Influences of sulfuric acid treatment on thermo-chemical properties of rice husk-pine sawdust heterogeneous briquettes

Abstract

Heterogeneous briquettes made from rice husk-pine sawdust blend treated with sulfuric acid have the potential to satisfy the increasing global demand for sustainable energy. However, untreated blends are constrained by unfavorable thermochemical properties. In this study, a blend comprising 10 wt% rice husk and 90 wt% pine sawdust was subjected to H₂SO₄ treatment at 1%, 2%, 3%, 4%, and 5% (v/v) for 1-h at ambient temperature. Briquettes were subsequently formulated, and thermochemical properties were assessed to identify treatment conditions enhancing energy and combustion performance. The briquettes were analyzed for lignocellulose using acid detergent fiber and lignin, and neutral detergent fiber; higher heating value HHV using calorimetry; and thermal efficiency and emissions using a water boiling test (WBT) coupled with a portable emission monitoring system (PEMS) 4000-series sensor. The results demonstrated that sulfuric acid treatment increased briquettes’ HHV from 18.21 ± 0.09 MJ/kg (untreated) up to 18.82 ± 0.03 MJ/kg. The 1% concentration yielded the highest thermal efficiency (71.49 ± 2.05%), accompanied by the lowest CO emissions (0.88 ± 0.03 g/MJd), meeting the ISO 19,867–1:2018 Tier 5 limits, and the lowest PM_2.5 emissions (100.67 ± 1.53 mg/MJd), within the Tier 3 limit. Overall, the optimal sulfuric acid treatment condition for effective combustion performance and energy delivery was 1%. However, cost–benefit analysis should be incorporated into future studies for practical real-world application.

Introduction

The rising global energy demand is one of the major challenges facing the World population. This energy demand is projected to grow by 34% between 2022 and 2050¹. This has exacerbated the already severe lack of access to reliable energy, especially in sub-Saharan Africa, where 730 million people depend on traditional biomass for household energy². One promising solution to this challenge is to produce and utilize heterogeneous briquettes from the vast amounts of forest and crop wastes generated. Heterogeneous briquettes are solid fuels generated by compressing a mixture of loose biomass materials, such as rice husks and pine sawdust³. However, briquettes from a blend of rice husks and sawdust have inferior thermochemical properties. For example, briquettes made at a proportion of 50%:50% blending ratio exhibited a calorific value of 16.68 MJ/kg and ash contents of 14.60%, which limits their usefulness in energy applications⁴. This is due to the high ash and low energy content of the constituent materials. For instance, the calorific value of rice husk ranges from 12.8 MJ/kg to 14.5 MJ/kg, and the ash content ranges from 15.87%_db to 25.56%_db, respectively⁵. Sulfuric acid can be used to treat these residues, thereby reducing ash content and increasing calorific value. This treatment involves soaking biomass materials in sulfuric acid at a specific concentration, reaction time, and temperature. The material is then thoroughly washed with water and dried. The dried biomass is then compressed to produce briquettes. This treatment removes microelements such as zinc, copper, and iron, as well as macroelements such as magnesium, calcium, and potassium, which are responsible for ash⁶. This therefore reduces the ash content, thus improving interparticle bonding during densification and thermal properties^7,8. Further, during this treatment, there is a hydrolytic breakdown of the polymer molecules, disintegrating the cellular matrix, accomplished by bond breakage in the lignin-carbohydrate complex, which removes the hemicellulose and condenses the lignin^9,10.

Some studies have examined the effect of sulfuric acid treatment on ash content and calorific value. Key among these is by Wistara et al., 2021⁷ where the ash content of oil palm biomass decreased from a range of 2.03% to 6.43% to a range of 0.62% to 4.27% and 0.4 2% to 2.15% at sulfuric acid concentrations of 1% and 5%, respectively. Another study was by Bamisaye et al., 2024¹¹, where the calorific value of watermelon peels was enhanced to 14.62 ± 0.21 MJ/kg and ash content reduced to 11.2 ± 0.13% after treatment with sulfuric acid. Similarly, the calorific value of watermelon waste increased by 31.83%, and the ash content decreased by 1.77% when treated with 0.5% sulfuric acid at 120 °C for 3 h, as reported by Adeleke et al., 2025¹². Further, the calorific value increased by 22.64% with sulfuric acid at 29.5% concentration at 120 °C for 3 h, as reported by Roy et al., 2022¹³.

These studies have focused primarily on the sulfuric acid treatment of single biomass feedstock and subsequent briquettes. The present study introduces a novel heterogeneous blend of rice husks and pine sawdust that has been subjected to sulfuric acid treatment. The mixed feedstock approach leverages the balance between the structural and chemical characteristics of agricultural residues (rice husks) and lignocellulosic wood waste (pine sawdust), thereby addressing waste valorization and performance enhancement. Further, the study goes beyond conventional analysis by employing a multi-criteria optimization framework that combines thermochemical properties, thermal efficiency, and emission behavior. This not only differentiates the work from the stated sulfuric acid-treated studies that emphasize isolated parameters but also establishes a comprehensive pathway for designing high-performance, environmentally friendly, sustainable briquettes from heterogeneous biomass blends.

The choice of blending rice husks and pine sawdust is based on availability and complementary fuel properties. Rice husks are an abundant agricultural residue in the region, including Uganda, with an annual rice production of 350,000 metric tons, of which 20% is rice husk¹⁴. However, its use as a fuel is limited by its high ash content, low bulk density, and low HHV of 15.87–25.56%_db, 88.82–124.26 kg/m³, and 12.77–14.54 MJ/kg, respectively⁵. Pine sawdust is widely generated as a by-product of timber processing, with a higher availability since it is 86% grown in Uganda compared to eucalyptus¹⁵. In addition, pine sawdust offers a lower ash content 1.72% and a higher HHV of 17.04 MJ/kg¹⁶. Combining these wastes contributes to their respective strengths^17,18, like rice husks, which enhance resource utilization for energy recovery, while pine contributes to combustion enhancement. Blending these feedstock materials addresses practical considerations of feedstock supply and production volumes, thereby ensuring that the study’s findings are relevant to real-world applications where the residues are readily available but singularly constrained.

To address this gap, an experimental study was conducted to investigate the influence of varying sulfuric concentrations (1%, 2%, 3%, 4%, and 5% by volume) on the thermal-chemical properties of briquettes from a blend of rice husks and pine sawdust. The objective of this study was to formulate and determine the effect of sulfuric acid treatment at varying concentrations on the thermochemical properties, and to identify the treatment conditions that yield the best energy and combustion efficiency for briquettes made from a blend of rice husks and pine sawdust.

Briquettes were formulated from a treated blend of rice husks and pine sawdust at constant compression pressure and particle size. The effect of treatment on the briquettes’ thermochemical properties was studied using a thermogravimetric analyzer (TGA) for proximate analysis; acid detergent fiber (ADF), acid detergent lignin (ADL) and neutral detergent fiber (NDF) for lignocellulose; oxygen bomb calorimetry for higher heating value (HHV) and water boiling test (WBT) incorporated with a portable emission monitoring system (PEMS) 4000-series sensor box for thermal efficiency and emission analysis.

Methodology

Preparation of raw materials

Rice husks and pine sawdust samples were collected from Tirinyi Central Ward, Tirinyi Town Council, Kibuku District, Eastern Uganda, and from K.W Timber Center, a local timber processing facility, in Kawanda, Wakiso District. The husks were produced from the Kaiso rice variety, which is among the most widely cultivated in Eastern Uganda and is known for its heavy grains, high spikelet fertility, and plentiful panicles, according to Hong et al., 2021¹⁹. The pine logs at the facility were processed using a locally fabricated circular blade power sawmill. The freshly collected materials were manually cleaned to remove dirt, stones, metals, and plastics, a method similar to that reported by Ajimotokan et al., 2019²⁰. The rice husk and pine sawdust materials were dried to a moisture content of 8.25 ± 0.16%_db for standardization. These were then blended with 10 wt% rice husk and 90 wt% pine sawdust, for a total weight of 10 kg, five times. A solution of sulfuric acid was prepared using 98% sulfuric acid at concentrations of 1%, 2%, 3%, 4%, and 5% to achieve a total volume of 25 L. The solution volume of 25 L was prepared at a solid-to-liquid ratio of 1:0.0025 g/l (1:2.5 kg/l), which is lower than the 1:0.025 g/l ratio reported by Nwosu-Obieogu et al., 2022²¹, thereby ensuring efficient utilization and minimizing acid consumption. Table 1 presents the specific quantities of acid and water used in the formulation. The acid volume required to make the solution was calculated using Eq. (1), where 98% is the concentration of the sulfuric acid reagent, and the required acid concentration is 1%, 2%, 3%, 4%, and 5%.

