Assessing thermal hazards and toxicity of raw biomass particles from prevalent agricultural crops in China

Abstract

Fire and explosion hazards pose significant safety concerns in the processing and storage of biomass particles, warranting the safe utilization of these particles. This study employed scanning electron microscopy, thermogravimetric analysis, and cone calorimetry to investigate the thermal hazards and toxicity of raw biomass particles from four prevalent agricultural crops in China: rice, sorghum, corn, and reed. Among the samples, corn exhibited the highest heat output of 8006.82 J/g throughout the thermal decomposition process. The quantitative evaluation of critical heat flux, heat release rate intensity, fire growth rate index (FIGRA), post-ignition fire acceleration (PIFA) and flashover potential (X) revealed a substantial fire risk inherent to all the examined straw samples. Notably, corn displayed the lowest FIGRA value of 8.30 kW/m² s, while rice demonstrated the minimum PIFA value of 16.11 kW/m² s. Moreover, the X values for all four biomass particle types exceeded 10 under varying external heat flux levels, indicating their high propensity for fire hazards. Analysis of CO and CO₂ emissions during combustion showed all four biomass samples exhibited high concentrations throughout, from the initial stages to the end. The present study offers crucial insights for formulating comprehensive fire safety guidelines tailored to the storage and processing of biomass particles.

Introduction

As fossil fuel resources are gradually depleting and their detrimental environmental impacts are more evident day after day, there is an urgent need to establish and implement alternative, clean, and renewable energy resources¹. In this regard, biomass energy, as one of the clean and renewable energy resources, has received wide interest, and multiple countries worldwide have established plans and development initiatives towards promoting the adoption of biomass resources to satisfy energy needs². Biomass particles, which are produced from agricultural and forestry waste, are a key component in the biomass energy domain. They are characterized by their low cost, storage convenience, and ease of transportation³. Crop straw is considered a fundamental feedstock for biomass particles; however, it is frequently not fully utilized and incinerated⁴. When used as the basis for biomass energy, straw materials require a grinding process to reduce their size and increase their respective density⁵.

With the worldwide development in the production of biomass fuels and the utilization of these fuels as a major energy supply source, the occurrences of self-heating⁶, dust explosion⁷ and off-gassing⁸ during their storage and transportation processes have catastrophic consequences. This results in major damage to facilities and even poses risks to the personal safety of workers involved in various stages and processes. In recent years, significant incidents have been reported related to bioenergy⁹. These are summarized in Table 1^10,11.

Table 1 Events related to biomass disasters in recent years.

A large number of researchers have investigated different aspects related to biomass fire. For instance, Li et al¹². investigated the effect of straw stack thickness and size by exploring the burning characteristics of straw. They concluded that the rate of mass loss is majorly affected by the thickness and diameter of straw. Àgueda et al¹³. investigated the effect of the slope- or wind-aided on the spread of flame by employing straw. They established an equation relating the wind velocity, slope of the combustion table, and flame spread rate in their study. In another study, Meng et al¹⁴. employed a fixed-bed reactor to numerically simulate and carry out experiments considering corn stalks of different lengths. They reported that corn stalks with shorter lengths have shorter ignition times. While previous studies have explored the behavior of straw fires, they have primarily studied and examined intact straw rather than fragmented biomass particles¹⁵. Compared to whole biomass fuels, fragmented biomass comprises small and loose particles that are highly flammable and burn vigorously¹⁶. In their work, Restuccia et al¹⁷. used wheat biomass particles as the research object and analyzed the self-heating ignition of biomass as influenced by particles size. The analysis revealed no significant influence of particles diameter variation on self-heating ignition behavior. Liu et al¹⁸. provided guidance regarding the safe storage of biomass particles by investigating the effects of the stacking density and ventilation characteristics of corn straw particles on the smoldering process. It was concluded that the rate of smoldering propagation velocity decreased with the increase in stack density. In another study, Huang et al¹⁹. conducted experiments on oil-rich biomass particles using a TG-DSC-DTG analyzer, and concluded that rapeseed oil residue exhibited the highest susceptibility to spontaneous combustion. In contrast to previous studies that primarily focused on the combustion characteristics of intact biomass, this paper delves deeper by examining the combustion profiles and potential toxicity associated with biomass pellets derived from prevalent agricultural crops in China. This targeted approach offers valuable insights for practical applications.

