Verdigris was a pigment commonly used in oil painting from the 15th to the 18th century, but which fell into disuse after that, when modern green pigments began to be developed that were more dependable and nonfading. Verdigris oil paint is well known to have many drawbacks which the Old Masters certainly suspected: they wrote that it was incompatible with many other pigments, even to the extent of ruining brushes that had been dipped into it once; and more importantly, that it has a tendency to turn brown or black under certain conditions which cannot be predicted reliably. However, even the masters might not have fully realized the extent to which verdigris darkened, since changes may have taken place slowly over a period of centuries. As evidence, there are later paintings of galleries (a popular genre in the 18th and 19th centuries) which show known Renaissance paintings with green landscapes that are now brown!
Not all verdigris glazes have darkened; the grasses and herbs in Van Eyck’s Mystic Lamb remain as bright as day they were painted (see detail below, showing the glorious copper resinate glaze still intact after 500 years).
But other works, even by the same painters, have darkened, sometimes ruinously. Below, St. Francis Receiving the Stigmata, sometimes attributed to Jan van Eyck: The grass and trees behind St. Francis were almost certainly not originally painted in such subdued tones. Gothic and most Early Renaissance painters almost never favored a subdued palette. But in this case the faded landscape is quite harmonious with St. Francis’ robes and the warm brown rocks.
Various recipes, from Theophilus onwards, promise an “eternal” verdigris which will not “die,” but these claims were probably overblown. And many of these recipes are now disputed by some modern art historians. The chief question is whether Renaissance painters such as Van Eyck used verdigris ground in cold oil, or whether they cooked the verdigris is a coniferous resin such as Venice Turpentine, dissolving it into a glaze referred to as copper resinate. Some modern art historians claim that recipes for copper resinate glaze were only intended to color metals, not for painting, and that van Eyck and contemporaries never used copper resinate, but instead made glazes out of linseed oil, or oil mixed with cold resin. The question has not been resolved.
With this controversy in mind, I decided to experiment myself and attempt to make verdigris pigment and then copper resinate glaze. To manufacture the pigment itself, I used a recipe from Theophilus for “Salt Green:” “If you want to make a green pigment, take a piece of oakwood, as long and as wide as you wish, and hollow it into the shape of a little chest. Then take some copper and thin it out into sheets of any desired width but of a length sufficient to span the [inside] width of the chest. After this take a flat pan full of salt, and, pressing the salt down firmly, put the pan in the fire and cover it with [glowing] coals for the night. Next morning grind the salt very carefully on a dry stone. Get some thin twigs and place them in the above-mentioned chest in such a way that two thirds of the cavity are beneath [the twigs] and the other third is above them. Smear the copper sheets on both sides with pure honey and sprinkle the ground salt on them. Then lay them next to each other on the twigs and cover them carefully with another piece of wood, fitted for the purpose, so that no vapors can escape. Next, in the corner of this piece of wood drill a hole through which you can pour in heated vinegar or hot urine until a third of [the chest] is filled; then block up the hole. Put this chest in a place where you can pile dung all over it. After four weeks pry off the lid, scrape off whatever you find on the copper, and keep it. Put [the copper] back again and cover it as above.”
I used small glass casseroles with plastic lids for my “little chests.” Otherwise I followed Theophilus pretty closely: after making a porous scaffold of sticks in the casserole and covering the copper sheets with honey and salt, which I ground to a very fine texture (ok, I used a blender) I laid them on the scaffolding, then I heated several different kinds of vinegar (Theophilus didn’t specify what kind of vinegar to use) because I wanted to investigate if the verdigris was different using different types of vinegar.
With that in mind I used apple cider, red wine, and white wine vinegars, along with urine, since Theophilus mentioned that too. I warmed each of the vinegars and the urine in the microwave, and then poured them each into separate casseroles and labelled them, then added the copper sheets with honey and salt, propping the sheets over the liquid on a scaffolding of sticks. Theophilus didn’t specify what kind of wood to use so I used small sticks from my garden. I then closed the casserole and, instead of putting them under dung (which I assume was intended to heat gently), I put them on my deck in Tennessee’s blazing midsummer sun.
