DIY Molten Aluminum Air Atomization (pics of powder!)

[StevenRS11]

So, I was just messing around the other day with two straws and a cup of water. You know how you can blow air out of one straw over the other, and it sucks up water and sprays it out as a mist? Well, turns out you can do the same thing with molten aluminum.

I took a section of carbon gouging rod and drilled a small hole through it, making the "straw". In the picture, you can see the little bit of carving I did on the top to ensure there is a low pressure zone over the hole. Blowing compressed air back down into the molten aluminum would be bad for obvious reasons. For the compressed air, I just took some copper tubing and crimped it into a nozzle shape.

Using some sheet metal, I suspended the graphite tube in a crucible of molten aluminum and let it heat up. To my incredible surprise, when I blew the jet of air over the tube, it sucked the metal up and blew it out as a fine mist. I captured this in a metal watering can (just had it on hand) and it looks like the mist solidified in about 1 foot of space. Everything closer was coated in a layer of deposited aluminum spatter, which was surprisingly solid.

Unfortunately, I didn't make a video of this in action because I didn't expect it to work the first time, but here are some pictures of the aftermath.

https://flic.kr/p/Wr3aSz

https://flic.kr/p/Wr3aRn

https://flic.kr/p/V9FimY

This was done using my smallest furnace with a mullite crucible. These crucibles are disposable and this one broke when I dropped it trying to get the cold slag out . All said and done, it took an hour to set it all up, 5 min to melt, and 30 seconds to turn about half a pound of solid aluminum into powder. The powder is a mix of sizes, some as large as 60 mesh with others finer than 325. Average looks to be around 160 ish. Personally, I am going to use this for thermite so it's good as is, but it would be easy to ball mill at this point.

Now, I do have a fair bit of experience with molten metals and their dangers, so I can't recommend this process to people who have never worked with them before. That said, it is relatively safe because you are not having to move or pour a large amount of metal- when most accidents happen.

Edited July 5, 2017 by StevenRS11

[Arthur]

Now if you could borrow the feed mechanism from a MIG welder to keep the liquid level up.

[MrB]

Now if you could borrow the feed mechanism from a MIG welder to keep the liquid level up.

Just use a bigger crucible.

The venturi should be pretty easy to work with, until it can pull molten aluminum at least 4 inches straight up. If you stay on top of the system, you could stay well below the oxide layer, and possibly your flux of choice, and just add a little bit all the time.

[StevenRS11]

Just use a bigger crucible.

The venturi should be pretty easy to work with, until it can pull molten aluminum at least 4 inches straight up. If you stay on top of the system, you could stay well below the oxide layer, and possibly your flux of choice, and just add a little bit all the time.

 

The venturi is capable of pulling the aluminum up at least 8 inches. I didn't have a way to drill a hole any deeper, heh. I plan on setting the same thing up in my #12 crucible, and it holds around 15 pounds of aluminum. I am also going to modify the nozzle setup a bit, with multiple jets to provide better atomization. The capture method needs work too- I am thinking an L shaped length of rigid HVAC ducting. I might need some sort of cyclonic dust separation system if I get lots of fines blowing around, but honestly that's a good problem.

 

Hopefully I can manage to get at least 50% +325 mesh. Working out the costs, it should be pretty cheap too. I have several hundred pounds of heavy gauge Al wire scrap (no insulation), and my large furnace takes about 15 min to melt a charge. Aluminum scrap goes for like 50 cents a pound, add in the 5 min it would take to atomize the whole load and that's about 1/3 of a propane tank, so what, 10 bucks total? I can live with that.

 

Now, I wonder if I can do the same thing with cast iron and oxygen to make Fe2O3? That would be a firework show in and of itself- I can't even imagine what burning over 1/8 of a pound of iron per second would look like.

Edited July 5, 2017 by StevenRS11

[Arthur]

Possibly consider an inert gas! Nitrogen CO2 or Argon will be cheaper than a big fire.

[MrB]

Now, I wonder if I can do the same thing with cast iron and oxygen to make Fe2O3? That would be a firework show in and of itself- I can't even imagine what burning over 1/8 of a pound of iron per second would look like.

 

Start small, and set it up to be remote triggered. I suspect molten iron and oxygen (pure) to be a fairly violent reaction. There might be an issue with temperatures preventing iron oxides from being reliably formed as well (To high temperature makes it release the oxygen and revert to iron. Well, i think.)

[OldMarine]

Sounds like a mini Bessemer process. Might just get steel!

[pyrokid]

I am very interested in this process! Powder generation was the one element of home metal production that I could never achieve! Some comments:

 

1) The product doesn't seem to contain spherical or spheroidal particles. I wonder if the geometry of the atomizer could be varied to control particle shape. Maybe ramping up the pressure would break them apart more.

2) I wonder about the hazard of ignition when using standard air, especially if magnesium is added to the melt.

