A6. Scenario E: walls and barriers
Overview
Partners Sandia, FZK and KI have contributed to the work performed in this scenario. It should be noted that work performed in Scenario A by partner UU is also relevant to the effect of walls and barriers.
Jet flames resulting from unintended releases of hydrogen can be extensive in length and pose significant radiation and impingement hazards, resulting in consequence distances that can be unacceptably large. One possible mitigation strategy to potentially reduce the exposure to jet flames is to incorporate barriers around hydrogen storage equipment. While reducing the jet extent, the walls may introduce other hazards if not properly configured. The goal of the work performed by Sandia is to provide guidance on configuration and placement of these walls to minimize overall hazards. A combined experimental and modelling programme is being carried out by Sandia National Laboratories. The purpose of the present study is to extend the available database on barrier walls as a hazard mitigation strategy, and to provide technical data for risk-informed hydrogen codes and standards decisions regarding barrier wall design and implementation. The experimental effort is complemented by a parallel numerical modelling effort that considers the interaction of jet flames and un-ignited jets with barriers and the ignition overpressure. Results from the experiments are used to validate the Navier-Stokes simulations of barrier wall impingement and overpressure and provide a measure of confidence in the extension of the simulation capabilities to more complex barrier wall designs and release conditions.
Partner FZK considered two of the tests covered in the Sandia Experimental programme and investigated three ignition times for each case.
Partner KI used data provided by the University of Alabama to validate their modelling approach for small, un-ignited impinging helium jets, which will be used to model hydrogen jets in the future.
Summary and conclusions of WP4 activities in scenario E
Sandia jet barrier interaction work
A combined experimental and simulation study has been performed by Sandia in order to assess the effectiveness of barriers to reduce the hazard from unintended releases of hydrogen. For the conditions investigated, 13.79 MPa (2000 psi) source pressure and 3.175mm (1/8in) diameter round leak, the barrier configurations studied were found to:
- reduce horizontal jet flame impingement hazard by deflecting the jet flame;
- reduce radiation hazard distances for horizontal jet flames;
- reduce horizontal un-ignited jet flammability hazard distances;
- For the 1-wall vertical barrier and 3-wall barrier configurations, the simulations of the peak overpressures from ignition were found to be approximately 40kPa on the release side of the barrier while approximately 5-3 kPa on downstream backside of the barrier. The literature indicates (AICHE, 1994 [13]) that an overpressure of approximately 35 kPa (5 psi) results in a 10% chance of personal injury from eardrum rupture.
Partner FZK considered ignition of both a free jet and a jet barrier interaction (tests 3 and 1 in SNL’s barrier experiments). In the calculations carried out for both cases, three absolute ignition times were chosen, namely, 140, 260 and 640 ms. These represented early, most probable, and a later value of ignition.
It was found that simple correlations for engineering purposes could not be used to assess the danger of the hazards involved in impinging jets. A predictive deterministic approach (CFD) must be used, as small changes in the boundary conditions could degenerate in major qualitative changes. A comparison with the experimental data (SNL) and with theoretical predictions (Chen & Rodi for axial concentration in un-ignited test), of the simulation results shows show that overall representation of the phenomena, i.e. overpressure and distribution of the hydrogen, is of acceptable accuracy for safety analysis purposes.
The jet barrier interaction simulations showed that the highest risk is not simply proportional to the amount of hydrogen inventory, but can be connected with the formation of the mixture, which has appropriate conditions for local flame acceleration, involving the part of the hydrogen in the more energetic regime of combustion. Another reason for such behaviour can be connected with the deficit of oxygen in the area of hydrogen accumulation, resulting in earlier transition into the regime of a diffusion flame. Thus changes in concentration such as the accumulation of hydrogen in different areas could have a big influence. The risk is not simply proportional to the amount of hydrogen, but to the local conditions providing local high flame acceleration.
KI's simulation of small un-ignited jet-wall interaction
Partner KI performed simulations of the interaction of a small-scale helium jet with a wall, based on experiments carried out at the University of Alabama [14] within the framework of Sandia National Laboratories experimental programme on hydrogen codes and standards. The goal of these experiments was to study and visualize the flow structure for different jet configurations in a laboratory environment.
Numerical simulations of SNL's small-scale jet-wall interaction experiments were performed by KI to validate the in-house CADYC code against qualitative non-intrusive experimental data. Analysis of the CADYC's simulation results shows that all macroscopic features of jet-wall interaction (including large scale vertical structures bouncing off the wall) are reproduced.