Raman spectroscopy in art and archaeology
2010; Wiley; Volume: 41; Issue: 11 Linguagem: Inglês
10.1002/jrs.2850
ISSN1097-4555
Autores Tópico(s)Conservation Techniques and Studies
ResumoThe Fifth International Congress on the Application of Raman Spectroscopy in Art and Archaeology (RAA 2009) took place in Bilbao (Basque Country, North of Spain) from 14 to 18 September 2009. This series of conferences has now a large tradition, starting in London (2001),[1] and following in Ghent (2003),[2] Paris (2005)[3] and Modena (2007).[4] The scientific programme included all applications and studies performed with Raman spectroscopy on Cultural Heritage and related materials. The conference offered an outstanding and unique opportunity for exchanging knowledge on leading edge developments. Cultural Heritage studies, interpreted in a broad sense, included amongst others, pigments, inks, new materials, gemstones, stones, precious stones, glass, ceramics, chemometrics on artwork studies, palaeontology, resins, fibres, forensic applications in art and archaeology, industrial archaeology, etc. These studies were presented along 4 Plenary Lectures, 43 Oral Presentations and 65 Poster Presentations. The number of active participants was 146 delegates among the 408 authors from 32 countries that presented at least one work to the Congress. These figures confirm the wide acceptation of the RAA series of Congresses although this is a highly specific one because Raman spectroscopy in Art and Archaeology is the core of the conference. This is in consonance with the increasing number of papers currently published in the field of Raman spectroscopy applied to Cultural Heritage, not only in this journal but also in other relevant journals concerning analytical chemistry and more properly spectroscopy, such as Analytical Chemistry, Analytical and Bioanalytical Chemistry, Spectrochimica Acta and Applied Spectroscopy, among others. And the search for new applications continues because several disciplines and professionals are increasingly requiring our expertise. Some of the papers in this issue give a general perspective and new developments in the field of Art and Archaeology, presenting current problems for Raman spectroscopy, but the Scientific Committee would highlight in the present edition of the RAA Congress the applications of Raman spectroscopy on the study of the environmental conditions affecting the Cultural Heritage (decaying, corrosion, etc.); and the response was unsuspected because 27% of the presentations considered some of the different aspects of environmental impacts. That is why this special issue covers 8 selected manuscripts, over the 23 works included here, on that particular topic. The impacts of the environment on both movable and immovable items of the Cultural Heritage is of great concern to all the professionals working in the field of heritage conservation. The characterisation of art and archaeological materials in a period of time in which the Cultural Heritage is in a rapid process of deterioration is crucial. For this reason, nowadays, one of the trends in analytical chemistry is not only the development of techniques and methods that can reliably identify the components in specimens with artistic and/or archaeological value but also the coupling of these techniques with other tools providing methods for the elucidation of the mechanisms of degradation processes that affect the Cultural Heritage properties. Atmospheric acid gases, organic compounds and particulate matter are the major sources of environmental stressors affecting these materials; they mainly affect not only the outside parts of the built heritage but also the acid aerosols that can reach the indoor parts of the buildings where movable items are exposed or stored. The aerosols containing CO2 play an important role in the biodeterioration processes because the initial acidity required to start the reactions of decaying is supplemented by this greenhouse acid gas. Moreover, infiltration waters carrying soluble ions are another source of important problems if such waters reach or affect the body of the artworks or archaeological materials. Most of these effects are contained in the next eight contributions summarised below. The UNESCO World Heritage List included in 1998, under the generic denomination of Rock Art of the Mediterranean Basin of the Iberian Peninsula (eastern half of Spain), around 1500 open-air rock shelters with paintings on their walls. The research group leading by Prof. Hernanz is working from the last years on these paintings and here they present the study of the painted panel 3, of the Hoz de Vicente rock shelter.[5] Pictographs, accretions, substrata and alterations were analysed to conclude in one favourable environmental impact and another adverse. As the pigmented layer was systematically between two layers of these hydrated forms of calcium oxalate, produced by the activity of fungi and lichens, authors concluded the formation of the protective external oxalate for some years after the execution of the paintings. Gypsum and clayish minerals appear in the accretions covering several parts of the panel; these compounds can be crystallised/deposited due to the weathering of the upper strata of the shelter that reach the panel through percolation of water. Wall painting fragments from a Pompeian House (Italy), exposed to two different environments, were analysed and compared by Maguregui et al.