Heritage Science (Mar 2018)

The white of the 20th century: an explorative survey into Dutch modern art collections

  • B. A. van Driel,
  • K. J. van den Berg,
  • J. Gerretzen,
  • J. Dik

DOI
https://doi.org/10.1186/s40494-018-0183-4
Journal volume & issue
Vol. 6, no. 1
pp. 1 – 15

Abstract

Read online

Abstract White pigments were abundantly used in 20th century paintings, and relate to several degradation risks such as titanium white mediated photocatalytic binder degradation or zinc soap formation. Knowledge about the white pigments that were used is essential for risk assessments of 20th century collections. In this study, a representative set of 179 paintings in Dutch modern art collections were analyzed by portable X-ray fluorescence spectrometry. Subsequent explorative data analysis was performed, supported by qualitative findings from 140 home-made reference samples. Micro-samples were further investigated by X-ray diffraction and scanning transmission electron microscopy to gain information about crystal structure and surface coating. This study reveals that Dutch 20th century artists were hesitant to use titanium white. In the Netherlands, the use of this pigment is observed convincingly from the 1970s onward and initially mostly in non-oil binders. Additionally, lead white was used until late in the 20th century for paints and grounds but rarely mixed with titanium white in paints. The study also indicated that many CoBrA artists used zinc white, underlining the risk of degradation due to soap formation. Furthermore, this study highlights the different production processes of titanium white. Through the association of titanium with niobium, an impurity only present for sulfate processed pigments, the production process can be identified. The absence of niobium indicates the presence of rutile and is thus a sign of paint stability, as well as a ‘post 1959’ dating. This explorative study illustrates the value of a combination of data analysis approaches, which includes assessment of spectra as well as descriptive, bivariate and multivariate analysis, for medium-sized datasets gathered at similar conditions.

Keywords