Furthermore, China holds the distinction of being the world's largest producer of agricultural crop straw, contributing roughly one-fifth of the global straw resources²⁰. The “National Comprehensive Utilization Report of Agricultural Crop Straw” reported that China has realized a total comprehensive utilization of agricultural crop straw of 647 million metric tons²¹. In particular, gramineous crop residues are one of the largest agricultural wastes in China, thus providing great potential for resource further utilization²². Accordingly, in contrast to prior research that may have utilized randomly chosen biomass samples, this study strategically focuses on dominant agricultural biomass pellets cultivated in China. This targeted selection enhances the applicability and practical relevance of the findings by directly addressing the specific biomass resources most commonly encountered in the region. Given the limited data on the thermo-oxidative degradation and combustion characteristics of straw particles, this study aims to elucidate the underlying mechanisms and intensity profiles associated with straw combustion processes. To achieve this, a multifaceted approach utilizing various thermal analysis techniques was employed. The analysis considers multiple aspects, including elemental composition, microstructure, pyrolysis behavior, and fire characteristics. This includes identifying potential ignition temperature, assessing the combustibility of the pellets, and analyzing the composition of toxic gases released during combustion. Through quantitative analysis, this study provides crucial guidance for the development of strategies aimed at preventing and controlling potential fires that may occur during the storage, transportation, and processing stages of biomass particles.

Materials and methods

Samples

In this study, four types of gramineous agricultural straw biomass particles, namely rice, sorghum, corn, and reed, were considered for investigation. The four straw types were collected from Yichun City, Jiangxi Province, China. Prior to the experimental procedures, the four different straws underwent milling and sealing treatments to minimize the potential influence of sample geometry, particle size, and moisture content on the results (Fig. 1). To enhance experimental accuracy and data reliability, the entire experimental protocol was independently replicated four times under identical conditions.

Figure 1

Four types of biomass particles.

Apparatus

Scanning electron microscope

A scanning electron microscope (ZEISS GeminiSEM 500) was utilized to investigate the microphysical structure of four types of straws at magnifications of 100× and 1000×. The samples were fragmented, adhered to a conductive adhesive, and sputter-coated with a thin gold layer before examination under an acceleration voltage of 1 kV.

Thermogravimetry analysis

A TG-DSC-DTG analyzer (Netzsch SAT 449 F3 Jupiter) was utilized in a controlled atmosphere of dry air to examine the thermal stability of the four types of biomass particles considered. The temperature of the samples was increased from room temperature to 800 °C at a rate of 10 °C per min under a concurrent flow of dry air at 50 mL per min. For each test, a sample mass of 10 mg of the material was employed. Furthermore, an experiment was performed with an empty crucible to establish a baseline correction for the TG analysis.

Cone calorimeter

To investigate the combustion characteristics of the four straw types, a cone calorimeter (Fire Testing Technology, UK) was employed following ISO 5660 standards. Samples were exposed to external heat flux densities of 45, 50, 55, and 60 kW/m². For each test, 10 mg of sample was evenly spread in a 10 × 10 cm² tin foil holder, conforming to the ISO 5660 standard area. The key combustion parameters recorded by the cone calorimeter were ignition time, heat release rate, and the concentrations of CO and CO₂ in the combustion gases. To ensure data consistency and comparability, the scanning speed was standardized to 1 scan/s.

Infrared camera

During the experiments, an infrared camera (ThermaCAM, FLIR Systems, USA) was positioned at a distance of 1.5 m from the combustion furnace to monitor the flame temperature of the biomass particles. The camera has an infrared image acquisition frequency of 1 frame/s and a thermal sensitivity of 0.08 °C.

