Combustion, kinetics and thermodynamic characteristics of rice husks and rice husk-biocomposites using thermogravimetric analysis

Pyrolysis of rice husk (RH), alkali-treated cellulose-rich rice husk (RHC), chemically modified RHC (RHCM) and RH-biocomposites by thermogravimetric analysis was carried out to determine combustion and kinetic parameters at three different heating rates of 20, 40 and 50 °C min−1. Combustion performance was analyzed from results of ignition temperature, burnout temperature, combustion rates, flammability index and combustion characteristic index. Increase in heating rate from 20 to 40 and further to 50 °C min−1 increased the onset of degradation, burnout and peak temperatures as observed by curve shifts to the right. Maximum combustion rates were around 0.57–0.59% min−1, 1.03% min−1 and 0.63–0.69% min−1 for RH, RHC and RHCM, respectively. For the RH-biocomposites, the maximum combustion rates were in a 0.76–0.97% min−1 range. Their average pre-exponential factors using KAS method were in the 2.24E-03–8.07E-03 range, respectively, while those for OFW method were in the 7.75E + 04–4.55E + 06 range, respectively. Average activation energies of RH-biocomposites were in the 41.0–58.2 kJ mol−1 and 48.3–67.7 kJ mol−1 ranges for KAS and OFW methods, respectively. The data were well fitting with coefficient of determination (R2) values close to 1. Average ΔG value ranges for RH-biocomposites ranged between 148.2 and 161.7 kJ mol−1. The low-energy barrier (≤ 5.4 kJ mol−1) between activation energy and enthalpy changes indicated that reaction initiation occurs easily.


Introduction
It has been reported that over 1.2 million tons of agricultural wastes was generated annually in Uganda [1].One such agricultural waste material is rice husks [2,3].Rice (Oryza sativa) of lowland and upland species has become a staple food in Uganda, with an annual production growth rate of over 9% [4].The processing of paddy into rice produces enormous amounts of husks (over 20% by mass) [5,6].In Uganda, the main disposal means for rice husks are open burning, a method that is associated with negative environmental effects [3,7,8].Additionally, the burning practice of rice husks often leads to air pollution, and the airborne particles can cause respiratory diseases in humans [9].The conversion of specific rice husks species into useful products like bioplastics or biocomposites could be facilitated by knowledge of their combustion characteristics, kinetics and thermodynamic behaviors [2].This knowledge can guide the application of rice husk-derived materials to suitable fields as well as to suitable disposal methods [10].Moreover, to study the influence of modification of rice husks, by e.g., alkali or acid, is important for production of biocomposites, because it modifies the chemical composition and makes their surfaces rougher, which improves adhesion with plasticizers and biopolymer matrices [11][12][13].
Combustion characteristics and kinetic behavior analysis is an important approach to study the mechanisms of the thermochemical conversion of a biomass material and the ICTAC Kinetic Committee recommended use of multiple heating rates to obtain more reliable combustion characteristic and kinetic parameters instead of single heating rates [14].Combustion characteristics and kinetic parameters can be obtained by thermogravimetric analysis (TGA) [15][16][17][18][19]. Combustion characteristic parameters focus on combustion reaction mechanisms while kinetic parameters focus on activation energies of a given material [20].Activation energy is the minimum amount of energy that reaction compounds need to hold in order to initiate a reaction.Various kinetics study on rice husks from different origins have been carried out using various methods [16,[21][22][23][24][25].From these studies, it was observed that most works concentrated on obtaining activation energies, but only very few investigated rice husks' combustion and thermodynamic characteristics [22,26].A comparison of mean value of activation energy of rice husks utilized in the present study with those of rice husks from different origins is shown in Table 1.Table 1 illustrates that obtained activation energies vary depending on the kinetic/fitting method as well as heating rate utilized in their determination and that the obtained values are strongly variable depending on the hydrogeological differences in region or country.This further justifies the gap in determination of kinetic parameters of rice husks, especially for development of bioplastics because it is important for sustainability.
Although the body of research involving kinetics of rice husks from regions outside Uganda is sizable, it is still limited considering the large number of available species of rice husks.In fact, there exists differences in hydrogeological conditions from one region and country to another, which implies that physical properties of agricultural residues are expected to be geo-specific [33].Additionally, weather conditions in a particular season greatly affect the quality of rice and accruing rice husks [34].Moreover, there have been no attempts to determine the synergy of reaction combustion characteristics, kinetic and thermodynamic parameters of rice husks, and their different chemical modifications, aiming at production of biobased materials.Therefore, this research was aimed at characterizing K98 rice husk (RH) and products derived from the K98 RH, including two chemical modifications of rice husk and biocomposites with incorporated rice husk modifications.Especially, thermogravimetric analysis at different heating rates was utilized to determine their combustion, kinetic and thermodynamic

