Kinetic mechanisms of dust combustion
The classification of the dust explosions can be associated to the mechanisms that are established by the nature of the dust as well. Indeed, the mechanism of flame propagation for many dusts can be classified in two main different groups. The first category corresponds to the materials that are characterized by the combustion of flammable gases emitted by particles heated to the point of vaporization or pyrolysis (Dufaud et al., 2012; Eckhoff, 2003; Petit, 2006). Additionally, some other dusts can propagate a flame through a direct oxidation process developed at the particle surface (Cashdollar, 2000). The former is commonly associated to the combustion of organic dusts, which are made of vaporizable substances (volatile flames) and the latter is linked to the oxidation of metallic dusts or graphite (Nusselt-type flames).
The majority of the accidents have been associated to the volatile flames due to the widespread use organic compounds in various industries. Figure 1.3 describes the distribution of the events associated to a dust explosion that were reported in North America in 2004 (Stahl, 2004). This chart shows that the activities that are more prone to a dust explosion include notably branches of the food and wood processing industries, paper, synthetics production, and pharmaceuticals production.
The explosion of an organic dust can be divided into three phases: the particle heating; its devolatilization (pyrolysis) and the oxidation of the pyrolysis gases. In fact, the heating and the pyrolysis steps are very fast for particles having small diameters (generally < 30µm) (Di Benedetto et al., 2010). For these particles, the combustion kinetics can be reduced to the oxidation in the homogeneous gas phase (Dufaud et al., 2012). On the other hand, the devolatilization of the solid particles becomes the rate-controlling factor at high dust loadings, for large particle sizes or for refractory powders.
Previously, Dufaud et al. (2012) established the influence of the pyrolysis step on the kinetics of the combustion of the micrometric starch. For this purpose, their study determined the explosive behavior of the organic powder and a mixture composed by the gases that are released during the pyrolysis phase. The experimental results did not evidence a significant difference between the maximum pressures that were achieved by the two comparative tests. Indeed, the maximum overpressures were 7.5 bar and 8.5 for the gases and the starch respectively. This similarity is attributed to the thermodynamic characteristics of the combustion of each mixture. Nevertheless, a predominance of the pyrolysis phase on the combustion kinetics was observed when the experimental results showed that the value of the maximum rate of pressure rise was 2830 bar·s-1 for the pyrolysis gases whereas it was lower than 400 bar·s-1 for the starch.
Furthermore, some organic dyes are considered to be more reactive than some organic natural dusts such as the starch and the combustible proteins. This condition is evidenced because the reactivity of these materials depends notably on the products that are yielded during the pyrolysis phase. Eckhoff (2003) established that the combustion rate of dusts that devolatilize unsaturated gaseous compounds, which are more reactive, can be more affected by the presence of fine particles in their size distribution. Further details about this dependence will be discussed in section 1.6.1.
Most of the metallic dusts differ from the organic dusts because they do not volatilize or pyrolyze, but melt and burn as discrete entities. In other words, some metal powders such as magnesium and aluminum may vaporize rapidly when subjected to high temperatures and thus react mainly in gaseous phase. These characteristics were also evidenced experimentally by Gao et al. (2014). These authors identified a more complicated structure in the combustion of organic dusts. The flame zone of these dusts are composed by blue spot flames at the leading zone and luminous flames behind them. This arrangement clearly differs from the one of metallic dust clouds, which is defined by the combustion of every entity. This difference can be observed in Figure 1.4.
Deflagrations and detonations in dust explosions
Previously it was established that the explosion of a combustible dust can result in a detonation or a deflagration according to the confinement level of the mixture. This condition determines the negative effects according to the characteristics of the flame front developed during the explosion. In fact, the explosion will correspond to a deflagration if the front propagates below the speed of sound in the unreacted gases. On the contrary, when the flame velocity is greater than the speed of sound is known as detonation. This severe explosion is defined by a strong pressure wave, which causes the compression of the unreacted mixture above its autoignition temperature. At this point, the ignition process is defined by the temperature upstream the front flame and not the heat transfer. This process is very rapid and causes an abrupt pressure change or shock in front of the reaction front (Crowl & Louvar, 2011).