[6] Samples from the second style painting, buried in the ground and not exposed to open air, recovered from recent excavations (2004–2006), and samples from the fourth style, excavated about 150 years ago and exposed since then to the open air, gave completely different results on the composition of the decayed products found together with the original ones. Probable decaying pathways were proposed to explain the formation of some decaying products (magnetite mainly) of red pigments and for the original components of the mortars, specially the whole process leading to the formation of coquimbite (nonahydrate iron sulphate). The comparison of the conservation state of the mortars and pigments exposed to two different kinds of environments (outdoors and under a burial) concluded the importance of the SOx impacts on the open air artworks. The chemical compositions of materials used in the preparatory and pictorial layers of wall paintings from four different sites of the Roman Empire during the first and second century AD were analysed by Cristini et al.[7] The fresco technique, revealed by the presence of calcite in the Greek and Roman samples, was not used for the Gallo-Roman works. The studies indicated that the same compounds were used for red and green pigments throughout the Roman Empire: red ochre (haematite) and green earths (celadonite, in the case of the Gallo-Roman fragments). Authors concluded that the impact of the atmosphere on the sample surface is very important, as the only fragments to have suffered identifiable deterioration (massive presence of gypsum) are those from the ‘open-air humid room’ in the Kenchreai location. The presence of gypsum indicates the implication of SO2 in the deterioration but, as it was restricted to the unpigmented areas, makes it possible to suggest that the pigments in the pictorial layer can serve as a protection for the calcite preparatory layers against the atmospheric acid attack. A series of wall paintings dating from the period fifteenth to sixteenth century, uncovered during the restoration projects at the church of Santa Eulàlia of Unha in the Val d'Aran (Spain), were analysed by Clark et al.[8] The Val d'Aran is situated on the Atlantic slope of the central part of the Pyrenees where Romanesque works of art are known to be particularly remarkable both for their abundance and their quality. Traditional materials such as hematite, goethite, vermilion, red lead, carbon black and calcite were identified, but the blue pigment identified in the Romanesque frescoes was aerinite, a rare Fe(II)/Fe(III)-containing aluminosilicate found locally in the Pyrenees region. A remarkable feature was the presence of nitrates, together with sulphates and oxalates; authors suggest the incorporation of nitrates to the wall paintings through the upward seepage of ground waters because nitramite (NH4NO3) was mainly identified in most of the samples. Archaeological objects are exposed to the action of micro-organisms when they are in a biologically active environment. The presence of iron sulphides in the corrosion system testifies in most cases that the degradation was influenced by sulphate-reducing bacteria. Iron sulphides and other iron/sulphur containing compounds (mackinawite, FeS and greigite, Fe3S4, mainly) were detected by Rémazeilles et al.[9] in rust layers of archaeological ferrous objects and in wet wooden fragments contaminated by iron, extracted from ancient wrecks. The detection of Fe(III)-containing mackinawite or greigite indicates that the archaeological object was exposed in the burial to a small amount of oxygen; this testifies on one hand that the object settled in anoxic conditions and on the other hand that it suffered from microbiologically influenced degradation. The identification of different forms of iron sulphides is very important to estimate the degradation state of an archaeological object and the measures required to store it, in order to avoid new transformations inside the objects when due to the reactivity with oxygen they are removed from their anoxic environment. Copper carboxylates were identified by Aceto et al.[10] as degradation residues on the surface of a metallic pigment used in the illumination and in the text of a ninth century Italian manuscript containing the Homilies on the Gospels of Gregory the Great, belonging to the Archive and Chapter Library of Vercelli (Italy). These compounds are responsible for the greenish aspect that text lines and decorated initials have developed over time from their original golden fashion. The investigations carried out allowed to hypothesise their origin as due to interaction of copper with carboxylic acids coming from the wooden cardboards used to store the illuminated manuscript or from previous treatments or from the level of volatile pollutants present in the closed environment. Moreover similar compounds were found in three ninth to tenth century Italian manuscripts coming from the Bobbio abbey. Salt crystallisation is a principal deterioration factor in many stone monuments. Kramar et al.