Fading dyes: Crystal violet degradation under UV light

Downside of synthetic dyes

Side-by-side images of a van Gogh artefact showing how crystal violet dye faded over 43 years
Image: courtesy Van Gogh Museum

Crystal violet (CV) is a synthetic dye that was first made in 1883 and is one of the group of dyes based on the triarylmethane structure. Their introduction led to a rapid decline in the use of natural dyes in works of art as they were replaced by these novel, bright-coloured substances that were relatively inexpensive to manufacture. However, they came at a high price in the arts arena, because triphenylmethane dyes were subsequently found to be highly light sensitive.

Fading and discolouration occurred relatively quickly, so that paintings began to deteriorate and lose their original quality. Purple CV regions on the canvas turned into various shades of brown after exposure to UV light. In some cases, the dye faded completely, as on a menu drawn by van Gogh which is held in the van Gogh Museum in Amsterdam (see photo on the right). Here, the effects of light can be clearly seen by comparing the view on the left, photographed in 1958, with the view on the right which was taken in 2001.

In chemical terms, the dyes degrade by losing methyl groups from the amino groups that are attached to the phenyl rings and by adding oxygen to the same nitrogen atoms. Several studies on the UV degradation of CV have identified degradation products, including one by a team of researchers in Italy who employed HPLC-DAD and a little LC/MS in their structural studies.

Now, the same team has undertaken a more comprehensive study using LC/MS and LC-tandem-MS to reveal many more UV degradation products which should help in the conservation of works of art containing triarylmethane dyes. Gabriella Favaro, Daria Confortin, Paolo Pastore and Marinarosa Brustolon from the University of Padua published their findings in Journal of Mass Spectrometry.

UV exposure

A solution of CV was exposed to UV irradiation at 365 nm for 150 hours and samples were removed periodically for analysis. In addition, the dye was soaked into filter paper and dried before UV exposure for different intervals. The decomposition products were extracted with methanol for analysis.

A combination of HPLC-DAD and HPLC/MS/MS on the same instrument was carried out, using electrospray ionisation in positive-ion mode for compound detection. The time-of-flight mass spectrometer was operated at high resolution, which aided the identification of the fragment ions from their exact masses.

The structures of the degradation products and their fragment ions formed in the mass spectrometer were confirmed by the high-resolution measurements, the isotopic profile, and the characteristic losses in the fragmentation schemes.

Crystal violet photodegradation

The total ion current chromatogram revealed more than 30 different peaks, most of which were identified. There was a series of demethylated compounds, including the mono- to penta-demethyl derivatives of CV. The related diarylmethane compound Michler’s ketone was also identified as a breakdown product, along with its demethylated derivatives. Among the oxygenated compounds were some N-oxides and, identified for the first time as CV breakdown products, some N-imido oxides and hydroxylamine derivatives.

Some of the compounds identified from the irradiation of CV on paper absorbed in the visible region, so appear coloured, but their structures could not be determined. The researchers could tell that they were not derived from CV but their identities remained unclear at this stage.

All of these breakdown products would contribute to colour fading on historical artefacts. In particular, the demethylated CV compounds absorb at a shorter wavelength or do not absorb at all in the visible region, so that their overall effect is to change to a reddish-brown colouration. This is consistent with the appearance of CV inks on aged paper.

The researchers paid special attention to the dual presence of Michler’s ketone and CV. When both compounds were added to cellulose paper, the degradation rate of CV was greater than that in the absence of Michler’s ketone. The team hypothesised that this was caused by up to three different mechanisms of activation. The ketone could act as a sensitiser for CV decomposition. Alternatively, a radical transfer mechanism involving both compounds, or the generation of singlet oxygen species from Michler’s ketone under UV light took place.

The present data can be confirmed by experiments on the colour degradation in real artefacts in which CV was employed but this might involve removing some material, so experiments need to be designed with care. The results will help to confirm exactly how UV light affects CV on paintings and documents and could lead to better methods of care and preservation.