Chlorination of rhodium sponge

Gold Refining Forum

Help Support Gold Refining Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Lou

Moderator
Staff member
Moderator
Joined
Nov 4, 2007
Messages
5,203
Location
Rhodia Drive
The topic at hand is making a soluble rhodium salt from 99,95 pure rhodium sponge. Emphatically, this post is not intended as a how-to and is not something to conduct by unqualified individuals or the casual hobbyist/professional refiner. Chlorine gas is corrosive, poisonous, and a potent oxidizer when hot-- all appropriate safety considerations should be followed. It has been prepared for reference purposes only, and the general amusement of my friends here on the goldrefiningforum!

Introduction

Rhodium is very difficult to put into solution, and virtually impossible to do so when in a bulk form such as wire, sheet, or plates. Even a 10% Rh addition to platinum (i.e. in thermocouple wire, Types S,B, R etc.) significantly increases the dissolution time from hours to days under atmospheric conditions. Many methods for refining rhodium involve melting with other metals (Zn, Cu, Mn etc.) and leaching away the matrix to provide a rhodium black with the result being that the higher the surface area of the rhodium black allows for faster solubilization.

It is possible to microwave pressure digest rhodium sponge (or other alloys), but results are often sub-optimal.

Lixiviants for rhodium are fairly limited:

  • Aqua regia, boiling
    Hydrobromic acid, boiling
    HCl/Cl2, 120*C, pressure
    18 M + sulfuric acid; reflux, preferably under pressure and/or admixed with a Cl2 source

Melts can be made with any acidic sulfate salt to generate the red-brown rhodium sulfate as previously described by freechemist. This reaction proceeds faster when chlorine is produced in situ with the addition of NaCl.

The goal is to make Na3RhCl6, a highly soluble rhodium salt that gives a burgundy wine colored solution, even at low concentrations. To do this expediently and with a high first past conversion, acid-insoluble RhCl3 is formed as a dark brown powder. This is then mixed with considerable excess of sodium chloride, and rechlorinated to make the soluble salt.

Reactions governing the system are as follows:

1. Rh(s) + 3/2 Cl2 (g) --> RhCl3 (s), 500-650*C, excess Cl2
102.91 209.27
2. RhCl3 (s) + 3 NaCl (s) --> Na3RhCl6 (s), 500* C, excess Cl2
58.44 278.24


Chlorine generation:

Instead of lugging out a chlorine cylinder, it's easier to make it in the fume hood chemically.

C3N3O3Cl3 (s) + 3 HCl (aq) --> 3 Cl2 (g) + C3N3O3H3 (s)
232 g/mol 36.46 g/mol 70.91

Trichloroisocyanuric acid (TCCA) is used as the oxidizer and acts upon 4 M HCl to cleanly generate chlorine; no heating is required to drive the reaction as the production and removal of a gas favors the products; stirring the slurry helps. The 4 M HCl is slowly dripped from an equal pressure addition funnel into the TCCA slurry such that a smooth rate of evolution is produced as produced at the bubbler. The joints are greased with Krytox 240 AC to tolerate the chlorine, and the connections are made with FEP tubing. The gaseous chlorine so produced is ran through a column of 18 M sulfuric acid to remove any HCl from the gas stream. Excess Cl2 is scrubbed with two traps (always have an empty trap in line to prevent suck-back into the line. It would be disastrous to have that happen). The end trap is 8 M NaOH which will yield hypochlorite and other oxo-chlorine compounds. Quartz wool is stuffed into the gas output end of the gas wash bottle to prevent any H2SO4 mist from entering the tube.

Rhodium (III) chloride
Next, pure rhodium sponge is weighed out and loaded into a clean quartz tube (Photos 1,2,3). It is important not to touch the quartz with one's bare hands as salts from human sweat will eventually cloud it from devitrification. This tube is packed with quartz wool on the output end. This is done for several reasons: first, in the event that a large surge of chlorine forms, it would be bothersome to have to remove blown out Rh from the trap, second, to look for the deposit of RhCl3 indicating that it has started vapor transport and the reaction is finished. Overall setup in Photo 4.

