Ultramarine. One pigment, many qualities

Natural ultramarine was and it’s still today a very expensive pigment. The synthetic version of this pigment is called French Ultramarine and was developed in the nineteenth century. It almost replaced the super expensive natural ultramarine in the artists’ palette . This post regards only the natural pigment. I started this little experiments since I had a little issue with some natural ultramarine bought from Kremer Pigmente and I ended up knowing much more about the actual production of this pigment and its many qualities.

This is the image that brought me to ask Dr. Kremer to send me more samples of their ultramarine varieties. Ultramarine is supposed to be IR transparent while the Kremer 10510 came out to be pretty IR absorbent.

 

Indeed, in order to save my money I bought a medium quality natural ultramarine (Kremer product code 10510) and it wasn’t as much infrared transparent as it was supposed to be, see my post on historical pigments. So, I asked Dr Kremer and, as usual, he was so kind and proactive to send me samples of all their set of natural ultramarine pigments.

With all the set of Kremer different quality ultramarine pigments I could run some little experiments. The goal was essentially to evaluate the quality of the different products. Indeed, Kremer sells different quality of ultramarine pigment starting from 8 euro/10gr until 200 euro/10gr.

Natural Ultramarine, varieties and quality

Natural ultramarine is made from the semi-precious stone, lapis lazuli which is a mixture of the blue mineral lazurite and colorless calcite.

At Kremer the ultramarine mineral comes in different qualities. The Afghan people can differentiate at least six different qualities!  But for producing the pigment,  at Kremer they need just to differentiate among three qualities of the raw material.

Cennino Cennini describes the method to separate the precious lazurite  from the ground powder, which contains calcite, pyrite and other impurities. At Kremer, I was told, they use now a high tech process for eliminate the pyrite. Though, pigment 10530 is still produced using the Cennini’s method and the pigment is more expensive because this process  is a lot of extra work which, at Kremer point out,  is all done by hand!  All the other Kremer ultramarine pigments use modern separation methods. Cennini instructed that in order to separate the blue lazurite from the colorless calcite and have a strong pigment, the lapis lazuli stone is grounded and the powder is incorporated into a mixture of melted wax, resins and oils and wrapped around a cloth. Blue particles of lazurite collect by settling at the bottom of the vessel while calcite and pyrites mostly remain in the mass.

As in the Cennini’s time, the residue material in the wax-resins-oils was still too precious to be thrown away and indeed it was still used as a much cheaper pigment called ultramarine ash (Kremer 10580). This pigment has the higher impurities content. Other than anisotropic minerals, such as calcite, it has also the highest content of pyrites.

Compound microscope, mounted slide, 10X. Ultramarine ash (10580) is the residue of the Cennini’s method. It is rich in anisotropic impurities and pyrites. Pyrite is opaque in reflected, transmitted and, consequently dark in crossed polars light.

The chemistry of ultramarine and its impurities is actually a bit more complicated. Indeed, beside the pyrite, most of the impurities are slightly blueish or grayish colored complex silicates the nature of which is not yet fully determined.

Other than 10530, the Cennini’s method ultramarine, Kremer has another high quality ultramarine pigment, crystalline ultramarine (10540). Indeed, there is one quality in Afghanistan which is lazurite in crystals and doesn’t contain any pyrite and only very small amounts of impurities.

Compound microscope, 20X. unmounted sample. The crystalline variety boosts deep blue colored lazurite grains.

Essentially, evaluating the quality of ultramarine means evaluate the lazurite/calcite ratio. Since this is one of the most important and widely used pigments in art history there is plenty of literature [1, 2, 3, 4].

Testing ultramarine with Multispectral Imaging

I chose to analyze three representative natural ultramarine Kremer products: 10500 (low quality), 10520 (good quality) and 10530 (Cennini’s method). 10500 it’s the lowest quality and goes for only about 8 euro / 10 gr. 10520 it’s described as good quality and it’s priced at about 76 euro / 10 gr. While 10530 is the one produced with the hand-made Cennini’s method and it’s therefore really expensive: 200 euro / 10 gr.

I did take an infrared image of the samples swatches (gum Arabic) and the Cennini quality looks more IR transparent, as expected and actually documented in early Italian paintings [5].

The finest ultramarine quality (10530) is the most IR transparent, as expected.

Testing ultramarine with Polarizing Microscopy

Though, once I had all those samples I wanted to have a bit more insight. The difference between the lowest and higher quality roughly resides in the ratio lazurite /  calcite. Polarized microscopy can easily differentiate between ultramarine (isotrope) and calcite (anisotrope). In this post I gave some indication and literature on how to identify pigments by polarized microscopy. I mounted the three selected pigment varieties in microscope slides. In the examination by crossed polars light with my normal (compound) microscope ultramarine looks dark, while calcite looks bright.

Compound microscope. Objective 20X microscope mounted slides. Only a small area can be imaged with a microscope, so there is little we can conclude on the average size of pigment grains and lazurite/calcite ratio. Lazurite mineral looks dark in crossed polars light while calcite is bright.

Testing ultramarine with Polarized USB microscopy

The high magnification of a compound microscope doesn’t allow an average observation, since we can image just a few grains at a time. So, in order to document a bigger surface I used the polarized USB microscope I introduced here.

I  observed the selected samples both in Reflected light and Crossed Polars Light. The decreasing calcite content is suggested by the reduction of bright particles in the crossed polars images, as well as the decreasing and homogeneous size of lazurite grains in the Cennini’s method pigment (10530). A similar small and homogeneous ultramarine pigment was observed in an Perugino’s altarpiece [6].

Polarized USB Microscope. Reflected light images of the three pigments mounted on a microscope slide. The Cennini’s method pigment (10530) has smaller and more homogeneous grains size, as actually documented in Perugino’s paintings [6]. Polarized USB Microscope. Crossed polars light images of the three pigments mounted on microscope slides. The bright spots are anisotropic minerals, presumably calcite, ultramarine main impurity. 10520 is higher quality than 10500 and indeed the anisotropic mineral content is much less. Cennini’s method ultramarine (10530) seems to be even more pure.

References

[1] A. Roy (editor) “Artist’s Pigments Vol.2: A Handbook of Their History and Characteristics”   National Gallery of Art, Washington, pp 37-65.

[2] R. J. Gettens, G. L. Stout   “Painting Materials. A short Encyclopedia” Dover Publication, pp.163-167, 1966.

[3] R. D Harley “Artists’ pigments, c. 1600-1835” Butterworths, pp 43-46, 1970.

[4] M. Doerner “The materials of the artists and their use in paintings with notes on the techniques of the old masters” pp 78-80, 1949.

[5]  C. Hoeniger “The identification of  blue pigments in early Sienese paintings by color infrared photography” JAIC 1991, Volume 30, Number 2, Article 1 (pp. 115 to 124).

[6] D. Bomford, J. Brough, A. Roy “Three Panels from Perugino’s Certosa di Pavia Altarpiece” National Gallery Bulletin, volume 4 1980.