Comparative Analysis of Stove Pyrolysis Performance of Donut Briquettes from Bus Wood Waste at Density Variations with and Without Used Oil Injection

The use of biomass as a fuel source for cooking in developing countries remains very common and has a significant impact on pollutant emissions and health problems due to kitchen fumes. The development of better biomass cooking furnaces is rooted in the principle of improving thermal efficiency through airflow regulation, combustion chamber design, and separation between pyrolysis and combustion zones. This aims to make energy release more controlled and particle emissions reduced. Recent research shows that furnaces using gasification and pyrolysis technologies (such as TLUD and downdraft types) can improve efficiency by up to two to three times compared to traditional furnaces, as well as significantly reduce carbon monoxide (CO) and PM₂ emissions, when their design and operation are optimized [32-34]. The theoretical basis for this involves the transfer of heat through convection and radiation from the fire to the pan, as well as the gasification/pyrolysis process in the biomass layer that is affected by the heating rate, oxygen availability, and physical characteristics of the fuel.

Densification of biomass into briquettes is an important step in increasing energy density, uniformity, and ease of handling fuel. Theoretically, increasing compaction pressure can increase the density as well as the mechanical strength of the briquettes, reduce the open porosity, and affect the diffusion of oxygen into the particles. This will have an impact on the combustion rate and heat release pattern. Recent research on briquettes and biochar shows that factors such as compaction pressure, adhesive type and content, and carbonization rate have a direct influence on calorific value, mechanical resistance, and emissions during the combustion process [35]. In briquettes that have holes (holeys or donuts), the addition of an air duct in the middle of the briquette serves to increase the surface area of the reaction and support airflow through the briquette core. Thus, it can theoretically increase the combustion power and improve the temperature distribution within the stove when combined with the right combustion chamber design.

The development of composite fuels through co-pyrolysis or co-combustion of biomass combined with waste oil, such as used cooking oil or other hydrocarbon fractions, is based on the principle of thermochemical synergy between solid and liquid fractions. Waste oil generally has a higher calorific value as well as greater volatility compared to biomass. Therefore, when this material is added in the form of drops or mixtures, it can increase the flame intensity, speed up the ignition process, and modify the distribution of pyrolysis products. Various studies on co-pyrolysis have shown that the addition of liquid or plastic hydrocarbon fractions to biomass can increase the proportion of high-energy products and improve the quality of the liquid or solid fuels produced [36, 37]. In the context of stove use, the injection of used oil in the flame zone can be considered a special example of co-combustion. In this case, the flow of oil droplets serves as an additional source of fuel that is able to increase temperature and firepower. But theoretically, this can also have an impact on emissions and flame stability if not managed properly.

The performance of biomass stoves is assessed based on a number of standard parameters, including thermal efficiency, usable energy, firepower, specific fuel consumption, and pollutant gas emissions. Theoretically, thermal efficiency is defined as the ratio between the energy absorbed by water (the increase in the enthalpy of water in the pot) and the chemical energy of the fuel used. Water Boiling Test (WBT) protocol to standardize efficiency, fuel consumption, and emissions under controlled laboratory testing conditions [38, 39]. Sensitivity analysis conducted in various studies shows that changes in test parameters, such as water volume, use of pot lids, and cold/hot start conditions, can affect the measured efficiency value. Therefore, for the purpose of comparison between briquette density and mode of operation (no vs with waste oil injection), it is necessary to apply a consistent test protocol so that the results can be accurately interpreted in the context of biomass stove performance theory.

This study aims to conduct a proximate analysis and performance of biomass pyrolysis stoves. The main focus of this study is to explore the temperature profile generated during the pyrolysis process, as well as the rate of energy that can be produced, both under conditions with and without the injection of used oil in the arrangement of donut briquettes (A, B, C) with densities of 30 bar, 40 bar, and 50 bar.

By comparing the two conditions, it is hoped that this study can provide broader insights into the effect of briquette density and waste oil injection on the performance of pyrolysis stoves, as well as provide useful recommendations for the development of biomass-based renewable energy technology in the future.

