Author Options:

Jewellery made from both silver and gold.? Answered

I plan to make (ideally through lost wax casting) a (fine) silver necklace pendant with (fine) gold lettering. I have made items using this process out of both metals, but have never combined both metals in the same piece. How would I go about doing this?
Thanks in advance.



Best Answer 8 years ago

(Warning: Long. For those who know me, long even for me. :)

I'm not a metalsmith, but I am a metallurgical engineer who used to work for the jewelry industry, doing research on jewelry metals.

Re melting points:

The values Kelsey mentioned are for the pure metals. You will probably be working with alloys of gold and silver, referred to in the industry as "karat gold" and "alloy silver."

Alloying a metal lowers its melting point, for the same reason that spreading salt on winter ice lowers its melting point.

The bad news is that I can't just tell you what the melting point of 18 karat gold or sterling silver is, because the MP depends on just exactly what metals the gold or silver is alloyed with.

14 kt gold, for example, is legally required to be  58.33% gold; but the remaining 41.67% can be anything at all. Usually the alloy elements are some combination of silver, copper, nickel, zinc and sometimes tin; but exactly what that combination is varies with the individual refiner and the alloy's intended purpose - casting, cold work, wire, etc.

The good news is twofold:
a) your jewelry metal supplier (or his refiner) should be able to tell you the composition and MP of any given alloy, and -

2) You can probably find a comparatively high-MP karat gold and a significantly lower-MP alloy silver.

A good jewelry metal supplier may even be able to hook you up with gold & silver alloys that are formulated to be cast with each other.

Thermal conductivity:

Pure gold is the most thermally (and electrically) conductive metal. Pure silver is second-most: this is why leaving a silver spoon in a cup of hot tea tends to cool it off.* Copper comes in third.

(*This doesn't work nearly so well with a stainless steel spoon - stainless is one of the least thermally conductive metals.)

In all likelihood, the alloying metals won't significantly change the relative thermal conductivities of the gold & silver. (I think) you can (probably) assume that gold in good, full contact with silver will (tend to) be at the same temperature as the adjoining silver.
(Parenthesized weasel words provided to prove that I really am an engineer. :)

Coefficient of thermal expansion:

Again, these will vary with the alloys. If you can find gold and silver alloys made to be cast with each other, they will probably have thermal expansion coefficients as close as the refiner can get them.

There will still be some significant difference, but I think that that may be to your advantage. Remember that as the silver contracts more than the gold does upon cooling, any holes in the silver will also get smaller, forcing the silver into good tight contact with any gold in that "hole."
(That's the theory anyway - I've never done this in practice. Like I said, I'm an engineer. :)

That should mean that the "two-mold" process suggested by Kelsey and Raving could work.

Some practical stuff:

I'd suggest that you sketch out your ideas, or maybe even carve initial versions of the wax (waddyacall'ems) you'd be using, and take them in to a good jewelry metal supplier (not "Big Box Hobby and Crafts," but someplace more like "Gold & Silversmith's Equipment and Supply"). A good shop should have someone who can tell you about which karat gold and alloy silvers will work best, and may even be able to help you with design & casting parameters.

Good luck with your project!

PS. You probably already know this, but Casting + Sharp Corners = Not So Much.  So a nice flowing cursive font for your lettering is a better bet than, say, Roman-style block lettering.

Not certain if this is a dead thread or not, but here are my thoughts as a senior Jewelry and Metalsmith student at Purdue :

[EASY]-If you are going for the inlay, script text look: Just do a "solder inlay" with hard gold solder or 10/14k yellow gold. (Either carve your text recesses in your wax before casting in silver or use an engraver to cut channels for text after casting. Fill with solder. Sand flush.)

[HARD]-If you want raised or large letters: Cast your silver piece with or without recesses for text. Then take your silver piece and add wax directly to (or in the recesses) where you want text. Sprue, invest, and burn out as usual. The two castings will most likely NOT be fixed together and will require soldering to finish the piece. Directly casting onto your pendant gives you almost the tightest fit between the two metals, great for contours and textures.

