More information
Description
The proposed project aims to develop a 3D physical model of paintings with craquelure patterns and through this decisively contribute to the development of evidence-based environmental specifications in museums. Collaborative efforts between partners from Poland and Norway, supported by research groups and museums in France, the UK and the USA will advance knowledge in this cross-sectoral area, blending natural and engineering sciences with conservation and art history. In the first phase of the project, a database of material properties for historical materials used in paintings as well as more modern natural and synthetic materials used in conservation practice for consolidation will be compiled. In particular, fracture toughness will be determined for painting materials containing animal glues such as sizing, grounds and paint layers. Additional investigations of material properties that will inform the 3D model include a quantum chemical study of the molecular level interactions between the organic oil medium and metal ions contained in common pigments to gain insight into the long-term mechanical properties of oil paint layers. As it is the ambition of the project to significantly advance sustainable care and conservation practice for paintings, the global dissemination of the project outcomes to users and stakeholders in the conservation and museum sector is an important task. The international network of institutions involved in the project will aid this effort. Most notably, the project outcomes will support clear recommendations for standardized practices to facilitate institutional adoption. This work will also expand the applicability of HERIe, a web-based decision-supporting tool for assessing risk of physical damage to works of art co-developed by the Applicant. HERIe is uniquely positioned to impact the adoption of sustainable practices since it is offered freely to the cultural heritage community and requires no specialized training
Summary of project results
Looking at paintings, we admire colours, shapes formed, the brushwork … but we see also crack patterns – the craquelures that give historical ‘patina’ to painted surfaces and enrich the viewer’s aesthetical perception. Variations in the craquelure pattern geometry have been recognized – spacings between cracks, shapes and sizes of ‘islands’ created by cracks, or crack orientation in relation to wooden or canvas substrates. Since these geometries depend on the substrate of the pictorial layer, thicknesses of ground or paint layers, and binders and pigments used in their manufacturing, the crack typologies have been correlated with different – geographically and chronologically - artistic traditions. Therefore, craquelures have been an important element in the art historical analysis of paintings and judgment of their authenticity. Further, craquelure patterns were known to diminish the vulnerability of paintings to temperature and humidity variations in their environment. However, the mechanisms along which the stress fields in paintings are generated and dissipated in the crack development were understood in heritage science only in very general terms.
Therefore, the strategic aim of the project was to develop a three-dimensional physical model of original cracked pictorial layers and through this create new knowledge at the interface between natural sciences, history of art, and conservation. From the perspective of art history, the project aimed to understand pictorial layers as physical systems and base analysis of their fracturing on a rigorous approach of fracture mechanics. In this way, the project had the ambition to add a new tool to the classification of craquelure patterns so far based on image analysis, more recently supported by computer recognition of pattern’s specific features. From the perspective of conservation, the project aimed at establishing relationships between crack densities and the vulnerability of pictorial layers to further fracture to determine quantitatively the levels and ranges of microclimatic variations at which the risk of physical damage of such vulnerable objects stays in the tolerable domain. The dissemination of the project’s outcome and reaching users and stakeholders in the conservation and museum sector globally to support the sustainable care and conservation practice for paintings was an important additional objective.
Throughout the project, a database of material properties for historic materials found in paintings was considerably enlarged through extensive experimental work on new groups of materials: tempera paints, oil paints after long-term natural ageing, varnishes, and consolidants used in the conservation of paintings. The project team was particularly satisfied with the successful determination of fracture toughness in thin films of paints, the key parameter characterising the material’s ability to resist crack propagation. The parameters allowed the modelling of stress development and crack propagation in paintings to be carried out reliably.
To this aim a three-dimensional physical model of painted objects – wooden or canvas substrates covered with the two-layer structure of the pictorial layer ‒ the gesso and the paint, including the developed crack networks ‒ was developed. Two paint types were considered — egg tempera and oil paints, laid on animal glue-based ground. Two scenarios of stress development were analysed: cumulative drying shrinkage of paints or ground, owing to gradual loss of water or evolution of the molecular composition of the binders, and moisture-induced expansion of the wood substrate. The tension between cracks was modelled for increasing magnitudes of materials’ dimensional change. Critical ratios of distances between cracks to the paint or ground layer thickness for which stress between the cracks dropped below the value inducing fracture in the materials and saturation of the crack patterns occurred were estimated. The modelling revealed that crack saturation occurred at relatively low shrinkage of ground or paints attained after the initial phase of drying. In turn, humidity variations typically encountered in spaces where paintings are displayed cannot lead to new cracks in paintings with developed patterns of these early ‘drying’ cracks.
Model predictions were validated by subjecting specimens mimicking historic paintings on wood and canvas in the laboratory to controlled humidity variations and tracing crack development in the pictorial layers. The dynamical response of the specimens induced by the variations agreed with the model. Regular sets of cracks of varying crack density were formed in agreement with the modelling. Also, diverse craquelure patterns extracted from high-resolution images of real paintings were interpreted by the modelling developed.
The project has created a new tool for analysing crack patterns, especially crack spacings, to gain information on the composition and stratigraphy of pictorial layers in paintings. For paintings on panels, analysis of cracks perpendicular to the wood’s grain was found to be particularly robust owing to the dimensional stability of wood in the longitudinal direction, and crack patterns of such geometry will be the target of future research linking fracture mechanics and art history.
Using the model, tolerable magnitudes of variations of microclimate parameters in spaces where paintings are displayed, have been derived. The project has confirmed decisively that paintings with developed craquelure patterns are significantly less vulnerable to climate variations than indicated by earlier studies of fracture risk for pictorial layers based on laboratory testing of undamaged materials. As the relationships between crack densities and the vulnerability of pictorial layers to further fracture have been quantitatively established, a perspective of expressing the risk of physical damage to paintings as expected visual change has been opened. This is vital for loss-of-value judgements necessary in professional decision-making and public engagement with cultural heritage.
The project team has reached diverse audiences of scientists and conservation practitioners by publishing outputs in several high-impact publications and engaging directly through dedicated museum workshops, conferences, and educational activities. The models developed within the Craquelure project have been embedded in the digital decision-supporting platform HERIe that provides remote and free access to quantitative assessment of risks to heritage assets at herie.pl.
The outcomes of Craquelure have the potential to form the basis of enhanced environmental guidelines for the cultural heritage sector, within the American Society of Heating, Refrigerating, and Air-Conditioning Engineers ASHRAE which are currently widely used in Europe and elsewhere as an expert of the beneficiary is a member of the team updating the guidelines.
The beneficiary joined as the widening partner of the Horizon Europe GoGreen project ‘Green strategies to conserve the past and preserve the future of cultural heritage’, implemented between 2022 and 2026. The group will complement the current focus of GoGreen on chemical degradation processes in heritage materials with physical degradation models for vulnerable objects.
Summary of bilateral results
The project was planned and implemented in international collaboration. It involved Polish and Norwegian researchers, one of whom was a young researcher (post-doc). The number of Norwegian researchers and their contribution to the project are in line with those planned in the project.