- Email: [email protected]
The use of radiation in the study of cultural heritage artefacts
Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Radiation Physics and Chemistry journal homepage: www.elsevier.com/locate/radphyschem
The use of radiation in the study of cultural heritage artefacts Dudley Creagh a,n, Vincent Otieno-Alego b, Alana Treasure c, Maria Kubik d, David Hallam e a
Faculty of Education Science Technology and Mathematics, University of Canberra, Canberra, ACT 2601 Australia Australian Forensic Police Forensic Laboratory, GPO Box 40, Canberra, ACT 2601, Australia Australian War Memorial, GPO Box 345, Canberra, ACT, Australia d West Australia Gallery, PO Box 8363 Perth Business Centre, Perth, Western Australia 6849, Australia e RM Tait and Associates, 18 Hayes Street, Queanbeyan, NSW 2620, Australia b c
H I G H L I G H T S
We describe a diverse range of techniques used to study cultural heritage artefacts. IR X-ray and particle beam techniques were used to study: The structure and composition of Australian Indigenous bark paintings. The effects of iron-gall inks on parchment. The results of corrosion and corrosion protection in machinery and vehicles.
ar t ic l e i nf o
a b s t r a c t
Article history: Received 4 October 2015 Received in revised form 19 January 2016 Accepted 29 January 2016
Patrons of art galleries and museums, tourists visiting historic buildings, and sightseers viewing archaeological sites are generally unaware of the extent to which science and technology has contributed to the value of what they see. Many countries rely on cultural tourism to generate national wealth. The use of radiation of many kinds to assist in the conservation/restoration of cultural heritage artefacts is described in this paper. In particular, the paper will describe studies of the pigments used in historic Australian Indigenous art, the degradation of manuscripts written using iron-gall inks, the protection of statues against corrosion and the selection of lubricants for use in old machinery. & 2016 Elsevier Ltd. All rights reserved.
Keywords: Cultural heritage Australia Indigenous art Iron-gall inks Parchment Corrosion prevention
1. Introduction Patrons of art galleries and museums, tourists visiting historic buildings, and sightseers viewing archaeological sites are generally unaware of the extent to which science and technology has contributed to the value of what they see. Many nations rely on cultural heritage tourism to assist in generating foreign exchange. For example: in Australia the nett value of the service industries to which tourism is a significant contributor is comparable to the earnings of the mining industry. Finding appropriate scientific strategies for the solution of problems is therefore very important. The scale of objects and artefacts which comprise our cultural heritage is extremely large. In size it ranges from large archaeological installations such as stone henges and burial mounds, the n
Corresponding author. E-mail address: [email protected] (D. Creagh).
Great Wall, the pyramids, prehistoric towns, and forts, medium scale constructions such as tombs (the Ming Tombs and the Terracotta Warriors burial at Sian), to cave paintings, and to everyday objects such as paintings, statues, and so on. The time span is from 4000 BCE to the present day. Every conceivable type of material has been used. All this makes the task of protecting and conserving our cultural heritage extremely difficult. The methods used in cultural heritage investigations should in principle be non-destructive. The first tenet of conservation is “primum non nocere: first do no harm”. Almost all modern methods involve the use of photon-inphoton-out techniques and therefore meet this criterion. The range of photon wavelengths used to illuminate the artefacts can be from the THz region to the γ-ray region (1000 nm to 0.01 nm). Photon-material interactions may include: elastic scattering, inelastic scattering, Raman scattering, photo-electric absorption and fluorescence.
http://dx.doi.org/10.1016/j.radphyschem.2016.01.040 0969-806X/& 2016 Elsevier Ltd. All rights reserved.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
Other techniques can be used when it is permissible to take samples from the artefacts. These destructive tests can include photon-in photon-out interactions as well as techniques involving particle interactions. Particle interactions include electron diffraction, neutron scattering, particle (usually proton) induced X-ray and γ-ray emission (PIXE; PIGME), Particle accelerators can be used to date specimens found in archaeological sites (See, for example: Creagh and Bradley, 2000). Other equipment used in Cultural Heritage Investigations include: Ground Penetrating Radar (GPR) and satellite remote sensing systems used extensively in archaeological research; portable Infrared and Raman spectroscopy systems used in the study of the composition of small inaccessible objects and images in museums, galleries, caves, and the outdoors; portable and laboratory-based X-ray diffraction (XRD) and X-ray fluorescence (XRF) (Creagh and Bradley, 2000). Synchrotron radiation (SR) facilities are hosts to most of the foregoing techniques (Creagh, 2007). Neutron radiation facilities (both reactor and spallation) provide support to Cultural Heritage Conservation groups. This paper describes how the authors have used a diverse range of the physical techniques to solve particular problems in the field of cultural heritage conservation in the past 25 years. Examples have been chosen to illustrate how radiation of different kinds is used to examine different artefacts.
2. Projects undertaken 2.1. Aboriginal indigenous art Aboriginal communities in Arnhem Land and two major collecting institutions were concerned about the stability of traditionally produced bark paintings. The collecting institutions were concerned about maintaining their collections in good condition and the traditional owners were concerned that paintings which European and Japanese collectors had problems when located in an air conditioned environment. A project was set up to trace the causes of the problem. Barks of the eucalyptus tetradonta tree which grows in the clan regions, and they were prepared in the usual way, namely: they were
placed on a fire and heated on both sides for a time to dessicate the bark. Optical microscopy was then performed on cross sections of the bark. This showed that the heating destroyed the starch grains in the bark leaving a much more homogeneous surface for the painters to paint on. Time lapse photography of the barks in a chamber in which both temperature and humidity could be controlled showed that the barks moved warped significantly when temperature and humidity were cycled. This led to the development of mounts which would gently restrain the warping and to the development of guidelines for the storage and display of the paintings. Arising from this project was the question of whether the provenance of traditional bark paintings could be established by analysing the paintings. The question had arisen because copying of paintings has become a problem, and the indigenous communities depend on the sale of paintings for their livelihood. The principal pigments in bark paintings are ochres, white pigments, black pigments. The provenance of old and new bark paintings is generally known: new bark paintings are sold by the community, and old bark paintings have generally been acquired by collecting agencies from the traditional owners or their representatives. But problems exist when there is a break in the custodial chain. This is most common for old (450 years) paintings. Curators then assign provenance on the basis of painting style. The ochres used in traditionally produced old bark paintings come from a few mine sites (Fig. 1). Each mine produced ochre of its own special character. Some locations produce ochres are red (due to a dominance of Fe2O3), some blood red (due to inter-mixture of Fe3O4), some glisten with particles of muscovite (mica), some have admixtures of goethite (FeO(OH)). And so on. Samples of old ochres taken from these mine sites were given to the project by Dr Mike Smith (National Museum of Australia (NMA)) for reference purposes. These samples, and samples of bark paintings, were analysed using laboratory-based X-ray diffraction and Synchrotron Radiation X-ray diffraction (SRXRD) (Creagh et al., 2007). Because SRXRD requires a much smaller sample size it is the more commonly used analytical technique (Creagh, 2007). Fig. 2 shows the diffraction pattern for a pink ochre. The light red colour is due to an admixture of kaolinite (Al2Si2O5(OH)4).
