Period Silver Stain Trials

Finally, an update on my silver stain research. I'm entering this into A&S, and have enough information and pictures to make a worthwhile blog post.

I'll skip the long train of how I found my sources, following reference to reference until I narrowed them down. I found seven recipes for silver stains. One ended up being a poor translation of another. One was incredibly complicated and is intended as a ceramics glaze; while the recipe would probably work on glass, I really can't call it a glass stain.

The recipes and my notes follow:

Mappae Clavicula
This is a compilation of recipes for a tremendous variety of crafts. The core of the text was compiled around 600 AD. Versions have been found from 800, 1130s, 1200s, and a variety of fragments of the text.

Amongst other recipes for colored glasses is this:

"For staining [glass] sapphire, silver and sulphur should be set on fire together; afterwards from 2 pounds of clear mosaic glass and 3 ounces of the above-mentioned sapphire cooked together, you will make jacinth stones."

I initially missed this when I read my copy of Mappae. I didn't want blue glass. It wasn't until I read "European Stained Glass Around 1300 - The Introduction of Silver Stain" in which Meredith Lillich explains that in period people knew "sapphires" to be both blue and orange varieties of the same stone. Although this appears to be a recipe for a pot metal orange she provides a compelling argument that it may have been used as a stain, and that the furnaces available at the time were capable of reaching the appropriate temperatures.

I purchased a few small vials of powdered silver for my experiments. I also bought 2 pounds of 99.5% pure sulfur prills (pellets), apparently used for livestock typically. I ground the sulfur to a powder in my mortar and pestle. In the absence of proportions, I mixed them equally.

I wasn't sure what "set on fire together" should mean precisely. I put them in a pan I was planning on throwing away anyway, and set them on low heat. Holy cow! The sulfur melted quickly and the whole thing became a black tar-like mess. I couldn't figure out precisely how I was to use it, and decided I would grind it back into a powder.

I believe the sulfur and silver are reacting to create silver sulfate. I'm not sure what reaction would happen alone that couldn't happen in the kiln, so I went ahead and made a second sample that would not be heated until in the kiln.

In my previous post I tested pure silver sulfate and the effects were apparent, but not this strong!

Slide 1: "cooked" MC formula. Slide 2: "raw" MC formula. Fired to 1000 degrees.

A closeup of the "raw" MC formula, 1000 degrees. 

This formula created a beautiful orange stain, exactly as the formula described and Lillich predicted. I believe the irregular coloring has more to do with my crude preparation than with the mix itself.

The Lapidario of Alponse the Wise (Alphonso X) of Spain

The Lapidario is the text by which Lillich (and Ken Leap, possibly from her research) speculate that silver stain truly became known in the West. It is a text describing stones and astrological properties. 

The stone "Ecce" is given it's own section. Lillich explains the text in depth. An English translation of the recipe:

"The treatise also describes a stone called ecce, which was used in glassmaking, saying that it was found in Spain, “in a mountain, not of great height, which overlooks the town of Arraca, and is called Secludes. And the stone is of an intense black colour, spotted with yellow drops. It is shiny and porous, brittle, and of light weight …; and if it be ground up with honey, and the glass be smeared with it and submitted to the fire, it dyes the glass of a beautiful gold colour, and makes it stronger than it was before, so that it does not melt so readily, or snap asunder with such ease.”

The Arts and Crafts of Older Spain, Volume II (of 3), by Leonard Williams

Lillich explains that the names given are incorrect and explains this is actually the city Guadalajara. She identifies the stone as silver antimony sulfide, or pyrargyrite (Ag3SbS3). This is known as red silver.

Further research on my part shows that this is not the exclusive mineral source. Proustite is a mineral that is visually indistinguishable from pyrarygyrite. They are isomorphous, meaning they have the same crystalline structure and essential composition. The difference is that where pyrargyrite has antimony, proustite has arsenic. The test needed to distinguish the two substances was not developed until 200 years post period. 

I purchased samples of both rocks. Further research and pleading from friends turned up that antimony is much safer to powder and heat up. Even though I was using tiny amounts of the mineral and it was VERY unlikely to be dangerous, I used proustite for my test.

I smashed the red crystals into what I initially thought was a sufficient amount of good size. The powder stained quite well, but my sample is very small and isn't visually impressive:

Slide 3: Lapidario formula featuring proustite, fired to 1000 degrees. 

The Marciana Manuscript

This painting text comes from the early to mid 1500s. The formula given is:

"If you wish to make a beautiful yellow colour which may penetrate into the glass, grind some silver leaf with a little honey and water, that it may hold together; then wash it in water with the fingers until it is well purified, in the same manner as powdered gold is treated. Distemper this silver with gum water made with soft water, dry it, then heat it in a furnace as before, and it will become very beautiful, &c"

I took silver leaf and tore it up a bit before starting to grind it in my mortar and pestle. My initial thought is that the honey is there to help give the leaf some body and grind it up. It makes sense that you would then wash it off. I used gum arabic and water to apply it to the slide.

