Sawdust (emanating from the processing of different tree species) and discarded polyethylene terephthalate (PET) plastic bottles are the primary materials utilized in this study for briquette production. The PET bottles were collected from waste receptacles in Benin City, Edo State, Nigeria. The tops and labels of the bottles were removed to ensure uniformity before heating them at 230°C to facilitate melting. The molten PET was held at that temperature for 5 minutes before being allowed to cool for 10 minutes, and thereafter pulverized into powder using a locally fabricated pulverizer. The powder was sieved using a lab sieve with a 50 μm mesh to obtain a uniform grain size (see Figure 1a). The sawdust was obtained from the sawmill's dumpsite in Benin City, Nigeria. It was sieved using a lab sieve with a 50 μm mesh to obtain a uniform particle size (see Figure 1b).

The PET powder (PP) and sieved sawdust (SD) were mixed at four mass ratios as illustrated in Table 1 to yield four types of briquettes: BR-A, BR-B, BR-C, and BR-D. No binder was used in this study.

One hundred grams of each blend was carefully placed in a plastic bucket and thoroughly stirred by hand with a wooden spatula until a uniform, consistent blend was achieved. The homogeneous mixture was transferred into a high-pressure hydraulic press briquetting machine (~17 MPa), where it was compressed to form compact, densified briquettes, as shown in Figure 2. The briquetting machine produces 50 briquettes per hour. The briquettes, which are 4.5 cm in diameter, have heights ranging from 1.5 cm to 3.7 cm.

The carbon (C), hydrogen (H), nitrogen (N), and sulphur (S) contents of both the raw materials and briquettes were determined using an elemental analyser (Exeter, EA-1110) following ASTM D5373-02 standards, while the oxygen content was calculated using Equation (1).

The proximate properties of the biomass samples were determined using a thermogravimetric analyzer (NETZSCH-STA449F5) with a sample amount of 12 ± 0.5 mg. following the ASTM E1131-08 standard. The analyser was run until constant weights were recorded. This residual weight is the ash content. Percentage ash content was calculated using Equation (2).

An oxygen bomb calorimeter (CAL3K-A) was used to measure the samples’ higher heating values (HHV). Each experiment was repeated three times.

The ash composition of the briquettes was analyzed using an energy-dispersive X-ray fluorescence (XRF) spectroscopy (Rigaku RIX-3000, Japan). The concentrations of heavy metals in the briquettes were determined using an atomic absorption spectrometer (GBC, Avanta-PM) and compared with European standards. Table 2 shows the ash composition of the briquettes.

To evaluate the structural integrity of the briquettes, they were lifted to a height of 2 meters and then dropped onto a sturdy concrete floor to simulate a standardized, controlled method to assess their mechanical strength and durability under impact conditions similar to those encountered in real-world handling stresses, such as transportation, handling, and storage. The mechanical strength was calculated by dividing the briquette’s weight after dropping by its weight before dropping. Additionally, the briquettes’ particle density was determined using standard methods based on the mass-to-volume ratio. Each characterization experiment was performed three times.

The morphologies of the briquettes were analyzed using Scanning electron microscopy (SEM) microphotographs, obtained with a Thermo Scientific Phenom ProX scanning electron microscope, PhenomWorld, Eindhoven, Netherlands. The briquette samples were placed on the aluminum holder stub using sticky carbon tape, then vacuum-dried and scanned at an acceleration voltage of 15 kV. The surface functional groups of the samples were analyzed by Fourier transform infrared (FTIR) spectroscopy on a ThermoFisher Scientific iS50 FTIR spectrometer, which scans from 4000 to 400 cm^-1 at a resolution of 4 cm^-1 and a scanning speed of 2 mm/sec. The briquettes were ground into fine powders before the FTIR analysis.

The combustion properties of the produced briquettes were examined using a thermogravimetric analyzer (NETZSCH-STA449F5). The analyzer operates from room temperature to 1600°C and can operate with nitrogen, oxygen, or air. The briquettes were ground and sieved through a laboratory test sieve to particle sizes ranging from 50 to 70 μm. About 10 ± 0.5 mg of each briquette type was weighed, placed into a platinum crucible and inserted into the thermogravimetric analyser. The samples were heated from ambient temperature to 900°C at 10°C/min in an oxygen atmosphere at 100 mL/min. The resulting data were used to generate TG and DTG curves for all samples.

