I see the same obstacle popping up time and time again, a member has decided on a reaction that demands high temperatures and/or pressures and needs
to obtain or construct a reactor sufficient for the task. I probably see most of these issues when members need to contain molten lithium or molten
alkaline earths.
The organized and accessible sources of information pertaining to this topic [ of reasonable concessions] are limited, because we often mimic
preparations from the literature, and they have the resources and don't have to make concessions. Therefore, you see little discussion of "alternate"
materials.
A valuable resource is called "The liquid Metals Handbook" and is a 1950 publication which I can't find for dl but here is a link to the online
version... http://babel.hathitrust.org/cgi/pt?id=uc1.b3734838;view=1up;...
wherein they discuss the relative merits of different materials of construction. Such as for Lithium, below.
I'd like to invite members to share their experiences about what worked as well as what didn't. Please try to give good descriptions, costs, sources,
results because I think this could be useful to many of us.
Also, this topic really has two sub-headings. One is for things that worked well enough for an experiment, ala proof of concept or similar. The other
is for things that flat out work well by some objective standard.
I've also attached a short discussion of Resistance to Molten Metals, a stainless selection guide, a description of ferritic SS 446, as well as
resistance, or lack thereof, to molten lithium by non-metals. Hope you find something there of use.
In the chart, solid black means OK, partial filling shows less than perfect but generally usable for short periods, the lined areas are unacceptable
and white means no data.
[Edited on 24-10-2014 by Dan Vizine]j_sum1 - 23-10-2014 at 14:37
Cool idea.
An idea I have been floating around the last 24 hours is electrodepositing a refractory metal on the inside of a graphite crucible. These are
relatively cheap on eBay (used by precious metals people) and should make an ok cathode. I anticipate it is possible to make a versatile crucible to
meet home chemistry specifications. The links you posted will give a good start point for metal selection.
J.Dan Vizine - 23-10-2014 at 20:04
Hi j_sum1.
Oh Oh...it's pretty quiet. Luckily, that's no deterrent to me going on and on....
Here's how molten Mg/MgCl2 is handled in austenitic SS like 304, 316, 347...they use a mild steel liner.
[Edited on 24-10-2014 by Dan Vizine]Dan Vizine - 24-10-2014 at 05:41
monolog? I can do that.
Yesterday I contacted a British SS firm to ask about the suggested material of construction to stand up to molten Ca/CaCl2 at 1050 C for 5 hours.
Today, I received this reply from their resident metals expert. This is in relation to my ongoing calciothermic reduction of ThO2 project.
Interestingly, SS 446 is what the original Sylvania researchers used...
"The combination of molten calcium and calcium chloride salt is a difficult combination for most materials. Molten calcium can cause cracking of RA330
and nickel based alloys. On the other hand calcium chloride molten salts in the presence of oxygen fluxes the scale eating away at the chromium oxide
coating. In these applications RA330 or 600 are alloys of choice. Since you have both, you now need a compromise, If the temperature were lower, 309
would be a good choice, but at 1050 C, you are above the oxidation limit. RA 253 MA could be a reasonable compromise, given the sublayer of silica and
RA330 is the only other potential compromise. Keep in mind that the oxidation limit of RA 253 MA is 1090, a temperature surge that takes the metal
above this temperature, could have adverse affects of RA 253 MA, in terms of breakdown of the oxide and leaching. RA 253 MA also has better creep
strength at this temperature. So the question is just how well can you control the temperature of the metal of construction. Regardless of the
material, there is not certainty that either will work, only the possibility. The 446 that you mention has not strength or ductility at this
temperature. I am attaching an article on materials for salt bath construction. Pay close attention to the comments on properly maintaining the salt
as the maintenance is the key to attaining a reasonable life."
Attachment: SALT POTS.pdf (308kB) This file has been downloaded 1106 times
I replied as follows:
Thank you for your detailed answer. I was aware that molten Ca was aggressive toward Ni based alloys, hence the worry that the 304 SS bomb that I have
may not work.
I’m attempting to duplicate work performed by Sylvania researchers 50+ years ago. They employed 446 for the bomb reactor. I was hoping a more
accessible alloy might work well enough. If I can maintain the level of leached metallic impurities below 1 % I would be happy. I plan to have the
product powder vacuum arc re-melted to give beads, which I understand will sometimes actually improve purity slightly.
-
Here are a few more details: A metal oxide (ThO2) was combined with Ca + CaCl2 in a SS 446 bomb. The bomb was heated in a furnace while being
maintained at 1050 C under an atmosphere of UHP argon supplied at a continual pressure (via external pipe) of 5 in Hg.
