Back to Ganoksin | FAQ | Contact

Lab scale casting & refining questions


Dear Orchid Folks,

I’m a metallurgist who is much more used to dealing with steel, but
I find myself needing to use some gold and gold alloy as “model
materials” to study their recrystallisation & grain growth
(annealing) behaviour on a microscopic level (its part of my PhD). I
need to work with these metals on a very small lab/craft scale, so
I’ve come to you…

There are two things I need to do.

  1. I have about 3 grammes of 99.99mass% gold. I need to melt this
    and then cold work it as much as possible (preferrably by rolling)
    to produce approx 1mm thick “fully hard” sheet/slab, from which I
    can cut a 3-4mm square or round specimen. I will then polish this to
    at least a one micron finish and finally anneal it in the electron
    microscope (so that I can watch the new crystals form and grow).

Its the melting part that I’m having trouble with. I have a number
of electric furnaces available that would do the job, plus a little
gas torch that might do the job.

The first problem is what to melt it in? I tried some scrap 9K gold
in an electric furnace, but the “gold” (a silver-gold-copper-zinc
alloy) attacked the crucible that I had borrowed from the chemistry
department, causing the bottom to drop out and my gold to vanish
into the bricks in the bottom of the furnace.

I really don’t want this to happen with the 99.99% stuff.

What materials can be used to contain molten gold?

Similarly I may need to cast my gold to some shape more specific
than a button before working it - what are the candidate mould

I need to maintain the four-nines purity level in my gold. I’m
assuming that as gold is so unreactive that this should be no
problem so long as I keep it away from copper, silver, zinc etc. Am
I right in that assumption or do I need to take precautions against

Any points to look out for on rolling gold? I’m assuming that (being
one of the most malleable metals in the world) it will be very well-
behaved in cold rolling.

What kind of reduction per rolling pass is practical? and about how
much reduction before annealling is necessary?

I don’t know how jewellers measure reductions on cold work, but
metallurgists find % reduction of area the most useful, so If I
started with a 2mm thick ingot and rolled it to 1mm thick (keeping
the width the same) that is a 50% reduction.

  1. The second material I have is scrap 9K gold (so that I can
    compare the response of a pure metal with a highly alloyed one).

Unfortunately the scrap 9k gold contains Zinc, which causes huge
problems when heated in the vaccuum of an electron microscope.

Is there a way that I can reliably remove all the zinc from this
alloy (by melting and oxidising it perhaps?) or will it be much
simpler to get some higher karat zinc-free material and re-melt it
with additional silver and/or copper?

I understand that melting and oxidising the zinc could produce some
noxious metal fumes. That’s OK as I have access to suitable fume

I do not care exactly what composition I finish up with (I can
easily check the analysis anyway), so its OK if the zinc removing
treatment also takes out a proportion of the other alloy elements

Having removed the zinc, I then need to cast and work this alloy as

Thanks in advance for any advice you can give me and appologies for
the long post.

Iain Fielden
Sheffield, UK


You’ll no doubt get more erudite replies than mine - but some basic
things - I would think the very best container for melting both of
your metals would be a porcelain crucible (or melting dish -
different shapes) In the US, I would refer you to or
Scientific supplies have them, too - usually triple the price. The
common method for casting an ingot is to use an ingot mold. 1) I
don’t know if you need to buy one for a couple of pours, and 2) They
are steel, which will theoretically not contaminate your pour, but it
IS metal.

I would suggest getting a charcoal block (also at those links) and
chiseling a shape into it-it’s very soft - wrapping wire tightly
around the perimeter will help keep it from cleaving into pieces,
which tends to happen. I do not know the math of your
rolling/.9999/hardness question, but in practice, to get full hard
24kt. you’re going to need to roll more like 6mm to 1mm - alloy gold
considerable less, but 24kt. doesn’t work harden nearly as much as
Cu/Ag/Zn and the like. Lastly, if I were doing your experiment, I
wouldn’t want to use an “off the shelf, who knows what’s in it alloy”

  • I’d make my own. And zinc is filthy, too, in many ways - it can
    burn out of your alloy in melting, and change the alloy %. Multiply
    your gold weight by 1.667 to get the amount of alloy to add to get
    9kt. Thus 10gm. Au x 1.667 = 16.67gm alloy. 10gm. Au + 16.67gm Alloy
    = 26.67gm 9Kt. What the alloy is, is up to you, it’ll still be 9Kt.
    Melting point 24kt. =1945F., by the way. Sounds like an interesting
    experiment, too