The pressure fronts produced by detonations and deflagrations pose some remarkable differences. On the one hand, a detonation is distinguished by a shock front with an abrupt pressure rise that might reach velocities of the order of 2000 – 3000 m/s (Lees, 2005). For this reason, time duration of a detonation is typically less than 1 ms. On the other hand, a deflagration has a wide flat pressure front that reaches subsonic velocities and does not have an abrupt shock front; hence this front can last many milliseconds (Crowl & Louvar, 2011). Besides, a deflagration usually achieves lower values of the maximum pressure than that obtained after a detonation.
The speed of the flame front developed in a dust explosion is usually comparable with that in gas deflagrations (Lees, 2005). Thus, the combustion of solid compounds is commonly associated to deflagrations rather than detonations. For this type of explosions, the energy from the chemical reaction is assumed to be transferred to the unreacted mixture by heat conduction and molecular diffusion. However, Eckhoff (2003) has discussed certain theories that pose that detonations in dust explosions might be developed due to several factors that might increase the flame speed. For instance, the presence of an explosive charge or similar external means of generating the initial shock may increase the strength of the shock wave. Moreover, Proust (1996) also posed that long canalizations or vessels that have a length/diameter ratio greater than 5 can be submitted to an acceleration phenomenon that might result in a detonation as well. In addition, a rising flame can increase the dust concentration in the reactive zone and stretch its front during the propagation process. This fact might also augment the possibility of a detonation. For these reasons, it is possible to establish that the geometry and conditions of the confined volume can determine if a dust explosion can result in a detonation as occurred with gases.
In the same way, some factors determine the impossibility to produce a detonation with a combustible cloud. For instance, a dust/air mixture that is unconfined has ignition delays that can be at least one order magnitude greater than gas/air mixtures. For this reason, the mass of dispersed dust must be extremely high to provide the energy transfer rate that is necessary for the cloud to constitute a detonation. As discussed above, an unconfined mixture will only constitute a flash fire.
EXPERIMENTAL DETERMINATION OF THE DUST IGNITABILITY
The chemical composition of a combustible dust allows establishing the mechanism associated to its combustion process. This analysis can provide an insight for the estimation of the effects of an eventual dust explosion. Nevertheless, this analysis is just a preliminary characterization of the dust flammability because some aspects associated to the flow dynamics of a dust cloud must also be considered. For this reason, it is necessary to perform a set of laboratory tests that determine the minimum requirements that must be fulfilled to ignite a combustible dust-air mixture. The experimental results of these tests constitute an assessment of the ignitability of the dust by taking into account the elements that compose the fire hexagon.
The sensitivity for dust ignition is an important aspect for the design of the industrial process equipment that is exposed to explosible atmospheres. With regard to the dissimilarities that were described in section 1.3 for the combustion of dust and gases, the interpretation of the flammability parameters must be different from the data of a combustible gas. The significant differences that are found in the heterogeneity of the mixtures and in the mechanisms of combustion demand the acquisition of the flammability data in a different way. They must be acquired according, not only to the procedures that have been standardized, but also according to the physical characteristics of the dust cloud.
This section briefly describes the main flammability parameters that characterize the ignitability of a combustible dust cloud:
• Minimum ignition temperature of a dust cloud.
• Minimum explosible concentration.
• Minimum ignition energy.
• Minimum oxygen concentration.
These ignitability parameters are complemented by the determination of the risk parameters associated to dust layers. The main risk that can be associated to dust layers corresponds to the development of secondary explosions. This condition corresponds to the suspension of a combustible dust that is caused by the turbulence generated by a preliminary explosion. This fact represents an escalation of the negative consequences through the development of a domino effect (Yuan et al., 2016). However, the experimental parameter that is considered to characterize the flammability of settled dusts is the minimum ignition temperature of dust layers. This parameter is envisaged as a basis of the determination of the safe operating conditions of locations of material usage and storage. Further information about this parameter and its relevance for the prevention systems of dust deposits is discussed in the international standard ASTM E2021-09.
