Nickel-Iron Battery/Prototype

Suggested contributions

We could use the documents below.

Then these can be legally shared by e.g. zippyshare.com with other developers who ask for a copy under the fair use doctrine. See the library section for what we already have.

We basically want all papers that mention nickel iron specifically and most of the others that relate to battery electrodes made from nickel oxyhydroxide(very frequently referred to as only "hydroxide" in the context of NiMH especially), and metallic iron and/or iron oxides. The electrode ones may not mention "nickel iron" per se because e.g. a good iron electrode can also be used in several other battery chemistries.

Most important:

6V, 60Ah nickel-iron battery. (http://md1.csa.com/partners/viewrecord.php?requester=gs&collection=TRD&recid=2274533EA) Periasamy, P | Ramesh Babu, B | Jegannathan, S | Muralidharan, S | Chakkravarthy, C | Vasu, K I Bulletin of Electrochemistry. Vol. 6, no. 2, pp. 263-265. 1990

Nickel--Iron Battery Development in CECRI. (Abstract Only) ,Periasamy, P | Babu, B R | Jegannathan, S Trans. SAEST. Vol. 24, no. 3, pp. 6.19. July-Sept. 1989

Alkaline Ni--Fe battery development is undertaken in CECRI under the sponsorship of the Defence Ministry (India). Dry powder sintering technique using the concerned metal powder is followed to fabricate porous Fe negatives and Ni positives. With carbonyl-nickel powder as the starting material, porous Ni positive plates (17.4 x 13.8 x 0.2 cm) were prepared by sintering in hydrogen atmosphere to get the Ni matrix of porosity 8-85%, followed by impregnation of nickel hydroxide into the pores of the Ni matrix. Porous Fe electrode was fabricated from electrolytic Fe powder by sintering in H atmosphere and activation of the sintered porous Fe plate (17.4 x 13.8 x 0.15 cm). A 6 V, 60 A/h Ni/Fe battery consisting of five individual cells in series is assembled with five positives and six negatives in each cell. The electrolyte is 30% KOH solution containing 50 g/l of LiOH. The separator is woven nylon fabric. The 6 V, 60 A/h Ni/Fe battery is charged at the 2 h (C/2) rate and discharged at different rates ranging from 1 h (C) to 5 h (C/5) rate to realise the A/h capacity in each case. In addition to carrying out life cycle test on the battery, self-discharge and effect of temperature on the output have been carried out. Special features of the Ni/Fe battery developed in CECRI are: high charging efficiency (80%), approx 50% capacity output at 0 deg C, high rate of charge and discharge and deep discharge up to 90-95%, without affecting the battery.--AA

Less important but still highly desirable:

• Westinghouse nickel-iron battery performance, 1981 Yeki
• Creator/Author Rosey, R.
• Publication Date 1981 Jan 01
• OSTI Identifier OSTI ID: 6394222; Legacy ID: DE83008901
• Report Number(s) CONF-811010-9
• DOE Contract Number W-31-109-ENG-38
• Other Number(s) Other: ON: DE83008901
• Resource Type Conference/Event
• Specific Type Technical Report
• Resource Relation 6. electric vehicle council symposium, Baltimore, MD, USA, 21 Oct 1981; Other Information: Portions are illegible in microfiche products
• Research Org Westinghouse Electric Corp., Pittsburgh, PA (USA). Advanced Energy Systems Div.
• Subject 33 ADVANCED PROPULSION SYSTEMS; 25 ENERGY STORAGE; ELECTRIC-POWERED VEHICLES; IRON-NICKEL BATTERIES; LIFE-CYCLE COST; PERFORMANCE; CAPACITY; DESIGN; ELECTRODES; ENERGY DENSITY; COST; ELECTRIC BATTERIES; ELECTROCHEMICAL CELLS; METAL-METAL OXIDE BATTERIES; VEHICLES
• Description/Abstract An advanced nickel-iron battery system is currently being developed by Westinghouse for energy storage applications which include on and off road electric vehicles, emergency standby power systems and deep water submersibles. The thrust of a present development program, sponsored by the Department of Energy under the Electric/Hybrid Vehicle Act, is to demonstrate battery system performance characteristics in an electric vehicle to achieve a 100 mile range on the SAE J227a D cycle. The 1981 nickel-iron battery performance objectives established by Westinghouse required to meet this range are: 54 wh/kg gravimetric energy density; 120 wh/l volumetric energy density; and 150 w/kg peak power density. Additional requirements are > 60% charge efficiency, selling price of $80/kWh, and 1000 cycles life to provide a system with acceptable operating life cycle cost. Demonstrated results for electrodes, cells, and batteries will be presented. These include charge/discharge voltage profiles, thermal effects on performance, power characteristics, cyclic stability, and vehicular mission profiles. The design and operating features of the battery system will also be reviewed.

