Articles originally published in Chemistry in Britain

Restoring old masters: why do paintings turn yellow?
Making make-up in Ancient Egypt
Revealing ancient fats
Fatty Acids in schistosome parasites
Insects' own mothballs

Return to credits

Restoring old masters: why do paintings turn yellow?
Chemistry in Britain, November 1999

Removing aged varnish from treasured old paintings is a highly skilled operation, and yet current procedures are not based on any detailed understanding of structural changes in the material. Applied as a protective coating, varnish - which is derived from tree resin - may enhance a painting's colour at first, but it gradually degrades, and typically forms a cracked and yellow crust.

Prompted by how little was known about the chemistry involved, and with the aim of improving restoration procedures, Jaap Boon and Gisela van der Doelen, from the FOM Institute for Atomic and Molecular Physics in Amsterdam have been studying the ageing process in commonly-used varnishes.

Using a range of chromatography and mass spectrometry techniques, the scientists compared fresh varnish with fragments from over 50 paintings by Dutch masters, including Rembrandt and Van Gogh. Having confirmed that tetra and pentacyclic triterpenes are principal constituents of fresh varnish, they showed that the composition changes radically over time with the appearance of a diverse range of triterpene oxidation products and higher molecular weight compounds.

They conclude that degradation follows a progression of light-driven oxidation and cross-linking reactions, with the varnish passing through several distinct stages as it ages. "Initially," says Boon, "triterpene side chains and functional groups are oxidised to more polar derivatives. These either break down further to smaller, volatile compounds, or, most significantly, combine to form cross-linked polymers. We suggest that this cross-linked network, extending over time, gives very aged varnish its characteristic texture."

But what of the yellow discolouration? The scientists reasoned that if oxidised triterpenes are not responsible for the yellowing, then the polymers may be responsible. Support for this theory came from simulation experiments in which fresh varnish, in solution, was subjected to accelerated, light-induced ageing.

"We found that triterpenes undergo cross-linking reactions, producing larger complexes with a distinctive yellow colour", reported van der Doelen. She and her colleagues are now investigating "whether they have the same chemical structure as the substances which discolour aged varnish on paintings."

Varnish is removed using polar solvents, which easily dissolve the smaller triterpene oxidation products. The Dutch scientists suggest, however, that because the polymers are less soluble, thinning down the varnish layer in this way may cause them to concentrate on the surface of the painting. Given the polymers' contribution to discolouration, the practice of varnish removal clearly needs more detailed study.
Russ Clare
Return to top


Making make-up in Ancient Egypt
Chemistry in Britain, April 1999

While colourful adornment with cosmetics is an age-old practice, there is, perhaps, an assumption that sophisticated chemistry used in the preparation of beauty products is a relatively recent innovation. However, Philippe Walter and his colleagues at the Research Laboratory of the Museums of France [Laboratoire de Recherche des Musées de France] have now shown that wet chemistry was employed to make widely used cosmetics in Ancient Egypt as long ago as 2,000 BC.

The Egyptians made a huge variety of medicines and cosmetics from organic and mineral sources.  Several examples are housed in the Louvre museum in Paris, still in their original stone, ceramic, wood or reed containers.  Quantitative crystallographic and chemical analysis of samples dating from 2,000 to 1,200 BC  revealed the common, respectively black and white, lead minerals galena (PbS) and cerussite (PbCO3) mixed in a lipid base.

A much more surprising discovery, however, was the white lead chloride compounds, laurionite (PbOHCl) and phosgenite (Pb2Cl2CO3).  As Walter explains:  "these lead mineral oxidation products are only formed in the presence of carbonated and chlorinated water," which means that extraction is unlikely to account for their extensive use over at least eight centuries.

Chemical weathering as a source of the compounds in such well preserved powders and containers was also discounted by the scientists -  they found neither contamination by foreign cations, nor evidence from scanning electron microscopy (SEM) of surface damage to crystals.

With synthesis remaining as the only explanation for the chemicals' presence, the researchers examined recipes for laurionite and phosgenite recorded from the first century AD in Greco-Roman texts, including Pliny the Elder's Natural History.  To support their conclusion that such methods originated from much earlier discoveries in Egypt, they replicated a recipe for silver foam (PbO) and rock salt (NaCl) to produce laurionite which SEM showed had a morphology that was remarkably similar to the Egyptian preparations (Nature, 1999, 397 483).

