Self ignition of hydrogen/air mixtures

In this work shock tube experiments have been performed for the hydrogen/air-system in the range 3 bar <p <50 bar and 700 K <T <1200 K using the reflected shock technique. A mixture composition of 15% hydrogen in air has been chosen. Different shock tubes have been used to cover the whole pressure range and to apply different measuring techniques and observation methods, such as pressure and OH-band emission measurement as well as shadow photography (Cranz-Schardin camera). The experimentally obtained ignition delay times tau(i) have been compared to chemical kinetic calculations using a well established hydrogen oxidation mechanism. At high temperatures (e.g. T >1050 K for p <4 bar) the experiments and zero dimensional, homogeneous reactor calculations yield more or less exactly the same ignition delay times. At lower temperatures-for all pressures under investigation-the so called ''hot spot'' ignition has been observed. The origin of this ''hot spot'' ignition is somewhere close to but not directly at the end wall of the shock tube. Ignition delay times measured for these ''hot spots'' are up to 1000 times shorter than the calculated tau(i). The reasons for this type of ignition may be inhomogeneities in temperature and mixture composition and catalytically effective particles. This ''hot spot'' ignition leads to independent inflamed regions inducing a pressure rise in the unburnt mixture. The mixture prepared in such a way may ignite in the form of a ''secondary explosion'' yielding a sudden pressure rise. In contrast to the inhomogeneous '' hot spot'' ignition, this secondary explosion can be regarded as a homogeneous self ignition and should be directly comparable to kinetic calculations. Hence, this experimentally observed instant of the secondary explosion has been compared to a calculated ignition delay time using the pressure history measured in the experiment. There seems to be agreement at relatively low pressures (p <15 bar) but not at high pressures.