Dust and Particulate Matter Generated during Handling and Pelletization of Herbaceous Biomass: A Review

Abstract

Using straw and herbaceous crops to replace or augment fossil fuels is becoming popular as access to forest biomass becomes environmentally stricter and more expensive. The low bulk density raw biomass is pre-processed and densified into pellets to facilitate handling and use. Dust is generated during collection, baling and debaling, grinding, drying, and densifying processed herbaceous biomass. Abundant literature deals with dust generated during the industrial handling of woody biomass, grains, and other crops like cotton. But the information on handling herbaceous biomass in the open literature is scarce. This paper reviews the available literature on dust generation from handling and processing of herbaceous biomass. Limited available data shows that herbaceous biomass species have a lower ignition temperature than woody biomass. The paper identifies several crucial pieces of information needed to ensure safety in the handling and pelleting of herbaceous crops.

1. Introduction

The increase in global demand for pellets and the limited availability of wood resources shift the focus towards herbaceous biomass utilization and strategies to promote herbaceous biomass pelletization technology. Many researchers experimented with numerous feedstocks to produce high-quality pellets from herbaceous biomass, as summed up by Stelte et al. [1], showing enormous interest in this sector.

A recent joint study by Agriculture and Agri-Food Canada (AAFC) and Natural Resources Canada (NRCan) [2] estimates the annual agricultural biomass produced in Canada at 80 Mt (million oven dry Mg). This number includes approximately 50 Mt of grains/seeds and about 30 Mt of crop residue. Roughly 71 Mt of annual forestry residue and 9.4 Mt of municipal waste can also be included in the inventory of biomass in Canada [2]. Wheat and barley are grown in all agricultural areas of Canada. Corn has a high yield and is primarily grown in Ontario and Quebec. Moving the industry forward by producing pellets from excess crop residue provides additional income for the farmer. Storage and handling of a pelletized biomass are much more accessible, safer, and cheaper than the activities for other forms of solid biomass.

After harvest, herbaceous biomass is baled and transported to a pellet plant where multiple steps take place: bale receiving, drying, grinding, pelletization, cooling and screening, packing and storage. The block diagram in Figure 1 shows the flow of herbaceous biomass from the field to the pellet plant. Dust is generated at every step of the process leading to the gradual accumulation of dust on the ceiling and other parts of the infrastructure and equipment, forming a dust layer. In 1997, measurements conducted on three alfalfa dehydrating plants in Alberta indicated that they emitted particulate matter at higher than 0.60 g of particulates/kg of effluent [3]. Alfalfa dehydrating plants produce alfalfa pellets and cubes primarily for export to Asia Pacific countries. Alberta Environmental Protection standards for particulates emission are (a) 0.60 g particulates/kg of air for urban areas with a population of less than 50,000 or rural areas; or (b) 0.20 g particulates/kg of air for urban areas with a population of more than 50,000 [4]. A study in Australia [5] indicated that farmworkers were exposed to high grain and soil dust levels for 12–16 h shifts per day during harvest. It was estimated that 22% of workers inhaled elevated dust concentrations of 2 mg/m³ over 12 h, and 40% were exposed to higher levels than the adjusted standard of 1 mg/m³ if they worked a 16-h day.

A pellet becomes the source of dust during processing and moments after it exits the pellet mill. The dustiness depends upon feedstock composition, production process, material handling and post-production handling of pellets. Chawla et al. [6] reported that dust is generated by the drying, grinding, and pneumatic conveying of alfalfa particles within alfalfa dehydrating plants. It was also reported that an essential source of dust in the alfalfa dehydrating plant is the hammer mill grinder, wherein the grinds collection cyclone suctions the ground alfalfa. Dust is highly combustible, leading to explosions and imposing health hazards [7]. Recently, Drax, the UK’s largest power station in North Yorkshire, faced criminal prosecution concerning employees’ health and safety due to dust exposure from wood pellets used to generate electricity [8]. Fire, dust explosion, respiratory, and harvest hazards associated with wood have been studied to minimize dust avoidance and particulate matter emission [9,10]. Herbaceous biomass dust and its effect on human health and safety, on the other hand, has not been researched extensively. Most of the studies on agricultural dust are either associated with dust generated in confined livestock buildings or grain and cotton processing centers. Several studies discuss the health aspects [11] and combustibility [12] of dust in cotton ginning plants. A recent study in Denmark [13] on the use of straw bales for heat and power application overlooks the seriousness of straw dust to health and safety. Yoder et al. [14] concluded that countries with established agricultural biomass systems had not researched safety hazards that focus on herbaceous biomass. Out of thirty articles identified through the Google search engine, Penn State University Library, and the American Society of Agricultural and Biological Engineers (ASABE) archives, the authors pointed out that no papers were explicitly found addressing the range of hazards associated with planting, production, and pre-processing of biomass [14]. In an earlier publication, Yoder et al. [15] developed a matrix in which the events “fires and explosions” were associated with harvesting and threshing activities.

