Pigment Analysis by Optical Polarising Light Microscopy

This submission is a response to an existing submission: 
Your name: 
Ruth Siddall
UCL department: 
Earth Sciences

Object Retrieval: Enamel-painted toy car; c. 1960-63

Traces of paint occur on the car, but much has disappeared, however enough remains to indicate how it would have originally appeared. The car has a zinc chassis, which is painted with enamels; white bodywork and a red representing a plush, upholstered interior. Enamel paints are Twentieth Century paints which combine the properties of a high gloss finish with a flexible paint surface (so named from their resemblance to glass-based colours), ideal for coating the intricate morphology of the car. All paints are made up two main components; the pigment which provides the colour and the medium, which binds together the pigment articles and also defines the finished surface of the paint. In the case of enamels, the medium is an alkyd. Alkyds are made from a polyester modified by fatty acids, the latter providing the flexibility typical of these paints.

Flakes of paint were collected from both red and white areas, and each flake, less than a square millimetre in size, was divided into two aliquots. The first aliquots were used for a spot lead test. They were placed in separate Petri dishes and doused with acetic acid. A small crystal of potassium iodide was then added. If lead is present and abundant, an immediate and strong reaction occurs, forming bright yellow lead iodide. No reaction was observed with either of the samples. However a weak yellow stain developed after the samples had been left for ten minutes or so, indicating trace amounts of lead to be present.

The second aliquot was mounted in MeltMount™ with a refractive index of 1.662 on a standard glass microscope slide, covered by a cover slip. The samples were both observed using polarising light microscopy, on a Leitz Orthoplan Pol microscope, with a magnification of up to 1000 x, close to the optimum for optical microscopy. This technique may be used to identify the pigment particles present, it cannot be used to identify the binding medium,. A problem encountered with both paint samples was that the pigment particles remained bonded together as a flake, and did not disperse in the mounting medium as hoped. However, the edges of the flakes were sufficiently thin that individual particles could be identified.

The pigment particles are so small that they were close to the resolution of the optical microscope, ~ 5 µm in diameter. Individual particle shapes cannot be easily identified in such small particles, and typically they appear to be spherical. Particles were all the same size.

In plane polarised light (PPL), the pigments had a high relief, and therefore an 100 x oil objective was used to clarify the image. The overall colour appeared brownish, but due to the small particle size and high relief, this is likely to be an optical phenomena.

When viewed under cross polars, the paint flake appeared bright and white when observed under low power. At high power the individual grains are observed to be strongly birefringent with third order colours on Newton’s Scale.

The optical and physical properties of this pigment are typical of precipitated, synthetic titanium dioxide, with a rutile structure; “titanium dioxide white”.

As in the white paint, the particle size of the pigment grains was ~ 5µm and therefore too small to be identified. This is further compounded by the fact that the particles were coated with a strong, pink-red stain that provides the pigment colour. Under cross-polars, the sample glowed a strong vivid red.

This is a lake pigment whereby a red organic dye is combined with an inorganic white substrate. It is impossible to identify the substrate by optical means alone in this sample.

Titanium dioxide white pigments can exhibit two different and distinct crystal forms, which are analogous to the two commonly occurring mineral forms of titanium dioxide, rutile and anatase. Both forms were manufactured commercially on an industrial scale. The so-called chloride-process of manufacture was developed in the US in the mid 1950s, but this was not manufactured in Europe until the mid ‘60s. Rutile-type forms are produced by this process and represent the commonest type of titanium dioxide whites. Laver (1997) states that lead may be present in titanium dioxide pigments in trace amounts, up to 8,000 parts per billion (0.0008%) - this is pretty much insignificant and is unlikely to cause a risk to health. Titanium dioxide, rutile-type is well known in alkyd enamel paints due to its high opacity and good covering properties.

The red pigment is a modern synthetic organic pigment and I am afraid I have no idea what it is! It could be synthetic alizarin or another anthraquinone, or an azo red or…?

If lead is present, it is in trace amounts and not contributing to the pigment colour. The presence of lead was indicated by the spot test.

Where next?

Analytical techniques employed in organic chemistry would be required to confirm identification of the binder and also identify the organic red. Raman analysis might help here, and a set of reference spectra are available in Scherrer et al (2009).

Further Reading
Laver, M., 1997, Titanium Dioxide Whites. In: FitzHugh, E. W., Artists’ Pigments, Volume 3., National Gallery of Art, Washington & Oxford University Press., 295-339.

Scherrer, N. C., Zumbuehl, S., Delavy, F., Fritsch, A., Kuehnen, R., 2009, Synthetic organic pigments of the 20th and 21st century relevant to artist’s paints: Raman spectra reference collection. Spectrochimica Acta Part A 73 (2009) 505–524

Ruth Siddall, UCL Earth Sciences, October 2009.