Table 1 Mean activation energies of rice husks from different origins
Rice husk origin Heating rate/°C min −1  Mean activation energy/kJ mol −1  parameters.The kinetic parameters were obtained from TGA experimental data using Kissinger-Akahira-Sunose (KAS) and Ozawa-Flynn-Wall (OFW) methods.

Processing
RH were washed with distilled water to remove any contaminants present in them.They were then oven-dried at 70 °C for 16 h.The dried RH were then crushed in a grinder and sieved through a 60-mesh screen before treatment with KOH alkali solution (6% w/v concentration and liquor ratio of 3:1) at 85 °C for 2 h [35][36][37].This treatment is performed to remove some silica, lignin and hemicellulose.The obtained solid was filtered and washed using distilled water before new oven treatment at 70 °C for 16 h.Acid hydrolysis was then performed using sulphuric acid (4% v/v concentration), and the mixture was heat treated at 70 °C for 2 h.The aim of this treatment is to remove silica, amorphous parts of cellulose and remaining lignin and impurities.The obtained residue was again oven-dried at 70 °C overnight to recover a RH fraction rich in cellulose, abbreviated as RHC.The fraction was bleached by mixing with sodium chlorite (12% w/v) at 80 °C for 4 h before oven-drying at 70 °C overnight [35,36].Part of the RHC was modified through mercerization to increase its strength and for a lustrous appearance, by adding 100 mL of isopropanol and 20% w/v sodium hydroxide (NaOH) to 10 g of bleached RHC [35,36,38].The mixture was stirred for 2 h at 60 °C, and the obtained residue was oven-dried at 70 °C for 4 h.The residue was then mixed with 100 mL of 20% w/v mono-chloroacetic acid and the solution and stirred for 4 h at 70 °C.The residue was washed with ethanol and HCl (37%) mixture at a 1:1 ratio to remove potassium ions, before oven-drying at 60 °C overnight to recover a RHC product, which was further chemically modified and abbreviated as RHCM.

RHC-starch and RHCM-starch based biocomposites
1 g of RHC or RHCM was added to 40 mL of distilled water at 80 °C and stirred continuously.1 g of gelatinized starch powder (for retrogradation) was then added in the solution and stirred to until homogenous solution was obtained [35,36,39].This was followed by addition of first 7 g of glycerol plasticizer and then 4 g of acetic acid.The mixture was stirred for 40 min at 80 °C in a water bath until a precipitate was formed [40].The precipitate was poured into a petri dish, spread to obtain the films, and left to cool at room temperature for 10 min before oven-drying at 55 °C for 16 h [41].

RHCM-gelatin-based biocomposites
1 g of RHCM was added to 40 mL of distilled water at 80 °C and stirred until the solution looked homogeneous. 1 g of gelatin powder (for gelatinization) was added, and the mixture was again stirred until it looked homogeneous.7 g of glycerol plasticizer was added, and the mixture was stirred at 80 °C in a water bath for 40 min until a precipitate was formed [40].The precipitate was poured into a petri dish, spread to obtain the films, and left to cool at room temperature for 10 min before oven-drying at 55 °C for 16 h [41].

Scanning electron microscopy (SEM) and energy-dispersive X-ray (EDX)
Morphology of RH, RHC, RHCM as well as developed RHbiocomposites was observed under a field emission scanning electron microscope (FESEM) coupled with energydispersive X-ray (EDX) spectroscopy, Carl Zeiss Sigma 300 VP model.Samples were attached on double-sided adhesive carbon tape, coated with a 0.2-nm chromium layer using a vacuum sputter coater (model Q 150 T ES), vacuum-dried and scanned at an acceleration voltage of 5.0 kV.EDX spectra were obtained at an acceleration voltage of 20.0 kV and collected for 19 s.

