In order to highlight connections between crack’s by means of infrared thermography in museum environments where the artefacts are most of the time in thermal equilibrium an active approach is necessary to create temperature differences of a few degrees.
The fact that different materials are used in the composition of the “Ratto delle Sabine” model made our study very demanding because of the different thermo-physics properties (thermal inertia and conductivity) of these materials. The radiographic survey campaigns showed that the inner structure of the model is composed out of organic materials, in particular wooden elements (of different dimensions and types) with metal joints in some parts, while in other parts the junctions are made out of lighter materials, like straw, or by means of dovetail and/or bonding with bandages. The external parts are completely out of raw clay, which is a very fragile material and delaminates easily.
For the reasons described above, this statue of Giambologna is a peculiar case of an artefact composed out of internal materials that could be subject to more degradation than the outer parts.
It is known that the organic material, wood in this case, needs well controlled environmental conditions for a good conservation, above all relative humidity (RH). Indeed, humidity affects the equilibrium moisture content of the wood. The temperature variations, influencing the RH variations, have a secondary effect on the wood behaviour, but not necessarily less important. Moreover wood is a material characterized by a long response time to the thermo-hygrometric variation of the environment. Table 1 shows the indications for safe daily temperature variations of ±1,5°C and ±5°C suggested by UNI 10829 – 1999 and Thomson respectively [22,23].
Taking into account these temperature limits for wood, it was decided to heat the places where the cracks were visible only for a short time in order to show the interconnections between the cracks, but avoiding at the same time temperature increases higher than ±5°C as indicated by Thomson.
The method consisted in introducing warm air (about 50°C) directly in the cracks considered critical, in particular from the static point of view, and checking by thermal camera the corresponding outflow of the warm air flow at the opposite side of the infiltration point. The heating system used was a conventional thermal convective portable device with a power of 1200 W. In order to heat only localized parts corresponding to the crack a soft plastic tube covered with an aluminum foil and directly connected to the thermal convective device was used. The aluminum foil avoids the influence of the surrounding radiation on the heating process. The thermal measurements were performed with at AVIO TVS 700 thermal camera operating in a spectral range (8–12 μm), with a microbolometer detector (320x240 pixel) – not cooled. The temperature range of the camera was set to (−20; + 400) °C with a thermal sensitivity of 0,07°C (at 30°C). Because the “Ratto delle Sabine” model is mainly composed out of raw clay and wood the emissivity value inserted in the thermal camera was 0,9.
During the local heating process the increase of the surface temperature around the target crack has been monitored in order to control eventual overheating of the other parts of the structure.
Laboratory experiment
In order to evaluate the effectiveness of the method several trials were first performed in laboratory using samples built with materials (e.g. raw clay, wood, metal nails and straw), characterized by thermo-physical properties close to the ones of the statue.
The laboratory test model was made by using hollow bricks, of which the thermal properties are close to the ones of raw clay of the statue. The holes of these bricks were filled with different combinations of the previously mentioned materials. Filling was done in such a way that a free pathway (average diameter of a few millimetres similar to those previously observed in the “Ratto delle Sabine” model) is left to allow for the flow of hot air. Warm air passed through these pathways and allowed us to determine the heating times required for the warming up of the cracks at the outer end whilst avoiding significant and dangerous increases of temperature nearby the inlet of the hot air inflow. Before manufacturing the laboratory samples the characteristics of the free paths (depth and length) to be copied were defined in the field by comparing the radiographic results and the cracking framework, obtained from other studies made within the restoration project.
The bricks simulated the raw clay envelope of the “Ratto delle Sabine” model because both are composed out of the same material: clay (thermal conductivity 0.383 [Wm−1K−1] specific heat 2190 [JKg−1K−1] and density 1.29 [Kgm−3] [24]). The only difference is the procedure of drying: the brick is dried in a furnace (between 800°C and 1200°C [25]) whilst the statue envelope was left to dry at environmental conditions under the natural heating of the sun. Therefore we can consider that the brick simulates the statue envelope in terms of thermo-physical properties.
The results reported in this paper are related to the samples with following dimensions : 15x31.5x4.5 cm. Two types of filling were used: mixed materials (wood, straw, raw clay) and raw clay only. In this way, the two extreme scenarios existing in in- situ are simulated. In fact, on one side the raw clay only sample simulates the more superficial cracks of the statue whilst the sample composed of mixed materials simulates the cracks crossing the parts of the structure composed mainly out of organic materials. The free paths in these samples were realized at a depth less than 2 cm from the surface.