This stain ended up being tricky, and I tried it three times. It's a known fact that stain takes differently with different glasses, and that you have to experiment with each glass you want to use. I also knew from Leap's book that pure silver requires higher temperatures than silver salts 

The initial slide with this formula, fired to 1000, came back clean. The silver did nothing.

Slide 4: Marciana formula, fired to 1000. Aka, clean slide!

I had similar difficulties with the last two slides, and decided to try again. I fired them to 1300 degrees.

Left slide: Marciana formula at 1300 degrees....

It definitely stained, you can see the pretty yellow color in the picture (that's pure silver leaf underneath, not gold leaf). Unfortunately at this temperature the silver leaf fused to the glass and I haven't been able to remove it. So, I tried again, at 1150 degrees. The silver didn't stain.

I may continue to try and get this to work with this glass. It MAY end up being that this glass gets soft at too low a temperature for any other result, however. I won't waste much more of my materials chasing it down with this particular glass, though. It appears when it works it will be a beautiful pale yellow.

Brussels

The Brussels manuscript is actually slightly post-period, from the early 1600s. It was very easy, however, so I included it:

""The yellow is made with silver, copper, and a little yellow ochre, the whole ground up together on a plate of copper.""

Hah. Easy. HAHAHA. Foolish glass geek.

This formula is the second slide in both of the two preceding pictures. You will notice it makes a beautiful color as well. Unfortunately the minerals fused to the glass in most of my samples. 

Bottom center: Brussels at 1150. Above it, Brussels at 1300.

Center/slide 5: Brussels at 1000. Clearly staining the edges!

This stain is proven to work, but I'm having difficulty getting the mineral residue to come off the glass. Lower temperatures for longer time makes sense, seems to be how it would have been done in period (they certainly didn't have an electronic controller on an electric kiln!) That should allow the stain to set in the glass without the glass getting to tack fusing temperatures.

Antonio da Pisa

I was able to try this recipe thanks to Mistress Kirsten allowing me to use her copy of the expensive and rare French translation (there is no English translation of the original Italian). My translation:

"The yellow color

To get the yellow color, take the fine filings of money, that is to say, the Venetian money and grind this on a chip of porphyry, until it is fluid like water, then when you begin to paint, apply it on the white glass where you want it to become yellow, and mix it in the least with the liquid egg tempera."

I collect ancient Roman coins. There are millions of them out there, they are very common. Many of them are little better than slugs, and if they hadn't been found with better samples you would not know they once had any value. They are also very cheap. On that basis I initially considered buying a Venetian coin and grinding it as described. Then I discovered Venetian coins are not as common as the Roman ones, and if I were to do as planned I would likely be destroying a valuable historical artifact. Instead I researched further and discovered that a common coin had a composition of 1 part silver to 6-to-9-parts copper. I mixed a batch of stain in those proportions, and gathered egg yolk for tempera. I've never worked with tempera before, it makes a wonderful paint/stain base:

A great many painters are smacking themselves and saying "Well, duh." 

Some Venetian coins had much higher ratios of silver. Because of that I doped the mix with more silver. In the picture above the top compound has considerably more silver than the one below. Knowing that the visual differences might not be as obvious after firing, I wrote an "L" ("lower") into the original mix.

Unfortunately, as with the preceding two, I had issues getting this to fire at a temperature that didn't fuse the metals to the glass. BUT, it DID stain!

Ignore the thermal shock cracks, I got impatient around 400 degrees.. Slide on the right, da Pisa at 1150

Questions/Lessons Learned

My six samples, drying before being fired.

All six of the period formulas worked as a silver stain. This particular glass made it difficult to show some of them in all of their glory, however. Interestingly, to me, the earlier formulas (Lapidario, Mappae) were the easiest. They make or use silver salts which, if I understand Leap correctly, fire at a lower temperature. The later period mixes (Marciana, Brussels, da Pisa) rely on actual metallic silver and are more complicated. However, I think they give a yellow color instead of the orange/ambers of the earlier mixes. 

Many glass recipes I see suggest a methodology of long slow cooking which would be impractical in my electric kiln but make sense in a clay and wood-fired oven. That may help the later period formulas. I would love to try some test slides if I ever have access to, or make, a period glass kiln. Something I am mildly considering at Pennsic this year.

For sake of discussion, the "mistaken" formula comes from Timothy Husband's book cited in my previous post on silver stain. It seems clear to me he is quoting a translation of the Lapidario, but identifies the wrong rock:

"antimony sulfide and silver in a liquid suspension, diluted 1 part to 6-10 parts by some form of clay (brick dust, ochres, etc)"

Antimony sulfide is called stibnite. I believe this has been confused with pyrargyrite, however Husband didn't cite that passage or explain the source. Later in the chapter, however, he cites a translation of the Lapidario, suggesting that might be the source of his recipe as well. Antimony sulfide and silver is, essentially, pyrargyrite. I purchased a sample of stibite initially, before I read Lillich's text on silver stain. I may still try "stibnite and silver" in the future.

It may be after A&S season, but I am going to continue to work on these formulas, probably with other glasses and temperatures and with finer grinding of some of the minerals. I am pleased to know, however, that I have an immediately workable orange stain that is BEAUTIFUL and that I can give out at my silver stain class at Pennsic this year.