In all of these linear models, Ea was estimated based on a curve fitting concept, followed by the estimation of their slopes as -Ea/8.31(KAS and Friedman) and − 1.0516Ea/8.31(FWO) (Abdelnaby et al. 2020;Eimontas et al. 2021). Figure (7b-d) displays the fitted curves obtained from each linear model, which consist of nine fitted linear relationships referring to conversion rate in the regions specified above.As shown, the plots of KAS and FWO consisted of parallel lines from the beginning of the reaction to its end; only a slight deviation was observed at the first conversion rate in both models due to production of several parallel interactions at this level leading to this discrepancy.On the contrary, in case of the Friedman model, randomness dominated the fitting relationship in the majority of conversion areas.These features were observed before with other feedstocks, and the authors hypothesize that this is due to the nature of the kinetic model that is very sensitive to noise measurements derived from experimental TGA data, hence affecting distribution of the fitting curves, slope values, and accuracy of the calculated Ea (Striūgas et al. 2020).The mathematical calculations showed that Ea changes with the increase of conversion, and their distribution is shown in Figure 8 (8).This means that the reaction was slightly more complex than those of other WTBs [24- (Kiminaitė et al. 2022).Lastly, the deliberated value of Ea and A from each model and their average values and R 2 are illustrated in Table 5.As indicated, the average Ea was estimated at 172 kJ mol −1 (KAS), 222 kJ mol −1 (FWO), 194 kJ mol −1 (Friedman) with R 2 values at 0.95 (KAS), 0.99 (FWO), and 0.96 (Friedman), respectively.In addition, FWO provided the highest R 2 meaning, so it is the most suitable linear model to study pyrolysis kinetic behavior of WTBs.However, this model showed high Ea (222 kJ mol −1 ) compared to Ea of pyrolysis of unsaturated polyester resin due to the involvement of carbon fibers in the reaction, which makes the reaction more complex and needs more energy to overcome the strong mechanical and chemical bonds between fibers and resin (Bai et al.  stability and can decompose at lower temperatures (Subadra, Griskevicius, and Yousef 2020).
Compared with WTBs with different composition in the literature (125 kJ/mol) (Chen et al. 2023;Lichao et al. 2023), carbon fiber-reinforced polyester resin WTB composites have a higher Ea (upto 172 kJ/mol), which means that the reaction has become more complex.

Nonlinear modeling methods.
Such approach is classified as a non-isothermal kinetic method, and Ea at each conversion region of pyrolysis of WTBs can be numerically captured after several iterations using integration algorithms (Abdelnaby et al. 2020) built by Matlab program based on the formules listed above.The algorithms were solved using 200 kJ/mol as an initial iteration to start the optimization process until Ea values became fixed, which was achieved after four iterations in the present research.All Ea received during the whole iteration process are presented in Table 6.As indicated, both models have similar values of average Ea estimated at 159 kJ/mol (Vyazovkin and Cai) with R 2 values at 0.94 (Vyazovkin) and 0.93 (Cai).Also, Vyazovkin and Cai fitting relationships of pyrolysis of WTBs are shown in Figure 6(6E, F), and their axes are defined by formulas (11,12).

Plotting of TG-DTG curves of pyrolysis of WTBs
Figure 9(9) illustrates the simulated TG-DTG practice curves of pyrolysis of WTBs using DAEM and IPR.As shown, DAEM succeeded to mathematically identify the thermal degradation zones and features of pyrolysis of WTBs with deviation below 0.5% (Dev.: calculated using Eq. ( 13)) at all heating rates (only curves obtained at lower and higher heating conditions were presented).In addition, the DTG curved were fully modeled with high prediction.The optimum factors used in the fitting process for particular activation energies: E1 (219.19 and 200.28 kJ/mol) and E2 (271.79 and 256.69 kJ/mol) and pre-exponential: A1 (4.09E + 13 and 1.61E + 09) and A2 (4.48E + 13 and 1.72E + 10) were determined.

Figure 1 .
Figure 1.FTIR spectroscopy of a) polymer resin and b) fibres fraction of WTBs.

Figure 2 .
Figure 2. a) TGA and b) DTG profiles of WTBs samples.

Figure 8 .
Figure 8. Distribution of activation energy of WTB samples in all conversion regions.

Table 2 .
Ultimate and proximate properties of WTBs.

Table 5 .
Activation energy and pre-exponential factor at each conversion rate of pyrolysis of WTBs.

Table 6 .
Ea values calculated using Vyazovkin and Cai models after many iterations.