Calculations

Thermal decomposition parameters

A quantitative assessment of the characteristics of the oxidation decomposition of agricultural biomass particles was carried out. To aid this, parameters such as the initial decomposition index (${C}_{i}$)²³, the volatile matter release index (${D}_{v}$)²⁴, the final decomposition index (${C}_{f}$)²³, and the comprehensive decomposition index ($S$)²⁵ are introduced. The corresponding equations are listed below.

$${C}_{i}=-{m}_{p1}/({t}_{p1}\times {t}_{i})$$

(1)

In Eq. (1), ${C}_{i}$ represents the potential for self-ignition of the four types of biomass particles. $-{m}_{p1}$ represents the maximum mass loss rate and ${t}_{p1}$ denotes the corresponding time of the first peak observed in the DTG curve. Also, the initial decomposition time, denoted as ${t}_{i}$ was derived from the TG-DSC-DTG software analysis.

$${D}_{v}={-m}_{\text{p}2}/({T}_{\text{d}}\times {T}_{\text{p}2}\times {\Delta T}_{1/2})$$

(2)

In Eq. (2), ${D}_{v}$ represents the flammable gas released during the pyrolysis of biomass particle. $-{m}_{p2}$ represents the maximum mass loss rate observed during the second peak in the DTG curve. ${T}_{\text{d}}$ is the initial temperature at which the volatile compounds begin to be released, while ${T}_{\text{p}2}$ corresponds to the temperature at which the second DTG peak becomes apparent. Additionally, $\Delta {T}_{1/2}$ denotes the temperature interval over which half of the maximum mass loss rate value is attained.

$${C}_{f}=-{m}_{\text{p}3}/({\Delta t}_{1/2}\times {t}_{\text{p}3}\times {t}_{\text{e}})$$

(3)

Similarly, ${C}_{f}$ in Eq. (3) is a parameter characterizing the biomass particle burnout performance. $-{m}_{p3}$ representsthe maximum mass loss rate observed at the third peak in the DTG curve. In this case, ${\Delta t}_{1/2}$ denotes the time interval at which half of the maximum mass loss rate is attained. Also, ${t}_{\text{p}3}$ is the time at which the third DTG peak occurs, while ${t}_{\text{e}}$ is the time at which the decomposition was completed.

$$S=\left[({-m}_{\text{p}4})\times ({-m}_{\text{av}})/\left({T}_{\text{i}}^{2}\times {T}_{\text{e}}\right)\right]$$

(4)

Moreover, In Eq. (4), $S$ is a parameter characterizing the degree of difficulty for the biomass particle to complete the whole decomposition process. ${-m}_{\text{p}4}$ and ${-m}_{\text{av}}$ denote the maximum and average rate of mass loss of straw throughout the pyrolysis process, respectively. ${T}_{\text{i}}$ represents the initial decomposition temperature, was determined through the analysis conducted using the TG-DSC-DTG software, and ${T}_{\text{e}}$ is the temperature at which the decomposition process was completed.

Critical heat flux

Critical heat flux (CHF) refers to the highest heat flux density a material’s surface can tolerate before combustion occurs. Adhering to the principle of thermal equilibrium, the CHF absorbed by a material is equivalent to the heat dissipated from its surface to the surrounding environment²⁶. The formulation is:

$$CHF=\frac{1}{\varepsilon }\left[{h}_{c}\left({T}_{ig}-{T}_{0}\right)+\varepsilon \sigma {T}^{4}\right]$$

(5)

In Eq. (5), ε is the emissivity of the sample, ${h}_{c}$ is the convective heat transfer coefficient, ${T}_{ig}$ is the ignition temperature, ${T}_{0}$ is the ambient temperature, and σ is the Stefan–Boltzmann constant.

For thermally thin materials¹⁵, there is an inverse relationship between the external heat flux density (${q}_{rad}$) and the ignition time (${t}_{ig}$), based on previous research^27,28.

$${t}_{ig}^{-1}=\frac{\varepsilon }{\rho c{L}_{0}({T}_{ig}-{T}_{0})}{q}_{rad}-\frac{{h}_{c}({T}_{ig}-{T}_{0})+\varepsilon \sigma {T}^{4}}{\rho c{L}_{0}({T}_{ig}-{T}_{0})}$$

(6)

In the above equation, $\rho$ is the density of the sample, $c$ is the specific heat capacity of the biomass particles, and ${L}_{0}$ is the sample thickness.

Fire hazard assessment

The biomass particle fire hazards are characterized using parameters such as the fire growth index ($\text{FIGRA}$)²⁹, the flashover potential ($\text{X}$)³⁰, and the post-ignition fire acceleration (PIFA)³¹.

