Investigation of the Lower Concentration Flammability Limits of Hydrogen–Oxygen and Hydrogen–Air Mixtures at Reduced Pressures

V. F. Tsenin, O. A. Zakharov, V. A. Maraganov, V. B. Nemirovsky,
L. K. Parfenov

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

In developing measures to ensure the explosion safety of certain closed volumes in which explosive hydrogen–oxygen and hydrogen–air mixtures may form, it is important to know how a change in the total pressure in the mixture affects its lower concentration flammability limit—in particular, for a flame propagating from bottom to top at pressures below atmospheric.

It is known that, at atmospheric pressure, hydrogen flames are characterized by two lower flammability limits: 9% vol. H₂, corresponding to flame propagation from top to bottom, and 4% vol. H₂, corresponding to a nonisothermal flame propagating only from bottom to top [7].

In the case of 4% vol. H₂, the possibility of flame propagation in the mixture is associated with the phenomena of selective diffusion and convection [7].

A change in the pressure in the mixture generally leads to a definite change in the conditions of both diffusion and convection and may affect the value of the LCL H₂. A decrease in pressure has a definite effect on the kinetics of the combustion reaction and may also be the cause of a change in the LCL H₂.

The isolated data known from the literature on the results of measuring the LCL H₂ at pressures different from atmospheric either do not relate to the range of reduced pressures [1], or—

are of a qualitative character [2], [3], [4].

The absence of such data was pointed out by the authors of [5], who compiled a review of studies on hydrogen combustion carried out before 1959.

The experiments were conducted in an apparatus consisting of a steel explosion chamber 5 cm in diameter and 30 cm high, a mixing vessel with a mechanical stirrer, valves, and manometers. The ignition source for the mixture was a single electric spark of duration \(10^{-3}\) sec, with an energy content of \(\sim 100\) mJ. The length of the spark gap was 7 mm, which corresponds to the quenching distance for limiting hydrogen mixtures [6].

In all experiments, before being admitted into the explosion chamber, the mixture was stirred for 5 minutes; the explosion chamber and all lines were evacuated to a pressure of \(10^{-1}\) mm Hg.

The mixture prepared in the mixer was sufficient to carry out 6–8 admissions into the chamber. As a rule, in each series of admissions the pressure in the explosion chamber was gradually decreased (i.e., a section was taken over pressures for a mixture of constant composition).

The criterion for ignition of the mixtures was the flame front reaching the upper end of the explosion chamber. Establishment of ignition and observation of combustion were carried out visually (only under conditions of complete darkening), and also with the aid of an open mercury manometer. It was specially established that, in the absence of combustion, from the action of a single spark alone, oscillation of the mercury meniscus is practically not observed. When the mixture burned, the amplitude of the pressure pulse was 5–30 mm Hg and more.

At reduced pressures, the main source of information was the glow of the propagating flame front (the glow is sufficiently contrasty for observation, and the amplitude of meniscus oscillation is very small); at pressures close to atmospheric—

... to the sensor; on the contrary (the glow disappears while still small, and the oscillation amplitude of the mercury meniscus is maximal).

The range of pressures investigated was 760–160 mm Hg. First of all, several series of experiments were carried out to establish the region in which the lower concentration limit is located. It was found that in the region from 300–400 mm Hg to atmospheric pressure, ignitions began at 4.1–4.2% vol. hydrogen; at pressures below 300 mm Hg, ignition of the mixture and propagation of the flame over the entire length of the chamber were observed at 4.0% vol. hydrogen and even at 3.9% vol.

In order to systematize the investigation of the boundary zone—between the explosion-hazardous and explosion-safe regions—subsequent experiments were carried out for the following selected pressures: 760, 600, 480, 370, 270, 210, and 160 mm Hg, and for concentrations: 3.8, 3.9, 4.0, 4.1, 4.2, 4.3% vol. H₂; thus, for one point corresponding to a certain pressure (for example, 470 mm Hg) and a certain hydrogen concentration (for example, 4.1% vol. H₂), one had to perform a fairly large number of experiments, on the order of ten.

The results of the experiments are shown in Figs. 1 and 2. The numbers placed next to circles indicate the number of failures in each of the selected points; the numbers next to crosses indicate the number of ignitions at these points.