There are some even less important ones on the research page.

See the Library section for a list of documents that we already have (these lists of what we need and what we have share no documents in common at this point on june 30 2011.)

factors affecting cost

1. Active material utilization fraction.

Not all of the active material actually gets used in the electrode. Obviously we want it to be high in the nickel electrode in particular because the cost of the material is high.

In the Edison cell the fraction of mass utilized can be calculated from the weights and composition of the electrodes Edison gives in the patents and this should be done. It can be increased with the addition of conductive carbon (like graphite) particles to the active material. Documents indicate that the homogeneity of the particle sizes is important in pocket cells to increase mass utilization, which should be no surprise as it would entail a smaller number of particles in between the gaps of large particles that are not compressed in any way against their neighbors, leading to high contact resistance and therefore low electrical coupling to the current collector.

For pasted electrodes it can be very high for both the iron and the nickel electrodes, patents indicate that it could be 80% without cobalt additive and almost 100% with it for the nickel. Smaller particle sizes and finer denser mesh helps too. Similar figures apply to the iron but since iron is cheap that will probably not be a deciding factor in the iron electrode design.

For sintered iron electrodes it is typically 1/3 or 1/4 of the total mass. The rest forms the conductive matrix of the current collector.

If the active material or a precursor that is later converted the the active material (like nickel oxide, probably NiO) is added before sintering the current collector that would need to be checked, but it would likely be high.

With electrodeposition and molten salt it should be very high.

For the nickel electrode the amount of conductive metallic material used (if metal is used, carbon works too) is typically 30% to 10 by volume of the electrode, so this will have a significant impact on cost if it is solid nickel. This may be a good reason to use a nickel plated steel of some sort or some other material as a current collector.

From patent number 00880978 Edison indicates that about 8 grams of the material the composition of which is described in number 839,371 is used per tube. The composition could be checked and the other patents searched (see research page for and archive of searchable pdfs of edison's patents) to find what the ampere hour capacity of each tube was, which I remember seeing but now forget, and then the efficiency of material utilization can be calculated. It is probably fairly low but can be improved by mixing in conductive particles of carbon and using a more uniform particle size. Mostly likely a pocket plate battery would be relatively expensive when made at a small scale, for these and other reasons.

The homogeneity of the materials is important apparently according to one of the journal articles and may be cheaper than additives.

2.Charge/discharge (round trip) efficiency (distinct from charge only efficiency).

This has a major impact on the cost of a solar power system that the batteries are included in. See [Impact of battery efficiency on cost]. To a rough approximation it can be represented as an addition to the $/Wh cost of the battery, and could add an additional $0.2 per Wh or more, which given that $0.2 per Wh is about the cost of lead acid AGM batteries (the normal type used in solar installations) that is a real problem.

However lead acid batteries are not that much better, actually, in real-world conditions. Nevertheless, the efficiency is one of the major considerations that should be optimized for. References indicate this can range from 50 to 65% depending on battery design, so it is not a given. Methods of enhancement so far identified include the addition of activators to the iron electrode like sulfur, selenium or tellurium compounds (which is best? Sulfur is the only one references seem to describe being used so experimentation or in depth knowledge of chemistry may be required here).

