Pyrolytic and Kinetic Characteristics of the Thermal Decomposition of Perilla frutescens Polysaccharide

The thermal decomposition of Perilla frutescens polysaccharide was examined by thermogravimetry, differential thermogravimetry, and differential thermal analysis. The results showed that the mass loss of the substance proceeded in three steps. The first stage can be attributed to the expulsion of the water from ambient temperature to 182°C. The second stage corresponded to devolatilization from 182°C to 439°C. The residue slowly degraded in the third stage. The weight loss in air is faster than that in nitrogen, because the oxygen in air accelerated the pyrolytic reaction speed reaction. The heating rate significantly affected the pyrolysis of the sample. Similar activation energies of the degradation process (210–211 kJ mol−1) were obtained by the FWO, KAS, and Popescu techniques. According to Popescu mechanism functions, the possible kinetic model was estimated to be Avrami–Erofeev 20 g(α) = [−ln(1–α)]4.


Introduction
Perilla frutescens (L.) Britton is an annual herbaceous plant belonging to the Labiatae family [1]. P. frutescens is an important cash crop; its flowers, leaves, stems, and fruits are highly valuable [2,3]. The dry ripe fruits of P. frutescens and perilla seeds are rich in oil, having an oil content of 36%-50%, which is higher than cottonseed, rapeseed, and castor seed [4].
After oil manufacture, seed cakes are mainly treated as wastes and abandoned or sold as feed because their use has not been fully studied or developed [5]. Seed cakes such as rapeseed cake, safflower cake, cottonseed, and soybean cake have been investigated as thermal decomposition materials for the production of bio-oil and chemical products [6][7][8][9][10]. However, the pyrolysis of P. frutescens cake has not been studied. The composition and content of a polysaccharide significantly influence its thermal decomposition. Polysaccharides are the major constituents of biomass. With the growing interest in utilizing bio-oil obtained from fast pyrolysis of biomass for fuels and chemicals, understanding the polysaccharide pyrolysis behavior has gained particular importance. Therefore, the thermal decomposition characteristics of the polysaccharide content of P. frutescens cake need to be studied. Polysaccharides are commonly found in biological organisms and exhibit bioactivities such as immunity enhancement, anti-tumor, antiinflammation, and anti-ulcer properties [11,12]. Hence, a study on the thermal stability of P. frutescens polysaccharide, a potential macromolecular drug, is considerably important. The thermal and kinetic analyses are also essential for the identification of polysaccharides and their three-dimensional structure [3][4][5][6][7]. Previous studies have investigated the thermal cracking characteristics of cellulose [13][14][15][16][17], hemicelluloses [18][19][20], and lignin [21][22][23]. However, no study has focused on such characteristics of P. frutescens polysaccharide. Hence, thermogravimetry (TG) and other methods are adopted in the current work to analyze the thermal characteristics and kinetics of P. frutescens polysaccharide. The thermal stability and thermal decomposition of the substance are discussed. A reference for the thermal decomposition of seed cakes and development of polysaccharides is also provided.

Experimental Materials
P. frutescens seeds were purchased from the Boshan Market in Zibo. P. frutescens polysaccharide was prepared in our laboratory according to the following process: raw P. frutescensRpulveriza-tionR40 mesh sievingRde-oiling with 60-90uC petroleum etherR80uC water extraction for 4 hRconcentration with rotary evaporatorRdeposition with 70% alcoholRfreeze dryingRP. frutescens polysaccharide.

Instruments
A TG/DSC STA449C-QMS403C thermal analyzer manufactured by NETZSCH Company (Germany) was used. Experiment P. frutescens polysaccharide weighing 20 mg was placed in an alumina crucible. The substance was subjected to TG, differential TG (DTG), and differential thermal analysis (DTA) in air and nitrogen atmosphere, under the following conditions: flow rate of 80 mL min 21 ; heating rates of 10, 30, and 50uC min 21 ; and from room temperature to 800uC.