3) Could the droplets be blown into a water tray for collection?

4) Could a water jet be used in conjunction with this method to more effectively break apart the droplets?

[MrB]

4) Could a water jet be used in conjunction with this method to more effectively break apart the droplets?

 

Sure. Ongoing steam explosions sounds like fun

Actually, it should be fine, as long as one takes the massive increase in volume in to account. I suppose you could skip the air all together, and just go with a water-based venturi, but working it out so that the metal doesn't solidify in the raiser pipe... I'm thinking ceramics, for the isolating properties.

[Arthur]

Hopefully the OP is better versed in the risks and hazards of molten metal than some others, and has sufficient space, separation and safety equipment.

[StevenRS11]

Ok, quick update- I added two jets on either side of the graphite tube that come together at a shallow angle, intersecting about 5mm into the molten aluminum spray. Based off of the patents I found, the particle size should decrease with increasing velocity of these jets, so I attempted to crimp the copper tubing into a shape resembling a de laval nozzle. I am certain that I have choked flow, and I *think* I am getting supersonic flow with 120 psi. Either way, I learned that I was just messing around earlier. This is completely different- the resulting atomized powder is dust. I did not succeed in capturing any meaningful quantity of it, so a cyclonic dust separator is necessary.

 

Hopefully the OP is better versed in the risks and hazards of molten metal than some others, and has sufficient space, separation and safety equipment.

 

While I am not a professional by any means, I have taught several metalcasting classes at local highschools and have all the necessary PPE. Granted, this is a rather unusual thing to be doing with molten metal, and I appreciate the concern.

[pyrokid]

This is completely different- the resulting atomized powder is dust. I did not succeed in capturing any meaningful quantity of it, so a cyclonic dust separator is necessary.

 

What did you mean by this? Did the dust blow away?

 

Please keep us updated. This is intriguing!

[StevenRS11]

Yea, the dust just blew away. Party because I was using an order of magnitude more compressed air, and partly because the atomized aluminum was so much finer. Don't forget I was blowing this into a watering can lying on its side:). Not the ideal capture device. I only ran that last test for a few tenths of a second because it was just blowing powder everywhere. I'd rather not have an aluminum powder fuel air explosion on my hands.

 

I have one of those dust deputy cyclonic things for my shop, and the thing is amazing. I'll just build one out of some sheet metal and use that to collect the aluminum if a long section of HVAC ducting doesn't work. Either way, I'm not going to do this again until I have a collection system that works with a handful of bread flower. I really think that, at least for someone who needs a whole lot of Al powder, this is a pretty practical method.

 

I am out of town for 8 days, but when I get back in I'll post some videos of the whole process.

[StevenRS11]

Ok, back in town and back to experimenting. I'm about to run it again, this time with a much less cobbled together setup.

 

We have-

 

2 pounds of clean, electrical grade aluminum

a modified venturi siphon nozzle based off a patent for spray coating things with molten metal

2 supersonic jets of air for additional atomization based off of the "air knife" excavation tool

HVAC ducting to capture the powder

 

The venturi is running at 80 PSI and the jet's at 180 PSI, aluminum at 1300F.

 

Apparently, magnesium significantly decreases the surface tension of molten aluminum. That *should* give finer particles, but I am worried about excessive oxidation during atomization. Ill try that next time. The supersonic jets are really cool too; when you get them working, the converging section of the nozzle actually gets warm.

[pyrokid]

My fabrication skills are quite limited but I am intent on reproducing your results!

 

1) Can you please provide a link to the spray coating patent? What is the angle of the primary jet relative to the gouging rod?

2) How exactly did you modify the top of the gouging rod to ensure a zone of low pressure?

3) Does the collection system consist of a length of straight HVAC tubing or is there an elbow at the end that points to a collection tray? I'm curious about the need and practicality of implementing a commercial dust collector. Harbor Freight offers one for $200 that claims to filter everything 5 microns and larger. The obvious concerns are longevity and the temperature of the particles being fed into the collector.

4) What is the state of the gouging rod after the procedure is finished? Are you left with a core of solidified aluminum? Can this be remelted during the next trial or does the tube need to be remade?

[StevenRS11]

My fabrication skills are quite limited but I am intent on reproducing your results!

 

1) Can you please provide a link to the spray coating patent? What is the angle of the primary jet relative to the gouging rod?

2) How exactly did you modify the top of the gouging rod to ensure a zone of low pressure?

3) Does the collection system consist of a length of straight HVAC tubing or is there an elbow at the end that points to a collection tray? I'm curious about the need and practicality of implementing a commercial dust collector. Harbor Freight offers one for $200 that claims to filter everything 5 microns and larger. The obvious concerns are longevity and the temperature of the particles being fed into the collector.