Forum posts and other information indicates that the efficiency actually rises over time to 80% or so after a few years of use, but this needs to be verified with other sources. Ideally the mechanisms could be identified and used to get this performance right off the bat.

This seems like one of the things an individual highly knowledgeable about chemistry could be particularly helpful with.

As the cost of the collector goes down over time this will become less important.

3. Cycle life. Although they get much longer cycle life than lead acid batteries, and it is to some degree inherent to the chemistry, it is not a given, and references indicate it can vary from 1500 to 3500 cycles, depending on the design of the battery, and up to 4000 with "good care" (issues which cause cycle life loss due to lack of good maintenance should be identified but this refers to non-sealed batteries so some of them might not apply to a sealed battery).

Cycle life unsurprisingly has a major impact on long term cost, so it should be maximized. It would be hard to test due to the time requirements, so this is a case where the higher quality references and perhaps in-depth knowledge of chemistry will be particularly needed.

Financial calculations should be done to determine how valuable very long lives are, although this is also important as part of OSE specs.

4. Calendar life.

Marketing material[www.nickel-iron-batteries.com] indicates that there are significant mechanisms that causes capacity loss independent of charge/discharge cycling. Some ideas on what it/they might be are included in the other reactions section.

In Edison cells one of the problems was the make-up electrolyte had impurities in it, and dust from the air would get into the battery and cover the electrode material particles. A sealed battery would of course avoid this, so there is another reason to do a sealed battery.

Problems with an unsealed cell

Ingress of stuff: Dust, carbon dioxide in the air which reacts with the electrolyte to form carbonate salts using up the electrolyte and possible being a problem chemically (need access to better references to know) Contaminants from the replacement electrolytes added, but from the solutes and that less than perfectly pure water is used. To some extent these will be expelled along with the electrolyte. May include some particulate matter.

Need to calculate how much of a problem these things would be without any sort of precautions, e.g. just a hole in the case.

Egress of stuff: Electrolyte solute, a mist is produced when bubbles break the surface of the water and this can result in loss of electrolyte solute as the water droplets that escape contain it.

Electrolyte solvent, due to the escape of mist and the conversion of the water to gas.

fortunately the active materials stay put, since they are hardly soluble in the electrolyte.

Prevention methods assuming lack of gas recombination ability: Gas permeable plastic film (polyethylene might do) allows only gasses to pass, keeping solute in and particulates out. Can be chosen to reduce the permeability of co2 relative to the other gasses.

Pressure regulated valve helps with ingress by only opening when the pressure inside the battery is higher than the atmosphere, thus preventing dust etc getting in by brownian motion as would occur with a simple hole in the battery case. Could also maybe be just a one way valve. In VRLA batteries they do not activate until a substantial pressure has built up because the hope is that most of the gasses will recombine in the electrode separator or if there is a catalytic recombiner, and the valve is more of a safety device. If there is no hope of recombination at all there may be little point in pressure regulated valve instead of just a one way valve. Maybe could be a liquid valve but then over the years the liquid might escape as mist etc. even if a nonvolatile liquid is used.

Design the vent hole to have a path that makes dust getting in harder, and as small as possible.

A porous material like the material used in doulton ceramic water filters, or plastic microporous material film (or a mini sterilizing filter as used in home brewing on the aerator air lines which use them) or fibrous mat could help filter dust and mist out of the air to prevent ingress and egress of those things.

An anti-mist filter like used on air pumps to remove oil mist could help prevent escape of mist.

Use high purity materials when replenishing electrolyte to reduce the addition of contaminants, could cost extra and/or be harder to obtain though.