Fig. 1. Traditional ochre mines in Australia pre-1900.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Fig. 2. SRXRD diffraction pattern of a pink ochre. The Y-axis shows the count-rate of X-rays detected at a particular angle. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 4. SRXRD diffraction pattern for a white pigment.
Fig. 3. PIXE spectrum of a sample taken from NMA42. The Y-axis shows the number counts in 10 s detected at a particular energy.
Very often, when a sample is given for analysis, only a small amount is given. And the question has to be asked: what is the smallest sample size which can be used to give meaningful results. Conventional μ-X-ray fluorescence spectrometry can be used to give atomic compositions. But more compelling evidence can be found by using Particle induced X-ray Emission (PIXE). Fig. 3 shows a typical PIXE spectrum of an ochre (NMA42). The elemental composition is deduced from the areas under of each of the peaks. The variation of the deduced composition with sample size is shown in Table 1. The elemental composition remains substantially the same even for 67 μg of material. The question as to whether or not provenance can be established solely by analysing the ochres was, however, answered in the negative. No particular ochre was unique to a particular clan region. The reason for this is that ochres were highly prized and were traded between the clans. South Australian newspapers of the 1880s and 1890s record large aboriginal groups from several hundred kilometres to the north coming to acquire the ochres form Bookatu. Some of this would have been traded with clans as they returned north. Anthropologists are now working with the Table 1 The percentage variation of elemental composition with sample size using PIXE for the sample NMA42. Mass (mg)
Ca%
Fe%
Mn%
Si%
P%
Al%
0.067 0.512 0.772 1.727
78.4 74.2 78.0 78.2
10.6 9.7 10.0 9.4
0.3 0.3 0.3 0.3
6.2 7.6 6.6 5.8
1.7 2.5 1.6 1.3
0.57 0.46 0.62 0.42
indigenous communities to rediscover the ancient trade routes by interpreting their oral tradition. But is there another way? Ochre was prized, so it became a trade good. White pigment was not prized, and in fact it was the task of the women to mine the white pigment from the banks of watercourses. Each family has its own location from which the pigment was extracted. White pigment was collected from a number of locations and analysed using SRXRD. Fig. 4 shows the diffraction pattern for one of several white pigments selected. Each is unique to a particular family group. No two pigments have the same mineral composition. Minerals found include kaolinite, huntite, quatz, talc and hydroxyapatite. If sufficient resources could be found it would be possible to extend this small pilot project and produce a “provenance map” for the white pigments. The mineral composition of the pigments analysed is shown in Table 2. 2.2. Iron gall inks on parchment Much of the Western World's history has been recorded using iron-gall inks on parchment. Iron gall inks were used from at least the 5th Century AD until the 20th century. Since the invention of the printing press by Gutenberg in 1440 AD manuscripts (documents written by hand) were largely supplanted by the Table 2 Mineral composition for three of the white pigments. Sample
Mineral
%
1
Kaolinite Huntite Talc Muscovite Talc Hydroxyapatite Huntite Quatz (Amorphous)
74 18 2 3 98 2 81 4 15
2 3
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
4
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
printed page. However many legal documents (contracts, deeds, and so on) were still written well into the 20th century. The Australian Constitution was written using iron-gall ink. And it was the need to conserve this important document which led to our interest in determining the processes which leads to the degradation of documents written using iron-gall inks. A great difference exists, for example, between the physical state of some manuscripts (for example: the Belles Heures de Jean de France, Duc de Berry (produced by the Limbourg Brothers in 1405 AD and which is held in the Metropolitan Museum of Art, New York (Husband, 2008)) and some of manuscripts, for example: music scores written by Johann Sebastian Bach some 300 years later (Bach Museum, Leipsig). The former seems to be as bright and new as the day it was written (iron-gall ink on vellum). In some of the latter (iron-gall ink on paper) the iron-gall ink has attacked the paper to such an extent it is possible to see through through the writing when the page is held up to the light. The work discussed here is confined in scope to a study of irongall ink on parchment. We wished initially to study the structure of parchment, its preparation, and the manner in which it ages. Parchment is manufactured from the skins of calves, sheep, or goats. The skins are soaked and stretched on frames. Skin is predominately composed of collagen which is a triply stranded polypeptide chain which self-organises to form fibrils which then organise to form a 2D-mesh-like structure (Fig. 5). Collagen fibres are typically 50–300 μm in diameter. The stretching process orients the collagen fibres along the direction of stretching and in this state the collagen is like a glued structure of fibres (κολλα means “glue” in Greek). The hides are then scraped to remove hair and blemishes. The hides are rubbed down with pumice to produce smooth surfaces which will readily accept the ink. In the last phase of preparation the hides are treated with a “white paste” (lime, flour, egg whites, milk) to prevent the ink from spreading. Small Angle and Wide Angle X-ray Scattering (SAXS, WAXS) patterns of collagen from a new skin show a ring pattern which illustrates the axial ordering of the collagen fibres along the direction of scattering (Kennedy and Wess, 2006) (Fig. 6). As the parchment degrades, by dehydration, exposure to sunlight, mould activity, and other environmental factors, the oriented structure loses definition. At this stage the “glue” is being lost and the capacity for the ink to adhere to the parchment is diminished. Until the 19th century iron-gall inks were made locally, very
Fig. 6. WAXS pattern for gently supported parchment.
often by households, and there are very many variants of the recipe. Iron-gall inks are organo-metallic inks made by dissolving oak galls in FeSO4 and mixing the filtered solution with gum Arabic. Fe2 þ SO4 þtannin⇒Fe3 þ tannin complex þsulphuric acid The nature of the Fe3 þ tannin complex depends on the composition of the oak galls to some extent. The existence of the sulphuric acid possibly assists in the binding of the ink to the parchment initially, but could perhaps be a problem in the long term. Initially studies of the adhesion of the inks to the parchment were undertaken. Changes in the distribution of atomic species have been studied by: laboratory-based micro-XRF spectroscopy; scanning electron microscopy/EDS; synchrotron radiation-micro-XRF Structural changes to molecular entities have been studied by vibrational spectroscopy techniques: Near Infrared Raman micro-
Fig. 5. Schematic diagram of the development of a collagen fibre.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
5
Fig. 7. Examples of heavily damaged lettering.