Thermogravimetric analysis (TGA)
Thermogravimetric analysis (TGA) (Eltra Thermostep nonisothermal Thermogravimetric analyzer, Haan, Germany) was used to determine mass loss of RH, RHC, RHCM as well as the developed RH-biocomposites as a function of temperature [19,42,43].TGA was carried out according to the ASTM E1131 standard.The samples were heated from 40 to 600 °C at three different heating rates of 20, 40 and 50 °C min −1 .High-purity compressed air (Oxygen: Nitrogen = 21:79, > 99.99%) was used for cleaning the crucibles and chamber prior to the TGA experiments.Nitrogen gas was used as the purge gas for pyrolysis experiments.The flow rate was maintained at 20 mL min −1 and the sample masses averaged 1.2 g.TGA also provided combustion explanations in terms of differential thermogravimetry (DTG), peak temperatures and mean reactivity of the RH, RHC, RHCM and the developed RH-derived biocomposites.

Combustion characteristic parameters
For each heating rate, ignition temperature, T i is a measure of difficulty to ignite.A low ignition temperature suggests easy ignition, and vice versa.Burnout temperature, T f refers to the temperature of the combustible substances in the completely burnt state.The ignition and burnout temperatures of RH, RHC, RHCM and the developed RH-derived biocomposites were determined according to the tangent method proposed by Liu et al. [44].The flammability index, F, reflects the reactivity of the early stages of combustion and is expressed by Eq. ( 1). where is the maximum combustion rate.The larger the value of flammability index, the more combustible a material is.The combustion characteristic index, S, reflects how quick the combustion rate is and is expressed by Eq. ( 2) [20]. where is the average combustion rate.The larger the combustion characteristic index, the better the combustion characteristics, and the more intense combustion is [45].

Kinetics modeling
From the Arrhenius equation ( k = Ae −E RT ), the rate constant in the kinetic equation is closely related to temperature.A is a pre-exponential factor and T is the absolute temperature.According to Xiao et al. [46], a non-isothermal process can be regarded as an isothermal process within an infinitesimal short time interval.As such, the kinetic equation for an isothermal homogeneous phase reaction can be adopted for the pyrolysis of RH, RHC, RHCM as well as developed RH-derived biocomposites under a programmed heating condition by the means of thermal analysis (see Eq. ( 3)).
where k is the rate constant of the reaction, is the immedi- ate mass loss ratio during pyrolysis and is given by Eq. ( 4), f ( ) is a function that reflects the apparent kinetics of pyrol- ysis of substances, which is expressed as shown in Eq. (5).
where m 0 , m, m 0 are the initial, instantaneous and final masses of the sample, respectively.
where n is the order of the reaction.Combining Eqs. ( 3)-( 5), we obtain: For a given heating rate, = dT dt , Eq. ( 7) can be obtained as the non-isothermal reaction rate.Therefore, Substituting Eq. ( 6) into Eq.(7) gives: In current study, the data obtained from TGA/DTG were used to determine the kinetic parameters (activation energy (E) and pre-exponential factor (A)), based on Arrhenius equation.Further, these kinetic parameters can be estimated graphically by integrating Eq. ( 8) and then applying mathematical approximation for exponential term.Based on previous literatures, five different approximations were carried out and correspondingly, four iso-conversional (model free) models and a model fitting method were adopted to calculate the apparent activation energy (E) at specific conversion time ( ) [17,47].Among the above five models, two integral forms namely Kissinger-Akahira-Sunose (KAS) model and Ozawa-Flynn-Wall (OFW) model were used in this study.
(a) Kissinger-Akahira-Sunose (KAS) method In the KAS method, mathematical approximation for exponential term is assumed, and after approximation and rearrangement, the solution is given by Eq. ( 9) [48]. (3) for a given value of conversion, yields a straight line with slope

R and an intercept ln
The OFW method uses the Doyle's approximation [49] and OFW equation is expressed by Eq. ( 10) [50].
where g( ) is constant at a given value of conversion.A plot of ln i against 1 T α i for a given value of conversion, yields a straight line with slope and an intercept ln

Thermodynamic analysis
The OFW method was used to obtain thermodynamic characteristics of RH, RHC, RHCM and the developed RH-derived biocomposites at a given heating rate, including change in Gibbs free energy (ΔG), change in enthalpy (ΔH) and change in entropy (ΔS) using Eqs.( 11)-( 13) [15,[51][52][53][54]. where where K B is the Boltzmann constant, h is the Plank's con- stant, and T m is the peak temperature during combustion.

