ZOLTÁN KIS AND ESZTER SZ. BAKONYI
ZOLTÁN KIS AND ESZTER SZ. BAKONYI
Plate 1 A false colour visualization of the horse based on neutron tomographic greyscale values. The 3D image of the horse is cut into two distinct parts, which were rotated apart from each other. This technique gives a good view into the inside of the object, making the typical sign of indirect lost wax casting visible; the drips indicating the application of a wax intermodel.
Plate 2 Neutron tomographic visualization of the horse in plate 2a and of the rider in plate 2b. The greyscale values characteristic to neutron attenuation are segmented into two distinct groups; bronze material (greyish) and other materials for which the attenuation power is higher than that of the bronze (green). This latter group could contain the remnants of the core, the transfixing and supporting pins, and the organic materials used for patination or repairing the tiny holes in the bronze. These images show the spatial distribution of the two groups, indicating only bronze in the wall of the statuettes and the areas of enrichment of the other materials at other parts of the statuettes; e.g., in the head in the upper insert of plate 2a or near the legs in the lower insert of plate 2a of the horse, and in the shoulder of the rider in plate 2b. Inserted images in plate 2a show a scattered structure, which is layered in the head of the horse. For both the horse and the rider numerical calculations result in a value of about 2% for the ratio of the volume of the second group of materials (apart from transfixing and supporting pins) and the volume of the bronze. In the image of the horse one can clearly see that one of the supporting pins (the green one on the right rear hoof) is made of different material as compared to the left rear hoof.
Plate 3 In a tomographic image with higher resolution in plate 3a, more details of the inner structure of the rider’s upper body are visible. E.g., the casting defects in the bronze wall and the transfixing straight bars or the mounting method of the shield can be seen more clearly. Unfortunately, based only on the neutron attenuation values, materials cannot be distinguished unambiguously. A good example is the case of the core remnants and the black organic patina used on the surface to make the sculptures look more antique in plate 3b. In the planar cuts of the tomographic image of plate 3c, however, the locations are clearly separable, therefore the identification of the materials is more straightforward.
Plate 4 Although their neutron attenuation power is similar, in some visible parts of the rider and the horse the different materials could be identified more easily. In plate 4a showing the rider’s leg one can see a corroded transfixing core pin through an opening. It is shown in the 3D image of plate 4b as a horizontal green bar. With the use of a small magnet its material was proved to be iron. In the photo of the horse’s nose in plate 4c one can see the organic black patina used for making the surface appear more antique. It is shown in the 3D image of plate 4d as two separate green volumes. In the image of the horse’s tail in plate 4e one can see a piece of gypsum plaster through an opening (cf. Shelley Sturman, Katherine May, and Alison Luchs, The Budapest Horse. Beyond the Leonardo da Vinci Question, in Making and Moving Sculpture in Early Modern Italy, Kelley Helmstutler Di Dio [ed.], Farnham 2015, pp. 25–46). As it is not visible in the 3D image of plate 4f we can suppose that it is a moving part of the core material. Unfortunately, there are other interior parts, which are not visible, and therefore the identification of their material is difficult.
Plate 5 The neutron tomographic image (plate 5a) shows an interesting and somewhat puzzling situation inside the head of the horse making visible a unique phenomenon of the casting technique. We can clearly distinguish two adjacent regions; a more homogeneous and solid upper part, and a more heterogeneous lower one. There is a visible border surface between them. The coloured lines in the three planar sections of plate 5b correlate to the coloured planes cutting the head of the horse. Based on the greyscale values, a segmentation of the material into two groups shows that the top and back of the horse’s neck is a solid bronze volume. Moreover, the segmentation pointed out that the heterogeneous lower part is a mixture of bronze (brown) with core residuals or other materials (green). The layered structure of the green material is clearly recognizable in the planar sections of plate 5b.
Plate 6 The comparison of the neutron tomographic and X-ray radiographic images highlights the complementary nature of these two imaging modalities. In both images, the holes from the transfixing pins can be seen. The X-ray image of plate 6a shows e.g., two holes as if they were void regions (black colour). The neutron image of plate 6b shows, however, that one of these holes is filled with a material for which the neutron attenuation power is higher than that of the bronze. It could be a metal or a hydrogen-containing material (e.g., wax) which was used to repair the hole. The insets show tomographic slices at the planes of the two holes.
Plate 7 The spatial distribution of the wall thickness from the neutron tomographic results. Most parts of the rider have a thickness between 1 and 3 mm. Parts with a thickness of more than 3 mm are indicated as magenta regions. This type of analysis gives evidence on the type and quality of casting. We can see the traces of how the wax was slushed during the production of the wax intermodel. It resulted in larger wall thickness at the ends of the mould (head, hands and feet). We can declare that it requires great professional knowledge to cast such a thin-walled sculpture.
Plate 8 The spatial distribution of the porosity from the neutron tomographic results. Most parts of the rider have a rather low porosity, but a few places with porosity above 0.1% are shown as colour coded regions. This type of analysis hints e.g., to the process of casting, to regions with gas bubbles or to inhomogeneous filling of the wall. In this case, the location of the most porous areas indicates the direction of casting; it was executed upside down, from the bottom of the figure, in a position where the feet were held lower than the bottom, and the melted bronze filled the bottom of the model for last. The porosity visible at the neck and the shoulder is due to the construction of the form. Nevertheless, in general the low porosity values prove the high quality of casting.