STUDY OF THE COMBUSTION PROCESS OF A COMPLEX GAS MIXTURE

G. D. Speshchii, N. N. Norkin

(Presented by Prof. I. V. Gebler, Doctor of Technical Sciences)

Questions connected with the sequence of combustion of components in complex gas mixtures have been studied by many researchers [1].

The method of analyzing intercone products to determine the sequence of combustion of mixture components was first used by Smithells and Ingle [1].

This method consists in separating the inner and outer cones of the flame by means of a glass or quartz tube connected to a gas analyzer for the analysis of intermediate substances.

This method had already been used by us in the study of a water-gas flame [2]. The present work was carried out with the aim of investigating the combustion process of a complex gas mixture, determining the order of combustion of the mixture components, the influence of excess or deficiency of air on the order of combustion, and studying the changes occurring between the flame cones.

The results of our investigation may serve as additional factual material supporting a definite point of view on the combustion process of a complex gas mixture. Since the work touches upon questions of the chemical thermodynamics of gas reactions, it should have a certain theoretical interest. The results of the work may also have practical significance in those cases where two-cone combustion of gas occurs, that is, under conditions of use of atmospheric burners.

For the work, illuminating gas was used, obtained at the gas plant of the Tomsk Polytechnic Institute; its average composition was as follows: CO₂—5.7%, CₙH₂ₙ—2.7%, O₂—0.7%, H₂—49.3%, CH₄—25.7%, CO—9.3%, N₂—6.6%.

The scheme of the experimental setup is presented in Fig. 1. Gas and air are supplied respectively by gas blower 1 and air blower 2 through a system for purification, drying, elimination of pressure fluctuations, measurement of flow rate, and equalization of temperature to ejector-type mixer 17 and burner with nozzle 18. The burner is a brass tube with an internal diameter of 8 mm. Onto the end of the tube, closed with a screen, is fitted a glass attachment with an internal diameter of 28.5 mm, with three side outlets at distances of 8, 40, and 70 mm from the burner mouth for sampling intercone gases for analysis.

Fig. 1. 1—gas blower; 2—air blower; 3—transformer; 4—manometers; 5—column with concentrated sulfuric acid; 6—column with caustic soda; 7—droplet trap; 8—drying column; 9—gas vessel; 10—air vessel; 11, 12—valves for regulating the flow rate of air and gas; 13, 14—rotameters; 15—thermostat; 16—temperature regulator; 17—mixer; 18—burner with attachment; 19—aspirator; 20—valve for taking a sample of the mixture for analysis.

The outlets are connected to wet aspirator 19. The gas analysis was carried out with a VTI gas analyzer with a platinum capillary for afterburning CO, H₂, and CH₄, instead of a loop with copper oxide and a cylinder with a coil. After preliminary experiments, the lower outlet was not used, since the composition of the gases taken at the level of the lower cone proved to be nonconstant. Evidently, this is explained by the formation of “dead” zones in the lower part of the tube, where gases can stagnate, and by phenomena associated with instability of the flame front.

The experiments were carried out with different amounts of air in the mixture. After adjustment of the gas and air flow rates, when the divided combustion became steady, the intercone gases, the mixture, and the gas were sampled; subsequently, when it had been established that the ratio between the amounts of gas and air in the mixture, determined from the readings of the rheometers and from the results of analyses of the gas and mixture, practically coincided, it proved possible to restrict the sampling to the gas alone (or to the mixture alone).

It was possible to bring the fraction of primary air up to 1.53; with any further increase the combustion became unstable, the cones joined inside the separating tube, and “flame breakthrough” occurred.

For comparison of the compositions of the intercone gases determined by analysis with equilibrium compositions, equilibrium compositions of the intercone gases were determined by calculation for experiments with different excess-air coefficients $\alpha$ ($\alpha > 1$; $\alpha \approx 1$; $\alpha < 1$). In order to obtain comparable results, experiments most similar in the quantitative ratio of the components in the initial combustible mixtures were selected for the calculation. The calculations of equilibrium compositions were carried out using expressions for the equilibrium constants and equations of material balance for each element. The initial data used for the calculation were: the composition of the initial mixture, the pressure of the mixture, and the gas temperatures inside the flame-separating tube [3].

