Determination of the Pyrolytic Characteristics of Various Biomass Pellets

Abstract

Biomass pellets are widely used for combustion but can also serve as sustainable feedstocks for pyrolysis. This study examined wood (WP), palm-pruning (PP), reed (RD), and daphne (DP) pellets. We present a compact framework linking composition (proximate/ultimate and lignocellulosic fractions) with TG/DTG, FTIR, TGA-derived indices (CPI, D_dev, R_w), T_pmax and R_av to predict product selectivity and temperature ranges. TG/DTG showed the following sequence: hemicellulose (≈200–315 °C) first, cellulose (≈315–400 °C) with a sharp maximum, and lignin ≈200–600 °C. Low-ash WP and DP had sharper, higher peaks, favoring concentrated devolatilization and condensables. Mineral-rich PP and RD began earlier and showed depressed peaks from AAEM catalysis, shifting toward gases and ash-richer chars. Composition shaped these patterns: higher cellulose increased R_av and CPI; links to T_pmax were moderated by ash. Lignin strengthened a high-T shoulder, while hemicellulose promoted early deacetylation (RD’s 1730 cm⁻¹ acetyl C=O) and release of CO₂ and acids. Correlations (|r| ≥ 0.70) supported these links: VM with total (m_∞) and second stage mass loss; cellulose with R_av and CPI (T_pmax moderated by ash); lignin and O/C with T_f and last stage mass loss; ash negatively with T_i, T_pmax, and m_∞. The obtained results guide the sustainable valorization of biomass pellets by selecting temperatures for liquids, H₂/CO-rich gases or low-ash aromatic chars.

1. Introduction

In recent periods, biomass has emerged as a vital component in the quest for sustainable energy solutions, offering a renewable alternative to fossil fuels. As a carbon-neutral resource, biomass has the potential to significantly reduce greenhouse gas emissions while providing a reliable energy supply. The utilization of biomass not only contributes to energy security but also promotes rural development and waste management by converting agricultural and forestry residues into valuable energy products [1,2].

Among the various forms of biomass, biomass pellets have gained considerable attention due to their high energy density, uniformity, and ease of handling. The pellets are produced from a variety of feedstocks, including agricultural residues, forest residues, and other lignocellulosic materials. The densification process enhances the physical properties of biomass, rendering it more suitable for handling and combustion in energy generation systems [3]. Biomass pellets are widely used in automatic combustion systems worldwide to meet household and industrial heat demand. Due to their favorable fuel properties and low ash content, pellets produced mainly from forest products, known as wood pellets, are the most common type. However, in recent years, pellets have also been produced from various biomass resources with significant waste potential. Secondary sources such as agricultural residues and landscape residues usually have higher ash content and lower energy content [4]. As a result, they often cause inefficient combustion in pellet stoves and lead to higher flue gas emissions [5,6]. Because of these drawbacks, agricultural and landscape residues are unlikely to be true alternatives or competitors to wood pellets. Nevertheless, large amounts of agricultural and landscape residues are generated regionally worldwide. Therefore, another sustainable way to utilize these residues in pellet form is through the application of thermochemical conversion technologies.

Thermochemical conversion technologies, including pyrolysis, gasification, and combustion, play a crucial role in transforming biomass into usable energy forms [7]. Pyrolysis, in particular, is a thermochemical process that decomposes organic materials at elevated temperatures in the absence of oxygen, resulting in the production of bio-oil, syngas, and char [8]. This process not only facilitates the efficient conversion of biomass into energy but also allows for the recovery of valuable chemicals and materials, thereby enhancing the overall sustainability of biomass utilization [9]. However, the environmental trade-offs of pyrolysis processes—such as energy input demands, emissions, and life-cycle implications—must also be considered when evaluating their true sustainability [10]. The products obtained from biomass pyrolysis include bio-oil, syngas, and char, each of which has a distinct set of applications and benefits. Bio-oil, a liquid product, can be further refined into renewable fuels or used as a chemical feedstock [11]. The yield and quality of bio-oil can vary significantly depending on the nature of the feedstock and the conditions of the pyrolysis process. For example, Pilon and Lavoie [11] reported that the pyrolysis of switchgrass at low temperatures produced a range of liquid condensable products, resulting from the depolymerization and fragmentation of the biomass polymers. Char, the solid residue, can be utilized as a soil amendment or as a carbon sequestration method, while syngas can be converted into electricity or utilized as a precursor for synthetic fuels [12,13,14]. Beyond this, the production of supercapacitors used in lithium-ion batteries through biomass pyrolysis has become a focus in recent years. In this process, biochar obtained from pyrolysis is converted into activated carbon by physical, chemical, or physicochemical methods [15]. On the other hand, hydrogen, which is one of the components of syngas produced from biomass pyrolysis, is considered a promising energy carrier for next generation fuels [16]. At this stage, the utilization of agricultural residues and other biomass resources is of particular interest [17,18,19].