FIGRA, is useful in the prediction of the fire growth rate as well as the fire hazard. It provides a basis for accurate fire modeling, preemptive hazard assessment, and emergency response planning. It is defined as the ratio of the peak heat release rate ($PHRR$) to the time needed to attain the peak ${(T}_{p})$ and can be expressed as follows:

$$FIGRA=PHRR/{T}_{P}$$

(7)

$\text{X}$, is a parameter utilized to quantify the transition of a fire into a “flashover” state. As the fire progresses to the “flashover” stage, it undergoes rapid propagation. This, in turn, leads to a significant escalation in the fire severity level. From a conceptual perspective, X is defined as the ratio of the $PHRR$ to the ignition time (${T}_{i}$) and can be represented in Eq. (8).

$$X=PHRR/{T}_{i}$$

(8)

PIFA, is a parameter representing the sustained growth rate of the flame after a fire is ignited. PIFA characterizes fire speed and intensity during the early stages of development. Thus, the evaluation of this parameter is critical for understanding the behavior of the fire and serves as a basis for taking timely fire extinguishing measures. PIFA can be represented by Eq. (9).

$$PIFA=PHRR/\left({T}_{P}-{T}_{i}\right)$$

(9)

Results and discussion

Elemental analysis

Results of the ultimate and proximate analyses, along with standard deviations, conducted on the four types of biomass particles are listed in Table 2. The ultimate analysis data are obtained from the CHNS Elemental Analyzer (ELEMEATAR, GER) according to ASTM D5373³². The proximate analysis data were obtained by analyzing the thermal gravimetric curves according to ASTM D7582³³. As shown in the table, the four biomass types have a relatively low moisture content (7.88~10.38%) and a high volatile matter content (74.17~85.89%). These characteristics highlight that the four types of biomass particle have a comparatively low moisture content. This, in turn, indicates favorable conditions for swift ignition and heat release. Additionally, it was found that the four biomass types have reduced levels of ash, nitrogen, and sulfur content compared to coal³⁴. In this regard, the combustion of these agricultural straws will lead to decreased emission of harmful gases such as NO_x and SO_x. Overall, they can be considered more environmental friendly compared to other energy resources.

Table 2 Ultimate and proximate analyses of the four types of agricultural straws.

Microstructures

In this study, a scanning electron microscopy (SEM) was used to examine the microstructure of the four types of biomass particles. As shown in Fig. 2a, under a 100× microscope, rice exhibited the largest conduit pore size, whereas reed had the smallest, which can be attributed to the differences in their physical structure and the respective nutrient absorption capacity³⁵. In addition, at 1000× magnification, intricately arranged elevated structures were observed at the surface of the rice straw. These structures can be attributed to the deposition of silicic acid within leaf epidermal cells³⁶. The silicified cell formation can increase the thermal stability of the biomass, potentially delaying the onset of pyrolysis but ensuring a more sustained release of volatile compounds. This can lead to a more controlled combustion process with fewer fluctuations in the combustion rate³⁷. Sorghum have a smooth surface with elongated, flat bumps in addition to a dense, compact structure. These characteristics are due to the interaction between cellulose, hemicellulose, and lignin at the cell wall³⁸. This will stabilize the combustion of this biomass, as illustrated by the TG-DSC-DTG curves, which show relatively minimal fluctuation. Furthermore, the surface of the corn exhibited a distinctive groove-like texture. This is primarily related to the arrangement of its epidermal cells and the formation of cell walls during its growth process³⁹. The unique structure facilitates enhanced heat transfer and gas emission, leading to unstable combustion patterns, as depicted in the DSC curve. For reed straw, vertical pores were observed on its surface in relation to the organization of its ducts⁴⁰, leading to a quick release of volatiles and a fast pyrolysis process. This is reflected in the TG-DTG curves by distinct stages of mass loss, sharp peaks in the rate of weight loss. To determine the particle size distributions of the four samples, the statistical analysis software Nano-Measure was utilized. Particle size distributions of four types of biomass particles, including their standard deviations, are shown in Fig. 2b. Approximately 80% of the biomass particles were distributed within a diameter range of 90.31 to 158.65 μm. Mean diameters for Rice, Sorghum, Corn, and Reed were 126.82, 133.38, 118.90 and 115.35 μm.