Under the conditions of the experiments performed, the combustion process proceeded at a constant pressure equal to atmospheric pressure. The temperature of the gases immediately surrounding the surface of the inner cone of the flame was 930°C, and the temperature at the level of the middle and upper gas-sampling tubes was 330°C, differing only very little, by no more than 15–20°.

The largest number of equations is required for the case with $\alpha < 1$; the following were taken as such:

$$ \frac{p_{\mathrm{H_2}} + p_{\mathrm{H_2O}} + 2p_{\mathrm{CH_4}}} {p_{\mathrm{CO_2}} + p_{\mathrm{CO}} + p_{\mathrm{CH_4}}} = \frac{[m]_{\mathrm{H}}}{[m]_{\mathrm{C}}}; \tag{1} $$

$$ \frac{ p_{\mathrm{CO_2}} + \frac{1}{2}p_{\mathrm{CO}} + \frac{1}{2}p_{\mathrm{H_2O}} } {p_{\mathrm{CO_2}} + p_{\mathrm{CO}} + p_{\mathrm{CH_4}}} = \frac{[m]_{\mathrm{O_2}}}{[m]_{\mathrm{C}}}; \tag{2} $$

$$ p_{\mathrm{CO}} + p_{\mathrm{CO_2}} + p_{\mathrm{CH_4}} + p_{\mathrm{H_2O}} + p_{\mathrm{H_2}} + p_{\mathrm{N_2}} = p_0; \tag{3} $$

$$ K_5 = \frac{p_{\mathrm{CO_2}} \cdot p_{\mathrm{H_2}}} {p_{\mathrm{CO}} \cdot p_{\mathrm{H_2O}}}; \tag{4} $$

$$ K_6 = \frac{p_{\mathrm{CH_4}} \cdot p_{\mathrm{H_2O}}} {p_{\mathrm{CO}} \cdot p_{\mathrm{H_2}}^{3}}. \tag{5} $$

\[ K_8=\frac{p_{\mathrm{CO}_2}\cdot p_{\mathrm{H}_2}^{4}}{p_{\mathrm{CH}_4}\cdot p_{\mathrm{H}_2\mathrm{O}}^{2}}. \tag{6} \]

In these equations:

\(p_{\mathrm{H}_2}, p_{\mathrm{H}_2\mathrm{O}}, p_{\mathrm{CO}}, p_{\mathrm{CO}_2}, p_{\mathrm{CH}_4}, p_{\mathrm{N}_2}\) are the partial pressures of the corresponding components in the mixture;

\(p_0\) is the total pressure of the mixture;

\([m]_i\) is the number of gram-atoms of each element in the system;

\(K_5, K_6, K_8\) are the equilibrium constants, respectively, of the reactions:

\[ \mathrm{CO}+\mathrm{H}_2\mathrm{O}\rightleftarrows \mathrm{CO}_2+\mathrm{H}_2 \tag{4} \]

\[ \mathrm{CO}+3\mathrm{H}_2\rightleftarrows \mathrm{CH}_4+\mathrm{H}_2\mathrm{O}; \tag{5} \]

\[ \mathrm{CH}_4+2\mathrm{H}_2\mathrm{O}\rightleftarrows \mathrm{CO}_2+4\mathrm{H}_2. \tag{6} \]

The results of the calculations are presented in the form of a table, in which, for comparison, the compositions of the intercone gases determined by analysis for the same experiments are given.

Table 1

From Table 1 it is evident that the actual composition of the intercone gases is far from equilibrium. The high rate of occurrence of thermodynamically possible reactions in gases at high temperatures gives grounds to assume that combustion should lead to the establishment of chemical equilibrium in the mixture of products formed. Under our conditions combustion takes place in a stream; the temperature changes along the height of the separating tube, and this apparently prevents the establishment of an equilibrium state.

