Sciencemadness Discussion Board

Li/MnO2 batteries

DerAlte - 14-6-2007 at 23:13

Recently I was given a small battery of this type with the news that it was ‘dead’, at least according to the electronic device (camera) it was designed for. I decided to dismantle it. I consulted the Web for the manufacturer’s description and determined that this type was made with lithium metal using organic electrolytes (forget what they were, have a note somewhere – smells ethereal) with an MnO2 anode..

The result, to my surprise, was that the battery still had considerable lithium in it. And it was not dead - the hacksaw produced sparks as it shorted the cell, it became hot. Should have tested the OC voltage.

I extracted the lithium as a rolled sheet, blackened on one side, and stored it under SAE 30 oil ( don’t use multigrade or any oil with additives – SAE 30 seems to be straight hydrocarbon, though I did notice initial bubbling, probably because the oil had some water in it). Never having played with elemental Li, I burned some and tried a bit with water, etc. Fun!

The anode is solid MnO2 contained in a rubbery compound, supported on an expanded metal substrate. My question is, what is this metal? It did not dissolve readily in HCl and is faintly magnetic. It dissolved in a mixture of strong HCl/sodium nitrate (the poor man’s aqua regia – don’t have any nitric acid at present). Iron is present – ferrocyanide test. What else? I expected chromium for a stainless steel but haven’t yet managed to detect it. Does anyone know what this mesh is?

Regards,

DerAlte

not_important - 15-6-2007 at 00:56

Electrolyte is propylene carbonate, possibly with 1,2 dimethoxyethane, and lithium perchlorate.

Hard to say what the expanded mes is. In the research developing Li/MnO2 batteries a number of metals were used. Try testing for nickel and copper, too.

DerAlte - 15-6-2007 at 22:23

Ni I can understand, but Cu? With perchlorate?

Thse cells are a work of art compared with alkaine or Zn/C types. The electrolyte composition returns to me now you mention it. I originally got the composition off the MSDS for the Duracell battery. MSDS sheets can be quite useful sometimes.

The organics must have a fair dielectic constant - I assume the perchlorate does the work.

I'll test for Ni, Cu, then. I was hoping it might be something more exotic. The mesh is certainly very resistant to HCl. But so are some Fe/Cr stainless steels. I was wondering if this mesh might make a useful anode for something.

Regards,

DerAlte

jtkelectroman - 16-6-2007 at 11:44

I highly doubt the mesh is made of copper. I opened one of those batteries up once and removed the lithium and the strange wire mesh also. I took some of the wire mesh and placed it in a hot torch and I got a deep dark reddish purple emission spectra. so it's certainly not copper. It could be some sort of lanthanide though.

olmpiad - 24-6-2007 at 10:29

I would expect your red color to be from lithium contamination, as it doesn't take much Li to color a flame.

not_important - 24-6-2007 at 18:35

The lanthanides are pretty reactive, unless only a minor part of an alloy, so I doubt the mesh is made of that.

olmpiad - 24-6-2007 at 19:23

I think that a lot would have to do with the brand and size of the battery. Although Li batteries are common, there are many different chemicals used in them, so most of them vary slightly from each other.

bio2 - 24-6-2007 at 20:56

I have a Lithium Ion battery from a laptop computer that is
discharged, it has 8 cells and is quite large.

If this battery is fully charged before disassembly may I then assume that the Li electrode is then more or less pure lithium?

S.C. Wack - 24-6-2007 at 21:56

No. The original poster is speaking of a quite different primary battery. It would be interesting to see what could be done with the Li-ion batteries, though.

DerAlte - 25-6-2007 at 09:53

Absolutely, S.C. Wack. IIRC only LiMnO4 cells contain naked lithium.

I mislabelled this thread. I did not intend to provoke a discussion on Li batteries. I was interested in the metal substrate used to hold the MnO2.

I now regret removing the MnO2 from the mesh. I could use an MnO2 anode!

The mesh produces a deep lime green solution when treated with HCl + nitrate, assumedly ferric and Cr(III) ions. Aqueous ammonia gives a flocculent hydroxide ppt., looks like ferric but may also contain Cr(III). The solution becomes crystal clear. I have this as a sideline waiting while I do other things.

The existing thread https://sciencemadness.org/talk/viewthread.php?tid=8368 is also investigation into stainless steel alloys, which I feel this mesh must be.