Figure 2

(a) SEM images and (b) particle size distributions of four types of biomass particles.

TG-DSC-DTG

Understanding the pyrolysis behavior of biomass particles is crucial for assessing fire risks, particularly in industrial and agricultural applications. The thermal degradation characteristics of biomass can significantly influence their flammability and combustion properties. This section discusses the pyrolysis behavior of four different biomass particles—rice, sorghum, reed, and corn—as observed in the TG-DSC-DTG analysis, and explores their implications for fire risk assessment.

During the initial volatilization phase, from 30 to 100 °C, all four biomass particles underwent moisture evaporation and release of volatile compounds. As shown in Fig. 3, the TG-DSC-DTG curves indicate a similar trend for all samples. Rice displayed the most significant mass loss rate at 1.80% per min, while sorghum showed the least at 1.17% per min, followed by a subsequent stabilization period. This phase primarily involves the loss of free water and some loosely bound water, which can affect the ignition temperature and initial flammability of the biomass. At a higher temperature range of 230 to 450 °C, they all experienced notable pyrolysis. During this stage, the cellulose, hemicellulose, and lignin underwent decomposition, leading to a substantial reduction in the overall weight. Among them, the reed exhibited the highest weight loss rate, with 10.47% per min at a temperature of 311 °C, while sorghum continues to display the least with 1.17% per min at a temperature of 280 °C. The DSC curves show two peaks, representing the respective thermal decomposition of cellulose and lignin in straw⁴¹. Thus, corn possesses the greatest lignin/cellulose ratio compared to the other three biomass particles, displaying a pronounced peak within the temperature range of 400 to 450 °C. The differences in weight loss rates and decomposition temperatures highlight the varying thermal stabilities and combustion properties of the biomass particles. Reed’s higher weight loss rate suggests a more aggressive pyrolysis reaction, leading to higher combustion intensity. The thermal decompositions of the four straw types are shown to be essentially complete within the temperature range of 450 to 800 °C. This stage involves the final breakdown of residual carbonaceous materials, leaving behind ash and other inert residues. Based on the results, the exothermic behavior evaluation can be carried out by integrating the DSC curve obtained⁴². It was shown that throughout the thermal decomposition process, corn produced the highest heat output of 8006.82 J/g, followed by reed (7470.57 J/g), rice (7295.71 J/g), and finally sorghum (7034.04 J/g). The pyrolysis behavior observed in the TG-DSC-DTG analysis provides valuable insights into the fire risk associated with each type of biomass.

Figure 3

TG-DSC-DTG curves of the four types of biomass particles in dry air.

Table 3 lists the thermal decomposition parameters along with standard deviations, obtained for the four types of biomass particles using the methods described in Section "Thermal decomposition parameters". Among them, sorghum is found to have the highest ${C}_{i}$ value. This indicates its superior flammability characteristics compared to the other types. On the other hand, reed is characterized by the highest ${D}_{v}$ value, suggesting a higher tendency for explosion during decomposition. This suggests that sorghum is well-suited for energy production processes that benefit from a quick release of energy. Nevertheless, this also implies that managing and processing sorghum necessitates strict safety protocols to avert uncontrolled combustion or explosions.

Table 3 Parameters of four types of biomass particle decomposition characteristics.

Thermal hazards analysis

Critical heat flux

CHF is a critical parameter that indicates the threshold at which thermal runaway occurs under the influence of heat flux. The variation in CHF values among different biomass particles suggests that their sensitivity to external heat flow varies. The CHF values for the four types of biomass particles—rice (10.22 kW/m²), sorghum (22.61 kW/m²), corn (19.65 kW/m²), and reed (5.28 kW/m²)—determined using the methods described in Section "Critical heat flux", are shown as the intercept values on the X-axis in Fig. 4. The analysis of these findings highlights that reed has the highest sensitivity to external heat flow. This is primarily related to the reed straw has low moisture content and its fragmented and porous structure. On the other hand, the results showed that sorghum has the highest thermal flux demand for combustion. This could be due to its elevated moisture content and the corresponding unique composition of the cell wall that hampers the combustion process, as reported in Sects. (3.1) and (3.2).