Biomass pyrolysis typically occurs in an inert atmosphere at temperatures between 300 and 600 °C. The thermal behavior and product distribution are strongly governed by the lignocellulosic composition. The three major constituents hemicellulose, cellulose, and lignin decompose over distinct temperature ranges, leading to different molecular breakdown pathways [8]. The extent of decomposition, which depends on process variables, influences the characteristics of solid and gaseous products as well as the overall conversion efficiency. Previous studies report that hemicellulose degrades at 190–350 °C, cellulose at 240–400 °C, while lignin decomposes over a much broader range starting above ~250 °C [20,21]. On average, biomass consists of approximately 10–25% lignin, 20–30% hemicellulose, and 40–50% cellulose [22]. The structural and chemical features of these polymers largely determine the diversity and quality of pyrolysis products. Fourier Transform Infrared Spectroscopy (FT-IR) is widely applied to identify functional groups in biomass, providing both qualitative and quantitative insights. Characteristic absorption peaks in the 400–4000 cm⁻¹ region allow the detection of structural changes between untreated and treated samples, thereby revealing their response to pyrolysis [23,24]. Specific peaks associated with lignocellulosic fractions and extractives enable predictions about the type and quality of products that may be obtained from pyrolysis.

Thermogravimetric analysis (TGA) is widely used to assess the pyrolytic characteristics of biomass. This technique records the weight loss of samples as temperature increases, thereby providing valuable information on their thermal stability and decomposition behavior [25,26]. Several key parameters derived from TGA, including the comprehensive pyrolysis index, pyrolysis stability, and devolatilization index, are crucial for evaluating the energy potential and conversion efficiency of biomass feedstocks. The comprehensive pyrolysis index reflects the overall potential for bio-oil and gas production, whereas pyrolysis stability indicates the resistance of a material to thermal degradation [20]. The devolatilization index, in turn, provides insights into the rate of volatile release, which is particularly important for optimizing pyrolysis processes [27,28]. Zhai et al., (2016) [28] further highlighted the role of TGA in elucidating the kinetic mechanisms of biomass pyrolysis, offering a better understanding of devolatilization under different atmospheric conditions. For lignocellulosic biomass, the TGA profile typically follows three major stages: removal of moisture and light volatiles (<120 °C), decomposition of hemicellulose (220–320 °C), decomposition of cellulose and partial lignin degradation (315–400 °C), and extensive degradation of lignin above 450 °C [20,29,30]. During the thermal decomposition of biomass via pyrolysis, a variety of compounds are generated in solid, liquid, and gaseous forms depending on the temperature range and the distribution of lignocellulosic constituents. Considering the decomposition temperature windows of the main components, the corresponding FTIR absorption bands, and the characteristic TGA peaks, the properties of solid, liquid, and gas products were compiled from experimental literature data and are summarized in Figure 1 [31,32,33,34,35,36,37].

Previous studies on biomass pellets often recommend the use of secondary resources, such as agricultural and landscape residues, as solid fuels alongside forest-based feedstocks. However, their high ash content, poor combustion performance, and associated emissions pose significant challenges for both users and the environment. Contrary to this prevailing view, we propose that pellets derived from secondary biomass should be regarded not primarily as direct fuels for pellet stoves, but rather as feedstocks for pyrolysis. Accordingly, this study investigates the pyrolytic characteristics of conventional wood pellets and regionally sourced secondary biomass pellets through thermogravimetric analysis under inert conditions. By examining their thermal behavior and deriving key pyrolysis indices, the research aims to provide deeper insights into biomass utilization as a renewable energy source and to support the advancement of more efficient thermochemical conversion technologies.

2. Materials and Methods

The biomass materials used in the study include yellow pine (WP), palm pruning residue (PP), reed canary grass (RD), and daphne branch (DP) pellets. The yellow pine pellets and daphne pellets were purchased from a commercial pellet producer, while the PP and RD pellets were produced in the laboratory of the Department of Agricultural Machinery and Technologies Engineering, Faculty of Agriculture, Akdeniz University, Antalya, Türkiye. Palm pruning residues and reeds were collected from university campus landscape area. Wood pellets are the most widely used type of pellet worldwide [38]. Daphne leaves are frequently utilized in the food [39] and medical [40] sectors, while its branches constitute a potential biomass source on a global scale [41]. Palm pruning residues are particularly abundant in Mediterranean climate conditions, where palm trees are commonly cultivated as landscape plants, and their regular annual maintenance generates a sustainable and substantial amount of pruning waste [42]. On the other hand, the reed plant is a fast-growing landscape and energy crop that naturally thrives in both urban and rural wetland areas [43]. The selected feedstocks can be regarded as sustainable biomass resources because they are derived from abundant forestry products and continuously generated agricultural and landscape residues.

