Why nitrogen dilution does not make compressed acetylene safe and what nitrogen is really for

Adding nitrogen to acetylene raises the pressure at which a decomposition can travel through it, but it cannot suppress the hazard, and it is a weak diluent compared with most alternatives. Worse, if a nitrogen-diluted mixture does detonate, the final pressure is higher than it would have been with no nitrogen at all. Nitrogen has an essential role in an acetylene plant, but that role is purging, not making compressed acetylene safe.

The most complete set of measurements is by Strizhevskii and Zaitseva (Tr. Vses. Nauchn.-Issled. Inst. Avtogen. Obrabotki Metallov, 1959). They measured the minimum decomposition pressure of acetylene mixed with seven different gases. The table gives the percentage of acetylene in the mixture at the decomposition limit for four of them; a mixture with more acetylene than this, at that total pressure, can decompose.

The 4.4 atm row is shown for the trend; its individual digits are still to be checked against the original table.

Read across each row and the pattern is clear. At 2.7 atm absolute a mixture that is almost one-third nitrogen is still at its decomposition limit, and the nitrogen needed keeps rising with pressure. Helium and hydrogen behaved much like nitrogen in the same work; butane, propane and natural gas allowed far higher acetylene percentages for the same total pressure.

Summarising these results for the 5–20 atm range, the compilation by S. A. Miller (1965) gives a working rule: nitrogen needed to stabilise an acetylene mixture is about twice the propane content, and natural gas about one and a half times the propane content, by volume. Even heavy dilution does not allow acetylene to be operated safely at high partial pressure, and detonation is believed to remain possible at appreciable dilution if the length-to-diameter ratio of the pipe is high enough.

The pressure reached in a detonation is a multiple of the total initial pressure. In the Chapman–Jouguet plane it is about 20–24 times the initial pressure, and two to two and a half times that again where the wave reflects off a closed end (dead legs and predetonation distance).

Suppose a process needs acetylene at a partial pressure of 2 atm absolute and nitrogen is added to bring the total to 4 atm absolute, in the hope of stabilising it. If the mixture still detonates, the multiplier applies to 4 atm, not 2. The same acetylene now produces a final pressure roughly twice as high. The compilation by S. A. Miller (1965) states this consequence directly: with nitrogen as diluent, a detonation from a given acetylene partial pressure ends at a higher final pressure than if no nitrogen were present.

So dilution that falls short is worse than no dilution. A designer who adds nitrogen and then raises the total pressure to recover the acetylene partial pressure has made the plant more dangerous, not less.

Two diluents behave quite differently, as compiled by S. A. Miller (1965).

• Ammonia gave acetylene partial-pressure limits of 4.3 atm absolute at 50 % acetylene, 7.2 atm at 45 % and 10.0 atm at 20 %.
• Carbon monoxide gave what the compilation calls striking stabilisation. Wet mixtures with about 20 % acetylene at 19 °C resisted platinum-wire initiation at acetylene partial pressures up to 42 atm, which is the vapour pressure of acetylene at that temperature.

Neither is a practical stabiliser for an industrial gas plant: ammonia is toxic and an acetylide promoter with copper, and carbon monoxide is toxic. They matter here because they show that stabilisation is chemical, and that nitrogen's inertness is exactly why it does so little.

Nitrogen left in the system after purging travels to the cylinders. It is not absorbed by the acetone the way acetylene is, so it occupies gas space and raises pressure. EIGA Doc 240/22 notes that settled cylinder pressure rises by about 0.4 bar for every 0.1 % by volume of air or nitrogen. The C.P.I. (Safety in the Production and Use of Acetylene, Paris 1968) records that nitrogen in the gas phase also reduces acetylene's solubility in acetone. Purges must therefore be completed and displaced with acetylene before filling, not left in the lines. The pressure does not show contents page explains how an abnormal settled pressure is read.

Nitrogen is also an asphyxiant. Purge vents should discharge at height, and anyone entering a vessel that has been purged needs the oxygen check described on the nitrogen asphyxiation page.