4.1 Result

The results of proximate testing on bus wood waste showed a moisture content of 10.83%, ash content of 1.79%, volatile 67.89%, fixed carbon 19.49%, and a calorific value of 4,446 cal/g. The high content of volatile matter indicates good thermal reactivity, while the low moisture and ash content generally support combustion efficiency and reduce fouling. Biomass with higher volatile matter and fixed carbon, accordingly, shows better thermal behavior [62, 63]. Tropical hardwoods often display HHVs in the range of 19-22 MJ/kg when moisture and ash contents are minimized [63, 64], placing the bus wood energy quality in a comparable class.

To assess the position of bus wood as a biomass fuel, these results were compared with several other types of biomass that have been extensively researched.

Table 1. Results of lignocellulose content

Table 2. Results of proximate testing of bus wood waste

Table 3. Comparison of the proximate analysis of several types of biomass

Note: Bus wood: Result; Rice husk: [65-67]; EFB of oil palm: [68]; Tropical Hardwood: [64, 69]; Corn Cob residue: [70].

From Tables 1-3, Figure 2, and Figure 3, it is evident that the properties of bus wood closely resemble those of tropical hardwood biomass. The relatively low levels of moisture and ash suggest its appropriateness for combustion and pyrolysis applications. Furthermore, the high volatile matter content of bus wood, measured at 67.89%, indicates excellent thermal reactivity. This figure is comparable to certain agricultural biomasses, including corn cob, which has been reported to have approximately 68%–72% volatile matter, around 4%–6% ash content, and a higher heating value (HHV) of about 18.00 MJ/kg (4,300 cal/g) [68]. In contrast, rice husks typically contain higher ash levels (15%–20%), which increase the risk of fouling and slagging. Meanwhile, several tropical hardwoods exhibit volatile matter in the range 65%–70% and calorific values between 18.84–20.09 MJ/kg (4,500–4,800 cal/g), confirming that bus wood has energy quality comparable to other hardwood species [64, 69].

Figure 2. Comparison of the proximate composition of several biomass

Figure 3. Comparison of calorific values of some biomass

The calorific value of bus wood (4,446 cal/g) puts it in the middle to upper-middle position in the biomass classification. This value is higher than rice husks, but slightly lower than tropical hardwood. This can be explained by the fairly high lignin content of bus wood (31.62%), contributing to long-term energy during combustion.

The following is a graph of the temperature distribution of test results (Figures 4-12).

Figure 4. Temperature distribution (T1) on the pyrolysis stove

Figure 5. Temperature distribution (T2) on the pyrolysis stove

Figure 6. Temperature distribution (T3) on the pyrolysis stove

Figure 7. Temperature distribution (T4) on the pyrolysis stove

Figure 8. Temperature distribution (T5) on the pyrolysis stove

Figure 9. Temperature distribution (T6) on the pyrolysis stove

Figure 10. Temperature distribution (T7) on the pyrolysis stove

Figure 11. Temperature distribution (T8) on the pyrolysis stove

Figure 12. Temperature distribution (T9) on the pyrolysis stove

4.1.1 Temperature profile on briquettes density 30 bar (T1–T3)

Based on Figures 4-6, with a density of 30 bar, the temperature development on the outer edge of the briquettes (T1) shows a gradual rise from about 30–40 ℃ at the beginning of the test to about 150–170 ℃ at the 220th minute, then a sharp increase near 300 ℃ towards the end of the process. The temperature curve for conditions without and with the injection of used oil is almost constant all the time, so that in the outer edge zone, the effect of the addition of used oil on the temperature is relatively small. This pattern corresponds to the character of a fixed-bed stove where the main reaction zone is in the core part of the bed, while the part near the wall tends to be at lower temperatures [71, 72].