The moisture content (MC), ash content (AC), and volatile matter (VM) of the pellets were analyzed in accordance with the methods set forth in ASTM E871-82, ASTM D1102-84 and ASTM E872-82, respectively [44,45,46]. The fixed carbon (FC) value was calculated using the difference method. The elemental content (CHNS) was determined using the Thermo Scientific Flash 2000 (ThermoFisher Scientific, Waltham, MA, USA) elemental analyzer. The higher heating value (HHV) of the pellets was calculated using Equation (1) developed by Ozyuguran et al. (2018) for comprehensive biomass materials [47]. The lower heating value (LHV) was determined using Equation (2) [48].

$$HH{V}_{db}=-4.9140+0.2611N+0.4114C+0.6114H+0.3888S+0.02097O$$

(1)

$$LHV=HHV-24.43[\left(MC+8.94H\right)/1000]$$

(2)

Thermogravimetric analysis was performed using a HITACHI STA 7300 thermal analyzer (Hitachi High-Tech Corporation, Tokyo, Japan). Prior to the analysis, each pellet was ground using a laboratory-type grinder to a particle size smaller than 100 microns. In the TGA analysis, the temperature was set between room temperature and 650 °C, with a heating rate of 10 °C min⁻¹, and the process was carried out in a nitrogen gas atmosphere. TG/DTG curves and critical points on the curves were determined using OriginPro 2025b software based on the obtained results.

The pyrolytic characteristics of the pellets, including the pyrolysis index ($CPI$), devolatilization index ($D_{dev}$), and pyrolysis stability index ($R_w$), were calculated in accordance with the methodology in Equation (3), Equation (4), and Equation (5), respectively [49,50].

$$CPI={\displaystyle \frac{\left(R_{p, max}\times R_{av}\right)\times m_{\infty }}{T_i\times T_{p, max}\times \mathsf{\Delta }{T}_{1/2}}}$$

(3)

$$D_{dev}={\displaystyle \frac{{\mathrm{R}}_{\mathrm{P},\max }}{{\mathrm{T}}_{\mathrm{i}}\times {\mathrm{T}}_{\mathrm{p},\max }\times \mathsf{\Delta }{\mathrm{T}}_{1/2}}}$$

(4)

$$R_w=8.5875\times {10}^7\times {\displaystyle \frac{-{\mathrm{R}}_{\mathrm{P},\max }}{{\mathrm{T}}_{\mathrm{i}}\times {\mathrm{T}}_{\mathrm{p},\max }}}$$

(5)

where ${\mathrm{T}}_{\mathrm{i}}$ is the temperature at which volatiles are first released (°C), $R_{p,max}$ is the mass loss rate peak (% min⁻¹), $T_{p,max}$ is peak temperature, $R_{P,max}$ is the maximum mass loss rate (% min⁻¹), $R_{av}$ is the average mass loss rate between $T_i$ and $T_f$ (% min⁻¹), $m_{\infty }$ mass loss during pyrolysis (%), $\mathsf{\Delta }{T}_{1/2}$ is the temperature range where $R/R_{P,max}$ equals 0.5, ${\mathrm{m}}_{\mathrm{f}}$ is weight loss (%), $CPI$ is the comprehensive pyrolysis index (%² min⁻² °C³) and $D_{dev}$ is the devolatilization index (% min⁻¹ °C³).

Fourier Transform Infrared Spectroscopy (FTIR) was employed to analyze the organic bonds and specific molecular structures of the pellets. Before analysis, the pellets were finely ground to a particle size below 0.5 mm. The measurements were carried out using an FTIR spectrometer (PerkinElmer, Spectrum 400 FT-IR/FT-NIR, Shelton, CT, USA) operating in the 650–4000 cm⁻¹ range. During the procedure, the powdered samples were placed on the instrument stage and firmly pressed against the ATR crystal accessory. Each spectrum was obtained by averaging forty scans, with the resolution adjusted to 4 cm⁻¹ in absorbance mode.

3. Results and Discussion

3.1. Pellet Fuel Properties

The ultimate and proximate properties of the pellets are presented in

Table 1. Pellet fuel and chemical properties.