This review highlights the available published data on dust generation due to harvest and postharvest operations, including pelletization of herbaceous biomass. The research focuses on health hazards due to dust and particulate matter and potential fires and combustion from handling dusty materials. The handbook published by Obernberger and Thek [16] reviews wood dust associated with wood pellet production, handling, and other safety concerns. Dust generation during baling and bale handling (Section 3.1), grinding operation (Section 3.2), and dust from pellets themselves (Section 3.3) for herbaceous biomass, as shown in Figure 1, are covered in this review. The available literature on dust generation from herbaceous biomass pelletization only focuses on the previously mentioned operations.

2. Dust Characterization and Classifying Methods

The size threshold to call it “dust” is industry-based, and it is hard to find a definition that everybody agrees on. This paper reviews different definitions presented in the open literature. As a general term, dust is a small particle ranging from submicrometers to millimeters. Figure 2 shows the range of small particle sizes from 0.001 μm to 100 μm compared to bacteria, viruses and spores. The U.S. Environmental Protection Agency (EPA) [17] compares the particle size to human hair (50–70 μm). Fine beach sand at 90 μm diameter is larger than human hair. Mechanically generated particles range from a fraction of micrometer to particles larger than 100 μm (0.1 mm). Dust particles are classified as particulate matter (PM) at 2.5 μm, 10 μm, and 100 μm, symbolically designated as PM_2.5, PM₁₀, and PM₁₀₀, respectively. PM₁₀₀ is a fine inhalable particle that can enter the human respiratory system during normal breathing. PM_10, known as thoracic dust, can get deep into the lungs, while PM_2.5 (respirable dust) can get to the bloodstream and pose the most significant health risk. Kwon et al. [18] discussed the characteristics of ultrafine particles (UFP) smaller than 1 mm. The mass fraction of these particles is small compared to the larger micron size particles. UFPs are characterized by large surface area and number per unit volume. UFPs can adsorb toxic chemicals and enter the blood circulation system and even into cellular organelles due to their high surface area that can cause their deposition in various parts of airways.

According to the National Fire Protection Association, dust is any finely divided solids with a diameter of 420 μm or less [19,20]. This size is larger than cement dust with particle size ranging from 2–100 μm, diesel exhaust 0.001–1 μm and a human hair 20–180 μm. The Manufacturing Safety Alliance of BC [7] characterizes combustible dust with a moisture content below 33% and 420 μm or less in size. The particle size is difficult to measure because the size and shape of particles depend on the source plant species, as cellulosic biomass particles have irregular geometries [21]. Plant moisture content and the anatomical part of the plant like leaves, stalks, bark, fruit generate a wide range of particles size and shapes. The British Columbia Ministry of Agriculture [22] characterizes the nuisance of agricultural dust by its detectability, intensity, and acceptability. Detectable dust is particles of which 90% pass through the 44 μm sieve. According to BCMA [22], two parameters quantify the intensity and dispersion behavior of the dust. Opacity is the degree of light transmittance through a cloud of dust. The opacity ranges from 0 percent (transparent) to 100 percent (opaque). Opacity is measured by changes in the light intensity of a beam of light across a known distance. The second parameter is visibility. BCMA [22] defines visibility as the maximum distance the human eye can distinguish an object against its background. It has been mentioned that fine particles are the leading cause of reduced visibility (haze).

Particle size and shape characterize the dispersion of the dust particle. The consequence of small particles and dust must be considered as their behaviour can be different from that of large particles. For example, a smaller particle facilitates flame propagation. Mechanical sieving is commonly used in industry to determine the particle size, with results affected by the duration of the vibrations of the sieving equipment. Many studies have adapted mechanical sieving where the mesh number corresponds to the range of particle sizes retained in that sieve. [20,23,24]. A stack of standard sieve is used to fractionate the dust from biomass samples into different particle size ranges from millimeters to micrometers. This method is highly debatable as particles bounce off the sieve surface and sometimes pass-through sieve openings widthwise, even though sieve openings are smaller than the particle’s length in mechanical sieving operations [25]. Nevertheless, Plumier et al. [26] suggested wet sieving using alcohol to remove dust particles from the corn kernels. Machine vision using a high-resolution camera and image processing techniques is the alternative method to classify particle shapes. Biomass powder samples of each particle size range were photographed with a scanning electron microscope (SEM) or laser particle size analyzer following sieving [20,27,28] to capture the complexity of particles shapes. Mazzoli and Favoni [28] demonstrated using SEM and ImageJ processing programs to study wood particles’ shape and size distribution with a diameter of less than 20 μm. The particulate matter from the air or exhaust can be sampled through a volume sampler such as MCV CAV-A/MSb that enables the collection of samples over quartz microfiber filters [29,30,31]. The deposited airborne particle matter on the fibrous filter observed through scanning electronic microscopy with energy dispersive X-ray spectroscopy (SEM-EDX) is used to characterize various morphologies [30].

3. Dust Generation from Herbaceous Biomass

Herbaceous biomass is sourced from plants with a non-woody stem and shrivels at the end of the growing season [32]. This biomass includes cereal crop residues, grasses (like alfalfa, timothy, and switchgrass), oilseed crop residues, tubers and legumes, flowers, herbaceous biomass of gardens, parks, pruning, vineyards, orchards, and mixtures of all these which can be used raw (direct residues of the field) or processed (from the food industry) [32,33]. These crops are annual or perennial. Depending upon the growth stage and the type of harvest equipment used, herbaceous crops may break, and dust will be detached from the kernel. Morphological structure and moisture content are the two main biomass properties that define the material’s integrity during handling.