Fig. 8. Atomic species map for portion of the right hand image in Fig. 7. Clockwise from the top left: calcium, iron, potassium, sulphur, phosphorus, copper.
spectroscopy (Lee et al., 2006). Synchrotron radiation-IR-microscopy (SRIR) (Treasure et al., 2012). It is of interest to determine the spread of an ink stroke both laterally across the parchment and into the depth of the parchment. The research was carried out on a 19th century indenture document. Letters were chosen from which the ink stroke had completely or partially detached. (Fig. 7). Fig. 8 show atomic species maps taken using a Kevex Omnichrom micro-XRF spectrometer operated at 14 kV. The calcium signal is weak and evenly distributed. The final stages of the smoothing of the parchment with “white powder”. The iron and sulphur distributions are well defined but the width of the sulphur signal is wider and more diffuse than the iron signal
indicating that the sulphate ion has migrated away from the iron to some extent. Details of the surface structure in the region of a letter from which the some of the ink has been lost were examined using a Cambridge Stereoscan Scanning Electron Microscope with an Energy Dispersive X-ray Analysis (EDAX) attachment. Fig. 9 shows a region from which the ink has been lost. The box highlights a region where collagen a tangled mesh of collagen fibres can be seen. In the corroded region outlined by a rectangle in Fig. 10 EDAX scans show that there has been a 120 times increase in the percentage of sulphur, This supports the belief that the migration of sulphuric acid has weakened the bonding between the collagen fibres.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
Fig. 9. Region of a (Magnification ¼ 80).
pen
stroke
from
which
the
ink
has
been
lost
Further evidence of the effect of the ink on the collagen structure can be elicited by undertaking line scans across a pen stroke using SRIR. Degradation of collagen is displayed by changes in the Amide I and Amide II bands. These are related to the C ¼O stretch which is sensitive to hydrolysis. Fig. 10 shows the result of a scan across a pen stroke taken using the IR spectroscopy beamline using a Bruker Hyperion using an ATR objective at a magnification of 100. The IR beam diameter was 5 μm. The Amide I peak at 1654 cm 1 remains constant but its subband (shown here as a yellow colour) varies significantly during the scan. The group of bands (3000–3500 cm 1) which are associated with N–H and –OH stretching modes exhibit a similar behaviour. Cross sections have to be cut across a pen stroke if the depth of penetration of the ink is to be established and any migration of atomic species normal to the surface is to be observed. This was studied using the Scanning X-ray Fluorescence Microscopy beamline at the Australian Synchrotron (Treasure et al., 2012). Fig. 11 shows the analysis grid used to scan the cross section underlying a pen stroke. As well it shows the spatial distributions of the two principal atomic species: iron and sulphur. The spatial distribution for the iron is much more uniform than for the sulphur. As well, the sulphur has migrated further into the parchment than the iron. 2.3. Study of corrosion and corrosion prevention
Fig. 10. SRIR spectral scans across a pen stroke.
Another research programme addressed the need to protect statues from corrosion and to develop lubricating oils to protect old vehicles which are still in operational use in museums and galleries [5]. This work involves the use electrochemical impedance spectroscopy and Synchrotron Radiation Grazing Incidence Diffraction (SRGIXD) [2]. Photographs of the two extremes for which techniques of protection against corrosion have to be found are shown in Fig. 12a and b. Outdoor statues must be protected against: the climate (sun, rain, frost, snow, etc.); the environment (man-made pollution, wind-borne particulates; and natural hazards (bird-lime, windborne sap, etc.). They need a protective coating which cling closely and strongly to the surface and act like a blanket to exclude moisture and pollutants. And the protective layer must maintain its integrity for years. But it must be capable of being removed from the surface without harming the surface.
Fig. 11. Spatial distributions of sulphur and iron concentrations for the sampling grid on the left for the cross section of a pen stroke.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
7
a. Schematic diagram of an X-ray beam interacting with a film deposited on a substrate.
sem image of a wax layer brushed into a bronze substrate before heat treatment (15 kV)
b. Keissig fringes arising from a monolayer on sodium dodecylsulphonate on an aluminium substrate at a wavelength of 0.1nm. Fig. 13. a. Schematic diagram of an X-ray beam interacting with a film deposited on a substrate. b. Keissig fringes arising from a monolayer on sodium dodecylsulphonate on an aluminium substrate at a wavelength of 0.1 nm.
sem image of a wax layer brushed into a bronze substrate after heat treatment. (15 kV) Fig. 12. a: sem image of a wax layer brushed into a bronze substrate before heat treatment (15 kV), b: sem image of a wax layer brushed into a bronze substrate after heat treatment (15 kV).
Machinery which is required to be operated or driven must have lubricating oils for their moving parts which can: withstand the heat and mechanical stresses generated whilst the equipment is in operation; immediately form a protective barrier when the equipment ceases operation; does not contain additives which will adversely affect the leather, cork and other gasket materials used in old engines, gear boxes, and differentials. The protective coatings have to have the characteristics of selforganising aliphatic compounds (laurates, stearates, palmates, etc.). These aliphates are long chain compounds consisting of many CH-bonds ending with a chemically active headgroup. For statues number of different preparations are sold and claims are made about their efficacy are made by the manufacturers. To test these claims test surfaces of different bronzes and patinated bronzes were coated with a variety of anti-corrosion products. Fig. 12a shows a scanning electron microscope image of a microcrystalline wav which has been dissolved in solvent, brushed onto the surface, in this case a bronze surface. The wax has not completely covered the surface, and would be ineffective as a protection against corrosion. Fig. 13b shows an image taken after the surface had been heated. The clumps of wax have been distributed more evenly across the surface, and the protective coating is acting more like a protective blanket (Creagh et al., 1995). Sulphonates have been the prime chemical building blocks
used by manufacturers to create corrosion prevention products for the automotive industry. They are made by the sulphonation of, and the removal of alkylated aromatic hydrocarbons from refined lubricating oils. In 1941 alkaline earth salts were added to create detergents. Later research led to their development as water displacing corrosion preventatives (WDCPs). Initial research into these WDPCs was undertaken to gain an understanding of how they interacted with the surface on which they were deposited. To study the thickness and other properties of the layer the techniques of Synchrotron Radiation X-ray reflectivity (SRXR) and Grazing incidence X-ray Diffraction GIDX) were used (Hallam et al., 1997). Fig. 13a shows a schematic diagram showing how XRR can be used to measure film thickness. The difference in path length between the X-ray beam which is externally reflected at the surface of the protective layer and the beam which is reflected by the substrate. As the sample is rotated through the angle of total reflection (the refractive index for X-rays for an air-material interface is less than 1) a series of fringes (Keissig fringes) can be observed. The film thickness can be inferred from the periodicity of the fringes. (θ2 ¼ m2 (λ/2d)2 þ αc2; where θ is the angle of reflection; λ is the X-ray wavelength; d is the film thickness; αc is the critical angle; m is an integer related to the order of the interference pattern). Plotting θ2 versus m2 gives a straight line with slope (λ/d)2 and intercept αc2. Fig. 13b shows the Keissig fringes which result from the interaction of an X-ray beam with a monolayer film of sodium dodecyl sulphonate on an aluminium substrate in a GIXD experiment (Creagh et al., 2000). GIXD can determine whether any in-plane organisation exists for this system. Layers of a sodium stearate, a similar were deposited on a silicon substrate using the Langmuir–Blodgett process. Fig. 14a shows that diffraction spots exist, and this indicates
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
8
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
a. GIXD pattern for stearate layers deposited on a silicon substrate at a wavelength of 1nm.
