A typical Ni-MH cell will loose its cycling ability due to excess of charge, and/or excess of temperature.
The dumb charge at 0.1 C for 16 hours provides, on average, 40 -50 % more charge at every cycle. This kills the cell faster, even if there is no raise in temperature.
Please note that the situation with Ni-CD and dumb charge is completely different. Ni-Cd accepts dumb charge and overtemperature much better than Ni-MH.
Standard Ni-MH cells are built with a negative electrode which has 20% more capacity than the positive electrode. At the end of charge,the positive electrode stop nickel oxidation, while the negative has still capacity in storing hydrogen.
The drop in voltage rise (-dV/dt) is due to the end of charge at the positive electrode. Actually, when the voltage stops rising, the cell is already slightly overcharged.
An excess of charge creates nickel dendrides that damage the fiberglass separator. In the new "hybrid" technology, this behavior is reduced considerably.
At C+20% overcharge, the negative electrode overcharges too, and hydrogen gas is being released within the cell. It migrates to the positive electrode, where is reacts reforming water. The combination of hydrogen and oxygen produces a lot of heat, it is like a "flameless combustion". This heat dries the electrolyte and pulverize the negative electrode, which is made of an alloy of lanthanium and nickel.
Ni-CD has no known limitations on overcharge. If the seal of the cell is perfectly tight, they can take a moderate overcharge almost forever.
The key in long life cycle is to provide the cell with the right amount of charge. More the overcharge, less the cycles. If the cell is overheated too, it last even less cycles.
I tried to use a Li-Ion type charger with Ni-MH. Those chargers typically measures the power provided from each cell during discharge, and recharges them with the same amount of energy. They also keeps a table in the flash memory for each cycle. In this setup, a set of Ni-MH (charged at 1 C) lasts over a 1000 cycles.
Another consideration. The negative electrode of the cell is more delicate and degrades faster than the positive one. At a certain point of the cell life, its capacity will be less than the positive one.
The overcharged negative electrode will start releasing hydrogen that cannot be oxydized, so the cell will vent hydrogen. And if the cell is being fast charged, the charger will miss the termination signal since the cell voltage will keep raising nonetheless.
This will lead to a catastrophic failure of the cell. The end-of-life for a Ni-MH is when the capacity of negative electrode becomes less than the positive, or when the electrolite dries up... whichever comes first.
I want to mention another thing. In critical or dangerous applications, Ni-CD are still preferred to Ni-MH, even if they may have only a third of the initial capacity. In case of failure or improper design of the charger, and in case of short circuit for that matter, they vent oxygen, not hydrogen. I have some underwater lights that, based on this consideration, I still feed with Ni-CD.
1. Any overcharge will decrease the lifecycle of Ni-MH. You're right that faster is worse. The 0.1 C /16 hrs was OK for Ni-CD but produced increased damages on Ni-MH due to the mulecular interstitial structure on the NH electrode. Basically overcharge will pulverize it at any rate.
2. The voltage plateau in Ni-MH is a combined effect between the ongoing reactions at positive and negative electrode. There is no free oxygen in NiMH cells, only free hydrogen. When I say the cell is fully charged right before the plateau is reached, I meant the positive electrode (nickel metal) has already reached its full oxidated potential. Any further charge will create super-oxidized states of nickel, that disappears after few hours the charge has been terminated. Creation of super-oxydes produces a jump forward of the termination voltage.
Manufacturer are not honest to include this extra capacity in the nominal cell capacity. Since the nickel superoxydes are dangerous to the separator, they had to limit their formation in the new "hybrid" LSD cells, which are more honestly rated.
3. I have seen separators made with everything. Glass is still the king material for superior cells, but its thickness limit the capacity of the cell. Polypropilene is the second best. I believe separators have greatly benefited from advances in the sport textile industry.
Cheap china NiMH cells uses PVC. It melts with slight overtemperature.
There are two alloys that can store hydrogen. The first is the AB5. While the composition of the alloy is known generically, the EXACT mix and the its processing are somewhat kept secret.
For AB5, manufacturers declare:
A= lanthanium, cerium, neodymium, praseodymium.
B= nickel, cobalt, manganese, aluminium.
Please note that the definition of AB5 alloy, for each A and B part, is "and/or".
I can add to this that A (the rare earth) is the active portion of the alloy, while B is a metallurgic co-formulant.
For AB2, another alloy of higher capacity but more prone to disgregate under electrochemical stress, the manufacturers declare:
A = titanium and/or vanadium
B = zirconium or nickel, chromium, cobalt, iron, and/or manganese.
Consider that Vanadium is one of the preferred catalyst used in the oil and plastic industry, and in the form of VnO5 was the first catalyst used in fuel cells. VnO5 has many of the catalytic properties of the platinium, at a fraction of the cost.
The metallurgic co-formulant of AB alloys, coupled with a specific cooling behavior after it is molten, is only required to form the right spatial orientation of alloy mulecules, to permit part A to "operate". More sophisticate is che cooling cycle, less co-formulant are required.
I'm not deep inside metallurgy. All I can tell you is the specialist tries to replicate the behaviour of palladium, at a fraction of the cost. Palladium can store up to 900 times of its volume in hydrogen. It is a physical phenomena called "adsorbtion".
From more first hand information, I can tell that AB5 is made, for the majority of it, of nickel and lanthanium. Cobalt, if present, is one of the non-active co-formulant.
e per quest'estate da leggere sotto l'ombrellone vi consiglio "Battery reference book terza edizione" :D è un po datato come libro ma contiene delle informazioni interessanti, poi sono appena 770 pagine :D