Sniffing out Decay
Classifying Degrading Plastic Artworks Using Their Smell
- Fig. 1: Tate plastics-based object analysed as part of this research with the SPME fibre in place, as packed in August 2017. Antoine Pevsner, Head 1923-24 (Tate T02241).
- Fig. 2: Analysis of plastic samples using SPME-GC/MS analysis in the Heritage Science Laboratory at the UCL Institute for Sustainable Heritage.
Many people think it’s strange to regard plastics as ‘heritage’, but objects made from plastics are found more and more often in museum collections nowadays. They are present in works of modern art and classic design objects such as the famous Panton chair.
Other museums have 20th century social history artefacts made of plastics, such as handbags, clothing and furniture that help us to understand the development of modern life and our consumer society. Plastics are also found in archives, for example in film reel and photographs. Despite being some of the youngest artefacts in museums, plastics can often be the least stable. Some plastic-based artworks have experienced catastrophic collapses after just a few decades in museums, due to chemical and physical deterioration. These problems are often most severe in objects made from early plastics, as they date from a time when plastics technology was in its infancy and formulations were somewhat experimental. Research in the field has identified four materials as particularly problematic – cellulose acetate, cellulose nitrate (sometimes referred to as celluloid), poly(vinyl chloride) (or PVC) and polyurethane foams.
VOC analysis has been studied a lot in medical research, where breath analysis can be used to diagnose conditions such as lung cancer and Alzheimer’s disease. The advantages of this approach are that it can extract useful diagnostic information from a complex chemical mixture of VOCs, while also being completely non-invasive for the patient.
Heritage objects are a lot like hospital patients – they are all different from each other, chemically very complex, and need to be treated with care. So, the factors that make VOC analysis useful for medical diagnosis are also advantages when analysing museum artefacts. This work aims to translate the success of VOC analysis from a medical context into museums.
The researchers from UCL and the University of Strathclyde used solid-phase micro-extraction gas chromatography/mass spectrometry (SPME-GC/MS) to analyse VOC emissions from over 200 samples of plastic objects. SPME-GC/MS uses a fibre with an adsorbent coating that picks up VOCs from the headspace of an object. The samples were sealed in vials for 1 week at room temperature, so that VOCs released from them could be trapped. The fibre was then inserted into the vial and left for an hour at room temperature to collect any VOCs emitted by the sample. The fibre was desorbed into a GC/MS so that the VOCs collected could be separated and identified. For the laboratory-based work, this was done on small samples of plastics objects, which were chosen to represent the plastics that are commonly found in museums, including the most vulnerable ones listed above. Some samples were also artificially degraded, by exposing them to high temperature and humidity, so that the way in which the mixture of emitted VOCs changes as a plastic sample degrades could be tracked.
Some of the compounds detected provide evidence of chemical or physical deterioration processes. For example, cellulose acetate, an early plastic which is found in photographs, film reel and works of modern art is known to deteriorate in museum conditions to produce acetic acid, a process known in the museum community as “vinegar syndrome”. This can be a real problem, because the acid then acts as a catalyst, accelerating the decay of the problematic object, and potentially spreading through the air to ‘infect’ other objects. Acetic acid was detected from the cellulose acetate samples analysed in this work, and the levels of detected acids were found to increase, the more degraded the sample was. Other degradation products such as aldehydes and other carboxylic acids were also detected, and in some cases found to increase in more degraded samples. Plasticisers such as phthalates were also detected, and often found to decrease in more degraded samples. This makes sense as it’s well-known in museums that over time, plasticiser will leach from objects, causing them to become brittle and develop sticky residues on their surfaces.
Classification Accuracies of 50-83%
However, it was difficult to understand something as complicated as the degradation of a plastic sample by identifying and tracking individual VOCs. As well as the base polymer, plastic objects can contain additives such as pigments, anti-oxidants and fillers, meaning that their degradation involves multiple chemical and physical processes. Using an approach that has been successful in medical research, the researchers developed a model which allowed them to use combinations of the detected VOCs to classify the samples according to how long they had been artificially degraded. The samples were divided into two groups, based on how degraded they were, and the presence and relative quantities of particular VOC mixtures were used to predict to which group a sample should belong. Classification accuracies of 50-83% were achieved for different polymer types. The combinations of VOCs that gave the best predictions often made intuitive chemical sense, for example an increase in oxidation products such as acetophenone from polystyrene, in combination with a decrease in VOCs that are manufacturing residues were found to indicate that a sample was more degraded.
The researchers then applied this model to the analysis of 3 artworks from the collections at Tate in London. SPME-GC/MS was used at the museum site to detect VOC emissions from 3 pieces dating from the 1920s and 1930s. In the case of the museum-based work, samples were not taken from the objects, the SPME fibres were simply placed in the storage containers of the artefacts and left in place for 1 week. This makes the analysis method non-invasive, a great benefit in the museum sector where sampling such unique and precious objects is often not permitted, while still providing detailed chemical analysis. Using the results, and the model they had developed with the laboratory samples, the researchers identified one of the artworks in particular as being at a more advanced state of degradation than the others, although it does not visibly appear to be in a worse condition.
These preliminary results demonstrate the possibility of using VOC analysis as a non-invasive tool to better understand the degradation of plastic museum objects and to identify deteriorating artefacts before damage becomes too severe. They give museum conservators evidence that they can use to make decisions about how best to store or display an object.
Work is continuing at the UCL Institute for Sustainable Heritage to develop this analysis into a practical method that can be used easily in museums. The researchers hope that this will be a diagnostic tool that is as useful in heritage as it is in medicine.
Dr. Katherine Curran
UCL Institute for Sustainable Heritage
London, United Kingdom
Katherine Curran, Mark Underhill, Josep Grau‐Bové, Tom Fearn, Lorraine T. Gibson, Matija Strlič: Classifying Degraded Modern Polymeric Museum Artefacts by Their Smell, Angewandte Chemie (2018); DOI: 10.1002/ange.201712278.
Further Materials Science articles: