6th World Conference on NDT and Microanalysis in Diagnostics and Conservation of Cultural and Environmental Heritage, Rome, 1999 May. Published by AIPnD, email: aipnd@numerica.it

TABLE OF CONTENTS

Abstract Introduction Materials and methods Results Conclusions Acknowledgement References

Summary

As a nondestructive research method for works of art X-ray emissiography is less familiar to conservators than X-radiography. This is probably why the method needs intense X-ray source. It affords only surface information of the object, which is different from X-radiography. In the case of paintings we can obtain image information related to oil colors used in the paint layer, because difference in intensities of secondary electrons emitted from adjacent colors records difference on a X-ray film as an emissiogram.

There have been studied on X-ray emissiography from the points of view of possibility and application of the method to works of art. However, very little has been reported on the fundamental study for understanding of emissiograms. In this paper, to improve the method practically we tried to make clear the relationships between the various colors or the structures in paint ings and emissiogram.

As the results, we reconfirmed and found out the following evidences. (1) The photosensitivity to emissiogram is closely related to most heavy atom which consists of the pigment or the other components. (2) The mixture of color with a heavy atom and that with light atoms, the color with a heavy atom dominates the emissiograph image, even if it is small amounts. (3) It was proved that there is no relation between the thickness of paint layer and the emission intensity of secondary electrons. (4) It was proved that the information of ground or underneath colors is able to catch under the condition of ground and/or paint layer.

Introduction

Scientific researches for works of art prior to the conservation practice is indispensable for modern conservators. Electromagnetic waves such as X-rays, ultraviolet rays or infrared rays are useful and available for nondestructive way of the scientific researches. X-radiography can reveal hidden inner information of works of art easily, therefore it has been used very commonly.

On the other hand the X-ray emissiography, we discuss in this paper, is not familiar to conservators. This method is different from X-radiography in some points: it affords only surface information of works of art nondestructively; intense X-rays (200-250KV) and X-ray filter are needed to take a p icture; X-ray films are exposed to not X-rays but secondary electrons from constituent materials of works of art. In the case of oil paintings we can obtain image information related to pigments in the paint layer, because varieties in emission intensities of secondary electrons from constituent elements of pigments brings out varieties in photosensitivity on X-ray film.

There have been studied on X-ray emissiography from the points of view of possibility and application of the method to works of art1,2). However, very little has been reported on the fundamental study for understanding of emissiograms. Through the basic experiments we have understand the characteristic of the X-ray emissiography and have sought a way of application to id entify materials used in a painting. In this paper, to improve the method practically we tried to make clear the relationships between the various colors or the structures in paintings and emissiogram through the following four experiments.

Theme 1 Twenty typical oil colors including inorganic, organic and complex organic pigments painted on raw canvas were examined to clear respective photosensitivity on a film and to draw a general principle in them.

Theme 2 Two combinations of the mixtures consists of two different colors were investigated to determine their photosensitivity and to compare the combination with the single use.

Theme 3 Effect of the thickness of paint layers on emissiograms, in other words, relation between the thickness of paint layers and the emission intensity of secondary electrons was studied.

Theme 4 How about the secondary electrons from the ground color strike the paint layer out and reach to the film? If possible, how about the thickness of the paint layer, and the conditions?

The Theory of X-ray emissiography

In the followings, the theory of X-ray emissiography is explained briefly comparing with that of X-radiography.

Fig 1: Diagramm of X-radiography (left) and X-ray emissiography (right) processes

Fig 2: X-radiogram (left) and X-ray emissiography (right) of a Japanese painting

Fig.1 (left) and Fig.2 (left) show a diagram of X-radiography and X-radiogram of a Japanese painting, respectively. In the method, in order to make a X-radiograph image, X-rays are projected through the painting's surface to strike a sheet of film placed behind the canvas. The selection of X-ray's energy are carried out in accordance with conditions of objects. The X-rays record on the film the various densities of materials through which they pass. All the inner information of the objects are recorded on the film sheet. In the case of painting, the various pigments and other components used create the image of the radiograph. Different materials have different densities. For example, Silver White (basic lead carbonate) is extremely dense, while Lamp Black (carbon) is sparse. In the area of a composition where Silver White is painted, the material so dense that the X-rays can hardly pass through to the film. In other words, the material absorbs almost all the X-rays. The result is a white area on the image. On the other hand, in the area of a composition where Lamp Black is painted, the material so sparse that the X-rays can pass through to the film. The result is a black area on the image. The X-radiogram therefore shows the composition of the painting relative to the positioning of materials.