Parameter	WP	PP	RD	DP
Moisture (%, wb 1)	8.32	6.34	6.40	6.90
Ash (%, db 1)	1.73	11.65	13.12	1.93
Volatile matter (%, db)	82.47	69.27	68.38	80.51
Fixed carbon (%, db)	15.80	19.08	18.50	17.56
C (%, db)	49.68	44.69	41.59	47.26
H (%, db)	6.14	5.59	5.49	5.88
N (%, db)	0.10	1.37	1.63	0.34
S (%, db)	0.00	0.35	0.12	0.01
O (%, db)	42.35	36.35	38.05	44.58
O/C	0.64	0.61	0.69	0.71
H/C	1.47	1.49	1.57	1.48
HHV (MJ kg−1)	20.19	18.15	16.82	19.15
LHV (MJ mkg−1)	18.65	16.77	15.47	17.70
Hemicellulose (%) 2	17.50	22.00	26.85	17.58
Cellulose (%) 2	46.90	44.00	51.53	30.84
Lignin (%) 2	27.30	28.00	11.91	22.31

¹ wb: wet basis, db: dry basis. ² [55,56,57,58].

. Wood pellets (WP) and daphne pellets (DP) stand out with notably low ash contents of 1.73% and 1.93%, respectively. On the other hand, palm pruning residue pellets (PP) and reed (RD) pellets have higher ash contents. In terms of pyrolytic performance, it has been reported that high ash content leads to higher residual mass (Hopa et al., 2019) [51] Furthermore, Mullen et al. (2014) [52] and Al-Rahbi and Williams, (2019) [53] noted that the presence of alkali metals in biomass ash can catalyze reactions that shift the product distribution from the desired bio-oil towards char and gases. In addition, Wei et al., (2019) [54] reported that the relationship between ash content and pyrolytic reactions is complex. Therefore, instead of focusing solely on the ash content, the presence and variety of inorganic elements in the ash should also be considered in the pyrolytic process.

In proximate analysis, volatile matter (VM) is the fraction of organics released as gases and condensable vapors during inert heating; it excludes moisture and ash. VM strongly influences pyrolysis behavior by governing the overall amount of volatiles (and thus the char left at a fixed endpoint), while the gas–liquid split is further shaped by mineral ash (AAEM) and temperature. From Table 1, the VM ranking is WP > DP >> PP ≈ RD (82.47%, 80.51%, 69.27%, 68.38%), indicating that WP and DP are more prone to devolatilization and are expected to yield more total volatiles (and less solid residue) than PP and RD at comparable conditions. This general trend is consistent with prior observations that higher volatile content favors volatile release during pyrolysis [59], with the caveat that ash-catalyzed cracking and operating temperature will modulate the proportions of gases vs. condensables. This devolatilization tendency is not governed by volatile matter alone; it is significantly modulated by the ash content and, more specifically, the presence of catalytically active alkali and alkaline earth metals (AAEMs) within the ash fraction. High-ash pellets (e.g., PP and RD) typically contain greater quantities of AAEM species such as K, Na, Ca, and Mg, which are known to catalyze thermal cracking and shift the pyrolysis pathway toward increased gas formation and residual char. This catalytic behavior of AAEMs has been well-documented in previous studies [60,61,62,63,64], and is more pronounced in biomass types with elevated ash content.

3.2. Pyrolytic Characterization of Pellets

The analysis of thermogravimetric (TG) and derivative thermogravimetric (DTG) curves is crucial in evaluating the pyrolytic performance of biomass. These curves provide detailed insights into the thermal decomposition behavior of biomass, allowing researchers to gain a comprehensive understanding of the kinetics, product yields, and overall efficiency of the pyrolysis process. The TG/DTG curves and pyrolytic behaviors of the pellets under investigation in the study were presented in Figure 2.

The thermal degradation behaviors of the main lignocellulosic fractions (Table 1) overlap distinctively in the TG/DTG analysis in Figure 2. In agreement with the literature, hemicellulose decomposes first (200–315 °C), cellulose follows (315–400 °C, usually with a sharp DTG maximum), and lignin degrades slowly over a broad range (200–600 °C), producing the high-temperature shoulder. For the present pellets, the hemicellulose response is best captured by the DTG activity between 200–300 °C. The reed pellet (RD) gives the strongest peak in this window; palm pruning pellet (PP) is evident but weaker than RD, while wood pellet (WP) and daphne pellet (DP) show comparatively modest peaks. This order agrees with the Table 1 hemicellulose contents (RD 26.85% > PP 22.00% > DP 17.58% ≈ WP 17.50%).

The lignin contribution is reflected by the shoulder extending into 400–600 °C. Here PP and WP display the most pronounced shoulders, DP is moderate, and RD the weakest fully consistent with their lignin fractions (PP 28.00% ≈ WP 27.30% > DP 22.31% > RD 11.91%). The strong shoulder of WP also matches its high carbon content and highest HHV (C 49.68%; HHV 20.19 MJ kg⁻¹). For PP, the pronounced shoulder arises from its high lignin fraction despite its lower C and HHV (C 44.69%; HHV 18.15 MJ kg⁻¹), which are depressed by the high ash content.