MINIMUM EXPLOSIBLE CONCENTRATION (MEC) AND MINIMUM IGNITION ENERGY (MIE)
These ignitability parameters are usually determined with the same test apparatus under very similar procedures. The conditions of the flammability test establish that some characteristics of the system such as the temperature, pressure and humidity do not differ significantly from the environmental conditions that are usually considered as a reference. Thus, the capability of the dust cloud of initiating and propagating an explosion flame depends on the instantaneous condition of the mixture as well as certain physical properties of the combustible dust.
One of the flammability parameters that assess this characteristic of a combustible dust is its minimum explosible concentration (MEC). This limit value determines the lowest quantity of dust per unit volume that is capable of propagating a deflagration through a well dispersed mixture of the dust and air under the specified conditions of test (ASTM E1515 − 14). This parameter is adopted for the design of process equipment and prevention systems. Indeed, the implementation of the necessary procedures to keep a low dust concentration in fluidization and storage units.
The maximum dust concentration is not usually determined for two important reasons. At first, a high concentration in a dust cloud is not stable for long time periods. This fact is attributed to the agglomeration and sedimentation phenomena that is observed for the disperse particles. These occurrences reduce the current concentration and put the dust cloud within the explosible range of the solid material. For this reason, it is not possible to consider a system with a high load of dust as a safe unit. The second reason relies on the high amount of energy that can be stored by a dispersed solid during the explosion. The experimental evidences observed for various combustible powders pose that the energy balance between the energy released by the chemical reaction and the energy absorbed by the biphasic mixture is affected when there is an overload of the dust. This fact represents a diminution of the severity parameters of the dust and a low capability to propagate an explosion flame. Thus, the maximum explosible concentration is not considered as an important factor for the characterization of a combustible dust.
Furthermore, another parameter that assesses the likelihood of ignition of a dust cloud during the processing and handling is the Minimum Ignition Energy (MIE). This parameter is also used in the industry to evaluate the need for precautions such as explosion prevention systems.
The experimental determination of the MIE is performed in a similar way in which the MEC is obtained. These parameters are ascertained by producing a confined dust cloud that is ignited under controlled conditions with an ignitor whose energy release has been established previously. The ignition source can be defined from different systems according to the purpose of the characterization and the available apparatus. In accordance with this statement, one of the suitable sources arises from the electrical energy discharged from a capacitor. The MIE corresponds to the minimum electrical energy that is sufficient to effect ignition of the most easily ignitable concentration of fuel in air under the specific test conditions (ASTM E201λ − 03).
MINIMUM OXYGEN CONCENTRATION (MOC)
This parameter is also known as Limiting Oxygen Concentration (LOC). It corresponds to the oxygen (oxidant) concentration at the limit of flammability for the worst case (most flammable) fuel concentration. In addition, it can also be defined as the minimum concentration of the oxidant gas that can cause the flame propagation in an atmosphere composed by a dust cloud (Amyotte & Eckhoff, 2010). The knowledge of MOC is needed for safe operation of some chemical processes. This information may be needed in order to start up, shut down or operate a process while avoiding the creation of flammable dust-gas atmospheres therein, or to pneumatically transport materials safely. For this purpose, NFPA 69 provides guidance for the practical use of MOC data, including the appropriate safety margin to use.
The minimum oxygen concentration of combustible dust cloud constitutes an important aspect in the determination of the inerting levels that are usually considered for a process unit. Eckhoff (2004) posed the possibility to implement an evacuation system to reduce the concentration of the oxidant gas in the process equipment. Hence, this system can become an integral part of a combined solution to dust explosion protection technique (venting, automatic suppression, full confinement). The direct influence of the percentage of O2 on the other flammability parameters has been discussed by Eckhoff (2004). This author affirmed that both the ignition sensitivity and the explosion violence of the dust cloud are reduced when the oxygen content of the atmosphere is reduced by mixing inert gas with the air within the test apparatus. For this purpose, nitrogen is commonly used because other gases (e.g. carbon dioxide) might react with some metals such as aluminum, magnesium, titanium and zirconium (ASTM E2λ31 − 13).
This advantageous reduction of the severity of a dust explosion allows reducing the vent area of a specific enclosure. This fact is due to the decrease of the maximum pressure that is achieved when the oxygen concentration is diminished. In fact, the experiments discussed by Eckhoff (2004) pose that a diminution of the oxygen content from 21% to 18% and 16% reduced the required vent areas by factors of 0.6 and 0.37 respectively. However, it should be underlined that the efficiency of the inerting process greatly depends on the nature of the gas.