The middle radial temperature profile (T2) shows a clearer difference between the two treatments. Without the injection of used oil, the temperature rises relatively quickly after the initial heating phase, reaching about 250 ℃ at the 70th minute and continuing to rise to the range of 450–500 ℃ with some local fluctuations, before approaching 600–650 ℃. With used oil injection, the temperature rise in the same position is held longer; The T2 curve flattens at around 90–110 ℃ until about the 120th minute, only then gradually increases and persists in the range of 380–420 ℃ until the end. This suggests that at low densities, the addition of waste oil tends to lower the maximum temperature in the central zone and prolong the heating phase, although the temperature remains within the commonly reported biomass slow pyrolysis range [73, 74].

In the position of the inner edge bordering the central hole of the briquette (T3), both treatments produced the highest temperature compared to T1 and T2. Without the injection of used oil, the temperature of the T3 rises gradually and only reaches 450–500 ℃ after the 60th minute, then shows some sharp peaks that can approach 800–900 ℃ before slowly decreasing. With used oil injection, the temperature rise of T3 is much faster; within ±20 minutes after start-up, the temperature has already passed 600 ℃ and then persists in the range of 650–700 ℃ with more subtle fluctuations. This temperature range of 600–800 ℃ is consistent with the reactor core conditions in pyrolysis and gasification of high-energy biomass [35, 73].

4.1.2 Temperature profile on briquettes density 40 bar (T4–T6)

Based on Figures 7-9, with a medium density (40 bar), the outer edge sensor (T4) again shows the lowest temperature among the three radial positions. Without the injection of used oil, the temperature of the T4 increases from about 30–40 ℃ to 70–80 ℃ in the 60th minute, then gradually reaches 200–220 ℃ and peaks at around 280–300 ℃. With used oil injection, the curve shapes similarly, but the peak is slightly lower, around 190–210 ℃, with no sharp spikes at the end. This indicates that in the outer edge zone, the addition of used oil tends to dampen the rise in peak temperatures and produce a smoother heating profile, in line with the finding that changes in fuel composition more predominantly affect the core zone of the reactor [71, 75].

The difference in the effect of used oil injection is clearly visible on the central radial sensor (T5). Without oil injection, the temperature rises gradually to about 450–500 ℃ and then decreases slightly towards the end. With oil injection, the temperature rise after the 120th minute becomes faster and continues until it reaches the range of 650–700 ℃ before descending; The average temperature in this zone is higher overall than in pure biomass mode. This increase is consistent with the concepts of co-combustion and co-pyrolysis, where the addition of a fraction of high-calorific liquid fuel can increase flame intensity and local temperature [27, 76].

The inner-edge sensor (T6) at a density of 40 bar records the highest temperature value in layer B. Without oil injection, the temperature rises slowly, only exceeding 400–500 ℃ after the 60th minute and subsequently showing peaks that can approach 850–900 ℃. With used oil injection, the temperature rises much faster; in ±20–30 minutes, the temperature has been in the range of 650–750 ℃ and has remained relatively stable within that range before declining. This confirms that at medium density, the injection of used oil accelerates the achievement of high core temperatures and reduces the amplitude of peak temperature fluctuations, which is in line with the results of the model and co-combustion experiments of oil on biomass-based systems [27, 75].

4.1.3 Temperature profile on briquettes density 50 bar (T7–T9)

Based on Figures 9-12, at the highest density layer (50 bar), the temperature at the outer edge (T7) increases from about 30–40 ℃ to almost 200 ℃ at about 120 minutes, then rises sharply to the range of 400–450 ℃ before declining. With used oil injection, the T7 temperature rise is initially slightly delayed, but subsequently the curve with oil goes beyond the oilless curve and reaches a peak of about 450–470 ℃, with a more sloping descent. This shows that at high density, the influence of co-fuel begins to be felt even in the outer edge zone, as the heat distribution inside the bed is more uniform.

In the middle radial (T8), the differences between treatments become more pronounced. Without oil injection, the temperature rises to about 400–450 ℃ and persists in this range with little fluctuation. With oil injection, the temperature rises faster after the 130th minute and peaks at about 600–630 ℃, then gradually decreases but remains above the oilless curve for most of the rest of the test. This increase in temperature at the highest density is in line with reports that high-density briquettes with a compact structure can maintain a more intense flame when combined with high-calorific liquid fuels [77, 78].

