A total of 6 large-scale experiments were carried out with a fire in an electric scooter, 3 of the experiments were carried out in a 55 m2 large room corresponding to a classroom, and 3 of the experiments were carried out in a 15 m long corridor (38 m2). The ceiling height in the building was around 3 m.

The concentrations of the gases CO2, CO, O2, HCl, HF, HCN, SO2, CH2O, NO and NO2 were measured in the experiments. The measurements are used to establish an experimental basis for evaluating whether and when critical gas values (according to ISO 13571:2012 "Lifethreatening components of fire") are achieved and thus lead to reduced ability to escape. The temperature change caused by the fire was measured at different heights in the room. In addition, video documentation is used to assess how the spread of smoke affects escape in a situation where there is a fire in an electric scooter in an escape route.

The study has shown that a thermal runaway in a lithium-ion battery leads to a rapid fire development where the battery essentially bursts into flames, with jet fires and potential ejections of burning battery cells far away from where the fire started. The duration of this fire behavior with jet fires and flying debris was between 3 and 7 minutes. In the fire experiments, the emitted energy was not high enough to raise the room temperature to a critical level. Near the fire, however, there is a hazard of fire spread to other combustible materials in the room due to the behavior of the fire and high temperature of the jet flame. Ejection of burning battery cells poses a hazard of fire spread even to areas far away from the start location.

Fires in an electric scooter battery or similar lithium-ion batteries can cause a rapid spread of smoke to the entire room. In the conducted experiments, the fire room was no longer smoke-free at the height of 1.9 m already after 1-2 minutes. Due to this rapid spread of smoke, visibility in the room will be affected after a short time and make escape more difficult. In the corridor, the smoke spread was relatively evenly distributed in height, while the smoke in the large room ("classroom") spread in a layer under the roof. Both forms of dispersion are thus possible, depending on the room and ventilation configuration.

The gas measurements in the fire experiments detected both asphyxiant and irritant gases. Due to the battery size, which affects how much gas is formed, in relation to room size and ventilation conditions, the calculated FEC, i.e., the critical concentration of irritant gases, was below the selected limit value of 0.1 in all experiments. Although the FEC value was below 0.1 in all the experiments, people in the fire room would have begun to feel an effect from some of the toxic gases. However, this effect would not have been disabling. The FED, that is, the critical dose for asphyxiant gases, was only obtained after 23 to 30 minutes. It is important to remember that the concentration of toxic gases in a room due to a fire in a lithium-ion battery depends on the ratio of battery size, room size, and ventilation conditions. This means the limit values could have been exceeded for a larger battery or in a smaller room.

The most important recommendation from this study is: Avoid storing and charging electric scooters and similar in living areas and escape routes.

The study is commissioned by the Norwegian Directorate for Civil Protection (DSB) and the Norwegian Building Authority (DiBK).

Read the full report in Norwegian with English summary: (pdf)