Discussion of whites [Archive] - Artist Forum

: Discussion of whites


M Winther
05-21-2016, 11:46 PM
Zinc white has very fine properties and is arguably the best white. But the dire facts about zinc white ought to be recognized by all oil painters. According to the article "Zinc White: Problems in Oil Paint" (2014, here (http://www.naturalpigments.com/art-supply-education/zinc-white-oil-paint-color/)), paints containing zinc oxide become extremely brittle in as little as three years. Since most paints that are marketed as "Titanium White" contain zinc oxide (to prevent yellowing), this poses a big problem. The scariest thing is that it causes delamination(!) if the canvas has been primed with acrylic gesso, which is the case with most modern canvases. In the test, the paint layer containing zinc oxide could with little effort be peeled away from the acrylic ground.

To solve the delamination problem one could use Gamblin Oil Painting Ground, which is a titanium white alkyd. The text on the can says: "Hint: Add a layer of Gamblin Ground to canvas pre-primed with acrylic gesso to make sure canvas is properly sealed or to make a smoother surface." Alternatively, one could add a layer of Griffin Titanium White (alkyd) or Holbein Fast Drying White (alkyd + poppy oil), which are two of the few pure titaniums around. Since neither of these titanium products contain zinc, one can without much risk paint with one's regular whites on this titanium alkyd surface, because alkyd forms a very strong chemical bond with oils.

Pure titanium white has two problems. It tends to create chalky tints and, unlike zinc white, it is prone to photo-yellowing. This is why many brands of "titanium white" is really mixed with up to 20% zinc white, which is photo-resistant and also the whitest of whites. Yet, somewhat surprisingly, Griffin Titanium White (alkyd) fares well in the "The White Test...5 Years in the Making" (here (http://blog.jonathanlinton.com/2010/05/white-test-5-years-in-making.html)). The yellowing is negligible. (I find this inexplicable. Is it really pure titanium?)

So that's one of the reasons why many painters still use lead white, despite the fact that lead is a hideous poison. However, Holbein has invented an alternative, namely Ceramic White, which is made of strontium-titanate. I own a couple of tubes that I ordered from Rakuten, Japan. Its characteristics are similar to lead white and it does not create chalky tints. Strontium-titanate is entirely inert and won't cause any problems.

Alkyd mediums, and alkyd whites, have the advantage that they retain a degree of flexibility, which would serve to diminish the cracking problem. So a titanium-zinc alkyd would be better than a titanium-zinc oil. However, when using alkyds in oil painting one must take care not to paint a fast-drying alkyd layer on a slow-drying oil layer, because this is likely to cause cracking, despite the fact that the alkyd layer has better flexibility.

Whether it is ground in oil or alkyd, zinc oxide never becomes chemically inert. It continues to induce changes in the paint layer and, among other things, produce saponins. The linked article spells it out frankly:

"Should artists toss out their tubes of white oil color containing zinc white? If artists are concerned about the long-term prospects of their paintings, the answer is affirmative. Does this mean that zinc white should be removed from the lists of pigments suitable for serious artistic use? Maybe. For the time being, it should only be used in paint where the objective is not permanence."

Leo Mancini-Hresko (White Test 2013, here (http://leomancinihresko.com/white-test-2013/)) writes:

"People have written for years about the great tensile strength that Lead White gives your painting, and in modern days folks been talking about the dangers of Zinc White delaminating. A simple paint test like this is a great way to check a paint's durability. Cracking is a big concern if you want your paintings to last, especially if you like to paint on stretched canvas.
Five or six years ago, before I had ever tested Zinc, I bent a piece of canvas that had a swatch of Robert Doak's suspiciously light and bright 'Lead White' on it. Not only did the swatch break and fall off the canvas but it hit the floor and broke into a million pieces. I later made the Zinc connection, and have since included Zinc on all my white tests. I also stopped using Doak's white.
[...]
The above sheet had a couple of Zinc Whites and Zinc/Titanium blends on it. I bring these sheets occasionally when teaching, and after bending it a couple times to demonstrate none of the Zinc stuff remains on the canvas. Pretty shocking, but important to remember that people don't often go around bending paintings. It will take some pretty significant mistreatment to get zinc to fall off a stretched painting. Philip de László reportedly used Zinc for all his impastos, and his paintings that I've seen are in great shape. Personally, though, I avoid the stuff entirely." (here (http://leomancinihresko.com/white-test-2013/))

Mats Winther

TerryCurley
05-22-2016, 07:26 AM
This is a lot to digest.

Right now I'm not too concern about the longevity of my paintings, I just want to make them look as good as I am capable of and hopefully become more capable as time goes on. I'm only in it for the fun.

M Winther
05-22-2016, 08:21 AM
That is a common attitude, I believe. But amateur oil painters should take this problem seriously, otherwise their paintings will look like this in a hundred years. It is kind of tragic. The alternative is to go over to acrylics, which is perfectly archival, it seems.

http://s32.postimg.org/el0812dkl/flaking_paint1.jpg
http://graysrestoration.co.uk/oils-on-canvas/

Below is a painting from the Altamira cave, some 20,000 years old, or so. It's in perfect condition. So our modern technique is worse, which is embarrassing. Many Picasso paintings are strongly affected by cracquelure. In certain Matisse paintings, the yellow has turned white.

http://s32.postimg.org/emq711jqt/Altamira_Bison.jpg
By Rameessos - Own work, Public Domain, https://commons.wikimedia.org/w/index.php?curid=5569986

Mats