Magnesium stearate as an alternative to potassium dichromate for short term protection of AP / Mg mixtures, and Shimizu's long term / improved passivation method

[PyroGnome]

Although dichromate isn't stable long term, it's used fairly often for protection of Mg in AP-oxidizer rockets since many people press & fire them in a short time period anyway.  Stearic acid has been used for a long time for the coating of some aluminums and as one of the many ways of coating iron / steel powder for fountain mixes and such, as well as being a low temperature fuel for some chlorate blue star mixes.  The attached paper tested magnesium stearate coating (by mixing 1:5 Mg stearate / magnesium with a weight of water equal to weight of magnesium and allowing to sit for over 12 hours (until dry) in a sealed vessel with silica gel. 

They then tested degradation of the raw mixture sitting in a container (not pressed, just mixed) with humidity held at 75% by wet NaCl at 40C for 24.  Their results were:

Unprotected Mg:  ~16% degradation

Mg coated with Mg stearate: ~2% degradation

Mg coated with dichromate:  <1% degradation

Not fantastic, but it'll make things safe enough if you're pressing and firing that day or the next.  MgSt didn't affect color, but it does decrease burning rate by 15%.  Directly using stearic acid reduced burn rate slightly more than the stearate salts.  Potassium dichromate has a strong catalytic effect with AP which promotes continous burning and lowered ignition temperature (as does Copper Oxide) and shimizu's pure guanidine nitrate strobes needed it to function, although the Jennings-White paper seems to indicate that a wider variety of materials will cause the strobe effect in AP oxidized systems and sulfate salts are usually sufficient.   catalytic amount needed in the mixture (hopefully not) but the less dichromate you need to use the better so even if it's still necessary in the mix you've cut out needing to store a solution.   It sounded like the other stearates worked as well, but caused additional  spectral lines to appear depending on the metal.  If you wanted to get really specialized you could probably coat the Mg in the stearate of a metal corresponding to the color you're going for although that seems like a lot of work for what will amount to an equivalent of maybe 1% additional oxide / carbonate / sulfate and they didn't explicitly test degradation with other metals, only spectral lines. 

Depending on the intended effect of the composition formula ratios will probably need to be adjusted if the same speed is to be maintained;  I'd imagine that strobes with low Hz strobing rates could be rendered unreliable, and the oxidizer and or Mg content (or that of a secondary fuel or oxidizer) may need to go up so unfortunately it probably won't function as a drop-in replacement except in mixes that are too hot already. 

This seems like a good balance to me.  Your conditions are unlikely to involve humidity and temperatures that high unless you live in Florida

I've also attached Shimizu's 3 part coating method for Mg which can be used if you need stars or rockets to still function 8 years later, this still uses dichromate (ammonium is recommended, I believe others can be used as well) but is effectively permanent and the other passivation chemicals, (guanidine nitrate, and any of the commonly or less commonly used sulfates) are probably already around your storeroom...  maybe not the guanidine nitrate but it isn't very expensive.

 

Effect of magnesium stearate coating on degradation of Mg with NH4ClO4.pdf Coating Mg for use with NH4ClO4 - Shimizu's improved and long-term stable Dichromate method.pdf

Edited October 13 by PyroGnome

[Mumbles]

That seems like a pretty high amount of stabilizer at ~16wt% in the Mg. For something like flake aluminum, the end stearin content is quite a bit lower, around 1-3wt% I believe. When coating iron or steel granules/turnings with wax or something similar, I believe ~3-5wt% is common.  It's a shame they didn't look at the loading to see if there is a lower value that is still effective, or perhaps also inhibits the burn rate less aggressively.  Water is also kind of a strange "solvent" to use for the coating considering most of the tested additives are insoluble in water.  Not that magnesium stearate has much solubility in anything else common, outside of acid.  Anyone cleaning soap scum out of a shower may be familiar with this.     

[PyroGnome]

Don't I know it with the stearate / water thing.  The water around here is so hard at certain times of the year you could hammer nails with it.  I made a weak bleach solution in the middle of the summer one time and the dissolved limestone crashed out into a nearly opaque suspension within about 8 hours.  Soap doesn't really lather at first, it just reacts with the water.     It has the unholy habit of shredding those melamine "magic eraser" pads that work great on nearly everything else, too. 

I'm guessing the high loading is responsible for the burn rate drops.  The layer of stearate is probably pretty thick.  Since they were only examining AP protection it's possible lower amounts were tried and not documented based on higher corrosion rates since it looked like they were shooting for something on the level of dichromate.  If you only need things to be ok for rockets you're pressing and firing that day it could probably be  lowered further without much change.

Or it might have been similar protocol to the dichromate coating technique where the common one uses far too much dichromate and never removes the excess, it's just that dichromate has an insane catalytic effect on ammonium perchlorate combustion (even compared to regular perchlorates) so it isn't really an issue.  I've noticed that most of the commercial chromate coating solutions were used at concentrations on the order of <20 grams per liter of water, not an ultra-saturated mess.  They were mostly used on plate magnesium so the surface area was less, but they could still treat a lot of it before needing to be replenished.   

