MITRA Forum Question Details

Really fading is not the issue in this instance. Increased transparency of oil films over time is a result of a couple of factors but primarily it is the conversion of certain pigments, primarily lead white and zinc white, into metal carboxylates (more commonly known as metal soaps). This occurs because of the interaction of fatty acids and mobile metal ions in the paint film.  This same can happen with other pigments but these are the most effected. In essence, basic lead carbonate has a refractive index of 1.94 and 2.09, which makes it relatively opaque in linseed oil which starts out with a refractive index of 1.48 -1.49. Lead carboxylates have a lower refractive index and if enough of the lead white is converted the film become less opaque. It is believed that this mechanism is exacerbated when the film is exposed to moisture and heat.

Another less important factor is that aged drying oil films have a slightly higher refractive index than do newer films. This may cause a very slight increased transparency but the above mechanism creates a far more pronounced impact.  

Given the above, I am not sure that your layering would really have any effect one way or the other. If it were an issue of fading the pigments at the surface would be far more sensitive than those below. We know this from cross-sections taken from historic paintings where the uppermost gradient of a fugitive color has little color but that same pigment below still retains its original hue.

What you describe is one of the reasons why it was always problematic to work on a dark ground. A transparent toning layer is probably a better choice but unless the highlights were applied quite heavily, they could lower in tone over time.

I do not know of any studies on titanium dioxide converting to a soap over time. Certainly, zinc oxide does and it appears to cause more deleterious effects.