It follows from Fig. 1 that in the region from 760 to 400 mm Hg, ignition of the mixture is unlikely at hydrogen concentrations below 4.1% vol. On the other hand, a tendency is observed toward improved conditions for ignition of the mixture as the pressure is decreased (at 4.1% vol. H₂, at \(P = 760\) mm Hg there were 4 failures; at \(P = 600\) mm Hg, 2 failures; at \(P = 480\) mm Hg, 3 failures; at \(P = 370\) mm Hg, 1 failure). Taking into account that the average frequency of ignition of the mixture at 4.1% vol. hydrogen in the indicated region is \( \frac{7}{10} \div \frac{8}{10} \), it may be considered that for the region 760–350 mm Hg, LCL\(_{H_2}\) is about 4.1% vol., which ...

Fig. 1. Dependence of the LCF of H₂ on the total pressure in a hydrogen–oxygen mixture.

Fig. 2. Dependence of the LCF of H₂ on the total pressure in a hydrogen–oxygen–nitrogen mixture
\((50\ \mathrm{vol.\%}\ \mathrm{H}_2)\).

corresponds to the data of many literature sources for the hydrogen–oxygen mixture investigated at atmospheric pressure, [4, 5], etc.

The tendency toward improved flammability of the mixture with decreasing pressure becomes stronger when the pressure is reduced to 160 mm Hg: in this region, at 4.1 vol.% H₂, no failures were observed. Moreover, in the region \(P = 300\text{–}160\) mm Hg, ignitions were quite often observed at 4.0 vol.% and 3.9 vol.% hydrogen. From Fig. 1 it follows that, in the region of about 350 mm Hg, the LFL\(_{H2}\) shifts toward a decrease, by \(\sim 0.1\) vol.%.

From Fig. 2 it follows that the dependence of LFL\(_{H2}\) on pressure for the hydrogen–air mixture (the mixture contained somewhat more nitrogen than air, namely: 80 vol.%) remains essentially the same as for the hydrogen–oxygen mixture: in the region of 760–350 mm Hg the value of LFL\(_{H2}\) is observed to be 4.1 vol.%; in the region of 300–160 mm Hg, about 4.0 vol.%, at approximately the same ignition frequency as in the case of the H₂–O₂ system.

As for the magnitude of LFL\(_{H2}\) in the region of atmospheric pressure (4.1 vol.%), this was to be expected, according to the literature data, [3], etc. With respect to LFL\(_{H2}\) in the region of reduced pressures, it can thus be stated that dilution of the H₂–O₂ system with nitrogen to 80 vol.% has practically no effect on the course of the dependence of LFL\(_{H2}\) on pressure: as in the case of the pure hydrogen–oxygen mixture, the LFL in the pressure region 300–160 mm Hg decreases by approximately 0.1 vol.% H₂ and is \(\sim 4.0\) vol.%.

CONCLUSIONS

1. The effect of lowering the pressure below 760 mm Hg on the concentration flammability limits of H₂–O₂ and H₂–O₂–N₂ mixtures has been investigated.

2. It has been established that, when the pressure is reduced to a certain value (400–350 mm Hg), the lower flammability limit

remains the same as at atmospheric pressure (\(\sim 4.1\%\) vol. \(\mathrm{H_2}\)); with a further decrease in pressure to 160 mm Hg, the limit decreases somewhat and becomes equal to \(4.0\%\ \mathrm{H_2}\).

1. Dilution of the limiting hydrogen–oxygen mixture with nitrogen to 80% vol. has practically no effect on the course of the dependence of the mixture’s LCFL\(_2\) on pressure in the pressure range investigated.

REFERENCES

1. S. A. Yantovskii, M. V. Chernyak, “Concentration range of explosibility of hydrogen–oxygen and hydrogen–air mixtures at pressures above atmospheric,” ZhFKh, vol. X, issue II, 1966, p. 2899.

2. L. N. Khitrin, Physics of Combustion and Explosion, 1957, Moscow State University Press.

3. Coward H. F. and Jones G. W., “U.S. Bureau of Mines Bull.”, 503, 1952.

4. A. Egerton, “Limits of Inflammability,” 4th International Symposium on Combustion and Detonation Waves, Oboronizdat, 1958.

5. J. L. Drell, F. E. Belles, “Survey of hydrogen combustion properties,” Rep. NACA, 1383.

6. B. Lewis and G. Elbe, Combustion, Flames, and Explosions of Gases, Moscow, 1968.

7. R. Markstein, “The phenomenon of flame-front instability,” 4th International Symposium on Combustion and Detonation Waves, Oboronizdat, 1958.