The presence of combustible components in the intercone gases, provided that the initial mixture contains an amount of air sufficient for complete combustion, is evidently explained by the course of secondary reactions, as a result of which combustible components are formed again.

In the intercone gases, in all experiments, unsaturated hydrocarbons were not detected, which indicates their preferential burnout.

In all experiments oxygen was found in small amounts in the intercone gases, both at \(a > 1\) and at \(a < 1\).

To determine the sequence and completeness of burnout of the components, comparisons were made of the ratios of their amounts in the initial mixture and in the intercone gases. The quantity characterizing the change in the content of a given component in the combustion products, as compared with its content in the initial mixture, we shall conventionally call “combustibility.”

To compare the “combustibility” of \(\mathrm{CH}_4\) and \(\mathrm{CO}\), the ratio

\[ \frac{\mathrm{CH}_4}{\mathrm{CO}} \]

was taken for the initial mixture and for the intercone gases from the middle and upper taps. From analysis of these quantities for all experiments it is seen:

a) the content of CO in the initial mixture is, as a rule, less than that of \(\mathrm{CH}_4\)

\[ \left(\frac{\mathrm{CH}_4}{\mathrm{CO}} > 1\right); \]

b) in the intercone gases from the middle tap, in most cases,

\[ \frac{\mathrm{CH}_4}{\mathrm{CO}} < 1; \]

c) in the intercone gases from the upper tap the CO content is still greater than in the initial mixture;

d) with decreasing \(a\),

\[ \frac{\mathrm{CH}_4}{\mathrm{CO}}\,\text{initial} \bigg/ \frac{\mathrm{CH}_4}{\mathrm{CO}}\,\text{middle} \]

increases

\[ \left( \frac{\mathrm{CH}_4}{\mathrm{CO}}\,\text{initial} \bigg/ \frac{\mathrm{CH}_4}{\mathrm{CO}}\,\text{upper} \right. \]

increases still more), the \(\mathrm{CH}_4\) content decreases, hence, with decreasing \(a\), the combustibility of \(\mathrm{CH}_4\) increases and the combustibility of \(\mathrm{CH}_4\) is greater than that of CO.

The ratio

\[ \frac{\mathrm{CH}_4}{\mathrm{CO}+\mathrm{H}_2} \]

for the intercone gases is, as a rule, smaller than for the mixture, and for the intercone gases from the upper tap it is smaller than for the intercone gases from the middle tap.

Thus, in the direction from the middle to the upper tapping tube in the intercone space, the content of \(\mathrm{CH}_4\) decreases and the content of CO and \(\mathrm{H}_2\) increases.

The ratio

\[ \frac{\mathrm{CH}_4}{\mathrm{CO}+\mathrm{H}_2}\,\text{initial} \bigg/ \frac{\mathrm{CH}_4}{\mathrm{CO}+\mathrm{H}_2}\,\text{upper} > \frac{\mathrm{CH}_4}{\mathrm{CO}+\mathrm{H}_2}\,\text{initial} \bigg/ \frac{\mathrm{CH}_4}{\mathrm{CO}+\mathrm{H}_2}\,\text{middle} \]

also grows with decreasing \(a\), which again indicates the high combustibility of \(\mathrm{CH}_4\) and its increase with decreasing \(a\).

From a comparison of the values of the ratio \(\dfrac{\mathrm{H}_2}{\mathrm{CH}_4}\) for all experiments it is evident:

a) for all initial mixtures and intercone gases from the upper tap,

\[ \frac{\mathrm{H}_2}{\mathrm{CH}_4} > 1, \]

while in the intercone gases from the middle tap,

\[ \frac{\mathrm{H}_2}{\mathrm{CH}_4} \gtrless 1. \]

For all experiments the methane content in the middle of the tube is less than in the mixture, and at the top it is less than in the middle, while the content of \(\mathrm{H}_2\) and CO in the middle is less than in the mixture, and increases toward the top;

b) in the lower part of the flame-separating tube a greater combustibility of \(\mathrm{H}_2\) is observed, possibly owing to the higher burning velocity of hydrogen;

c) ultimately \(\mathrm{CH}_4\) burns out to a greater extent, and its combustibility increases with decreasing \(a\).