I assumed that the protective oxide on the bomb interior would be destroyed quickly by these strenuously reducing conditions and couldn’t be counted
on to provide any protection whatsoever. The oxide would, of course, protect the outside somewhat. Is this assumption wrong?
Would being above the oxidation temperature of SS 309 matter under these circumstances? Oxidation will only attack the outside of the bomb and past
experience in metal melting in SS 304 has shown that despite heavy scale formation, the 304 metal will maintain structural integrity over many melts
of copper and brass. So my concern really centers on the question of what effect will be felt by the bomb interior from this Ca/CaCl2 mixture over the
projected 2 heating cycles. I plan two 5 hour runs.
Upon re-visiting the different papers that I have gathered regarding the reduction, I realize that I quoted the wrong conditions.
Both of these variants worked:
1) 5 hours at 950 C
OR
2) 1 hour at 1050 C.
Does a 950 C reaction temp bode any better for my existing reactor of 1/8 “ SS 304?
My temperature control should be sufficiently good that I can keep things at desired temperature +/- 10 C. The reduction of ThO2 is barely exothermic
and so it shouldn’t be a big factor. The original workers didn’t even mention one.
And lastly, even though the Th raises eyebrows, this is obviously not a “terrorist” type project. If I were planning something nasty, the powdery
ThO2 that I have already would be the material of choice. I want to turn this into small beads of metal for sale to element collectors, thorium being
one of the hardest samples to obtain. I have a deadline, too. After Jan 1, 2015 I won’t be able to sell without a license in the US and this is just
a basement project by a retired chemist.
And the answer I received:
Dan:
The 950 C temperature is still well above the limit of 304 SS. However, at the 950 temperature, 309 stainless would be a reasonable alternative that
could withstand both of these conditions. 446 is a ferritic stainless steel and will be hard to find, as quality ferritic stainless steels are not
being produced in the US anymore unless one is willing to buy a full heat quantity.
So, the news is mixed (for me anyway). SS 309 can be found but I will need to construct a whole new reactor from it.
[Edited on 24-10-2014 by Dan Vizine]Dan Vizine - 24-10-2014 at 07:11
An idea I have been floating around the last 24 hours is electrodepositing a refractory metal on the inside of a graphite crucible. These are
relatively cheap on eBay (used by precious metals people) and should make an ok cathode. I anticipate it is possible to make a versatile crucible to
meet home chemistry specifications. The links you posted will give a good start point for metal selection.
J.
I really like that idea. I've toyed with the idea of electroplating Mo on the inside of my SS 304 reactor. Let me know if you proceed, I'd love to
hear your results.MrHomeScientist - 24-10-2014 at 07:33
Another wealth of information, Dan. Thanks for posting! I'll review when I have more time.Oxirane - 24-10-2014 at 09:11
Applying teflon coating or enamel on top of a steel reactor vessel is something that has interested me for several years. Especially the latter would
allow to turn very cheap carbon steel containers into all-suitable chemical vessels because they would stand against everything that common glass
will. They are used ubiquitously in chemical industry.Praxichys - 24-10-2014 at 09:55
Great information!
What about making the reactor from quartz? CaCl2 might diffuse Ca into it but IIRC they temper glass through ion exchange in molten salt baths. Quartz
should be very resistant to most chemical processes, even at high temperatures. Ace Glass specs limited use of quartzware at 1100C. The only issue I
can see is a potential SiO2-fueled reduction reaction happening when using molten Mg, Al, etc.
I imagine a hand-blown thick all-quartz bomb with a ground stopper, to which a stainless steel screw clamp is added to hold the stopper in place under
pressure. Bands, etc could also be used to reinforce the quartz reactor. IIRC xenon arc lamps have quartz envelopes which can withstand something like
30 bar at very high temperatures.
I don't know if parts made from exotic stainless steel would be any less cheap than hand-manufactured quartz vessels.
Alternatively, a quartz/porcelain vessel (like a quartz RBF with a stopper maybe) could be reinforced to withstand pressure by packing it tightly
inside a stainless steel vessel using quartz sand or something. Of course, this mandates batch processing, is tedious to set up, and requires a lot of
heating, but for the amateur it may be of some use.
High pressure reactions might take place in ordinary glass which is completely inside of an external pressure vessel maintained with inert gas
pressure sufficient to balance the internal pressure of the flask. Using temperature to fuel pressure could be simplified by charging the pressure
vessel with an inert fluid with a similar vapor pressure curve as the reaction inside of the flask. A simple pressure transducer measuring delta-P
could control a regulator to a high-pressure N2 bottle or something to keep the pressure balanced.