Fig.1 (right) and Fig.2 (right) shows a diagram of X-ray emissiography and X-ray emissiogram of a Japanese painting, respectively. In X-ray emissiography, only high energy (200-250KV) X-rays are projected to a sheet of film placed above the painting surface through the X-ray filter (Sn sheet) which absorbs the soft X-rays. The result is that only the components of high energy X-rays strike into the painting surface through the film. The film is not exposed by the high energy X-rays because the photosensitivity is very low at the X-rays. Nevertheless, the high energy X-rays can make excited conditions of the component atoms of the materials in the painting. After the excitation the secondary electrons (photoelectrons) are emitted to outside from the atoms which consist of the surface area. The film in contact with the surface of the painting is exposed by these secondary electrons. The amount of emitted electrons is dependent of the atomic number of the component elements of the materials. The atoms having high atomic number emit large amount of secondary electrons. For example, Silver White mainly constituted of basic lead carbonate (Pb/atomic number is 82) shows the intense emission of secondary electrons. The result is that the area where Silver White is painted corresponds a black area on the image. On the other hand, the area where Lamp Black (C/atomic number is 6) is painted corresponds a white area on the image. The X-ray emissiography therefore shows composition of the sur face in the painting relative to distribution of elements of materials. An elaborate understanding of emissiograms is needed to various fundamental information because the black and white on the film is a only scale of information.

Experiments

Through this study the following conditions were adopted to take a picture of emissiogram: Phillips MG-225 type was used for X-ray source; exposure conditions were on 225kV-10mA-3min.; distance between the X-ray source and the film was 70cm; a Sn small sheet (3mm in thickness) was used for X-ray filter which fixed to the source; Fuji FR type industrial X-ray films and the Fuji automatic developing system were used.

In each theme the following original samples were made with oil colors and raw canvases by authors.

• Sample 1: twenty different oil colors were painted smoothly on the raw canvas. The colors used were analyzed by SEM-EDXRF to clear their component elements.

• Sample 2: mixture of two different colors for three concentration ratios (1 /1, 1/3 1/7) were painted on the raw canvas. There were two combinations such as Silver White and Lamp Black, and Cadmium Yellow and Oxide of Chrom, respectively.

• Sample 3: four degrees of thickness (0.1~0.4mm) in their paint layer for six different colors (Lamp Black, Emerald Green, Cadmium Yellow, Cerulean Blue, Vermilion, and Silver White) were painted on the raw canvas.

• Sample 4: three different colors (Emerald Green, Oxide of Chrome, and Permanent Green) were painted over two kinds of the grounds (Silver White and Titanium White). The variation of thickness in paint layers is from 0.1 to 0.4 mm.

Results and Discussion

Theme 1
In order to determine relationship between kinds of colors and photosensitivity to emissiogram, 20 colors were examined. The details of Sample 1 is shown in Fig.3. Every color has own characteristic pigment as material itself or in other components. We could not confirm the difference between inorganic and organic pigments from the emissiograph images.

SILVER WHITE (2PbCO3 - Pb(OH)2)	VERMILION (HgS)	CADMIUMYELLOW (CdS)	OXIDE OF CHROM (Cro)	COBALT BLUE (CoO - nAl2O2)
TITANIUM WHITE (TiO2)	LIGHT RED (Fe2O2)	STRONTIAN YELLOW (TiO2+ organic pigment /Ba Body)	CELADON GREEN (CrO+complex organic pigment /Ca Body)	CERULEAN BLUE HUE (TiO2+complex organic pigment/ Ba Body)
TITANIUM WHITE (Watercolor)	BURNT UMBER (FeO2+MnO2)	PERMANENT YELLOW (Sb205+Ni+Ti+organic pigment)	EMERALD GREEN TiO2+complex organic pigment/Ba Body)	COMPOSE BLUE (Tio2+complex organic pigment/Ba Body)
TITANIUM WHITE (Acrylic Color)	CHINESE RED (organic pigment/Ba Body)	YELLOW OCHRE (Fe2O3 - H2O+Al2O3 - SiO2)	PERMANENT GREEN (complex organic pigment /Ca Body)	ULTRAMARINE (Na2Al6Si6O24S4)
Fig 3: Appearance of Sample 1 (up) and its arrangement of 20 kinds of pigments (down)
Fig 4: X-ray emissiogram of Sample 1

From the emissiogram (Fig.4) of Sample 1, some facts were found out:.

• Pigments including a heavy atom showed black on the emissiograph image, which was detected in Silver White and Vermilion (Type A). They have Pb (atomic number 82) and Hg (80) respectively. Therefore the emission of the secondary electrons is intense in pigments including a heavy atom.

• On the other hand, pigments consists of light atoms (Fe (atomic number 26) , Co (27), Cr (24), Ti (22), Ca (20), Si (14), Al (13), and Na (11) etc.) showed white on the emissiogram, which was detected in Yellow Ocher, Oxide of Chrom, Cobalt Blue and Ultramarine etc. (Type B).

