Synthetic · Mid-to-late 19th century
Low toxicity
Varies by specific compound; some early aniline dyes raised health and safety concerns for factory workers, prompting later industrial regulation.
Modern synthetic dyes are subject to industrial safety and environmental regulation not present in the 19th century; individual compounds vary and are not itemised here.
'Aniline dyes' is a loose historical name for the first synthetic dyes made from coal-tar chemicals. Most early ones belong to a few classes: triarylmethanes such as fuchsine (magenta), methyl violet and aniline blues, whose three linked aromatic rings carry a delocalised positive charge; xanthenes such as eosin; and, from the late 1870s, azo dyes built around a nitrogen–nitrogen double bond. Long conjugated systems with strong donor and acceptor groups give them intense, saturated colour at very low concentrations.
After mauveine in 1856, chemists in Britain, France, Switzerland and above all Germany turned coal-tar by-products from gasworks into a stream of new colours: fuchsine in 1859, followed by violets, blues, greens and the first azo dyes. Companies such as BASF, Hoechst and Cassella grew from this trade, and dye firms published pattern books of dyed samples to sell their ranges. Within a few decades synthetic dyes had displaced most natural dyes except indigo, which fell in turn after 1897.
Early synthetic dyes were bright but often fugitive. Basic triarylmethane violets fade quickly in light, and chemical analysis can now gauge how far a sample has faded. Eosin, used in 'geranium lake' pigments, is notorious: SERS found eosin in Van Gogh's Irises, suggesting the now whitish background was once pink, and imaging studies map how geranium lake mixed with lead white discolours. Many early dyes were also poor to washing, so bleeding and staining are further signs conservators look for in textiles of the period.
These dyes contain no distinctive heavy elements, so identification relies on molecular methods: HPLC and mass spectrometry for extracts, SERS for microscopic samples from paintings, prints and paper, and ATR-FTIR with statistical modelling for dyed fibres. Dated pattern books from dye companies act as reference collections, and identifying a dye class can itself help date an object, since each became available at a known time.
Most early aniline dyes are no longer made. They were replaced by better-performing acid, direct, disperse and reactive dyes for textiles and by synthetic organic pigments for paint and print, chosen for light- and wash-fastness that is tested against standard scales. Some descendants — crystal violet and eosin, for example — survive mainly in laboratory stains and specialist inks rather than in cloth.
FactStrong evidence
SERS of a sample from Van Gogh's Irises found eosin Y, suggesting the now whitish background was originally pink; faded methyl violet was identified in Matisse cut-outs and a 19th-century silk.
Caveat: Microscopic samples from a few objects.
FactStrong evidence
Mass-spectrometry imaging of geranium lake mixed with lead white maps the chemical changes behind its discoloration in paint layers.
Caveat: Model paint samples rather than paintings.
Source: Chemical Mapping of the Degradation of Geranium Lake in Paint Cross Sections by MALDI-MSI
FactStrong evidence
Early synthetic dyes include triphenylmethane dyes (fuchsine from 1859, crystal violet from 1883), azo and xanthene dyes; most are no longer made and can be distinguished on wool by ATR-FTIR with chemometrics.