Gold Forms A Substitutional Solid Solution With Silver

Working With Gold-Silver Solid Solutions

The Au-Ag system is one of the simplest phase diagrams you'll encounter, but that simplicity hides some genuinely annoying behavior if you're trying to cast or solder these alloys without thinking about what's actually happening. Gold and silver have essentially identical atomic radii—both sit right around 144 picometers—and they share the same FCC crystal structure, which is why they dissolve into each other completely across the entire composition range. This means you get a single-phase solid solution from pure gold all the way to pure silver, with no intermetallic compounds, no eutectic, and no peritectic reactions to confuse things. When gold forms a substitutional solid solution with silver, individual silver atoms simply replace gold atoms in the FCC lattice and vice versa. There's no long-range ordering, no clustering at room temperature, and the resulting alloy is completely homogeneous if you cool it slowly enough. The melting point varies linearly between the two pure components—gold melts at 1064°C and silver at 961°C—so an alloy in the middle will melt somewhere in that range, but not exactly in the middle. The liquidus and solidus lines bow slightly toward the lower-melting component, which matters more than most people realize. I spent a couple of days last year troubleshooting a batch of 14-karat rose gold that kept cracking during cold working. The issue wasn't the copper content—it was that the casting had solidified through a two-phase region during cooling, and the segregation was subtle enough that a standard spectrographic analysis didn't flag it. The fix was a homogenization anneal at 900°C for four hours followed by a water quench, which dissolved the compositional gradients before I tried any mechanical working. If you're casting anything in the 30-70% silver range, don't skip the homogenization step. The difference between a workable ingot and one that tears apart at the rolling mill is usually just whether you gave the atoms enough time to shuffle into uniform positions.

There's also the color shift to consider, and it's not as straightforward as you'd think. Most jewelers assume that adding silver to gold just lightens the yellow, but the relationship isn't linear either. A 50-50 Au-Ag alloy isn't simply halfway between gold yellow and silver white in appearance. The reflectance spectrum changes in a way that makes mid-range compositions look slightly greenish under certain lighting conditions, which is why higher-silver jewelry alloys almost always include a small amount of copper to neutralize that tone. I've seen people spec out 18k white gold with only silver as the whitening agent and then wonder why the final piece looks like pale champagne instead of actual white gold. The copper addition of maybe 2-3% is what makes the difference, and it's not negotiable if you want a neutral color.

Phase Diagram Behavior and Processing Implications

The complete miscibility of gold and silver means you don't have to worry about phase transformations during heat treatment in the way you do with systems like Cu-Zn or Au-Cu. You can anneal at any temperature below the solidus and stay in a single-phase region, which makes this alloy system forgiving from a processing standpoint. But that doesn't mean it's easy to work with. The main practical issue is that solid solution strengthening in the Au-Ag system is modest compared to other precious metal alloys. A 50-50 Au-Ag alloy might reach a hardness of around 120-140 HV in the annealed state and climb to maybe 200-220 HV after significant cold work, which is nowhere near what you'd get from something like Au-Cu or Au-Ni systems. If you need hardness for a wear application, silver alone isn't going to cut it. Another thing people underestimate is the effect of silver on gold's corrosion resistance. Pure gold doesn't oxidize at any practical temperature, and it's immune to most acids. Add enough silver and you've introduced a metal that readily oxidizes and dissolves in nitric acid. This is the basis of the old parting process, but it also means that any Au-Ag alloy with significant silver content will patina and tarnish over time in certain environments. The tarnish isn't the sulfide darkening you get from sterling silver—it's more of a surface enrichment phenomenon where silver selectively dissolves or oxidizes, leaving a gold-enriched layer behind. It's generally superficial, but if you're selling jewelry and the customer starts seeing a dull gray surface after a few months, that's what's happening. Welding and soldering these alloys also deserves a mention. The flux chemistry you'd use for pure gold won't necessarily work optimally for high-silver content Au-Ag alloys because the silver oxide formation changes the surface tension and wetting behavior. I found this out the hard way when a client brought me a repaired gold-silver alloy frame that had been soldered with standard gold solder and flux. The joint looked fine visually but failed under a gentle bend test. Switching to a borax-based flux with a slightly higher fluoride content and using a solder with a marginally lower melting point resolved it. The joint held for years after that. Small detail, big difference.

Compounding and Casting Notes

If you're making these alloys from scratch, the compounding process is deceptively simple. Melt the gold, add the silver, stir, cast. But there are a few things that go wrong if you're not careful. Silver has a lower density than gold—10.49 g/cm³ versus 19.32 g/cm³—so if you're adding bulk silver to a gold melt, make sure it's fully dissolved before you pour. An undermixed charge will segregate during solidification even though the system is theoretically completely miscible, because convection currents in the melt aren't strong enough to maintain homogeneity in a large pour. I've seen ingots where the center was nearly pure silver and the outer shell was much higher in gold, and it took three remelts and extensive homogenization annealing to fix it. The oxidation behavior during melting is another factor. Silver oxidizes more readily than gold, and while the silver oxide decomposes back to metal at melting temperatures, any prolonged heating in air will cause surface oxidation and dross formation. A reducing atmosphere or a cover flux helps, though most people working with these alloys just melt quickly and move on, which is usually sufficient for small batches. For larger productions, I'd recommend a controlled atmosphere furnace. The equipment cost is higher but the yield improvement is noticeable—you're losing less silver to oxidation and dross, and the compositional accuracy is much better. One last thing that comes up occasionally: the thermal conductivity of Au-Ag solid solutions decreases as you move away from the pure endpoints. Pure gold has a thermal conductivity of about 310 W/(m·K), and pure silver is even higher at roughly 430 W/(m·K). A 50-50 alloy drops to somewhere in the 250-280 range, which matters if you're using these alloys for electronic contacts or thermal management applications rather than jewelry. The drop isn't dramatic, but it's real and it follows the general rule that solid solutions conduct heat worse than pure metals because the atomic disorder scatters phonons. If thermal performance is a design constraint, you need to account for it explicitly rather than assuming the alloy behaves like whichever component you're more familiar with.