Several key combustion parameters, including the ignition temperature (T_i), peak temperature (T_p), and burnout temperature (T_b), were extracted from the TG and DTG curves to evaluate the overall combustion performance of the briquettes. T_i is typically determined using the tangent method on the TG-DTG curves [23]. Furthermore, three combustion indices, namely, the ignition index (D_i), burnout index (D_b), and combustion intensity index (H_f), were evaluated during the combustion analysis. These indices were calculated using Equations (3)-(5) [23,24].

where DTG_max = highest weight loss rate (wt.%/°C), t_i = time taken to ignite (min), t_b = time taken to burnout (min), ∆T_1/2 = temperature range at half of DTG_max value (°C), ∆t_1/2 = time range at half of DTG_max value (min), t_p = time to reach peak temperature (min).

Kinetic analysis is very important in the study of the combustion process of solid fuels, including briquettes. In this study, the Coats-Redfern method was used to estimate kinetic parameters from TGA data. It was selected for its strength, ease, and efficiency in non-isothermal conditions characteristic of biomass thermal degradation [25]. It is especially appropriate for analyzing data obtained at a single heating rate, which aligns with our experimental design. The reaction rate constant (k) is dependent on absolute temperature (T), which is expressed by the Arrhenius equation [26]:

T indicates the temperature (K), A signifies the pre-exponential factor (min^-1), E_a indicates the activation energy (kJ/mol), and R is the universal gas constant (0.008314 kJ/mol K). The solid-gas reaction rate is defined as follows [26]:

dxdt represents the rate of degradation, which is a linear function of the temperature rate constant. x represents the briquettes’ weight loss ratio during the combustion, t signifies the time for the combustion, and g(x) is the differential form of the reaction model, which depends on the reaction mechanism. The weight loss ratio (x) can be obtained from thermogravimetric analysis data, as follows [27]:

m₀, m_t, and m_f are the initial, instantaneous, and residual mass of samples during the combustion process, respectively.

Combining Equations (6) and (7) at the heating rate (β = dT/dt) and rearranging gives Equation (9):

The integral expression of g(x) is presented in Equation (10).

A = pre-exponential factor (min^-1), E_a = activation energy (kJ/mol), k(T) = reaction rate constant, R = universal gas constant, T = reaction temperature (K), x = conversion rate, and g(x) = kinetic model, f(x) = the integral form of the kinetic model.

Kinetic parameters, such as the activation energy, can be determined by integrating Equation (10) and approximating the exponential term. For the Coats-Redfern method, after approximation and rearrangement, the solution is expressed as Equation (11) [27]:

In this study, the briquettes’ combustion process was approximated as a first-order kinetic reaction; hence, according to [27],

Substituting in Equation (10) and integrating, Equation (11) translates to:

Straight lines were fitted from the plots of ln⁡[−ln⁡(1−x)T2] against 1T, and the activation energy (E_a) was estimated from the slope (−EaR) of the plots.

Version 4.2.3 of the Water Boiling Test (WBT) protocol was used to ascertain the thermal performance of the developed briquettes practically. An upgraded cookstove was used, and the performance evaluation involved boiling 5 litres of water in a flat-bottomed, cylindrical aluminium cooking pot with a maximum capacity of 7 litres. The tests were carried out using the briquette samples, like typical household cooking methods. Specific operational parameters, such as thermal efficiency, time to boil, specific fuel consumption, burning rate, and firepower, were evaluated for all three WBT phases: cold-start high-power phase, hot-start high-power phase, and simmer low-power phase [28]. During high-power (cold start) conditions, 5 litres of water were heated to its boiling point in the 7-litre pot using the cookstove at room temperature. For the high-power (hot start) phase, 5 litres of water were heated to boiling point temperature in a 7-litre pot with the cookstove pre-heated, and during the simmer phase, the water temperature was kept at about 3°C below boiling point for 45 minutes [28]. Kerosene was used for igniting the briquettes during the test.

The energetic densities (energy per unit volume) of the developed briquettes were determined by multiplying their particle density (PD) with their higher heating value (HHV), as shown in Equation (14) [2].

where ED is the energetic density (MJ/m³), PD is the particle density (kg/m³), and HHV is the higher heating value (MJ/kg).

The Fuel Value Index (FVI) of the developed briquettes was determined using Equation (15) [2].

where FVI is fuel value index (MJ/m³%), and Ash is the Ash content of the briquettes (%).

Table 3 presents the physicochemical properties of both the raw materials and the briquettes (BR-A to BR-D) produced in this study. PET plastic is known to contain a high proportion of volatile matter (VM) due to its manufacturing process and chemical composition [29,30]. Consequently, the VM content of PET waste was substantially higher than that of sawdust, thereby directly impacting the overall VM levels in the briquettes. Specifically, the VM percentage increased with increasing plastic content, rising from approximately 83.19% in BR-B to 85.19% in BR-D, as indicated in Table 3. These observations are consistent with findings from previous studies [3,17].

