A further step towards EOSC implementation: research data on ultramarine blue disease in masterpieces made open via @ESRF data portal
The European scientific community is working in the development of the European Open Science Cloud (EOSC), which will allow researchers to share all their data with their counterparts. A team working at the ESRF on how blue pigment was prepared for use in historical paintings is one of the first to share its data. Their results are published in Science Advances.
According to a survey led by Nature in 2016, 70% of scientists admitted they could not reproduce another scientist’s experiments and more than half could not reproduce their own. In order to improve sharing and, in turn, enhancing innovation, the European Union is working on implementing the European Open Science Cloud (EOSC), a kind of “library” of all experimental raw data and methods, available to everyone.
PaNOSC is one of the European H2020 projects currently contributing to this goal, and the ESRF, as leading institution of the project. “We are in the process of implementing the ESRF Data Policy to organise the data from experiments in an archive, which ultimately everyone will be able to access. The scientific teams will have three years to keep their data closed to the public, and after that any other scientist can try to repeat or do new data analysis of the very same experiment if he or she wishes to”, explains Andy Götz, coordinator of the project. The final goal of PaNOSC and the EOSC is to make data from publicly funded research in Europe Findable, Accessible, Interoperable and Reusable (FAIR).
One of the first ones to contribute to this new open data archive is Alessa Gambardella, from the Rijksmuseum (sponsored by Akzo Nobel), and her colleagues, with their research on ultramarine blue, which is a highly-valued brilliant blue pigment used in masterpieces. Comprising the blue mineral lazurite, ultramarine blue pigment historically was obtained from the semi-precious stone lapis lazuli. Unfortunately, paint containing ultramarine blue pigment is susceptible to a degradation phenomenon known as ‘ultramarine disease’.
Watch our past interview with Alessa Gambardella:
Gambardella carried out the experiments on ID21 with the project team and requested that their raw and processed data would be available to everybody: “I think the EOSC project is the way of the future, and I wanted our research to be open and available. If someone wants to build upon our research, they can do so by re-using the data – whether it be the raw or processed data – that we have stored in the ESRF Data Portal with a unique DOI. Further, the related full processing protocol is available in GitHub. While going ‘open’ was quite time-consuming and required reaching a new level of vulnerability, I feel that such transparency will ultimately serve as the best remedy for safeguarding against the ‘reproducibility crisis’, especially as data size and complexity continue to grow while also helping to improve innovation”, she explains. Götz adds: “Gambardella’s work is an illustration of what EOSC and PaNOSC will enable and help to carry out in the future”.
In this particular study, Gambardella and the team of scientists used X-ray absorption near-edge structure spectroscopy (XANES) at the sulfur K-edge in microbeam and full-field modes to measure how the sulfur speciation within lazurite depends on the heat-treatment of lapis lazuli from which it was extracted. The influence of heat was of interest because the exposure of lapis lazuli to heat is mentioned in historical documents describing pigment preparation. The findings showed that sulfur signatures in lazurite depend on the heat-treatment of lapis lazuli and give a unique signature for high temperatures. The team further found heat-treatment sulfur signatures in lazurite particles from five historical paint samples originating from artworks in the Louvre Museum (Paris), the Rijksmuseum (Amsterdam) and the Mauritshuis (The Hague, NL). This provided them with a marker for artists’ pigments that had been extracted from heat-treated lapis lazuli. As an example, the method has been applied to a very small fragment from Johannes Vermeer’s Girl with a Pearl Earring (c. 1665, Mauritshuis). The results, and those obtained with other techniques, have been published in Heritage Science. “Understanding how the properties of ultramarine blue are affected by preparation is a first step towards determining their influence in the aging of the paint and, ultimately, the cause of ‘ultramarine disease’. Identifying a marker for such pigment preparation was very fortuitous”, explains Gambardella.
Marine Cotte, scientist in charge of the beamline ID21 and co-author of the study, adds: “The open data policy at the ESRF is making it much easier for all scientists (those who acquired the data, but later, anyone possibly interested) to find back raw data and metadata, in a very organized way. Considering the unique and precious nature of cultural and natural heritage, such efforts to share raw data are essential. Any data obtained on an unreproducible object must be preserved and made accessible, as part of our scientific heritage”.