The inner edge sensor (T9) shows the highest temperature in layer C. Without oil injection, the temperature rises rapidly to 550–600 ℃ around the 100th minute and then fluctuates in the range of 500–650 ℃. With used oil injection, the temperature rise is slightly delayed at the beginning, but after the 120th minute, the curve continues to rise until it reaches around 730–760 ℃ before slowly dropping towards the 500 ℃ range. The high core temperature range and longer duration of the temperature plateau in the waste oil treatment indicate the presence of a significant additional energy supply in the main reaction zone, consistent with the characteristics of biomass–oil co-pyrolysis and co-combustion reported in the literature [27, 74, 76].

4.1.4 Radial gradient patterns and effects of density–injection combination of used oil

Overall, the entire treatment combination showed a consistent radial temperature gradient, where the inner edge sensor (T3, T6, T9) always recorded the highest temperature, followed by the middle radial (T2, T5, T8), while the outer edge (T1, T4, T7) had the lowest temperature. This pattern is consistent with the characteristics of fixed-bed reactors and cylinder gasifier stoves, where the dominant oxidation and pyrolysis zones are near the fuel core and produce higher temperatures than the outer walls [71, 79].

In terms of density, an increase in compaction pressure from 30 to 40 and 50 bar tends to raise the maximum temperature and prolong the duration of high temperatures in the middle and deep radial positions, especially in treatment with used oil injection. This is consistent with the results of densification studies, which report that higher density briquettes generally result in more stable combustion, more uniform fire temperatures, and longer flame durations, although the total combustion rate decreases [23, 35, 78].

Regarding the effect of used oil injection, the results showed that the effect depended on the combination of density and radial position. At a density of 30 bar, waste oil mainly increases the temperature in the core (T3) but tends to decrease or delay the temperature rise in the middle radial (T2). At densities of 40 and 50 bar, oil injection consistently increases the temperature in the middle radial and core (T5, T6, T8, T9), accelerates the achievement of high-temperature plateaus, and dampens temperature spikes on the outer edges. This pattern is in line with the basic principles of co-combustion and co-pyrolysis, in which the addition of oil fractions increases flame intensity and local temperature, while modifying the heat distribution within the reactor [27, 75, 80].

4.1.5 Energy calculation using the Water Boiling Test (WBT)

Stove performance measurement is carried out by WBT, where energy is expressed as the rate of useful energy (MJ/h) transferred to the water in the pot. This method is widely used to compare the heat transfer efficiency between biomass stove designs in the laboratory and field [58].

Example of energy calculation using the WBT method: In the first test (no used oil injection) to heat water by 2 kg (initial temperature 28 ℃) until it boils, it takes 9.58 minutes, then the energy rate per hour is calculated as follows: Eq. (2). Table 4 shows a table of observations and the results of the calculation of energy rates.

$Energi per hour = \frac{Q}{t} \times 60$     (2)

where:

$Q=m \times c \times \Delta T$

$Q=2 \times 4186 \times(100-28)$

$Q=602784 \mathrm{~J}=0.6028 \mathrm{MJ}$

so:

$Energi\ per\ hour =\frac{Q}{t} \times 60$

$Energi\ per\ hour =\frac{0.6028}{9.58} \times 60=3.77$

The results of the WBT test (Figure 13) of 16 times show that the Kayu Bus briquette pyrolysis stove is able to produce relatively stable useful energy, both in operation without or with the injection of used oil. The hourly useful energy value is in the range of about 3.7–4.2 MJ/hour for non-injection conditions and around 3.8–4.3 MJ/h for conditions with used oil injection. The fluctuations between repetitions are relatively small; the energy value spread in each test mode was generally less than ±0.2 MJ/hr of its average inclination, indicating that the stove-briquette system performed fairly repeatably throughout the test series, in line with the performance stability findings of the stove gasifier and TLUD tested using WBT and ISO 19867-1 protocols [38, 58, 71].