Results and Discussion
Influence of different atmospheres on the thermal decomposition of P. frutescens polysaccharide Figures 1A-C illustrate the thermal decomposition of P. frutescens polysaccharide in air and nitrogen. Similar TG and DTG curves are obtained. Absorbed water is lost in the first stage under both atmospheres, with temperatures ranging from room temperature to 189 and 191uC, respectively, with corresponding total weight losses of 13% and 11% as well as maximum weight losses of 5.55% and 6.03% at 96 and 97uC, respectively. The curves are similar for air and nitrogen without weight loss differences, although the weight loss starts 2uC earlier in air than in nitrogen. These conditions demonstrate that the weight loss is independent of the type of atmosphere in this stage and does not depend on the substances being combusted. As shown in the DTA diagram, more thermal discharges occur in nitrogen. The weight loss in the second stage, which corresponds to 189-432 and 190-554uC, is mainly caused by the decomposition of volatile substances and the polysaccharide. The weight losses in air and nitrogen are fastest at 273 (25.2%) and 278uC (22.8%), respectively. In this stage, a considerable difference between the total weight losses of the polysaccharide under the two atmospheres is found (52.8% vs. 60.9%). This finding shows that a fierce decomposition reaction occurs in P. frutescens polysaccharide within this temperature range, and that this app:ds:perillapolysaccharide is comparatively stable below 190uC. The weight loss in air is faster than that in nitrogen, with more abundant and intense thermal discharges, because the oxygen in air is involved in the decomposition reaction and greatly accelerated the pyrolytic reaction speed reaction in air. This phenomenon was consist with conclusion in reference [24]. Consequently, thermal discharge by the combustion of P. frutescens polysaccharide occurs. The total weight loss of P. frutescens polysaccharide is higher in nitrogen because of the wider temperature range for decomposition than that in air. The weight loss within the same temperature range under nitrogen decomposition is 53.3%, which is smaller than that in air. The wider temperature range of the decomposition in nitrogen is based on the slower volatilization or evaporation of weightless substances with low boiling points. The third stage involves the slow decomposition of the remaining materials, generating porous residues. The masses of final residues in air and nitrogen are 19.0% and 33.6%, respectively. In air, a rapid weight loss in P. frutescens polysaccharide appears at 537uC with a weight loss of 64.1%, which is smaller than that under nitrogen environment at 305uC. This phenomenon is due to the fact that P. frutescens polysaccharide intensely burns in air and causes residue polymerization, which covers the surface of materials and hinders the volatilization of internal substances. With increased temperature, the substances on the surface are depolymerized, which results in another remarkable weight loss. Therefore, P. frutescens polysaccharide is more stable in nitrogen. A similar phenomenon was also reported in other literature [25].
As shown in Figure 1, two peaks representing thermal discharges can be found in the pyrolysis of P. frutescens polysaccharide in air at 398 and 562uC, in contrast to the thermal discharge peak in nitrogen at 560uC. The thermal discharge in air is higher than that in nitrogen because the thermal discharge in air is based on oxidative combustion, whereas that in nitrogen depends on the volatilization and evaporation of substances or cracking of the polysaccharide. These phenomenons were consist with conclusions in reference [24]. The DTA peaks are smaller than the DTG peaks, which indicate that the thermal discharge reaches the maximum at the maximum rate of mass loss, during which the substances start to absorb heat. The increase in temperature, substance decomposition, as well as escape and thermal discharge of the products lead to the maximum values for the rate of mass loss and thermal discharge. The escape of the products decreases with significantly decreased pyrolysis products, slower mass loss of substances, polymerization of pyrolysis products, and thermal absorption. Consequently, the DTA peak flattens.

Comparison of the pyrolysis characteristics under different heating rates
As shown in Figures 2A-C, the starting temperature for pyrolysis in stages 1 and 2, the temperature for the maximum mass loss rate, and the maximum mass loss rate are all enhanced with increased heating rate. The temperatures for the maximum mass loss rates are 263, 278, and 290uC corresponding to the heating rates of 10, 30, and 50uC min 21 . These phenomena are caused by the higher thermal inertias at higher heating rates. The temperature difference is larger for higher thermal discharge value, consistent with the maximum and second highest thermal discharges at 50 and 30uC min 21 , respectively, in the DTA curve. Hence, the heating rate significantly influences the pyrolysis of P. frutescens polysaccharide. As shown in Table 1, P. frutescens polysaccharide exhibits better thermal stability than other carbohydrates, with a higher residue mass at 800uC. The amount of residues and peak of the second stage differ among various carbohydrates. This finding is related to the source and type of polysaccharide. Table 1. Decomposition temperatures and residues of seven carbohydrates [11,12].
The FWO equation is written as The KAS equation is written as The Popescu equation is written as Tj~T where b is the linear heating rate (uCNmin 21 ); E is the activation energy; (JNmol 21 ); A is the pre-exponential factor; R is the molar gas constant; g(a) is the integral mechanism function; and T, T n , T m , and T j are temperatures at the time of n, m and j.

Determination of the parameters of thermal analysis kinetics
The activation energy can be determined without confirmation of the reaction mechanism based on the FWO and KAS methods. Temperature integration is not adopted in the Popescu method to avoid the Arrhenius equation and compensation action, among others [27]. Therefore, the mechanism function can be reliably confirmed using the Popescu method. FWO and KAS are adopted to verify the calculation results of the Popescu equation. The linear fitting of the change rates based on the three methods is shown in Figure 3 (a: FWO, b:KAS, c: Popescu), and the kinetic parameters obtained are shown in Table 2. The activation energies calculated from the three methods are similar to the lnA value. Therefore, the activation energies calculated by the three methods are generally all effective and similar to one another. However, the activation energy fluctuates under different conversion rates, which can be attributed to the complicated sample composition and complex reaction during pyrolysis. Table 3 compares various kinetic parameters of pyrolysis between different biomass sources [25,[33][34][35]. None of these published results are similar to ours, which suggests that thermal behavior is influenced greatly by the type of feedstock.

Determination of the thermal analysis kinetics mechanism
The pyrolysis mechanism is confirmed by the different conversion rates under different heating rates and temperatures. Table 4 shows that the mechanism function is more accurate when the relation coefficient r is higher and the standard deviation is smaller [11]. The mechanism function Avrami-Erofeev 20 g(a) = [2ln(1-a)] 4 is the most suitable for the kinetic mechanism function for the pyrolysis of P. frutescens polysaccharide. The core is randomly formed and then predominates in the pyrolysis of P. frutescens polysaccharide (n = 4). This result can be attributed to the heterogeneous nucleation of the volatile substances caused by the other components within the polysaccharide. This kinetics mechanism is different with that in literature 24. This is probably due to the difference of polysaccharide component and content.

Conclusions
The pyrolysis of P. frutescens polysaccharide can be divided into three stages, namely, the water evaporation of cells, the cracking of the polysaccharide, and the slow decomposition of residues. The heating rate plays a significant role in the pyrolysis of P. frutescens polysaccharide. Nitrogen favors the stability and storage of this compound. The Popescu, FWO, and KAS techniques, which yield similar activation energies, are suitable for the determination of the kinetic parameters of P. frutescens polysaccharide. The mechanism function Avrami-Erofeev 20 g(a) = [2ln(1-a)] 4 is the most suitable kinetic mechanism function in the second stage of the pyrolysis of P. frutescens polysaccharide.