Handbook of Batteries, 4th edition (more details available in that reference):
Fe + 2NiOOH + 2H2O <---> 2Ni(OH)2 + Fe(OH)2 (first plateau)
3Fe(OH)2 + 2NiOOH <---> 2Ni(OH)2 + Fe3O4 + 2H2O (second plateau)
The overall reaction:
3Fe + 8NiOOH + 4H2O <---> 8Ni(OH)2 + Fe3O4
Discharging is left to right, charging is right to left.
Plateau refers to a region on the voltage vs state of charge of the iron electrode.
In practical batteries the second plateau is not normally used for various reasons even though avoiding it entails adding a lot of extra iron.
Theoretical specific energy using only the first plateau:
55.8+32+2 = atomic mass of discharged iron = 89.8
58.69+32+2 = atomic mass of discharged nickel electrode = 92.69
1 electrons per molecule nickel hydroxide, so roughly 10^5 coloumbs per mole of electrons, 27.7 Ah at 1.2 volts (33.3 Wh) or so per mole, 182.49 g, so 182.48 Wh / kg theoretical with 100 percent utilization, including the water as a reactant but not the electrolyte (of which only a vanishingly small amount is theoretically needed).
Cobalt hydroxide is used to improve conductivity of the nickel electrode either by adding it to the nickel oxyhydroxide at manufacture or into the powder after, by e.g. coating the particle electrolytically to give them a higher effective conductivity without interfering with their reactivity too much (it does reduce it a little though so again there will be an optimum amount). It can also help to increase the fraction of active material that is utilized. 1 to 5% of the active material mass may be used for this. We probably want to avoid bothering with this.
Metallic cobalt to the metallic mesh, function needs to be checked, probably to improve contact resistance and reduce oxidization of the mesh and maybe reduce corrosion rate by electrolyte.
Sulfur, tellurium or selenium compounds are used to activate the electrode, see other reactions section
Carbon black is sometimes added to pocket plate electrodes to improve bulk conductivity of the active material mass
Graphite particles can also be similarly added for pocket plate electrodes
Preliminary Figures for a 12V, 1kWh pile
NiFe cells produce a working potential of 1.2V, and charge at 1.4V. A 12V battery would then consist of 10 cells, and charge at 14V which is typical for most "12V" batteries. (Some charts suggest each cell may charge at 1.6V to 1.8V, indicating at 18V charge for a 12V battery)
To achieve 1kWh capacity, we will need 1000W/12V = ~85Ah. This means that each cell will need to provide 85Ah capacity. This corresponds to 1 coloumbs per second*3600 seconds per hour*85Ah*1/10^5 coloumbs per mole, which is approximately 3.17 moles of electrons. Iron is cheap and so is not a limiting factor. Nickel's electrochemistry in an NiFe battery indicates we get an electron per nickel atom, so 3.17 moles of nickel will be required with a hypothetical 100% utilization of active material. This is about 185g of Ni at a molar mass of 58.69g/mole. For a 'safety' margin, we will round up to 200g. At 200g Ni per cell, a total of 2kg of nickel will be needed for a 1kWh unit. The actual capacity of this cell based on the rounded values above would be 1095.8Wh.
Other important reactions in the cell
If you want a good battery rather than a poor one, the side reactions and other reactions in the cell are the most critical factor, with the only second being electrode geometry/design. They are a fundamental part of battery design. These exclude the reactions that contribute to the output electrical energy to the cell.
These are what sap energy away, leading to poor efficiency and high self discharge, and partly what limits battery life. Some of them are used to counteract the undesirable ones.
There are others that need to be learned from the documents listed in the sticking points section :
- Tellurium dioxide is a recommended additive but never seems to be used in commercial batteries, improves efficiency by reducing hydrogen evolution.
- Mucic acid is another additive that can be used to reduce the hydrogen evolution when integrated into the iron electrode, by a factor of 10 during storage (reducing self discharge greatly). Not clear if it helps improve efficiency as the patent does not mention that, but hydrogen evolution is the main reason the energy efficiency is so low so maybe worth further investigation.
- Self discharge due to oxidization by dissolved o2 (pretty small in magnitude)
- The iron being anaerobically attacked by water, about 1000 times larger in magnitude than the oxidation due to dissolved o2. This is the main reason the battery has such a high self discharge rate. Basically the conversion of iron plus water to iron hydroxide plus hydrogen gas.