4) What is the state of the gouging rod after the procedure is finished? Are you left with a core of solidified aluminum? Can this be remelted during the next trial or does the tube need to be remade?

 

The only tricky part is getting some molten aluminum to play with, really. Past that, it's easier than making rcandy rockets CATO every. single. time.

 

Patent link- https://www.google.com/patents/US3776462

 

By the way, the thing described in this patent is insane. It's essentially a spray paint can, except it sprays molten metal. From your hand.

 

The primary jet is angled at 90 degrees, but I filed a groove on the top of the rod at a 15 degree angle. This simply makes it more difficult for me to blow air back down the rod into the molten aluminum. On the back side of the rod, I cut away material so molten aluminum would not accumulate and solidify, blocking the jet. A traditional venturi siphon is completely enclosed, but I felt this could clog and blow air back down. The top of the rod looks sort of like the bottom half a venturi siphon, I guess. I played around with water until I got a profile that did what I wanted it to do.

 

Right now, my collection system is a horizontal 4 foot length of 7 inch diameter HVAC ducting followed by a tee and a 4 foot vertical section. The 'collection pan' is just the capped off bottom of the tee. I plan on bending a sheet of aluminum flashing into a cone and placing it in the top, with air entering tangent to the side of the cone at the widest point through multiple slots. The fines that didn't collect in the ducting will fall out the bottom of the cone into a jar or something.Finally, air exits though a 2 inch diameter opening in the top of the cone. No vacuum source needed, the compressed air is what is driving it all. Ill post a picture of that too, but it looks like this- https://www.amazon.com/dp/B002JP315K?psc=1&smid=A2ZANZX0RHL3OT. I actually have one of these things for my CNC router at work, and it is amazing. Sawdust, metal shavings, the odd paper towel, MDF dust, anything.

 

The gouging rod is fine, the aluminum doesn't stick to it at all. If you run it to completion, you end up siphoning all the aluminum out anyway. Also, I preheat the rod with a torch for a few seconds before I turn the air on. The carbon does erode a bit, and I imagine one rod would be good for 2 or 3 fifteen pound runs before it needs to be replaced.

 

This is another good resource- http://engj.org/index.php/ej/article/download/1054/527/

 

It talks about the impinging jets that actually do the atomization work. The primary jet in my design is really just an easy way to pump the aluminum out of the crucible. While it does produce a powder like in my first test, the powder produced by the supersonic jets feels like commercial 32 micron atomized. Plus, I am still loosing the finest fraction until I build the cyclonic thing. Honestly, you could run it just fine with only the primary jet and the HVAC tubing. It's the secondary jets that really blow things around- they consume a very, very large amount of compressed air.

 

Hopefully I'll get time to run it all tmrw and get a good video.

Edited July 19, 2017 by StevenRS11

[AllisterF]

Very cool, concept, can't say I would bother but it's always fun to explore...

 

My brother and I have been doing 'backyard' metal melting and casting for nearly two decades as a hobby, we started out with a small home built charcoal foundry for aluminum, then moved on to waste oil and propane based ones to get cleaner and higher temps... We set out with a goal to see if we could achieve melting and casting cast iron, but it proves to be quite a challenge to get to those temps for the DIYer in the backyard... Never did get reliable cast iron melting, but we can readily melt and cast copper and bronze so we are real close to the temps needed for cast iron...

 

Also had the pleasure of creating several mini lava flows in the backyard as the lower heat cements/concretes have liquefied as we pushed the limits of some of our earlier foundries... And since there was a mention of safety, let me state that when we were pushing the limits trying to see how much the foundries could handle and ended up creating our very own 'lava flows', several safety precautions were taken, as we were aware that the cements/concretes were likely going to fail before the cast iron gave in...

 

 

The only tricky part is getting some molten aluminum to play with, really. Past that, it's easier than making rcandy rockets CATO every. single. time.

 

 

For the mechanically inclined that isn't even that hard or tricky, my first aluminum melting foundry was built from a scrap disposable 'party' helium tank, lined with Structo-Lite plaster (50lb bags available at many hardware stores for about $12) a section of iron pipe, a used hair dryer, some duct tape and was fueled by common BBQ charcoal... I used an empty steel soup can for the crucible initially, before upgrading to a stainless steel can then to a more proper graphite crucible... It wasn't an eloquent design but it melted soda cans as fast as I could toss them in, and cost less then $20 to make... I did a lot of lost foam sand casting with that little foundry in the early days...

[Arthur]

From some literature research a while ago, the two methods used to make fine aluminium involved air jet spaying or inpinging the flow of liquid metal onto a fast spinning ceramic disc or cone. The product is usually approximately spherical and the size distribution can be changed with the detailed operating conditions of the system.

 

Getting the definitely non spheroidal ultrafine particles -usually referred as "Dark" or "Black" is a whole further processing stage, they are not simply sieved from mixed size spherical powders.