Ignition of hydrogen-containing atmospheres by mechanically generated sparks

Status:

completed 03/2023

Aims:

Hydrogen is very sensitive to ignition compared to many other fuel gases. However, observations by BAM showed that the ignition effectiveness of mechanically generated sparks is different for hydrogen and acetylene, the reference gases of explosion group IIC. Therefore, knowledge of the ignition effectiveness for hydrogen as a function of the material pairing is essential for the assessment of the necessary measures. Furthermore, it is important for the evaluation of protective measures in practice how the ignition effectiveness of mechanical sparks changes if hydrogen is not present as a pure gas but as a component in a gas mixture. Facilitations resulting from the investigations would be of immediate considerable practical benefit for many applications. The values are also to be included in TRGS 723 (Technical rules for hazardous substances) when it is revised.

Activities/Methods:

To investigate the reproducibility of the relative frequencies of ignition by mechanical impact with the apparatus used, a total of four preliminary test series were carried out.

In atmospheres with a hydrogen content of 10%, 8 test series were carried out in which impacts with impact pins made of high-alloy steel of the grade 1.4571 and the non-ferrous metal alloys copper-beryllium (CuBe) and aluminium-bronze (AlBr) with a kinetic impact energy of 61 J were carried out in each case against impact plates made of high-alloy chromium steel of the grade 1.4571, of low-alloy steel of the grade 1.0579 and of screed concrete. Furthermore, it was investigated how a reduction of the impact energy to 31 J affects the relative frequencies of ignition.

In order to assess at which proportions an admixture of hydrogen to methane has a recognisable effect on the ignitability by mechanical impact, further test series were carried out with methane-hydrogen fuel gas mixtures, in which the hydrogen proportion in the fuel gas mixture was gradually increased from 25 mol-% to 50 mol-% to 75 mol-%. To confirm the results, the impact energy was increased at 25 mol-% and the total fuel gas content was varied at 75 mol-%.

Results:

The tests confirm that the use of non-ferrous metals that are difficult to oxidise instead of stainless steels can generally reduce the ignition probabilities. Thus the use of low-sparking tools can be a suitable explosion protection measure if the kinetic impact energy can be limited at the same time. For impacts against screed concrete, the kinetic impact energy must generally be limited more than for impacts against metallic surfaces, even when low-sparking tools are used, in order to prevent effective ignition. A corresponding addendum to the regulations is recommended.

Furthermore, the ignition effectiveness of mechanical impacts in explosive atmospheres with hydrogen-methane mixtures was investigated. For hydrogen contents of up to 25 mol-%, no difference was found in the tests carried out in this work compared with methane without hydrogen admixture. With a limitation of the kinetic impact energy to 61 J, ignitions were only observed at all with fuel gas mixtures of 75 mol-% hydrogen and 25 mol-% methane. Overall, the tests also confirm that in principle an assessment of the ignition effectiveness of mechanical impacts in explosive atmospheres with various fuel gas mixtures according to the explosion group classification is possible.

Last Update: