Li-plating: The Baba Yaga for batteries…
By Apoorv Shaligram
It has been a while since my last post. I have been keeping busy with work and have not found enough time to complete my blog posts, though I now have a list of topics written down. Still, as I had promised some time back, today I will try to explore what Lithium plating is, what causes it to happen and how it affects Li-ion batteries.
Lithium plating is the boogeyman of batteries. Few truly know why it occurs, fewer still can predict when it occurs, but speak about battery failures and safety incidents and the first thing to get the blame from all quarters is lithium plating. So, it is something that everyone is afraid of in their hearts, but hardly anyone truly understands… Here, I will try to explore this topic based on my understanding of the working principles of batteries and some first principles based analysis.
To begin with, why does lithium plating occur?
Or rather, what are the conditions that lead to lithium plating? Basically, lithium plating occurs when there is more lithium being forced to the anode than the amount which can be accepted via the desired anode reaction. E.g., If there is 8g of active lithium content in the cathode and anode has capability of accepting only 7g of lithium in charging, and the cell is fully charged until all 8g of active lithium is removed from the cathode, 1g of lithium will end up plated on the anode surface.
Sound so simple. If the issue of lithium plating is just a balance between the amount of anode and cathode active materials, why haven’t people figured out a solution to this?
This is because the amount of active lithium in an electrode is not a static number which can be balanced easily. The amount of active lithium in an electrode material is a function of the rate at which lithium is being transferred, the direction in which lithium is being transferred, the thickness of the electrodes (diffusion distance for lithium) and the porosity of the electrodes, temperature etc.. as can be seen below in some old data of mine.
On top of this, the amount of active lithium is also dependent on the life of the electrode, or rather specifically, the change in the value of electrode resistance over time due to irreversible reactions of the electrolyte and the changes in the nature of the solid-electrolyte-interphase (SEI) layer on the anode surface. Add to this, the electrode capacity imbalance resulting from manufacturing process variances. Thus, it is dependent on design and manufacturing parameters (electrode loading & porosity, starting resistance of electrodes, uniformity & homogeneity of electrode fabrication), operational parameters (rate of charging, temperature of cell/electrode) and also the state-of-health (SOH) of the battery cell! Thus, cell designers can only design cells to prevent lithium plating under certain conditions. The moment the conditions change (and eventually they will), lithium plating WILL occur.
But if lithium plating is inevitable, does that mean all lithium ion batteries are bound to explode? That doesn’t sound right. I have had a cellphone that never exploded…
This is where lithium plating becomes Baba Yaga or the boogeyman for Lithium ion batteries. Since it is so difficult to understand and control the limits of lithium plating, over the years it has generated a certain level of fear in the minds of battery engineers. To answer the above question, we need to understand the impact of lithium plating on battery cells. To clarify, not all lithium plating leads to fires or explosions. In fact, only a very very small fraction of cases of lithium plating result in battery fires. In most cases, the lithium plating is only marginal and not enough to cause enough dendritic growth to rupture the separator and short-circuit the cell internally.
Think about it. Even if overcompensating with the anode capacity leads to a trade-off on cell capacity and energy density, good cell designers and engineers will always keep some factor of safety to account for small transgressions beyond the stated limits on the operating conditions. Thus, lithium plating will only occur when the transgressions are beyond even that factor of safety. Either that, or when the battery cells age to the point where there is capacity imbalance created due to irreversible side reactions. Note that, these irreversible reactions may not be uniformly spread about the cell or electrode volume and almost certainly be localized at locations where conditions favor their occurrence.
Lithium plating can also occur when we try to fully charge cells to their theoretical capacity and there is significant potential drop between different locations on the electrode. If the potential drop is significant enough, the part which gets charged to 100% capacity first will see lithium plating by the time the last bit of charging is completed. Remember that almost every cathode chemistry has some excess lithium available. This potential drop is equal to the current flow * resistance of electron path between the two points (ΔV = I*R). If this potential drop exceeds the lowest plateau voltage for anode potential, it will result in lithium plating. Note that this depends on electronic resistance of the electrode as well as charging current and hence is another reason why fast charging can cause lithium plating.
Coming back, the plating that occurs because of cell ageing is never sudden, and will always occur is very small quantities. Also, since this lithium that is plated on the surface is not sufficient to short-circuit the cell internally, it will not cause catastrophic damage. Instead, the lithium creates new active surface, which causes the electrolyte to decompose anew and form fresh SEI. The lithium itself probably partakes in the decomposition of the electrolyte liquid in some part, while the rest lithiates the anode active material it is in contact with over time (if the material has some active capacity to absorb that lithium). The formation of fresh SEI layers lead to sharp increase in resistance to ionic transport locally, and hence lead to more and more plating in the subsequent cycles, each leading to further electrolyte decomposition and a localized depletion of active lithium from the system. Hence, we see a sudden downturn in the capacity profile over cycle life. This knee-point in the profile is actually a pointer that lithium plating started at that point.
P.S.: One last point: We saw that the internal potential drops between different points in a cell lead to lithium plating. This potential drop between points plays another role in the occurrence of lithium plating; that of heating. Cell ageing, or the process of electrolyte decomposition; and like all chemical reactions, the rate of electrolyte decomposition (kinetics) increases with temperature. When cells heat due to passage of current, this heat generation is also not uniform across the cell, but is rather concentrated around the part where current enters the electrodes (the tabs that connect the current collectors to the terminals). This is also the zone where the electrode active material gets lithiated first before the rest of the electrode. Thus, the part of the electrode that is most likely to see lithium plating, is also the part that sees maximum heat generation in its vicinity — a double whammy!!
In the next part, we shall take a look at ways to prevent lithium plating in cells. I keep it as a separate part, since it needs a technical deep dive as opposed to the more philosophical approach I took here to understand Li-plating.
