Ignition Sources of Gas Atmospheres in the Study of Their Flammability

V. A. Marasanov, D. A. Zakharov, V. F. Panin, A. V. Melnik

(Presented at the scientific seminar of the Department of Radiation Chemistry)

On the question of choosing a method for testing gas atmospheres for explosion hazard.

The explosion and fire hazard of atmospheres in enclosed volumes is one of the most pressing issues relating to the operating practice of many objects of new technology. During operation, combustible vapors and gases are released into the initially explosion-safe atmosphere of such objects. Over time, the atmosphere of the object becomes explosion-hazardous.

For the purposes of safety engineering and the design of explosion-protection equipment operating in such atmospheres, it is necessary to know the values of the concentration limits (CL) and the minimum ignition energies of each ignitable composition. Quite often the composition of the atmosphere is complex, and the nature of the influence of combustible impurities on the CL and on the value of the minimum ignition energy of the object’s atmosphere is difficult to predict on the basis of literature data, and in some cases impossible. Therefore, as a rule, the CL and minimum ignition energies are determined experimentally. To carry out experiments, it is necessary to choose an ignition method that would make it possible to obtain minimum values of the CL and ignition energies, i.e., that would characterize the degree of greatest hazard of the system.

Ignition of gases is usually considered as a process,

…causing and ensuring the propagation of flame in the combustible mixture. This process passes through two stages: in the first, the energy of the source is supplied to the explosive medium; in the second, such development of the flame front takes place that the rate of its propagation no longer depends on the parameters of the source [1]. The determining factors in ignition are associated with the properties of the source and with the conditions of flame propagation. To ensure ignition at energies comparable with the theoretical ones, it is necessary, in volumes comparable with the width of the combustion zone, to create the conditions for the front of a self-propagating flame. With sufficient power of the ignition source for a gas mixture, the ignition limits become the concentration limits of flame propagation [2].

Depending on the method of ignition, there are many methods for testing gas atmospheres for flammability. As the ignition source it is possible to use: a heated surface (incandescent wires, balls, rods), a flame, inductive and capacitive sparks, etc. All of them differ in their energy capabilities and in the nature of energy release. The task of the study is reduced to the selection of such methods that would ensure obtaining the minimum values of the CP and ignition energies both at normal and at reduced pressures.

The most probable source of ignition of atmospheres in industry is electrical sparks from static discharges, normal switching operations, emergency conditions, and overheated surfaces. The appearance of a torch may also occur as a consequence of the ignition of overheated flammable materials.

Let us dwell in more detail on the characterization of overheated surfaces and electrical sparks as the most probable sources of ignition of flammable atmospheres in industry. In ignition from heated surfaces, it is necessary that the heat release exceed the heat losses and that the limiting temperature of the source be no lower than the autoignition temperature [3]. The temperature of a heated surface introduced into a gaseous medium must be the higher, the smaller the dimensions…

of this surface [4]. As the dimensions of the source increase, the temperature gradient decreases, which promotes ignition, but at the same time it becomes difficult to account for the energy dissipated in the body and in the supply line due to heat capacity and thermal conductivity, and the energy expended directly on heating the gas mixture [5].

In this connection a contradiction arises: in order to obtain the minimum values of the FLC, it is necessary to increase the power of the source (the heated body), which leads to an increase in the size of the heated surface, and this in turn creates difficulties in determining the minimum ignition energies.

When an electric spark is used as the ignition source, these difficulties do not arise. The power of a source of electric sparks can be varied over very wide limits, and the energy released in the electric spark is determined by oscillography. Studies carried out in [6] showed that, in determining the FLC at reduced pressures, the most effective ignition source is a pulse generator. On the basis of the foregoing, we come to the conclusion that, in determining the FLC and ignition energies over a broad range of gas-mixture pressures, the electric-spark ignition method is universal, making it possible simultaneously to determine the FLC and the minimum ignition energies of gas mixtures. Therefore we shall dwell in greater detail on the characteristics of electric-spark ignition of gas mixtures.

SPARK IGNITION

Ignition by a spark is a complex process accompanied by many chemical and physical phenomena. In simplified form, a spark channel may be represented as a kind of heated gas-like body with a very high temperature, reaching up to \(20000^\circ\mathrm{C}\) [7,8]. In the spark channel there arises intense local—

excitation of molecules and their ionization. All this promotes the formation of an initial flame kernel. The initial kernel of hot gas is regarded \([9,10]\) as a sphere. The amount of heat in the sphere from the burned gas is proportional to its volume. Heat loss by radiation and heat transfer is proportional to the surface of the sphere. The surface temperature will be determined by the balance between the heat obtained in the sphere and the heat lost through the surface. The kernel develops into a self-propagating flame when the surface temperature is high enough to impart to the surrounding gas molecules the energy necessary to sustain the reaction. Therefore, not every electric discharge causes ignition of the combustible mixture around the spark; the development of the flame kernel that has arisen after the discharge ceases may fail to occur because of heat losses. Only when the flame kernel reaches a certain critical volume, at which the losses are compensated and the necessary combustion temperature is maintained, can it prove capable of self-propagation. The energy required to form a sphere of critical volume determines the minimum amount of spark energy leading to ignition of the gas mixture. The existence of a “critical” kernel has been confirmed experimentally by means of shadow photographs \([11]\).