Thoughts on the above:
- Electroplating requires a cyanide or aqua regia solution and a full set-up of equipment. Potentially hazardous, toxic, and expensive.

- Keum-boo is thin and not easily detailed.

- Precious Metal Clay shrinks a little bit when fired and lacks sharp detail, but would save time on doing a second casting.

- And, the metallurgical stuff above is interesting and accurate, but not worth knowing for this project unless for your own edification. Melting temps between base metals and solder are all you need to worry about.

You could do it mechanically (brazing or fastening the parts to one another). 

You could also do it with two molds, but you need to be sure (by design) that you cast the gold (MP 1064 C) first before doing the silver (MP 962 C).  The idea would be to do a lost-wax mold for just the gold part fabricate it, then build the wax for the final piece using the actual gold where it belongs.  You end up with a lost-wax mold where the gold (higher MP) is already positioned, and put the silver in "around" it.

My suggestion is strictly theoretical, based on conversations I've had with our jeweler, who did my wife's re-engagement :-) ring in gold and platinum.  Hopefully and actual metalsmith will give you the right answer!

First and foremost, congratulations on your re-engagement =]

Secondly, as for the two mold method, is the difference in melting points sufficiently high for that to work?
The slight problem with brazing, is that I wanted the lettering inlaid. so there is no surface joint between them. I suppose I could make them separately, with the silver recessed. And then solder them in...
Any thoughts? =P

Solder inlay? That is, just using solder of a different color than the base metal, then polish flat.

I think doing it as an inlay is probably the best.  Can you do the part-counterpart shapes precise enough to get a tight friction fit?  Or maybe make the gold inlay a few tens of microns larger, and warm the silver enough to get a thermal-expansion fit?

The closeness of the melting points is something that concerned me as I was writing my first response.

A rule of thumb I learned in freshman geology is that metals and minerals generally become plastic (deformable) when their absolute temperature (i.e., in kelvins) is above 50% of their melting temperature.  With only a 100C == 100K difference in melting points, it's certainly plausible that pouring the silver onto the gold could get you into that regime.

However, this immediately leads to the quantiative question.  If you start with gold at room temperature (300K) and pour a relatively small volume of molten silver (at 1235K), the gold will start to warm up as the silver cools down.  Will the gold be able to get above about 650K (380C) while the silver is cooling, before the whole system is at a temperature below that point?

One very stupid question I have for you.  My understanding of the casting process is that you allow the system to cool by radiation (it sits there in air until the metal hardens), not by immersion in water.  Is that right?  The latter would obviously keep the temperatures down, but not the former.  Also, if you have to heat the mold all the way to the silver's melting point, then you've explicitly brought the gold up that hot too, and softened it.

Whether you choose to allow the mold to cool by radiation or by quenching depends on what metal you are casting, and what you want to do with it. Quenching would harden the metal, making it brittle, this can be desirable in certain applications.
For this, I would allow the mold to cool by radiation.
The other thing that worried me was given that both metals have (presumably) different temperature coefficients of expansion, trying to heat and cool them together would lead either to the lettering falling out, or to the silver deforming on even fracturing.

Quenching hardens SOME metals, and softens others. Most famously, iron is hardened, as is nickle (in some white gold-but I'm guessing the author is from Europe where white gold is now Palladium instead). Yellow gold and copper are recommended to quench to soften. Silver can go either way, depending on the temp reached (precipitation hardening).

jtbako said:
"Quenching hardens SOME metals, and softens others. Most famously, iron is hardened, as is nickle...
...Yellow gold and copper are recommended to quench to soften. Silver can go either way, depending on the temp reached (precipitation hardening).

Um, not exactly.

For any pure metal and some alloys, it depends on what's happened to the metal before you start your heat-&-quench cycle.