1

Table 1.

Quantities of acid and water for acid solution preparation.

Acid centration (%)	Acid (liter)	Water (liter)
1	0.26	24.74
2	0.52	24.48
3	0.78	24.22
4	1.04	23.96
5	1.30	23.70

Each of the five blends was independently introduced into acid solutions at 1%, 2%, 3%, 4%, and 5% concentrations at ambient temperature, and the mixtures were manually mixed for 5 min to ensure uniform distribution of the solid in the liquid. The mixtures were manually agitated at 10-min intervals throughout the one-hour reaction period to ensure uniform interaction of the material blend and the acid solution. The treated material blends were then thoroughly washed with water to remove any residual acid, thereby reducing the risk of corrosion in burner equipment during briquette use. Subsequently, the treated blends were dried in the sun for approximately 9 days (52 h) to a final moisture content of 8.34 ± 0.16%_db. The moisture content before and after washing and drying was determined in accordance with the Association of Official Analytical Chemists (AOAC, 2000) standard method, using a drying oven (Oven Plus II OPL225.WD1.C, Gallenkamp, United Kingdom). The particle size distribution was determined in accordance with BS ISO 3310–1:2000 using standard sieves with mesh sizes of 2.36 mm, 1.18 mm, and 0.425 mm.

Production of treated heterogeneous briquette samples

Treated and dried blends of rice husks and pine sawdust were compacted separately using a briquetting press (BHP 3/7 A0402, CHEMA Anlagen-und Apparatebau GmbH, German) at the Agricultural Engineering and Appropriate Technology Research Center (AEATREC)-Namalere to produce cylindrical treated heterogeneous briquettes of diameter 70mm and height ranging from 50 to 60 mm. Due to their hygroscopic nature, the briquette samples were packed in airtight bags and labeled THBS1, THBS2, THBS3, THBS4, and THBS5 corresponding to blends treated with sulfuric acid at concentrations of 1%, 2%, 3%, 4%, and 5%, respectively. Similarly, an untreated, dried mixture of rice husk and pine sawdust was compacted, packed, and labeled THBS0 as a control. This provided a baseline reference for the experiment, ensuring accurate comparison and confirming that the observed effects in briquette properties are due to sulfuric acid treatment. The sample was prepared without subjecting the rice husk and pine sawdust blend to sulfuric acid treatment.

Determination of properties of the briquettes

The treated heterogeneous briquette samples were evaluated for proximate, chemical, thermal, emission and mechanical properties to determine how these properties are influenced by sulfuric acid concentration, as described in the following subsections.

Proximate properties

The proximate properties measured were moisture content, volatile matter, fixed carbon, and ash content. These were determined using a Thermogravimetric analyzer (ETRA THERMOSTEP, ELTRA GmbH, Germany) in accordance with the Deutsches Institut für Normung (DIN) 51,718–20 standard, similar to the method used by George et al., 2023¹⁶, and Lubwama et al., 2018²². The measurement process involved heating 1 g of each briquette sample at 103 °C for 6 h in a thermogravimetric analyzer to determine moisture content. This was followed by connecting a Nuffle furnace to an inert gas and further heating the sample for 5 s at 750 °C to determine volatile matter. Finally, the furnace containing the sample was connected to high-purity oxygen and burned to ash. Weight loss is the fixed carbon, and the remaining ash is the ash content.

Chemical properties

The chemical properties measured were only lignocellulose composition (cellulose, lignin, and hemicellulose). Lignocellulose was determined by measuring acid detergent fiber (ADF) and acid detergent lignin (ADL), and neutral detergent fiber (NDF), following the International Organization for Standardization (ISO) 13,906:2008 and ISO 16,472:2006 standard methods, respectively²³. To determine ADF, ADL, and NDF, 20g of each sample was treated with neutral detergent, acid detergent, and 72% sulfuric acid. The cellulose and hemicellulose were then determined by taking the difference between ADL and ADF, and NDF and ADF, respectively⁵.

Thermal properties and emissions

The thermal properties assessed were higher heating value (HHV), burning rate, cooking power, and thermal efficiency. HHV was measured using an oxygen bomb calorimeter (IKA C 2000, IKA®-Werke GmbH & Co.KG, Germany), which is integrated with software that automatically computes HHV, a method also used by Olupot et al. 2016⁵. This was determined by burning 1g of treated heterogeneous briquettes under adiabatic conditions. Burning rate, cooking power, and thermal efficiency, as well as emissions, were measured using a water boiling test (WBT). These were determined in accordance with the ISO 19,867–1:2018 protocol for wood-based stove and fuel performance and emission testing, and following the Harmonized Laboratory Test Sequence²⁴. This system comprises the Aprovecho Laboratory Emission Monitoring System (LEMS), a portable emission monitoring system (PEMS) 4000-series sensor box, and a mini moto TLUD gasifier. It uses the MusaPlotter software and the Aprovecho Excel Analysis Tool (ISO 19,867 module) to acquire real-time data and process standardized data, respectively. The PM_2.5 was collected using 47 mm binderless borosilicate glass fiber filter (Part NO. FPAE-47, HI-Q Environmental Products Company, San Diego, CA, USA). These filters are first stabilized to meet the rigorous requirements of ISO 19,867, preventing moisture from affecting the weight. Calibration and initialization were performed using high-precision sensors for CO, CO₂, and PM_2.5 in the PEMS 4000 sensor box. This is followed by initialization of the MusaPlotter for data acquisition, and a dedicated CSV file was created for the test run. The sensors were given a 5-min warm-up, and the LEM blower was turned on for 10 min to stabilize the flow rate. To ensure a high-quality background filter was run before the test to measure the ambient laboratory emission, which was later subtracted from the experimental results. During the experiment, 200 g of treated heterogeneous briquettes were placed in the mini moto TLUD gasifier and ignited with 20 g of a briquette soaked in bioethanol for 2 min. A test pot containing seven liters of water was placed on the gasifier, and the experiment was run. This was followed by a standard test sequence involving cold-start high power to evaluate performance from a cold stove, hot-start high power to evaluate performance once the stove has reached thermal equilibrium, and low power to evaluate the ability to maintain a low boil with minimal emissions, each carried out for 30 min.

Finally, a post-test data processing was conducted by importing the recorded CSV data from the PEMS 4000 into the Aprovecho Excel analysis tool to compute thermal efficiency and specific emissions and assign ISO Tiers (0–5) for both emissions based on the ISO 19,867 framework. In addition, the analysis tool was used to compute the burning rate and cooking power. The tests were conducted at room temperature (25 ^°C), pressure (25.04 Hg), and humidity (60%) for 90 min. Burning rate, thermal efficiency, and cooking power were computed by Eqs.( 2, 3, and 4), which are incorporated in the Aprovecho Excel Analysis tool.

2

where M_{fd, cold}, M_{fd, hot}, and M_{fd, low} and Δt_cold, Δt_hot, and Δt_low are the mass of dry fuel consumed and time taken during the cold, hot and low power test phases. For this experiment, each test phase lasted 30 min (generating a total time of 90 min).

3

where M_w is the mass of water boiled (kg), C_pw is the specific heat capacity of water (0.004186 MJ/(kg.^oC), ΔT is the temperature difference (^oC), M_e is mass of vapour (kg) and λ_v is latent heat of vaporization (2.257 MJ/kg), M_f,dry is the mass of dry fuel consumed (kg) and LHV is lower heating value (MJ/kg) determined using Eq. (4).

4

where HHV is the higher heating value, h_v is the latent heat and is the mass of water formed

5

where M_w is the mass of water boiled, C_pw is the specific heat capacity of water (4.186 kJ/(kg. ^oC), ΔT is the change in temperature (^oC), M_e is the mass of vapour (kg), and λv is the latent heat of vaporization (2,257kJ/kg), M_f,dry is the mass of dry fuel consumed, and t is time (min).

Mechanical properties

The mechanical properties determined were bulk density and breakage resistance. The bulk density was measured following the American Society for Testing and Materials (ASTM) D7481-09 (2009), and density was calculated using Eq. (6), similar to that used by Orhadahwe et al., 2024²⁵. The height and diameter were measured with a Vernier caliper, and the mass was measured with a precision balance. The breakage resistance was determined using the breakage index, which was obtained by dropping a known mass M₁ from each briquette sample from a height of 1 m onto a hard surface and measuring the mass M₂ of the remaining briquette. Each briquette sample was dropped once. The breakage index was then calculated using Eq. (7), as used by Mulindwa et al.²⁶.