In \([12]\) it was shown that the radii of kernels increase with time; moreover, the radii of kernels that cause ignition of the entire gas mixture are, at any instant, greater than the radii of kernels that do not cause ignition (Fig. 1). In the same work it was shown that if, at some instant \(t_0\), the radius of the kernels is greater than a certain critical value, then they will grow; otherwise the kernels gradually die out. The time \(t_0\) is of the order of \(10^{-4}\) sec, and it is important that the release of energy occur precisely during this time.

For each gas mixture there is a limiting minimum energy of the electric spark that causes ignition—

of the mixture. The existence of a true minimum energy is considered established [13].

Studies of the minimum energy have shown that the character of the dependence of the minimum energy on mixture composition is the same for all combustible mixtures. As the spark power increases, the ignition limits expand. But the expansion of the ignition limits is asymptotic in character. There is a limit to the increase in spark power at which any further increase in power does not lead to an expansion of the ignition limits. A spark of such power is commonly called a saturating spark. A conclusion as to whether one is dealing with a saturating spark can be made only on the basis of special experiments [14].

Theoretical calculation of the minimum energy does not give satisfactory agreement with experiment. Discrepancies amount to several tens of times [5], which is unacceptable for practice. A more reliable method for determining the minimum energy is experimental; however, in doing so it is necessary to take into account the influence of various factors on the magnitude of the minimum ignition energy, such as: pressure, the dimensions of the spark gap, the material and shape of the electrodes, the composition and temperature of the mixture, the type of discharge, and the time over which the energy is introduced [11].

An analysis of results concerning the influence of pressure on the magnitude of the ignition energy, carried out in [2], shows that as the pressure is lowered the minimum energy and the quenching distance increase. However, at each fixed pressure there is an interval between the electrodes at which the ignition energy remains unchanged. The character of the dependence of the energy on the distance between the electrodes is the same for all gas systems. Figure 2 presents the dependence of the ignition energy on the distance between the electrodes for a hydrogen–air mixture [15]. It is clear from the graphs that the electrodes affect the magnitude of the ignition energy up to certain distances. With a further increase in the distance between the electrodes, the ignition energy

remains approximately at the same level. This value of the energy is taken as the minimum ignition energy. The extinguishing effect in electrodes with dielectric flanges (curve 1) is expressed more strongly than in electrodes without flanges (curve 2). The diameter of the electrodes has no substantial effect on the magnitude of the minimum energy beyond the extinguishing distance [5]. As the initial temperature of the ignitable mixture is increased, the ignition energy decreases [16].

The question of the influence of the duration of the electric spark on the ignition energy has not been covered sufficiently fully in the literature to permit any conclusions to be drawn. It should be noted, however, that the optimum pulse duration for igniting gas mixtures when determining minimum energies, as noted above, is of the order of \(10^{-4}\) sec.

In [11] it is noted that the form of electrical discharges also affects the magnitude of the ignition energy. Capacitive discharges have a lower ignition energy; with arc discharges the ignition energy is 1.5 times greater, and with glow discharges it is 2–3 times greater.

Fig. 1. Growth of the flame kernel with time

Labels in the figure:

• \(R\), cm
• \(t\), ms
• \(\varphi = 1.0\), propane–air
• chemical radius
• gases that cause ignition
• gases that do not cause ignition

Fig. 2. Dependence of the ignition energy of a hydrogen–air mixture on the distance between electrodes.

Labels in the figure:

• \(E\), mJ
• \(d\), mm

Literature

1. L. N. Khitrin, S. A. Goldenberg. Izd. AN SSSR, OTN, 1957, No. 3.

2. E. S. Shchetinkov. Physics of Gas Combustion. Moscow, “Nauka,” 1965.

3. H. Elbe. The problem of ignition. Fourth International Symposium on Combustion and Detonation Waves, Moscow, 1958.

4. B. Lewis. Explosions and Combustion of Gases. Moscow, IL, 1952.

5. B. Lewis, G. Elbe. Combustion, Flames, and Explosions of Gases. Moscow, “Mir,” 1968.

6. M. Zabetakis, J. Richmond. Problems of combustion and detonation waves. Oborongiz, 1958.

7. N. Sobolev. Investigation of excitation of atomic spectra in a condensed spark. ZhTF, vol. XIII, No. 5, 1953.

8. Yu. M. Pchelkin. Combustion Chambers of Gas-Turbine Engines. Moscow, 1967.

9. Ya. B. Zel’dovich, N. N. Simonov. ZhTF, vol. XII, No. 11, 1949.

10. G. Elbe, B. Lewis. In the collection Problems of Ignition and Flame Stabilization. Edited by S. A. Goldenberg. Moscow, 1963.

11. V. S. Kravchenko. Scientific Research in England in the Field of Intrinsically Safe Application of Electrical Energy in an Ignitable Atmosphere. Moscow, 1964.

12. D. Arnold, R. Sherburn. Fourth International Symposium on Combustion and Detonation Waves. Moscow, 1958.

13. D. J. Whittington. Ignition of methane–air mixtures under the action of electric discharges. Report at the 3rd International Congress, Salzburg, 1963.

14. L. N. Khitrin. Physics of Combustion and Explosion. Izd. MGU, 1957.

15. V. S. Kravchenko, V. S. Komarov, T. V. Belikova. Industrial Occupational Safety, 1964, No. 7.

16. K. S. Svetts. In the collection Problems of Ignition and Flame Stabilization. Edited by S. A. Goldenberg. Moscow, 1963.