Handling damaged lithium batteries

Risks of quarantine in water

Lithium-ion batteries are one of the biggest challenges nowadays when recycling waste electrical and electronic equipment. The batteries can easily be damaged during removal, which greatly increases the risk of short circuiting and of stored energy being released. One way to prevent damaged batteries from reacting is to ‘quarantine’ them in water for an extended period of time, which causes them to discharge. A series of experiments examined whether this practice entails risks and what they might be.

Before the emergence of lithium-ion batteries, removing batteries from waste electrical and electronic equipment was a simple matter. As consumable parts that already had to be replaced regularly during normal use, the batteries were easy to access and simple to remove. The first generations of lithium-ion batteries in mobile phones and laptops could also be removed with the flick of a hand. Now, lithium-ion batteries are usually glued into laptops, tablets and mobile phones. They are no longer protected by their own housing, but rather installed as soft ‘pouch cells’. Lithium-ion batteries are now also installed in a variety of small appliances such as electric toothbrushes and e-cigarettes. These batteries must be removed before mechanical processing. Poorly protected cells in particular can be easily damaged during removal, which is why quarantine containers filled with water, sand or vermiculite are usually available in workplaces where lithium-ion batteries are removed. Water has the benefit of discharging the battery during quarantine, which can theoretically be further aided by adding salt. However, because the battery reacts in water, harmful gases that pollute the water can be released. There is also a risk that lithium metal batteries will accidentally come into contact with water and the elemental lithium they contain will react with the water and release hydrogen.


To assess these risks, Empa carried out a simple series of experiments.

In a first experiment (see Figure 1), lithium-ion pouch cells from mobile phones and laptops were left in test containers with tap water or salt water (NaCl solution) for 48 hours. The  batteries used were from waste electrical or electronic equipment and individual cells were deliberately damaged. Eight cells were used per test container (5 l). The test containers were sealed so that they were airtight and fitted with a fermentation lock so that the gas development could be recorded in qualitative terms. After 48 hours, samples were taken and sent to the laboratory for analysis.

In a second experiment, new, fully charged lithium manganese oxide batteries were placed in test containers with tap water or salt water to observe the spontaneous reaction. The batteries were also left in water for 48 hours so as to also observe slower reactions. The experiments were carried out with one intact and one severely crushed button cell and AA battery. 

Figure 1: From left to right; used, partly swelled pouch cells/additional damage to individual cells/groups of eight cells were placed in five litres of tap or salt water/fermentation tank to monitor gas development

Discharging lithium-ion batteries

Within 48 hours, the intact pouch cells discharged from approx. 4.0 V to 0.4 V – 1.75 V. Since the residual energy content in lithium-ion battery cells is in the single-digit percentage range at voltages below 2 V, such a deeply discharged cell no longer poses a risk. It was also found that the discharge is no greater in salt water. On the contrary, the cells that were placed in salt water were, on average, less discharged. This can be explained by the fact that the contacts corrode more in salt water, which delays the discharge.

Gas formation

The movements of the fermentation locks were recorded and analysed during the first hour. It was clearly visible that gases form when the pouch cells are discharged in water. However, the gas formation is moderate (1–2 movements per minute) and visibly decreases over time (see video1). A greater deceleration in activity can be observed in the salt water. This suggests relatively rapid corrosion of the contacts and a corresponding decrease in discharge levels.

Pollutants in water

The water samples were analysed for 25 elements by Bachema AG. 13 of the examined elements were detected in at least one sample. The results are listed in Table 1. Instances where the thresholds set out in the Swiss Waters Protection Ordinance for introduction into the sewer system were exceeded are marked with an *. The stronger discharge in tap water results in higher values for cobalt, copper and lithium. However, the values for nickel and manganese are higher in salt water. Since the exact same cells were not used in the experiments, it is reasonable to assume that more NMC cells (lithium-nickel-manganese-cobalt oxide) were present in the salt water experiment. The analysis shows that heavy metals are released into the water during quarantine and the thresholds for introduction into the sewer system are quickly exceeded.

1 The video contains a time-lapse recording (60x acceleration) of the first 60 minutes of the experiment.

Video Pouch cells gas

The Video could not be loaded.
If you wish to benefit from this service, you have to agree to the use of third party cookies in your cookie settings .

Metallic lithium

In the experiments with lithium manganese oxide batteries, no strong immediate reaction was observed from either the intact or the damaged batteries. The voltage was checked after two hours. It was found that a discharge occurred and it was stronger in salt water. The larger contacts could potentially explain the different behaviour compared to the pouch cells. In addition to the decline in voltage, a slight formation of bubbles could also be observed. It was not possible to check whether these were released hydrogen. After 48 hours, the batteries had severely degraded (see Figure 2). However, gas formation remained relatively low.


The experiments showed that lithium-ion batteries can be easily discharged to a safe level in water. The addition of salt is not only unnecessary, but rather also slows the process down. If the quarantine is carried out in water, moderate amounts of gases are produced, especially at the beginning. The quarantine containers should therefore be removed from the workplace as soon as they contain batteries, and good ventilation must be ensured. The quarantine water must also be disposed of as hazardous waste due to the accumulation of heavy metals.

There is no risk of a strong reaction of elemental lithium from the anode of lithium metal batteries due to contact with water. In addition to the tested lithium manganese oxide batteries, lithium thionyl chloride batteries, which are characterised by their high energy density and very low self-discharge level, are also used. These batteries are primarily used as backup batteries, in consumption meters, alarm and security systems and other long-term applications. Since thionyl chloride can react violently with water, caution is advised here.

Figure 2: Lithium manganese oxide batteries after 48 hours of contact with water