Just speculating here.Dan Vizine - 24-10-2014 at 13:56
Quartz, even if it could withstand the Ca, would cost a lot more than metal. Any shop can fabricate SS 309. Quartz is a specialty.
The 309 SS isn't really exotic. It's just that 302, 304 & 316 make up 98 - 99% of all offered sheet and round products. eBay currently has SS 309
rounds and a few sheets. And hundreds of welding wires and rods. Nothing for me there....now. The plates are too thin. There are even housewares made
of 309.
There are also equivalents to 309: Nitronic® 60 (Alloy 218) is truly an all purpose metal. A fully austenitic alloy originally designed as a wear alloy,
considered to have the best galling resistance over the entire range of stainless steels. Other characteristics include good high temperature
properties around 1800 F, crevice corrosion and chloride pitting resistance (both greater than Stainless Steel 316) , as well as, oxidation resistance
comparable to Stainless Steel 309 and far better than Stainless Steels 304 and 316.Alloy 21800 is composed of between 16% to 18% chromium, 8% to 9%
nickel, 7% manganese, 3.5% to 4.5% silicon, with the balance being iron. It can be formed using common forming techniques, and preheating is not
required in the welding process. Common welding methods such as gas metal arc, gas tungsten arc and submerged arc can be used to weld this
grade.
The thing that nags at me is that I don't really care much about oxidation resistance inside the bomb where the reaction takes place. My question was
never answered directly.
Notice that this is just another ~ 18-8 SS. Mn and Si contribute to thermal stability of the alloy toward O2 on the outside of the reactor. It doesn't
mean it's necessarily much less likely to leach Ni and Fe than 304. And I know from my own metal smelting experiences that SS 304 lecture bottles cut
in half make crucibles for Cu that last for many runs. Sure, some scale flakes off, but running at 950 C for 5 hrs seems easy enough. Melting Cu was
hotter and some brasses are in that 900 range. The "cuprostatic" pressure at the bottom of a full crucible exceeds the 5 in Hg pressure I'll be
running at. I'm still debating internally.
That last idea of yours is a very good one. So good that that's exactly how chemists frequently run high pressure reactions in glass. Just seal the
[strong] glass ampoule and put it in an autoclave containing some of the same solvent contained inside the glass ampoule.
P.S. Interestingly, eBay sells SS 309 foil to protect tools during heat treatment. I think its just over 0.01 in. Could have some applications....
[Edited on 24-10-2014 by Dan Vizine]macckone - 24-10-2014 at 14:55
I want to turn this into small beads of metal for sale to element collectors, thorium being one of the hardest samples to obtain. I have a deadline,
too. After Jan 1, 2015 I won’t be able to sell without a license in the US and this is just a basement project by a retired chemist.
[Edited on 24-10-2014 by Dan Vizine]
What new regulations are these? I haven't been able to find a reference.Dan Vizine - 24-10-2014 at 15:23
I was told this by eBay seller Billy the Chemist. He was selling ThO2 in large amounts. Feds visited him "just to check"...
He no longer sells it. Sent it back to his supplier, Green Mountain (I think). They don't sell it either.
I've never checked, myself. I take his word for it.macckone - 24-10-2014 at 16:00
You may want to check. My understanding is that unless
it is specific exempted (ore, standard sources, smoke detectors,
etc) any radioactive compounds require a license to sell and has
since 1946 or so.macckone - 24-10-2014 at 16:13
maccckone, If that's the case, then enforcement is fairly lax. Slightly radioactive elements like U and, rarely, Th are sold all the time on eBay by
individuals. Or maybe "they" can see when something is a true danger vs. a non-issue, like a big chunk of U-238, and allocate their efforts
accordingly. Oxirane - 25-10-2014 at 07:19
Metal electroplating is a good way to coat some equipment, but as we all know, many, or even most metals act as catalysts in proper conditions, and
even the most durable and inert elements like platinum group are actually one of the most efficient catalysts in hydrogenation, and so like.
That's why I think enameling is still one option that really should be investigated thoroughly. It can be applied to common steel and as a glass, it
stands against everything except maybe fluorine and strong hot bases, and it can hold up pretty much as high pressures as the steel vessel structure
will, and it should be way more cheaper and easier to apply than metallic coatings, since enameling has been done for a hundred years or something
like that. Can anybody find any articles or technical details how it is done? I remember that one would pretty much simply add powdered glass to metal
and heat it to melt so it solders onto the metal, forming a coating.