MAXIMUM PRESSURE AND MAXIMUM RATE OF PRESSURE RISE
The deflagration of a combustible dust is characterized at laboratory scale by analyzing the evolution of the pressure that is produced by the ignition of a well-dispersed dust cloud. This analysis evaluates the pressure increase that is produced in a confined test vessel by the impacting shock wave that is generated by an explosion. In this way, it is possible to determine the first two parameters that establish the severity of a dust explosion. The analysis of this information is a deciding factor for the characterization of the performance of its protection system (Bartknecht, 1989) .
Initially, it is possible to determine the maximum pressure ( Pmax ) that is achieved by the combustion process. This severity parameter is usually considered for the design of pressure resistant vessels and relief systems such as vents of rupture disks, but it can also be considered for the definition of other Standardized methods for the quantitative evaluation of the flammability and explosibility of the combustible dusts operating parameters such as the equipment isolation or the inerting procedures. This parameter is mainly defined by the thermodynamic properties of the combustible mixture.
The second explosibility parameter that is determined during the course of the flammability test is the maximum rate of pressure rise ( dP / dtmax ). This parameter is mainly associated to the kinetics of the combustion and the evolution of deflagration in the test vessel and is considered for the same purposes than the maximum pressure rise.
Both parameters are defined not only by the physical and chemical properties of the mixture, but also by the other components of the apparatus and the operating conditions of the test (e.g. ignition delay). This condition is due to the balance between the energy that is produced by the chemical reaction and the energy that is transferred by the reactive system to the cooling fluid the test apparatus. Besides, it is also necessary to take into account the effects of the ignition sources in order to correct the experimental results by neglecting the pressure increase that is caused by their activation.
The peak values of the explosion pressure and its rate of increase are obtained for the optimum concentration of the combustible dust. This value is usually unknown; hence the two flammability parameters are determined by a series of tests performed over a large range of concentrations. After determining the maximum rate of pressure increase ( ( dP / dt)max ), this parameter is normalized to a 1.0 m3 volume with the calculation of a third severity parameter. This additional data is denominated as the deflagration index (KSt) and is calculated in accordance with the ‘cube-root law’. This law defines a simple relationship that depends on the volume of the vessel to power of 1/3: 1/3 dP K St Vvessel 1.2 dtmax.
This law is used to scale up the standard test results from laboratory-sized vessels to plant-sized equipment. However, it is continuously controverted because of the inaccuracies that arise when the thickness of the propagation flame is significant with respect to the vessel radius (Dahoe et al., 1996).
Table of contents :
