Borax and Cryolite flux for MLCC's the best?

[4metals]

Irons is right, as he always seems to be, don't let that Gabby Hayes look fool you!

Milling the material in with the flux is the best way to assure uniform mixing. Mix it all with your standard flux mixture in the correct proportions and do a test melt.

 

[Irons]

Irons is right, as he always seems to be, don't let that Gabby Hayes look fool you!

Milling the material in with the flux is the best way to assure uniform mixing. Mix it all with your standard flux mixture in the correct proportions and do a test melt.

Dag Nabbit. :mrgreen:

 

[samuel-a]

4metals
You are correct, by 'dec' i indeed meant decahydrate, and i do agree anhydrous is preferable.

-------------

However, with respect... i disagree with the suggestion of adding soda ash and the result is somewhat evidente by the picture uploaded by Kevin. From my experianc, the slag does not seem like it should - i.e. glassy and thin without shot embedded in it.
There could be number of reasons other than the flux mixture of course; temp', time in furnace, mixing, metallic constituents.

Kevin, You mentioned you included resistors, those are composed mainly from an alumina substrate and have some organic encapsulation. This can aslo lead to troublesome smelting.

I would encourage you to first incinerate (if resistors are included) and try and replace the 1 part soda ash with silica or 50:50 Li2CO3:SiO2, rise the smelting temp' , have a longer smelt time. Then pour and obseve the result.

My 2 cents.

 

[Lou]

Kevin needs to be pouring into a conical mold.

Fluorides are best for alumina/zirconia/titanates; carbonate works well for silicaceous materials.
SiC seems to handle fluoride baths OK.

 

[4metals]

I think a 60 gram sample is too small to base your formulations on in the first place. Second for what he is trying to accomplish the melt was not long enough for the cryolite to effectively do its job. Cryolite melts at a high temperature and that is where it does its work on the substrate ceramic.

A cone mold is always preferred for test melts and slag remelts but for melting the actual lot, molds are required. That is because I have seen countless times a refiner pour a cone mold (a classic conical cone mold) too full and end up with a bar that does not fit into any crucibles. Whoops!

I prefer cone molds made in a conical form over the vee shaped bar molds, the conical shape forces the molten metal down from all sides and effects better collection, especially when it is a small metallic fraction to begin with. For years I used a square mold tipped on end to form a vee and had results like shown in the picture, (beads not in one metal mass but un-joined) then I switched to a true cone mold (conical) and the same flux scenarios resulted in solid uniform beads.

Silica alone, or in combination with only borax, is not a suitable flux because of the high melting point of the silica. Mixing silica 1:1 with soda ash will lower the viscosity and the melting temperature. If you want to try Samuel's suggestion, use a 50:50 mix of silica and soda ash in lieu of straight silica.

 

[Irons]

Use Acidic Flux to attack Basic substrates and vice versa.

Alumina= basic
Silica= acidic.

 

[4metals]

Well, mixing the acid silica with the basic soda ash yields a neutral flux component.

The borax will make the melt acidic and the borax will function to liquify the melt and lower the slag formation temperature. With high melting components like cryolite and alumina that really makes a difference.

Cryolite is a neutral flux. So you will still end up with an acidic flux. The alumina acts as a basic component. The theory here is to add enough alumina material to keep the end pH alkaline although measurement is all but impossible. This is because an alkaline slag will reduce the temperature required for complete liquidity.

Unlike assay fusions where a flux is measured by its reducing power on the lead we want to avoid lead as a collector or any other metal additions if possible.

If you notice a scum like gangue floating on the melt that indicates you have not burned the material completely so additions of sodium nitrate will burn that away for you by generously donating its oxygen to the combustion. This again is something you don't want to go overboard with. Better to burn completely up front. I have melted material that I could never seem to get rid of the scummy layer in pretreatment by adding 5% sodium nitrate to the mix. It behaves better as a component in the flux than it does being tossed on top.

As you can imagine it is important to learn the behavior of individual flux components and how they react with your blend. For this reason processing large homogeneous blends where you can mess with the flux components on a sample basis and get it right for the production run is critical.

Unfortunately this kind of refining comes from more "been there done that" than reading. I cant tell you how many times I've singed eyebrows, burned off my arm hairs, and melted my shoe soles just to get a sample of the molten slag.

You have to arm yourself with the knowledge of the make up of what you are melting and apply the knowledge of what different flux components do and work out some test scenarios. After a while it gets easier. (and the hair grows back too!)

 

[kjavanb123]

All,

Thanks so much for your valueble responses. I got them after I remelted the material from my yesterday smelting, and this time I cooked it for an hour and swirled it few times during that time. Poured the melt into angeled mold and it looks like this,


And closer look reveals that tiny beads had time to consolidate into this,


There was a hole in crucible right in the middle of my second batch, thankfully the level of melt was below it, so nothing was loss, I am going to use your advises tommorow.

Regards,
Kevin

 

[4metals]

One of these would work wonders now.


And when you really get going, these are for slag remelt day.