a. Right: schematic diagram of an EIS system. Left:
cartoon illustrating the processes taking place at the anode.
b. Chemical modelling of how the sodium dod ecylsulphona te molecules might attach to a bronze substrate. Fig. 14. a. GIXD pattern for stearate layers deposited on a silicon substrate at a wavelength of 1 nm. b. Chemical modelling of how the sodium dodecylsulphonate molecules might attach to a bronze substrate.
that long range ordering in the diffraction plane is occurring. X-ray intensity is observed between the spots indicating the ordering is not perfect over the area probed by the X-ray beam. Fig. 14b shows a depiction using chemical modelling which demonstrates how the sodium dodecylsulphonate layers might organise on a bronze substrate. The head-group has attached itself to the surface and the molecules have packed themselves together forming a hexagonal array on the surface. The stearate chains are angled with respect to the surface because of the stereochemistry of the sodium sulphonate-group These insights into the behaviour of protective coatings, however, do not assist with the problem of how one deals with the assessment of assess the deterioration of coatings and which products are most suitable for use. Electrochemical Impedance Spectroscopy (EIS) is able to provide a solution. In EIS the surface under investigation is made the anode of an electrochemical cell. Fig. 15a shows a schematic diagram of the region adjoining the anode of a basic EIS system. As well it shows a cartoon illustrating the processes taking place at the interface and the electrical equivalent circuit. For a bare metal inserted into an electrolyte ions which are absorbed at the metal–solution interface and create a charged double layer This behaves like a leaky capacitor (a resistance in parallel with a capacitance) whilst the electrolyte is purely resistive. When a coating is added to the surface an additional capactiance (due to the non-conducting coating) is added which is large compared to the dipole layer capacitance, and is in series with the resistance of the electrolyte. So long as the coating maintains its integrity the low frequency impedance is high. At the frequency increases the impedance reduces as E 1/ω2. If pores exist in the coating resistive leakage paths exist, which bypass the
b. Bode impedance plots for a number of products alleged to provide good corosion protection. Fig. 15. a. Right: schematic diagram of an EIS system. Left: cartoon illustrating the processes taking place at the anode. b. Bode impedance plots for a number of products alleged to provide good corosion protection.
capacitances. The pores may exist because of poor application of the coating initially, or more likely, through degradation of the layer. At high frequencies the impedance becomes purely resistive. In practice the testing of statues and other static objects is made using a portable cell in which contact with the surface is made using a moisture retaining pad containing the electrolyte (usually non-gaseous mineral water) (Letardi, 2000). Fig. 15b shows a Bode impedance plot in which the corrosive resistance has been measured for a number of lubricant samples on steel. A good lubricant is characterized by a high low frequency impedance and a smooth high frequency roll-off. Only one of the products shown has the desired specifications. All the others exhibit varying degrees of porosity.
3. Conclusions In this paper three of the many projects the authors have been associated with have been outlined. There have been many more. Notable other projects include: the restoration of a Japanese fighter, study of important medals (the Victoria Cross and the Lusitania medal), studies of the degradation of elastomers, the effect of moisture on composite materials, the study of the armour of the Australian outlaw Joe Burns, et cetera. All have required the use of a wide range of radiation-based technologies to solve the questions posed by curators and conservators.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Acknowledgements The authors are indebted to the Australian Research Council (LE0346515), the National Capital Development Commission, the Australian Synchrotron, the Australian Synchrotron Research Program, and the Australian Nuclear Science and Technology Organisation for direct funding and to the Australian War Memorial, the National Gallery of Australia, the National Film and Sound Archive, the National Museum of Australia for in kind assistance. Many scientists have assisted us, and we are very appreciative of their time and effort. They include: Milan Stoilovic (AFP Forensic Laboratory); Mark Tobin, Ljiliana Puskar, David Paterson, Martin de Jonge (ASP); James Hester, Gary Foran (ASRP); Australian National University (Research School of Chemistry – Graham Heath; Engineering – Adrian Lowe; Research School of Earth Sciences).
References Creagh, D.C., 2007. Synchrotron radiation and its use in art, archaeometry, and cultural heritage. In: Creagh, D.C., Bradley, D.A. (Eds.), Physical Principles in Art, Archaeometry and Cultural Heritage vol. 2. Elsevier, The Netherlands, pp. 1–91, ISBN:13-978-0-444-52856-8. Creagh, D.C., Bradley, D.A. (Eds.), 2000. Radiation in Art and Archaeometry. Elsevier,
9
Amsterdam, The Netherlands, ISBN: 0 444 050487 7. Creagh, D.C., Hallam, D.L., Price, C.L., 1995. Testing of Waxes for the Protecting of Memorials. National Capiltal Planning Authority, Canberra, Australia. Creagh, D.C., Kubik, M.E., Sterns, M., 2007. On the feasibility of establishing the provenance of Australian Aboriginal artefacts using synchrotron radiation X-ray diffraction and proton induced X-ray emission. NIM A580, 721–724. Creagh, D.C., Otieno-Alego, V., O’Neill, P.M., 2000. X ray reflectivity and grazing incidence diffraction studies of the adhesion of protective wax coatings on metal surfaces. In: Proceedings of the Eighth Meeting of the International Radiation Physics Society, Prague, Czech Republic. 4–10 June, 2000. Hallam, D.L., Adams, C.D., Creagh, D.C., Holt, S.A., Wanless, E.J., Senden, T.J., Heath, G. A., 1997. Petroleum sulphonate inhibitors. In: McLeod, I.D., Pennec, S.L., Robbiola., L. (Eds.), Metal 95. James and James, London, UK, pp. 159–161, ISBN: 1 873936 67 2. Husband, T.B., 2008. The Art of Illumination. The Metropolitan Museum of Art, New York, ISBN 987-1-58839-294-7. Kennedy, C.J., Wess, T.J., 2006. The use of X-ray scattering to analyze parchment and degradation. In: Bradley, D.A., Creagh, D.C. (Eds.), Principles in Art, Archaeometry and Cultural Heritage vol. 1. Elsevier, The Netherlands, pp. 159–173, ISBN13: 978-0-444-52131-6. Lee, A.S., Mahon, P.J., Creagh, D.C., 2006. Raman analysis of iron gall inks on parchment. J. Vib. Spectrosc. 40, 170–175. Letardi, P., 2000. Electrochemical methods in the conservation of metals. In: Creagh, D.C., Bradley, D.A. (Eds.), Radiation in Art and Archaeometry. Elsevier, Amsterdam, The Netherlands, pp. 15–39, ISBN: 0 444 050487 7. Treasure, A.E., Creagh, D.C., Whitely, C., Howard, D., de Jonge, M., 2012. SR-μXRF in the analysis of iron-gall inks on parchment. In: Streli, C., Mantler, M. (Eds.), Proceedings of the European X-ray Radiation Spectroscopy Conference. Vienna, Austria. 17–22 June, 2012.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
Contents lists available at ScienceDirect
Radiation Physics and Chemistry journal homepage: www.elsevier.com/locate/radphyschem
The use of radiation in the study of cultural heritage artefacts Dudley Creagh a,n, Vincent Otieno-Alego b, Alana Treasure c, Maria Kubik d, David Hallam e a
Faculty of Education Science Technology and Mathematics, University of Canberra, Canberra, ACT 2601 Australia Australian Forensic Police Forensic Laboratory, GPO Box 40, Canberra, ACT 2601, Australia Australian War Memorial, GPO Box 345, Canberra, ACT, Australia d West Australia Gallery, PO Box 8363 Perth Business Centre, Perth, Western Australia 6849, Australia e RM Tait and Associates, 18 Hayes Street, Queanbeyan, NSW 2620, Australia b c
H I G H L I G H T S
We describe a diverse range of techniques used to study cultural heritage artefacts. IR X-ray and particle beam techniques were used to study: The structure and composition of Australian Indigenous bark paintings. The effects of iron-gall inks on parchment. The results of corrosion and corrosion protection in machinery and vehicles.
