PlasCO2 - Green CO2 Technologies for the Cleaning of Plastics in Museums and Heritage Collections is a 3-year multidisciplinary research project (2018 - 2021) funded by the Portuguese Foundation for Science and Technology.

The project aims at assessing the suitability of Carbon Dioxide (in liquid or supercritical phase) as an alternative, environmentally friendly, cleaning technology for plastic-based works of art. Poly(methyl methacrylate) (PMMA), Polyurethanes (PUR), and other foam/rubber-based materials will be primarily investigated.

Extensive tests will be performed on new, unsoiled and artificially soiled mock-up samples prepared using contemporary material equivalents. The samples will be characterised before and after cleaning trials, as well as after accelerate ageing, to evaluate the impact of Carbon Dioxide (CO2) on the selected plastics.

The assessment will include empirical observations and data collected from several analytical techniques to guarantee a full characterisation of the samples from the macroscale to the nanoscale. In particular, changes in the visual appearance of the samples (dimensions, colour, gloss, surface texture), as well as modifications on their physical-chemical and mechanical properties will be monitored. For the CO2 cleaning procedure, different parameters and conditions (i.e. CO2 in Liquid or Supercritical phase, pressure, temperature, exposure time, use of a co-solvent) will be taken into account and tested to outline efficient and safe cleaning protocol(s) for PMMA, PUR and other foam/rubber-based materials.

The most promising CO2 conditions will be compared to traditional cleaning methods for plastic objects, and discretely tested on sacrificial historical samples either bought at flea markets or donated by partner institutions like NEÂdS - Núcleo de Estudos Ângelo de Sousa and Museu Benfica - Cosme Damião.

Data collected from the cleaning trials, on both mock-up and historical samples, will be further used for the development of a Machine Learning model. Using this strategy will help to understand the relationships between changes in the physical-chemical and mechanical properties of the selected materials and the conditions of the cleaning procedures, as well as to foreseen outcomes for future tests. Based on the Machine Learning system, a user-friendly prototype of a computational tool will also be designed, with the aim of assisting conservators with the selection of the optimal CO2 treatment conditions.

Due to its unique properties, CO2 might prove to have the potential for other applications for cultural heritage. The use of liquid or supercritical CO2 as a carrier of consolidant materials for PUR-conservation treatments is also currently under research.