The intensity of the cellulose peak (315–400 °C) tracks composition but is clearly modulated by ash catalysis [65]. Although RD has the highest cellulose fraction (51.53%), its peak appears at lower temperature and is broader, consistent with its very high ash (13.12%) and that of PP (11.65%), whose alkali minerals lower devolatilization temperatures and flatten DTG maxima [66]. In contrast, WP (1.73% ash) and DP (1.93%) present sharper, higher-temperature peaks, even though DP contains the least cellulose (30.84%); low mineral content reduces catalytic effects and concentrates the main devolatilization event.

Mass balance trends in the TG traces are also coherent with the proximate/elemental data. Samples richer in volatile matter (WP 82.47% and DP 80.51%) exhibit a more intense Stage 2 mass loss band and leave less char, whereas PP 69.27% and RD 68.38%, combined with their high ash (11.65% and 13.12%), retain more residue at the end of heating. Elemental ratios of pellets are consistent as well, a lower O/C generally supports a higher HHV; WP (O/C 0.64) shows a higher HHV than RD (0.69) and DP (0.71), while PP, despite the lowest O/C (0.61), has a lower HHV because ash dilutes the energy density and enhances catalytic devolatilization.

The comprehensive pyrolysis index ($CPI$), devolatilization index ($D_{dev}$), and pyrolysis stability index ($R_w$) are great significance in the pyrolysis of biomass, as these indices give invaluable insights into the efficiency, product yield, and stability of the pyrolysis process. Each index serves a unique purpose in evaluating the pyrolysis behavior of biomass, allowing for optimized process conditions and improved product quality. The data regarding the pyrolytic characterization of the pellets are presented in

Table 2. Pyrolytic characterization of pellets.

Parameter 1	Unit	WP	PP	RD	DP
$T_i$	(°C)	190	163	128	165
$T_f$	(°C)	427	407	473	486
$m_{\infty }$	(%)	75.74	65.4	66.81	76.06
$T_{p, max}$	(°C)	365	334	310	352
$R_{av}$	(% min−1)	55.9	56.8	61.01	46.19
$\mathsf{\Delta }{T}_{1/2}$	(°C)	59	90	59	88
$CPI$	(10−3%3 °C−3 min−2)	10.9	5.4	10.8	6.2
$D_{dev}$	(103% °C min−3)	1.2	1.4	0.9	1.4
$R_w$	(109% min−1 °C−2)	1.7	1.3	1.3	1.4

¹ T_i: Initial devolatilization temperature; T_f: Final devolatilization temperature; m_∞: Mass loss during pyrolysis; T_pmax: Peak temperature; R_av: Average mass loss rate between T_i and T_f; ΔT_1/2: Temperature range where R/R_pmax equals 0.5 (°C); CPI: Comprehensive pyrolysis index; D_dev: Devolatilization index; R_w: Pyrolysis stability index.

.

The $CPI$ is a comparative indicator of pyrolytic performance and reflects how strong and compact the main devolatilization event is when rate and temperature are considered together [67]. In our dataset, the order WP (10.9) ≈ RD (10.8) > DP (6.2) > PP (5.4) aligns with Table 1 and Figure 2. WP attains the top $CPI$ because its very low ash (1.73%) enables a sharp, high-temperature cellulose peak ($T_{p,max}$= 365 °C) with a narrow event ($\mathsf{\Delta }{T}_{1/2}$ = 59 °C) and a high average rate ($R_{av}$= 55.9% min⁻¹). Although RD has lower volatile matter (68.38%), its high cellulose fraction (51.53%) and sustained average rate ($R_{av}$ = 61.01% min⁻¹) keep $CPI$ high. Abundant AAEM minerals (alkali/alkaline-earth; ash = 13.12%) shift $T_i$ and $T_{p, max}$ to lower values (128–310 °C) but do not prevent a concentrated main event [51,68]. By contrast, DP and PP show lower $CPI$ because their main devolatilization spans are broader ($\mathsf{\Delta }{T}_{1/2}$ = 88 and 90 °C); for PP, high ash (11.65%) disperses and catalytically overlaps reactions, while DP despite high volatile matter (80.51%) exhibits a lower average rate ($R_{av}$ = 46.19% min⁻¹), which reduces $CPI$. The $D_{dev}$ ranking PP = DP (1.4) > WP (1.2) > RD (0.9) indicates a stronger per-degree volatile release for PP and DP arising primarily from AAEM catalyzed reactions and low $T_i$ in PP (163 °C) and from the high volatile/low ash profile of DP (80.51%/1.93%). Finally, the R_w sequence WP 1.7 > DP 1.4 > PP = RD 1.3 denotes the most thermally stable course for WP, moderate stability for DP, and relatively lower stability for PP and RD owing to mineral induced overlap. The literature also shows that $CPI$ and thermal stability tend to increase together [50]. Our results follow this trend for WP and DP, while RD is a partial exception because ash catalysis lowers peak temperatures yet its high cellulose keeps $CPI$ elevated [54,69].