Regards,

DerAlte

bio2 - 25-6-2007 at 12:24

This is the mechanism of Li-Ion from Wikipedia.

So how to get Lithium and/or Cobalt element from this.

....The anode of a conventional Li-ion cell is made from carbon, the cathode is a metal oxide, and the electrolyte is a lithium salt in an organic solvent.

The underlying chemical reaction that allows Li-ion cells to provide electricity is:

Li(x-1)CoO2 + Li(x) C6 = C6 + LiCoO2

It is important to note that lithium ions themselves are not being oxidized; rather, in a lithium-ion battery the lithium ions are transported to and from the cathode or anode, with the transition metal, Co, in LixCoO2 being oxidized from Co3+ to Co4+ during charging, and reduced from Co4+ to Co3+ during discharge..

DerAlte - 13-7-2007 at 21:11

To wrap this thread up, I have finally got round to the analysis of the composition of the mesh. As far as I can determine, it is merely Fe/Cr stainless steel, no Ni or anything else. Prosaic! Only used mg quantities. Being almost non-magnetic but not quite, the ratio Fe:Cr can be estimated.

DerAlte - 19-11-2008 at 19:51

S.C Wack said, about a year ago (above)
Quote:
It would be interesting to see what could be done with the Li-ion batteries, though


Well, I have, so to speak, now been there and done that. Here’s a report.

For the construction and other details of Li Ion cells, Google and see Wiki. The one I had was an ex-laptop Sony, of which I took one flat cell, weighing about 50g. There are no toxic substances in such a cell (except in the toxic state of California, where cobalt is known to to be possibly carcinogenic, as determined by painting over shaven genetically chosen rats with cobalt salts every four hours until they croak).

The cell consisted of a flattened spiral of 8 layers; a black layer of Lithium cobalt oxide Li(1-x)CoO2 weakly adherent to both sides of an Al foil strip, followed by a permeable polyethylene separator, then a Cu foil coated on both sides with graphite and intercalated Li, (CLi(x)), and finally another separator.

Electrolyte is, I believe, LiPF6 dissolved in cyclic ethylene carbonate (1,3-dioxacyclopentane, 2-one). It has a ketone smell but there does not seem much of it, just enough to wet the cell. It is highly water soluble. Of LiPF6 I have no idea of its properties but it must ionize in the solvent, which has a dielectric constant of ~65 due to the high polarity of the solvent molecule.

On discharge, the Li+ ions move to the cathode and fill the interstices of the CoO2 crystalline structure; on charge they move to between the graphite planes. A fully-charged cell should thus have most of it Li content interstitially on the carbon anode. The Co would then be in its highest oxidation state, near IV. A discharged cell should have the Li content mainly in the cathode of CoO2, and the Co should be in the II oxidation state, as LiO.CoO, while the anode is mainly graphite.

Note:Warning: The cell must be discharged first. The fact that the battery is considered dead does not guarantee the cell hasno charge. Do not short it directly but through about 10 ohms or a suitable lamp, eg 6v, 2w. Otherwise shorting out may cause the cell to get very hot and possibly burst

Cut off the plasic protected Aluminum case.

The spiral should be carefully unwound and the layers separated, taking note of the coating next to the Al foil, if it adheres to the polyethylene film. I found the Cu film tended to hold the graphite better. The cobalt oxide did not stick to the Al but to the separator. The films will show a white deposit once opened to the air.

Remove the solvent and electrolyte(?) by steeping the electrodes and their seperators separately in water. I assume the LiPF6 either dissolves or decomposes. I could obtain nothing identifiable from the solution but on evaporation a small amount of white precipitate was formed, which would not redissolve in water.

Carefully scrape the black coating off the Al foil where adherent, and off the separators adjacent to it. If the coating does not come off the seperator cut it up as I did, into small pieces. There is an organic binder involved here. Place all the pieces of black material and the cut up separator in a vessel and pour on concentrated HCl. It will quickly turn blue due to formation of CoCl2.

Warning: Chlorine gas is evolved at this stage due to presence of higher oxides of Co – do outside or in a fume cupboard.

The amount of acid to be added must be judged – do not overdo it, the excess has to be got rid of by evaporation. Let it soak for 24-48 hours, at a slightly elevated temperature (say 40C).

