A5. Catastrophic release scenarios

In more than 200 experiments performed in the frames of Pro-Science (PS) experimental programme, the hazard potential of a severe hydrogen leakage inside a fuel cell cabinet was investigated, using a generic fuel cell enclosure model with an internal volume of approximately 560 litres. In all experiments, 120 litres of this total volume were blocked by a solid cube representing large internals of the fuel cell. Based on the description of a commercially available fuel cell unit, the maximum hydrogen release rate possible in the case of a rupture of the hydrogen feed line inside the enclosure was evaluated to 15 g/s, so in the experiments hydrogen release rates of 1.5 to 15 g/s were used. A security mechanism, based on the pressure drop inside the hydrogen lines of the fuel cell after the rupture of the feed line, was assumed to shut down the hydrogen supply after 1 second. With this assumption and the release rates evaluated, the following three cases with different venting characteristics of the model enclosure were investigated:

• In case 1, two vent openings were arranged diagonally on two opposite sides of the enclosure, passive (case 1a) and active venting (case 1b, two fans mounted at the vent openings) through the openings was investigated.
• In case 2, enlarged vent openings with doubled size at same positions were used (passive venting), and
• In case 3, an additional chimney was fixed on the top of the enclosure with the smaller vent openings in the sidewalls (passive venting).
  

The experimental work started with distribution experiments, where hydrogen concentrations were determined at certain positions inside and outside the model enclosure during fixed time periods after the hydrogen release. Additionally, the distribution behaviour of the injected hydrogen was monitored using the BOS technique, an optical method for the visualisation of density gradients. In the distribution experiments two internal geometries of the model were investigated:

• a low obstructed internal geometry where a grid cube consisting of intersecting obstacles (blockage ratio 50%) was mounted at the top of the model enclosure, reducing the gas volume inside the enclosure to approximately 380 l (in this configuration about 32% of the internal volume is occupied by obstacles; such a degree of blocking corresponds to the obstruction of the fuel cell which was used as a prototype for the experiments); and
• a highly obstructed internal geometry in which the entire free space inside the enclosure was occupied by grids (blockage ratio 50%), reducing the gas volume inside the enclosure to approximately 240 l.
  

Due to the buoyancy of the released hydrogen, forced ventilation (so called chimney effect) was observed in all experiments with low obstructed internal geometry. Here the lower vent opening acted as an inlet for the fresh air while the upper one was the exit for the hydrogen/air mixture. In the experiments on case 3, the additional chimney took the role of the upper vent opening. In the experiments with the highly obstructed internal geometry the mentioned chimney effect was not observed, and compared to the experiments with low obstruction outside the enclosure, only small hydrogen concentrations were measured. At the same time inside it an inhomogeneous mixture distribution with very high hydrogen concentrations close to the walls and beneath the top was observed. Due to the high hydrogen concentrations found inside the model enclosure in the distribution experiments with the highly obstructed internal geometry, it was decided not to perform combustion experiments with such internal geometry.

In the combustion experiments two scenarios differed by the location of the ignition point were investigated. In scenario C the ignition position was inside the enclosure, while in scenario D it was located outside the enclosure. Three ignition positions and two ignition times were used in the experiments:

• Ignition position CMS inside the enclosure close to the upper left front edge,
• Ignition position OUT outside the enclosure above the centre of the upper vent opening,
• Ignition position EXT outside the enclosure above the centre of the chimney,
• Delayed ignition where the ignition source was turned on 4 s after the beginning of an experiment for 300 ms; this ignition time was chosen since in the distribution experiments the highest H2-concentrations most often were observed between 2 and 6 s, and
• Durable ignition where the ignition source was turned on simultaneously with the beginning of an experiment for a duration of 5 s to take into account possible ignition sources that are permanently present (e.g. hot surfaces).
  

From the results of the experiments the following conclusions can be drawn:

Obstructions: With the highly obstructed internal geometry the chimney effect inside the enclosure was inhibited, leading to high hydrogen concentrations inside the model enclosure. Another reason for the high H2concentrations is the partial blocking of the vent openings by the grid. Compacting the obstacles would minimise the hazard of possible flame acceleration in distributed obstructed areas and would also allow to avoid an obstruction of the vent openings.

Venting: The number and arrangement of the vent openings used in the experiments enabled an appearance of a chimney effect inside the enclosure. The three main venting characteristics investigated showed little difference between the maximum H2 concentrations measured during the experiments; however differences in the hydrogen transport to the outside of the enclosure and in the homogeneity of the hydrogen distribution were recognised. In cases 1a and 2, an inhomogeneous hydrogen distribution inside the enclosure and rather slow transport to the outside was observed, while in the cases 1b and 3 a more homogeneous hydrogen distribution and faster transport to the outside was reached. In the combustion experiments, cases 1a and 2 generated similar loads, but different combustion behaviours were found in some of the experiments on cases 1b and 3.

• Case 1b:  The active venting used in case 1b is responsible for a comparably slow combustion with a released hydrogen amount of 1.5 g for durable internal ignition. In this experiment similar transport ratios of hydrogen release and fan- induced flow were reached. With the higher hydrogen amount of 3 g, the flame velocities determined inside the enclosure model were similar to the ones of the corresponding experiments on the cases 1a and 2. The opposite effect of the active venting was found in experiments with durable outside ignitions, where stronger combustions were observed for released hydrogen amounts of 3 and 4 gH2 compared to the other cases. For delayed ignitions, similar combustion behaviours as in the other cases were observed.
• Case 3: An effect of the additional chimney used in the experiments on case 3 was observed with delayed external ignition, where even released hydrogen amounts of 4 g did not cause any detectable pressure wave. For the other ignition settings, similar burning behaviours as in the other cases investigated were observed.
  

Ignition position and time: In combustion experiments with the same hydrogen amount, a durable ignition generated higher loads to the ambience than a delayed one, and also an internal ignition produced stronger combustions than an external one. The highest loads were detected with a durable internal ignition, while the slowest combustions were observed for delayed outside ignitions.

Hydrogen amount: The highest combustion loads for a given amount of released hydrogen were observed in the experiments with a durable internal ignition. With these ignition settings combustion was detected by all the sensors, even in experiments with a hydrogen amount of 1.5 g. The ignition of 3 g H2 resulted in pressure waves with a maximum amplitude of 40 mbar inside the model enclosure; such pressure loads can cause glass breakage of large windows. In the experiment with a hydrogen release of 4 g and durable internal ignition pressure waves with a maximum amplitude of approximately 100 mbar were observed inside the enclosure, which could even lead to injuries to human beings. Although the enclosure model survived this test, other enclosure types may well be damaged or missiles may be generated during the explosion.

The findings of the experiments have demonstrated that for the diminishing of possible hazards, it is necessary to reduce the hydrogen amount that can be released from a ruptured pipe inside the fuel cell enclosure by constructional means to below 1.5 g. In most of the experiments with this hydrogen amount no combustion was observed, and even with the ignition settings producing the highest combustion loads, no dangerous pressure waves were detected inside the enclosure. Furthermore, the distribution experiments with the enclosure model showed that, with this release rate and the low obstructed internal geometry, hydrogen concentrations only slightly above the flammability level for hydrogen/air-mixtures were found.