6

7

The proximate, lignocellulosic, thermal, and mechanical properties were measured in triplicate to ensure reproducibility. The mean, standard deviation, and p-values at the 5% significance level (ANOVA single-factor) were calculated using the data analysis tools in Microsoft Excel. Additionally, the Pearson correlation coefficient was calculated using the data analysis tools in Microsoft Excel. Pearson offers greater statistical power and accuracy for assessing linearity and normality.

Cofunding variables included material heterogeneity in particle size, variability in acid distribution, sunshine exposure, distribution of rice husks and pine sawdust, moisture levels at biomass collection, moisture absorption by briquettes, biomass contaminants, and measurement errors in weight. These were addressed by sieving biomass through a 3 mm sieve to minimize heterogeneity in particle size, continuously mixing blends in acid solution at 10-min intervals during the 1-h experiment to ensure uniform acid distribution, measuring daily sunshine hours to reduce variability, thoroughly mixing rice husks and pine sawdust for five cycles during blending for even distribution, drying freshly collected biomass to a uniform moisture content before blending and treatment, carefully removing foreign materials such as metals, stones, and plastics to prevent contamination, weighing biomass three times with a high-precision scale to eliminate measurement errors in weights, and packing briquettes in airtight bags to prevent air absorption from the environment.

Visual analysis was conducted using multi-series line charts incorporated with error bars generated in R version 4.5.2. Figure 1 presents the study’s methodological flow.

Fig. 1.

Methodological flow of the study.

Results and discussions

Moisture content and particle size of raw materials during the process

The moisture content of the rice husk-pine sawdust blend before treatment was 8.25 ± 0.16%_db, and after treatment with H₂SO₄ was 57.84 ± 0.29%_db before drying and 8.34 ± 0.16%_db after drying. The moisture contents of the untreated and treated blends after drying were below 13%, as reported by Lubwama et al. 2020²⁷ for successful briquetting. Particle size distribution of the untreated blend was 2.16%, 36.73%, 55.25%, and 5.86% for 2.36mm, 1.18mm, 0.425 mm, and less than 0.425 mm, respectively. For the treated blend, the values were 0%, 7.10%, 69.16%, and 23.74% for 2.36 mm, 1.18 mm, 0.425 mm, and less than 0.425 mm, respectively. The treated blends have a higher proportion of particles below 1.18 mm (92.9%) than the untreated blends (61.11%), indicating that treatment renders particles brittle and prone to shattering, thereby reducing particle size.

Effect of sulfuric acid concentration on proximate properties

The proximate properties investigated were moisture, ash content, volatile matter, and fixed carbon. The effects of H₂SO₄ at varying concentrations on these properties are presented in Fig. 2.

Fig. 2.

Effect of sulfuric acid concentration on (a) moisture, (b) ash content, (c) volatile matter, and (d) fixed carbon on briquettes.

In Fig. 2, the moisture content of treated rice husk-pine sawdust briquettes ranged from 7.79 ± 0.11% to 9.34 ± 0.15%. The briquettes were within the recommended range of less than 10% as defined by the German Institute for Standardization Deutsches Institut für Normung (DIN) 51,731¹⁸. This moisture content can reduce briquette decomposition and rotting²⁸. In addition, the exhibited moisture content can improve the combustion efficiency of briquettes, as less energy is required to drive off moisture during combustion²⁹. To maintain moisture at the recommended levels, the briquettes should be stored in an airtight bag to prevent moisture absorption from the environment, given their hygroscopic nature.

The ash content of briquettes from the treated blend ranged from 2.00 ± 0.64%_db to 3.53 ± 0.29%_db, with THBS1 the highest and THBS5 the lowest, all lower than 11.57 ± 0.53%_db from the untreated blend, indicating a reduction with increasing concentration. This is similar to the reduction reported by Cuvilas et al., 2012³⁰ for spruce from 2.82%_db to 1.45%_db when treated with 0.1 M H₂SO₄ at 180 ± 2 °C, for 150 min. Also similar to a reduction from 12.97 ± 0.06% to 11.20 ± 0.13% for briquettes made from watermelon peels treated with 0.5% H₂SO₄ at 120 °C for three hours, as reported by Bamisaye et al., 2024¹¹. Furthermore, it is similar to the reduction from 2.1% to 0.02% for melon seed husks treated with 3.5 M H₂SO₄ at 90 °C for one hour, reported by Nwosu-Obieogu et al., 2024²¹. The reduction is due to the removal of alkali and alkaline earth metals, which are responsible for ash, by H₂SO₄^6,7,31. Ash is transformed into inorganic oxides, such as alkali oxides, including K₂O, Na₂O, and alkaline earth oxides, including CaO, and MgO, and Si₂O²⁹. These oxides cause slagging due to ash diffusion as a result of the low deformation temperature of 900–1000 °C for alkali oxides³². The ash content should be kept low, as high ash content also downplays combustion rate and ignition due to limited transfer of heat during briquette combustion^17,18. These briquettes’ ash content is < 4%, meeting the ISO 17,225 class B for standardized domestic heating and class I1-I3 for industrial application³³. The correlation analysis between ash content and H₂SO₄ concentration yielded a coefficient of -0.984; increasing the concentration consistently reduced briquette ash content. In addition, there is a limited association between ash content and lignin (r = -0.338), cellulose (r = 0.137), and hemicellulose (r = 0.02).

The volatile matter ranged from 64.76 ± 4.61% to 78.25 ± 1.61%, with THBS4 the lowest and THBS5 the highest, versus 61.5 ± 0.57% for the untreated blend. Generally, it decreased with increasing acid concentration, except at 5%, where it increased. It is lower than for the untreated, which differs from findings by Cuvilas et al., 2012³⁰ where a decrease from 70.2% to 65.3% for spruce treated with 0.1 M H₂SO₄ at 180 ± 2 °C for 150 min and Bamisaye et al., 2024¹¹ where it decreased from 67.17 to 65.10 ± 0.04% for watermelon peels treated with 0.5% H₂SO₄ at 120 °C for three hours. The increase in volatile matter relative to the untreated blend is due to the hydrolysis of hemicellulose into arabinose and xylose, partial hydrolysis of cellulose into sugars, and partial modification of lignin into phenols^9,10. These by-products generate more volatiles compared to hemicellulose, cellulose, and lignin during combustion. An increase in volatile matter is also an indicator of potassium removal, as observed in ash reduction, since potassium is one of the elements responsible for ash³⁴. The general increase in volatile matter for the treated briquettes across all concentrations indicates their readiness to ignite with a proportionate flame and enhanced efficiency^20,29. The correlation between acid concentration and volatile matter yielded a coefficient of -0.170, indicating that an increase in acid concentration has a marginal effect on reducing volatile matter. In contrast, a correlation between cellulose and volatile matter yields a coefficient of 0.867, suggesting that volatile matter is more strongly associated with increases in cellulose. In addition, there is a strong association of volatile matter and reducing lignin (r = -0.750), increasing cellulose (r = 0.867), and decreasing hemicellulose (r = -0.926).