After a few weeks, I could see that a green crust was building up on the copper.
However, when I tried to lift out the copper plates, I was appalled to see that in addition to the green crust, a thick, bright red pigment had oozed out of the copper beneath the salt green!
Apparently, the Made in China copper sheets I had purchased at Hobby Lobby contained something other than copper, possibly a red pigment. (I even wondered if I had accidentally made Minium! All I know is it stung my fingers when I was foolish enough to touch it barehanded.) At this point, I feared the whole project lay in ruins, and even thought about throwing the whole, probably toxic, mess out. There seemed to be no hope I would ever get the red pigment off the green verdigris. But I was too lazy to haul it to the special city dump for toxic materials, so instead just left the copper sheets lying in the sun to dry out.
After a few more weeks, I checked the copper sheets, and amazingly, the red pigment that had leached out had almost completely receded back into the copper, leaving the green pigment intact and almost untouched! I also noticed that the different vinegars had produced slightly different greens. The urine had produced less green but it was of a deeper color. But since the fiasco with the red pigment had deterred my enthusiasm somewhat, and there didn’t seem to be a whole lot of difference, I scraped the green off all the copper sheets and mixed it together in a bowl. And I was able to do this without getting any of the residual red pigment, which had dried into flakes, into the mix. It was almost too much to have hoped for.
I ground the fresh verdigris a little bit. Then I put all the green pigment into a glass container and covered it with fresh vinegar, stirred it, and let it settle.
As the vinegar evaporated, I stirred it, mashed up lumps, and let it settle more.
My plan was to let it evaporate and produce the distilled and purified verdigris written about by Merrifield. This process took a lot longer than I expected, a couple of months, but it finally (mostly) evaporated away until I had a smaller bowl of verdigris sludge. I attempted to grind it additionally and, although some reached a fine powdery state, some small clumps remained. There was also a mixture of colors, from paler to dark, although this might have had more to do with the fineness of the grind. The finely ground verdigris seemed to be paler.
When dry, I had about 1/4 cup of verdigris. It was probably a mixture of basic and neutral copper salts, along with copper chloride, but since I didn’t check the chemical composition I don’t know for sure. Probably the verdigris used in the Middle Ages and Renaissance was similarly mixed in composition.
At this point, I attempted to grind the verdigris to use as pigment, which proved more difficult than I expected. I never was able to get all the chunks and lumps out no matter how long I ground it. For grinding I used a glass pestle and a bowl, which didn’t work as well as I’d hoped. Next time I’ll use a glass muller. Cennini spoke of grinding verdigris in vinegar, which made sense, but that didn’t seem to help much either. The verdigris was surprisingly hard, and the honey I’d used made it somewhat slippery and gooey. Next time, I’m forgetting the honey and salt, since I couldn’t see why it made a better product.
Finally, I was ready to grind the verdigris into various media: glair made from egg whites; egg tempera medium made from egg yolk; linseed oil; and Venice Turpentine, a coniferous resin made from the European Larch tree. I painted samples of all these media ground with verdigris on a slightly defective wood panel gessoed with rabbit skin glue and marble dust. The results were varied. Some visible clumps remained in the various media. And it dripped a little.
I was not able to produce a glaze similar to the cooked resinate glaze with either cold resin or linseed oil.The verdigris ground in boiled linseed oil turned dark almost immediately, and remained gritty and opaque. The cold resin also failed to yield a really transparent glaze since the verdigris never really dissolved. It was much paler than the cooked resin glaze.
The glair was surprisingly good, although opaque and pale. I wondered if the honey had improved the verdigris for watercolor. On the other hand, verdigris in egg yolk turned into a translucent mayonnaise that did not appear to be suitable for tempera painting at all, although it was quite pretty.
