Dipping beneath the surface: Scientific study of Turner's painting has revealed his obsession with new pigments. Dalya Alberge reports

'PIROGEN. Opogene. Lipigene . . . Molybdic acid. 3 dracm 1 black. 2 blue oxide more oxigene Green then Acid . . .' This confused formula is not from a chemistry book but from the pen of J M W Turner. We might assume that the 19th-century Romantic painter was driven only by passion and natural forces but he was so fascinated by the scientific properties of pigments and paints that he filled sketchbooks with notes on them.

Although these have long been available to historians, no one bothered to transcribe them fully, let alone study them, and Turner's lifelong obsession with finding new pigments remained unappreciated. That is, until Dr Joyce Townsend of the Tate Gallery's conservation department embarked on a scientific analysis of his oils and watercolours. 'Previously, we hadn't realised Turner experimented so much,' she says. 'He never stopped experimenting, from the earliest to the latest works . . . That Turner's methods were innovative was noted while he was finishing paintings during the varnishing days at the Royal Academy, yet few knew what his materials were. He disliked being overlooked while he worked.' Townsend's research adds valuable insight not only into his technique but into the man himself. It led her to conclude that Turner was never satisfied, that in seeking to express his ever-evolving artistic ideas, he impatiently awaited the development of new materials. For example, chrome yellow was first produced in 1814 and Townsend discovered that Turner was using it that very year. 'He didn't wait until everyone else had tried something out. And there is no sign of hesitancy in the way he applied his new materials.'

Her research took some four years; there are 300 paintings and more than 20,000 watercolours and sketchbooks from which to choose in the Turner Bequest at the Tate. The study (focusing on 50 paintings, 200 watercolours and a range of sketchbooks) culminates next month with an exhibition, Turner's Painting Techniques. The project was made possible by a grant from the Leverhulme Trust to enable the Tate to set up a laboratory for examining artists' materials.

Most of Townsend's findings required the help of the most advanced scientific equipment: from Turner's chemistry notes she learnt relatively little, his writing being barely legible. But even when the names of chemicals are written clearly, today's scientists can barely make head nor tail of them.

Most of his surviving notes are just dry lists of chemicals. 'We lack a laboratory notebook saying what worked and didn't,' she says, 'though Turner wasn't the kind of person who would have noted that.' In some cases Turner intersperses names of chemicals to be mixed for art with those for medical conditions: for example, rhubarb and morphia, recommended to the artist for travellers' ailments. Some passages read like medieval witches' potions: 'the Wood from Worms / Cotton / Water and Cow Dung . . .' What Turner intended to do with these is unclear.

With a stereo microscope, Townsend scanned the surface of each painting: under varying lighting levels, she examined the texture, for example. Where there seemed to be evidence of reworking (usually more thickly applied layers of paint were a clue), she used X-rays. In Two Women and a Letter, an unfinished work, she found that Turner had made several alterations to the original composition. With an electro-microscope, she analysed pigment elements within each painting. (This part of her research has helped to date certain works more accurately.) In some works, she detected evidence of experiments to achieve intensely bright tones before manufacturers had invented such colours. 'He was ready and waiting for new colours before they were available . . .'

From these tests, she could also tell the order in which he applied the different paint tones, and which elements of a composition he painted first. She found, in works like Apollo and Daphne, that Turner tended to work up his figures at a late stage, 'from a blob of one or two colours put in earlier', and that, clearly dissatisfied, he would keep returning to a work.

Extensive analysis of the paint base showed the extent to which Turner varied his materials within a single composition: for example, she found he used linseed oil, heat-treated, for the sky; bitumen and oil for shadowed landscape; pure resin and oil / resin mixtures for certain glazes; oil / wax emulsions (including beeswax and spermaceti wax) and layers of almost pure wax medium for water. She also discovered that Turner used an absorbent ground, one that would absorb oil from the paint to make the paint dry faster.

In works such as Pembury Mill, she could identify where he had scratched out areas of dark paint with his thumbnail. Indeed, there are contemporary reports of Turner keeping his thumbnail long for this purpose. And she observed, too, how in sketches like Shipping at the Mouth of the Thames, rather than a pencil underdrawing, he used lean oil paint thinned with turpentine to set out the composition. 'The composition must have been clear in Turner's mind before he put brush to canvas,' she says, 'since these areas of colour interlock but do not overlap.'

Her research will prove invaluable for restoration. As Townsend explains, 'The conservation of Turner's paintings is very hazardous due to the sometimes extreme solubility and temperature sensitivity of his paint.' In his own day, his Opening of the Wallhalla cracked after eight days at the Royal Academy. Ruskin, Turner's contemporary, noted: 'The fates by which Turner's later pictures perish are as various as they are cruel.'

Turner's Painting Techniques, Tate Gallery, London, SW1 (071-821 1313) 22 June-10 Oct; free. Sponsored by Thorn EMI

(Photograph omitted)