The fixed carbon ranged from 10.62 ± 2.65% to 25.09 ± 4.55%, with THBS1 the lowest and THBS4 the highest. The values at THBS3 and THBS4 were higher than 17.50 ± 0.69% for the untreated blend. This increase is similar to that reported by Cuvilas et al., 2012³⁰ where it increased from 19.82 ± 0.11% to 22.70 ± 0.15%for spruce treated with 0.1 M H₂SO₄ treated at 180 ± 2 °C for 150 min, and Bamisaye et al., 2024¹¹ where it increased from 19.82 ± 0.11% to 22.70 ± 0.15% for watermelon peels treated with 0.5% H₂SO₄ at 120 °C for three hours. The fixed carbon increased with increasing H₂SO₄ concentration up to 4%, then decreased at 5%. This is because, at less than 5%, partial hydrolysis of hemicellulose and removal of inorganic elements occur, thereby increasing volatile matter and reducing ash content, respectively, thereby increasing fixed carbon. In addition, hemicellulose is more susceptible to acid hydrolysis than lignin, which has a condensed, aromatic structure that makes it resistant to acid attack; thus, lignin is retained, which substantially contributes to fixed carbon³⁵. However, at 5% concentration, there is potential for excessive degradation of hemicellulose and cellulose and partial degradation of lignin, thereby reducing subsequent lignin and, consequently, fixed carbon. A correlation between H₂SO₄ acid concentration and fixed carbon produced a coefficient of 0.351, suggesting that concentration marginally increases fixed carbon. In contrast, that between lignin and fixed carbon was 0.727, which is much stronger, indicating that fixed carbon is strongly associated with increases in lignin. Further, there is a strong association of fixed carbon with increasing lignin (r = 0.727), decreasing cellulose (r = -0.815), and increasing hemicellulose (r = 0.855). Furthermore, a strong negative association between fixed carbon and volatile matter (r = -0.979) is observed, suggesting that an increase in fixed carbon is closely associated with a reduction in volatile matter. Similarly, a limited association occurs between ash content with volatile matter (r = 0.230), and a moderate negative correlation with fixed carbon (r = -0412). This relationship indicates a trade-off in treatment, in which lignin and carbon stabilization occur at the expense of volatile matter. Table 3 presents the summary of the ANOVA analysis for concentration, ash, volatile matter, and fixed carbon, while Fig. 3 presents the graphical representation for Pearson correlation coefficients of concentration, lignin, hemicellulose, and cellulose with ash, volatile matter, and fixed carbon.

Table 3.

Pearson correlation coefficient and ANOVA analysis for lignocellulose composition.

	Pearson correlation coefficient (r)			ANOVA single-factor analysis
	Lignin	Hemicellulose	Cellulose	F	P-value	F crit
Concentration	0.252	− 0.122	− 0.026	NA	NA	NA
Lignin	1.000	0.780	− 0.928	1.819	0.214	5.318
Cellulose	− 0.928	− 0.950	1.000	0.201	0.666	5.318
Hemicellulose	0.780	1.000	-0.950	1.151	0.315	5.318

NA: Not applicable.

Fig. 3.

Graphical representation of the Pearson Correlation coefficient for ash content, volatile matter, and fixed carbon with concentration and lignocellulose composition. NA: Not applicable.

From Table 2, the ANOVA analysis indicates a statistically significant difference between treated and untreated briquettes in ash content and volatile matter, as indicated by F > F crit and a P-value < 0.05. However, there is no statistically significant difference in the fixed carbon of untreated and treated briquettes since F < F crit and P-value > 0.05.

Table 2.

Pearson correlation and ANOVA single-factor analysis for proximate properties.

Factor	ANOVA single-factor analysis
	F	P-value	F crit
Concentration	NA	NA	NA
Ash content	980.669	1.18 × 10–9	5.318
Volatile matter	21.998	0.002	5.318
Fixed carbon	0.313	0.591	5.318

NA: Not applicable.

The values of moisture content, ash content, volatile matter, and fixed carbon in the samples THBS1-THBS5 are presented in Fig. 2, along with their respective uncertainties as mean and standard deviation. The uncertainties propagate through comparative analysis of these properties, influencing the confidence with which samples are interpreted. The uncertainty in moisture content ranged from ± 0.03 to ± 0.38%, with the smallest indicating relatively high precision in distinguishing between the samples, while that of ash content ranged from ± 0.17 to ± 0.63%, resulting in greater overlap among the samples and reducing the statistical significance of small differences. In the same way, the large uncertainties in volatile matter (± 0.42–4.61%) and fixed carbon (± 0.61–4.55%) show that the exact magnitudes of variation must be interpreted within the bounds of these propagated errors, much as the general trend is robust. Overall, error margins indicate that the observed differences are reliable within the stated confidence intervals, though caution is warranted when comparing samples with overlapping uncertainty ranges.

Effect of sulfuric acid concentration on chemical properties

The lignocellulose composition was the only chemical property that was determined. Lignocellulose composition included cellulose, hemicellulose, and lignin (true lignin). The effect of sulfuric acid 1% (THBS1), 2% (THBS2), 3% (THBS3), 4% (THBS4), and 5% (THBS5) sulfuric acid concentrations on lignocellulose composition of produced briquettes, as well as for untreated, is presented in Fig. 4.

Fig. 4.

Effect of sulfuric acid concentration on (a) Lignin, (b) Hemicellulose, and (c) Cellulose.

The true lignin ranged 8.36 ± 0.23% to 13.34 ± 0.65%, with THBS1 being the lowest and THBS3 the highest. The trend shows an increase between acid concentrations of 1–3%, followed by a drop between 3–5%. At a low acid concentration (1%), partial cleavage of lignin-carbohydrate linkage leads to solubilization of lignin fragments, thereby resulting in a reduction of measurable lignin content compared to untreated briquettes (10.14 ± 0.34%). At moderate concentrations (2–3%), lignin undergoes condensation and reprecipitates, leading to an apparent increase in its measured content in the solid fraction. This is due to the formation of Pseudo-lignin and the redeposition of lignin fragments on the biomass matrix³⁶. At higher concentrations (4–5%), there is increased hydrolytic cleavage of lignin-carbohydrate linkages and possible depolymerization of lignin structures, resulting in partial solubilization and lignin removal, thus reducing the overall content. Lignin, though resistant, can undergo structural changes during acid treatment, including solubilization and condensation, thereby producing a non-linear relationship³⁷.

Hemicellulose ranged from 11.61 ± 1.12% to 24.04 ± 0.08% at 4%, with THBS4 being the highest and THBS5 the lowest. It increased between 1–4% acid concentration, then decreased between 4–5%. At a lower acid concentration (1%), possible partial hydrolysis into xylose and arabinose, which are reducing sugars, as reported by Monlau et al., 2013³⁸, reduces hemicellulose compared to the untreated briquettes (16.21 ± 0.30%). At 1–4%, the initial increase in hemicellulose is attributed to relative enrichment rather than absolute formation, due to possible removal of extractives and ash, leaving hemicellulose largely intact because of insufficient hydrolytic severity at lower acid concentrations, enhancing fractional separation rather than complete solubilization of hemicellulose³⁹. The continued increase in hemicellulose before reaching a peak at 4% acid concentration is due to a transitional phase between partial preservation and progressive depolymerization. The decline in hemicellulose after 4% acid concentration is due to increased hydrolytic severity, breaking down hemicellulose into xylose and arabinose, reducing, and possibly further degradation of the sugar into furfural and 5-hydroxymethyl furfural⁴⁰.

Cellulose ranged from 34.75 ± 1.64% to 39.64 ± 1.28%, with THBS3 the lowest and THBS5 the highest. It decreased between 1–3% acid concentration and then increased between 3–5%. The higher cellulose content at 1% concentration compared to the untreated briquettes (37.43 ± 1.24%) is due to reduced lignin and hemicellulose. The decrease between 1–3% acid concentration is due to hydrolysis of the amorphous region, which is more accessible and susceptible to acid attack by increased hydrogen ions (H⁺)⁴¹. This results in partial polymerization by disrupting the bonding structural organization, reducing cellulose integrity, and hence loss of cellulose mass at mild acid treatment⁴². The observed increase in cellulose beyond 3% is due to relative enrichment resulting from possible extensive hemicellulose degradation and partial lignin removal. This leaves a solid fraction that is increasingly rich in cellulose. Further, the removal of amorphous cellulose leaves a more acid-resistant crystalline cellulose, with increased crystallinity as severity increases⁴³.

Generally, the apparent increase in cellulose and a reduction in hemicellulose after 4% acid concentration is possibly due to the degradation of hemicellulose into reducing sugars and further into furfural and 5-hydroxymethylfurfural, while cellulose undergoes further depolymerization due to the attack of its amorphous region by the H⁺. This results in a relative increase in cellulose proportion as the loss of hemicellulose shifts to balance the solid composition, hence the observed trend in cellulose and hemicellulose.

Overall, the compositional trend shows that moderate acid treatment concentration (2–3%), yields optimal lignin levels (12.68 ± 0.28–13.34 ± 0.65%). This enriches the bonding of biomass during compaction^44,45 through the exhibition of thermoplastic behavior at higher temperatures, and by acting as a lubricant⁴⁵. The non-linear response of lignin, hemicellulose, and cellulose shows the competing processes of solubilization, condensation, hydrolysis, and pseudo-lignin formation. The drop in lignin and hemicellulose for THBS1 and THBS5 below the untreated blends is similar to the results reported by Nwosu-Obieogu et al., 2022²¹ where the reductions were from 0.29% to 0.003% and from 23.0% to 5.0%, respectively, and the cellulose reduction for THBS1-THBS3 samples was similar to the reduction from 71.8% to 21.4% for melon seed husks treated with 3.5 M H₂SO₄ at 90 °C for 1 h. This results discussion points to an optimal treatment range of 2–3% where compositional and structural integrity balance can favour briquette performance.