FC was observed to be the highest in BR-A, probably due to the high amount of FC in SD. It was, However, it was observed to increase from 6.84% in BR-B to 8.05% in BR-D as the PET content increased. All samples had lower FC than BR-A, probably because the raw sawdust contained higher FC. Table 3 also summarises the elemental composition analysis. The carbon content and heating value of PET waste were higher than those of the sawdust, which was probably attributable to the chemical composition and structure of plastics [14]. This had a significant impact on the carbon content and HHV of the plastic-containing briquettes, as indicated in Table 3. The ratio of plastic in the briquette blends was positively correlated with the carbon content and heating value in the samples BR-B to BR-D, with carbon content increasing from 48.55 to 57.87% and heating value rising from 20 to 23.65 MJ/kg. These findings are consistent with previous studies [5,17,31]. The hydrogen and oxygen content of the briquettes was comparable among samples, as indicated in Table 3. No significant amounts of nitrogen were detected, and no sulphur was detected at all, which means that these briquettes are not expected to generate significant amounts of nitrogen oxides (NO_x) during the combustion or co-combustion process [5,32].

As indicated in Table 3, the atomic ratios of the hydrogen to carbon (H/C) and oxygen to carbon (O/C) in the briquettes were between 0.086 and 0.118 and 0.828 to 0.596, respectively. These ratios were plotted on the Van Krevelen diagram (Figure 3), which graphically illustrated the evolution of the composition and fuel properties of the briquettes [33]. Reduced ratios usually indicate improved fuel quality, as reflected by the proximity to the origin of the diagram [33,34]. The briquette produced from sawdust alone (BR-A) was initially located in the upper right-hand corner, as is typical of biomass. Upon mixing with PET plastics, the samples moved toward the origin, indicating greater levels of coalification, higher energy density, and better solid-fuel prospects [35]. The diagram also indicates that the rise in plastic content shifts briquettes from the lignite area (BR-B) to the bituminous coal area (BR-C and BR-D). Comprehensively, the addition of PET plastic increases the coalification and consequently the quality of the fuel of the briquettes.

From Table 3, the O/C ratios of the briquettes decreased from 0.828 in BR-A to 0.596 in BR-D, indicating significant de-oxygenation, which leads to an increase in higher energy density [4,36]. Hence, BR-D shows the highest HHV. The lower O/C ratio in BR-D (0.596) indicates it is the most hydrophobic of the briquettes [37,38]. It will be the least likely to absorb moisture from the air, making it the most stable for long-term storage in humid environments without crumbling or losing quality. The H/C ratios dropped from 0.11 to 0.08 from BR-A to BR-D, as seen in Table 3. These remarkably low ratios suggest that the briquettes will burn with very little flame and minimal smoke. They will provide a steady, glowing heat rather than a rapid, volatile-driven fire [38,39].

Table 4 presents the drop strengths and densities of the briquettes produced in this study. The data reveal that the drop strength and density of the briquettes increase with increasing plastic content. Specifically, the drop strength increases from 77.8% in BR-A (sawdust only) to 82.9% in BR-D (highest plastic content), implying improved mechanical durability and resistance to breakage during handling and transportation. The bulk density also shows a significant upward trend, increasing from 648.12 kg/m³ in BR-A to 715.83 kg/m³ in BR-D, which indicates enhanced compactness and energy density of the briquettes. Yiga et al. [2] observed a rise in density and mechanical strength for briquettes produced from carbonised rice husk.

Table 5 presents the concentrations of heavy metals in the produced briquettes. The results confirm that the concentrations of heavy metals in the briquettes comply with the standards set by the European Pellet Council (EPC) [40] for densified solid fuels. However, the cadmium content in BR-D was slightly above the set limit.

Figure 4 shows the FTIR spectra of the developed briquettes.

The FTIR spectra of the briquettes provide a detailed view of their chemical composition by showing characteristic absorbance bands associated with specific functional groups and molecular structures. For all samples, there’s a broad band near 3400 cm^-1, which comes from O-H stretching vibrations, mainly due to moisture or hydroxyl groups present in the biomass. Interestingly, as the PET content increases, this band decreases in intensity, possibly indicating fewer hydroxyl groups in PET than in the biomass. In the 2900-2800 cm^-1 range, signals related to aliphatic C-H stretching vibrations are observed, which are common in components like cellulose, hemicellulose, lignin, and the hydrocarbon chains in PET. These signals remain fairly consistent across all spectra, with minor variations that suggest changes in aliphatic content as PET content increases. A key feature appears around 1700 cm^-1 in the spectra of briquettes with higher PET levels (BR-B, BR-C, BR-D). This peak corresponds to ester carbonyl groups (C=O), which are a hallmark of PET. Its increasing intensity with higher PET ratios confirms the presence of ester functionalities typical of PET plastics.