For most metals in a fully annealed state (e.g. previously heat-soaked at a temp roughly 50-80% of its melting temperature and then allowed to cool slowly), reheating followed by a rapid quench from the annealing temperature will resulting in grain boundary hardening.

But if the metal has already been quenched once to create grain boundary hardening, a second heat-&-quench cycle may soften or harden the metal; depending on rate of temperature, the heat soak temperature itself, the quench rate, and the grain-boundary formation characteristics of the specific metal involved.

If the metal has been worked (bent, hammered, pounded, stretched or other subjected to a change in shape) since its last full anneal, it will soften upon a slow quench due loss of the work-hardening effect - but harden on a rapid quench due to the influence of prior cold work on grain boundary formation. Repeated cycles of cold work followed by heat-&-quench cycles can be used to tailor the hardness and ductility of a metal to desired parameters - if you know what parameters to use for that particular metal.

But for steel, stainless steel, aluminum-copper superalloys, and other alloys subject to precipitation hardening, precipitation softening, grain boundary embrittlement, and similar quench effects; it can get a bit complicated. :)

jtbako also said:
"... nickle (in some white gold - but I'm guessing the author is from Europe where white gold is now Palladium instead)."

White gold is a karat gold alloy typically including silver, copper, nickel, zinc. The "white" color comes from using much less "red" copper; and proportionately more "white" metals such as silver, nickel, and zinc; than in yellow karat gold alloys.

Palladium is one of the Platinum Group Metals, and is a naturally occuring element.

As of market closing on June 4, 2010, the spot price on the New York Commodities exchange for silver was USD $17.41/oz. The spot price for Palladium was USD $425/oz.

Alloys of palladium, platinum, rhodium, and other platinum group metals are often used as jewelry metals; but attempting to sell white gold as palladium would constitute criminal fraud in most parts of the world, including the EU.

I'm not a metallurgist, I just use the stuff ; ) I have looked at the explanations about steel crystallization at different temps, and at precipitation/age hardening (specifically sterling silver) and can make them work. Sorry if the white gold alloy info was misleading-the primary additive to the (yellow) gold has been nickle, but, due to allergies and leaching problems, palladium is replacing nickle in the EU.

Well, I guess we're both really saying the same thing - how a given piece of metal is going to respond to quenching depends on the metal exact composition and heat treatment/cold work history.

Sorry to jump all over you like that, but I am a metallurgist* - and I'm  persnickety enough that I just couldn't let those statements stand when they  seemed to me to be so potentially misleading.

(*Ironically, I don't use the stuff that much - my machinist and smithing skills leave a great deal to be desired.)

I see what you mean about the white gold; and I hadn't known about European jewelers replacing nickel with palladium in white karat golds - thanks.  :)

I'm one of the few people who like to be corrected-at least when more information is provided : )


Yes, they do have different thermal expansion (thanks, Wikipedia!): gold is 14.2 um/m/K, where silver is 18.9; that's nearly 30% different.  Sounds to me like after-the-fact inlay is best.

Another nice theory shot down by inconvenient facts :-)

Yes, I think I'm going to carve the letters as recesses in the silver and then inlay gold wire (or perhaps gold leaf), possibly using the keum-boo method suggested by RavingMadStudios. And yeah, isn't that always the way? Theories are often so beautiful before they get ruined by the facts =P Thanks for all of your help =]

Make sure the silver is harder than the gold when inlaying-or cheat and use some of the Precious Metal Clay (or even the painted on gold accent color or plating...).

Depending on exactly what you're going for, the easiest solution might be electroplating the gold onto the silver. Or possibly inlaying gold wire into recessed letters, or maybe soldering separately cast gold letters onto the pendant. Or making the gold pieces and embedding them in the wax model so that they are held in the mold by the investment when the wax in burned out, then incorporated into the silver piece via inclusion. You might also Google "keum-boo" and "depletion gilding".

As I said, it depends on what you're going for.