[11] have investigated limestone samples and weathering products of two baroque monuments in Ljubljana (Slovenia) in order to constrain the minerals produced during the weathering process. Calcite, dolomite, quartz, anatase, goethite, haematite and phyllosilicates were determined as original minerals. But limestone was found to be extensively deteriorated in outdoor and indoor environments, showing flaking, sub-florescence, efflorescence and black and white crusts as deterioration phenomena. Among these weathering forms, efflorescence was found to be more complex in terms of mineral assemblage (gypsum was associated to hexahydrite, pentahydrite and niter). In contrast, the mineralogy of sub-florescence was rather simple being composed only by gypsum. Results showed that gypsum also crystallised under the surface as sub-florescence, which eventually led to the flaking and crumbling of the limestone. Here also sulphate and nitrate compounds have been detected, highlighting the influence of urban atmospheres on the conservation state of buildings. In the Northwest of Spain and in the North of Portugal, where the architecture utilises mainly granite stones since centuries, the principal mechanism of stone deterioration is feldspars hydrolysis. Materials such as wax, oils and natural epoxi and acrylic resins have been used for the past 100 years as protective coatings that might act as consolidants and as hydrofugants of stone monuments. In this work, Pan et al.[12] show the advantages of using Raman spectroscopy to study the evolution of a protective treatment (walnut oil over Roan granite in this case) once applied, as well as the determination of the curing (oxidative process) time and chemical transformations. Five different excitation wavelengths, from 1064 to 488 nm, were compared, being the intensity of the spectrum recorded at 532 nm; the higher for Raman measurements of oil on a granite substrate. The development of new instruments, methods and data treatment of the spectral information will be a field of continuous research despite the type of real samples that are going to be studied. Real samples in the field of Cultural Heritage are always complex mixtures of original and degradation compounds that require new approach to be implemented in the daily practice of Raman spectroscopy. Here we have some examples of these new developments, including Surface Enhanced Raman Spectroscopy (SERS) methods, Raman mapping, elucidation of different compounds in old and synthetic pigments, the use of chemometric tools to help in the research of historical painting layers and the development of easy to use flow chart to ascertain the identification of natural organic compounds in old varnishes. A new SERS method carried out directly on the fibre of historical textiles, without the need of the standard hydrolysis of the mordant-dye complex, has been developed by Jurasekova et al.[13] to detect historical mordant dyes. In this case, alizarin and carminic acid have been detected in reference wool and linen fibres dyed with madder and cochineal respectively; the method has been applied to one archaeological Coptic textile (sixth to eighth century AD) of Egyptian origin, where alizarin has been clearly identified. Other mordant dyes of the flavonoids family (luteolin, apigenin and flavonols), have also been identified by the same SERS method in wool fibres dyed with natural plants used in Central and South America (Dyer's greenweed, old fustic, onion and chilca) following pre-Columbian dying recipes. Raman mapping can provide molecular information to complement data derived from other analytical techniques in works of art and other objects of cultural significance. The classical Raman mapping is performed using a motorised microscope stage that moves a sample or an object point-by-point in two spatial directions. The paper from Ropret et al.[14] reports on the development of a new Raman mapping approach based on a set of scanning mirrors that direct the laser beam in two spatial directions, vertically through the microscope head or through a horizontal exit on the Raman micro-spectrometer. The advantages and limitations of these two Raman mapping approaches are compared and discussed based on an example of a contemporary oil painting on canvas. Both methods can be used non-invasively in works of art that fit under a microscope objective or in microsamples when, for example, obtaining information on the samples' layering structure is necessary. Verdigris is among the most important synthetic pigments in painting history, being subjected to a large number of investigations. The reproduction of an old recipe to obtain the pigment, developed by San Andrés et al.,[15] leads to a