Table 1. Crop classifications are based on physical characteristics for handling and processing [34].

Characteristics	Examples
Dry, thin, brittle stalks have low shear and bending resistance.	Wheat, soybean, grass, dry rice straw
Dry, thick, stiff stalks have high shear and bending resistance.	Dry corn stover, dry sorghum stover, sunflower
Wet, thin stalks and leaves have low shear and bending resistance.	Green forage, wet cereals, rice, grass straw, peanuts, tomatoes, other vegetable residue
Wet, thick, stiff stalks and leaves have low shear but high bending resistance.	Green corn stalk, sorghum, sunflower, sugar beet, pineapple
Hard, wooden stalks and branches have high shear and bending resistance, are less flexible.	Cotton, peach branches, pruned vines, trees, hedges, shrub willows, shrub poplars

[34] lists these morphological properties for a few of the herbaceous biomass that affects the harvest and handling of these materials. Dry, thin cell wall herbaceous plants break easily during harvest and handling. The thick cell wall and moist plants are less prone to shatter [35]. Moisture content influences particle generation, with higher moisture decreasing overall particle release. The straw with the lowest water content released more particles than wood chips with high moisture content [36].

Madsen et al. [36] studied the levels of dustiness generated by a small pilot-scale rotating drum as a dust generator from wheat straw, wood chips, wood pellets, and wood briquettes. They assessed different microbial indicators, including bacteria, actinomycetes, fungi, lipopolysaccharide, endotoxin, and muramic acid. They reported that straw was dustier, with two to fifty times more respirable particles released than wood chips, wood pellets and wood briquettes. Sebastian et al. [37] conducted a follow-up study to measure microbial dustiness of baled straw, and wood chips. The bacterial dustiness of baled straw was significantly higher than wood chips, while dust from ecological straw contained fewer bacterial components than dust from conventional straw. Moreover, storing biofuels outdoors over the summer resulted in increased microbial dustiness, and workers handling the biofuels were exposed to high levels of dust rich in microorganisms [37]. Plumier et al. [26] showed that the amount of dust released increases per unit mass of grain handling with repeated handling as a portion of the dust detached from the corn kernel.

Cohn et al. [38] characterized airborne particulate matter (PM) derived from two sources: (a) the air from agitating the wheat straw pellets and wood pellets in a rotating drum, and (b) the air at an electricity generating facility burning straw in Denmark. PM generated in the rotating drum from the source material was more prominent in diameter than PM collected within the facility from the straw storage hall and the boiler room, which likely included post-combustion ash. Straws produce more respirable PM as the particle size distribution of the pure biomass PM had a range of 3.5–5.0 µm for the straw samples and 5.0–7.5 µm for the wood pellet samples. Interestingly, only 30–58% of the PM generated in the rotating drum air was of respirable size, whereas 98% of the PM collected in the electricity generating facility were respirable. The author mentioned that this could be due to larger PM in the rotating drum that grinds faster than smaller PM at the biomass facility [38]. Grebot et al. [39] reported that straw used as fuel in a biomass CHP (combined heat and power) system produced UFP with a concentration of 2.21×10⁶ particles per cm³ after the filter. The concentration of wood pellet particles was reported roughly ten times larger at 6.3 × 10⁷ particles per cm³ after a multi-cyclone separator.

There have been considerable changes in technology and work practices in Britain’s grain industry, where exposure to inhalable dust among grain workers is estimated around 3 mg m⁻³ and only 15-20% of individual personal exposures being >10 mg m⁻³ compared to 8-h average inhalable dust levels of >10 mg m⁻³ in the 1990s [40]. These levels are yet to achieve the long-term limit of 1.5 mg m⁻³ for inhalable grain dust health-based exposure limit published by the Dutch Expert Committee on Occupational Safety where further work is necessary.

3.1. Biomass Collection

Dust and other impurities are generated early on during crop harvesting [41]. A multi-pass harvester introduces dirt particles and other contaminants ranging from 6 to 14%, with an average level of 8% for common agricultural residues [42]. Baling involves picking up the cut crop from the ground [43]. When needed, the bales are transported to a pellet plant, where the bales are chopped or ground by a tub grinder. The chopped biomass is dried in rotary drum dryers if its moisture content is more than 15%. The dried biomass is ground to less than 4 mm for making pellets and subsequently pressed into pellets.

Bonner et al. [44] investigated the influence of five baling methods on the total ash content and variability of ash content within baled corn stover in southwest Kansas in the U.S. The results showed the mean ash content to range from 11.5 to 28.2% of the bale mass, depending on operational choice. Kenney et al. [45] reported the ash content of hand-harvested whole-plant corn stover typically in the range of 5–7% and single-pass corn stover (usually lacking lower stalk) as low as 2–4% (

Table 2. The biogenic ash content of several herbaceous-biomass crops [45].