The intense blue color is said to be due to (CoCl4)2- ion complex, which changes to Co2+ (pink) on dilution or reduction of Cl- concentration. Pour off the mother liquor, wash off the residue and let crystallize as ruby colored CoCl2.6H2O. Evaporate as necessary. My yield was 22g of cobalt chloride hexahydrate.

The remaining liquor contains CoCl2 and lithium chloride. Treat with sodium carbonate. I had hoped to precipitate the cobalt carbonate first by carefully titrating, leaving the slightly soluble Li2Co3 (~1g/100g aq.) in solution. The pink-purple precipitate was very flocculent and slow to settle and this proved impossible, so excess Na2Co3 was added to precipitate both carbonates (~8g). I have still to separate them.

The aluminum and copper electrodes were 69 cm x 9 cm, the Cu a calculated 0.0016 cm thick, weight 9g approx. A useful piece of copper foil. The electrode active area was about 900 cm^2, which accounts for the very low internal resistance. Burning the carbon from the anode showed lithium still present in the flame spectrum.

Regards, Der Alte

blogfast25 - 20-11-2008 at 08:49

Der Alte:

Just one minor detail:
Quote:
I assume the LiPF6 either dissolves or decomposes.


I believe HPF<sub>6</sub> is one of the strongest mineral acids known. Combined with a strong alkali (say LiOH) the resulting salt should be extremely stable.

Good work though!

DerAlte - 4-2-2009 at 22:24

Extractions from NiCd batteries.

I have put this here because the thread appears to have metamorphosed into ‘what can be salvaged from batteries’. Precaution: Cadmium is quite poisonous - more than lead and also cumulative, and a suspected carcinogen. All group IIb metals are toxic. Zinc is required in man in small amounts, used in conjunction with enzymes/co-enzymes - assigned RDA is 15 mg/day. Cd replaces Zn in the enzymatic conjugate removing its catalytic effect and hence poisons the catalyst.
The rat-oral LD 50 of Zn++: Cd++: Hg++ is of the order 200:10:1. One rapid effect of Cd poisoning is impairment of lung action if inhaled. Care must be taken not to heat Cd compounds in an enclosed space without adequate ventilation. Use latex or neoprene gloves.
Because of toxicity NiCd cells are being phased out in favor of the far less toxic Nickel-metal hydride type. They are no longer freely available in many countries, esp in EU and US. Cadmium plating, once common, is also on the way out. Nickel compounds are less toxic. I find them a skin irritant.

NiCd Cell Chemistry: This is alleged to be:

Cd + 2OH- —> Cd(OH)2 + 2e- 0.81v anode
NiO2 + 2H2O + 2e- —> Ni(OH)2 + 2OH- 0.49v cathode

<--- Charge
Overall cell : Cd +NiO2 + 2H2O —> Cd(OH)2 + Ni(OH)2 1.30v
Discharge - - - >

{Or, (Wiki), NiO(OH) + 2H2O + 2e- —> Ni(OH)2 + 2OH- , where Ni is Ni(III); elsewhere I have seen the cathode described as hydrated IV oxide}
Electrolyte, aqueous hydroxide (KOH or NaOH): it merely serves to transport OH- ions, just as the Li salt does for Li+ in a Li ion cell; it is not consumed. But Wiki points out that overcharging electrolyses the water of solution, and will deplete it. To avoid production of explosive hydrogen, the Cd anode is designed for a higher capacity than the cathode.

Process:

Straightforward simple chemistry. The cell was approx. C size, ex a very old laptop computer. Weight, ~50g. It appeared fully discharged and had been for ages. The cell had leaked at some time. The internal structure was a crude spiral of a few turns. The separator appeared to be paper or cellulose material – almost dry and friable. Both anode and cathode strips were very dark, almost black, with adherent white material. I had expected, per the above, greenish for the Ni(OH)2 and white for the Cd(OH)2. Anode and cathode materials were supported on metal meshes, woven (I believe) for the Ni and expanded for the Cd side. They were covered both sides (because of the spiral construction). These compounds were obviously not the hydroxides of the discharged cell.

Addition of water to the fibrous separator produced a soapy feel, and a flame test showed predominant K+ ion. A small amount of separator fizzled in dilute HCl – assumedly the KOH had absorbed CO2 from the air.