Sealing the battery

The evolution of hydrogen and oxygen gas is a problem to a greater or lesser extent in all batteries with an aqueous electrolyte. This includes lead acid, NiMH and nicad, but the problem is particularly bad in nickel iron due to the quantity of gas produced being higher and for some reason the chemistry apparently not being amenable to the cheaper approaches used in the other types, possibly including that hydrogen is evolved even when the battery is not being overcharged.

This problem can be dealt with in a number of ways to the extent the battery can be considered "sealed" in that there should be no need to add electrolyte over the course of the batteries intended lifetime, which includes:

-Reducing it to an acceptable level so low that the supply of electrolyte is not the limiting factor in battery life. Especially helps if the battery life is not that long. Depends on clever chemistry, making the iron electrode (where hydrogen is evolved on overcharge) extra large to prevent formation of hydrogen gas until well after the battery is fully charged (because the nickel electrode is fully charged well before the iron one is, preventing the battery from storing any more energy after that point).

-Recombining the gasses There are a wide range of approaches used to do this in batteries of various sizes and types, see the research page for links and patents and notes. In sealed lead acid batteries they can get the gases to recombine in the porous electrode separator by getting the oxygen evolved to the other electrode, where it promptly recombines with the hydrogen for some reason.

The other common method is catalytic combiner caps, also called "hydro caps" which catalyze the conversion of hydrogen to oxygen. The big problem with this is that they usually use platinum group metals which are extremely rare and expensive, albeit in small amounts. Research needs to be done to determine if the relevant metals might be recoverable in the very small amounts needed locally. Patents indicate it might be possible to use the cheaper more common metals like osmium.

Research also needs to be done to determine if different more readily available catalysts like Raney nickel or something might be made to work.

The design of such caps is also not entirely trivial, as the catalyst surface must be kept free of liquid water and any mist from the electrolyte. A lot of heat can also be produced, especially during overcharge conditions.

It might be possible to ignite the mixture in a controlled and safe way with a spark gap or heated nichrome element, thereby eliminating the catalyst metals.

Whatever method is chosen there is also the choice of shared gas space and non-shared. In shared gas space the cells share the same atmosphere and so the water that is produced by recombination needs to be distributed equitably back to the cells again, or water or electrolyte solutes will accumulate in one cell or another. And thermal gradient will also result in redistribution of the water by evaporation and condensation. Non-shared gas space is the simplest in this regard, with each cell having it's own atmosphere, but then you need a separate combiner mechanism for each cell (although of fractional capacity), which may be more complex to manufacture.

Combiner caps made for lead acid batteries might work for us for now, and are readily available. But there may or may not be some minor snags with e.g. access to the catalyst being blocked by the solutes in the electrolyte accumulating, whereas in lead acid the electrolyte solute is also a volatile liquid itself rather than a solid in NiFe. Purchasing ones made for NiFe specifically has the problem that they would likely not be readily available and the supply might dry up in the future if the manufacturer stops making them or puts the price up.

However the battery could be designed, with some compromises, to work in either sealed or non-sealed mode.

Also, sometimes the gasses are not produced in stoichiometric proportions, usually there is an excess of hydrogen because the oxygen reacts with other components in the cell or electrolyte instead of going to O2 gas. Obviously this can't continue forever or it would cause major problems with the battery chemistry and limit battery life. It usually only happens for a while until equilibrium is (nearly) achieved, with the ratio of evolved gasses getting closer and closer to stoichiometric for recombination to water as time goes on, either during the charge cycle or with time after the battery is built. Thus there still needs to be provisions to allow some gas escape, even aside from the safety issues.

If it occurs cyclically (not clear yet form information available) then if venting the excess occurs every cycle, it would cause some loss of electrolyte. Unless it is quite low this is bad. Instead the container will probably have to be made to stand these transient pressures and contain the gas until the other of the pair is also evolved and recombination can proceed.