Crop Type	Average (%)	Range (%)
Corn cob	2.9	1.0–8.0
Corn stover	6.6	2.9–11.4
Miscanthus straw	3.3	1.1–9.3
Reed canary Grass	6.7	3.0–9.2
Rice straw	17.5	7.6–25.5
Sorghum straw	6.6	4.7–8.7
Sugarcane bagasse	5.6	1.0–15.2
Switchgrass straw	5.8	2.7–10.6
Wheat straw	8.0	3.5–22.8

). Although the biogenic ash content of crops can vary widely, the authors concluded that the higher ash was due to the contamination of plant material during baling. Cherney et al. [46] measured the fibre contents in soil contaminated baled grass. They concluded that elemental aluminum is the best indicator of soil contamination of biomass. The ash has incorporated the content and composition of substances derived from the soil, and extrinsic contamination during the harvest and collection process elevated the ash concentration [46]. At the same time, coarse particles with a diameter bigger than 1 µm and aerosol particles with a diameter less than 1 µm are generated from fly ashes [46].

The occupational exposure to dust in agricultural harvesting was evaluated based on two main agricultural areas and seven private farms while focusing on a group of tractor drivers and a group of private farmers from the sampled area in Poland [47]. The study showed that 7.6% in the group of tractor drivers and 24.2% in the group of private farmers exceeded the annual obligatory working limit of 2104 h. Meanwhile, the highest amounts of dust were observed during threshing and combined grain harvesting, which was 57.5 mg m⁻³ on average and 35.7 mg m⁻³ on average, respectively [47]. Thus, this resulted in a high annual calculated mean level of dust exposure in the worksite of 10 examined tractor drivers within the range of 5.3–10.8 mg m⁻³, while that for seven examined private farmers was within the values of 3.6–10.7 mg m⁻³, which was over the maximum allowable value of 2 mg m⁻³.

3.2. Biomass Size-Reduction

Particle size reduction is the most dust-generating operation in the processing of herbaceous biomass. Chawla et al. [3] measured particulate concentration in hammer mill cyclone stacks among three alfalfa dehydrators in Alberta that process alfalfa into pellets or cubes for export. They reported that hammer mill grinding of dried chopped alfalfa (fresh cut dehy, sun-cured) or re-grinding of alfalfa cubes emit large quantities of particulate matter into the atmospheric air, 1.143 g/kg (for Plant 1 grinding fresh cut dehy) to 4.779 g/kg (for Plant 2 grinding fresh cut dehy) (

Table 3. Particulates emission from hammer mill cyclone stacks and particle size distribution of alfalfa grinds (geometric mean diameter d_gw and geometric standard deviation S_gw) sampled at the same time as the assessment of particulates emission levels in three dehy plants in Alberta (data taken from Chawla et al. [3]).

Plant (Sample)	Particulates Emission (g/kg Dry Basis)	dgw (mm)	Sgw (mm)
Plant 1 (dehy)	1.143	0.305	2.215
Plant 2 (dehy)	4.779	0.279	1.209
Plant 3 (re-grind)	2.035	0.300	1.905
Plant 3 (sun-cured)	2.232	0.256	0.989

, all values are reported on a dry basis). These source emission test results were way over the 0.60 g/kg effluent limit set by Alberta Environmental Protection.

Table 3 also lists the geometric particle diameters of ground alfalfa from the dehy plants, analyzed at the University of Saskatchewan. Plant 1 used a 2.8 mm (7/64 in) hammer mill screen with the largest geometric mean diameter (d_gw) of 0.305 mm. Plant 3 ground sun-cured alfalfa using a 3.2 mm (1/8 in) hammer mill screen had the smallest d_gw at 0.256 mm. Particulates emitted by the hammer mill cyclone stack may be affected by the efficiency of the cyclone and the fineness of the alfalfa particles being separated from the air in the pneumatic handling system.

Sokhansanj [48] published Figure 3, showing the size distributions after grinding several herbaceous biomasses using a hammer mill equipped with a 6.4 mm (1/4 in.) screen. The stack of sieve sizes for size distribution analysis was 0.425 mm to 2 mm. A mass fraction between 7% to 9% ended in the pan (particles passing through 0.425 mm sieve). Corn stover had the largest fraction of small particles in the pan. Ground willow had the most significant particles remaining on the top 2 mm screen. Arthur et al. [49] ground rectangular bales of wheat straw, round rice straw bales and corn stover bales using two tub grinders equipped with a 2.7 mm screen. The mass fraction of particles passing through 0.590 mm sieve was 24% for wheat straw, 13% for rice straw, and 23% for corn stover. Moisture content ranged from 8% for rice straw and 10–11% for corn stover and wheat straw. Kaliyan et al. [50] tested a tub grinder with a larger screen size than Arthur et al. [49]. The size of particles in the pan reduced considerably.

Sobczak et al. [51] measured the average concentration of PM₁₀, PM_4.0 and PM1_.0 dust generated from wheat grinding using a hammer mill and roller mill with three different moisture content levels. Hammermill produced a higher percentage of particles smaller than 0.1 mm than the roller mill, yet the obtained concentration of PM₁₀ for both mills exceeded most European countries’ acceptable level of 100 µg m⁻³. The 3 mm screen mesh hammer mill generated the highest organic dust during grinding the material with 9% moisture content [51]. Figure 4 shows the median particle d50 at the end of each milling of wheat straw with different choices of mill equipment and sieving grid [52].