The separated strips were soaked in water for 48hrs to dissolve any KOH or K2CO3 adherent. A piece of the Cd strip was cut off and heated to red heat with a small torch flame. It ignited and produced red-brown fumes, characteristic of metallic cadmium burning to oxide (brown) – the oxide is not volatile but the metal is. (See warning above re vapors of Cd and fumes). Conclusion: the black substance was mainly Cd metal.

The cell seemed to have failed by overcharging & depleting the electrolyte solvent. This would account for the nickel side looking black too, as (hydrated) oxide. The adherent white stuff was possibly Cd(OH)2.

Dilute H2SO4 (c. 30%) was added to the cut up Cd strip pieces. Copious amounts of H2 evolved (tested by burning), confirming it was mainly metal. The resulting liquid was turbid but colorless with white stuff, which I concluded must be disintegrated separator material: Cd(OH)2 would have dissolved. The mesh appeared to be untouched – no colored ions were seen, e.g. from Fe, Cr, Ni (i.e. stainless steel). I suspect the mesh was at least Ni plated. (Ni is very slow to react even with strong acids).
After filtering off bits of mesh and crud and evaporating (care! Cd++ ions may be carried over in vapor), an unsatisfactory attempt was made to crystallize the sulphate. CdSO4 is rather soluble (>70 g/100g aq.) and decreases solubility with temperature. The clear liquor was put aside to deal with the Ni compound.

The mesh holding the Ni compound was severely disintegrated and broke up easily. It was reduced to powder in a mortar. The granular power was magnetic but no efficient separation by these means was attained – the oxide (?) may also be magnetic. It was mixed with 30% HCl (because I had NiCl2 already) and gently heated for several hours. A very dark green solution was obtained {Warning: a moderate quantity of Cl2 gas was evolved, showing some of the Ni compound is Ni(III) if not Ni(IV)}

NiCl2 is also fairly soluble, varying from around 50 at 0C to ~90g/100g aq at 100C. After filtering, to get rid of the mesh metal (Ni, I suspect again) and white crud, the solution was evaporated to give green crystals of the hexahydrate. However, little was obtained and much remained in solution.

I decided to convert the metal ions to insoluble carbonates or hydroxides, but the hydroxides and carbonates have close solubility products. Na2CO3 is highly alkaline, so I suspected that basic carbonates of unknown composition might result and opted for the hydroxides. The solutions of the Cd sulphate and Ni chloride were diluted with ~20 times volume of water and 10% NaOH added. Unlike zinc, Cd is not amphoteric: excess NaOH does not redissolve the Cd hydroxide – ‘cadmates’ do not form..

The cadmium hydroxide produced was a flocculent heavy white cloud which settled well. The Ni(OH)2 was initially a turquoise color, no doubt due to hydration, which turned green upon standing. Both were filtered, washed and dried carefully.

According to CRC, both hydroxides are somewhat unstable on heating. Cd(OH)2 begins to decompose (to oxide) as low as 130C, so controlled drying is needed. Ni(OH)2 is stable to 200C and easily dried in an oven.

Both hydroxides easily react with dilute acids allowing any desired salts to be made. Yield was 13g Cd(OH)2 (pure white) and 10g Ni(OH)2 (dark green).

{I did a few experiments with the Cd. Cadmium is rather uninteresting. All salts are colorless with a colorless anion. CdS is yellow and CdO brown. It forms a few complexes (e.g. with ammonia, cyanide) but less than zinc, which it otherwise greatly resembles. It electroplates easily from acidified sulphate. The deposit on a copper wire was easily scraped off – it is a bit harder than lead. It melts easily (at 321C, but the vapor pressure is as high as 1mm Hg at only 394C (CRC)). It burns much like zinc but less energetically, giving a red/brown cloud of oxide instead of yellow/white. Remember toxicity if you try this!}

{Note: Other cells may be found in a more discharged state than the one I used. The same process could be employed if the electrode material is predominantly hydroxide.}
Der Alte

Rosco Bodine - 5-2-2009 at 00:39

Fructose or glucose will reportedly reduce the nickel, copper, ect. to the free metal on boiling a concentrated mixture of the cadmium and nickel hydroxides keeping pH above 3.5...but leaving the cadmium which should not be reduced
by the sugar.

See US4818280 attached.

[Edited on 5-2-2009 by Rosco Bodine]

Attachment: US4818280 reduced 3 micron_nonferrous_metal_p.pdf (274kB)
This file has been downloaded 670 times

DerAlte - 5-2-2009 at 12:50

Fascinating, Rosco. Finely divided Nickel is a useful catalyst.