I should probably point out that Shimizu's 3-component method doesn't require heating anything, either, and although he didn't rinse it off after drying it could be rinsed off.  Swapping out ammonium molybdate or tungstate should work as well, although tungsten is getting up there in weight to be a coating material for something as light as Mg.  Various references mentioned permangenates and vanadates as well, I think vanadium is just generally too expensive but permangenate seems to have seen common use.  I couldn't find a procedure for any of those but the molybdate which was basically the same process as chromium with ammonium heptamolybdate tetrahydrate (and another with ammonium phosphomolybdate hexahydrate which is apparently superior) but they should look incredibly similar to the chromate procedure aside from that.  Ammonium molybdate costs roughly 10x as much as dichromate (which isn't awful considering the amount actually required) but I'm going to hold off buying any until I find something more concrete or rule out the apatite coatings which are cheap and easy to work with, and apparently self-healing.

The wikipedia page on chromate coating actually explains what's happening fairly well.  https://en.wikipedia.org/wiki/Chromate_conversion_coating#Aluminium_and_its_alloys

It looks like industrially in situations when it the water impurity doesn't matter and bi-metal interfaces between pipes are involved they just dump 100-200ppm of sodium molybdate in the water supply and are done with it.  Apparently the plain molybdate coating (and probably the plain chromate coating) normally has an issue of cracks in the passivation layer which limit effectiveness in normal situations but since they're keeping some in solution it doesn't matter in practice. 

The experimental state-of-the-art is apparently bi or tri-layer coatings with some chemical passivation, a sol-gel thin film coating where doped mixtures of silanes (many are based on  tetraethoxysilane (TEOS)) hydrolyzed and condensed in solution bond to either the passivation layer or MgOH on the surface of the coating and are polymerized with acid or base, and possibly another layer coating that.  These guys are exploring things usable on medical implants so it gets a bit more strict.  Most of these are very easy room temperature baths / reactions, but the chemicals aside from possibly TEOS might not be cheap or easy to work with.  I saw at least a couple that needed different nanomaterials as dopants;  the only nanomaterial in any use in pyrotechnics is Cab-O-Sil M5 which might look like snowflakes (if you've ever bothered with it) but that's just the large scale structure that the 1-20nm pieces of SiO2 self-assemble into (it starts doing this mid-air while still molten during the fuming process or it'd be smaller than that, it's actually kinda interesting, the manufacturer site has an outline of it) since it attracts itself pretty strongly.  That stuff is probably about as healthy to inhale as asbestos.    Anyway, if you want to read one of the largest research review papers ever published with a ton of references to the listed procedures, here you go https://pmc.ncbi.nlm.nih.gov/articles/PMC10055842/

Here's something else that's much nicer than the chromium alternative and meant for medical bone implants so likely to be resistant to fairly harsh environments.  Apparently calcium/zinc hydroxyapatite are already in use, but strontium is biocompatible in small amounts.  The coating process is cheap and everyone likely already has the strontium source (Sr(NO3)2) required. 

Quote

High purity (>99.9%) Mg specimens (640 ppm Fe) with dimensions 10x10x5 mm were ground to 1200 grit finish and used as substrate materials. Prior to the coating process, Mg blocks were ultrasonicated in acetone at room temperature for 15 min, followed by absolute ethanol for 10 min and then rinsed with deionized water. In  this study, SrP coatings were only applied to pure Mg and, as a result, coating pre-treatment, which is critical in the case of (any) conversion coating for Mg alloys,  was not discussed [44,45]. A coating solution contained 0.1 M Sr(NO3)2 and 0.06 M NH4H2PO4 and pH was adjusted to 3.0 by HNO3. A one-step conversion coating process was conducted by immersing Mg specimens in the coating solution at various temperatures from, 40 to 80 °C.  Coated specimens are assigned names  according to the coating temperature (i.e. Mg40C was coated at 40 °C); while uncoated Mg is used as the ‘‘control’’. Coated Mg samples were dried in air and kept in a desiccator for further characterization. All chemicals used in this work are analytical grade from Sigma–Aldrich (Sydney, Australia). 
 

https://pubmed.ncbi.nlm.nih.gov/24291328/

But the fulltext paper unfortunately isn't public and I couldn't locate a copy that was.

That's it, the worst thing in the process is nitric acid and it can probably be purchased / used at very low concentration to get the pH down to 3 here if you don't want a jug of 70% laying around.   They found it effective but I only have the first couple of pages of the PDF.  Note the mention of Mg alloys requiring a pre-treatment as well.  For 50/50 Mg/Al this probably isn't the case since it retains the acid reactivity of Mg and a near-identical process can be used on aluminum with various other alkalai earths.