ar t ic l e i nf o
a b s t r a c t
Article history: Received 4 October 2015 Received in revised form 19 January 2016 Accepted 29 January 2016
Patrons of art galleries and museums, tourists visiting historic buildings, and sightseers viewing archaeological sites are generally unaware of the extent to which science and technology has contributed to the value of what they see. Many countries rely on cultural tourism to generate national wealth. The use of radiation of many kinds to assist in the conservation/restoration of cultural heritage artefacts is described in this paper. In particular, the paper will describe studies of the pigments used in historic Australian Indigenous art, the degradation of manuscripts written using iron-gall inks, the protection of statues against corrosion and the selection of lubricants for use in old machinery. & 2016 Elsevier Ltd. All rights reserved.
Keywords: Cultural heritage Australia Indigenous art Iron-gall inks Parchment Corrosion prevention
1. Introduction Patrons of art galleries and museums, tourists visiting historic buildings, and sightseers viewing archaeological sites are generally unaware of the extent to which science and technology has contributed to the value of what they see. Many nations rely on cultural heritage tourism to assist in generating foreign exchange. For example: in Australia the nett value of the service industries to which tourism is a significant contributor is comparable to the earnings of the mining industry. Finding appropriate scientific strategies for the solution of problems is therefore very important. The scale of objects and artefacts which comprise our cultural heritage is extremely large. In size it ranges from large archaeological installations such as stone henges and burial mounds, the n
Corresponding author. E-mail address: [email protected] (D. Creagh).
Great Wall, the pyramids, prehistoric towns, and forts, medium scale constructions such as tombs (the Ming Tombs and the Terracotta Warriors burial at Sian), to cave paintings, and to everyday objects such as paintings, statues, and so on. The time span is from 4000 BCE to the present day. Every conceivable type of material has been used. All this makes the task of protecting and conserving our cultural heritage extremely difficult. The methods used in cultural heritage investigations should in principle be non-destructive. The first tenet of conservation is “primum non nocere: first do no harm”. Almost all modern methods involve the use of photon-inphoton-out techniques and therefore meet this criterion. The range of photon wavelengths used to illuminate the artefacts can be from the THz region to the γ-ray region (1000 nm to 0.01 nm). Photon-material interactions may include: elastic scattering, inelastic scattering, Raman scattering, photo-electric absorption and fluorescence.
http://dx.doi.org/10.1016/j.radphyschem.2016.01.040 0969-806X/& 2016 Elsevier Ltd. All rights reserved.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
Other techniques can be used when it is permissible to take samples from the artefacts. These destructive tests can include photon-in photon-out interactions as well as techniques involving particle interactions. Particle interactions include electron diffraction, neutron scattering, particle (usually proton) induced X-ray and γ-ray emission (PIXE; PIGME), Particle accelerators can be used to date specimens found in archaeological sites (See, for example: Creagh and Bradley, 2000). Other equipment used in Cultural Heritage Investigations include: Ground Penetrating Radar (GPR) and satellite remote sensing systems used extensively in archaeological research; portable Infrared and Raman spectroscopy systems used in the study of the composition of small inaccessible objects and images in museums, galleries, caves, and the outdoors; portable and laboratory-based X-ray diffraction (XRD) and X-ray fluorescence (XRF) (Creagh and Bradley, 2000). Synchrotron radiation (SR) facilities are hosts to most of the foregoing techniques (Creagh, 2007). Neutron radiation facilities (both reactor and spallation) provide support to Cultural Heritage Conservation groups. This paper describes how the authors have used a diverse range of the physical techniques to solve particular problems in the field of cultural heritage conservation in the past 25 years. Examples have been chosen to illustrate how radiation of different kinds is used to examine different artefacts.
2. Projects undertaken 2.1. Aboriginal indigenous art Aboriginal communities in Arnhem Land and two major collecting institutions were concerned about the stability of traditionally produced bark paintings. The collecting institutions were concerned about maintaining their collections in good condition and the traditional owners were concerned that paintings which European and Japanese collectors had problems when located in an air conditioned environment. A project was set up to trace the causes of the problem. Barks of the eucalyptus tetradonta tree which grows in the clan regions, and they were prepared in the usual way, namely: they were
placed on a fire and heated on both sides for a time to dessicate the bark. Optical microscopy was then performed on cross sections of the bark. This showed that the heating destroyed the starch grains in the bark leaving a much more homogeneous surface for the painters to paint on. Time lapse photography of the barks in a chamber in which both temperature and humidity could be controlled showed that the barks moved warped significantly when temperature and humidity were cycled. This led to the development of mounts which would gently restrain the warping and to the development of guidelines for the storage and display of the paintings. Arising from this project was the question of whether the provenance of traditional bark paintings could be established by analysing the paintings. The question had arisen because copying of paintings has become a problem, and the indigenous communities depend on the sale of paintings for their livelihood. The principal pigments in bark paintings are ochres, white pigments, black pigments. The provenance of old and new bark paintings is generally known: new bark paintings are sold by the community, and old bark paintings have generally been acquired by collecting agencies from the traditional owners or their representatives. But problems exist when there is a break in the custodial chain. This is most common for old (450 years) paintings. Curators then assign provenance on the basis of painting style. The ochres used in traditionally produced old bark paintings come from a few mine sites (Fig. 1). Each mine produced ochre of its own special character. Some locations produce ochres are red (due to a dominance of Fe2O3), some blood red (due to inter-mixture of Fe3O4), some glisten with particles of muscovite (mica), some have admixtures of goethite (FeO(OH)). And so on. Samples of old ochres taken from these mine sites were given to the project by Dr Mike Smith (National Museum of Australia (NMA)) for reference purposes. These samples, and samples of bark paintings, were analysed using laboratory-based X-ray diffraction and Synchrotron Radiation X-ray diffraction (SRXRD) (Creagh et al., 2007). Because SRXRD requires a much smaller sample size it is the more commonly used analytical technique (Creagh, 2007). Fig. 2 shows the diffraction pattern for a pink ochre. The light red colour is due to an admixture of kaolinite (Al2Si2O5(OH)4).