3.3. Pelletization and Pellet Durability

Dust emissions from crop residue pelletization operations themselves have not been extensively studied. Beauchemin and Tampier [53] reviewed EPA on wood processing (woodworking) emissions, but none described dust generation from wood pellet operations. The authors reported that the airborne dust generated during the manufacturing of wood products typically shaving, sawdust, chips or other sawmill residues. The contaminants emitted from wood pellet plants and dust are total organic compounds (TOC). At the same time, small amounts of carbon monoxide, carbon dioxide, and nitrogen oxides are also emitted from the dryer burners [53].

Pellets tend to disintegrate quickly either in storage or transportation conditions due to moisture adsorption and impact on hard surfaces. The small-size broken pellets can turn into dust.

Table 4. International standards (ISO 17225) for straw pellets and commercial, residential wood pellets [54,55].

Pellet Quality Index	ISO 17225-6 Straw Pellets	ISO 17225-2 Wood Pellets
Durability (%)	≥97.5	≥98.0
Fines content (%)	≤1.0	≤1.0
Bulk density (kg/m3)	≥600	600–750
Ash content (%)	≤6.0	≤0.7

is an extract from International Standards 17225 [54,55] on the top physical quality of pellets that shows two different standards for woody and agricultural pellets. Durability measures the strength of pellets against impact and frictional forces—the pellets with lower durability exhibit a higher chance of breakage. The minimum durability requirement for wood pellets is higher than the minimum requirement for agricultural pellets based on ISO 17225 [54,55]. The minimum bulk density is specified chiefly at a higher value for agricultural pellets. Ash content is the most distinct difference between the minimum specified for agricultural pellets at 6% vs. 0.7% for wood pellets.

Theerarattananoon et al. [56] investigated the effect of moisture and die geometry on the durability and bulk density of a few agricultural and purpose-grown crops. The maximum durability value obtained was 96.8%. A further increase in moisture content value resulted in decreased pellet durability. For sorghum stalk pellets, the durability value increased initially with increased moisture content and reached 89.5%. Using a larger hammer mill screen size from 3.2 mm to 6.3 mm resulted in increased bulk density and durability of biomass pellets.

Castellano et al. [57] showed that pellets from herbaceous materials had lower mechanical durability due to lower lignin and higher extractive contents, as shown in

Table 5. Lignocellulosic materials composition [57].

Lignocellulosic Materials	Lignin (w% d.b.)	Hemicellulose (w% d.b.)	Cellulose (w% d.b.)	Extractives (w% d.b.)
Pine (Softwood)	28.46	44.16	23.56	6
Oak (Hardwood)	30.31	34.33	21.94	5.32
Oat (Herbaceous)	10.59	26.86	24.53	30.87
Triticale (Herbaceous)	11.07	34.75	20.86	28.39
Rice straw (Herbaceous)	13.82	37.09	24.08	13.98

. Woody plants owe their hard surfaces to a high lignin content, which binds their fibres together. At the same time, non-woody biomass has a lower proportion of lignin which loosely binds their cellulose fibres, giving them more pliable surfaces. Cellulose is responsible for strength in wood fibre due to its high degree of polymerization and crystallinity. Hemicellulose acts as a matrix for the cellulose and acts as a link between the fibrous cellulose and amorphous lignin. By contrast, agricultural biomass has a high concentration of hydrophobic waxes, which impacts pellet quality (weak and powdery pellets) [57]. These experimental results indicate that the overall durability of agricultural pellets is expected to be lower than wood pellets.

Higher durability pellets are produced by finer grinding, as shown in Figure 5 [58]. It has been reported that nearly 30% of fines less than 250 µm is desirable to produce high-quality pellets [58]. As discussed in Section 3.2, the finer grinding of biomass generated dust in more significant amounts. Questions arise about whether wood pellets or pellets from herbaceous biomass should be handled differently due to their chemical compositions and particle size distributions. Consequently, the dust created from herbaceous biomass pellets needs additional attention due to their low durability and high fines content.

4. Human Health and Herbaceous Feedstock

The literature discusses the human health aspects of agricultural dust in terms of organic, non-organic dust, and nuisance dust. Nordgren and Baily [59] reported that prolonged exposure to dust, vapors, and fumes might lead to chronic obstructive pulmonary disease (COPD), asthma, hypersensitivity pneumonitis, and possibly interstitial lung disease and cancer. The U.S. Occupational Safety and Health [60] publication defines organic dust as molds, pollens, bacteria, pesticides, chemicals, feed and bedding particles, and animal particles. The workers in the wood pellets production facilities were exposed to wood dust and bioaerosols, which resulted in increased dry cough, wheezing, and nose, eye, and throat irritation [61]. Hay, grain, fuel chips, straw, and livestock are organic dust sources. Over time, exposure to organic dust can result in diseases like organic dust toxic syndrome (ODTS) and farmer’s lung disease (FLD). Generally, FLD and allergic alveolitis contemporary with chronic cough, dyspnea on exertion, fatigue, anorexia, and weight loss.