From a comparison of the values of the ratio \(\dfrac{\mathrm{H}_2}{\mathrm{CO}}\) for all experiments it is evident:

a) the ratio

\[ \frac{\mathrm{H}_2}{\mathrm{CO}} \]

for the initial mixture is considerably greater than unity, while in the combustion products it is about unity. This undoubtedly indicates a greater combustibility of \(\mathrm{H}_2\) in comparison with CO;

b) the ratio

\[ \frac{\mathrm{H}_2}{\mathrm{CO}}_{\text{mid.}} \gtrless \frac{\mathrm{H}_2}{\mathrm{CO}}_{\text{upper}}, \]

but in the upper part of the flame-separating tube, as a rule, both CO and \(\mathrm{H}_2\) are greater;

Fig. 2. Dependence of the ratios of the amounts of components in the mixtures on the excess-air coefficient.

\[ 1 — \frac{\mathrm{CH}_4}{\mathrm{CO}}\ \text{init.}\ \bigg|\ \frac{\mathrm{CH}_4}{\mathrm{CO}}\ \text{upper}, \qquad 2 — \frac{\mathrm{CH}_4}{\mathrm{CO}+\mathrm{H}_2}\ \text{init.}\ \bigg|\ \frac{\mathrm{CH}_4}{\mathrm{CO}+\mathrm{H}_2}\ \text{upper}, \]

\[ 3 — \frac{\mathrm{H}_2}{\mathrm{CH}_4}\ \text{init.}\ \bigg|\ \frac{\mathrm{H}_2}{\mathrm{CH}_4}\ \text{mid.}, \qquad 4 — \frac{\mathrm{H}_2}{\mathrm{CH}_4}\ \text{init.}\ \bigg|\ \frac{\mathrm{H}_2}{\mathrm{CH}_4}\ \text{upper}. \qquad 5 — \frac{\mathrm{H}_2}{\mathrm{CO}}\ \text{mid.} \]

c) the ratio $\dfrac{\mathrm{H}_2}{\mathrm{CO}}$ avg. at $\alpha \gg 1$ is less than the ratio $\dfrac{\mathrm{H}_2}{\mathrm{CO}}$ avg. at $\alpha < 1$, which indicates a decrease in the combustibility of $\mathrm{H}_2$ with decreasing $\alpha$.

Analysis of the results of the comparison leads to the conclusions:

1. The combustibility of $\mathrm{CH}_4$ is greater than that of $\mathrm{CO}$ and increases as $\alpha$ decreases.
2. The combustibility of $\mathrm{CH}_4$ is greater than that of $\mathrm{H}_2$.
3. The combustibility of $\mathrm{H}_2$ is greater than that of $\mathrm{CO}$.

The dependencies given are presented graphically in Fig. 2. If one judges by the compositions of the mixtures taken from the middle tap, the combustibility of $\mathrm{H}_2$ is greater than the combustibility of $\mathrm{CH}_4$. Upward, the content of $\mathrm{CH}_4$ becomes smaller, and that of $\mathrm{CO}$ and $\mathrm{H}_2$ greater, evidently owing to the reaction

\[ \mathrm{CH}_4 + \frac{1}{2}\mathrm{O}_2 \rightleftharpoons \mathrm{CO} + 2\mathrm{H}_2. \]

The greater the content of $\mathrm{CH}_4$ in the initial mixture, the greater proves to be the amount of $\mathrm{H}_2$ and $\mathrm{CO}$ in the upper part of the flame-separating tube. Proceeding from this, the following sequence of combustion may be proposed: $\mathrm{C}_2\mathrm{H}_4—\mathrm{CH}_4—\mathrm{H}_2—\mathrm{CO}$.