1 STANDARDIZED METHODS FOR THE QUANTITATIVE EVALUATION OF THE FLAMMABILITY AND EXPLOSIBILITY OF THE COMBUSTIBLE DUSTS
1.1 DUST EXPLOSIONS
1.2 THE DUST EXPLOSION HEXAGON
1.3 MECHANISMS OF DUST EXPLOSIONS
1.3.1 Kinetic mechanisms of dust combustion
1.3.2 Deflagrations and detonations in dust explosions
1.4 EXPERIMENTAL DETERMINATION OF THE DUST IGNITABILITY
1.4.1 MINIMUM IGNITION TEMPERATURE (MIT) OF A DUST CLOUD
1.4.2 MINIMUM EXPLOSIBLE CONCENTRATION (MEC) AND MINIMUM IGNITION ENERGY (MIE)
1.4.3 MINIMUM OXYGEN CONCENTRATION (MOC)
1.4.4 INTERNATIONAL STANDARDS
1.5 EXPERIMENTAL DETERMINATION OF THE EXPLOSIVITY OF A DUST CLOUD 22
1.5.1 MAXIMUM PRESSURE AND MAXIMUM RATE OF PRESSURE RISE
1.5.2 LAMINAR BURNING VELOCITY
1.5.3 INTERNATIONAL STANDARDS
1.6 INFLUENTIAL PARAMETERS
1.6.1 PARTICLE SIZE DISTRIBUTION (PSD)
1.6.2 INITIAL TURBULENCE
1.6.3 DUST FRAGMENTATION AND AGGLOMERATION
1.6.4 MOISTURE CONTENT
1.6.5 ADDITIONAL PARAMETERS
1.7 CRITICAL ANALYSIS OF THE STANDARDS
1.7.1 APPLICABILITY OF THE STANDARDS IN THE CHARACTERIZATION OF MICROMETRIC PARTICLES
1.7.2 EMERGING TOPICS: NANOMETRIC PARTICLES AND HYBRID MIXTURES .
1.9 LIST OF VARIABLES
2 NUMERICAL SIMULATIONS OF SOLIDS DISPERSIONS
2.1 COMPUTATIONAL STUDY OF A GAS-SOLID FLOW
2.1.1 Numerical methods applied for the description of a homogeneous or heterogeneous system
2.1.2 Selection criteria for the computational approach
2.1.3 Numerical description of the gas flow turbulence
2.1.4 Implementation of the DES model in the CFD simulations
2.1.5 Lagrangian approach for the description of the dispersion process of the combustible dust
2.2 APPLICATION OF THE COMPUTATIONAL FLUID DYNAMICS ON THE DESCRIPTION OF GAS-SOLID MIXTURES
2.3 APPLICATION OF THE COMPUTATIONAL FLUID DYNAMICS ON DUST EXPLOSIONS
2.3.1 Characterization of the flame velocity
2.3.2 Description of dust explosions with the FLACS-DustEx code
2.5 LIST OF VARIABLES
3 EXPERIMENTAL STUDY OF THE DUST DISPERSION AND ITS EFFECTS ON THE EXPLOSIBILITY PARAMETERS
3.1 COMBUSTIBLE DUSTS ANALYZED
3.1.2 Wheat starch
3.1.3 Adjustment of the particle size distribution to the Rosin-Rammler equation
3.2 GRANULOMETRIC ANALYSES
3.3 DIGITAL PARTICLE IMAGE VELOCIMETRY (DPIV)
3.3.1 Continuous wave laser
3.3.2 Tracer particles
3.3.3 Image analysis
3.4 DETERMINATION OF THE CHARACTERISTICS OF THE PRESSURIZED GAS INJECTION
3.4.1 Mass balance
3.4.2 Analysis of the high-speed videos
3.4.3 Transient pressure of the vessel
3.4.4 Transient pressure of the vessel
3.4.5 Gas velocity and Reynolds number
3.4.6 Pressure drop of the gas flow
3.4.7 Description of the gas injection into the modified Hartmann tube
3.5 DETERMINATION OF THE DUST DISPERSION CHARACTERISTICS INSIDE THE MODIFIED HARTMANN TUBE
3.5.1 Experimental setup
3.5.2 Set of experiments
3.6 DETERMINATION OF THE CHARACTERISTICS OF THE DUST DISPERSION INSIDE THE 20 L SPHERE
3.6.1 Experimental setup
3.6.2 Dispersion nozzles
3.6.3 Set of experiments
3.6.4 Experimental analyses
3.6.5 Determination of the ignition delay
3.8 LIST OF VARIABLES
4 CONFRONTATION OF THE COMPUTATIONAL AND EXPERIMENTAL DESCRIPTION OF THE DUST DISPERSION PROCESS
4.1 DESCRIPTION OF THE DUST DISPERSION IN THE MODIFIED HARTMANN TUBE
4.1.1 Description of the mesh
4.1.2 Boundary and initial conditions
4.1.3 Numerical parameters associated to the physics of the gas flow
4.1.4 Numerical parameters associated to the physics of the micrometric wheat starch
4.1.5 Results and comparison with the experimental approach
4.1.6 Application to the determination of the minimum ignition energy of the micrometric aluminum and wheat starch
4.2 DESCRIPTION OF THE DUST DISPERSION IN THE 20 L SPHERE
4.2.1 Description of the mesh
4.2.2 Numerical parameters associated to the physics and the discretization
4.2.3 Initial and boundary conditions
4.2.4 Results and comparison with the experimental approach
4.2.5 Application to the determination of the explosibility parameters of the micrometric wheat starch
4.3 LIST OF VARIABLES