3N5 Rh.jpg
weighed Rh portion.jpg
Rh in quartz tube.JPG

In the small test hood:

setup.JPG

One can see in Photo 5 the production of RhCl3 and various subchlorides. Chlorine was used to displace air from the tube at lower temperatures (the rate of reaction is minimal at less than 400 C). The reaction depends upon diffusion of hot chlorine into the rhodium sponge, so periodically rotating the tube (caution! the glass will not appear hot in well-lit conditions!) helps complete the chlorination. Note the progression from left to right.
550C RhCl3 formed.JPG
In the next installment, we'll show sodium hexachlororhodate being made by intimately grinding the RhCl3 with NaCl and re-chlorinating.

And yes, that hood is dirty :-/
 
Thanks for the typo correction and interest from you both.


Back to it:

After several hours of temperatures ranging between 450-600*C, the RhCl3 is produced. This is evident when some of the material condenses upon the cool end of the quartz as a red film. It has been my experience that until the chlorination is complete, this sublimation product doesn't deposit. This can be seen in Photo 6. The chlorine valve at the one end of the tube is closed and the material is left to cool in a chlorine atmosphere. It is important to cease the addition of HCl to the TCCA at this point, and open the other valve slightly to allow it to vent so that the tube does not pressurize. The tube should remain slightly green-yellow from chlorine during cool down, or else the chlorination was incomplete and Rh is still depleting residual chlorine.

An issue to avoid is the competing reaction:

2 RhCl3 (s) + 3/2 O2 (g) <--> Rh2O3 (s) + 3 Cl2 (g)

This oxide is also acid insoluble and largely useless. To avoid this, one should not heat the powder up until chlorine has purged out the lines, nor should one close the system until this is accomplished. Rh2O3 looks quite similar to RhCl3. Keep air/oxygen from the system. Rh2O3 is more stable than RhCl3 and will take considerable amounts of chlorine to displace the oxygen.

The next step is to weigh the cooled chlorination product to ensure that its mass gain is appropriate to the stoichiometry of reaction 1. Do note that some residues always remain in the tube or boat but these can be cleaned out with certain solvents or sublimated off the glass at 880C with a fast chlorine stream. Photo 6 shows the finished product vs. the starting material.
RhCl3 vs Rh.jpg
Now that the material has been weighed, and the weight gain found to be appropriate, the material is put into a glass mortar and pestle (do not use metal for handling of this; scoopulas, spoons, mortars; all will contaminate the product). This mix is ground with 3 X its weight in ACS sodium chloride and re-loaded into the tube (Photo 7).
Salt + RhCl3.JPG
It is quickly brought up to 500-550 C and held under a brisk stream of chlorine for two hours. A final ramp to 800+C to melt the material and what's left is a black glass (see Photo 8).

Na3RhCl6.JPG
The material is then weighed, dissolved in water, filtered, the residue caught and ashed in open air and post reduced with a stream of hydrogen at 350 C. This is then heated to 800 C in argon to remove adsorbed hydrogen that would present a fire hazard when handling the cool Rh powder.

I'll attach a photo of the resulting Rh solution soon.
 
This is just amazing, seems like it was not that long ago I was reading about you doing experiments in your garage, now here you are in a lab chlorinating Rhodium successfully.
 
Very cool stuff Lou! 8)

Thank you so much for taking the time to take pictures and document that for us! I will probably never try this, but I love having at least a little insight into how it's done. I guess my Mason jars and plastic buckets aren't up to the task. :|

Dave
 
Everyone starts somewhere. When I joined this forum, I was already doing this type of work in a lab setting for a couple of years.

I'm just glad to show these type of processes despite the minimum applicability to most on this forum. Handling these precious metals on a daily basis has really given me perspective for how difficult this profession really is. It's almost like being a bank teller--handle money every day, you're accountable to the last penny, meet all types of personalities, keep all sorts of accounts straight, etc. Big difference is that any error made on their end is in mis-entering a value for a check and is a little more easily remedied. When refining gold for a percent or two, any mistake, and there goes the refiner's profit. Couple that with the reality that any physical operation has some loss, easily determinable or not and it really keeps a person on their toes.