Surface morphology
Surface morphology of K98 RH at different magnifications is shown in Fig. 1a-c.The rice husk appeared ( 9) irregular in shape with a size varying in micron order.They formed a well-organized, rigged and compact block-like structure with dented lumps and straight ridges [55][56][57].These structures are formed by cementing noncellulosic materials complexed with cellulose [58].The structures were punctuated with prominent domes [59,60].Figure 1d-i shows the surface morphology of RHC and RHCM produced by chemical modification, respectively.Cellulose fibers typically have long, small cavity and irregular fibrous structure with various thickness [61].Rice husks fiber bundles are separated into individual fibers in plant cells, as lignin and hemicellulose are deposited between the cellulosic micro-fibrils forming an interrupted lamellar structure [62].Chemical modifications to form RHC and RHCM were found to alter the surface structure of remaining larger RHC and RHCM particles, by revealing a more dominant fibrillary network on the outer surface [55,63].Surfaces were rougher after the modification because the chemical treatment is expected to remove particulated material (lignin, hemicellulose, fats and waxes) from the surface of the rice husks [11,64].Fiber bundles were also released, individualized and visualized as microfibrils.This result agrees with those reported Johar and Ahmad [65] and Nascimento et al. [66].
Figure 2a-c shows the surface morphology of RHCstarch biocomposites indicating relatively good adhesion between RHC and starch matrix although some voids in the interfacial boundary were observed in some locations.These voids will act as sites of stress concentration and as a load is applied, fracture at these sites is expected to occur.These regions (see Fig. 2a, b) therefore present some amount of areas with poor bonding between RHC and starch.This result is similar to those obtained by Yap [67] and Bisht et al. [68].
Surface morphologies seen in Fig. 2d-f for RHCMstarch biocomposites showed both large-sized voids and various small-sized voids, and also the surface appears to have more roughness and topological variation (see Fig. 2d) [69].Figure 2g-i presents surface morphologies for RHCM-gelatin-based biocomposite.This material appears to have the most homogeneous dispersion of RHCM compared to the other biocomposites.This can be attributed to the potentially more favorable interactions such as H-bonding interaction between RHCM and gelatin matrix [70].Most of the fibers were firmly attached to the matrix without voids.A smoother surface (see Fig. 2i) also indicates improved adhesion between RHCM and gelatin matrix as result of surface modification through introduction of carboxyl groups, which may favorably interact with amine groups in gelatin.Similar observations were obtained by Bisht et al. [68] and Marichelvan et al. [71].

Elemental analysis of surfaces by EDX
Elemental analysis by EDX revealed that the RH, RHC, RHCM and biocomposite samples had carbon and oxygen as the main constituents with other elements such as silicon, aluminum and sodium having very small percentages (see Fig. 3).As expected there was a high carbon-oxygen ratio (2.1:1) in the K98 RH.The carbon-oxygen ratio for RHC and RHCM were reduced to 1.1:1 and 1:1.1, respectively (see Table 2).The decreased carbon-oxygen ratio supports removal of lignin, which should have higher carbon-oxygen ratio compared to cellulose.These findings are similar to those obtained by Onoja [72] and could be explained by removal of more carbon-rich lignin from the materials.Na increased because of some entrapped Na + from alkali treatment.The samples had very small silica peaks, especially for the developed RH-derived biocomposites.The silica content in RH, RHC and RHCM was as expected higher compared to the RH-derived biocomposites (see Fig. 3d-f).It should be noted that to form RHC, alkali treatment did not significantly reduce the amount of silica, but in RHCM, the amount was greatly reduced.From EDX also, only trace amounts of Mg, Al, and K were detected with no significant changes due to alkali treatment.