complex mixture of hydrated copper(II) acetates containing neutral verdigris and two varieties of basic verdigris. As the characterisation of such a mixture was complicated by the lack of reference standards, authors complemented the Raman measurements with Fourier transform infrared (FTIR) and X-ray diffraction experiments to perform a multianalytical approach. The new results have been compared to available studies. The spectral analysis allowed authors to distinguish among the different verdigris varieties (chemical compounds in fact), assigning stretching and bending wavenumbers to the different chemical compounds. Both wavenumbers and Raman intensities were used to quantify the relative compositions of the mixture components. Another example of the Raman applicability on synthetic pigments is the work by Bouchard and Gambardella,[16] on the study of cobalt blue spinels in the field of works of art, aiming to provide a better understanding of the Raman spectra of cobalt-based pigments. Raman analysis of 21 widely used industrial blue cobalt-based pigments, completed with information from elemental energy dispersive spectroscopy and X-ray diffraction techniques, led to the identification of 17 spinel-type cobalt pigments, each with its characteristic mineral composition that are summarised in the paper. During the experiments, Raman microspectroscopy turned out to be a perfect tool for detecting the presence of doping agents in the spinel lattice. The use of only Raman spectroscopy information to solve problems in the analysis of Cultural Heritage items can be highly improved if chemometric tools are applied on the spectral information. The work presented by Navas et al.[17] explores the application of principal component analysis (PCA) on first-derivative Raman spectra to investigate historical tempera paint model samples. Nineteen model samples were prepared according to Old Master recipes to obtain egg yolk tempera painting standards (several pigments covering blue, red and white colours) similar to those used by medieval artists. Multivariate analysis on the Raman spectra showed the excellent ability of PCA, when applied to the derived Raman spectra, to discriminate the model samples according to compositions, opening a new strategy for identification of historical pigments and binders. The organic substances possibly used as ancient varnishing materials are quite well known, and seem to be comparable for musical instruments and for furniture. They are made of natural organic components such as resins, oils, waxes, glues or gums with, in some cases, inorganic additives, leading to highly complex mixtures of compounds whose identification in unknown samples is not an easy task. The work presented by Daher et al.[18] proposes the use of FT-Raman and infrared spectra to discriminate the different classes of organic media; a flowchart based on the different spectroscopic features obtained from 14 unaged reference samples representative of resins, oils, glues and gums is proposed to differentiate the four chemical families tested, and further, within the resins family, four subgroups were clearly set apart: triterpenic resins, shellacs, colophony and Venice turpentine, and finally copal and sandarac. The Raman characterisation of the materials used in works of Art is and will be a research subject of great interest. The number of contributions to this field presented at the RAA 2009 Congress represented around 30% of the total presentations. This special issue has selected four of them due to the specific contributions contained in each one. The applicability of two mobile non-destructive techniques, energy dispersive X-ray fluorescence spectroscopy (EDXRF), to ascertain the elemental composition, and fibre-coupled Raman spectroscopy, to get molecular information, has been demonstrated by Deneckere et al.[19] on the central panel of the Wyts triptych, after Jan van Eyck. Using a combination of these direct techniques, the most relevant pigments (vermilion, lead white, anatase, massicot, zinc white, lead-tin yellow type I, iron oxides, etc.) were identified. Moreover, the combination of these analytical techniques also gave information about restored parts and the layered structure of the panel. This research confirmed that the painting is a heavily restored copy after a lost original by Jan van Eyck. Japanese folding screens, called byobu, are one of the oldest and most highly refined forms of Japanese art, where paper and silk were used as materials for the artists to paint on. The two pairs of folding screens studied in the work presented by Pessanha et al.,[20] also exhibit a golden background to create a luminous effect. Presently only about 60 examples of this Namban genre remain, so the study of these two pairs is of most importance to the knowledge of this precious craft. The hand-painted Japanese folding screens were analysed by Raman, EDXRF and FTIR spectroscopy, in order to characterise the materials used in their manufacture. The materials identified, such as gold, silver, malachite, azurite, vermillion, red lead, red madder, yellow ochre, white oyster shell and carbon black are part of the traditional Japanese palette. Applied tin-relief brocade (commonly called applied brocade) refers to a decorative painting technique using tin leaf applied over a supporting relief mass (filling) which is glued to the artwork to simulate gold and silver textile brocades. This originated in Germany around 1415–1430 and spread across Europe from the mid-fifteenth to the mid-sixteenth century. The study presented by Rodríguez et al.