Colour code

Necessity is the mother of invention, and for conservator and PhD student Maria Kubik the cliché holds true.

In probably one of the more original doctoral projects underway in a university in Australia, Kubik has developed a portable reflectance imaging system that has the potential to change conservation in museums and galleries around the world.

Conserving artwork is involved and detailed work. One of the most important features of preserving a painting is knowing which pigments were used. Pigments provide the colour in paints, and each has a unique chemical property. If conservators need to add colour to a painting, it is important that the pigment matches almost exactly the one used originally.

But there are other, potentially more significant ramifications of this technology.
“A lot of the attraction in identifying pigments is preventative”, Kubik says. “Knowing which pigments have been used is important for managing artwork. For example, particular pigments are light sensitive, and if you know this as a conservator, you can best preserve or treat it.

“Knowledge of materials is an aid in forgery detection too. Authentication is obviously important for valuation reasons. One way of checking for obvious fakes is to establish the artist’s known palette, and check that no pigments of inappropriate dates of first manufacture or usage are present.”

Kubik, whose favourite paintings include those from 20th century Australia, is in a strong position to understand the issues faced by the conservator community, having studied conservation at an undergraduate level and worked in the industry, mostly in the ACT, for a number of years.

During this time she completed her Masters, but her appetite for further study was whetted. She proposed a doctoral project to look into the analytical side of conservation.

“But it all changed around. I had in mind a couple of analytical techniques that I wanted to play with. Through experimentation I actually found that they were either too limited in their application or it was impractical. This new imaging technique came out of the need for addressing those issues.”

Currently, conservators use two common techniques when analysing the pigments of a painting.

The first method involves the removal of a tiny section from the artwork, which is then mounted in resin and put under a Raman microscope, showing up the unique chemical structure of each pigment.

Alternatively, a painting can be placed under a reflectance spectrometer, which determines which pigment is which, depending on the spectral reflectance pattern of the sample.
Both methods contain an element of damage risk to pieces that are often delicate, ageing, and expensive.

“Even on the smallest scale, taking a cross-section is destructive, and because the spectrometer is a giant instrument you have to move the painting to the lab, which is a hassle. It also has a limited size capacity, so big paintings are out,” Kubik says.

“The challenge has been to come up with some non-destructive form of testing that can rapidly and accurately assess what pigments have been used in a paint mix.”

Kubik has built a small, portable reflectance spectrometer that can accurately assess the pigments used in any painting. It is compact, and, placed in front of a painting, can transfer pigment data to a laptop and match it to a database within minutes.

“I’ve been testing it under ambient light and it doesn’t seem to make any difference to its performance, which is great because if I just set this up on the floor of a gallery, they’re not going to want me to turn out all the lights.”

The imager is a combination of commercially available equipment and technology built at the Research School of Chemistry.

Initially, Kubik looked at integrating imaging spectroscopy, which is used extensively in remote sensing, where information is extracted from the light reflected from the Earth’s surface.

“Unfortunately, while the approach looked promising, it quickly became apparent that the equipment used in remote sensing was not suitable for my purposes. It might work at distances measured in kilometres, but assessing paints involves working at distances of less than a metre,” she says.

However, remote sensing did have something to contribute to the device. The data that Kubik’s imager generates is converted into spectra by software on a connected laptop, and is stored and recalled by a special database. This program comes from a combination of remote sensing and astronomy software products.

The real innovation comes direct from the laboratory on campus where Kubik worked with Laser and Optical Spectroscopy group leader Professor Elmars Krausz on special optics to filter the light reflectance.

Within the imaging system there are custom made carousels of these interference filters that are the ‘eyes’ of the system. They build up a data cube, or the layers of images taken by the equipment of the reflectance at different wavelengths. Each layer represents an image acquired at a particular wavelength band, which in turn provides a full reflectance spectrum at each point.

The reflected light is then measured using a charge coupled device (CCD) camera. “In its simplest form, a CCD consists of a collection of pixels that produce an electrical charge proportional to the amount of light received,” Kubik says.

Once the CCD has measured the light, the data is fed back into the specialised software and loaded into and compared with a database of pigments.

The conservator then knows the exact pigments used in the artwork. It’s a powerful tool that brings together the artistic, scientific and technological.

“By combining the three technologies of digital imaging, reflectance spectroscopy and multivariate computer analysis, it’s possible to determine the composition of paints in-situ,” Kubik says.

The interdisciplinary nature of the project has allowed her to work with and seek advice from researchers in various laboratories across campus.

“That’s the great thing about being here at ANU – you have access to lots of expertise, and everyone has been enthusiastic about this project,” Kubik says.

“I really had, and still have, a very small understanding of the intricacies of building optics and the potential they have. But here’s an example where my specialised skills as a conservator meet a specialised area of science and come up with an applied product from joint knowledge.”

“The optimal number of filters used can vary and depends on many factors, such as the imaging system, the subject, the technique for reconstructing the image, and the costs and benefits of using more of fewer channels for each application.”

Kubik has already tested the camera in a real gallery environment at the Australian War Memorial. She is also testing the imager’s robustness under different conditions and on artworks with different conservation issues. One work in her laboratory, for example, has had a yellowed varnish applied in the past; another has been scratched and gouged.

Kubik hopes that the trend in conservation to preserve a painting by means that don’t necessarily involve active treatment will lead to interest in the technology she has spent the last two years developing.

“The process of conservation now requires a full understanding of the original materials and processes used by artists. Research into materials and conservation processes is carried out in collaboration with museums, conservators and scientists,” she says.

The benefits may not just be practical for conservators, but also assist art historians in their analysis of artworks by shedding more light on the intentions of artists, Kubik says.

“Studying painting techniques gives a closer relationship to the practice of painting, may clarify the artists’ intent and reveal a series of tools used to achieve the final image.

“Technical art historical study and conservation rely heavily on accurate knowledge of materials present in an object in order to decide on a correct course of treatment and preservation. It most importantly allows for developing a sympathetic method of conservation.”

Back to top