Both available from Legend

Years back, when I had to slag off during a melt, I used the big cone mold and made a slight inclined ramp about 6" high and tapering out about 6 feet. Then after slagging off I poured the molten slag down the ramp where it cooled. Often a small puck of metal was at the bottom, it hardens long before the slag, that was thrown back into the melt. By pouring the slag on to the ramp it was much thinner and a hammer would break it into manageable pieces. The last thing you want to work with is huge cones of solid slag from the big mold.

I was melting in a small Cutter rotary furnace but its capacity was huge, The refractory lining (no crucible) was 3 feet in diameter and 6 feet deep. We had to pour out 5 o 6 cones to empty half of the slag. I probably could have stopped the rotation before slagging off and avoided any metal pour off but as it behaved nicely and collected in an easily recovered puck, stopping while in the process wasn't necessary.

 

[4metals]

I went hunting in my files to show you what a rotary tilt furnace looks like. Unfortunately these were from before the days when everyone had a camera in their phone so regretfully, I have no photos.





The unit operated in a horizontal position and it rotated slowly at about 1/2 a revolution per minute. The flame shot right down the center and the rear had a charging door which could be opened while in operation to add material. The unit tilted down to pour from the nose.

The entire unit was lined with refractory brick and mortar so care was always taken to assure a flux mixture with the least effect on the refractory. Even then we had to reline once a year. But that was melting every day, 5 days a week for 10 hours. I always blended and fluxed enough material to assure I could process for 10 days on one flux formulation.

In the assay lab, my favorite hang out, I tried different flux proportions and blends on the mix to see which had the lease effect on the refractory while doing the required reduction and pooling of the values. I did this by having thin slices of the refractory brick cut which looked like tongue depressors in both size and thickness. These were added to the fusions and relative comparisons of refractory wear were made to choose the best result with the least effect on the lining.

Because of the way this thing worked, with the flame shooting towards the opening in the front, we had troubles with fines blowing right out the front of the furnace. We overcame this with a briquette making press which pressed out flux sweep mix into pellets 8" around and a foot long. They melted as quickly as the powder and solved the dust issue.

This was one hot furnace to run. It required the operator to suit up in the "tin man suit".

 

[Irons]

I bet that furnace wasn't very efficient from a fuel consumption standpoint. That was back in the Day when gas was 20 cents a Gallon.

 

[4metals]

The newer designs have the pour out hole in the center so the heat can stay in a bit longer and be more efficient.



But gold was under $300 back then too!

 

[FrugalRefiner]

4metals,

Thanks for sharing that glimpse into the world of the commercial refiner.

Dave

 

[Irons]

The newer designs have the pour out hole in the center so the heat can stay in a bit longer and be more efficient.



But gold was under $300 back then too!

I like that Lazy Susan for the Cone Molds.

 

[4metals]

If you notice that Cone Mold Lazy Susan has hinged cone molds that flip up to a stop so the contents can slide out and it flips back down for a refill. Each of those molds holds about 5 gallons of slag.

Unlike this run of the mill boring Bar Mold Lazy Susan. (With pretty blue mold release)

 

[Irons]

That's still above my Pay Grade. :mrgreen:

OK, I see how that setup works now. Thanks for the explanation. :idea:

 

[kjavanb123]

All,

I dissolved the metal alloy produced from smelting mixture of SMD capacitors, resistors and few Ta capacitors, in dilute nitric acid, the solution has turned to green color as can be seen in the qtip, stannous chloride test showed a little of yellow color which could be Pd, and some undissolved residue that could it be Ru? Or Ta? Further testing will be conducted, as I read Ru is insoluble in nitric acid.

Here is the residue left from filtering the nitric solution,


And qtip result after adding stannous chloride to green solution,


Will post further updates soon.

Thanks
Kevin

 

[4metals]

Refining the alloy you produced in the smelting process is a totally different topic than this thread originally addressed. I suggest you will get more input by reserving this thread to follow up on the smelting process and open a new thread for refining the smelted alloys.

That will make future searches easier and more informative.

 

[924T]

Samuel-a,

I forgot to ask you how you measured the temperature of your gas furnace?

Type K thermocouple? Laser thermometer? Other?

Were you using the same furnace with the Kaowool that you recently posted a video of?

Cheers,

Mike

 

[924T]

Irons,

My budget for acquiring lab glass is temporarily depleted, so I'm back to the flux approach, and
will be ordering my first 5 lb. bag of Cryolite from Sheffield tomorrow.

I'm not sure there is an affordable graphite crucible for running a 1kg batch of ceramic capacitors,
since it would become 2 kg. with the flux mixture added, which would require quite a large crucible.

I feel like a just got a college education, reading the exchange between you and 4metals, about smelting.
Many thanks to the both of you!

It sounds like I might be better off utilizing a medium-sized crucible, and periodically pour off the
slag, and then add more capacitor/flux mixture to the melt.

If you have a favorite brand/type of graphite crucible, I'd sure like to hear about it.

Cheers,

Mike