“These colours give me extraordinary exaltation. I have no thought of fatigue; I shall do another picture this very night”. And so he did, time and time again. The famously mad Vincent Van Gogh produced many great works of art. Sunflowers, The Yellow House and even A Starry Night all bring to life Van Gogh’s particular love of the colour yellow. Some theorists even suggest that Van Gogh truly did see the world in shades of yellow – a side effect of remedies such as digitalis and absinthe, used to counteract his seizures.
Whatever the reason, Van Gogh adorned his paintings with every kind of yellow to hand: from Naples yellow and ochre to lemon and chrome yellow. These splashes of colour were created from a range of synthetic and organic yellow dyes, each bringing their own ‘origin story’ to the canvas. Art historians seek to uncover the composite dyes used to create a particular colour so they can better date a piece of art, interpret its meaning, preserve and restore it.
An entire field of research has consequently grown up around identifying the organic dyes a long-dead painter once used. The latest effort is a twist to a technique called Raman spectroscopy. Raman spectroscopy shines a beam of light - somewhere between ultraviolet and near infrared - onto a paint sample. Each component with the paint scatters the light differently, creating a unique ‘spectral fingerprint’. This fingerprint can then be used to identify the mystery components against a catalogue of fingerprints of known compounds.
The problem with regular Raman spectroscopy however is its power. More specifically, the beam of light shone onto the paint sample is powerful enough to degrade the very dye it is trying to detect. Raman version two – or rather Surface-enhanced Raman spectroscopy (SERS) - presents a way to overcome this.
SERS shows much greater sensitivity than regular Raman spectroscopy, and can therefore be used to identify mystery components with less destruction of the paint sample. The physics behind exactly how SERS achieves this is still a topic for debate. The trick lies in how the sample is presented. In SERS, unlike regular Raman spectroscopy, the sample is adsorbed onto a metal surface – this seemingly simple adjustment is what enhances its sensitivity.
All of this sounds great in theory but what about in practice? Pretreating a drop of oil paint with a concoction of acids and alcohols, before subjecting it to SERS, has already proven successful in detecting indigo and Prussian blues in oil paint. However as Van Gogh once said “There is no blue without yellow and without orange.” The vivid landscape of yellow oil paints had been largely unexplored by SERS – that is until now.
A team of US researchers have focused their efforts on producing clean, accurate spectral fingerprints of a range of natural yellow dyes. They then tested the usefulness of their technique by measuring the same spectra in the corresponding oil paints created from these dyes.
The dyes tested fell into two categories: organic dyes and lake pigments. The latter term, although suggesting a dye sourced from the depths of a body of water somewhere, actually refers to a dye bound to substance such as a metallic salt in order to create a thick, glossy paste. ‘Lake’ in this case actually refers to the word ‘lac’ meaning a resin-like secretion.
In order to successfully identify the lake pigments Still de Grain and Reseda Lake within microscopic amounts of their corresponding oil paints, the samples first had to be pretreated with a tiny amount of hydrochloric acid and methanol. Following this simple pretreatment, the spectra of these dyes were clearly detected using SERS. When it came to oil paints generated from the organic dyes turmeric, old fustic and Buckthorn berries, this pretreatment proved unnecessary – SERS alone was successful in detecting their unique spectra.
For the very first time SERS has been shown to be an eagle-eyed sleuth when it comes to detecting yellow dyes in oil paints. With the trials and tests now complete, its time to put SERS into the field and see what historical insights this new player can bring to the yellow world of art.
Mayhew HE, Fabian DM, Svoboda SA, Wustholz KL. Analyst. 2013 Aug 21;138(16):4493-9. doi: 10.1039/c3an00611e. Surface-enhanced Raman spectroscopy studies of yellow organic dyestuffs and lake pigments in oil paint.