You do not need to get anywhere near the melting point of the platinum to dissolve it in gold.
Right. Copper and gold will dissolve together even at room
temperature, given enough time. On shortlived test circuit boards,
we used to put gold plate over copper without a nickel guard layer,
and the gold would disappear in a year or so.
Al Balmer
Sun City, AZ
Platinum is soluble in gold, silver and copper and so when brazing platinum with gold solder you get a diffusion of gold, silver and copper into the platinum matrix.
Well, yes, Jim. Technically that’s correct and theoretical. All
platinum workers know, in practice, that a gold soldered platinum
joint will not hold even a little bit. It’s like using glue…
Well, yes, Jim. Technically that's correct and theoretical. All platinum workers know, in practice, that a gold soldered platinum joint will not hold even a little bit. It's like using glue....
Chuckle. Ok, John, now You’re the one slightly overstating things, I
think. Yes, gold soldered joints to platinum are sometimes not very
strong. But using higher melting gold solders, and allowing it to
“soak” a little, does give a usable joint, in many cases. If this
were not so, two tone pieces, or things like platinum prongs soldered
to gold rings, would not work.
There are some cases, though, were it becomes very difficult to get
a joint. One sometimes sees rings of gold with an outside band
wrapped around it in platinum, for a two tone band. If the outer
platinum band breaks, or is cut, or otherwise tempts one to try and
solder the two together, you can have a wild old time, because though
you get everything fitting just fine, when you heat it up to solder,
the platinum has a different rate of thermal expansion than the
gold, and gaps open up in your seam just as the solder wants to flow,
and then on cooling, not only do you have a poorly fitted seam held
in that gapped position by the solder, but then the different rates
of shrinkage can then crack the solder itself. Depends on what
solder, what alloys, and whether Mr. Murphy is watching you try and
fix the ring. But sometimes…
Just the other day, I needed to take the platinum wire prongs off an
18K yellow gold ring that was being scrapped out. Wanted the platinum
wire in the platinum scrap, not the gold. Rather than bothering to
fire up the torch to unsolder them, I figured I could just peel them
off with a round nose plier. Easy for the first two on one side, like
opening a can of sardines… The two on the other side apparently had
gotten hotter for longer, because those wouldn’t peel off at all. The
platinum broke before the solder gave way. As good as a solder joint
from gold to gold. So then I had to fire up the torch anyway to take
off those two remnants… So sometimes, you get a good joint, and
sometimes it’s not as good as it looks. The big problem for me is
simply that you can’t tell by just looking whether it’s any good or
not.
Peter
Peter,
So to a degree, both levels of statement have some truth. The accurate statement of a true metallurgical bond existing is correct, but doesn't completely address the bond strength, which can still vary based on the dissimilar types of metal. But the statements claiming merely a "glue" type level of bond, while not exactly correct, ARE correct in terms of the behavior the bonds appear to exhibit in real life.
This has more to do with the ingredients of the solder alloys than
the diffusion depth. The diffusion of the solder is not significantly
different in either case. Because there is a lot more zinc in the
low temp solders you have a more brittle crystal structure in part
because the zinc has a hexagonal crystal structure and the Pt, gold,
silver, and copper are all face centered cubic. The mismatch in
crystal structure makes for less ductility in the alloy and the more
zinc the worse it gets. Also the high strength of the platinum allows
you to peel it without the wire breaking.
There is another gotcha here, platinum has a significantly lower
thermal expansion rate than the gold or the solder alloy. That can
actually cause the bond to be under so much stress that it literally
self destructs. We did some calculations on diffusion bonded
platinum and 18k gold, when they cool the difference in contraction
creates enough strain on the bond that it exceeds the tensile
strength of the gold and it just peels apart as it cools. With a
liquid phase bond you get a little more leeway but the difference in
expansion rate can still put enough strain on the bond to end up with
cracks in the crystal lattice which may be the other reason for the
peeling.
James Binnion
James Binnion Metal Arts
Platinum is soluble in gold, silver and copper and so when brazing
platinum with gold solder you get a diffusion of gold, silver and
copper into the platinum matrix.
Well, yes, Jim. Technically that's correct and theoretical. All platinum workers know, in practice, that a gold soldered platinum joint will not hold even a little bit. It's like using glue....
That is why I like to use silver solder on platinum / gold joints it
does not peel because it is more ductile than gold solders and can
better deal with the strain from the mismatch in thermal expansion
rates which is most likely the main problem with the joints soldered
with gold solders. Try it sometime.
Jim
Peter
So to a degree, both levels of statement have some truth. The accurate statement of a true metallurgical bond existing is correct, but doesn't completely address the bond strength, which can still vary based on the dissimilar types of metal. But the statements claiming merely a "glue" type level of bond, while not exactly correct, ARE correct in terms of the behavior the bonds appear to exhibit in real life.