Thermal analysis by TG and DTG
Figure 4 presents the TGA curves of RH, RHC and RHCM (Fig. 4a-c) as well as accruing RH-derived biocomposites (Fig. 4d-f) at three different heating rates of 20, 40 and 50 °C min −1 .The curves show the typical appearance of pyrolysis of RH and from them, the thermal phases for each of the heating rates can be located [23,73].It is evident that the complete pyrolysis process for all the samples can be divided into three major stages.The first stage from 40 to 300 °C causes degradation and hence mass loss (≤ 8%) of some light components, including inbound water and light volatile components [16,17,23,30,69,74,75].
The second stage between 300 and 425 °C causes the maximum mass loss (40-50%) during thermal degradation.From the percentage loss in mass, it is clear that the degradation rate of the developed rice husk-derived biocomposites (see Fig. 4d-f) was higher compared to the RH, RHC and RHCM (see Fig. 4a-c) at the same temperature range under consideration.Moreover, for RH-derived biocomposites, the major mass loss in this range was possibly due to the decomposition of gelatin and starch [71].An increase in heating rate leads to a clear shift of this phase to the right-hand side, signaling that higher temperatures are required to cause decomposition of hemicellulose and cellulose [76].At lower heating rate, sufficient time is available for more heating, achieving linear temperature profile between the outer surface and inner core of the RH, while at the higher heating rate, there is shorter exposure time and possible temperature gradient between outer and inner core of the RH [22,77].
In the third stage, from 425 to 600 °C, any remaining lignin in the RH is decomposed and the mass loss continues until char residues are left at the maximum decomposition temperature [69,78].The char residues left were in ranges 44.4-46.1%,28.4-28.5% and 19.9-30.3%for RH, RHC and RHCM, respectively.For RHC-starch, RHCM-starch and RHCM-gelatin, the ranges for char residues were 6.5-7.2%,3.4-4.4% and 6.8-7.1%,respectively.The lower residues are deduced to the lower contents of inorganics and lignin in the biocomposites.The presence of these compounds can also promote the carbonization of polysaccharides leading to high char yields (see Table 2).
Figure 5 presents the differential thermogravimetric (DTG) curves of RH, RHC and RHCM (Fig. 5a-c) as well as accruing RH-derived biocomposites (Fig. 5d-f) at three different heating rates of 20, 40 and 50 °C min −1 .The rate of mass loss (% min −1 ) is very low during the first stage, whereas it is higher in the second stage of the decomposition process, where hemicellulose and cellulose are expected to be decomposed from the rice husks [20,79,80].
The second stage of degradation shows the peak temperature ranges at 3364.8-393.5 °C, 380.3-413.3°C and 365.1-380.8°C for RH, RHC and RHCM, respectively.These high peak temperatures are important in application, especially during processing of RH-biocomposite packaging, so that they are able to withstand processing temperatures before severe degradation.In confirmation, peak temperature ranges for RHC-starch, RHCM-starch and RHCM-gelatin biocomposites were 334.9-365.6 °C, 310.4-362.3°C and 301.1-337.4°C, respectively.These  temperatures are above thermal stability ranges reported in previous studies [71,74,81].It should be noted that increase in heating rate leads to increasing peak temperatures as observed by the shifts of the peaks to the right, due to shorter exposure time at higher heating rates, so the temperature where degradation is initiated will be higher [16,30,53,82,83].For example, for RHCM-gelatin, increase in heating rate from 20 to 40 °C min −1 led to increasing peak temperatures from 301.1 to 333.0 °C.Further increase in heating rate to 50 °C min −1 increased the peak temperature to 337.4 °C (see Fig. 5f).

