My previous varnish, while perfectly suitable for most interior applications, degrades quickly outdoors. When exposed to the weather, even with a coat of paste wax, the varnish will begin to show wear in about four months, and will have substantially failed within a year. As this is likely due to the inherent water absorption of linseed oil, I decided to make an improved varnish using tung oil, which is known to be essentially waterproof.

   Due to its unique fatty acid profile, tung oil (from the seeds of the tung tree) dries faster and harder than any other natural oil in common use. Unfortunately, this is due to its rapid oxygen absorption, which greatly expands the volume of the oil and leads to significant wrinkling of the oil film. Worse still, attempting to pre-polymerize the oil by heating (boiling) it beforehand causes it to solidify instantaneously instead of thickening gradually. However, in a fortunate natural coincidence, this problem can be solved by the addition of pine rosin. When dissolved into tung oil during the heating process, rosin prevents the solidification of the oil, which itself then confers hardness and water resistance to the rosin in the final varnish film; the two make a perfect pair. To make such a varnish, I began with the following ingredients:

•  50g Raw Tung Oil   (from The Real Milk Paint Co.)
•  50g Pine Rosin     (from Diamond G)
• 0.4g Ferric Acetate (from wood stain, air-dried in a shallow glass tray)

   Preparing the varnish was as straightforward as possible; to start, the three ingredients were added to my large pyrex dish (the ferric acetate being ground into the oil, as usual), and heated to 220°C while stirring. I then took a sample every half hour by placing a drop onto cold glass, spreading it by hand, and placing it in the sun to dry. After three hours of heating, the drops had become nearly solid upon cooling, and the dried samples were only wrinkled in the thickest areas. I then removed the varnish from the heat, and once it had cooled to 100°C I added 100g of turpentine as a thinner, then filtered the warm varnish into a jar for storage. These steps can all be seen below.

   The resulting varnish is slightly thicker than my original rosin varnish, but is also somewhat slippery, so both are equally brushable. Furthermore, the amount of iron drying catalyst is the same, so both varnishes also have the same amber tint, which can be seen at the top of the page. This is where the similarities end, however. The tung oil varnish dries extremely quickly, becoming rubber-hard overnight indoors, and within an hour under direct sunlight. Full curing is likewise accelerated, with thin layers taking less than three days to reach final hardness. This hardness is greatly improved as well, and is similar to that of my asphaltum enamel. The varnish is also absolutely waterproof, and a sample dried on glass showed no change after being submerged in water for a day. The characteristic friction of the rosin is eliminated as well, and cured varnish films are slick to the touch. Finally, the one-step preparation allows for the rapid production of significant quantities of varnish, as needed for large projects.

   Overall, this varnish appears to be superior in every way to its linseed oil equivalent, but weathering tests will have to be conducted to confirm this, and to identify any potential faults. Primarily, my concerns are that the impermeability of the varnish may trap moisture in the wood, and that its relative hardness may cause it to crack as the wood expands and contracts. These were areas where the linseed oil varnish actually performed well due to its permeability and softness, with its failure being due to gradual wear rather than catastrophic peeling. As tests and projects using this new varnish begin to age, it will be interesting to see how it degrades, and what its failure mode eventually will be.

   While this varnish performs adequately when brushed on, I have found that it produces much nicer results when applied with a rag, much like linseed oil. If wiped on thinly it dries in four hours, and three coats will produce a finish similar to french polished shellac (but of course, totally waterproof). Due to the rapid drying time and ease of application, this is now my go-to method for finishing wood.

   Although the color of this varnish can be varied by changing the proportion of ferric acetate used, this comes at the expense of undesirable drying characteristics. A better option that I've found is to tint the varnish with oil paint. If a transparent pigment (having a very fine particle size) is used in small quantities, the color can be shifted somewhat within the region of warm and neutral tones, without the varnish losing its transparency and obscuring the grain of the wood. One historical pigment commonly used for this purpose is madder lake, but a more lightfast option is to use transparent iron oxides, which I've found produce excellent results.