kilowatt
Hazard to Others
Posts: 322
Registered: 11-10-2007
Location: Montana
Member Is Offline
Mood: nitric
|
|
Purification of Spongy Alloy
I am interested in the production of alloys for forging via electroreduction at 750-900°C in fused CaCl2/CaF2 of sintered mixed metal oxides (and
CaCO3 or alkali metal carbonate to impart carbon). This is essentially the FFC Cambridge process.
This way a very uniform alloy is produced, similar to those made by powder metallurgy. However the raw metal is in the form of a highly porous
sponge, whose voids are filled with electrolyte consisting of CaCl2, CaF2, CaO, and Ca metal. CaO as well as oxygen content in the alloy itself can
be minimized if the electrolysis is carried out for an extensive time running the cell to the point of chlorine evolution, but the other electrolyte
constituents still remain.
I am looking for a way to effectively leech out these materials from the spongy matrix without causing side reactions with any of the metals involved.
It would be preferable to do so with the spongy material remaining in a mass so as to more easily allow forging into a dense stock, but grinding the
material into a powder may also be workable. The preferred forging process is by heating the metal in fused salt so as to form a protective layer
even when removed into the air for manual forging. I have just begun using this process to forge knife blades from commercial stock, but would like
to extend it to spongy metals formed by oxide reduction. The issue I see with the fused salt forging is of course that the salt will tend to be
absorbed by the spongy metal during heating, and so intermittent leeching stages may be needed if it is even possible.
Most of the alloys I am interested in can be obtained commercially but only with great difficulty if at all by an individual, and at very high cost.
In contrast the oxides and other compounds of all the involved metals are either quite cheap to obtain in multi-kilogram quantities or only needed in
small amounts. In short the price of these finished alloys is many times greater than the sum of their constituents.
The alloys I am currently most concerned with contain the following elements, which will need to remain unreactive toward whatever leeching methods
will be used to clean the metal sponge.
High Speed Tool Steels containing:
Iron
Carbon
Chromium
Vanadium
Molybdenum
Manganese
Tungsten
Stellite Alloys containing:
Cobalt
Chromium
Carbon
Nickel
Iron
Silicon
Manganese
Molybdenum
Tungsten
Titanium Beta Alloys containing:
Titanium
Aluminum
Zirconium
Molybdenum
Chromium
Tin
Iron
The mind cannot decide the truth; it can only find the truth.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by kilowatt  | | I am looking for a way to effectively leech out these materials from the spongy matrix without causing side reactions with any of the metals involved.
It would be preferable to do so with the spongy material remaining in a mass so as to more easily allow forging into a dense stock, but grinding the
material into a powder may also be workable. |
I have two immediate ideas, although neither is chemical in
nature.
The first is to hammer on the sponge while under the molten calcium salt flux and mechanically drive out the impurities. Immediately after the alloy
is reduced, it's got high chemical activity, and that includes bonding to itself. As for practicalities, the way to seal an anvil under flux is to use
solid flux as the sealant. This entails a thermal gradient, necessarily, so what's done is to incorporate cooling coils. The combination of solid salt
and the cooling coils themselves should be considered to constitute the base. This is essentially the same technique used to manufacture cubic
zirconia gemstones, which, for a refractory, uses a colder wall of zirconia.
The second idea is to build yourself a vacuum induction furnace and melt the sponge. OK, that's a lot of work, but it seems much more feasible to me
lately since I've been reading up on power electronics. Of course, if you had a vacuum induction furnace, you could just do direct melts.
