Analysis of the Visual

Photographing the visible fluorescence of ultraviolet radiation shows that the sunlit facade has been painted with zinc white. It turned out to be part of a later overpainting to hide a number of small losses in the underlying paintlayer, damage that had been particularly evident in the bright parts of the painting. The underlying lead white original colour does not fluoresce significantly. Some black spots show old repairs that have not yet aged and therefore do not fluoresce.

The Colour Zinc White – A Double Nature

On one hand praised because it does not yellow in oil paint but at the same time feared because it tends to crack and flake. The latest research has been able to show how the crystalline structure of the pigment interferes with the drying process of oil, hard and brittle layers are formed with weak bonds to one another.

Zinc white, which has a characteristic strongly yellow-green fluorescence, was introduced as a watercolour paint in the latter part of the 18th century but has been known as a mineral since ancient times. For oil painting, it became publicly available as an artists’ colour only from the mid-1830s, and this suggests later overpaintings for example of the area with the the sunlit facade.

1. Under the microscope, the darker small losses are visible through the yellow-white overpainting with zinc white. Micro photo at 30x magnification, image taken by using a Dino-Lite digital microscope.

2. On the second floor of the arcade’s gray-blue wall, the structure of the painting is visible in a small loss of colour. The brown canvas, the white primer, an intermediate layer in brown and the gray-blue wall colour appear under the microscope. Micro photo at 250x magnification, image taken by using a Dino-Lite digital microscope.

3. The red-pink colour from the lips of an over-painted face in the lower left part of the painting is visible in the colour peaks. Micro photo at 250x magnification, image taken by using a Dino-Lite digital microscope.

Inside The Painting

The painting is based on a drawing of the same motif that the artist transferred to canvas, the lines of which one perceives through colour, but which emerges even more clearly when the painting is studied under infrared radiation. With the same radiation, it has been possible to see a face behind the monastery courtyard, which indicates that the artist reused the canvas. There are tangible similarities between the face and one of the heads of three Spartan boys that Eckersberg painted in 1812, during his stay in Paris.

The infrared reflectogram shows a combination of Eckersberg's first motif painted in Paris and the later added architecture of Rome. The signature of the architecture motif from Rome in dry carbon-based medium is visible in the reflectogram because it is made on a light ground layer. It indicates the architecture in a sketchy manner as in contrast to the previous drawing which is much more detailed. In the upper part of the reflectogram there are some auxiliary lines that do not appear to refer to the architecture but probably originate from the earlier underlying motif.

Eckersberg painted this drawing for the study in oil in 1815 in Rome.

It belongs to Statens museum for Kunst. The painting is not exhibited. This digital reproduction is shown to illustrate this painting's connection to the research project around Eckersberg's painting The Cloister in S. Maria in Aracoeli.

Eckerberg painted this student work, Three Spartan Boys Practicing Archery, in David’s studio in Paris 1812. The head, seen in the IRR image, was painted before 1812 in Paris.

The painting is not exhibited. This digital reproduction is shown to illustrate this painting's connection to the research project around Eckersberg's painting The Cloister in S. Maria in Aracoeli.

Analysis of The Painting Within

To better understand the underpainting and the paint layers revealed by infrared reflectography (IRR), a combination of non-invasive and micro-invasive analysis were performed.

Approximate locations of cross-section and spots analysed by X-Ray Fluorescence (XRF).

1. Sample taken from the man’s face (cross-section). In order to analyse the paint layer, a microsample was taken from the area where the man’s face is. The sample was embedded in resin to prepare a so-called cross-section which was analysed by optical microscopy and scanning electron microscopy with energy dispersive X-Ray Spectroscopy (SEM-EDXS).

2. XRF measurement on the face of the man. An infrared reflectogram revealed a largescale underlying composition with a man’s head. X-Ray Fluorescence analysis (XRF) was carried out on selected areas (2, 3) to investigate possible markers on the face of the man below the architecture.

3. XRF measurement outside the face.

In the cross-section one can see the following layers: the ground or preparatory layer, then the paint layers of the man’s face, and finally the architecture on top, as well as a thin layer of retouching in the top left corner.

The thick, warm light pinkish ground layer contains mainly fine and medium size white particles as well as some coarse translucent grey and some small red and light brown particles. In between these, very fine black particles can also be found. Due to the fluorescence observed under UV radiation, we can conclude that the ground layer may consist of at least three applications.

The paint layers above the ground (presumably from the man`s face) are of a dark warm grey colour and contain, as well as white, also black, red and yellow particles.