• Neutral grayish tints on the emissiogram were confirmed in Cadmium Yellow , Strontian Yellow, Cerulean Blue, Emerald Green and Compose Blue (Type C). Because Cadmium Yellow has Cd atom (atomic number 48), it is considered to be positioned between Type A and B. As other pigments have Ba body (atomic number 56) as most heavy atom, they are also belonged Type C.

To speak generally, the photosensitivity to emissiogram is closely related to most heavy atom which consists of the pigment or the other components.

Theme 2
In order to clear the effect of the mixing and its ratios of two colors upon the photosensitivity to emissiogram three degrees of mixing ratios for two combinations of colors (Silver White and Lamp Black, and Cadmium Yellow and Oxide of Chrom) were examined. Moreover, relation between difference in the combination of the colors and the emissiograms was also investigated. The details of Sample 2 is shown in Fig.5.

SILVER WHITE	1 : 1	LAMP BLACK	CADMIUM YELLOW	1 : 1	OXIDE OF CHROM
	1 : 3			1 : 3
	1 : 7			1 : 7
Fig 5: Appearance of Sample 2 (up) and its arrangement and compositions of two pigments (down)
Fig 6: X-ray emissiogram of Sample 2

In the case of the combination of Silver White (Pb (atomic number 82) and Lamp Black (C (6)), the appearance of the mixtures were very similar to Lamp Black because the pigment is extremely strong in the coloring power. Reversely, as is detected in Fig.6, Silver White showed black on the emissiogram, while Lamp Black indicated white. Moreover, the mixture of them showed black, even at the mixture of 1/7. As is shown in this case, the mixture of color with a heavy atom and that with light atoms, the color with a heavy atom dominates the emissiograph image, even if it is small amounts.

In the case of the combination of Cadmium Yellow (Cd (atomic number 48) and Oxide of Chrom (Cr (24)), the appearance of the mixtures were depended equivalently on both pigments. Gray tints on emissiograms were observed for the mixtures of the colors. This fact is different from the case of the mixture of Silver White and Lamp Black, and it is considered to be affected by both of two colors. As is shown in this case, the emissiogram of the mixture of two colors consist of medium weight atoms (atomic number 20-50) is depended upon both colors and the mixing ratio.

Theme 3
Relationship between thickness of a paint layer and the emissiogram was investigated in Theme 3. The cross-section of Sample 3 is shown in Fig.7. From the X-radiograms in Fig.8, the thickness of paint layers in Sample 3 were proved to be four degrees. This result was answered the purpose. Generally speaking, thick layers absorb more X-rays than thin layers under the same conditions, which is a character of X-radiography. Reversely, differences in p hotosensitivity were not observed in Sample 3.

Fig 7: Diagram of cross-section of Sample 3

Fig 8: X-ray radiogram (up) and X-ray emissiogram (down) of Sample 3

It was proved that there is no relation between the thickness of paint layer and the emission intensity of secondary electrons.

Theme 4
How about the secondary electrons from the ground color strike the paint layer out and reach to the film? X-ray emissiography has been considered to be offered only surface information. This experiment concerns the thickness of paint layers and the atomic number of most heavy elements which consist of the ground (or underneath layer. Moreover, thickness of paint layer concerns the concealing power which is peculiar to respective oil color.

Fig 9: Arrangement of grounds and paint layers of Sample 4 Fig 10: Diagram of cross-section of Sample 4

Fig 11: X-ray emissiogram of Sample 4

From the results (Fig.11), difference in Silver White ground reflected the emissiogram. This was explained by the constituted element's atomic number, as mentioned above. the following evidences were detected. At some thin areas of the paint layer partially black on emissiogram was observed. The secondary electrons from the ground were considered to be struck the paint layer out and reach to the film. The phenomenon was easily confirmed in most thin area of the paint layer. On the other hand, as is observed in Titanium White ground, the secondary electrons from the ground were not to be struck the paint layer out. Therefore, the color of paint layer dominates the emissiogram completely.

It was proved that the information of ground or underneath colors is ab le to catch under the condition of ground and/or paint layer.

Conclusion

Through these experiments we found out several new evidence about the emissiography technology. We have a dream that emissiography will be applied as a identification of pigments in practical uses. Using the results above mentioned, we could design the "Emissiography Scale", which is made of several different oil colors on a small canvas showing several steps of exposure on emissiogram, to presume pigments used in a painting. These results can be estimated to be profitable to improve the nondestructive technology. In the near future we would like to use X-ray emissiography practically for various scientific researches in woks of art.

References

1. Miura,S.: Emissiography of the Painting by Kuroda, Scientific Papers on Japanese Antiques and Art Crafts (Kobunnkazainokagaku), 30 (1985), 21-27 (in Japanese)
2. Miura,S. and Kamba,N.: Emissiography of a Votive Tablet, Science for Conservation (Hozonnkagaku), 26 (1987), 12-30 (in Japanese)