A funny way to recycle dead batteries (part 1)
Let’s say you have a steel part in your project that needs to be protected from rusting and also has to have a nice look. In a hobby workshop you have a number of options that we’ve discussed in the previous article. One of them that I particularly like is phosphate conversion coating. In the simplest case you just dip the part in a solution called “rust converter” or something like that and in a couple of hours get a result like this:
The solution contains phosphoric acid that etches thin surface layer of steel and converts it into iron phosphate that, being insoluble, stays on the surface, firmly attached to the base metal. The resulting coating is relatively wear-resistant and behaves decently even on threaded connections. In its original state it is also mildly corrosion-resistant, but may be made even more resistant by dipping the part into linseed oil. The oil seeps into the cracks and pores of the phosphate and in some time polymerises there, creating a reliable barrier for water, oxygen or other corrosive factors. Polymerisation takes place due to oxidation of oil’s unsaturated fats by air and is catalysed by metal ions in the phosphate. This is a good example of synergetic behaviour in practice, when two coatings enhance each other’s performance. In order to accelerate the oil’s polymerisation the part may be mildly heated. Alternatively, it might be heated to >300°C before dipping into the oil to produce a thicker hybrid phosphate / black oxide coating.
The hybrid process is hardly applicable for the larger parts, and the vanilla phosphating, even though I do like how it works, still produces only a very thin film, probably in the range of microns. It may give a bit better coverage on rusty parts, but on shiny steel it is very hard to grow this protective layer at all, that’s why the thing is called “rust converter”. The difference between clean and rusty parts is, of course, in chemistry.
Reaction of pure metal with acid is a RedOx: it involves oxidation of neutral metal atoms into metal ions and simultaneous reduction of hydrogen ions into hydrogen gas. The rust (iron oxide), on other hand, undergoes a neutralisation reaction. Both reactions are thermodynamically favourable, but the RedOx implies electron transfer from iron to hydrogen and gas evolution and this path has higher activation energy and therefore slower rate. Apart from this, rust is a highly porous material that immediately exposes large surface area for the reaction to take place.
In industry to overcome the limitation of slow etching rate on clean steel parts and to grow a desirable thick coating, several improvements are made over the primitive etching with phosphoric acid. One is really obvious — rising the temperature to just-below-boiling (like 98°C) to increase the reaction rate. Another is adding sodium nitrite. Its function is to act as oxidiser to prevent formation of hydrogen gas and circumvent the kinetic barrier associated with this process. In layman’s terms such an oxidiser allows fresh metal to act like a rust upon contact with the acid. How can nitrites be oxidisers while they are known food-preservant anti-oxidants? Well, nitrites turn out to do both things, depending on the acidity. Lastly, a mixture of metal phosphates is added to the acidic solution to improve density of the coating.
Wait, didn’t I say just before that we are expecting a film of insoluble metal phosphates to form on the metal surface? How could it be then that we are dissolving metal phosphates in the bath beforehand? That’s an interesting point. The phosphoric acid has not 1 but 3 acidic hydrogen atoms, each of which may or may not be displaced by a metal ion. If only some of the hydrogens are displaced, compounds known as acidic salts are formed. To make things worse, iron is a complicated atom too, it can form ions with charge +2 or +3.
And if that was not enough, in solution there could be not “the” phosphoric acid either, but more like phosphoric acids.
And how are we supposed to make sense from all of it? Well, for practical purposes we might consider that we are dealing with just a mixture of metal, hydrogen and phosphate ions, taken in some proportion. The more hydrogen ions are in the mixture, the more soluble is the “compound”. When there are no hydrogens left, an insoluble metal phosphate film is created. Now it should be clear, what role play the metal phosphates, dissolved into the bath: with the free phosphoric acid they form those acidic slats in the solution. Acidic salts contain less hydrogen than pure acid and therefore on reaction with metal on the clean surface they form more phosphates and release less hydrogen, which translates into thicker and denser coating for the same amount of oxidised surface metal. An additional benefit may be extracted from this approach by partially replacing iron in the soluble salts with phosphates of other metals, typically manganese and zinc. The resulting mixed-metal-phosphate layer often offers better mechanical and aesthetic properties than pure iron phosphate.
So, enough theory let’s start some witchcrafting. Getting phosphoric acid from industrial suppliers was easy as it is widely used and is not in the lists of controlled substances. Next I needed the mixture of metal phosphates. In principle, there are such mixtures available to industry specifically for conversion coating processes. But those are not as popular and the suppliers did not agree to sell it in any fractions of a freight wagon. No problem, let’s etch some metals with the acid to get the needed mixture! What we need is finely divided pieces of iron, zinc and manganese, or their oxides. Wait a minute, I know the name of that mixture — dead batteries! Right, the usual alkaline battery consists of zinc and manganese dioxide powders, stuffed into a steel enclosure. You won’t believe how easy it is to get a few kilo’s of them if you just put an advertisement in the corporate’s flood chat.
I’ve cut them in halves with pipe scissors, and dumped the halves into a 3L glass jar. Then there is a small problem on the path to the desired phosphates: the alkaline batteries contain potassium hydroxide as electrolyte and it would be inefficient to mix it with acid as they would undergo neutralisation. However, just boiling the pieces with water and filtering out the solution (and repeating it 2–3 times) is enough to get rid of most of the hydroxide (and I’ve also got a highly alkaline solution suitable for technical purposes as a bonus). Then I added 1L of phosphoric acid and boiled (actually, heated on a water bath) for a couple of hours. The acid etches the metal shreds and what’s left in the solution is a mixture of acidic metal salts that is the product we need. After filtering out the liquid, the procedure may be repeated a couple of times with fresh acid, while there still are shreds left. One nasty nuance is that the batteries come in plastic wraps glued on them. The glue then makes its way into the solution and results in blobs of goo floating around. This goo tends to clog glass and paper filters and it probably would be wise to pass it through sand first. As a note, the batteries also contain small amount of brass in the central contact and nickel in the coating of the steel housing. Brass consists of zinc, which we already have in the mix, and copper, which does not react at all. Nickel is present in minor quantities and does not tend to react eagerly either. A chemist might notice that manganese does not form phosphates when it is in higher oxidation states (+4 for dioxide), yet in the dead batteries it is already partially reduced and will be reduced even further by a cementation-like process due to the presence of other metals being etched nearby.
If the boiling of shreds is done for enough time and there is an excess of shreds, then the resulting solution will be just slightly acidic (saturated with metals to a point where they are starting to fall out as insoluble phosphates). This is not a right composition for the actual bath and it needs to be corrected for two parameters: concentration of metal ions and concentration of hydrogen ions (acidity). There are practical recommendations in the literature about achieving proper balance of those. The procedure involves taking measurements by titration with two different indicators (for so called free and residual acidities) and adding water and acid according to the measurement’s results.
Ok, the solution is ready now and we need a vessel for the procedure. It has to be robust, accommodate the long parts I’m willing to process and endure repeated heating to 98°C. The obvious choice would be a capped thick-walled steel pipe, which I was fortunate to buy at a local scrap-yard. I would prefer, for sure, a stainless steel container, but cost-cutting, you know, is what happens these days all around. After trimming it to length and finding a pair of matching flanges, the welding may begin.
There is more material left that I’d like to describe in details about the construction of the bath and the achieved results but I’m already at the limit of the contemporary human’s 5min attention window, so I’m going to do that in the next part of this series. Stay tuned, subscribe, give a like, you know it!