Combustion characteristics parameters
The combustion characteristic parameters of the RH, RHC, RHCM as well as accruing RH-derived biocomposites are shown in Table 3. Increase in heating rate led to an increase in ignition, peak and burnout temperatures.This result confirms the results for thermal stability presented in Figs. 4 and 5.For example, increasing heating rate from 20 to 40 °C min −1 led to increase in ignition temperature from 290.6 to 309.3 °C for RH.Further increase in heating rate to 50 °C min −1 increased ignition temperature further to 326.7 °C.This trend is consistent with the results obtained by El-Sayed and Mostafa [18].In a similar trend, increasing heating rate from 20 to 40 °C min −1 led to increase in ignition temperature from 226.3 to 258.8 °C for RHC-starch biocomposite.Further increase in heating rate to 50 °C min −1 increased ignition temperature to 269.6 °C.Increasing ignition and burnout temperatures signal enhanced thermal stability since higher temperatures are required to onset degradation of the RH, RHC and RHCM [84].This result is again desired for packaging materials to provide wide processing window.Similar results were shown by El-Sayed [29] who reported that increase in heating rate of rice husks from 10 to 15 °C min −1 led to increase in ignition temperature from 258.0-262.0 to 258.0-264.0°C, respectively.Maximum combustion rates of raw rice husks were around 0.57-0.59%min −1 , 1.03% min −1 and 0.63-0.69%min −1 for RH, RHC and RHCM, respectively.RH-biocomposites' maximum combustion rates were at a 0.76-0.97%min −1 range.These low values signal high flame retardancy due to reducing flammability and combustion characteristic indices [85].A similar trend was observed for average combustion rates.For these, RH, RHC and RHCM had equal rates of 0.10% min −1 , 0.13% min −1 and 0.14% min −1 , respectively, while the developed RH-biocomposites reached higher rates of 0.17% min −1 .Flammability index and combustion characteristic index values were minimal.For example, maximum flammability index and combustion characteristic index were 1.73 × 10 −5 % min −1 °C−2 and 0.85 × 10-8% min −2 °C−3 for RH-biocomposites, respectively.The values generally decreased with increase in heating rate.This was expected since it takes longer to transfer heat from the external environment to the interior of the samples, thereby creating a hysteresis effect [20].Moreover, the low flammability and combustion characteristic indices obtained indicate that the samples generally have very poor combustion performance [86,87].Low flammability and combustion characteristic are desired for biocomposites, not least in packaging applications so that there is no conduction heat transfer between the packaged material and the packaging material.It was, however, clear that RHC and RHCM are more combustible than the original non-modified RH, since their flammability and combustion characteristic indices are higher [45].This is expected because during alkali modification to form RHC and RHCM, some thermally stable compounds like silica and lignin are erased from the surface of the rice husks [88].RH, RHC and RHCM had mean reactivities in the range of 1.48 × 10 −3 -2.71 × 10 −3 % min −1 °C−1 , while the produced biocomposites had reactivities around 2.14 × 10 −3 -2.96 × 10 −3 % min −1 °C−1 .Low mean reactivity is desired so that the RH, RHC and RHCM burn with less ability, thereby enhancing the flame retardancy of produced biocomposites [55].