Attention is given to the emission of particulate matter from the biofuel combustion process and the possibility of reducing the fine particles generated from the combustion chamber in the past decade [24,26,62]. Only a limited amount of research on particulate matter emission on the pre-combustion process, including handling and storage, was conducted. Rohr et al. [63] focused on reviewing available literature on assessing potential occupational health and safety (OH&S) concerns related to biomass combustion and included pre-combustion exposures comprising storage and handling operations of biomass fuel itself. Rohr et al. [63] also covered wood dust exposures and risks extensively with limited information on agricultural residues. The authors highlighted the existence of broad uncertainties related to health effects findings from studies on agricultural residues. The straw dust endotoxin correlated significantly with microbacterial activity and dustiness that caused chest tightness, fever, fatigue, and diarrhea [36,37,64,65].

Barrera et al. [64] investigated the severity of breathing in grain wheat dust on the respiratory health of two different groups of people: the farmworkers handling massive field wheat and the cattle raisers handling huge amounts of stored wheat. The hazard was not dependent on the type of wheat handled but on the duty type and the presence of collective protective equipment as the exposure from the machines used during harvesting or unloading was >4 mg m⁻³. Risk of field wheat dust correlated highly with acute symptoms such as cough, wheezing, dyspnea, runny/stuffy nose, scratchy throat, and systemic signs. In contrast, the level of exposure to stored wheat dust was associated only with cough [61]. The particles below 1 µm are more likely to reach the alveoli [61].

The National Dairy Database [66] states that opening grass bales in an enclosed area, such as a barn, can produce 1.6 × 10⁹ spores/m³ of thermophilic Actinomycetes. 7.5 × 10⁵ spores/min can be absorbed in A person’s lungs by carrying out easy chores. Concentrations of 4 × 10⁹ viable spores/m³ have been detected in silo openings. According to Smith [67], a new group of fungi and yeasts start to expand once the hay is baled. The growth of fungi is more prevalent when the moisture content of the baled hay is between 20% and 30%. The three main groups of fungi in hay bales are Aspergillus, Fusarium and Penicillium. A few distinct fungi that grow during bale storage (e.g., Aspergillus flavus) are known to produce mycotoxins. Each kind of fungi has its ideal temperature and moisture level where they grow best, but none grow well at low moisture levels (<15%).

The constituent of inorganic dust in agriculture usually describes the soil composition. Crystalline silica may represent up to 20% of particles, and silicates represent up to 80%. These very high concentrations of inorganic dust are likely to justify the rise in chronic bronchitis reported in many studies.

Gbaguidi-Haore et al. [68] evaluated numerous studies to discover the factors contributing to FLD among farmers handling hay. A detailed statistical analysis showed that dense packing of hay corresponded with increased concentration of hypersensitivity pneumonitis FLD–causing microorganisms (e.g., Absidia corymbifera, thermophilic Actinomycetes) in the hay. The concurrent analysis of batches of hay- and farm-level factors showed that bad climatic conditions of harvest, high-density hay-packing, mainly round bales, and high altitude were the main factors associated with high concentrations of these microorganisms in the hay. These findings show the need of better grass packing, drying and respiratory protection (e.g., NIOSH N95) when in contact with potentially moldy grass to avoid occupational hypersensitivity pneumonitis.

5. Thermal Characteristics of Agricultural Dust

Dust Safety Science has published that agricultural activity and food production has made up a large portion of the overall dust fire and explosion, ranging from 33 to 50% since 2017 [69]. The most recent report includes all incidents of the widespread dust fire and explosion involving all industries from 1 January to 30 June, 2021, as shown in Figure 6 [69]. The flaming of the dust cloud releases energy and generates gaseous reaction products such as carbon dioxide and water vapor, leading to dust explosion. This is highly possible since tiny biomass particles can become easily airborne when handled. At the same time, the lower explosive limit for many organic materials is between 10–50 g/m^3, and ignition sources can include sparks from equipment [70].

Figure 7 illustrates the five main components to initiate a dust explosion. Over the years, several researchers have studied the parameters that describe dust explosion hazards, categorized into three: explosion parameters, concentration limits, and ignition parameters [71]. The explosion severity can be evaluated by the maximum pressure of the explosion in a test vessel with the intent of achieving ignition in the center of a uniform dust cloud. The higher the explosion pressure, the more detrimental it is towards the surroundings. The understanding of the dust concentration is vital due to the explosibility limits. The minimum explosive concentration (MEC) is the smaller concentration at which the dust starts to be fierce. The minimum oxygen concentration (MOC) is the limiting percentage of oxygen to support combustion.

The lowest temperature to ignite the dusty sample is the minimum ignition temperature and is determined for both dust cloud and dust layer. The minimum autoignition temperature is measured for dust dispersed as a cloud. Dust layers’ hot surface ignition temperature represents the minimum temperature threshold for initiating dust fires that are self-sustaining. The minimum ignition energy (MIE) is the lowest energy required to ignite the most flammable mixture of a dust cloud subjected to an electric discharge. The temperature at which maximum volatile release (TMVR) indicates the temperature at which explosion might be initiated because of the significant production of volatiles [72]. Therefore, the propensity for the outbreak is a function of particle concentration, oxygen concentration, the energy of the ignition source or the temperature of the heat exerted on the dust. Testing standards are available to measure the explosion, concentration, and ignition parameters.