On top of these layers is a whitish layer with some brown and black particles, which is followed by a darker and cooler grey layer mainly composed of white and black particles (these two are presumably from the architecture of the current composition). Due to the bright greenish fluorescence observed under UV radiation, zinc white is most probably present. Zinc white, together with titanium white, indicates that this layer could be a retouching or a later addition, along with the very thin light grey layer on the left edge of the cross-section.

By identifying the various elements found in the cross-section, we can determine what the sample's different paint layers consist of.

Ground Layer

Calcium and lead are the main elements present in the ground together with small amounts of iron, silicon and aluminium. In some of the very dark coarse particles, the occurrence of calcium and sulphur suggests the presence of calcium sulphate, while the widespread presence of calcium alone shows that calcium carbonate is the dominant component.

Underpainting, the Man's Face

Lead (lead white), antimony (Naples yellow), mercury (vermilion) and iron (earth pigments) are the elements present in the underpainting.

The Layers in the Current Painting

The bottom white layer in the overpainting is mainly composed of lead (lead white) together with a lower amount of aluminium, silicon and very low amounts of iron (probably indicating sporadic earth pigment particles).The top dark grey layer, on the other hand, contains high amounts of lead (lead white) together with lower amounts of zinc (zinc white) and very low amounts of iron, aluminium and silicon.

Top Layer – Retouching

Titanium (titanium white) is the main element in the thin retouching. Traces of zinc.

The complexity of the underpainting and the application of several paint layers are very clear in the backscattered-electron (BSE) images. Heavy elements with high atomic numbers appear bright, while light elements with low atomic numbers look dark. Therefore, BSE images provide preliminary information about the material’s composition and highlight differences and similarities in the paint layer’s structure.

In the analysed cross-section, almost all the layers are primarily composed of white particles, which are bright in BSE image, suggesting the presence – later confirmed by EDX analysis – of lead white.

The X-ray Fluorescence Spectroscopy (XRF) was performed noninvasively and directly on selected points of the painting to investigate the face of the man below the visible architecture. Following the indication of the IRR images, measurements were carried out both on (2) and outside the man’s face (3). The overall compositions were very similar apart from a slightly higher amount of mercury on the areas measured on the face, probably related to the use of vermilion in combination with lead white to create the skin colour.

The elemental analysis also detected variable amounts of zinc and titanium, suggesting that zinc white and titanium white were probably used in the paint layers. The two pigments – industrially produced respectively since the 1830s and 1910s – indicate retouching and later additions to the painting which is dated c. 1815.

The two spectra indicate that zinc (Zn), mercury (Hg) and lead (Pb) are present in the two investigated areas.

The Painting's Story

1810–1813

Eckerberg’s student work, Three Spartan Boys Practicing Archery, painted in David’s studio in Paris 1812. The head, seen in the IRR image, was painted before 1812 in Paris.

1813–1816

Study drawing, at the National Gallery of Denmark 1815. Nationalmuseum’s study in oil is painted in 1815 in Rome. In Eckersberg’s diary, the entry for March 4, 1815 says: ”painted a small portion of the courtyard in the Kapuzin monastery Ara-celi.”

1824

Eckersberg paints the motive of the courtyard on a larger scale and with five figures. (See picture above. This painting is not exhibited but we show a digital reproduction to illustrate the painting's connection to the research project around Eckersberg's painting Klostergården in S. Maria in Aracoeli.)

1854

In Eckersberg’s estate catalogue from February 17, 1854, No. 33, page 4, the painting is described. It says it was sold without a frame. ”Some sketches and incomplete paintings from Rome without frames.”

1886

The Cloister in Rome was demolished in 1886 to make room for the Vittorio Emanuele II monument.

1895

In the catalogue for the Eckersberg exhibition in Copenhagen, October to November 1895, Emil Hannover wrote that the painting was owned by the artist Otto Haslund. The study was probably restored before the exhibition as the stretcher frame can be dated to the late 1800s.

The sketch is mounted on a stretcher frame of a type that came into use in the late 1800s. This makes it likely that the painting was repaired before the Eckersberg exhibition in Copenhagen, October to November 1895. A new support cloth was glued to the original, which resulted in the picture having a thin painted frame around it, which is common for oil sketches.

1916

The study, gift to the NM from the Friends of Nationalmuseum, 1916. It was purchased for 2000 SEK from Winkel & Magnussen, an art dealer in Copenhagen. The museum’s Superintendent Richard Bergh declared that ”it would be highly desirable” to purchase the painting.