Fig. 1. Traditional ochre mines in Australia pre-1900.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
3
Fig. 2. SRXRD diffraction pattern of a pink ochre. The Y-axis shows the count-rate of X-rays detected at a particular angle. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 4. SRXRD diffraction pattern for a white pigment.
Fig. 3. PIXE spectrum of a sample taken from NMA42. The Y-axis shows the number counts in 10 s detected at a particular energy.
Very often, when a sample is given for analysis, only a small amount is given. And the question has to be asked: what is the smallest sample size which can be used to give meaningful results. Conventional μ-X-ray fluorescence spectrometry can be used to give atomic compositions. But more compelling evidence can be found by using Particle induced X-ray Emission (PIXE). Fig. 3 shows a typical PIXE spectrum of an ochre (NMA42). The elemental composition is deduced from the areas under of each of the peaks. The variation of the deduced composition with sample size is shown in Table 1. The elemental composition remains substantially the same even for 67 μg of material. The question as to whether or not provenance can be established solely by analysing the ochres was, however, answered in the negative. No particular ochre was unique to a particular clan region. The reason for this is that ochres were highly prized and were traded between the clans. South Australian newspapers of the 1880s and 1890s record large aboriginal groups from several hundred kilometres to the north coming to acquire the ochres form Bookatu. Some of this would have been traded with clans as they returned north. Anthropologists are now working with the Table 1 The percentage variation of elemental composition with sample size using PIXE for the sample NMA42. Mass (mg)
Ca%
Fe%
Mn%
Si%
P%
Al%
0.067 0.512 0.772 1.727
78.4 74.2 78.0 78.2
10.6 9.7 10.0 9.4
0.3 0.3 0.3 0.3
6.2 7.6 6.6 5.8
1.7 2.5 1.6 1.3
0.57 0.46 0.62 0.42
indigenous communities to rediscover the ancient trade routes by interpreting their oral tradition. But is there another way? Ochre was prized, so it became a trade good. White pigment was not prized, and in fact it was the task of the women to mine the white pigment from the banks of watercourses. Each family has its own location from which the pigment was extracted. White pigment was collected from a number of locations and analysed using SRXRD. Fig. 4 shows the diffraction pattern for one of several white pigments selected. Each is unique to a particular family group. No two pigments have the same mineral composition. Minerals found include kaolinite, huntite, quatz, talc and hydroxyapatite. If sufficient resources could be found it would be possible to extend this small pilot project and produce a “provenance map” for the white pigments. The mineral composition of the pigments analysed is shown in Table 2. 2.2. Iron gall inks on parchment Much of the Western World's history has been recorded using iron-gall inks on parchment. Iron gall inks were used from at least the 5th Century AD until the 20th century. Since the invention of the printing press by Gutenberg in 1440 AD manuscripts (documents written by hand) were largely supplanted by the Table 2 Mineral composition for three of the white pigments. Sample
Mineral
%
1
Kaolinite Huntite Talc Muscovite Talc Hydroxyapatite Huntite Quatz (Amorphous)
74 18 2 3 98 2 81 4 15
2 3
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
4
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
printed page. However many legal documents (contracts, deeds, and so on) were still written well into the 20th century. The Australian Constitution was written using iron-gall ink. And it was the need to conserve this important document which led to our interest in determining the processes which leads to the degradation of documents written using iron-gall inks. A great difference exists, for example, between the physical state of some manuscripts (for example: the Belles Heures de Jean de France, Duc de Berry (produced by the Limbourg Brothers in 1405 AD and which is held in the Metropolitan Museum of Art, New York (Husband, 2008)) and some of manuscripts, for example: music scores written by Johann Sebastian Bach some 300 years later (Bach Museum, Leipsig). The former seems to be as bright and new as the day it was written (iron-gall ink on vellum). In some of the latter (iron-gall ink on paper) the iron-gall ink has attacked the paper to such an extent it is possible to see through through the writing when the page is held up to the light. The work discussed here is confined in scope to a study of irongall ink on parchment. We wished initially to study the structure of parchment, its preparation, and the manner in which it ages. Parchment is manufactured from the skins of calves, sheep, or goats. The skins are soaked and stretched on frames. Skin is predominately composed of collagen which is a triply stranded polypeptide chain which self-organises to form fibrils which then organise to form a 2D-mesh-like structure (Fig. 5). Collagen fibres are typically 50–300 μm in diameter. The stretching process orients the collagen fibres along the direction of stretching and in this state the collagen is like a glued structure of fibres (κολλα means “glue” in Greek). The hides are then scraped to remove hair and blemishes. The hides are rubbed down with pumice to produce smooth surfaces which will readily accept the ink. In the last phase of preparation the hides are treated with a “white paste” (lime, flour, egg whites, milk) to prevent the ink from spreading. Small Angle and Wide Angle X-ray Scattering (SAXS, WAXS) patterns of collagen from a new skin show a ring pattern which illustrates the axial ordering of the collagen fibres along the direction of scattering (Kennedy and Wess, 2006) (Fig. 6). As the parchment degrades, by dehydration, exposure to sunlight, mould activity, and other environmental factors, the oriented structure loses definition. At this stage the “glue” is being lost and the capacity for the ink to adhere to the parchment is diminished. Until the 19th century iron-gall inks were made locally, very
Fig. 6. WAXS pattern for gently supported parchment.
often by households, and there are very many variants of the recipe. Iron-gall inks are organo-metallic inks made by dissolving oak galls in FeSO4 and mixing the filtered solution with gum Arabic. Fe2 þ SO4 þtannin⇒Fe3 þ tannin complex þsulphuric acid The nature of the Fe3 þ tannin complex depends on the composition of the oak galls to some extent. The existence of the sulphuric acid possibly assists in the binding of the ink to the parchment initially, but could perhaps be a problem in the long term. Initially studies of the adhesion of the inks to the parchment were undertaken. Changes in the distribution of atomic species have been studied by: laboratory-based micro-XRF spectroscopy; scanning electron microscopy/EDS; synchrotron radiation-micro-XRF Structural changes to molecular entities have been studied by vibrational spectroscopy techniques: Near Infrared Raman micro-
Fig. 5. Schematic diagram of the development of a collagen fibre.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
5
Fig. 7. Examples of heavily damaged lettering.
