Benchmarking the Oddy Test: Development of Damage Thresholds and a New Tool for Standardizing Material Tests for Cultural Heritage Institutions

A variety of manufactured materials such as foams, fabrics and adhesives are used in the conservation and display of cultural heritage objects. Many institutions around the world rely on the Oddy Test ^[1] to evaluate the suitability of such materials for potential use in collections care. The Oddy Test is an accelerated aging test that exposes silver, copper, and lead coupons to conservation materials at 60°C and approximately 100% relative humidity for 28 days (Figure 1). However, there are several limitations that exist when conducting and interpreting the Oddy Test. The test is time, labor, and resource intensive, requiring a month to complete. It also relies on the use of silver, copper, and lead surfaces as proxies for collection objects, potentially overlooking impacts on other artist’s materials such as paper, paints, or plastics. The Oddy Test is also sensitive to differences in experimental procedure across institutions, where metal purity and surface polishing techniques or airtightness and type of testing vessels vary. It’s equally important that the evaluation of the metal coupons at the end of the test is subjective, as a trained individual must visually assess the material as suitable for permanent (P) or temporary (T) use, or unsuitable (U) for use with collection objects.

Figure 1. Photographs of duplicate copper, silver, and lead control coupons (left) and sample coupons (right) from the Oddy Test protocol at The Metropolitan Museum of Art. The sample coupons were aged with a hot melt adhesive and rated U (unsuitable) for copper and lead, and T (temporary) for silver.

To address these limitations, past work in the Preventive Conservation Science Laboratory (PCSL) focused on optimizing The Met’s Oddy Test methodology to minimize experimental variations.^[2] Efforts also continue to assess the impact of conservation materials on cellulosic objects using paper rather than metal coupons.^[3] Most recently, we published a standardized method for photographing metal coupons and applied a machine learning model to assign P/T/U ratings to reduce subjectivity.^[4] The Oddy Benchmarking Project aims to continue this work by creating standardized mixtures of reactive chemicals acting as a calibration tool to produce reliable and reproducible Oddy Test results across different testing protocols.

In the first phase of the project, a team of conservators surveyed and interviewed conservators, scientists, and collections care staff from 134 institutions to generate a list of approximately 40 materials of interest with well-documented use in the field. The collected data includes Oddy Test results, microchemical test results, and personal accounts of the field performance of commonly used conservation materials. Research scientists in the PCSL are now using direct thermal desorption (DTD) paired with gas-chromatography mass-spectrometry (GC-MS) to identify the volatile compounds that are emitted from these materials with the goal of determining which components have the potential to react with the surface of silver, lead, or copper during an Oddy Test. Identification is followed by quantitative analyses to determine the concentrations of reactive compounds within the conservation materials.

To complement the DTD-GC-MS data, our team is also performing direct analyses of the metal coupons’ surfaces after exposure to the conservation materials during the Oddy Test. X-Ray Diffraction (XRD) (Figure 2), Raman spectroscopy, Scanning Electron Microscopy (SEM), and electrochemical techniques are being employed to characterize the corrosion products on the coupon surface. The identification of corrosion layers can help to elucidate the chemical pathways of degradation, aiding in our understanding of the chemical reactions occurring during the Oddy Test.

Figure 2. (A) 2-Dimensional X-Ray Diffraction pattern collected from the surface of a copper coupon aged with polymer adhesive tape. The copper coupon showed red tarnish and was rated as U (unsuitable) at the MMA. (B) Integrated XRD pattern showing library match to cuprite (vertical blue lines).

Ultimately, we aim to relate the reactive components and their concentrations to the amount and type of corrosion detected on Oddy coupons, and more broadly to the performance of the conservation materials in the field. This will inform the development of standard “benchmarking” solutions that clearly delineate the borders between permanent and temporary, and temporary and unsuitable ratings. The goal is to minimize and hopefully eliminate the subjective nature of the approach to materials testing in collections care, and to improve access to unbiased information regarding the suitability of conservation materials.

This project has been made possible through a National Leadership Grant from the Institute of Museum and Library Services (IMLS), grant number MG-249353-OMS-21.

[1] Oddy, W.A., An unsuspected danger in display, ‘Museums Journal’ 73 (1973), 27-28. Return

[2] Stephens, C.H., Buscarino, I., Breitung, E.M., Updating the Oddy Test: Comparison with Volatiles Identified Using Chromatographic Techniques, ‘Studies in Conservation’ 63 (2018), 425-427. Return

[3] Volpi, F., Stephens, C.H., Potthast, A., Breitung, E.M., Ongoing development of a semi-quantitative protocol for assessing the suitability of commercial materials used to store or exhibit cellulose-based artworks, ‘The European Physical Journal Plus’ 136 (2021), 1084. Return

[4] Long, E.R., Bone, A., Breitung, E.M., Thickett, D., Grau-Bové, J., Automated corrosion detection in Oddy test coupons using convolutional neural networks, ‘Heritage Science’ 10 (2022), 150. Return