Dust explosion

A Dust explosion is the rapid combustion of a cloud of combustible particles suspended in air, producing heat and a potentially damaging pressure rise. The hazard arises when particle size and concentration support flame propagation and an ignition source is present. Confinement can prevent combustion products from expanding freely and greatly increase the consequences.

Dust explosions can occur in process equipment, silos, rooms and industrial buildings. Settled dust can also be lifted by an earlier event and form a new cloud, creating a secondary explosion. Evaluation therefore needs to address dispersed dust, accumulated layers, ignition sources and the strength and geometry of the enclosure.

Explosion protection combines prevention of explosive clouds, control of ignition, inerting and pressure-limiting measures. No single measure is suitable for every dust or installation. Test results must retain their particle, concentration, ignition and turbulence conditions.

Formation of an explosive dust cloud

Zalosh describes combustible dusts and powders as potentially explosive when their characteristic particles are small enough and suspended between lower and upper explosive concentrations. The cited particle-size range is 100 to 400 micrometres depending on the material. Those values are guidance from the 2002 handbook, not a universal threshold for every dust.

For dusts with characteristic particles below about 100 micrometres, typical lower explosive concentrations are reported as 30 to 60 grams per cubic metre. Typical upper values of 2,000 to 6,000 grams per cubic metre are difficult to measure and apply in practice. These ranges must not replace material-specific testing.

A cloud forms when particles become suspended by pouring, air movement, vibration or another disturbance. Normal operation may not produce the highest concentration, while upset conditions can dislodge accumulated material. The relevant assessment therefore includes credible abnormal as well as steady operation.

A dust explosion requires dispersion, ignition and flame propagation. Based on Zalosh (2002), pp. 3-402–3-406.

An Explosion follows if the cloud ignites and flame propagates through it. The fuel and oxidant are already distributed within the cloud, and the combustion products heat and expand. In a closed or partly confined volume, pressure then develops according to flame propagation, burned-gas conditions and heat losses.

Uniform cloud assumptions can be misleading. Industrial dust clouds may be stratified or localised, and concentration can change during dispersion and settling. Measurement and modelling must represent where the cloud is and how it changes over time.

Particle properties and concentration

Particle size affects the surface available for devolatilisation and heat transfer. Smaller particles generally reduce the minimum explosive concentration until a limiting size is reached. Below that limiting size, other combustion processes can control the result, so the trend does not continue indefinitely.

Particle size also affects maximum pressure and rate of pressure rise. Zalosh gives examples where pressure changes little below 100 micrometres, but a less combustible material can become difficult or impossible to ignite above a particular size. Those observations are dust-specific rather than universal.

Concentration influences both ignition and pressure. The maximum pressure often occurs around 500 grams per cubic metre in the cited organic-dust tests, several times the minimum explosive concentration. Using the minimum explosive concentration as the design worst case would therefore miss the concentration associated with greatest pressure.

Dust chemistry and moisture, temperature and turbulence also affect behaviour. The chapter reports decreasing minimum explosive concentration with increasing temperature for a given material and particle size. These dependencies must be established with representative material and test conditions.

Measured industrial concentration is itself uncertain. A sample at one point and time may not represent another region of the vessel or a later upset. A safety case based on a normal low reading can be challenged by local deposits and later resuspension.

Ignition energy and temperature

Minimum ignition energy is the smallest spark energy that can ignite a combustible cloud under specified test conditions. The cited method uses a modified Hartmann tube and supports a range of spark energies. Results depend on discharge type and duration as well as the dust.

Capacitive and inductive discharges should not be treated as equivalent. Zalosh reports that long-duration inductive conditions can reduce the energy required by an order of magnitude in the cited comparison. The test circuit and spark duration must therefore accompany any minimum-energy value.

Minimum ignition energy also varies with concentration, particle size, temperature and dust-air velocity. The table values in the handbook span less than 1 to more than 2,000 millijoules for the listed samples and conditions. A generic spark-energy threshold cannot be assigned to every dust cloud.

Cloud ignition temperature is usually measured by injecting dust through a heated furnace tube. Layer ignition temperature is instead measured using a dust layer on a controlled hot plate. These are different thermal conditions and can produce different results.

Layer temperatures are often lower than cloud ignition temperatures because the layer dissipates heat differently. Local hot spots and nonuniform wall temperatures further complicate the comparison. A mean equipment-surface temperature may not reveal the small region capable of ignition.

Deflagration pressure and KSt

A dust deflagration is flame propagation at subsonic speed through the combustible cloud. It can produce high enclosure pressures even though it is not a detonation. Zalosh reports worst-case pressures for listed dusts of 6.7 to 11.2 bar gauge under their test conditions.

Deflagration index (KSt) is a size-normalised measure derived from the maximum rate of pressure rise and test-vessel volume. It is used to compare dust reactivity in explosion-venting methods. KSt must be linked to the standard test method and conditions rather than treated as a material constant independent of testing.

Rate of pressure rise depends on flame propagation, flame temperature and vessel size. Turbulence has a particularly strong effect in dust explosions because particle dispersion itself creates turbulent motion. The delay between dust injection and ignition changes the turbulence at ignition.

Zalosh reports measured maximum pressures for many flammable gases and dusts in similar ranges under their respective tests. This does not mean their ignition difficulty, flame propagation or protection design is identical. Dust dispersion and finite particle burning introduce additional processes.

Closed-vessel calculations assume particular conditions, including flame propagation through the mixture and burned-gas behaviour. Near flammability limits, incomplete combustion and heat losses can reduce measured pressure below theoretical values. Those limits matter when interpreting pressure estimates.

Secondary explosions and accumulated dust

A primary explosion inside equipment can breach its enclosure and create a blast wave and strong air movement. That disturbance can lift dust deposited on floors, walls and equipment. The resulting cloud may then ignite and produce a secondary explosion in the occupied building.