I prepared a cooked copper resinate glaze by combining Venice Turpentine and verdigris and heating it gently. This produced a much greener, very transparent glazing medium, although once again, clumps remained. After a day or so, however, I noticed that the cooked Venice Turpentine medium had settled and thickened a lot and the color and texture had improved. Yet another fortuitous accident! Most of the pigment had sunk to the bottom of the bottle, and the resin was now darker, very smooth and glassy with no clumps remaining. When I tried it on the gessoed panel, it produced a beautiful, perfectly transparent, glassy green glaze.
I experimented to find modern substitutes for verdigris: I thought thalo green and viridian might be most similar, but they were much brighter to my surprise. I painted samples of both pigments in oil and also mixed with the Venice Turpentine. Viridian and a small amount of yellow ocher in Venice Turpentine was the most like the verdigris copper resinate and might make an acceptable modern substitute, although nothing seemed as transparent and perfectly grassy-green as the verdigris.
At this point, I entered the project in Stella Nova Arts and Sciences Fair (a medieval re-enactment arts event for those readers not in the SCA), and received a number of helpful hints from the judges.
I continue to observe the painted samples to see if they deteriorate. Thus far the verdigris dissolved in hot Venice Turpentine is still bright and glassy, although a bit of Christmas glitter is sticking to it now. Sorry. The verdigris in linseed oil is dark and opaque and seems to have darkened further and spread. The verdigris in glair is surprisingly fresh looking , although quite opaque and lighter than the other samples.
Trying to see if I could improve the ratio of undissolved verdigris in the copper resinate, I re-ground the verdigris in the cooked resin and managed, with a lot of elbow grease, to get the pigment ground into a much finer powder. I then recooked the Venice Turpentine with the pigment and walked away from the hot plate for a minute with almost disastrous results: it boiled up and turned a coppery brown! Ruined yet again, I thought, but keeping my cool, or being lazy, I just put it back into the bottle to let it settle. But my luck held once again. After a day or two, a miracle occurred: the coppery brown (which rather resembles the brownish verdigris on old paintings) settled in a layer between shades of various green and the glaze floating on top returned to its former green glory. Whew! If you look closely, below, you can see the brownish overcooked layer.
But still most of the verdigris remains undissolved, and the transparent copper resinate is really no greener. After re-reading one researcher’s results making copper resinate, I realized that I probably didn’t use nearly enough resin, although I followed the instructions in Merrifield. It’s not a matter of mixing, but of actually dissolving the pure pigment molecules into the resin. One modern researcher used the same recipe and had to add powdered rosin to the mix, and managed to dissolve more of her verdigris. I will try that next time.
Then more reading led me to a new conclusion: Noting the description of Venice Turpentine as a “soft resin,” and remembering that modern research hasn’t pinpointed exactly which resinous substance Van Eyck used anyway, I realized he wouldn’t have used a soft resin like Venice Turpentine. I suspect he used amber varnish to dissolve his verdigris. This might explain his (usually) superior results. Many writers on painting mediums of the masters have long suspected he used amber, anyway.
Per the judges’ recommendations, I’m using the copper resinate I’ve made thus far to paint several little samples of Van Eyck landscapes with underpainting and green glazes to show how it would have been used. Here is a little patch of Heaven from The Mystic Lamb which I plan to try to copy:
It shouldn’t be too hard but I think I’ve already done it wrong in my first layer:
Perhaps the underpainting had verdigris mixed into the lead tin yellow, instead of the dull green made from black and lead tin yellow I used. Don’t worry, I’ll figure it out. Sigh. (Note: I’ve been in my studio trying to re-create the scene, and it’s harder than I expected, as usual. I’ve wiped it off twice already. Extra long sigh.)
Plans this summer include making another batch of verdigris using pure copper sheets from Jerry’s Artarama, leaving off the salt and honey and simply using white wine vinegar. After that, I’m making some amber varnish, which ought to be a much harder resin, and dissolving the finely powdered verdigris in that.
The other day I measured an already gessoed panel in my studio, and something made me look up online the exact dimensions of The Arnolfini Wedding. I knew it. My panel is within a quarter of an inch. And the bride’s dress is glorious, copper resinate green. I’ve always loved that painting.
