Although a great deal of information is available on artists' pigments, knowledge of the oil binder and mediums has been much slower to develop. This is not surprising seeing that methods to investigate the ingredients from small samples (small enough to allow detailed analysis from actual paintings) have become available only in the last few decades. These new techniques have awakened great interest in the function of these materials.

In 1999,¹ I was involved in making historically accurate reconstructions of oil processing and paint mediums using recipes from artists' instruction books that had been published in Britain between 1808 and 1845. These oils and mediums were then used to make hand-ground paints using lead white pigments, umber, and vegetable black. One goal of this work was to discover what components contribute to defects commonly seen in old paintings, e.g. film-formation problems such as drying-craqueleure and wrinkling as well as the tendency of lead white paints to yellow in the dark and to grow increasingly transparent with age.

To ensure that all steps in this study followed historical practice, linseed oil² was freshly pressed from organically grown flaxseed using a custom-built oil press. The subsequent oil was treated with and without lead driers [lead (II) oxide (litharge), lead acetate, and lead subacetate], and with and without heat. The oil was also treated by washing it in water over a period of weeks. In some cases, untreated freshly pressed oil was used for comparison.

Lead white paint made with oil heated to 300^o C (top). Lead white paint made with fresh oil, drier added during paint grinding (bottom).

Oils and pigments were ground together using a granite slab (5 cm thick, 50 x 50 cm square) and a granite muller (8.5 cm diameter, 4.7 cm high); this equipment was necessary because grinding lead white requires substantial force. The paint was applied to several different substrates (artists' boards, black and white opacity charts, and Mylar) with different application techniques: a brush and a palette knife on the artists' boards, and a draw-down bar that deposits a thin film of uniform thickness on the opacity charts and Mylar.

It was evident, even in the initial stages of grinding, that the method of oil processing has a profound effect on the texture, consistency, and flow properties of the paint. This was especially true of lead white paints, and less so with umber and vegetable black.

Another interesting finding was the relationship between the oil content and the paint flow. One might expect that a paint containing less oil would be more stiff than a paint made with more oil, yet the most fluid lead white paints were made with the least amount of oil. The type of drier also had a striking influence, e.g. the addition of lead acetate to the oil resulted in a thick, fluffy, and sticky paint with a consistency not unlike whipped butter. In these examples the deciding factor in the rheology of the lead white paint was the oil processing method rather than the amount of oil present or the presence of other mediums.

When Megilp or Gumtion³ was added to paint, the characteristics of these gelled mediums generally overshadowed the effect produced by the oil processing method; the addition of either resulted in a highly manageable paint. Copal varnish, another 19th-century addition to paint, initially behaved similarly but it had a very short working time and tended to slump after application. Full details of the preparation of the oils and mediums and the working characteristics of the paints can be found in CCI Report No. 72894.⁴

These oils and paint samples have been part of several studies in the Netherlands to investigate their chemistry before and after artificial aging, and the development of transparency as the paint ages. At CCI, a project is underway to investigate the response of the lead white samples to light exposure and storage in the dark. Lead white paint is well known for its tendency to yellow in the dark and to bleach when exposed to light. A previous project demonstrated that this phenomenon still occurs in oil paintings that are more than 100 years old. By subjecting the lead white paint samples to cycles of dark and light and reading their response with a spectrophotometer, it is possible to determine what constituents lead to greater or lesser yellowing in the dark. Initial results indicate that temperature during heat treatment can play a role in this, e.g. oils heated to 300^o C yellow less in the dark than oils heated to only 150^o C. The choice of the drier and the amount used are also important in the yellowing behaviour of the paint.

It is anticipated that these historically accurate paint samples will not only contribute significantly to the understanding of artists' oil processing and mediums in the studies mentioned above, but they will continue to be useful to researchers far into the future.

1. This work was undertaken during a 14-week fellowship in Amsterdam with Molecular Aspects of Aging in Art, a 5-year project (launched in February 1995) funded by the Netherlands Organisation for Scientific Research. The object of this project was to develop a scientific framework for the conservation of painted art from the 15th to the 20th century on the molecular level. Art historians, restorers, analytical chemists, and technical physicists were involved.
2. Artists' linseed oil is expressed from flaxseed (linseed), and must be refined or processed to remove water-soluble components (the mucilage) and hasten drying. Mucilage can be removed simply by allowing the oil to stand for long periods (since it gradually separates out) or it can be removed by washing the oil with water or treating it with heat and/or driers (metallic compounds).
3. Megilp and Gumtion were popular artists' mediums. When these gels were added to paint it became quite liquid and flowing under the brush, but when the brush was lifted the paint was more solid and did not run. John Scott Taylor described the buttery effect of adding Megilp, explaining that such paints "keep their place in working, with a flimsy firmness that is perfectly delightful" (p. 33 in Taylor, J.S. Modes of Painting Described and Classified. London: Winsor & Newton Limited, 1890).
4. Carlyle, L. Molart Fellowship, Historical Reconstructions of Artist's Oil Paint: An Investigation of Oil Processing Methods and the Use of Medium-Modifiers. CCI Report No. 72894. Ottawa: Canadian Conservation Institute, April 2000.