Fig. 8. Atomic species map for portion of the right hand image in Fig. 7. Clockwise from the top left: calcium, iron, potassium, sulphur, phosphorus, copper.
spectroscopy (Lee et al., 2006). Synchrotron radiation-IR-microscopy (SRIR) (Treasure et al., 2012). It is of interest to determine the spread of an ink stroke both laterally across the parchment and into the depth of the parchment. The research was carried out on a 19th century indenture document. Letters were chosen from which the ink stroke had completely or partially detached. (Fig. 7). Fig. 8 show atomic species maps taken using a Kevex Omnichrom micro-XRF spectrometer operated at 14 kV. The calcium signal is weak and evenly distributed. The final stages of the smoothing of the parchment with “white powder”. The iron and sulphur distributions are well defined but the width of the sulphur signal is wider and more diffuse than the iron signal
indicating that the sulphate ion has migrated away from the iron to some extent. Details of the surface structure in the region of a letter from which the some of the ink has been lost were examined using a Cambridge Stereoscan Scanning Electron Microscope with an Energy Dispersive X-ray Analysis (EDAX) attachment. Fig. 9 shows a region from which the ink has been lost. The box highlights a region where collagen a tangled mesh of collagen fibres can be seen. In the corroded region outlined by a rectangle in Fig. 10 EDAX scans show that there has been a 120 times increase in the percentage of sulphur, This supports the belief that the migration of sulphuric acid has weakened the bonding between the collagen fibres.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
Fig. 9. Region of a (Magnification ¼ 80).
pen
stroke
from
which
the
ink
has
been
lost
Further evidence of the effect of the ink on the collagen structure can be elicited by undertaking line scans across a pen stroke using SRIR. Degradation of collagen is displayed by changes in the Amide I and Amide II bands. These are related to the C ¼O stretch which is sensitive to hydrolysis. Fig. 10 shows the result of a scan across a pen stroke taken using the IR spectroscopy beamline using a Bruker Hyperion using an ATR objective at a magnification of 100. The IR beam diameter was 5 μm. The Amide I peak at 1654 cm 1 remains constant but its subband (shown here as a yellow colour) varies significantly during the scan. The group of bands (3000–3500 cm 1) which are associated with N–H and –OH stretching modes exhibit a similar behaviour. Cross sections have to be cut across a pen stroke if the depth of penetration of the ink is to be established and any migration of atomic species normal to the surface is to be observed. This was studied using the Scanning X-ray Fluorescence Microscopy beamline at the Australian Synchrotron (Treasure et al., 2012). Fig. 11 shows the analysis grid used to scan the cross section underlying a pen stroke. As well it shows the spatial distributions of the two principal atomic species: iron and sulphur. The spatial distribution for the iron is much more uniform than for the sulphur. As well, the sulphur has migrated further into the parchment than the iron. 2.3. Study of corrosion and corrosion prevention
Fig. 10. SRIR spectral scans across a pen stroke.
Another research programme addressed the need to protect statues from corrosion and to develop lubricating oils to protect old vehicles which are still in operational use in museums and galleries [5]. This work involves the use electrochemical impedance spectroscopy and Synchrotron Radiation Grazing Incidence Diffraction (SRGIXD) [2]. Photographs of the two extremes for which techniques of protection against corrosion have to be found are shown in Fig. 12a and b. Outdoor statues must be protected against: the climate (sun, rain, frost, snow, etc.); the environment (man-made pollution, wind-borne particulates; and natural hazards (bird-lime, windborne sap, etc.). They need a protective coating which cling closely and strongly to the surface and act like a blanket to exclude moisture and pollutants. And the protective layer must maintain its integrity for years. But it must be capable of being removed from the surface without harming the surface.
Fig. 11. Spatial distributions of sulphur and iron concentrations for the sampling grid on the left for the cross section of a pen stroke.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
7
a. Schematic diagram of an X-ray beam interacting with a film deposited on a substrate.
sem image of a wax layer brushed into a bronze substrate before heat treatment (15 kV)
b. Keissig fringes arising from a monolayer on sodium dodecylsulphonate on an aluminium substrate at a wavelength of 0.1nm. Fig. 13. a. Schematic diagram of an X-ray beam interacting with a film deposited on a substrate. b. Keissig fringes arising from a monolayer on sodium dodecylsulphonate on an aluminium substrate at a wavelength of 0.1 nm.
sem image of a wax layer brushed into a bronze substrate after heat treatment. (15 kV) Fig. 12. a: sem image of a wax layer brushed into a bronze substrate before heat treatment (15 kV), b: sem image of a wax layer brushed into a bronze substrate after heat treatment (15 kV).
Machinery which is required to be operated or driven must have lubricating oils for their moving parts which can: withstand the heat and mechanical stresses generated whilst the equipment is in operation; immediately form a protective barrier when the equipment ceases operation; does not contain additives which will adversely affect the leather, cork and other gasket materials used in old engines, gear boxes, and differentials. The protective coatings have to have the characteristics of selforganising aliphatic compounds (laurates, stearates, palmates, etc.). These aliphates are long chain compounds consisting of many CH-bonds ending with a chemically active headgroup. For statues number of different preparations are sold and claims are made about their efficacy are made by the manufacturers. To test these claims test surfaces of different bronzes and patinated bronzes were coated with a variety of anti-corrosion products. Fig. 12a shows a scanning electron microscope image of a microcrystalline wav which has been dissolved in solvent, brushed onto the surface, in this case a bronze surface. The wax has not completely covered the surface, and would be ineffective as a protection against corrosion. Fig. 13b shows an image taken after the surface had been heated. The clumps of wax have been distributed more evenly across the surface, and the protective coating is acting more like a protective blanket (Creagh et al., 1995). Sulphonates have been the prime chemical building blocks
used by manufacturers to create corrosion prevention products for the automotive industry. They are made by the sulphonation of, and the removal of alkylated aromatic hydrocarbons from refined lubricating oils. In 1941 alkaline earth salts were added to create detergents. Later research led to their development as water displacing corrosion preventatives (WDCPs). Initial research into these WDPCs was undertaken to gain an understanding of how they interacted with the surface on which they were deposited. To study the thickness and other properties of the layer the techniques of Synchrotron Radiation X-ray reflectivity (SRXR) and Grazing incidence X-ray Diffraction GIDX) were used (Hallam et al., 1997). Fig. 13a shows a schematic diagram showing how XRR can be used to measure film thickness. The difference in path length between the X-ray beam which is externally reflected at the surface of the protective layer and the beam which is reflected by the substrate. As the sample is rotated through the angle of total reflection (the refractive index for X-rays for an air-material interface is less than 1) a series of fringes (Keissig fringes) can be observed. The film thickness can be inferred from the periodicity of the fringes. (θ2 ¼ m2 (λ/2d)2 þ αc2; where θ is the angle of reflection; λ is the X-ray wavelength; d is the film thickness; αc is the critical angle; m is an integer related to the order of the interference pattern). Plotting θ2 versus m2 gives a straight line with slope (λ/d)2 and intercept αc2. Fig. 13b shows the Keissig fringes which result from the interaction of an X-ray beam with a monolayer film of sodium dodecyl sulphonate on an aluminium substrate in a GIXD experiment (Creagh et al., 2000). GIXD can determine whether any in-plane organisation exists for this system. Layers of a sodium stearate, a similar were deposited on a silicon substrate using the Langmuir–Blodgett process. Fig. 14a shows that diffraction spots exist, and this indicates
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
8
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
a. GIXD pattern for stearate layers deposited on a silicon substrate at a wavelength of 1nm.
a. Right: schematic diagram of an EIS system. Left:
cartoon illustrating the processes taking place at the anode.