The sequence and completeness of combustion can be determined theoretically from the chemical affinity for oxygen, that is, in other words, from the maximum work of the reactions. Since under our conditions all experiments were carried out at constant pressure and temperature, the expression $A_{\mathrm{M}} = -\Delta Z_0$ was used, where $\Delta Z_0$ is the change in the standard isobaric potential.

Since the affinity for oxygen is manifested most noticeably when oxygen is deficient, the experiments with the smallest $\alpha$ were taken.

Table 2

From a comparison of the magnitudes, given in Table 2, of the maximum works for reactions thermodynamically possible at $1200^\circ\mathrm{K}$, it is seen that reactions (1) and (2) should proceed to the greatest extent, but when oxygen is deficient, evidently, it is mainly not (2) but (3) that occurs, and as a result $\mathrm{CO}$ and $\mathrm{H}_2$ are formed. Then, in decreasing order of maximum work, follow reactions (4), (5), (6), (7), and (8).

Thus, in the equilibrium mixture of combustion products there is no $\mathrm{CH}_4$, there is $\mathrm{H}_2$, and there is even more $\mathrm{CO}$ than in the initial mixture. In terms of the magnitude of chemical affinity for oxygen the components are arranged as follows: $\mathrm{C}_2\mathrm{H}_4—\mathrm{CH}_4—\mathrm{H}_2—\mathrm{CO}$, that is, the previously proposed arrangement is confirmed.

When the values of the maximum works at \(600^\circ\mathrm{K}\) are compared, a different arrangement is obtained. For the attainment of equilibrium, it is chemical affinity, rather than the rate of reactions, that is of greater importance; and at \(600^\circ\mathrm{K}\) the chemical affinity for oxygen is greater for CO than for \(\mathrm{H}_2\).

M. Kh. Karapet’yants (4) points out that the values of equilibrium constants obtained by calculation cannot be used to explain the magnitudes of yields in complex processes without taking into account the rate, which may determine a greater or lesser yield of particular products. As is known, \(\mathrm{H}_2\) and CO are distinguished by a high combustion rate; for \(\mathrm{H}_2\) the rate is approximately 3 times greater than for CO. The combustion rate of hydrocarbons is lower than the combustion rates of \(\mathrm{H}_2\) and CO. Undoubtedly this affects the results, as does, possibly, an insufficient completeness of mixing of the gas with air.

The following course of the process may be assumed: hydrogen in the lower part of the flame-separating tube, at high temperature, burns very rapidly; CO also burns; \(\mathrm{H}_2\mathrm{O}\) and \(\mathrm{CO}_2\) are formed; a certain amount of the latter is also formed as a result of the combustion of \(\mathrm{C}_2\mathrm{H}_4\). \(\mathrm{H}_2\) and CO are then formed again in the course of reactions (3), (6), (7), and (8). In the region of lower temperatures (in the upper part of the flame-separating tube) the combustion rate decreases (in comparison with CO); the composition of the intercone gases changes in the direction of an increase in the content of \(\mathrm{H}_2\) and CO and a further decrease in the content of \(\mathrm{CH}_4\), whose affinity for oxygen is considerably greater.

REFERENCES

1. W. Bone and D. Townend. Flame and combustion in gases. Lnd. 1927.
2. N. N. Norkin and G. D. Spekts. Study of the structure of a water-gas flame. Izvestiya TPI, vol. 66, no. 3, 1948.
3. A. M. Gurvich and Yu. Kh. Shaulov. Thermodynamic investigations by the explosion method and calculations of combustion processes. Moscow University Press, 1955.
4. M. Kh. Karapet’yants. Chemical Thermodynamics. Goskhimizdat, Moscow–Leningrad, 1953.