Figure 4

Linear fitting to obtain the CHF of the four types of biomass particles.

Heat release rate intensity

Heat release rate intensity (HRRI) is a fundamental measure of fire combustion intensity, defined as the rate of heat release per unit area. Figure 5 provide valuable insights into the combustion behavior and overall fire risk associated with different biomass particles under various heating conditions. For all materials, the HRRI exhibits a sharp peak around 20 s, corresponding to the rapid decomposition and volatilization of cellulose and hemicellulose components during the initial pyrolysis stage. This rapid release of volatiles leads to a sudden spike in the heat release rate. A higher peak HRRI value indicates a more intense and rapid volatilization, potentially leading to flashover conditions and increasing the risk of rapid fire spread or explosion, making the biomass more hazardous. After the initial peak, the HRRI curves show a decline, attributed to the slower pyrolysis of the more resistant lignin component. A slower decline in HRRI may imply a longer-lasting release of volatiles, potentially prolonging the fire duration and increasing overall heat exposure. The peak HRRI values and curve shapes vary across biomass types, likely due to differences in their compositions and ratios of cellulose, hemicellulose, and lignin. This information aids in assessing the relative fire risks associated with different biomass sources.

Figure 5

Transient variation of HRRI of different biomass particles under different external heat fluxes.

Additionally, higher external heat flux levels (e.g., 60 kW/m²) generally lead to higher peak HRRI values and more pronounced initial spikes, indicating more intense and rapid volatilization reactions. This sensitivity to external heating conditions can exacerbate fire risk in scenarios with higher heat exposure or preheating, such as in wildland fires or industrial accidents.

Furthermore, the linear fitting of the peak HRRI (PHRRI) values of the four types of biomass particles to the external heat flux values is presented in Fig. 6a. Based on the results, sorghum exhibited the most significant variation in the PHRRI value with a change in the external heat flux. Reed has the maximum PHRRI value. A higher PHRRI indicates a more intense fire, which can be more challenging to control and extinguish. Figure 6b presents the variation in the total heat release (THE) of the four types of biomass particles subjected to different heat flux intensities. Incomplete combustion can lead to the production of smoke and toxic gases, adding to the fire’s hazard. As shown in the figure, all four types of biomass particles underwent a critical transition from incomplete combustion to complete combustion when exposed to an external heat flux of 55 kW/m².

Figure 6

(a) Variation of PHRRI with different external heat flux, (b) total heat release of the four types of biomass particles.

Fire hazard assessment

In this section the biomass particle fire hazards are characterized using parameters such as the fire growth index ($\text{FIGRA}$)²⁹, the flashover potential ($\text{X}$)⁴³, and the post-ignition fire acceleration (PIFA)³¹.

As shown in Fig. 7, all three parameters linearly increases with increasing external heat flux. In particular, under different external heat flux levels, corn has the lowest $\text{FIGRA}$ value of 8.30 kW/m² s (Fig. 7a), indicating a relatively slower fire growth rate compared to the other biomass types studied. While rice exhibits the lowest $\text{PIFA}$ value of 16.11 kW/m² s (Fig. 7b). These relatively high values reported suggest that the four types of biomass particles investigated in this study are characterized by a greater fire hazard compared to wood particles⁴⁴ According to the Petrella rating system⁴³, the $\text{X}$ values of the four types of biomass particles exceeded 10 under different external heat flux levels (Fig. 7c). This indicates that these four samples have a high fire hazard and and necessitate appropriate precautions and safety measures when handling and utilizing these biomass fuels. The linear increasing trends of these fire hazard parameters with external heat flux highlight the sensitivity of the biomass materials to heating conditions, potentially exacerbating the fire risk in scenarios with higher heat exposure or preheating.

Figure 7

Thermal fire hazard assessment parameters: (a) FIGRA; (b) PIFA; (c) X.

Flame temperature

Figure 8 presents the flame morphology (Fig. 8a), the evolution of the flame temperature (Fig. 8b), and the maximum flame temperature (Fig. 8c) of the four types of biomass particles when subjected to a heat flux of 55 kW/m².

Figure 8

(a) Thermal infrared images, (b) temperature variations, and (c) maximum temperature values of the four types of biomass particles.