Kinetics analysis
The variation in degree of conversion of RH, RHC and RHCM (Fig. 6a-c) as well as accruing RH-biocomposites (Fig. 6d-f) as a function of temperature at different heating rates of 20, 40 and 50 °C min −1 are shown in Fig. 6.In order to compute the kinetic parameters, the same values of conversion rate (α) in the range of 0.2-0.8 were considered for the three heating rates [15,17].This range was considered because during the fitting of data using the KAS and OFW methods, the lowest conversion value (0.1) and the highest conversion value (0.9) did not fit well because of lower correlation values [51,89].As the temperature rises, conversion rates increased due to reduction in original mass of the RH (see Eq. ( 4)).The trend followed by the conversion curves is similar to that presented in TG and DTG curves, which present a decomposition in original mass (see Figs. 4 and 5).Moreover, increase in heating rate tends to shift the conversion curve to the right, signaling higher temperatures are required to decompose constituents of RH, RHC, RHCM and RH-biocomposites at higher heating rates.Similar depictions have been presented elsewhere [52,53,76].
Kinetic parameters of RH, RHC and RHCM and RHbiocomposites, such as activation energy and pre-exponential factor for pyrolysis, were determined using KAS and OFW methods based on the thermogravimetric analysis data obtained under different heating rates, based on the thermogravimetric analysis data obtained under different heating rates [15].According to Liu et al. [20], activation energy is the energy required to transform molecules from normality to an active state in which reaction can easily occur.Using the KAS method, activation energies for progressive conversion values need to be calculated according to Eq. ( 9). Figure 7 shows the KAS linear plots of ln (β/T 2 ) versus 1/T for the conversion values within 0.2-0.8 for RH, RHC, RHCM (Fig. 7a-c) and RH-biocomposites (Fig. 7d-f).
The calculated activation energies and pre-exponential factors from the slopes and intercepts of the KAS plots are listed in Table 4.
Using the OFW method, activation energies for progressive conversion values need to be calculated according to Eq. (10). Figure 8 shows the OFW linear plots of ln (β) versus 1/T for the conversion values within 0.2-0.8 for RH, RHC and RHCM (Fig. 7a-c) as well as accruing RH-biocomposites (Fig. 7d-f).The calculated activation energies and pre-exponential factors from the slopes and intercepts of the OFW plots are listed in Table 4.
The apparent activation energies, corresponding preexponential factors as well as respective R 2 values that were obtained from the slopes in each method are shown in Table 4.For RH, RHC, RHCM and accruing RH-biocomposites, R 2 values were close to 1, signaling accuracy and reasonability of the fitting method for the values of the three heating rates [76,90].Similar results were reported by Yuan et al. [91] who found that all the correlation coefficients away from straight line were > 0.95, which indicated the reliability of kinetic parameters evaluated by linear fitting.Both activation energy and pre-exponential factors generally increased with increase in conversion rate because as temperatures increased, changes in mass of the sample became minimal and therefore more energy was required for a reaction or for transformation to occur.Moreover, high pre-exponential factors depict low frequency of molecular collisions in the reaction mixture [52].Additionally, the range for the obtained pre-exponential factors for each method had quite narrow ranges, which indicated the reliability of computed activation energy values [53].
Average activation energy of RH, RHC and RHCM was 95.9 kJ mol −1 , 94.0 kJ mol −1 and 123.4 kJ mol −1 , respectively, using the KAS method.The respective activation energies using the OFW method were 101.2 kJ mol −1 , 99.7 kJ mol −1 and 127.1 kJ mol −1 .These high activation energies are attractive for material to be used as packaging [20].In fact, for RHC-starch, RHCM-starch and RHCM-gelatin, activation energies were 58.2 kJ mol −1 , 41.0 kJ mol −1 and 49.5 kJ mol −1 or 67.7 kJ mol −1 , 48.3 kJ mol −1 and 56.1 kJ mol −1 for KAS and OFW methods, respectively.Average pre-exponential factors using KAS method for RH, RHC and RHCM were in the 0.24E + 00-1.10E+ 03 range, respectively, while those for OFW method were in the 1.57E + 08-7.04E+ 11 range, respectively.For developed RH-biocomposites, average pre-exponential factors using KAS method were in the 2.24E-03-8.07E-03range, respectively, while those for OFW method were in the 7.75E + 04-4.55E+ 06 range, respectively.From this, it was clear that the reaction during KAS was slower than that during OFW [52].The variation of activation energy with increase in conversion rate (α) is shown in Fig. 9, depicting that the activation energy of RH, RHC and RHCM (Fig. 9a-c) as well as accruing RH-biocomposites (Fig. 9d-f) is highly dependent on the conversion rate.This also confirmed the complexity of the process of RH, RHC, RHCM pyrolysis, since their pyrolysis reaction is not a one-step reaction [17,20,26,92].The low correlation coefficient (R 2 ) of conversion rate values less than 0.2 and greater than 0.8 is not considered.From Fig. 9, it is clear that increase in conversion rate led to increasing activation energy, due to thermal degradation of different components of RH, RHC, RHCM and RH-biocomposites with increasing temperature.This trend is similar to other studies based on rice husks [23].The weaker bonds (lower molecular mass compounds) decayed at moderate energy and lesser temperatures, while degradation of stronger bonds (higher molecular mass compounds) needed more energy at greater temperature [20].It should be noted that OFW method presented higher activation energy values compared to the KAS method, but the variations among these methods at each conversion rate were minimal.This was consistent with other literature such as rice husks [30], hazelnut husks [93] and sawdust as well as rice husks [94].Lower activation energies for RH-biocomposites could be due to chemical reaction which washed away some thermally stable compounds form the surfaces of the rice husks [95].