As far as extraction is concerned, it appears that there is little metal ion migration in the NiCd cell. The metal appear to be held in place since there is no metal ion transferred on charge or discharge. The stuff I got seemed of fairly high purity - the Cd(OH)2 was a pure white color, the Ni(OH)2 a nice green.

Incidentally, where do you find all these obscure patents?

On the subject of derivation of reagents from the crud of failed cells, I am waiting for a dead NiMH cell. Lanthanum, Praeseodymium etc. are used as the M.

This is recycling for fun, not for political correctness! So far I have extracted fairly pure - say reagent grade - Pb metal, MnO2, carbon electrodes galore, Co, Ni, Cd salts, Li metal, Li salts, Zn metal. All you need is HCl or H2SO4, NaOCl and an alkali, plus Na2CO3 and a source of heat.

Der Alte

[Edited on 5-2-2009 by DerAlte]

DerAlte - 5-9-2009 at 11:31

What about the lithium?

I intend to repeat the process for extraction cobalt and this time separate the lithium. First, I’ll restate a paragraph from the earlier post:
Quote:
On discharge, the Li+ ions move to the cathode and fill the interstices of the CoO2 crystalline structure; on charge they move to between the graphite planes. A fully-charged cell should thus have most of it Li content interstitially on the carbon anode. The Co would then be in its highest oxidation state, near IV. A discharged cell should have the Li content mainly in the cathode of CoO2, and the Co should be in the II oxidation state, as LiO.CoO, while the anode is mainly graphite.


I believe most of the Li will be in the cathode in a discharged cell. There is some as LiPF6 in the electrolyte, but the amount of the electrolyte is quite miniscule – it seems to merely wet the cell. It is not consumed, merely a source of Li+ ions. This LiPF6 will probably hydrolyze in water to LiF and PF5 which will then further hydrolyze to HF and a phosphorus acid. Li3PO4 is highly insoluble. So we’ll ignore the part in the electrolyte.

Use the same procedure as before and scape off the cathode material. This time I will ignite it to get rid of the organic binder. This should make for easier and faster acid solution, and allow a better estimate of the amount of acid needed by weighing it.

The cobalt will next be precipitated as (hydrated?) Co2O3 using hypochlorite, again as little as possible. This should leave LiCl and NaCl in solution only. Separating by filtration, the addition of the estimated amount of Na2CO3 and boiling should precipitate most of the Li as carbonate (Li2CO3 is least soluble in hot water ~0.7g/100g aq). At most 5-6 gms are expected per cell! Not worth the bother but an interesting excercise.

The cobalt can be treated as above.

Regards, Der Alte

DerAlte - 5-9-2009 at 11:32

Silver from silver/zinc cells

While I’m at it, how to get the silver from these. These buttons are still used in watches and calculators. The positive terminal uses Ag2O. This gets reduced to Ag in the discharged cell. So all you have to do is remove the cathode material, place it on a charcoal block and apply a flame from a blow torch. Any oxide is reduced, the silver melts. Dead easy!

Regards, Der Alte

IrC - 10-9-2009 at 00:21

Quote: Originally posted by not_important  
The lanthanides are pretty reactive, unless only a minor part of an alloy, so I doubt the mesh is made of that.


That depends upon which one. I have lots of every one except of course for the non existent Pm and some you can blast away with a torch doing very little. Yb, Tm, Er, Ho, Tb, Gd, and Dy take a serious amount of heat to get to oxidize, more than you might think when reading how they react in air.

Lu is the best example and Dy, Gd take very high heat as well. Sm reacts slowly, glowing red and oxidizing after you apply more heat than you might think is needed, very strange stringy looking metal to work with for sure. While they claim it burns in air it is more like a slow red glow traveling through the metal, nothing like burning magnesium. I have played around making various oxides and carbonates for ceramic experiments from them all and Lu will sit it an alumina boat which is glowing bright red doing very little.

Lu is probably the hardest lanthanide I have had to cut and you need an oxy-acetylene to get it to react much (very hard on Alumina boats especially when trying to keep the flame and gasses away from the metal), even an oxy-mapp has little effect on Lu. Er, Sm and Dy are very hard to cut as well.


[Edited on 9-10-2009 by IrC]