b. Chemical modelling of how the sodium dod ecylsulphona te molecules might attach to a bronze substrate. Fig. 14. a. GIXD pattern for stearate layers deposited on a silicon substrate at a wavelength of 1 nm. b. Chemical modelling of how the sodium dodecylsulphonate molecules might attach to a bronze substrate.
that long range ordering in the diffraction plane is occurring. X-ray intensity is observed between the spots indicating the ordering is not perfect over the area probed by the X-ray beam. Fig. 14b shows a depiction using chemical modelling which demonstrates how the sodium dodecylsulphonate layers might organise on a bronze substrate. The head-group has attached itself to the surface and the molecules have packed themselves together forming a hexagonal array on the surface. The stearate chains are angled with respect to the surface because of the stereochemistry of the sodium sulphonate-group These insights into the behaviour of protective coatings, however, do not assist with the problem of how one deals with the assessment of assess the deterioration of coatings and which products are most suitable for use. Electrochemical Impedance Spectroscopy (EIS) is able to provide a solution. In EIS the surface under investigation is made the anode of an electrochemical cell. Fig. 15a shows a schematic diagram of the region adjoining the anode of a basic EIS system. As well it shows a cartoon illustrating the processes taking place at the interface and the electrical equivalent circuit. For a bare metal inserted into an electrolyte ions which are absorbed at the metal–solution interface and create a charged double layer This behaves like a leaky capacitor (a resistance in parallel with a capacitance) whilst the electrolyte is purely resistive. When a coating is added to the surface an additional capactiance (due to the non-conducting coating) is added which is large compared to the dipole layer capacitance, and is in series with the resistance of the electrolyte. So long as the coating maintains its integrity the low frequency impedance is high. At the frequency increases the impedance reduces as E 1/ω2. If pores exist in the coating resistive leakage paths exist, which bypass the
b. Bode impedance plots for a number of products alleged to provide good corosion protection. Fig. 15. a. Right: schematic diagram of an EIS system. Left: cartoon illustrating the processes taking place at the anode. b. Bode impedance plots for a number of products alleged to provide good corosion protection.
capacitances. The pores may exist because of poor application of the coating initially, or more likely, through degradation of the layer. At high frequencies the impedance becomes purely resistive. In practice the testing of statues and other static objects is made using a portable cell in which contact with the surface is made using a moisture retaining pad containing the electrolyte (usually non-gaseous mineral water) (Letardi, 2000). Fig. 15b shows a Bode impedance plot in which the corrosive resistance has been measured for a number of lubricant samples on steel. A good lubricant is characterized by a high low frequency impedance and a smooth high frequency roll-off. Only one of the products shown has the desired specifications. All the others exhibit varying degrees of porosity.
3. Conclusions In this paper three of the many projects the authors have been associated with have been outlined. There have been many more. Notable other projects include: the restoration of a Japanese fighter, study of important medals (the Victoria Cross and the Lusitania medal), studies of the degradation of elastomers, the effect of moisture on composite materials, the study of the armour of the Australian outlaw Joe Burns, et cetera. All have required the use of a wide range of radiation-based technologies to solve the questions posed by curators and conservators.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i
D. Creagh et al. / Radiation Physics and Chemistry ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Acknowledgements The authors are indebted to the Australian Research Council (LE0346515), the National Capital Development Commission, the Australian Synchrotron, the Australian Synchrotron Research Program, and the Australian Nuclear Science and Technology Organisation for direct funding and to the Australian War Memorial, the National Gallery of Australia, the National Film and Sound Archive, the National Museum of Australia for in kind assistance. Many scientists have assisted us, and we are very appreciative of their time and effort. They include: Milan Stoilovic (AFP Forensic Laboratory); Mark Tobin, Ljiliana Puskar, David Paterson, Martin de Jonge (ASP); James Hester, Gary Foran (ASRP); Australian National University (Research School of Chemistry – Graham Heath; Engineering – Adrian Lowe; Research School of Earth Sciences).
References Creagh, D.C., 2007. Synchrotron radiation and its use in art, archaeometry, and cultural heritage. In: Creagh, D.C., Bradley, D.A. (Eds.), Physical Principles in Art, Archaeometry and Cultural Heritage vol. 2. Elsevier, The Netherlands, pp. 1–91, ISBN:13-978-0-444-52856-8. Creagh, D.C., Bradley, D.A. (Eds.), 2000. Radiation in Art and Archaeometry. Elsevier,
9
Amsterdam, The Netherlands, ISBN: 0 444 050487 7. Creagh, D.C., Hallam, D.L., Price, C.L., 1995. Testing of Waxes for the Protecting of Memorials. National Capiltal Planning Authority, Canberra, Australia. Creagh, D.C., Kubik, M.E., Sterns, M., 2007. On the feasibility of establishing the provenance of Australian Aboriginal artefacts using synchrotron radiation X-ray diffraction and proton induced X-ray emission. NIM A580, 721–724. Creagh, D.C., Otieno-Alego, V., O’Neill, P.M., 2000. X ray reflectivity and grazing incidence diffraction studies of the adhesion of protective wax coatings on metal surfaces. In: Proceedings of the Eighth Meeting of the International Radiation Physics Society, Prague, Czech Republic. 4–10 June, 2000. Hallam, D.L., Adams, C.D., Creagh, D.C., Holt, S.A., Wanless, E.J., Senden, T.J., Heath, G. A., 1997. Petroleum sulphonate inhibitors. In: McLeod, I.D., Pennec, S.L., Robbiola., L. (Eds.), Metal 95. James and James, London, UK, pp. 159–161, ISBN: 1 873936 67 2. Husband, T.B., 2008. The Art of Illumination. The Metropolitan Museum of Art, New York, ISBN 987-1-58839-294-7. Kennedy, C.J., Wess, T.J., 2006. The use of X-ray scattering to analyze parchment and degradation. In: Bradley, D.A., Creagh, D.C. (Eds.), Principles in Art, Archaeometry and Cultural Heritage vol. 1. Elsevier, The Netherlands, pp. 159–173, ISBN13: 978-0-444-52131-6. Lee, A.S., Mahon, P.J., Creagh, D.C., 2006. Raman analysis of iron gall inks on parchment. J. Vib. Spectrosc. 40, 170–175. Letardi, P., 2000. Electrochemical methods in the conservation of metals. In: Creagh, D.C., Bradley, D.A. (Eds.), Radiation in Art and Archaeometry. Elsevier, Amsterdam, The Netherlands, pp. 15–39, ISBN: 0 444 050487 7. Treasure, A.E., Creagh, D.C., Whitely, C., Howard, D., de Jonge, M., 2012. SR-μXRF in the analysis of iron-gall inks on parchment. In: Streli, C., Mantler, M. (Eds.), Proceedings of the European X-ray Radiation Spectroscopy Conference. Vienna, Austria. 17–22 June, 2012.
Please cite this article as: Creagh, D., et al., The use of radiation in the study of cultural heritage artefacts. Radiat. Phys. Chem. (2016), http://dx.doi.org/10.1016/j.radphyschem.2016.01.040i