Table 3 presents the Pearson correlation coefficient and ANOVA single-factor analysis of acid concentration and lignocellulosic composition. From Table 3, it is observed that there is no statistically significant difference in the lignocellulose composition between untreated and treated briquettes, as the P-values are > 0.05 and the F-values are < F crit. Regression analysis of concentration and lignocellulose compositions generated r values close to 0, indicating that there is little effect of lower acid concentration on lignocellulose composition of briquettes.

The apparent drop in lignin and hemicellulose in THBS5 is likely due to an analytical limitation inherent to sugar-based quantification rather than to complete structural degradation⁴⁶. Lignin is resistant to sulfuric acid hydrolysis at increasing concentrations, partial solubilization, and redistribution of degraded products can influence the analytical balance, leading to an apparent increase in cellulose⁴⁷. This shift complicates the direct interpretation of compositional changes, thereby underscoring the need to consider methodological constraints when evaluating lignocellulose fractions in biomass.

There was inherent variability of replicate measurements as shown by the means and standard deviation representation of the true lignin, hemicellulose, and cellulose in the samples THBS1-THBS5 in Fig. 3. These uncertainties propagate into any comparative interpretation of data. The uncertainty in true lignin ranged from ± 0.16–0.65%, the smallest, indicating a relatively high precision in distinguishing between samples, compared to that in hemicellulose (± 0.08–2.16%) and cellulose (± 0.13–2.15%), which were larger, indicating greater overlap among the samples and reducing the statistical significance of small differences. The overlap in the error margins for true lignin in samples THBS2 (12.68 ± 0.28%) and THBS3 (13.34 ± 0.65%) indicates that the observed difference may not be statistically significant when measurement uncertainty is considered, compared to hemicellulose and cellulose for the same samples. The propagated uncertainties influence the confidence intervals of regression analyses and trend interpretation when evaluating the relationships among lignin, hemicellulose, and cellulose compositions in the briquettes.

Effect of sulfuric acid concentration on thermal properties of the produced briquettes

The thermal properties studied were higher heating value (HHV), thermal efficiency, cooking power, and burning rate. The effect of H₂SO₄ concentration on these properties is presented in subsections 3.4.1, 3.4.2, and 3.4.3.

Higher heating value

The effect of H₂SO₄ at varying concentrations on the higher heating value (HHV) of rice husk-pine sawdust briquettes is shown in Fig. 5. The HHV of treated briquettes ranged from 18.31 ± 0.02 MJ/kg to 18.82 ± 0.03 MJ/kg, with THBS1 the lowest and THBS4 the highest, compared with 18.21 ± 0.09 MJ/kg for the untreated (THBS0).

Fig. 5.

Effect of sulfuric acid concentration on HHV.

The optimal treatment was achieved at an acid concentration of 4%, at which the enhanced HHV was attributed to selective hemicellulose hydrolysis and partial lignin modification. At low acid concentrations (1–2%), the increase in HHV was minimal due to limited hemicellulose hydrolysis and lignin modification. Between 3–4%, the increase was more pronounced, consistent with greater hemicellulose hydrolysis and lignin modification, which enriched cellulose (34.75 ± 1.64% to 35.03 ± 0.20%) and lignin (12.70 ± 0.16% to 13.34 ± 0.65%). Both cellulose and lignin are contributors to thermal properties³⁹. Between 4–5%, however, the HHV decreased, suggesting over-hydrolysis of hemicellulose and cellulose degradation, leading to the formation of inhibitory compounds that reduce energy yield.

ANOVA single-factor test comparing HHV of untreated and treated briquettes yielded F-statistic (F), p-value (p), and critical F-value (F crit) of 6.36, 0.036, and 5.32, respectively. These results indicate that sulfuric acid treatment of the biomass significantly increases the HHV of resulting briquettes (F > Fcrit and p < 0.05).

The increase in HHV ranged from 0.51% to 3.3%, with THBS1 showing the smallest increase and THBS4 the largest. THBS4 is higher than the 1.72% (15.33MJ/kg) increment reported by Aslam et al.²⁷ for briquettes made from rice husks treated with 5% H₂SO₄ at ambient temperature for 1 h. However, the increment is lower than the 22.64% recorded for rice stalks treated with sulfuric acid at 29.5% concentration, 120 °C temperature, and three hours reaction time reported by Roy et al., 2022¹³. Furthermore, it is lower than the 31.83% increase reported for watermelon waste treated with H₂SO₄ at 0.5% concentration, 120 °C, and three hours of reaction time¹². It is also lower than the 22.07% increase in calorific value observed for watermelon peels treated with H₂SO₄ at 0.5 concentration, 120 °C, and a reaction time of three hours, as reported by Bamisaye et al., 2024¹¹.

HHV generated correlation coefficient of 0.463 with concentration, 0.487 with fixed carbon, 0.488 with lignin, -0.876 with volatile matter, -0.510 with ash. Suggesting that an increase in HHV is more strongly associated with a reduction in volatile matter than with a reduction in ash, and with increases in H₂SO₄ concentration, fixed carbon, and lignin. In the analysis of HHV and moisture, low moisture content did not necessarily translate into a higher heating value. This is due to the intrinsic chemistry of the biomass, particularly as induced by H₂SO₄ hydrolysis of the lignocellulose matrix. This treatment promotes absorption of alkali and alkaline earth metals while facilitating solubilization of hemicellulose, thereby altering the fuel’s energy characteristics. Table 4 presents a summary of the Pearson correlation coefficient and the ANOVA single-factor analysis.

Table 4.

Pearson correlation and ANOVA single-factor analysis for HHV.

	Pearson correlation coefficient (r)	ANOVA single-factor analysis
	HHV	F	P-value	F crit
Concentration	0.487	NA	NA	NA
Lignin	0.488	1.819	0.214	5.318
Cellulose	– 0.602	0.201	0.666	5.318
Hemicellulose	0.688	1.151	0.315	5.318
Ash content	– 0.510	980.669	1.18 × 10–9	5.318
Volatile matter	– 0.876	21.998	0.002	5.318
Fixed carbon	0.487	0.313	0.591	5.318
HHV	1.000	6.36	0.036	5.32

NA: Not applicable.

The uncertainty in HHV for the samples is presented as the mean and standard deviation in Fig. 4. The uncertainty for the samples ranged from 0.00% to 0.38%, indicating relatively high precision in distinguishing the HHV of the samples. The overlap of error margins for sample THBS3 (18.47 ± 0.38 MJ/kg) and THBS4 (18.82 ± 0.03 MJ/kg) indicates that the apparent difference between the HHV may not be statistically significant when the measurement uncertainty is put into consideration. The relatively small uncertainty of THBS1 (± 0.02 MJ/kg), THBS4 (± 0.03 MJ/kg), and THBS5 (± 0.00 MJ/kg) indicates high precision, while the larger uncertainty in THBS3 (0.38 MJ/kg) indicates overlap and less precision.

Thermal efficiency, burning rate, and cooking power

The effects of H₂SO₄ at varying concentrations on thermal efficiency, burning rate, and cooking power in briquettes are presented in Fig. 6. The thermal efficiency used in this work is the char-based thermal efficiency. The thermal efficiency ranged from 50.28 ± 1.71% to 71.49 ± 2.05%, with THBS1 the highest and THBS5 the lowest, all exceeding the 49.95 ± 0.58% of the untreated blend. Generally, thermal efficiency decreased with increasing acid concentration. The highest efficiency is higher than the 61.04–64.00% range reported by Abdulkareem et al., 2018⁴⁸ and Walozi et al. 2026¹⁸ for briquettes made from a blend of non-charred sawdust and cardboard waste. It is also close to the lower value of 75.06% reported by Elinge et al., 2021⁴⁹ and Walozi et al., 2026¹⁸ for briquettes made from non-charred coconut and groundnut shells, with a starch slurry binder. At 1% acid concentration, the briquette thermal efficiency was higher due to minimal distortion of the cellulose structure and compaction, thereby enhancing heat transfer and minimizing energy loss, as evidenced by the high cellulose content. When the concentration increased to 2–3%, a pronounced reduction in thermal efficiency was observed due to moderate hydrolysis, resulting in cellulose partial degradation and hemicellulose degradation, which reduces combustion stability. Further increases in concentration beyond 3% generate a plateau, indicating thermal efficiency stability due to persistent degradation of cellulose and hemicellulose, resulting in incomplete energy conversion.