Case example:
Say someone sends in 100 ounces of thermocouple wire. Perhaps 96 of those ounces are platinum, the rest is rhodium and perhaps traces of base metals from handling.
For processing that 100 ounces, competitively, one might gross 3.36 ounces of platinum and a half ounce or so of rhodium. Considering all the melting/digesting/front-end assaying/refining work/back end assaying and then making a deliverable product while hedging it and following traces in the processes all over the system to keep the mass balance and so on and so forth... it's a right proper bear. It's much easier to manage when doing more metal but volume through a refinery is anything but constant. Ultimately, it's a lot of work and involves juggling many different hats for not much of a return. Still, it's a living and it's generally fun. Has its ups and downs like anything else. At least I get to meet a lot of great people who have really interesting stories and moreover, it's great flattery to be trusted to handle their money.
 
Nice photos!. Does the "black glass product" stick at all to the quartz?. Or is it easy to remove?.

How much % of the original Rh remains unreacted?.

Do the impurities get flushed out?. (I imagine all, but Iridium, do).
 
The black glass delaminates itself from the quartz upon cooling. It makes loud/scary popping noises from the internal stress.

I took the material, ground it up, sonicated and then microwaved it into solution. A fine residue of rhodium black remains, in this instance, about 80% conversion to a soluble product.
 
Thanks Lou. I've never seen so much Rh sponge processed in my life, and probably will not in the future, but it's nice to see these processes and learn how they work.

I can imagine a variation of this GPC process for Rh can be used to produce extremely pure gold from gold sponge that has been already through the liquid chemistry cycle. I'd love to read whatever you can say about that. :shock:
 
It's not that much, a couple ounces.

As to Gas Phase Halogenation (GPH)...you can make ultra high purity gold with that method, and other methods beside. It takes a very, very long time and very stringent conditions. I use highly pure iodine, but one crystal to do the job. I start with spectroscopically pure gold and put in a quartz tube. The temperature is low.

The process is pretty straight forward--at one end is the pure gold, at the other is a hot zone. The Au forms AuI3 and disproportionates back into pure gold. The growth of these crystals is kinetically controlled more than anything else--high surface area gold powder reacts faster than the lower surface area gold crystals, so the longest part is the induction period. The net result is mass transfer.

The sluggishness of the process makes the gold very pure, as it goes on layer by layer and forms large crystals.

Ivan Timokhin (periodictable.ru) really kick started this process and has it figured for all of the vapor-transportable PMs. He's also a great guy.
 
Lou said:
It's not that much, a couple ounces.

As to Gas Phase Halogenation (GPH)...you can make ultra high purity gold with that method, and other methods beside. It takes a very, very long time and very stringent conditions. I use highly pure iodine, but one crystal to do the job. I start with spectroscopically pure gold and put in a quartz tube. The temperature is low.

The process is pretty straight forward--at one end is the pure gold, at the other is a hot zone. The Au forms AuI3 and disproportionates back into pure gold. The growth of these crystals is kinetically controlled more than anything else--high surface area gold powder reacts faster than the lower surface area gold crystals, so the longest part is the induction period. The net result is mass transfer.

The sluggishness of the process makes the gold very pure, as it goes on layer by layer and forms large crystals.

Ivan Timokhin (periodictable.ru) really kick started this process and has it figured for all of the vapor-transportable PMs. He's also a great guy.
I was thinking along the lines of sublimating AuCl3 from the hot end of the quartz tube, transporting the vapor under a stream of Chlorine, and depositing the sublimate at the other, cold end of the quartz tube. Temperature less than 200 C. I'm not sure it works.
 
Sure it does--I've made really nice AuX3 crystals that way (then dissolve in an ether and filter to remove any gold)


Here's some of the kinetics for M/Cl2 system.

Iodine and bromine are much softer and seem to work better on the softer metals at lower temperatures, in finely divided form. Chlorine is more aggressive insofar as oxidation goes. BrCl I find to be a good compromise.
 

Attachments

  • Screen Shot 2013-05-24 at 9.31.06 AM.png
    Screen Shot 2013-05-24 at 9.31.06 AM.png
    85.8 KB · Views: 323
Fascinating! Thanks for taking the time to write down this guide.

8)

Göran
 
Back
Top