Electrode production options

There are many different ways to make the physical electrodes given active material. The Edison process involved tubes packed with powdered active material, nickel or graphite flakes, molasses and some additives such as cobalt hydroxide (see other reactions and additives section) under about 4000 PSI and with about 8 grams of the mixture per tube, each of which was about a quarter of an inch in diameter and formed of nickel plated perforated sheet metal and reinforced with rings of metal in some areas. See his patents for details. These are then connected together to form the nickel electrode. He made the iron is in either a similar way or by compressing iron active material into a brick with copper crystals, mercury and other additives.

A very similar process is was still used as lately as the nineties by some manufactures, who may still be using it. For some reason they use smaller tubes only a few millimeters wide.

The changhong batteries are pocket-plate type. Verification is needed on exactly what pocket plate indicates. It can include tubular but there seems to be a distinct method as well in use which involves essentially denting the surface into pockets, much like a tv dinner tray. This could be easier and cheaper.

Other current collectors that are described as being or having been used commercially are: Nickel fiber mesh and paper

Sintered nickel powder

A foil or mesh on nickel or nickel plated steel which is covered in powder then sintered to bond the particles to it, increasing surface area

Foil (for teflon bonded) can be bend or scored or etched to increase surface area

Nickel foam- nickel powder is mixed with e.g. polyurethane foam particles the surface of which they coat, packed together and heated at greater than 700 degrees, removing the polyurethane and leaving the nickel.

Nickel plated steel wool

Others described in patents but which don't seem to be used:

Carbonized polymer material, heat some polymers to 900 degrees and they carbonize into a relatively durable and conductive material which is chemically compatible more or less with both the nickel and iron electrodes. Carbon particles of activated carbon are mixed in too.

There may be others as well

Other ideas:

conductive plastics?

Getting the material onto the substrate:

Active material includes additives etc.

Used commercially: Mix the active material with teflon or another hydrophobic plastic particles, graphite particles, and some elastomer and a thickener. Exact compositions described in patents. Apply to mesh or foil and press hard, like 700 kg per cm2.

Mix it into a paste and apply to a mesh or foam by running the foam or cloth through rollers along with the paste. Apparently it stays put in the foam without any additives like thickeners or binders though I would have though it would slowly fall out. de-emulsifiers, thickeners and other additives are sometimes used to make the details of production easier.

A common method called vacuum impregnation is to dip the mesh or foam in melted nickel nitrate and then into hot sodium hydroxide solution, converting the nitrate to hydroxide. Presumably it is done under a vacuum else the air in the foam would prevent entry of the liquids.

Electrochemical impregnation, there are several ways to do it and a chemist could probably come up with more, but basically the right nickel compound precipitates out of solution in fine particles on the surface of the nickel matrix. The source of the nickel can be either the metallic nickel matrix (beefed up with extra nickel of course) or the surrounding solution.

Described in patents but not apparently used: In some cases the active material can be mixed with e.g. nickel fibers or fibers of the polymer to be carbonized, then the whole thing heated, bonding the fibers, flakes, whatever, together into a continuous mesh. The heating does not damage the active material if it is formulated right and done in the right atmosphere (not hard though).

list of tasks that need to be completed

If you can do any of the things below please do, and put a link to the location where the completed work is, or promote that item to a new section on this page. When the Pivotal tracker comes online these can be moved there.

-Decide between pocket plate and nickel plated steel wool: Can we get away without the diffusion bonding? Diffusion bonding takes hours at very high temperature, leading to high capital and energy costs. Other methods could maybe be used: ultrasonic welding, resistive welding, the fibers may be connected during the electroplate process satisfactorily. Also, the diffusion bonding's purpose is to reduce the bulk resistance of the steel wool but maybe we don't absolutely need to do so.