It would help to know what permissible level of contamination is and how much product loss you're willing to tolerate. I'd also be interesting hearing
about open-cell vs. closed-cell in your sponge, because if you've got an appreciable percentage of closed cells, leaching may not work at all.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
This 2007 article from Chemical & Engineering News is relevant: Calcium Fluoride Goes Soluble. Reading between the lines, I'm guessing there's no way (short of the research reported) of chemically leaching out
the CaF2 without attacking everything else too. On the other hand, if you're willing, you could dissolve everything but the CaF2, filter, and then
precipitate the metal out in powder form.
|
|
|
UnintentionalChaos
International Hazard
Posts: 1454
Registered: 9-12-2006
Location: Mars
Member Is Offline
Mood: Nucleophilic
|
|
| Quote: Originally posted by watson.fawkes | On the other hand, if you're willing, you could dissolve everything but the CaF2, filter, and then precipitate the metal out in powder form.
|
But the whole point of the FCC cambridge process is to reduce the metal :p If you dissolve it back out, what have you accomplished?
Department of Redundancy Department - Now with paperwork!
'In organic synthesis, we call decomposition products "crap", however this is not a IUPAC approved nomenclature.' -Nicodem
|
|
|
kilowatt
Hazard to Others
Posts: 322
Registered: 11-10-2007
Location: Montana
Member Is Offline
Mood: nitric
|
|
The whole point of using any calcium fluoride was to obtain a lower melting eutectic. Industrial FFC Cambridge cells use pure CaCl2 and so have I
previously during my experiments on the process. It is just that I will be using the CaCl2/CaF2 eutectic for other electrolysis processes in the same
cell that leave solid calcium as a cathodic biproduct and had simply wanted to use the same electrolyte for this powder reduction too. I could use
pure CaCl2 for this though if it would make it easier, but I'd still have to worry about calcium metal in there. There could be further problems
since calcium is soluble with some of the alloying agents. A potassium chloride electrolyte might be more suitable due to the volatility of potassium
and solubility of its hydroxide; I do not see why that wouldn't work.
As for melting the sponge, that rather defeats the purpose of a powder process because the molten alloy recrystallizes in a less uniform fashion than
is found in compressed powder alloys. If you look at steels like CPM 3V and CPM M4 they are significantly superior to their melt processed
counterparts.
I'm afraid I don't quite follow your idea with the submerged anvil thing, but if it involves a "skull crucible" type setup then it will be orders of
magnitude more energy intensive than anything I can handle. Usually several megawatts are required for even small systems like that. It seems
possible that much of the electrolyte could be driven out by hammering in the air and reheating in a salt bath so long as the hammering was done at a
temperature greater than the melting point of the electrolyte. I don't think it would really end up very pure though, and once compressed it could no
longer be leeched.
If there is a solvent than can remove the calcium or potassium chloride electrolyte without attacking the alloy (maybe alcohol or something),
induction heating in an inert atmosphere may be possible for forging it to full density if I cannot use a salt bath. I have been planning on making a
glove box for all manners of things like this, handling reactive metals and such. Regarding salt baths though, boric oxide or some type of molten
glass may have enough surface tension so as to not enter the sponge too much during forging. Molten chloride salts on the other hand are tenaciously
penetrating and so would likely be a poor choice. I'm not concerned about losing the outer layer of the material, so long as the bulk of it can be
obtained pure enough to exhibit its full mechanical properties.
[Edited on 26-5-2009 by kilowatt]
The mind cannot decide the truth; it can only find the truth.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by kilowatt | | I'm afraid I don't quite follow your idea with the submerged anvil thing, but if it involves a "skull crucible" type setup then it will be orders of
magnitude more energy intensive than anything I can handle. Usually several megawatts are required for even small systems like that.
|
It's the same principle, yes, but with a very reduced energy load. The basic setup would be something like
the following. An anvil sits upon a post. Around the post is a pipe, a bit larger in diameter. The pipe connects to a working tank, heated, so that
the anvil sits near the bottom of the tank. The idea is to have some mechanical isolation between the anvil and the surrounding tank. Molten flux goes
down into the pipe, where it meets cooling coils at the bottom that keep a solid plug of flux. This is similar to a skull crucible, but the total
temperature difference is a lot less (a factor of four or so), and the distance between hot and cold sinks is much greater (a couple of cm to a couple
of m, two orders of magnitude), so the temperature gradients are likewise smaller. Power consumption is proportional to these gradients. It's also
proportional to the surface area, which is another couple of orders of magnitude smaller. By that proportion, we're down (conservatively) to a few
kilowatts, or about the consumption of an electric clothes dryer.