Thank you.
That was an excellent post, and a prime example of what can be
learned on Orchid from those experienced in their craft.
Wayne Emery
The Gemcutter
Well, yes, Jim. Technically that's correct and theoretical. All platinum workers know, in practice, that a gold soldered platinum joint will not hold even a little bit. It's like using glue....
Jim is absolutely correct. The concept he describes is not
theoretical, but actually has been in use for many years. Before
blacksmiths had learned how to do welding, they brazed parts together
using pure copper and were able to achieve strong and practical
joints. Just because some have problems with the technique, it does
not mean that technique is faulty, but the skills of the practitioner
needs improvement.
Leonid Surpin
All platinum workers know, in practice, that a gold soldered platinum joint will not hold even a little bit. It's like using glue....
hmmmm. well then I must have been using some killer glue all these
years
One time I did a little test. I made two similar joints in plat. one
was pt soldered the other was 20K white welding (don’t blame me for
trade names). I stressed the hell out of them and the 20K was
stronger, at least under the duress I applied at that time under
those conditions. Having glued 18Y to pt on a couple of occasions I
must conclude that i am indeed, not a platinum worker.
If you have a molten metal in contact with a solid one and the solid one is soluble in the molten one then there will be a metallurgical bond between them as the molten one dissolves the solid. You do not need to get anywhere near the melting point of the platinum to dissolve it in gold. Platinum is soluble in gold, silver and copper and so when brazing platinum with gold solder you get a diffusion of gold, silver and copper into the platinum matrix. If you were to take a small bead of platinum and place it in a crucible of molten gold the platinum would totally dissolve into the gold given enough time. You do not have to melt the highest melting point metal first to make an alloy it is just much much faster to do so.
Too true. Chemists and physicists who use thin films of metal make a
wafers of copper with a layer of gold on it that is exceedingly
thin, and need to use it in a reasonable amount of time. If it is
left in storage for too long (weeks to years depending on the
thicknesses) the atomically thick layer of gold will dissolve into
the copper substrate at room temperature.
Jason
Hello all, I have had a lot of luck using hard white gold solder on
platinum when working next to or on top of stones. It seems to hold
fine but needs rhodium to look right.
Have fun. Tom Arnold
Chuckle. Ok, John, now You're the one slightly overstating things,
Partly the nature of forums - sound bites. It may seem that Jim,
Peter and I are at odds here, but they are actually making the
point. When you braze iron or steel with brass or bronze “brazing
rod”, the temperatures of the two are close enough that they actually
alloy at the juncture. The melting point of platinum is so high
relative to gold solders (as in another post, on the order of 5,000F)
that most often it acts more like a soldered joint than a brazed one.
Solder a platinum earpost down with gold (or silver) solder and it
will just snap off… The workarounds that Jim and Peter talk about
are well known in the platinum business, but they are exactly that -
workarounds. And as Peter pointed out, you just never know by
looking if it’s a good join or not. In the trade what happens is best
practice - don’t depend on the solder at all, design the piece in a
way that takes the dissimilar metals into account to begin with, use
the fundamental structure for strength and not the solder. And do as
much as possible with the platinum before merging it with the gold…
This is really the whole issue, and although Jim and Peter both seem
to be at odds, they are actually saying the same thing in different
words… I’m like most people here - I don’t want or need to know if
the coefficient of linear expansion is.8 or.10 - if I solder a ring
seam and whack it with a hammer, is it going to hold? Platinum ring,
gold (or silver) solder - no, it won’t, 99% of the time… Been
there, done that…
Before blacksmiths had learned how to do welding, they brazed parts together using pure copper and were able to achieve strong and practical
We’re not talking about iron, we’re talking about platinum (doh),
which is double and triple the melting point. It has nothing
whatever to do with technique - the differential in temperature
between melting platinum and white gold solder is something like
5,000F.
This is, as Peter also pointed out, an everyday issue in shops that
actually work platinum. Practical knowlege, not theory.
but doesn't completely address the bond strength, which can still vary based on the dissimilar types of metal.
I thought I’d share a conversation I just had…
Standing outside, watching the world go by on a break, and a
colleague comes out and walks up to me. Like me, he’s a platinum
specialist for some decades… He asks me if I have a platinum
tester, which I don’t. He says he has a job to put an old setting
into a new ring, and it’s supposed to be platinum, but he doesn’t
think so… It’s to go on a platinum shank with 6 baguettes. He says,
“I have it here, want to see it?” Sure … And it’s around a 4
carat in an old gallery setting, nicely made. I say I don’t think
it’s platinum, it’s white gold… “Of course instead of waiting,
you could just do the work - doesn’t matter that much in the end.” I
say…
And he says, “Yes, but of course the platinum shank won’t stick to a
white gold setting, and I don’t want it coming back on me…”
Meaning: In which case we’ll have to redo the job with a white gold
shank. The attitude being, “But of course you understand that…”
Which of course I do…
Oh, and he also had some pithy words about the pre-platinum
technology guys using silver on platinum (and having potential
trouble there), but I won’t go there now…
It’s just the nature of the beast. Platinum is like no other metal.