Figure 13. Result Water Boiling Test (WBT)

Table 4. Time observation and energy rate calculation results

In general, the energy curve is useful for conditions with used oil injection (red line) being slightly above the no-injection curve (black line) on most test numbers. In the initial test (1st to 7th test), the difference between the two curves was relatively consistent: The useful energy with oil injection was about 0.05–0.2 MJ/hour higher than without injection. This shows that the addition of waste oil as a co-fuel is able to increase the utilization of heat energy from the pyrolysis process without disturbing the stability of stove operations, consistent with reports that the addition of a fraction of high-calorific liquid fuel can increase firepower and clean heat output in biomass conversion systems [27, 76].

In the mid-to-late tests (8th to 16th tests), the two curves showed a pattern of approximation, although at some test points (e.g., 9th and 13th tests), conditions with oil injection still provided slightly higher peak useful energy values than without oil. On the other hand, there are some repeats (e.g., the 8th and 10th tests) in which the useful energy with the oil injection drops close to or slightly below the conditions without oil, which indicates the presence of sensitivity to small variations in operating parameters, such as the rate of oil droplets or the distribution of briquettes. This phenomenon of inter-test variability is consistent with the sensitivity analysis of WBT and biomass stove test parameters reported in the literature, where small changes in test conditions can affect estimates of efficiency and useful energy [38, 81].

Overall, there was no systematic decline in performance from the first to the last repeat; both operating conditions maintained useful energy values in the range of 3.8–4.1 MJ/h with relatively narrow variations. This confirms that the combination of multilayer pyrolysis stoves and Bus Wood donut briquettes is capable of providing stable heat output in repeated use scenarios, and that the injection of used oil provides a small but consistent addition of useful energy. This energy range is within the range of efficient biomass stove performance reported in WBT studies and household stove performance tests in developing countries, suggesting that the stove-fuel configuration in this study is at a competitive level of performance for domestic cooking applications [71, 72, 79].

4.2 Discussion

The temperature distribution recorded at the nine measurement points (T1–T9) showed a consistent radial gradient in the annular pyrolysis reactor. The highest temperature was observed on the inside side of the briquettes (T3/T6/T9), followed by the middle (T2/T5/T8), while the outer side (T1/T4/T7) remained at the lowest temperature range. This pattern corresponds to the characteristics of fixed-bed downdraft, where intensive reaction zones typically form near hot gas flow channels and flame fronts, while the outer layer functions more as a thermal insulator [58, 79]. The maximum temperature reached inside, ranging from 800–900 ℃, suggests that the reactor can create pyrolysis conditions that are close to the high-temperature char production regime, which is important for the formation of high-quality biochar as well as tar decomposition [35, 82].

The impact of briquette density on the temperature profile becomes very clear when comparing layers A (30 bar), B (40 bar), and C (50 bar). At low pressure (30 bar), heating occurs more quickly on the inside, but temperature fluctuations are quite significant, especially at the T3 point, which reflects a shift in the fire front and variations in the rate of devolatilization. Meanwhile, at pressures of 40 and 50 bar, the increase in temperature at the T5/T6 and T8/T9 points takes place more gradually, but results in a more stable temperature plateau in the range of 650–800 ℃. These findings are in line with densification theory, which states that increased compaction pressure can shrink macro pores, improve interparticle interactions, and increase thermal conductivity. As a result, the movement of the pyrolysis front becomes more regular, and the combustion process becomes more stable [35, 83]. On the other hand, the temperature at the outer point (T1/T4/T7) remains below 300–320 ℃ for most of the process. This suggests a significant radial conduction gradient and reinforces the hypothesis that the briquette edge serves as an insulating "blanket" that reduces heat loss through convection to the reactor wall [58, 84].