|
|
|
markgollum
Hazard to Self
Posts: 53
Registered: 21-2-2004
Member Is Offline
Mood: No Mood
|
|
When I fused CaCl2 to use as a dessicant, I was surprised just how volatile it was. I think you should be able to vacuum distill it off the the sponge
rather simply at 900-1000 C. However, I think that you might have problems with the Calcium oxide promoting the formation of inclusions in your
finished products, it might help to rinse your sponge with cold dilute hydrochloric acid after the distillation step but the decreased reactivity of
the deadburnt lime will be working against you.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by kilowatt | | It is just that I will be using the CaCl2/CaF2 eutectic for other electrolysis processes in the same cell that leave solid calcium as a cathodic
biproduct and had simply wanted to use the same electrolyte for this powder reduction too. I could use pure CaCl2 for this though if it would make it
easier, but I'd still have to worry about calcium metal in there. |
You could wash out residual CaF2 with
plain CaCl2. CaCl2 dissolves in water. Metallic calcium dissolves in anhydrous ammonia. At least you can get your cell clean.
As for ammonia, CaCl2 is one of the few salts that's not highly soluble in it. If your alloy can tolerate ammonia, there are a number of salts to pick
from, well soluble in it. There are other polar solvents to consider, as well, such as acetic acid, sulfur dioxide, and especially, if your alloy can
tolerate it, sulfuric acid.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
It's silicates and
carbonates that are the biggest problem in dead-burnt lime, and neither of these should be present in a salt flux.
|
|
|
kilowatt
Hazard to Others
Posts: 322
Registered: 11-10-2007
Location: Montana
Member Is Offline
Mood: nitric
|
|
I have considered acetic acid but I'm not sure how the alloy would tolerate it. I believe glacial acetic acid would attack some of the metals while
aqueous anything definitely would form a passivating layer which I do not want. Sulfur dioxide would seem a good choice but pure SO2 is nasty stuff
as is anhydrous ammonia. Is there another amine I could use that would work as well as ammonia for some of these salts? I am now further considering
using KCl as the main electrolyte instead of CaCl2 for solubility reasons.
If I can get the sponge to an oxygen-free state I should only have the electrolyte chloride and metal to work with. Potassium is volatile enough that
it could be distilled out as could perhaps the chloride since it too is quite volatile at 900-1000C (I could actually hit the boiling point if I
wanted too since the alloy would not melt at 1420°C). Even calcium could possibly be removed by distillation. This could be done under vacuum (or
more realistically reduced pressure argon) in an induction furnace or simply with heating wire to make it even easier. It is starting to look like
the best answer so far. On the flip side though, some potential alloying constituents could be volatile too at such temperatures although it seems
none of those I have listed are. Even if they were, the alloy could likely be adjusted to compensate since most of the elements are only present at a
few percent.
[Edited on 26-5-2009 by kilowatt]
The mind cannot decide the truth; it can only find the truth.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by kilowatt | | [...]while aqueous anything definitely would form a passivating layer which I do not want. Sulfur dioxide would seem a good choice but pure SO2 is
nasty stuff as is anhydrous ammonia. Is there another amine I could use that would work as well as ammonia for some of these salts?
|
It's clear to me now why you're avoiding water. Any of the oxidizing acids, even mildly, would do
essentially the same thing. As for SO2, it's not high on my list, but it's fairly easy to scrub (if you have a scrubber). Ammonia is much easier to
deal with, since it dissolves so readily in water. I don't know from any direct experience whether these work. I've spent some time looking for
solubility data in these other solvents, and I've had some difficulty. For example, the IUPAC-NIST Solubility Database has no entries for anhydrous ammonia.
For other amines, there are plenty, although ethylamine and triethylamine seem decent starting points. Luckily this is a pretty easy thing to test. All you need is a few ml of candidate solvent and some
candidate salts.
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