Fig. 6.

Effect of sulfuric acid concentration on (a) thermal efficiency, (b) burning rate, and (c) cooking power for the briquettes.

Thermal efficiency shows correlations of -0.893 with concentration, -0.569 with HHV, -0.567 with fixed carbon, -0.617 with lignin, 0.431 with cellulose, 0.946 with ash, and 0.422 with volatile matter. This indicates that higher thermal efficiency is more strongly associated with ash than with volatile matter and cellulose, and with reduced H₂SO₄ concentration, fixed carbon, and lignin. However, the thermal efficiency is higher than that of briquettes from untreated blends with higher ash content, indicating that treatment enhances thermal efficiency.

In Fig. 6, the briquette burning rate increased with increasing acid concentration, ranging from 6.17 ± 0.09 g/min to 7.44 ± 0.01 g/min, with THBS1 being the lowest and THBS4 the highest. The burning rate of THBS1 is lower than that of the untreated blend (THBS0; 7.28 ± 0.01 g/min). At a lower concentration (1%), there was minimal degradation of hemicellulose and modification of lignin; thus, few pores were formed within the material. The briquettes burn slowly, making them suitable for slow-heating applications. As the concentration increased to 2–4%, the burning rate increased rapidly due to increased degradation of hemicellulose and amorphous cellulose, thereby creating more pores in the biomass blends. Further increases in concentration led to a slight decline due to over-degradation of cellulose and hemicellulose and to lignin modification. The burning rate yields coefficients of 0.459 with HHV, 0.635with fixed carbon, 0.984 with lignin, -0.677 with cellulose, -0.632 with ash, -0.592 with volatile matter and 0.604 with concentration on correlation analysis, indicating that the burning rate increases with an increase in HHV, fixed carbon, concentration and lignin, and with a reduction in cellulose, ash, and volatile matter.

Further, cooking power decreased with increasing acid concentration, ranging from 513 ± 3.51 kW to 1001.67 ± 8.50 kW, with THBS1 being the highest and THBS3 the lowest. The cooking power of THBS1 exceeds 872.00 ± 0.00 kW for the untreated blend briquettes. At 1% concentration, the cooking power is highest, as is the thermal efficiency, indicating that all energy released is efficiently transferred to the cooking process. When the concentration increased to 2–3%, the cooking power decreased markedly due to reduced thermal efficiency and accelerated combustion, leading to unstable combustion. When the concentration increased to 4%, a gain was observed due to a slight increase in thermal efficiency and burning rate, which improved the balance. However, due to further degradation of cellulose and hemicellulose, the cooking power decreased at an acid concentration of 5%. The cooking power yields a coefficient of -0.785 with acid concentration, -0.468 with HHV, -0.532 with fixed carbon, 0.414 with volatile matter, 0.533 with cellulose, -0.727 with lignin and 0.859 with ash content on correlation analysis, indicating that cooking power is increases with an increase in volatile matter, cellulose and ash content and with a reduction in acid concentration, HHV, lignin and fixed carbon. The cooking power shows a correlation coefficient of 0.973 with thermal efficiency and -0.858 with burning rate in the correlation analysis, indicating that cooking power increases with increasing thermal efficiency and decreases with increasing burning rate. Further, the correlation between thermal efficiency and burning rate yielded a coefficient of -0.808, indicating that thermal efficiency increases as burning rate decreases. Table 5 presents a summary of the Pearson correlation coefficient and ANOVA single-factor analysis for thermal efficiency, burning rate and cooking power.

Table 5.

Pearson correlation and ANOVA single-factor analysis for thermal efficiency, burning rate and cooking power.

	Pearson correlation coefficient (r)			ANOVA single-factor analysis
	Thermal efficiency	Burning rate	Cooking power	F	P-value	F crit
Concentration	− 0.893	0.604	− 0.785	NA	NA	NA
Lignin	− 0.617	0.984	− 0.727	1.819	0.214	5.318
Cellulose	0.431	− 0.677	0.533	0.201	0.666	5.318
Hemicellulose	− 0.242	0.477	− 0.308	1.151	0.315	5.318
Ash content	0.946	− 0.632	0.859	980.669	1.18 × 10–9	5.318
Volatile matter	0.422	− 0.592	0.414	21.998	0.002	5.318
Fixed carbon	− 0.567	0.635	− 0.532	0.313	0.591	5.318
HHV	− 0.569	0.459	− 0.468	6.36	0.036	5.32
Thermal efficiency	1.000	− 0.808	0.973	2.739	0.137	5.318
Burning rate	− 0.808	1.000	− 0.858	0.415	0.537	5.318
Cooking power	0.973	− 0.858	1.000	4.333	0.071	5.318

NA: Not applicable.

Overall, low-sulfuric-acid-concentration briquettes exhibited higher thermal efficiency and cooking power, though the combustion rate was lower. Although the combustion rate increased with increasing acid concentration, thermal efficiency and cooking power decreased. This makes low-acid-treated briquettes desirable for household cooking and small-scale operations.

From Table 5, it is evident that there was no statistically significant difference in the averages of all acid concentrations for thermal efficiency, burning rate, and cooking power between the treated and untreated briquettes, as the p-value > 0.05 and F < F crit.

The thermal efficiency, burning rate, and cooking power for the samples are presented in Fig. 5 as mean ± standard deviation to account for uncertainties. The uncertainty in burning rate was the least, ranging from ± 0.03 to ± 0.09 g/min, indicating high precision, followed by thermal efficiency (± 0.79–2.05%), while cooking power (± 3.51–14.00 kW) presented the largest uncertainty, indicating overlap and low precision, thus reducing the statistical significance of small differences. The high uncertainty in cooking power also underscores that the exact magnitude of variation must be interpreted within the bounds of these propagated errors, much as the overall output in TBHS1 is more robust than in THBS3-THBS5. These error margins indicate that the observed performance differences in thermal efficiency, burning rate, and cooking power are reliable within the stated confidence intervals, but caution is warranted when comparing samples with overlapping uncertainty ranges.

Effect of sulfuric acid concentration on emissions

The emissions studied were carbon dioxide (CO₂), carbon monoxide (CO), and particulate matter (PM_2.5). The effect of H₂SO₄ at varying concentrations on CO emissions, CO₂ emissions, and PM_2.5 emissions is presented in Fig. 7.

Fig. 7.

Effect of sulfuric acid concentration on (a) CO emissions, (b) CO₂ emissions, and (c) PM_2.5 emissions for the briquettes.

In Fig. 7, the CO emissions for briquettes made from treated rice husk and pine sawdust ranged from 0.88 ± 0.03 g/MJd to 3.87 ± 0.04 g/MJd, with THBS1 having the lowest and THBS3 the highest. All briquettes exhibited low CO emissions of ≤ 3 g/MJd (Tier 5), in accordance with International Organization for Standardization (ISO) 19,867–1:2018, except for THBS3 and THBS5 briquettes. The CO emissions for THBS1 are lower than 1.12 ± 0.01 g/MJd for the untreated blend (THBS0). At 1%, these emissions are lower due to the preservation of cellulose integrity and the partial degradation of hemicellulose, which enhances cellulose and improves combustion. As the concentration increased to 2–3%, a sharp rise in CO emissions was observed, attributable to moderate acid hydrolysis that partially degrades cellulose and hemicellulose, thereby increasing volatiles and promoting incomplete combustion. When the acid concentration increased to 4%, CO emissions decreased due to structural modifications resulting from partial cellulose stabilization and lignin enrichment. However, at 5%, the increase in CO emissions is attributable to cellulose degradation and partial lignin modification, which reduce combustion stability. The CO emissions yielded coefficients of 0.688 with concentration, 0.682 with lignin, -0.451 with cellulose, -0.767 with ash content, 0.319 with fixed carbon, -0.215 with volatile matter, 0.221 with HHV, -0.962 with cooking power, -0.892 with thermal efficiency, and 0.806 with burning rate in the correlation analysis. This indicates that the increase in CO emissions is associated with increased acid concentration, lignin and burning rate, a slight increase in HHV and fixed carbon, and decreased cellulose, ash content, cooking power, and thermal efficiency, and a slight decrease in volatile matter.