Furthermore, in spectra with more PET, especially those with higher ratios, there are noticeable peaks in the 1600-1500 cm^-1 range. These are associated with aromatic C=C stretching vibrations, reflecting the aromatic rings within PET molecules. This peak is less prominent in the pure sawdust sample (BR-A), which mainly contains lignin and other biomass components. Finally, the bands between 1200 and 1000 cm^-1 are linked to C-O stretching vibrations in esters, ethers, and carbohydrates. PET’s ester groups add or strengthen these peaks, highlighting its chemical influence on the briquettes. According to Carnaje et al. [41], the detection of both aliphatic and aromatic hydrocarbons suggests the briquettes contain fats and oils, which can help improve their ignition properties.

Figure 5 presents SEM micrographs showing the morphology and structure of the produced briquettes. The SEM micrographs reveal notable morphological differences between the briquettes, which are probably attributable to the increasing content of PET plastics, acting as an effective binder and densifying agent.

The micrograph (a) shows that the briquette made entirely of sawdust (BR-A) has a very porous and loose structure. The sawdust particles are clearly visible, irregular in shape, and many large gaps seperate them. This suggests that, without any binding agent, the particles don’t stick together very well.

From micrographs (b) to (d), it can be observed that the structure of briquettes with more PET, progressing from BR-B to BR-D, becomes noticeably denser, more tightly packed, and more uniform. The images reveal that the sawdust fibres are being compressed and wrapped in a smoother material, probably the melted and re-solidified PET plastic. This layer fills the gaps and helps bond the particles much more strongly. Overall, increasing the PET content from BR-A to BR-D results in a clear transition from a porous, loosely held-together structure to a denser, layered, and more cohesive one. This enhanced density and bonding make the briquettes more durable and less prone to breaking apart or disintegrating. The layered, fibrous composition also promotes more controlled and stable burning, which could reduce emissions and improve fuel efficiency.

In particular, briquettes with higher PET, like BR-C and BR-D, are likely to be stronger during handling and storage, and to burn more evenly and efficiently. This means they might burn longer and produce cleaner, more consistent emissions. The SEM images clearly show that adding PET improves internal cohesion and reduces porosity, thereby enhancing physical and thermal performance. On the other hand, pure biomass briquettes are more porous and fragile, which negatively affects their durability. Overall, as the PET content increases, the structure improves, making these PET-sawdust composites a promising option for sustainable, high-performance fuels.

Figure 6 presents the non-isothermal combustion behaviour of the briquettes (BR-A to BR-D) recorded using a thermogravimetric analyser (TGA). Table 6 presents key combustion parameters, including the ignition temperature (T_i), peak temperature (T_p), and burnout temperature (T_b). As shown in Figure 6, the briquettes initially experienced weight loss predominantly due to the evaporation of moisture and some volatiles, consistent with previous studies [7,21]. This initial weight loss occurred over approximately 26 to 344°C for BR-A, 28 to 324°C for BR-B, 26 to 330°C for BR-C, and 26 to 319°C for BR-D. The initial weight reduction was approximately 9.79% for BR-A, 10.25% for BR-B, 9.07% for BR-C, and 6.69% for BR-D.

The evaporation of moisture led to the formation of cracks on the briquettes’ surfaces, which subsequently facilitated the release of volatile compounds [42].

The combustion kinetics of the four briquettes (BR-A to BR-D) were determined using data from the thermogravimetric analysis. An Arrhenius-type kinetic model was employed to analyse the TG-DTG data. As shown in Figure 6, the DTG curves for the four briquettes, measured at a constant heating rate of 10°C/min, exhibit single, bell-shaped peaks without shoulders across the temperature range from ignition to burnout. This suggests that the combustion process for all four briquettes proceeded as a single-step reaction, consistent with established methods for identifying single-step kinetics [54,55]. The Coats-Redfern method was applied to analyse the combustion region, assuming the process follows a first-order kinetic reaction. While we acknowledge the concerns raised by the ICTAC Kinetics Committee regarding the limitations of single-heating-rate methods, such as their potential difficulty in accurately determining kinetic triplets due to the complex nature of biomass combustion, the Coats-Redfern method was chosen based on practical considerations. The Coats-Redfern technique is useful for initial assessments when approximate kinetic parameters surffice, enabling meaningful comparisons among different biomass types under similar conditions. Moreover, conducting experiments at various heating rates requires significant time, resources, and specialised equipment, which is quite taxing. Additionally, analysing data at a single heating rate mirrors a real-world briquette combustion scenario, where processes typically occur under constant heating conditions, thus increasing the relevance of the findings from this study. Despite its limitations, the Coats-Redfern method has been demonstrated to be reliable in numerous studies when applied with an awareness of its constraints [22,25,56,57,58]. Figure 8 presents the results of fitting the conversion rate data (x) as a function of reaction temperature (T) for briquettes BR-A to BR-D. The correlation coefficients ranged from 0.9598 to 0.9803.