Thermodynamic parameters
Thermodynamic parameters of RH, RHC, RHCM and RHbiocomposites at three heating rates of 20, 40 and 50 °C min −1 were calculated using the OFW method, as shown in Eqs. ( 11)- (13) and are presented in Table 5.It is clear that pre-exponential factors and ΔH (enthalpy of reaction) increased with increase in heating rate, while Gibb's free energy (ΔG) showed an opposite trend as increasing heating rates led to a decrease in the ΔG value.The ΔH (enthalpy of reaction) is the energy exchanged between reactants and products during the chemical reaction [53].The low-energy barrier (≤ 5.4 kJ mol −1 ) between the activation energies (see Fig. 9) and the ΔH values for the process indicated that reaction initiation can occur easily, due to the fact that the lower difference between ΔH and activation energy favors the complex formation [30,32,53,54].This result agrees well with works of Loy et al. [30] on rice husks pyrolysis as well as works of Rasool et al. [94] on rice husks and saw dust pyrolysis.Average ΔH value ranges (depending on heating rate) of RH, RHC and RHCM were 95.6-95.9kJ mol −1 ,    Gibbs free energy reveals the overall energy change of the system, and lower ΔG values indicates favorable decomposition [52].Average ΔG value ranges (depending on heating rate) of RH, RHC and RHCM were 163.0-166.0kJ mol −1 , 167.4-171.5 kJ mol −1 and 161.3-161.9kJ mol −1 , respectively.The high ΔG values obtained in this study indicated that thermal decomposition of RH, RHC and RHCM is a non-spontaneous process [96].The values of Gibbs free energy were found in the same range when compared to values by other authors for rice husks (165.0 kJ mol −1 ) [94] and rice bran (167.2 kJ mol −1 ) [97].In fact, the average ΔG values for developed RHC-starch, RHCM-starch and RHCMgelatin were 157.0-161.7 kJ mol −1 , 151.4-161.4kJ mol −1 , and 148.2-153.8kJ mol −1 , respectively.
The change in entropy (ΔS) is a measure of disorders, and a negative ΔS value depicts decreased disorders of the system, i.e., ordered product formation [52].The low and negative value obtained for RH, RHC, RHCM and RHbiocomposites revealed that extended time is required for thermal decomposition of active material in the rice husks [30,54,98].This characteristic makes rice husks an attractive component for biocomposites because for their application, and thermal decomposition is typically less desirable.Low and negative values were upheld, no matter the heating rate utilized, meaning that heating rate had no striking effect on the entropy change of the RH, RHC, RHCM and RH-biocomposites.

Conclusions
Pyrolysis of RH, RHC, RCM and three RH-biocomposites by thermogravimetric analysis (TGA) was carried out to determine combustion and kinetic parameters at three different heating rates of 20, 40 and 50 °C min −1 .As expected, the TGA established that with an increase in heating rates, the degradation curves moved toward upper-temperature regions, signaling enhanced decomposition temperature due to shorter exposure time.Maximum combustion rates were around 0.57-0.59%min −1 , 1.03% min −1 and 0.63-0.69%min −1 for RH, RHC and RHCM, respectively.The maximum combustion rates of RH-biocomposites were in a 0.76-0.97%min −1 range.Lower values for combustion rate signify enhanced flame retardancy.Average pre-exponential factors using KAS method for RH, RHC and RHCM were in the 0.24E + 00-1.10E+ 03 range, respectively, while those for OFW method were in the 1.57E + 08-7.04E+ 11 range, respectively.For developed RH-biocomposites, average pre-exponential factors using KAS method were in the 2.24E-03-8.07E-03range, respectively, while those for OFW method were in the 7.75E + 04-4.55E+ 06 range, respectively.Average activation energy of RH, RHC and RHCM was 95.9 kJ mol −1 , 94.0 kJ mol −1 and 123.4 kJ mol −1 , respectively, using the KAS method.The respective activation energies using the OFW method were 101.2 kJ mol −1 , 99.7 kJ mol −1 and 127.1 kJ mol −1 .For RHC-starch, RHCM-starch and RHCM-gelatin biocomposites, activation energies were 58.2 kJ mol −1 , 41.0 kJ mol −1 and 49.5 kJ mol −1 as well as 67.7 kJ mol −1 , 48.3 kJ mol −1 and 56.1 kJ mol −1 for KAS and OFW methods, respectively.Average ΔG value ranges (depending on heating rate) of RH, RHC and RHCM were 163.0-166.0kJ mol −1 , 167.4-171.5 kJ mol −1 and 161.3-161.9kJ mol −1 , respectively.The average ΔG values for developed RHC-starch, RHCM-starch and RHCM-gelatin biocomposites were 157.0-161.7 kJ mol −1 , 151.4-161.4kJ mol −1 , and 148.2-153.8kJ mol −1 , respectively.There was a clear relation between the conversion degree and activation energy, which indicated complexity of the rice husks combustion process.The low-energy barrier (≤ 5.4 kJ mol −1 ) between the activation energies and ΔH values for the process indicated that reaction initiation occurs easily.Generally, the obtained results from this study can guide the application of rice husk-derived materials to suitable fields as well as to suitable disposal methods.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material.If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.