Peter, James,
Regarding PT and AU joints. just out of curiosity, what is the
quality join if the gold is cast with the platinum set in the
investment?
Mike DeBurgh, GJG
Henderson, NV
We're not talking about iron, we're talking about platinum (doh), which is double and triple the melting point.
Platinum melting point - 3215 degrees F.
Iron melting point - 2800 degrees F.
3215 / 2800 = 1.15
Sorry, but 1.15 is not a double, and definitely not a triple.
Leonid Surpin
the differential in temperature between melting platinum and white gold solder is something like 5,000F.
John I have known you for a long time and I have great respect for
your skills and knowledge but, you are way off on the melting point
of platinum and the root of the problems in brazing platinum with
gold solders. Platinum melts at 3214.9 F white gold solders melt
between 1200 F for a typical easy and 1700F for the 20K welding
white so 1500F to 2000 F differential worst case. It is nothing to
sneeze at but again the melting point is not the problem it is the
thermal expansion mismatch tearing apart the joint. The gold solders
do diffuse into /alloy with the platinum but that joint can be broken
on cooling by the stresses involved. The more ductile your solder
the more likely it will survive the process.
James Binnion
James Binnion Metal Arts
John, I think the difference in melting points between plat. solder
and white gold solder is somewhere around 1500f. Gold solder will
indeed stick to platinum. I have built many two tone pieces,
combining gold with platinum.
They have stuck together. I can’t say they’ve all been pretty, but
they’ve stuck together.
Tom Arnold
If you were to take a small bead of platinum and place it in a crucible of molten gold the platinum would totally dissolve into the gold given enough time
And if you take powdered chocolate and put it in milk, it will
dissolve, too.
Since I started this mess, I’ll finish it. What I said was that gold
solders on platinum form a weak link because the temperature
differential is so great. The reason I said that is because it is
common knowlege. The hotter the flow point of the solder (or
non-solder metal) is, the better the bond becomes. Also common
knowlege. When you get to the point of “easy” platinum solder - 1200
(which is 1200C) then you start getting a pretty good bond. And yes,
1200 solder is actually a gold alloy - also common knowlege. 900
welding is 900C and is about as good as you’re going to get out of
the gold family of solders. We don’t need to discuss whether gold and
platinum dissolve in each other or whether iron can be brazed with
brass rod because we all know that. CAN platinum be soldered with
gold solders? I just did it 3 minutes ago on a diamond ring. Does it
form a strong, permanent joint - is it PROPER? Absolutely not.
Platinum needs to be worked on it’s own terms just like every other
metal, or else you’re just another hack.
Saying that I need to work on my technique is like saying Einstein
needs a math tutor…
Regarding PT and AU joints. just out of curiosity, what is the quality join if the gold is cast with the platinum set in the investment?
Mike, I’ve only done this a couple times, with 18K yellow gold and
platinum. It did work, but I can’t comment too much on the quality
of the bond, since I didn’t rely on it. Instead, I cut “keys” into
the platinum that the gold would flow into, giving a mechanically
interlocked joint. The seams looks good enough, but I have no way of
knowing if the apparently fused seam had any real strength. I
suspect that it did not, though, because the line between the two
metals/colors was very clean and crisp. It didn’t look like the gold
had thoroughly “wet” the platinum. But again, the models had also
been made with the idea of such a look to the seam, so it’s hard to
be sure. I should mention, though, that these were cast in a
centrifuge, so I could use a substantially lower flask casting temp
(as I recall, around 750F), so the gold was cooling fairly quickly.
Not as much time molten in contact with the platinum as a higher
flask temp might have allowed.
HTH
Peter
Since I started this mess, I'll finish it. What I said was that gold solders on platinum form a weak link because the temperature differential is so great. The reason I said that is because it is common knowlege. The hotter the flow point of the solder (or non-solder metal) is, the better the bond becomes.
Liquidus temperature is only one aspect of solder bonding and is not
any more or less important than several other factors. It is
certainly no guarantee of a strong bond.
Saying that I need to work on my technique is like saying Einstein needs a math tutor.....
I am not certain if this is aimed at me but since you quoted one of
my posts I will ask you to re-read my posts, I made no such
statement. In fact I said "I have great respect for your skills and
knowledge "
You are one of the first people that come to mind when I think of
master goldsmith.
Regards,
Jim
James Binnion
James Binnion Metal Arts