- Self discharge at the nickel electrode, need to check the mechanisms/reactions again probably relatively low since it is in nimh and nicad despite the same reactants present.
- Corrosion of the metallic mesh by electrolyte, this may be one of the lifetime-limiting reactions that increases with temperature.
- Oxidization of sulfide to sulfate and it's ensuing accumulation on the surface of the iron electrode, increasing internal resistance. Sulfide is also often used to keep the iron electrode active so when it is used up the battery dies.
- Sulfides like iron sulfide, or in some cases elemental sulfur (S8), or compounds of selenium or tellurium is used to "depassivate" the electrode. According to patents and reference sources this consists of reducing non-conductive, non-active oxides like FeO on the surface of the electrode back to active forms. They also indicate that without these additives charge efficiency drops greatly and hydrogen evolution goes way up, unless high purity iron like carbonyl iron is used and even then they don't work for long unless sulfur is added. No information has yet been found as of june 3rd 2011 on the performance of selenium and tellurium as mentioned above, if it might be superior to sulfur. Probably not.
- Accumulation of sulfur on the surface of the iron electrode esp. at lower temperatures and high discharge rates. If the sulfur or sulfides (or maybe other additives) are incorporated into the structure of the iron mass they are released faster when higher discharge rates occur and this can end up being too fast for them to diffuse away, resulting in them accumulating as a solid on the surface of the electrode, and since they are insulating and block access to the electrolyte this is a problem. At lower temps the solubility is lower so they are more prone to precipitate out of solution. This is not a permanent effect, the compounds redissolve if given some time. The exact temperatures and discharge rates depend on the quantities of additive involved and the surface are of the electrode etc. but references indicate this a problem more in the 0 deg C range than room temperature and at relatively high discharge rates, like two or three times the rated discharge rate (how high? In milliamps per sq meter is probably a reasonable measure).
- Deposition of iron in the nickel compound crystal structure <--needs more research could be important life limiting reaction.
- Oxidization of the surface of metallic flitts to low conductivity nickel oxide.
Edison had a problem with nonconductive layer on the flitts forming and didn't know what it was, maybe explained in later patents.
- Carbonate and the other one in the battery handbook undesirable ions, note that carbon dioxide from the air will react from the electrolyte to form carbonate salts <-- need to add those other ones. Some docs on the desired list may help inform here.
- Probably all kinds of minor undesirable contaminants, Edison mentions manganese, and high purity material like carbonyl iron is often used to avoid them, and similarly for nickel electrode. However this may be expensive or harder to produce.
- Lithium hydroxide apparently improves the thermodynamic reversibility of reactions (improves overall energy efficiency) and slows down the iron poisoning of the nickel electrode maybe this indicates longer life and higher charge/discharge efficiency, we need a chemist who can identify other ways of improving esp the charge/discharge efficiency by identifying other acceptable additives that might work. Also maybe analyze modern batteries if they have high performance.
- Evolution of gasses at the nickel electrodes during charging and also charge-stand, this could play into the low efficiency but may only be commensurate with other prerequisite reactions at the iron electrode and unpreventable in themselves.
- The air interface one that causes oxidization of the nickel metal and/or active material(?) of the electrode when there is an interface with the electrolyte and air, nickel electrode needs to be submerged. Another cause of self discharge if not designed right, plus can corrode the electrode or the nickel plating of it right through possibly leading to failure.
- A couple of abstracts of inaccessible documents seem to clearly indicate that accumulation of potassium/sodium carbonate and similar compounds (presumably produced by atmospheric co2 and impurities in makeup electrolyte) deposit on the electrodes and this is one of the first problems that limits cell lifetime for unsealed, conventional nife batteries. They may be rejuvenated if this is the case.
Need to add references. See additives section too.
Nickel itself that is the concern rather than any particular compound. Water soluble compounds are unsurprisingly much more of a concern than metal that is bound in solid objects like nickel plating or some types of stainless steel, since it is more bioavailable and can be spread and spilled more easily. At low levels of exposure talking in terms of nickel content is done. But of course the exact solubility and other factors has a substantial effect on the exact toxicity at higher levels of different water soluble compounds too, note the ld50 is 10x lower for nickel chloride as for nickel oxyhydroxide according to the documents below, so the expedient of speaking in terms of nickel content doesn't always work that well.