"The chemical reactions to make laurionite and phosgenite are simple," says Walter.  "However, the recipes require many repetitive operations including daily filtration for several weeks, so they must have been quite difficult to develop.  Coupled with the cosmetics' complex formulations of lipids and minerals, it shows that chemical technology in Ancient Egypt was far more sophisticated than we had previously supposed."
Russ Clare
Return to top


Revealing ancient fats
Chemistry in Britain, February 1999

Revealing ancient fats Chemistry in Britain, February 1999 Traces of preserved lipids on ancient pottery can provide us with a fascinating insight into our ancestors dietary habits. By analysing solvent extracts taken from ancient cooking and storage vessels, Richard Evershed and colleagues at Bristol University's School of Chemistry, have revealed traces of olive oil, leaf waxes, beeswax, and animal fats - including, most recently, milk fat (Science, 1998, 282, 1478). Significantly, however, the researchers also found that oxidation products of unsaturated fatty acids were notably absent.

Commenting on this absence, Evershed said: "We predicted, from our knowledge of chemical changes in modern food processing, that low molecular weight components like dicarboxylic acids (diacids) should arise in the high temperatures and oxidising conditions of cooking pots and lamps, but, with greater solubility, they are probably leached by groundwater during burial".

Using gas chromatography of solvent extractable lipids in pottery from waterlogged and arid sites, the Bristol researchers have now confirmed this assumption. They found that, while oxidation products were absent from Neolithic cooking vessels found at the Chalain lakeside settlement in France, they were abundant in unglazed lamps from Qasr IbrÓm, in North East Africa. The lamps, notably, had never been exposed to water since their burial in the sixth century AD (Proc. R. Soc. Lond. B, 1998, 265, 2027).

However, the scientists also found that oxidation products are detectable, despite leaching. In the alkaline conditions for saponification, solvent insoluble residues of Chalain pottery released series of diacids and hydroxy carboxylic acids that had been covalently bound in the ceramic matrix. "This is a significant finding", explains Helen Bland, a member of the team. "The range of oxidation products should reflect the original lipids - nine carbon azaleic acid, for example, is most likely derived from a fatty acid unsaturated at C9 . Leaching of oxidation products could present a distorted impression of usage, but knowing the chemicals are bound in the pottery should enable a more complete story to be revealed."

Experiments that simulate oxidising conditions in pottery are now in progress to establish the reliability of the products as markers for unsaturated fatty acids commonly found in plant oils.
Russ Clare
Return to top


Fatty Acids in schistosome parasites
Chemistry in Britain, January 1999

The phospholipid bilayers of biological membranes form highly effective barriers. This is shown to good effect by the parasitic blood flukes, or schistosomes, which live for years in the blood of mammals and birds, protected from the host immune system by a unique double membrane, the tegument, that envelops the worm. Schistosoma mansoni infects humans to cause the disease schistosomiasis, or bilharzia.

In seeking to understand the host-parasite relationship, and hence the nature of the disease, Aloysius Tielens and colleagues at Utrecht University's Laboratory of Veterinary Biochemistry, have recently characterised the parasite's tegumental lipids (Biochem. J., 1998, 334, 315).

Phosphatidylcholines predominate in the two lipid bilayers. Using high performance liquid chromatography and mass spectrometry the researchers found two isomers among the nine major constituents of the outer membrane phosphatidylcholine fraction. Phospholipase digestion showed that the two molecules are constructed identically, and the scientists attributed chromatographic separation to a difference in the mono-unsaturated octadecenoic acid, one of the constituent fatty acids. Tandem mass spectrometry then demonstrated that in one isomer the fatty acid had the double bond at carbon nine (9-octadecenoic acid, C18:1(9)), forming the common oleic acid, while in the other it was unsaturated at carbon five (5-octadecenoic acid, C18:1(5)).

The 5-octadecenoic fatty acid is rare in nature, and its distribution in the worm is almost entirely restricted to that single phosphatidylcholine species, 1-palmitoyl-2-(5-octadecenoyl) phosphatidylcholine, which is abundant in, and confined to the outer tegumental membrane. Although schistosomes do not synthesise fatty acids de novo, the parasite probably manufactures 5-octadecenoic acid in the cells beneath the tegument - it is absent from the host's blood so accumulation is unlikely.