In a report about a fatal explosion accident in a Denmark wood pellet manufacturing facility, Hedlund et al. [73] argued that the risks associated with wood dust were not fully understood. A closer analysis revealed that operations personnel knew what type of dust causes outbreaks and the critical importance of diligent housekeeping. Average particle size, the concentration of the dust-air mixture, pressure, temperature, oxygen in the atmosphere are the factors that affect the force and sensitivity to dust explosions [74]. By understanding the properties of a dust material, one can create inherently safer facility designs and attempts to prevent dust explosions from occurring in the first place.

One fuel characteristic that is particularly important for fire hazards is the ignition tendency. Ignition is required to initiate combustion. Previous studies focused on the factors that can cause explosions, such as the minimum ignition energy, minimum explosible dust concentration, and minimum ignition temperature. Melin [19] writes that outbreaks occur in dust suspended in the air and dust sediments on hot surfaces.

Table 6. Fire and explosion parameters for wood, coal, and Lycopodium ¹ particles (adapted from [19]).

Test Mode	Parameter (Dust < 63 μm)	White Wood	Bark	Coal	Lycopodium	Testing Standards
Dust cloud	Auto-ignition temperature, TC, °C	450	450	585	430	ASTM E1491
	Min. ignition energy, MIE mJoule	17	17	110	17	ASTM E2019
	Max. explosion pressure, Pmax, bar	8.1	8.4	7.3	7.4	ASTM E1226
	Max. explosion pressure rate, dP/dtmax, bar/s	537	595	426	511	ASTM E1226
	Deflagration index, Kst, bar.m/s	146	162	124	139	ASTM E1226
	Min. explosion concentration, MEC, g/m3	70	70	65	30	ASTM E1515
	Limiting oxygen concentration, LOC, %	10.5	10.5	12.5	14	ASTM E1515
Dust layer	Hot surface ignition temp. (5 mm), TS, °C	300	310			ASTM E2021
	Hot surface ignition temp. (19 mm), TS, °C	260	250			ASTM E2021
	Auto-ignition temperature, TL, °C	225	215			USBM RI 5624
	Dust class (>0 to 200 bar m/s)	St 1	St 1	St 1	St 1	ASTM E1226
	Dust class (Explosion severity ES > 0.5)	Class II	Class II			OSHA CPL 03-00-6

¹ Spores of Lycopodium is a yellow-tan dust-like powder, containing the dry spores of clubmoss plants, or various fern relatives. The spores are combustible and are used to build dust bombs during theatrical special effects (Wikipedia 2021).

summarizes tested explosion parameters of dust from white and bark pellets, coal, and the highly flammable Lycopodium spores. The data is a compilation of tests performed by various agencies and organizations and was reported by Melin [19]. A few decades ago, explosion and ignition parameters of agricultural dust passing a No. 200-sieve (74 µm) shown in

Table 7. Explosibility of agricultural dust [75].

Biomass	Ignition Temperature (°C)	Minimum Ignition Energy (J)	Minimum Explosive Concentration (kg/m3)	Maximum Explosion Pressure (kPa)
Alfalfa	460	0.32	0.1	455
Cereal grass	540	0.8	0.2	358.5
Corn	400	0.04	0.045	655
Corn starch	380	0.02	0.04	793
Wheat straw	470	0.05	0.055	682.6
Wheat starch	380	0.02	0.025	724

were published by Ball et al. [75]. The authors reported that the minimum energy ignition energy and the pressure rise are affected by the change in average particle diameter, where the explosibility of agricultural dust increases with a decrease in particle size. There is no comprehensive published study on the explosion parameters of dust generated from herbaceous biomass.

Dhiman et al. [72] investigated the thermal properties of lignocellulosic biomass, namely Bermuda grass, corn cobs, corn stover, Eucalyptus, loblolly pine, sugarcane bagasse, sweetgum, and switchgrass, as shown in

Table 8. Hot surface ignition temperature and volatilization temperature of herbaceous and woody biomass [72].

Biomass	Hot Surface Ignition Temperature (°C)	Onset Temperature for Rapid Volatile (°C)	Temperature for Max Volatile Rate (°C)	Activation Energy (kJ/mol)
Bermuda grass	275	266.1	312.4	48.3
Corn cobs	280	276.5	313.7	63.2
Corn stover	290	277.5	316.4	71.3
Eucalyptus	285	291.9	315.7	72.4
Loblolly pine	315	306.1	348.8	64.4
Bagasse	305	285.2	345.2	77.1
Sweetgum	305	290.2	338.8	95.2
Switchgrass	295	285.0	329.8	61.6

. The samples were air-dried and ground in a hammer mill equipped with a 3.18 mm screen. Particles that passed through the 437 μm screen were collected. The moisture contents of the dust samples varied between 6.1% and 8.3% w.b. (wet basis). The higher ash contents resulted in significantly higher bulk and particle densities and lower biomass dust’s volatile matter and energy contents. Table 8 shows that hot surface ignition temperature (HSIT) ranged between 275 °C and 315 °C and was found to be linearly related to the bulk density of the dust samples. Dust from grassy biomass feedstocks had lower HSIT compared to dust from woody biomass feedstocks. The temperatures at which volatiles was released were lower for grassy biomass dust at 266.1 °C to 285.2 °C compared to woody biomass dust at 290.2 °C to 306.1 °C.