From Fig. 7, it was also observed that CO₂ emissions ranged from 55.33 ± 2.52 g/MJd to 67.33 ± 2.08 g/MJd, generally higher than 55.67 ± 1.53 g/MJd for the untreated blends. These emissions increased at 1–3% acid concentrations due to increased oxidation of carbonaceous materials, due to access to cellulose and lignin arising from acid hydrolysis. As the concentration increased from 4–5% the, these emissions reduced due to over-hydrolysis, reducing the carbonaceous material available for combustion in lignin and cellulose. The CO₂ emissions yielded coefficients of 0.468 for acid concentration, 0.877 for lignin, -0.726 for cellulose, -0.576 for ash content, 0.497 for fixed carbon, -0.451 for volatile matter, 0.320 for HHV, -0.912 for cooking power, -0.800 for thermal efficiency, and 0.852 for burning rate in correlation analysis. This suggests that CO₂ emissions are associated with increases in acid concentration, lignin, burning rate, fixed carbon, and HHV, and with decreases in cellulose, cooking power, thermal efficiency, ash content, and volatile matter.

On the other hand, the figure shows that PM_2.5 ranged from 100.67 ± 1.53 mg/MJd to 679.00 ± 8.72 mg/MJd, with THBS1 the lowest and THBS5 the highest. THBS1 was in the Tier 3 range of ≤ 218 mg/MJd to > 62 mg/MJd, according to ISO 19,867–1:2018, while THBS2 and THBS4 were in Tier 2 (≤ 418 mg/MJd to > 218 mg/MJd) and THBS3 and THBS5 in Tier 1 (≤ 1030 mg/MJd to > 418 mg/MJd). The increase in PM_2.5 from 1–3% is due to the fragmentation of biomass materials during hydrolysis by acids, which increases the fine particles emitted during combustion. When the concentration increased to 4%, emissions reduced due to partial cellulose stabilization and improved compaction. However, at a 5% concentration, PM_2.5 emissions increased substantially due to excessive lignocellulose degradation, including cellulose and biomass fragmentation. The PM_2.5 emissions yielded coefficients of 0.906 with acid concentration, 0.388 with lignin, -0.140 with cellulose, -0.944 with ash, 0.234 with fixed carbon, -0.067 with volatile matter, 0.274 with HHV, -0.911 with cooking power, -0.932 with thermal efficiency, and 0.652 with burning rate in the correlation analysis. This shows that PM_2.5 emissions are associated with increases in acid concentration, burning rate, slight increases in lignin, fixed carbon, and HHV, and reductions in ash content, cooking power, and thermal efficiency, as well as slight decreases in volatile matter and cellulose.

The correlation coefficients for PM_2.5 were 0.904 with CO and 0.711 with CO₂, indicating that PM_2.5 emissions increase with increases in CO and CO₂ emissions. The correlation coefficient between CO and CO₂ was 0.934, indicating that CO increases with increasing CO₂.

ANOVA single-factor analysis, as shown in Table 6, indicated that the averages for all concentrations of treated and untreated briquettes are statistically significant for CO, CO₂, and PM_2.5 since p-value < 0.05 and F > F crit.

Table 6.

Pearson Correlation and ANOVA single-factor analysis for CO, CO₂ and PM_2.5 emissions.

	Pearson correlation coefficient (r)			ANOVA single-factor analysis
	CO	CO2	PM2.5	F	P-value	F crit
Concentration	0.688	0.468	0.906	NA	NA	NA
Lignin	0.682	0.877	0.388	1.819	0.214	5.318
Cellulose	− 0.451	− 0.726	− 0.140	0.201	0.666	5.318
Hemicellulose	0.178	0.488	− 0.093	1.151	0.315	5.318
Ash content	− 0.767	– 0.576	− 0.944	980.669	1.18 × 10–9	5.318
Volatile matter	− 0215	– 0.451	− 0.067	21.998	0.002	5.318
Fixed carbon	0.319	0.497	0.234	0.313	0.591	5.318
HHV	0.221	0.320	0.274	6.36	0.036	5.32
Thermal efficiency	− 0.892	– 0.800	− 0.932	2.739	0.137	5.318
Burning rate	0.806	0.852	0.652	0.415	0.537	5.318
Cooking power	− 0.962	– 0.912	− 0.911	4.333	0.071	5.318
CO	1.000	0.934	0.904	8.927	0.017	5.318
CO2	0.934	1.000	0.711	10.303	0.012	5.318
PM2.5	0.904	0.711	1.000	9.120	0.017	5.318

NA: Not applicable.

Overall, a 1% H₂SO₄ concentration yields cleaner briquettes with lower CO and PM_2.5 emissions; however, it also results in modest CO₂ emissions. Furthermore, CO, CO₂, and PM_2.5 emissions were balanced at an acid concentration of 4%. This indicates that, when using sulfuric acid to treat biomass blends for briquetting, the acid concentration must be carefully controlled, as lower concentrations promote better combustion, whereas excessive acid concentration compromises air quality by increasing toxic emissions (CO and PM_2.5). On the other hand, the focus on CO and PM_2.5 emissions without measuring sulfur-related impacts, such as sulfur and associated SO_x emissions, which also pose significant environmental and health concerns, limited the study. Therefore, future studies should incorporate the sulfur and associated emissions to provide a more comprehensive analysis of the sustainability of sulfuric acid-treated briquettes.

The measurement uncertainties for CO, CO₂, and PM_2.5 for the samples were presented as mean and standard deviation, as shown in Fig. 6. The uncertainty in CO emissions (± 0.03–0.07 g/MJd) was the smallest, indicating high precision and allowing clear differentiation, followed by CO₂ (± 1.53–2.52 g/MJd), while PM_2.5 (± 1.53–13.45 mg/MJd) was the largest, indicating low precision and overlap, thereby reducing differentiation between samples. The larger uncertainty in PM_2.5 also indicates that while the overall trend of higher particulate emission in THBS3 and THBS5 is robust, the precise magnitude must be interpreted within the bounds of these propagated errors. Therefore, the observed error margins for CO, CO₂, and PM_2.5 indicate that the observed differences are reliable within the stated confidence intervals, but caution is warranted when comparing samples with overlapping uncertainty ranges.

The briquettes under the 1% and 4% treatment conditions exhibit a clear trade-off between emissions and energy yield. At the 1% treatment condition, combustion is cleaner with CO of 0.88 ± 0.03 g/MJd and PM2.5 of 100.65 ± 1.53 mg/MJd, lower, accompanied by a higher thermal efficiency of 71.49 ± 2.05% and cooking power of 1001.67 ± 8.50 kW. On the other hand, the 4% treatment condition provides only an increase in the HHV of 18.82 ± 0.03 MJ/kg compared to 18.31 ± 0.02 MJ/kg at 1%, although this occurred at a cost of elevated emissions with CO of 2.56 ± 0.05 g/MJd and PM2.5 of 440 ± 6.08 mg/MJd, accompanied by reduced thermal efficiency of 50.70 ± 1.02% and cooking power of 612.00 ± 14.0 kW. For applications where reducing health-related pollutants and maximizing practical energy delivery are essential, the 1% treatment condition provides a more favorable balance; hence, it is recommended, since the increase in HHV at 4% does not surpass the decline in combustion quality for such applications. The 1% sulfuric acid concentration ensures minimal acid use, thereby reducing costs for acid acquisition, protective gear, and corrosion of burning equipment, thus increasing profitability. In addition, low-cost decentralized treatment systems, including buffering agents such as lime and ash, are recommended to reduce toxicity and corrosiveness, especially in rural settings, thereby enhancing profitability⁴³. However, the economic implications of the costs associated with sulfuric acid treatment, including acid acquisition, protective gear, and washing, and their impact on profitability, were not incorporated into this study, limiting its practical real-world application. It is recommended that future studies incorporate economic analysis to ensure practical, real-world application.

Effect of sulfuric acid concentration on mechanical properties

Mechanical properties investigated were breakage resistance and bulk density. The breakage resistance was determined using the breakage index. The effect of sulfuric acid at varying concentrations on bulk density and breakage index is presented in Fig. 8.

Fig. 8.

Effect of sulfuric acid concentration on (a) Bulk density and (b) Breakage index and.

In the figure, the breakage index obtained using Eq. (7), for briquettes made from a blend of treated rice husks and pine sawdust was lower, ranging from 0.27 ± 0.21% to 3.52 ± 2.23%, compared with that of untreated briquettes (7.36 ± 0.18%). Further, the breakage index decreases with increasing acid concentration, indicating the briquette’s ability to withstand shattering during transportation and use.