Membranes produce biologically active chemicals from the enzyme hydrolysis of their phospholipids, and the exclusive distribution of 5-octadecenoic acid certainly suggests an important regulatory function. The researchers suspect that the fatty acid is a signalling molecule - it is known that lipids act as intermediaries between external stimuli and the parasite's internal enzyme systems.

Such a function for the chemical would be particularly appropriate, as Jos Brouwers, one of the team members, explains: "in other membrane systems, stimulation selectively releases fatty acids unsaturated at the five position, such as arachidonic acid (C20:4), suggesting they have a role in signalling. However, these common fatty acids will normally be crossing the schistosome's tegumental membranes in the flow of nutrients from the host's blood. In contrast, information could be carried by 5-octadecenoic acid because of its unique structure and location."
Russ Clare
Return to top


Insects' own moth-balls
Chemistry in Britain, January 1998

No matter how innovative our exploitation of chemicals may seem, it often turns out that evolution in the natural world arrived first. Take, for instance, the once-popular pest repellent naphthalene in moth-balls; scientists have now discovered that a South American stick insect uses a remarkably similar solution for keeping unwelcome visitors at bay.

Being cumbersome and slow, stick insects are highly vulnerable to predation and so, to escape attention, many species - there are some 2,500 - are cryptically-coloured to resemble twigs or leaves. In contrast, the Peruvian fire stick, Oreophoetes peruana, advertises its presence with flamboyant colour schemes - males are black with red bands while females and nymphs have yellow, green and orange markings. The bright colours may help to ensure that predators associate the insect with a potentially unpleasant experience, for when disturbed the fire stick ejects a malodorous fluid from glands located just behind the head.

In the US, Thomas Eisner and his colleagues at the departments of Chemistry and of Neurobiology and Behaviour at Cornell University, Ithaca, have been working with entomologist Randy Morgan to investigate the chemistry of this secretion. Morgan has established a breeding colony of fire sticks at the Cincinnati Zoo and Botanical Garden, and, by 'milking' several individual insects, the scientists have been able to collect a sufficient quantity of the opaque fluid for analysis. Gas chromatography, followed by mass spectrometry and infrared spectrometry, revealed only a single volatile component - quinoline.

Eisner and his team went on to confirm the suspected biological activity of quinoline. They found that, in its pure form, the chemical's vapour was strongly repellent to both ants and cockroaches and, on contact, it caused local irritation to spiders and frogs. In cockroaches, the irritant effects of a sample of the secretion and of pure quinoline were indistinguishable. The researchers found that the fire stick secretes quinoline as the inner phase of an aqueous emulsion and this helps to spread the chemical over wet surfaces such as amphibian skin and, presumably, over the mouth and eye linings of predatory birds and mammals (J. Experi., Biol., 1997, 200, 2493).

When the scientists looked at the distribution of quinoline, they found it to be a very rare natural product - it is a minor constituent of coal tar and traces have been discovered in the bark of the South American tree, Galipea officinalis - the source of Angostura Bitters. Quinoline derivatives such as 1-methyl-2-quinolone have been found in several insects and are presumed to be used in defence, while other animals synthesise the very similar quinazolines from anthranilic acid (ortho-amino-benzoic acid). However, concluded Eisner, "the Peruvian fire stick appears to be the first known animal source of quinoline". The chemical was not found in the insects' diet of ferns, indicating that the animals synthesise rather than accumulate it. This suggests, according to Eisner, that the fire stick has diverged from the few other stick insect species that have evolved a chemical defence - using mevolonic acid, the common terpene precursor, they synthesise cyclopentanoid monoterpenes, such as anisomorphal, which are unrelated to quinoline.

The researchers suggest that napthalene may be the closest chemical relative of quinoline from a structural and electronic viewpoint. Napthalene, an abundant constituent of coal tar, has been used for many years as a household insect repellent. As Jerry Meinwald, another member of the team, points out, "the Peruvian fire stick appears to have hit upon the expedient of producing an insect analogue of moth-balls".
Russ Clare
Return to top

Return to credits