[21] focuses on six early sixteenth-century altarpieces in the Basque Country, Spain. Examination with Raman spectroscopy, FTIR and scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM/EDX) identified the inorganic and organic components of the various layers. Raman spectroscopic mapping was used to image the location of phases in selected cross sections. Five altarpieces had calcium sulphate grounds whilst only one, thought to come from Flanders, had a calcium carbonate ground; the grounds based on gypsum must be highlighted because this material is a high stable one in the Atlantic climate (low temperature with high humidity). However, the filling masses consisted of different mixtures of inorganic and organic (protein and/or oil or beeswax) materials. The fourth significant work in this subject of characterisation of works of art is a contribution by Steven Saverwyns[22] who used micro-Raman spectroscopy to study six paintings attributed to Liubov Popova, one of the most prominent figures of the Russian avant-garde artistic trend. With the onset of perestroika, Russian avant-garde art became very popular in the West, the demand was high, prices zoomed and soon the market was flooded by forgeries. Raman spectroscopy was evaluated as a tool for identifying pigments of chronological inconsistency (pigment anachronisms); the results especially those of the synthetic organic pigments identified the Terminus post quem date, the earliest point in time when the paintings could have been made. These anachronism studies concluded that none of the works could be attributed to Liubov Popova. The number of works where Raman spectroscopy is the key technique to analyse archaeological objects is continuously increasing. Most of the time the problem had been the difficulty in access to samples but nowadays the archaeologists are professionals who prefer to work together with scientists in order to characterise the objects even at the time of its excavation. This section includes five works covering different archaeological samples and materials where Raman spectroscopy has been the main instrumental technique used in the studies. The Gristhorpe Man is a well preserved skeleton, found in 1834 (Gristhorpe, North Yorkshire, UK) in an intact coffin fashioned from the hollowed-out trunk of an oak tree. It was stained black from the oak tannins, wrapped in an animal skin and buried with a range of grave artefacts, including a bronze dagger, flints and a bark vessel. As part of the project around these biomaterials and remains, Edwards et al.[23] have analysed in a non-destructive way several mysterious small spherical objects discovered in the coffin underneath the skeleton and initially believed to be ‘mistletoe berries’ associated with ancient burial customs. The interpretation of the Raman spectral data, microscopic analysis and comparison with modern specimens has led to the conclusion that the small spheres were phosphatic urinary stones, deposited in the coffin base during the degradation of the cadaver, proving that the supposed mistletoe berries were not correctly attributed originally. Powdered pigments found in bowls from the Pompeii archaeological site and some wall painting fragments from Vesuvian area (conserved in the National Archaeological Museum of Naples) were investigated by Aliatis et al.[24] using microscopic Raman and FTIR spectroscopies, XRD and SEM–EDX. Apart from the common pigment materials expected for the Roman wall paintings, some unexpected pigments and mixtures were identified, namely, (a) huntite (CaMg3(CO3)4) in a white powder found in a bowl, (b) Celadonite found in the green samples from the wall paintings, together with Egyptian blue and basic lead carbonate and (c) a mixture of malachite, goethite, Egyptian blue, hematite, carbon, cerussite and quartz in a heterogeneous green pigment in another bowl. Roman pottery from Oiasso harbour (nowadays Irun, Basque Country) was thoroughly studied by Olivares et al.[25] by using micro-Raman spectroscopy and X-ray diffraction analysis in order to identify and characterise the mineralogical composition of those samples and to get a deeper insight about technologies involved in the elaboration of ceramic artefacts. Authors conclude the use of high temperatures (above 1100 °C) in the firing due to the presence of mullite, trydimite, pseudowollastonite, diopside and rutile instead of anatase. Moreover, the presence of hematite (α-Fe2O3) and maghemite (γ-Fe2O3) instead of magnetite (Fe3O4) suggests the oxidising conditions during ceramic firing. Terra sigillata is certainly the most famous fine ware of the Roman period and for this reason it is a subject of interest for both the archaeology and archaeometry communities. Leon et al.