-look into electroless plating instead of electroplating. It looks as though it would require a lot of chemical and produce a lot of waste which would be somewhat toxic, compared with electroplating. For the iron electrode the pocket plate should be relatively easy because it doesn't have to be under such pressure, and it is more conductive. The battery handbook 3rd edition says they are made with stamped pockets from a steel sheet (much like a TV dinner tray). -Identify active material usage fraction for various electrode designs -come up with some reasonably good and workable battery designs that mesh with OSE specs - calculate materials and estimate manufacturing cost for them and pick what looks like would be the cheapest - hash out what would be the production process so have a reasonable idea of the complexity of what it will be before finalizing decision to use this design. If seems to be getting ungainly, go back and do likewise for a different design and maybe it will be more makeable. - produce bill of materials for prototype -identify suppliers -build and test it well -repeat thrice integrating knowledge gained in previous prototypes while hashing out the details of the eventual production line. -design production line -track down suppliers and produce bill of materials -prototype pieces of it -build the line and get it working.

Notes on interpreting patents and other documentation

Chemists have a terrible habit of interchanging the names of compounds as if they were synonyms based on family relations, and glossing over details. Nickel Oxyhydroxide if often referred to as nickel hydroxide in this context. Incidentally so is actual nickel hydroxide, which can get confusing for someone with limited knowledge of the art.

A nickel-iron battery likewise may use not just iron as the negative electrode but can also make use of the higher iron oxides like Fe3O4 (ferric oxide) called iron oxide as a reactant (need to check details on this). Similarly FeO (ferrous oxide or iron monoxide) is a problem in the batteries but is likewise called iron oxide. Similarly there are a couple nickel oxides and the name is often used to refer to several.

Note that often the individual electrodes are sometimes talked about (like with the iron plateau thing) in the absence of the other, which can happen because sometimes you can think of the battery to some degree as 2 half cells and many reactions proceed almost independently of what is going on at the other electrode.

Also it is a convenient way to talk about what happens if the electrode is completely charged/discharged or whatever which might not happen in a real battery. But they can never be fully separate, reactions at one usually end up influencing the other at least a bit, and sometimes important ones are basically prerequisite for certain reactions at the other. But it is interesting partly because they can get "out of sync" to some degree sometimes and be in different states of charge.

Basically you have to step away with the simplistic view of batteries that we are usually presented with and think of it in terms of the complex electrochemical system that is is.

See wikipedia "cathode" for an explanation of which is the cathode in an electrochemical cell.

Current project status

Basically, having gone through all those patents and documents and so much more (check the edit history) my recommendation as to how to proceed is:

1. Obtain a a changhong battery. 2. Performance test it with a battery analyzer fr maybe 20 cycles to collect a bunch of data regarding charge discharge efficiency at various states of charge etc. This should be available from the manufacturer, but I have found that it is not in fact. Probably for marketing reasons. They figure you will just cross your fingers and buy it anyway.

We need this info anyway so we know what we are competing against in the market.

3. Dismantle the electrodes to look at their construction. In all my travels I have not been able to determine how pocket plate electrodes are made exactly this [1] is about it. Although there are many different ways and the basic idea is just to pack the material in at 2000 to 4000 psi (it becomes a solid mass probably a bit like soapstone) and then hold it at a pressure of about 40 psi. The main purpose of the pressures is to decrease the contact resistance between the constituent parts.

4. If the performance is higher than expected they may have developed new additives, especially if the efficiency is better than 60%. Check the battery for patent numbers. Contact the company to see if the battery or any part of it is patented. Look up the patents.

Under US patent system, they cannot conceal this information in the slightest - it must be fully made known - or the patent is not enforceable. That is why stuff usually has "protected by patent x" on it.

5. If there are not patent problems, analyze the active material by microscopic and chemical analysis to determine what the additives are so we can copy them. Fear not, this is not unethical as this is SOP in manufacturing.

6. Design battery Open battery prototype I while keeping the production line in mind. To keep costs low there will probably need to be a bit of custom equipment for electroplating etc which can be made with the replab but not really be part of it.

7. Build, test.

I think it is most probably by a ways that pocket plate is the best, for a variety of reasons, mostly simplicity of manufacture, not tubular and not sintered and not mesh and not teflon bonded and not nickel plated steel wool. Changhong even uses pocket plates for their starting lighting ignition batteries so it should be good.