As previously discussed in detail, the pyrolytic behavior of the pellets was related to their ultimate/proximate composition, lignocellulosic fractions, and TG/DTG derived parameters. The fuel and chemical properties of the pellets used in this study, along with their relationships to pyrolytic parameters were computed pairwise Pearson correlation coefficients and summarized the results in the correlation matrix (Figure 3). Given the small sample size (n = 4), we report the coefficients as descriptive effect sizes without formal significance testing and, where relevant, provide Spearman correlations as a robustness check.

Based on the correlation coefficients, the strong correlations (|r| ≥ 0.70) in Figure 3 agree with the compositions in Table 1, the indices in Table 2, and the TG/DTG critical points and peak positions in Figure 2. Higher volatile matter (VM) aligns with greater total mass loss (m_∞) and a stronger Stage-2 loss (m_s2), which matches a stronger main DTG peak. Cellulose content correlates positively with the peak temperature (T_pmax), the average mass loss rate (R_av), and CPI, indicating a sharper, higher temperature main decomposition when mineral interference is low. Lignin correlates with late stage behavior, showing positive relations with the final devolatilization temperature (T_f) and the Stage-3 loss (m_s3). The O/C ratio also tracks late devolatilization and correlates with higher T_f and larger m_s3, because more oxygenated matrices sustain reactions at higher temperatures.

Negative strong correlations are mainly driven by mineral catalysis and early decomposition [70]. Ash content (AC) correlates negatively with T_i, T_pmax, and m_∞, indicating AAEM-induced earlier onset, lower peak temperatures, and more solid residue. Hemicellulose correlates negatively with T_i and T_h (temperature at hemicellulose shoulder maximum) and often with T_pmax, since it decomposes first and pulls the thermal response to lower temperatures. These relations explain the trends in Table 2: CPI rises with higher T_pmax, higher R_av, and larger m_∞ (low ash, cellulose); D_dev increases when the main release is triggered early and remains compact (AAEM assisted or high VM/low ash); and R_w reflects the stability of the thermal course, decreasing when mineral-induced overlap is strong [70,71]. The links between the pellets’ fuel properties, lignocellulosic content, and key TGA parameters are consistent. Future studies with larger sample sizes can develop and validate stronger equations for pyrolytic characterization.

3.3. Evaluation of the FTIR Spectral Bands of the Pellets

All four pellet spectra display the typical lignocellulosic pattern (Figure 4). A broad O-H stretch, aliphatic C–H stretches, a carbonyl/aromatic window, and a rich fingerprint region of C-O-C and C-OH vibrations. These bands are consistent with the proximate/ultimate data in Table 1 and the TG/DTG behavior in Figure 2, i.e., carbohydrates dominate the main devolatilization event while lignin feeds the high temperature shoulder.

In the single bond region (≈3700–2800 cm⁻¹), a broad O–H envelope at ~3400–3200 cm⁻¹ appears in all samples due to hydrogen bonded hydroxyls in cellulose/hemicellulose; it is more pronounced in RD and PP, and slightly lower in WP and DP, matching their compositions in Table 1. The C-H stretches at ~2930–2850 cm⁻¹ are clearer for WP and PP, consistent with their higher lignin/extractive-like aliphatics [51,72].

In the double-bond region (≈1800–1500 cm⁻¹), the carbonyl band near 1730 cm⁻¹ (acetyl/uronic C=O of hemicellulose) is strongest for RD, moderate for PP, and weaker for WP and DP, mirroring the hemicellulose order (RD > PP > WP ≈ DP) [73]. Aromatic skeletal bands at ~1600–1510 cm⁻¹ (guaiacyl/syringyl lignin units) are most pronounced in PP ≈ WP, moderate in DP, and weakest in RD, in line with lignin contents (PP 28.00% ≈ WP 27.30% > DP 22.31% > RD 11.91%) [74].

In the fingerprint region (≈1500–750 cm⁻¹), polysaccharide bands at ~1155–1030 cm⁻¹ (C-O-C/C-OH) and the cellulose marker at ~896 cm⁻¹ (β-glycosidic C-H) are stronger in RD and WP, consistent with their higher cellulose fractions (51.53% and 46.90%) [75,76]. Lignin related features around ~1260–1215 cm⁻¹ are clearer in PP and WP, while DP shows a more carbohydrate-leaning pattern aligned with its lower cellulose (30.84%) and high volatile matter [76,77].