Using the Coats-Redfern method at a heating rate of 10°C/min, the activation energies (E_a) for the four briquettes were calculated and are summarised in Table 8.

The results indicate that the E_a values for briquettes with plastic (BR-B to BR-D) range from 18.67 to 21.74 kJ/mol, and are lower than that of BR-A, without plastic (29.31 kJ/mol). The observed trend suggests that increasing the PET plastic content reduces the activation energy required for the briquettes’ combustion process. The key reason for the decrease in activation energy is the strong synergistic physical and chemical interactions between the decomposing biomass and plastic materials, which promote increased carbon conversion efficiency and a higher yield of volatile products compared to the combustion of biomass or plastic alone [59,60,61,62]. For instance, kinetic studies by Mishra et al. [61] demonstrated that blending Samanea saman seeds (SS) with PET plastic at a 3:1 ratio required a lower activation energy to initiate degradation than either component alone. Oyedun et al. [62] reported that increasing the plastic content in mixtures of sawdust and HDPE reduced the activation energy required to initiate the decomposition reaction. Similarly, Burra and Gupta [59] observed that blending polycarbonate with pinewood reduced the activation energy of the polycarbonate by about 50 kJ/mol, compared to the degradation of the individual polymer. The addition of PET plastic improved the reactivity of the briquettes during thermal decomposition, a trend that aligns well with observations from our study.

Table 9 presents the performance of briquettes BR-A to BR-D during the various WBT phases: cold start, hot start, and simmer, focusing on important parameters such as time to boil, burning rate, fuel consumption, energy efficiency, and firepower.

For the cold start high power phase, the time to boil reduced from BR-A to BR-D, implying enhanced heating performance. The burning rate and fuel consumption were lower, suggesting better combustion, and the thermal efficiency increased with the proportion of PET plastic, reaching above 54% in BR-D. Similar trends were observed in hot start, with reduced boiling time and improved efficiency. BR-D consistently exhibited the best performance, consuming less fuel and energy per liter, and having greater thermal efficiency. The burning rate and fuel consumption were much lower during simmering, indicating energy conservation. The thermal efficiency continues to increase to more than 73% in BR-D. The firepower was slightly reduced, which was appropriate to keep the heat at a constant low level.

In general, BR-D consistently outperforms the other briquettes across all the stages: it uses less fuel, has higher thermal efficiencies, and boils more quickly. This indicates that BR-D is more efficient in energy consumption, cheaper, and can heat quickly and simmer for a long time, hence it is more ideal for clean cooking [39,63].

Table 10 presents the energetic properties of the developed briquettes BR-A to BR-D. The data show a clear trend of increasing energetic density and Fuel Value Index (FVI) from BR-A to BR-D.

These parameters are essential indicators of the energy storage capacity, combustion efficiency, and general applicability of the briquettes across various applications. Energetic density increases to 16,929.38 MJ/m³ in BR-D as compared to 12,962.40 MJ/m³ in BR-A. An increase in the energetic density means that more energy is stored in a given volume. This increase enhances the efficiency of storage and transport since less volume is needed to produce the same energy output [39]. It also indicates better packing and densification of the briquettes, which may be because of the presence of PET particles or to superior production methods. The FVI rises from 2,365.40 MJ/m³% in BR-A to 6,461.59 MJ/m³% in BR-D. The Fuel Value Index (FVI) essentially measures how much energy a briquette contains based on its volume. The higher the FVI, the more efficient the briquette is at burning and the greater its potential to produce heat. Briquettes with a higher FVI are especially suitable for applications that require a lot of heat or long-lasting heat output [63]. The changes in energy properties observed are likely connected to physical changes in the briquettes’ structure (as shown in Figure 5). These include becoming denser, having fewer pores, and forming stronger bonds between particles. All of these factors help improve how well the briquettes burn and how much energy they retain during combustion.