Note that nickel hydroxide, oxide and oxyhydroxide are all considered "insoluble" although obviously the chemistry here depends on them being at least a bit soluble. But that increases the safety margin a bit. It may be substantially higher under these alkali conditions though.
Still all metals are toxic to some degree including iron and we need some numbers to make an informed decision. Preferably a solid evaluation of dose response relationship, and if the batteries are to be widely adopted now is the time to factor in chronic exposure too. It needs to be considered not in a vacuum but relative to other options available.
higher quality :
- http://www.hpa.org.uk/web/HPAwebFile/HPAweb_C/1236757324101 l
unknown due to lack of access:
- material safety datasheets with information regarding nickel oxyhydroxide, clearly it is sometimes used as a material in NiMH, in fact most of the hits are datasheets for such batteries rather than the material itself:
- http://www.batteriesplus.com/msds/Duracell_Nickel_Oxyhydroxide_%20Batteries_NorthAmericaMSDS.pdf (as if companies would give accurate information on their own products)
Note that the permissible exposure limits (PEL) for the material is about a fifth that of graphite, which we know is not too toxic though it might accumulate in the lungs I guess. Still there are a lot of variables there as most particles inhaled are not retained, depends on size etc etc.
Obviously the LD50 is quite low at 1000 mg/kg range. But that means little in terms of what happens at lower levels. For Chloride it seems to be in the 100 mg/kg range.
http://www.osha.gov/SLTC/healthguidelines/nickelsolublecompounds/recognition.html They fail to say dosages in critical places, this information is actually low quality.
nickel chloride: http://www.sciencelab.com/xMSDS-Nickel_chloride-9926213
From Dietary information perspective: http://iom.edu/Activities/Nutrition/SummaryDRIs/~/media/Files/Activity%20Files/Nutrition/DRIs/ULs%20for%20Vitamins%20and%20Elements.pdf
"A Tolerable Upper Intake Level (UL) is the highest level of daily nutrient intake that is likely to pose no risk of adverse health effects to almost all individuals in the general population. Unless otherwise specified, the UL represents total intake from food, water, and supplements."
1 milligram per day for nickel, 40 for iron. Well that's to totally zero adverse effect level and there's probably plenty of room for more there. Nickel is probably available as a dietary supplement, I wonder what happens when people take too much.
http://www.eoearth.org/article/Public_Health_Statement_for_Nickel case report little more than half way down of people drinking 250 mg/liter (250 ppm) nickel welding, relevant to pocket production: http://www.crios.be/Welding/toxicology.htm Maybe it woudl be better to make the pockets without welding. Folding perforated sheets cleverly, sewing with nickel wire etc. plenty of other options that should be fine .
As usual it is difficult to obtain quality information on a health issue and it is very time consuming to wade through the crap and get some answers. Case reports, vague qualitative statements, contradictions, and paywalls, are the norm. Ideally a table of dose-response relationships for a range of different people over a wide range of doses would be obtained but whatever.
In summary it looks like it is certainly less problematic than lead or mercury, gloves, ladling or scooping rather than pouring dry materials, and working outside would more than suffice for me personally. Spills of solutions and especially powders should be prevented.
All batteries (at least the ones in common use) are toxic to a greater or lesser extent. The NiFe Battery isn't an exception, but it is far less toxic than Cadmium-based batteries.
- Iron is non-toxic and commonly available.
- Nickel is slightly toxic.
- The electrolyte is caustic, but a so-called starved electrolyte cell can be used in a sealed battery without degradation in performance if potential spills are a perceived problem. In this design only the minimum amount of electrolyte is used, and it is absorbed in a porous mat so there is no liquid that can escape. This is what an Absorbed Glass Mat (AGM) lead acid battery is, and it can be used here too. The cost of the mat is not insignificant though. For photovoltaic there may be no need, flooded cells could be fine.
- Increasing the viscosity of the electrolyte with glycerin also could be used, similar in theory to a gel-cell battery. There is a cost of increased internal resistance for a higher viscosity electrolyte. It may also pose a problem with the amount of gas produced, which needs to escape from the region of the electrodes. The glycerin might reduce cycle life as it could degrade on one of the electrodes in the strong red/ox environment or contain contaminants.
- Glycerin is a by-product of creating bio-diesel, thus using a waste of a different process. It may have other uses elsewhere in producing the active materials etc. too.