Secondary explosions can involve dust accumulated outside the original equipment. Based on Zalosh (2002), pp. 3-410–3-411.

Secondary explosions can be more damaging than the original event. They involve larger occupied spaces and can cause casualties and structural damage. The assessment must therefore include accumulated dust outside the process vessel, not only the material normally inside it.

Even a cloud occupying a fraction of a building can produce severe consequences. Zalosh emphasises that dispersed layers can supply enough fuel for a secondary event. The quantity, location and ease of resuspension of those layers belong in the hazard review.

Layer and cloud hazards are linked but different. A layer can ignite through a hot surface and later act as a source for a cloud, while a cloud requires suspension and a suitable concentration. Controls must address both the initial fuel state and the transition between them.

The time sequence matters for protection. Preventing the primary explosion or limiting its pressure can reduce the disturbance that creates a secondary cloud. Housekeeping and ignition control reduce the fuel and initiation opportunities, while venting addresses pressure in the primary enclosure.

Prevention and inerting

Prevention starts by avoiding explosive concentrations and controlling ignition sources. Process ventilation, limiting resuspension and reducing accumulated layers can address cloud formation. Grounding equipment can prevent electrostatic charge from reaching ignition conditions where those conditions are understood.

Inerting adds an inert gas to reduce oxidant concentration below the level needed for combustion. Zalosh defines the limiting oxygen concentration as the maximum safe oxygen concentration for prevention by inerting. Its value depends on the dust and the inert gas.

For the organic dusts listed in the cited table, carbon-dioxide-inerting values are around 11 to 15 volume percent oxygen. Some metals have much lower values, including pyrophoric behaviour for the zirconium example. Those examples show why inerting cannot use one universal oxygen setpoint.

Monitoring accuracy, response time and nonuniform mixtures affect inerting reliability. The cited NFPA 69 guidance calls for a margin below the limiting oxygen concentration when continuously monitored, with a larger margin when not monitored. The applicable edition and process conditions must be checked.

Hazardous area classification can help organise ignition-control measures by where explosive atmospheres may occur. The supplied documents do not provide a detailed zone-classification method for dusts, so no zone numbers or dimensions are stated here. That absence must not be filled with an invented layout.

Venting and other protective measures

Explosion venting discharges combustion gases during a deflagration to keep pressure below the enclosure damage threshold. Vents commonly use panels, rupture discs or engineered devices that open after initiation. Their purpose is pressure limitation, not prevention of ignition.

Early deployment and large, unrestricted vent area improve the cited venting arrangements. Required area depends on enclosure volume, mixture reactivity, turbulence, geometry and vent characteristics. A vent-area value from a test or guide cannot be transferred without checking those conditions.

Explosion venting is not suitable for detonations. A detonation produces a local shock and maximum pressure almost instantaneously when it reaches a point, leaving no useful period for vent deployment. Deflagration and detonation must therefore be distinguished before selecting protection.

Deflagration suppression detects and arrests combustion during an incipient stage. It is another pressure-limiting approach distinct from inerting and venting. Its performance depends on timely detection and distribution of the suppressing action.

Dust-explosion protection acts at different points in the event sequence. Based on Zalosh (2002), pp. 3-402–3-403 and pp. 3-414–3-416.

Pressure containment instead designs the vessel and its appurtenances to withstand expected deflagration pressure. This addresses the enclosure's resistance rather than preventing combustion. The relevant pressure and structural response must be demonstrated for the actual arrangement.

Testing, interpretation and uncertainty

Explosibility tests use defined vessels, dust dispersion systems and ignition sources. Zalosh describes near-spherical vessels of 20 to 30 litres for common tests and larger vessels where ignition-energy effects require them. Results can change with vessel size and turbulence.

Minimum explosive concentration is sensitive to igniter energy and turbulence. The 20-litre method uses specified ignition devices, while a 1-cubic-metre vessel may reduce ignition-energy influence. A measured result must therefore identify the apparatus and ignition method.

KSt and maximum pressure results also depend on dispersion delay and ignition timing. Standardised injection and ignition procedures improve reproducibility between laboratories. Applying the values to an installation still requires an assessment of its actual turbulence and geometry.

Test data should be checked against the page image and units when used quantitatively. The extracted source text shows that some particle sizes and ignition temperatures need particular care because symbols can be lost. No values in this article should be used as a design table without that verification.

A full protection assessment maps the sequence from dust formation to ignition, pressure development and enclosure response. It identifies which measures interrupt each stage and what happens if one fails. This reveals secondary explosions and common dependencies that a simple combustibility label would miss.

Material-specific assessment should use dust representative of the installation. A table entry with the same generic material name can have different particle size, processing history and ignition behaviour. Selection should therefore explain how the tested sample compares with the dust actually produced and accumulated.

Protection layers should be evaluated for shared dependencies. An inerting system that fails can leave a vessel exposed to the same ignition sources that the vent is intended to address. Monitoring, maintenance and upset conditions belong in the evaluation of each protective layer.

The assessment record should distinguish cloud, layer and secondary-explosion evidence. It should identify the test methods used for ignition and reactivity, the expected dust dispersion paths and the enclosure's damage criterion. Those details make clear which parts of the protection strategy address each stage. They also help identify whether the available evidence describes the whole installation or only one test vessel and set of conditions.

See also

• Minimum ignition energy
• Self-heating
• Explosion venting

References

• U.S. Nuclear Regulatory Commission. (2005). Fire dynamics tools: Quantitative fire hazard analysis methods for the U.S. nuclear regulatory commission fire protection inspection program (NUREG-1805). Author.
• Zalosh, R. (2002). Explosion protection. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed.). National Fire Protection Association.