Figure 8b presents the variation in the combustion temperature of the four types of biomass particles. The flame temperature undergoes an initial peak at the initiation of combustion, followed by a swift decrease and subsequent stabilization, aligning with the observed pattern in the HRRI curves elucidated in Section "Heat release rate intensity". In addition, Fig. 8c shows the maximum temperatures of the four samples. Based on the reported results, rice consistently exhibits significantly lower temperatures compared to the remaining three samples. This indicates that, among the four types of biomass particles, rice has the lowest potential to cause thermal hazards.

Toxicological analysis

Concentrations of CO and CO₂ can provide key information about its combustion performance and safety characteristics⁴⁵. As Fig. 9 shows, the biomass particles combustion sufficiency was positively correlated with the radiation intensity. Inadequate combustion early on resulted in increased CO concentration. As combustion progressed, the carbon in the biomass particles was progressively oxidized to form more CO₂, accompanied by decreasing CO concentration. In addition, the reported CO and CO₂ concentrations of reed straw are consistently the highest among the four types of biomass particle at different heat fluxes. This is associated with the highest carbon content of reed, as shown in Sect. (3.1).

Figure 9

Concentrations of CO and CO₂ under different external heat fluxes.

Furthermore, the ratio of the CO concentration to the CO₂ concentration is a critical parameter in the assessment of the combustion product toxicity. The higher the CO/CO₂ ratio, the higher the potential hazard to the environment and human health. Figure 10 presents the variation of the CO/CO₂ ratios of the four types of biomass particle at different heat fluxes. The analysis revealed that all four biomass particles exhibited relatively high CO/CO₂ ratios during both the initial and final stages of combustion. Among them, reed displayed the highest CO/CO₂ ratio across various heat flux levels.

Figure 10

CO/CO₂ ratios of the four types of biomass particles under different external heat fluxes.

Conclusion

In this paper, a set of devices, including TG-DSC-DTG, cone calorimeter, and infrared camera, were used to investigate the thermal hazards and toxicity of biomass particles from four prevalent agricultural crops in China. The results of the TG-DSC-DTG curves revealed a consistent thermal oxidation decomposition pattern for all biomass particle. Additionally, it was shown that corn exhibited the highest heat output of 8006.82 J/g throughout the entire thermal decomposition process, followed by reed (7470.57 J/g), rice (7295.71 J/g), and sorghum (7034.04 J/g). Moreover, under different radiation heat fluxes in a cone calorimeter, the obtained CHF values of rice, sorghum, corn, and reed are 10.22, 22.61, 19.65, and 5.28 kW/m², respectively. A more comprehensive analysis of the fire hazard associated with biomass particles were conducted, utilizing parameters such as FIGRA, PIFA, and X. It was found that even at the lowest heat flux intensity, all parameters of the four types of biomass particles were significantly higher than those of wood particles. Furthermore, the evaluation of the CO/CO₂ ratio further indicates that these biomass materials produce higher levels of CO during the initial and final stages of combustion. In particular, reed combustion has the highest toxicity and requires additional attention.

Future research should include larger sample sizes and a wider variety of agricultural biomass particles to develop a comprehensive understanding of pyrolysis and combustion behaviors across different materials. Conducting scale-up experiments in conditions that mimic real-world fires will validate laboratory findings and provide practical insights into these materials' combustion behaviors. Incorporating advanced analytical techniques such as gas chromatography-mass spectrometry (GC–MS) and Fourier-transform infrared spectroscopy (FTIR) can offer detailed chemical analyses of combustion by-products, identifying specific toxic compounds. Investigating the long-term environmental impact of biomass particle combustion, including effects on air quality and soil health, is crucial for sustainable agricultural practices. Additionally, developing and testing safety management protocols tailored to the specific hazards associated with different biomass particles can improve fire prevention and response strategies. The study’s findings can be applied to perform accurate risk assessments for facilities storing or processing biomass particles, enabling targeted mitigation strategies to reduce fire hazards. Fire safety training programs for agricultural workers and first responders can be enhanced with specific information about the combustion characteristics and toxicity of different biomass materials. Revised guidelines for the storage and handling of biomass particles can be developed based on the identified thermal and toxic properties.