We hear you gasp in horror; the thought of your precious LFP battery bank being no longer sends shivers down your spine! Alas, all good things eventually must come to an end. What we want to prevent is an end of the premature (and possibly spectacular) kind, and to do that we have to understand how lithium-ion batteries die.
Battery manufacturers consider a battery “dead” when its capacity falls to 80% of what it should be. So, for a 100Ah battery, its end comes when its capacity is down to 80Ah. There are two mechanisms at work towards the demise of your battery: Cycling and aging. Each time you discharge and recharge the battery it does a little bit of damage, and you loose a little bit of capacity. But even if you put your precious battery in a beautiful glass-enclosed shrine, never to be cycled, it will still come to an end. That last one is called calendar life.
It is difficult to find hard data on calendar life for LiFePO4 batteries, very little is out there. Some scientific studies were done on the effect of extremes (in temperature, and SOC) on calendar life, and those help set limits. What we gather is that if you do not abuse your battery bank, avoid extremes, and generally just use your batteries within reasonable bounds, there is an upper limit of around 20 years on calendar life.
Besides the cells inside the battery, there is also the BMS, which is made out of electronic parts. When the BMS fails, so will your battery. Lithium-ion batteries with a build-in BMS are still too new, and we will have to see, but ultimately the Battery Management System has to survive for as long as the lithium-ion cells do as well.
Processes inside the battery conspire over time to coat the boundary layer between electrodes and electrolyte with chemical compounds that prevent the lithium ions from entering and leaving the electrodes. Processes also bind lithium ions into new chemical compounds, so they are no longer available to move from electrode to electrode. Those processes will happen no matter what we do, but they are very much dependent on temperature! Keep your batteries under 30 Centigrade and they are very slow. Go over 45 Centigrade and things speed up considerably! Public enemy no. 1 for lithium-ion batteries, by far, is heat!
There is more to calendar life and how quickly a LiFePO4 battery will age: State-Of-Charge has something to do with it as well. While high temperatures are bad, these batteries really, really do not like to sit at 0% SOC and very high temperatures! Also bad, though not quite as bad as 0% SOC, is for them to sit at 100% SOC and high temperatures. Very low temperatures have less of an effect. As we discussed, you cannot (and the BMS will not let you) charge LFP batteries below freezing. As it turns out, discharging them below freezing, while possible, does have an accelerated effect on aging as well. Nowhere near as bad as letting your battery sit at a high temperature, but if you are going to subject your battery to freezing temperatures it is better to do so while it is neither charging nor discharging, and with some gas in the tank (though not a full tank). In a more general sense, it is better to put away these batteries at around 50% – 60% SOC if they need longer-term storage.
If you really want to know, what happens when a lithium-ion battery gets charged below freezing is that metallic lithium is deposited on the negative (carbon) electrode. Not in a nice way either, it grows in sharp, needle-like structures, that eventually puncture the membrane and short out the battery (leading to a spectacular Rapid Unscheduled Disassembly Event as NASA calls it, involving smoke, extreme heat, and quite possibly flames as well). Lucky for us, this is something the BMS prevents from happening.
We are moving on to cycle life. It has become common to get thousands of cycles, even at a full 100% charge-discharge cycle, out of lithium-ion batteries. There are some things you can do though to maximize cycle life.
We talked about how LiFePO4 batteries work: They move lithium ions between the electrodes. It is important to understand that these are actual, physical particles, that have a size to them. They are yanked out of one electrode and stuffed into the other, each time you charge-discharge the battery. This causes damage, in particular to the carbon of the negative electrode. Each time the battery gets charged the electrode swells a bit, and each discharge it slims down again. Over time that causes microscopic cracks. It is because of this that charging to a little below 100% will give you more cycles, as will discharging to a little above 0%. Also, think of those ions as exerting “pressure”, and extreme State-Of-Charge numbers exert more pressure, causing chemical reactions that are not to the benefit of the battery. That is why LFP batteries do not like to be put away at 100% SOC, or put into float-charging at (near) 100%.
How fast those lithium ions get yanked hither and yon has an effect on cycle life as well. In light of the above that should be no surprise. While LFP batteries will routinely do charging and discharging at 1C (i.e. 100 Amp for a 100Ah battery), you will see more cycles out of your battery if you limit this to more reasonable values. Lead-acid batteries have a limit of around 20% of Ah rating, and staying within this for lithium-ion will have benefits for a longer battery life as well.
The last factor worth mentioning is Voltage, though this is really what the BMS is designed to keep in check. Lithium-ion batteries have a narrow Voltage window, for both charging and discharging. Going outside that window very quickly results in permanent damage, and on the high end a possible RUD Event (NASA-talk, as mentioned before). For LiFePO4 that window is about 8.0V (2.0V per cell) to 16.8 Volt (4.2V per cell). The build-in BMS should take care to keep the battery well within those limits.