- “Changing Pictures: Discoloration in 15th – 17th –Century Oil Paintings, by Margriet van Eikema Hommes, Archetype Publications, 2004.
- “Lost Secrets of Flemish Painting: including the first complete English translation of the De Mayerne Manuscript, B.M. Sloane”, by Theodore Turquet De Mayerne (author), Donald Fels Jr. (author), published by Alchemist, Inc., Floyd, VA, 2001.
- “Aspects of Painting Technique in the Use of Verdigris and Copper Resinate,” by Renate Woudhuysen-Keller, from “Historical Painting Techniques, Materials, and Studio Practice,” preprints of a Symposium, University of Leiden, the Netherlands, 26-29 June 1995, Edited by Arie Wallert, Erma Hermens, and Marja Peek, Published by The Getty Conservation Institute.
- “The Materials and Techniques of Medieval Painting,” by Daniel V. Thompson, Dover Publications, Inc., New York, 1956.
- Perspectives on the Painting Technique of Jan Van Eyck: Beyond the Ghent Altarpiece, by Noelle L.W. Streeton, Archetype Publications, London, 2013.
- “The Secret Formulas and Techniques of the Masters,” by Jacques Maroger, Hacker Art Books, New York, 1979.
- “The Materials of the Artist and Their Use in Painting with Notes on the Techniques of the Old Masters,” by Max Doerner, Harcourt, Brace and Company, New York, 1934.
- “Artists’ Pigments, Volume 2, edited by Ashok Roy, 1993, National Gallery of Art and Oxford University Press, New York.
10. Medieval and Renaissance Treatises on the Arts of Painting, by Mrs. Mary P. Merrifield, Dover Publications, NY. 1967.
11. Theophilus: On Divers Arts, Translated from the Latin with Introduction and Notes by John G. Hawthorne and Cyril Stanley Smith, Dover Publications, NY, 1963.
12. Methods and Materials of Painting of the Great Schools and Masters, by Sir Charles Lock Eastlake, Dover Publications, NY, 2001.
“Exposure of the verdigris in linseed oil to air begins the process of turning it brown almost immediately” … and they said it was an unpredictable pigment, hmm…. http://www.grosse.is-a-geek.com/paintings_pigments-2.html (end of page)
Julia Morgan Scott says
Verdigris seems tricky enough that I hesitate to use it on an actual painting. I see you used malachite powder instead of making the pigment from copper and vinegar. Perhaps the browning I saw was just the color shift, but it seemed excessive to me in oils. Also, it was very difficult to get the pigment ground finely enough to grind in the oil without clumps, but on the other hand, the old masters paints were grittier than ours. A puzzle. I look forward to reading more on your website which seems very interesting.
Doing a school research paper on varnishes and am delighted to read this extensive experimentation! And you cite your sources too, thank you! Would love to know how this worked out with your planned adjustments.
Julia Morgan Scott says
I’m so glad you enjoyed my article on making verdigris. I tried making it again, but it wasn’t during the summer and it apparently wasn’t warm enough to make very much. I think I left off the salt and honey, too. My attempts to make oil paint with my verdigris were disappointing – the verdigris in oil has turned quite brown already! This seems to contradict current research on Van Eyck’s technique; modern researchers seem quite certain that his wonderful green was nothing more than verdigris in linseed oil. I believe if this was true, he surely knew how to create an isolating layer of some sort, perhaps amber varnish, but the researchers say there was no amber varnish involved. I don’t believe this but I’m not enough of an expert to figure out the solution. Perhaps my verdigris in oil wasn’t dissolved enough. Exposure of the verdigris in linseed oil to air begins the process of turning it brown almost immediately. I decided not to paint anything with real verdigris because of this. Obviously, some bit of knowledge is missing, and I don’t want to get so bogged down in research I stop painting! I would prefer to use a modern substitute, such as Thalo Green, which is wonderfully transparent and more stable.
Julia Morgan Scott
Thank you for the article, and the valuable bibliography.