how does the extraction system?

How about making the aluminum super hot, it has a poling point of 2400 so could be an option

Aluminium with silicon would be an option, i think, tesla uses it for there mega castings for warp prevention. Also why not make a vacuum and skip the inert gas. Does the aluminium not stick when not heating the print plate?

Have you thought about using a really good insulator, like aerogel, in between the head and bed?

You could try using alumina capillary tubes as the nozzle. They are only a couple $ on aliexpress and have a VERY high melting point. That should solve any aluminum corrosion as well.

I would try to mount the feed mechanism as far away as possible and push the wire through the feed tube into the nozzle from a distance like a mig welder does. Also you will most likely need an argon or co2 filled chamber around the build zone or layers will be brittle.

Heat shielding is probably your best friend for this. There are reflective option up to a certain temp but Alumina fiber is also amazing at stopping heat transfer.

Back in the 1990s I worked on a project to direct 3D print metal using metal droplets.  Here are a few things to keep in mind.

1. Molten aluminum is an aggressive solvent - it dissolves stainless steel.  Not sure how your molybdenum rod orifice will hold up long term.  I doubt titanium or inconel will hold up long term.

2. Tin rod/wire should be available at any welding store and is easy to work with.  I think 97%  tin/3% antimony is the common nearly pure tin alloy available (pure tin deforms as it changes structure in normal temperature ranges, called "tin pest").  Tin/Silver is used for soldering/brazing copper water pipes (stronger than plain tin and non-toxic - unlike lead).

3.  Avoid Zinc!  It has a high vapor pressure, meaning that molten zinc has significant evaporation - it tends to accumulate on nearby surfaces (like your nozzle), possibly as zinc oxide.   I tried to melt a chunk of brass once in a small muffin furnace; I didn't succeed, but the surface became shiny copper (I concluded that the zinc evaporated off the surface).  I was probably trickling argon into the furnace to avoid oxidation.

***Do NOT melt zinc in an area with poor ventilation.*** Exposure will likely result in zinc chills/metal fume fever https://en.wikipedia.org/wiki/Metal_fume_fever.

4. The high thermal expansion coefficient of aluminum means it will warp and any finished product would likely have high internal stresses.  Annealing might help relax internal stresses.

5. For our orifice, we used a sapphire (single crystal alumina) orifice made for water jet cutting.  We had a custom orifice chamber manufactured from alumina.  The green alumina was shaped and the sapphire orifice was inserted in the end.  Then they were fired to cure the alumina.

6.  Without an inert atmosphere, aluminum will grab oxygen out of the air forming alumina layers/inclusions.  That may or may not be an issue for this project.

7.  For an inert gas, we used argon, available from any welding supply.

Could you use sodium silicate and cast/coat an orifice?

Nice one!

I recommend trying to find aluminun/zinc  alloys, they're usually used for brazing, so, low melting point. Dunno if there are sources of it on wire form, I just have seen rods.

For the coil, I suggest using capilary copper tube, the kind used in refrigeration, so you can run water in it while it's running.

Well, I haven't read anything about ZVS besides that it make coils melt metal inside of them, but if water is a problem (let's say the ZVS uses high voltage, or mains, or anything alike), you can run non conductive coolant, like oil. Suboptimal and messy, but will do the trick.

ZVS for heating uses low voltage and high current, almost all bigger ones use copper tube with water for cooling. Good idea on filler rods.

Had been thinking of possible low melting temperature alloys with Bismuth and Aluminum a while ago but gave up because metallurgy is a quite demanding topic.

So just bringing the nozzle to the melting point of pure aluminum is a very pragmatic approach of solving the problem. However there is a plethora of aluminum alloys available and maybe you have to try the one or other to find one that is easier to extrude.

Pure aluminum has it's melting point at 660°C while silicon alloys have lower melting points. One problem is the oxide layer, which forms very quickly and has melting point around 2000°C.. maybe you will have to add an inert gas nozzle ;-)

Finally someone used the idea I had. I would try steel welding wire, ceramic nozzle and smaller diameter coil, but this version is very intriguing too. Good job!

This is interesting, I think it's the first time I've seen metal FDM. Most of the time we see modified MIG welders in this space.

Looking forward to more developments. Keep us posted.