Fernandez-Anez and Garcia-Torrent [76] studied the effect of particle size and dust layer thickness on the ignitability of different solid fuels layers that include biomass, coal and cake samples, as shown in

Table 9. Minimum ignition temperature (MIT) of solid samples with different particle sizes [76].

Sample	d50 (µm)	MIT at 5 mm (°C)	MIT at 50 mm (°C)
Wood chips	146.6	360	270
	1100	390	270
Wood pellets	404.1	300	250
	3000	>400	250
Olive pit	68.2	300	230
	1300	>400	320
Wheat straw	171.2	320	250
Almond shell	44.3	290	230
Coke	77.4	360	240
	1200	>400	>400

. The authors measured the minimum ignition temperature of dust layers (MIT). They concluded that an increase in dust layer thickness and a decrease in particle size could reduce the temperatures needed to ignite a dust layer [76]. They also showed the need for complete characterization and operational safety measures of each new material. The standard practices well-known for fossil fuels could not be directly extrapolated to new ones.

Tannous et al. [77] categorized a blend of ground Douglas fir particles using a set of 14 wire mesh sieves. The calculated mean diameter of the mixture was 251 µm. The sphericity of particles increased with the sieve’s decreasing size, indicating that smaller particles also have a smaller aspect ratio. The dust samples from grassy samples have high aspect ratio particles because the needle-like particle exhibits d50 values more than 437 μm compared to woody biomass [72].

Polin et al. [23] collected sunflower dust from horizontal surfaces and flat parts of a combine harvester in South Dakota. They compared it with corn stover dust to study the differences between them. Overall, sunflower has a reduced ignition point by 30 °C and a higher surface area that allows more oxygen and heat transfer than corn stover at the same particle size. Very fine dust has a particle size in the 25–63 µm range that constituted 19.3% of the mass of collected dust overall and had the lowest ignition points for both biomasses. The auto-ignition point for corn stover and sunflower samples reduces concurrently with the dust particle size [23].

Building codes, fire codes, electrical codes, the National Fire Protection Association standards, and the Occupational Safety and Health Administration regulations are essential in the safety of biomass handling facilities [78]. The National Building Code of Canada points out minor conditions affecting human health, fire safety, and structural sufficiency. The British Columbia Building Code made all barns within municipal districts accommodate the national code [79]. The fire hazard assessment results are to find the possibility of unintended fire occurrence and indicate the potential frequency and consequences of those occurrences. The assessment approach is generally made by evaluating the existing effective methods associated with the activity or formulating a reasonable procedure. The latter method is chosen when insufficient exposure involves evaluating the activities or processes. There is backwardness in fire hazards safety and response in perennial grass-related agriculture businesses.

6. Summary and Recommendations

With the recent interest in herbaceous biomass pellets, there is an explicit need for research to expand our understanding of the risks associated with their pelletization, especially flammability risks, which may cause ignition and explosion on industrial facilities. Moreover, health hazards should be considered when designing the unit operations for particle size reduction, handling and drying herbaceous biomass. The following points summarize this review:

• Abundant data are available to characterize dust from woody biomass. But data on dust from herbaceous biomass (straw, stover, hay) are limited.
• The harvesting, collection, handling, size reduction and pelletization of herbaceous biomass generate a considerable amount of dust. But numerical data are lacking.
• The available information on dealing with the dust during the harvest and postharvest operations is mainly from general housekeeping recommendations.
• There is considerable data and regulations on working with grain dust, but their applicability to handling herbaceous biomass in pellet mills is unclear.
• Dust thermal properties, specifically those associated with dust explosion, are available for wood and coal but not crop residues.
• Compared to woody biomass, herbaceous biomass has lower lignin content and a high concentration of hydrophobic waxes, resulting in weak powdery pellets and potentially contributing to more dust generation. The wide range of sizes and shapes in particles of herbaceous biomass and its heterogeneous nature are responsible for most of the difficulties presented in its characterization for handling purposes.
• The building codes and processing centers for agricultural pellets are general and do not address specifics associated with pellet production processes.

We recommend a systematic study of the harvest and postharvest operations involved in the pelletization of herbaceous biomass to generate valid engineering data towards characterizing dust in terms of particle size during processing/handing and stages where particulate matter is generated. Besides that, studying the ignition sources present in the pelletization facility and the dust characteristics from herbaceous biomass is needed to design prevention and protection measures. Specific data on the explosion parameters, concentration limits and ignition parameters for crop residues’ dust cloud and dust layers need to be developed. There is a consensus to characterize dust from the herbaceous biomass pellets processing operations. This may be patterned after the testing standards (including experimental design) used for coal and wood dust.