[26] have made a comparative study of central Italian and southern Gaul productions. Authors suggest the use of a combination of XRD (for mineral composition analysis) and Raman (for distortions in the hematite bands) analysis, like in the previously referred work. The results suggested that the presence of the strong additional Raman band at around 680 cm−1 present in both clays and ancient slips could be associated to the re-crystallisation of hematite, occurring above 750 °C. It could also be related to magnetite content, but this hypothesis is not supported by X-ray diffraction data. Raman scattering is revealed as very sensitive to small variations in clay composition and firing temperature and it has been successfully used as an in situ sensitive probe for discriminating between Italic and South-Gallic productions. Pottery fragments belonging to the first and last production period of an old ceramic factory, recently found close to the walls of Parma (Italy), active from the fourteenth century until the seventeenth century, were analysed by Bersani et al.[27] using micro-Raman spectroscopy and SEM/EDX for the glazes and the painting materials, while time-of-flight neutron diffraction and X-ray diffraction were used to characterise the ceramic bodies. From the different mineral phases identified, authors propose the conditions of ceramic production from illitic calcareous clays and annealed in an oxidising atmosphere at an estimated temperature of 900–1000 °C, without any evidence of a change between the fourteenth and seventeenth century productions. The most controversial finding was related to the nature of the white decorations of the seventeenth century ceramics, where the white clay used contained not more than 1% titanium but in the anatase form. The 23 manuscripts included in this special issue are excellent examples of the innovative applications of Raman spectroscopy from prehistoric samples to present-day artefacts. Some of them make use of other non-destructive instrumental techniques to support the Raman information. Others incorporate chemical modelling and/or chemometric analysis to explain and interpret, to diagnose in practice, the presence of unexpected materials together with the original ones. But all of them have in common the Raman information as the core of the works. The instrumental improvements in the new portable Raman spectrometers (less weight, high power in the excitation laser source, better detectors and microscopy capacity) will be shown in future works where, the in situ field measurements will be so important than the current laboratory surveys on small number of microsamples. With the added advantage of the number of spectra a portable device can collect during a working day, the in situ mapping capacity will increase enormously even at the micron scale. The high quality and the amount of spectroscopic information from Raman and other techniques allow us to have a better knowledge of the items we are studying. But we need to go further, trying to interpret how the studied item interacts with its surrounding environment (polluted atmosphere, burial, indoor of a museum, etc.) and how this conditions its conservation state. This diagnosis will help restorers to design properly the cleaning and conservation actions on the Cultural Heritage artefacts. Several of the manuscripts collected in this special issue on the application of Raman spectroscopy in Art and Archaeology show research works developed with a high level of multidisciplinary interaction among different research groups. The future trends in the field of Cultural Heritage call for cooperation between people in the field of Humanities (historians, restorers, archaeologists, etc.) and those in the field of Science (spectroscopists, chemists, geologists, biologist, environmentalists, etc.). The contribution of those attending the RAA 2009 Congress to a research conducted in this collaborative manner has been clearly shown and we hope to increase such multidisciplinary cooperation in the works to be presented in the forthcoming RAA 2011 in Parma, Italy. We are extremely grateful to the participants and institutions who assisted in making the conference possible. In particular, we would like to thank the University of the Basque Country and the Bizkaiko Foru Aldundia for their financial support to accommodate the conference in the Museum of Fine Arts, Bilbao. The co-funding of the Spanish Ministry of Science and Innovation (MICINN, Acciones Complementarias 2009 - Modalidad A, ref. CTQ2009-06887-E/BQU) and the Basque Government (ref. RC-2009-2-116) is gratefully acknowledged, as well as the facilities given by the Museum of Archaeology, Bilbao. The organising committee expresses its thanks to the following companies who also supported the conference: Renishaw, Thermo-Fischer Scientific, HORIBA Jobin Yvon, Microbeam-BWtech and Jasco.
Referência(s)