A comparison between treatment with and without the injection of used oil showed that the effect of co-fuel was greatly influenced by the location of the thermocouple as well as the density of the briquettes. At a density of 30 bar, the injection of used oil slightly increases the rate of temperature rise at T1 and T3 in the early stages of the process. However, in T2, it actually produces a lower temperature with a smoother curve than the condition without injection. These findings indicate that the addition of oil can enrich the volatile phase and increase local heat capacity. On the other hand, it also has the potential to slow down oxidant diffusion as well as move the main reaction zone to the inside of the briquettes, thus causing the temperature in the middle zone not to reach the same level as in the case without the use of oil [85, 86]. The same trend—the acceleration of the heating process and the increase in core temperature—is observed more clearly at pressures of 40 and 50 bar, where the addition of oil causes the temperature to rise at the midpoint and depth (T5/T6 and T8/T9) to tens of degrees. This increase is mainly seen in the early to middle phase of pyrolysis, before gradually decreasing in the late phase due to reduced volatile content [27, 87].

These findings are consistent with research on co-pyrolysis between plastic waste and oil, which shows that the presence of oil can accelerate heat transfer, increase the rate of temperature rise, and support the formation of liquid and gas fractions with high calorific values through synergistic interactions between hydrocarbon components [27, 85]. However, when the effective porosity is too low as a result of the high density, the pyrolysis and oil results can be trapped. This leads to the formation of very local reaction zones with sharp temperature fluctuations, as seen at some peaks in the T3, T6, and T9 curves. A similar phenomenon has also been reported in high-pressure biomass briquettes, where a combination of high density and limited air supply can trigger periodic transitions between exothermic and endothermic reactions in the bed [35, 84].

Although the internal temperature gradient is quite extreme, the WBT test results show that the useful energy delivered to the pan is in a relatively narrow range, around 3.5–4.1 MJ/hr for all test variations, with the used oil injection providing an average increase of only about 0.1–0.2 MJ/hr. This suggests that most of the additional energy from the waste oil is channeled to accelerate internal pyrolysis and the formation of gaseous/liquid products, rather than directly increasing heat transfer to the water load [79, 88]. This condition is in line with research on WBT tests, which show that modifications to stove design or fuel characteristics do not necessarily result in significant improvements in thermal efficiency. This is due to several factors, such as the limitation of heat transfer areas, heat loss through walls, and fire instability, which is often the main barrier [58, 89].

Based on the comparison with the pyrolysis stove performance range described in the systematic review, the useful energy values obtained from this study were moderate but showed relatively consistent stability of both treatments [79, 81]. This stability is important from the perspective of the household user, since the variation in low useful energy is directly related to the ability to predict cooking time and comfort in operation, although the maximum level of efficiency has not yet been achieved. Previous research on biomass stoves using agricultural waste has also indicated that a combination of a reactor design that concentrates heat in the central area and the application of a high-density fuel can produce a more stable flame, albeit with only moderate WBT efficiency [35, 79].

Overall, the results of this study show that the use of medium to high density (40–50 bar) Bus donut briquettes combined with waste oil injection is more effective in creating a hot and stable pyrolysis environment in the reactor core. It provides benefits in the process of converting biomass into high-quality biochar and gas. However, the improvements obtained in cooking performance, measured through the WBT method, are limited. These findings support recent recommendations regarding the use of briquette-based stoves in developing countries, which suggest that the design of the technology should integrate two main objectives: efficiency in cooking and the production of value-added biochar, especially when utilizing biomass as well as local waste [35, 58].

From a systems engineering perspective, these findings open up opportunities for further development, for example, by optimizing the ratio of waste oil injection rate to airflow, increasing the heat transfer surface area above the reactor, and integrating emission measurements (CO, PM, and volatile organic compounds) to assess the trade-off between increased core temperature and potential increased emissions. The application of the latest ISO-based testing protocols and temperature distribution numerical modeling approaches will help to refine the stove design so that thermal and environmental performance can be improved simultaneously [58, 81].

The author would like to thank the Chancellor of Universitas Musamus, Lab. Materials and Lab. Biomass of Mechanical Engineering Education, Universitas Negeri Yogyakarta; Lab. Chem-Mix Pratama, D.I. Yogyakarta; Workshop Rayap Mesin Engineering, Kulon Progo Regency, D.I. Yogyakarta, and Workshop PT. Surya Kencana Gemilang Teknik, Sleman Regency, D.I. Yogyakarta, for providing support for this research.