The strong 1730 cm⁻¹ in RD and its high ash suggest early deacetylation and decarboxylation, thus more CO₂/acetic acid in the vapors and suppression of anhydrosugars [76,78]. PP and WP, with pronounced aromatic bands, are expected to give more aromatic/phenolic bio-oil fractions and char with higher aromaticity. WP’s very low ash supports a cleaner, low-ash char, whereas PP’s higher ash yields a more ash-rich char and catalyzes gas formation. DP, with high VM and low ash, should favor higher condensable yields and a moderate-aromatic char.

3.4. Integrated Product Distribution from Composition, TG/DTG and FTIR

Table 3. Product distribution and quality expected for each pellet from integrated composition, TG/DTG and FTIR data.

Pellet	Key Temperatures (°C)	1 Liquid-max range (°C)	Char @600 °C (100–m∞, wt%)	2 Expected Liquid Families (Bio-Oil)	3 Gas Tendency (Qual.)	4 Char Quality (Qual.)
WP	Ti: 190	335–395	24.3	Oxygenates (strong anhydrosugar/levoglucosan path) + phenolics (G/S)	Low–mid	Clean, low ash, higher aromaticity with high temp.
	Tpmax: 365
	Tf: 427
	ΔT1/2: 59
PP	Ti: 163	289–379	34.6	Phenolic rich (guaiacol/syringol); anhydrosugars suppressed; some light acids	Mid–high	Ash rich, aromatic; catalytically hardened
	Tpmax: 334
	Tf: 407
	ΔT1/2: 90
RD	Ti: 128	280–340	33.2	Light oxygenates (acetic acid, furfural); anhydrosugars suppressed	Mid–high (CO2/CO, some H2)	Ash rich, moderate aromaticity
	Tpmax: 310
	Tf: 473
	ΔT1/2: 59
DP	Ti: 165	308–396	23.9	Oxygenates (less anhydrosugar than WP) + some phenolics	Low–mid	Cleaner than PP and RD, moderate aromaticity
	Tpmax: 352
	Tf: 486
	ΔT1/2: 88

¹ Liquid-max range: Estimated as T_pmax ± ½ · ΔT_1/2 (Table 2); main devolatilization band before extensive secondary cracking. ² Expected liquid families: Inferred from FTIR + composition + TG/DTG: cellulose → anhydrosugars; hemicellulose/acetyls → light oxygenates/acetic acid/CO₂; lignin (G/S) → phenolics. ³ Gas tendency: Ranked from T_i, T_pmax, ΔT_1/2, m_s3 and ash (AAEM): lower Tᵢ, depressed T_pmax. Higher ash strong 1730 cm⁻¹ ⇒ more gas; opposite ⇒ less gas. ⁴ Char quality: Judged from ash, T_f and m_s3 (high-T persistence), lignin fraction, and indices (R_w, CPI): low ash ⇒ cleaner char; higher lignin/persistence ⇒ more aromatic; high ash ⇒ ash-rich/catalytically hardened.

integrates composition (Table 1), TG/DTG results (Table 2; Figure 2), and FTIR spectra (Figure 4) to estimate, for each pellet, the condensable favored temperature range, the likely bio-oil families, and the relative gas/char tendencies.

Across all pellets, Stage2 (≈200–400 °C) is the main devolatilization zone where liquids form, while Stage3 (≈400–600 °C) increases gas formation and makes the char more aromatic (Figure 2). Low-ash, cellulose-rich WP and DP show sharp, higher-temperature peaks (T_pmax 365/352 °C with narrow–moderate ΔT_1/2), so liquid yield is favoured around their condensable-favoured ranges (WP ≈ 335–395 °C; DP ≈ 308–396 °C). Their FTIR spectra are dominated by carbohydrate bands with moderate lignin signals, suggesting bio-oils richer in oxygenates with some phenolics and cleaner, low-ash chars [72,79]. Mineral-rich PP and RD start earlier (low T_i) and show AAEM-catalysed peak depression/overlap (T_p,max 334/310 °C). Together with the stronger 1730 cm⁻¹ band in RD (acetyl C=O), this points to more CO₂/CO (and some H₂), more light acids (e.g., acetic acid), lower liquid yields, and ash-richer chars; their condensable-favoured ranges are lower (PP ≈ 289–379 °C; RD ≈ 280–340 °C).

Final solid yield follows the total mass loss (m_∞) from TGA: WP 75.74% and DP 76.06% leave ~24 wt% char, while PP 65.4% and RD 66.81% leave ~34.6–33.2 wt% char. Thus, WP and DP favour more condensables and less residue, whereas PP and RD retain more solid due to minerals. In all cases, operating near Tₚ,max ± ½ΔT_1/2 maximises condensables; temperature 450–500 °C shifts selectivity toward cracking and gases (CO₂, CO, H₂, some CH₄) and produces harder, more aromatic char.