Quartz glass vessels would be nice, but I once asked for a small catalytic tube with ground joints and they asked half a grand from it. Mass produced
quartz glassware is sold around somewhat lower price but still they are expensive compared to borosilicate glass. Of course quartz reactor would be
nice because you can see the reactants all the time.macckone - 25-10-2014 at 07:43
maccckone, If that's the case, then enforcement is fairly lax. Slightly radioactive elements like U and, rarely, Th are sold all the time on eBay by
individuals. Or maybe "they" can see when something is a true danger vs. a non-issue, like a big chunk of U-238, and allocate their efforts
accordingly.
Depleted uranium is on the exempt list. There is also a list
For maximum exempt quantity. Thorium is technically not
Exempt but ore is. And thorium itself is not particularly
Dangerous and it is probably low priority. But if you label it
As a radiation standard then you are probably ok.Magpie - 25-10-2014 at 10:10
.
That's why I think enameling is still one option that really should be investigated thoroughly. It can be applied to common steel and as a glass, it
stands against everything except maybe fluorine and strong hot bases, and it can hold up pretty much as high pressures as the steel vessel structure
will, and it should be way more cheaper and easier to apply than metallic coatings, since enameling has been done for a hundred years or something
like that. Can anybody find any articles or technical details how it is done? I remember that one would pretty much simply add powdered glass to metal
and heat it to melt so it solders onto the metal, forming a coating.
I looked briefly into the possibility of enameling the inside of my homemade autoclave as described in Prepublication. I quote here my comment from
that thread:
"A further enhancement that might be possible for my autoclave is enameling. But I understand from Wiki that this is a tricky process and might
not be an option for pipe steel due to high carbon content. I would also need access to a kiln as used in pottery making. If I couldn't do it myself I
would think labor costs would be high. "Dan Vizine - 25-10-2014 at 10:47
Ceramics are harder to do "correctly" than many realize. There are just so many seemingly insignificant factors (to the uninitiated) that can trip you
up.
When you see the process being done (whatever it may be), its seeming simplicity doesn't give you a very good feel for all the tests and calculations
that went into selecting the correct conditions, materials & processing.
If you just read a little about how bathtubs are enameled, the heating requirements alone are formidable. But, if you can access a commercial kiln,
you may be able to pull this off.
I'd suggest buying the glass enameling powder from a commercial source. If you try to compound it yourself, it will likely be more difficult than you
think.
All that being said, I hope I haven't discouraged you, Oxirane. I'd love to see someone crack this nut. As you suggest, it could be very useful.Dan Vizine - 25-10-2014 at 10:57
I have summarized the resistances to molten Li by refractory metals, ceramics and non-metals that I previously submitted as separate documents. If any
of the resident Li fans want to neaten up their collected data, you may want this.
That reference is great! The idea of setting up the required electroplating equipment was at first a bit off-putting. The molten salt deposition
technique (without an external power supply) potentially simplifies things.
For those who may not have noticed this, this is a very useful technique to overlay a base metal with a thin, adherent layer layer of a refractory
metal. This is clearly in the running for my own current project, as this could give superior results to any SS that I could hope to find.Dan Vizine - 27-10-2014 at 19:21
Speaking about reasonable concessions, and just for the record, the use of commercial "iron" or steel pipes is nothing at all like doing a high
temperature reaction in an iron reactor, or maybe I should more correctly say that only to the first approximation are they similar. But most
reactions performed in these reactors will pick up P and perhaps some S from the metal. A lot of the low quality Chinese exports (like everything at
my Home Depot) weld quite poorly, with the welds full of porosity. Vacuum-tight construction can be tricky.Oscilllator - 28-10-2014 at 02:12
Looking at that reference however, it appears that graphite is an unsuitable choice for electroplating Taantalum (and therefore probably other) metals
in molten salt melts, as apparently its surface disintegrates.
This is unfortunate, as I have a bunch of molybdic acid lying around that I was going to try and plate onto the surface of a graphite crucible.
Perhaps an aqueous solution would yield better results. I seem to remember you posted a patent somewhere regarding the deposition of Mo using high
concentrations of acetates, but I can't recall where you put it.Dan Vizine - 28-10-2014 at 04:46
Oscillator,
I don't think I that posted that (I could be wrong), but I just don't recall that patent at all.Dan Vizine - 29-10-2014 at 14:34
Well, I guess it's worth noting that SS309, which is apparently the poor man's high temperature stainless steel can be purchased...where else?, but on
e-Bay.
For $6 or $7 dollars, I found a sheet of 1/8" thick metal which is 4" x 10". That is reasonably priced by nearly anybodies standards.
So in addition to the basic construction materials SS 302/304 and the more corrosion resistant 316 that we have for other projects, we have this as
the best low cost metallic way to work at 1000 C with molten metals.