The high breakage index at a low concentration (1%) is due to insufficient hemicellulose degradation, which reduces interparticle bonding and produces weaker, more prone-to-failure briquettes. At moderate concentrations (2–3%), there is a sharp decline in drop resistance due to partial hemicellulose degradation and lignin modification, which enhance interparticle adhesion and compaction. At higher concentrations (4–5%), the breakage index decreased further due to complete hemicellulose hydrolysis and partial lignin modification, thereby enhancing structural integrity.

The correlation coefficients for the breakage index were -0.883 with H₂SO₄ concentration, -0.630 with lignin, 0.361 with cellulose, 0.885 with ash content, and -0.436 with HHV. This suggests that a high breakage index is associated with increasing ash content and cellulose, and with reduced sulfuric acid concentration, lignin, and thermal efficiency. The results showed a significant difference in the breakage index between the treated and untreated briquette blends (p = 0.000006; F = 111.470; F_crit = 5.318).

The bulk density of those from a blend of treated materials was higher, ranging from 603.47 ± 16.53 kg/m³ to 858.94 ± 32.37 kg/m³, compared with those from untreated materials, which were 571.66 ± 29.65 kg/m³. In addition, it generally increases with H₂SO₄ concentration, with a slight drop at 4% (THBS4). At low concentration (1%), the bulk density was low due to limited structural modification. When the concentration was increased to a modest level (2–3%), bulk density increased due to partial hemicellulose degradation and lignin modification, thereby enhancing interparticle bonding. The reduction at 4% concentration was associated with reduced lignin and overhydrolysis, leading to weakened cellulose. However, at 5%, bulk density increased due to improved particle distribution and reduced voids.

The correlation coefficients for the bulk density were 0.845 with H₂SO₄ concentration, 0.625 with lignin, 0.339with cellulose, -0.866 with ash content, and 0.298with HHV. This suggests that high density is associated with increased acid concentration and lignin, slightly higher HHV, and reduced ash content and HHV. A correlation between bulk density and breakage index yields a coefficient of -0.978, suggesting that as bulk density increases, breakage index decreases. Generally, there is a significant difference in the bulk density of the treated and untreated briquette blends (p = 0.002; F = 20.313 and F_crit = 5.318). Table 7 presents the Pearson correlation and the single-factor ANOVA for bulk density and breakage index.

Table 7.

Pearson correlation and ANOVA single-factor analysis for bulk density and breakage index.

	Pearson correlation coefficient (r)		ANOVA single-factor analysis
	Bulk density	Breakage index	F	P-value	F crit
Concentration	0.845	− 0.883	NA	NA	NA
Lignin	0.625	0.630	1.819	0.214	5.318
Cellulose	0.339	0.361	0.201	0.666	5.318
Hemicellulose	0.085	− 0.148	1.151	0.315	5.318
Ash content	– 0.866	0.885	980.669	1.18 × 10–9	5.318
Volatile matter	– 0.245	0.365	21.998	0.002	5.318
Fixed carbon	0.366	− 0.486	0.313	0.591	5.318
HHV	0.298	− 0.436	6.36	0.036	5.32
Thermal efficiency	– 0.930	0.934	2.739	0.137	5.318
Burning rate	0.877	− 0.901	0.415	0.537	5.318
Cooking power	– 0.943	0.913	4.333	0.071	5.318
CO	0.932	− 0.852	8.927	0.017	5.318
CO2	0.815	− 0.746	10.303	0.012	5.318
PM2.5	0.924	− 0.876	9.120	0.017	5.318
Bulk density	1.000	− 0.978	20.313	0.002	5.318
Breakage index	– 0.978	1.000	111.470	0.000006	5.318

NA: Not applicable.

The uncertainty in bulk density and breakage index was presented in Fig. 7 as means and standard deviations. The uncertainty in breakage index (± 0.15–2.23%) was the smallest, indicating relatively higher precision, thus clear differentiation compared to bulk density (± 6.89–65.85 kg/m³), which was larger. The overlapping error margin for bulk density for THBS3 (± 6.89 kg/m³) and THBS4 (32.37 kg/m³) suggests that the differences should be viewed with caution. The larger uncertainty for the breakage index of THBS1 (± 2.23%) broadens the reasonable range of values, reducing the strength of comparisons with a smaller error margin. These indicate that measurement uncertainties in bulk density and breakage index propagate into comparative analyses and trend evaluations, and that careful consideration of error ranges is essential for robust conclusions.

Conclusions

The study treated a blend of 10% rice husk and pine sawdust with H₂SO₄ at 1%, 2%, 3%, 4%, and 5% concentrations, and formulated and analyzed briquettes from the treated blends to identify the most suitable treatment condition. It can be concluded that H₂SO₄ treatment effectively reduced the ash content to 2.00 ± 0.64—3.53 ± 0.29%_db, from 11.57 ± 0.53%_db for the briquettes from a rice husk and pine sawdust blend, thereby improving fuel quality. The ash reduction increased with increasing acid concentration (r = -0.984). In addition, the treatment increased the heating value (HHV) to a range of 18.31 ± 0.02 MJ/kg to 18.82 ± 0.03 MJ/kg of briquettes from rice husk and pine sawdust blend, versus 18.21 ± 0.09 MJ/kg for briquettes from the untreated blend, with 4% concentration yielding the highest HHV. The HHV increased with increasing acid concentration (r = 0.463). Further, H₂SO₄ treatment enhanced cooking power and thermal efficiency with char of briquettes at 1% acid concentration to 1001.67 ± 8.50 kW and 71.49 ± 2.05%, versus 872.00 ± 0.00 kW and 49.95 ± 0.58% for briquettes from the untreated blend. However, as HHV increased with increasing acid concentration, both thermal efficiency (r = -0.893) and cooking power (r = -0.785) declined, suggesting that excessive acid treatment reduces the effective conversion of energy. Similarly, CO emissions reduced to 0.88 ± 0.03 g/MJd at 1% H₂SO₄ treatment versus 1.12 ± 0.01 g/MJd for briquettes from the untreated blend meeting the Tier 5 range (≤ 3 g/MJd) under the ISO 19,867–1:2018. While PM_2.5 emissions reduced to 100.67 ± 1.53 mg/MJd, compared with 103.33 ± 1.53 mg/MJd for briquettes from the untreated blend, which is within the Tier 3 range (≤ 218 mg/MJd to > 62 mg/MJd). However, acid treatment at > 1% increased CO emissions (r = 0.688) and PM_2.5 emissions (r = 0.906). Overall the 4% treatment condition resulted in a higher HHV (18.82 ± 0.03 MJ/kg versus 18.31 ± 0.02 MJ/kg) though associated with poorer combustion performance, including lower thermal efficiency (50.70 ± 1.02% versus 71.49 ± 2.05%), reduced cooking power (612.00 ± 14.00 kW versus 1001.67 ± 8.50 kW), higher CO emissions (2.56 ± 0.05 g/MJd versus 0.88 ± 0.03 g/MJd), and increased PM_2.5 emissions (440 ± 6.08 mg/MJd versus 100.65 ± 1.53 mg/MJd) compared to the 1% treatment condition. The 1% treatment condition, on the other hand, provided a more balanced outcome, integrating lower pollutant emissions with higher thermal efficiency, a stable burning rate, and greater cooking power. Therefore, the 1% treatment condition is optimal, ensuring both effective combustion and practical energy delivery, and is recommended for adoption in briquette production.

Author contributions

Ronald Walozi: Conceptualization, Methodology, Data Collection, Investigation, Resources, Funding Acquisition, Formal Analysis, Writing—Original Draft Preparation, Visualization & Project Administration. George Samuel Onep : Supervision, Conceptualization, Formal Analysis, Methodology, Writing—Review & Editing. Yinka Sofihullahi Sanusi : Supervision, Visualization, Writing—Review & Editing. Omer Eisa Babiker : Conceptualization, Supervision, Writing—Review & Editing. Milon Selvam Dennison : Supervision, Conceptualization, Formal Analysis, Writing—Review & Editing. Alphonse Candia : Conceptualization, Fund Acquisition, Data Curation, Writing—Review & Editing. Samuel Okurut : Methodology, Fund Acquisition, Data Curation &Writing—Review & Editing. Benard Wamubirigwe: Methodology, data curation, and Writing— Review & Editing.

Data availability

Research data is contained in the manuscript, and any additional data will be available on request.

Declarations

Competing interests

The authors declare no competing interests.