3.5. Contribution of This Research to Sustainability

This study proposes pyrolysis as an alternative and sustainable conversion route for both woody and non-woody biomass pellets. Through TG/DTG and FTIR analyses, it demonstrates that such feedstocks, under optimized pyrolytic conditions, can yield solid (biochar), liquid (bio-oil precursors), and gaseous (syngas) products. Especially for high-ash agropellets, which may pose higher emission risks if directly combusted, routing them to pyrolysis yields value-adding streams: biochar, which retains carbon and can be used for soil amendment and carbon sequestration; bio-oil precursors, which can be upgraded to renewable fuels or chemical intermediates; and syngas (CO, CO₂, H₂), which can support clean energy generation.

From a broader sustainability perspective, transitioning agropelets from inefficient combustion to pyrolytic systems enables the generation of value-added products, supports circular bioeconomy principles, strengthens local energy systems, and promotes better resource efficiency. This approach also aligns with the United Nations Sustainable Development Goals (SDGs)—specifically SDG 7 (Affordable and Clean Energy), SDG 12 (Responsible Consumption and Production), and SDG 13 (Climate Action)—by enabling the production of renewable energy carriers, supporting waste valorization, and fostering carbon-retentive outputs [80]. Ultimately, the study calls for rethinking biomass pellet use—not as limited to combustion, but as part of a more integrated, versatile, and environmentally responsible bioenergy strategy.

4. Conclusions

Combustion is not the only route to evaluate biomass pellets. Pyrolysis is a sustainable option, especially for high-ash pellets that tend to give low efficiency and high emissions in combustion. Here, we combined composition with TG/DTG features, simple reactivity–stability indices, FTIR markers, and pairwise correlations to predict how pellets respond during pyrolysis and what product qualities are expected. Lignocellulosic composition (cellulose–hemicellulose–lignin) and ash content directly shape pyrolysis performance and product quality.

Feedstocks that are low in ash and richer in cellulose usually show a later, sharper main devolatilization step. This concentrates vapors, favors the formation of condensables (bio-oil precursors), and yields cleaner, low-ash chars. By contrast, biomass with higher mineral ash (AAEM) content tends to start devolatilizing earlier and shows broader, lower peaks due to catalytic effects, which shifts products toward gases and leaves more ash-rich, mineralized chars.

The mechanism matches composition: cellulose increases the intensity and speed of the main event; lignin sustains high-temperature mass loss and promotes aromatic char; hemicellulose drives early deacetylation and light acids. FTIR is consistent with TG/DTG and supports product predictions.

Pyrolysis indices compactly summarize behavior and align with the trends, and the correlation analysis links feedstock chemistry to thermal response: higher volatile matter aligns with total mass loss, higher ash aligns with earlier onset and lower peak temperature, higher cellulose aligns with a stronger and faster main event, and lignin and higher O/C align with late-stage loss. Despite the limited number of samples, the correlation coefficients obtained were notably high, indicating consistent relationships between chemical composition and thermal behavior. These results suggest that even small variations in feedstock chemistry can systematically influence pyrolysis dynamics. Future research should expand the dataset to include a larger number and broader diversity of biomass feedstocks. This will allow the development of more statistically robust models and provide stronger validation of the relationships observed in this study.

The wide variety of biomass sources that can be used as pellet feedstock poses a challenge in achieving feedstock consistency. In this regard, lignocellulosic composition, ash content, and especially the mineral composition of ash can introduce multiple uncertainties in determining pyrolysis behavior. Such variability is further influenced by external agricultural factors, such as exposure to fertilizers and pesticides, which can significantly alter the chemical and mineral characteristics of biomass. In this context, we recognize the importance of kinetic investigations. Future research will be directed toward multi-rate TGA-based kinetic modeling, which will allow a more comprehensive understanding of the pyrolytic behavior of diverse biomass pellets and strengthen the sustainable scaling-up of such processes to industrial applications.

While the TGA results obtained from ground samples offer valuable insights into thermal behavior, further investigations using actual pellet or chip forms in bench/lab scale systems are needed to better reflect realistic pyrolysis conditions. Moreover, this study did not include direct product yield measurements (bio-oil, gas, char) from bench-scale pyrolysis experiments. Incorporating such yield analyses in future work should incorporate such yield analyses to validate and refine the predictions derived from TG/DTG and FTIR data. Furthermore, while the sustainability potential of pyrolysis is well recognized, a comprehensive environmental assessment requires consideration of associated trade-offs, including energy input demands, emission profiles, and life-cycle impacts. Although these aspects were beyond the scope of the present study, future research will benefit from integrating life-cycle assessment (LCA) approaches to evaluate the environmental performance of pyrolysis processes more holistically. This is particularly important for scaling up biomass conversion technologies in alignment with climate and resource efficiency goals.