Ferric Chloride with a little Copper Chloride dissolved in it (like after use as a copper etchant) works like crazy on Aluminum. You might want to dilute it to keep things from overheating (literally). If you know anyone who etches circuit boards, they might have some old etchant laying around.
If you have ever used the Ferric Chloride to etch copper or copper-containing materials like brass, then it has the Copper Chloride already in it. It doesn't take much of it to make the reaction go on aluminum. I would think anything you mix with the Ferric Chloride that won't be consumed by it might work to thicken it. But, I'm not coming up with something specific. Maybe some sand or clay would make it into a paste.
In other words, used PC board etchant is better than fresh for the purpose.
[ ... ]
Before you consider doing this, remember -- Ferric Chloride when used to etch printed circuit boards *must* be agitated. That which is close to the copper gets exhausted, and you need to bring fresh etchant into the vicinity to keep the reaction going. This means liquid, and moving the board up and down frequently -- or standing the board on end in a beaker with a magnetic stirrer doing its thing. Or for serious work, with a pump recycling the etchant, and it hitting the board from spray nozzles. Double-sided boards are best done on end, with two bars of fan nozzles (all plastic, *not* most metals) oscillating up and down. (The spray etcher which I used at work had metal only in the screws acting as bearings between the links and the spray bars. And those had to be titanium, as nothing else would survive the environment. (I guess that ceramic would work nicely, but the vendor apparently didn't think of that idea. :-)
Now -- I don't know how deep you want to etch the aluminum, but you should bear in mind the etchant getting exhausted.
And (as you have given no details), are you perhaps planning to etch *anodized* aluminum, to make front panels or nameplates? For that, the etchant which I used in the past was sodium hydroxide solution. (Note that this only works with the resists like KPR (Kodak Photo Resist, which I think is no longer made by Kodak), which is developed by
1,1,1 Trichlor, *not* for the positive photoresists like that made by Shipley, which uses an alkali as a developer, and thus can't stand up to the KOH used as an anodized aluminum etchant.
[ ... ]
But -- you'll probably have to put it on, wait, scrape it off, put on more, wait, scrape that off -- a lot of that sort of work. If the KOH solution (lye) will work, you can use nylon tweezers holding cotton swabs to apply it.
The action with aluminum and the spent FeCl + CuCl solution can be ferocious, with massive foaming and extreme heating! That's why I suggested dilution. I would think you could dribble some spent etchant on a damp sponge and just wipe it on the aluminum and get some decent etching action. The action of the spent etchant on aluminum seems to be about 1000 times faster and VERY exothermic, compared to the action on copper, which is quite slow, even with warmed solution and agitation.
I learned the hard way that nylon screws fall apart after 5 minutes of exposure to FeCl! My ething machine started disassembling itself, with nylon screw heads popping off, more leaks, more screws popping! That was a big surprise to me.