Microsc. Microanal. 21, 56–62, 2015 doi:10.1017/S1431927614013129
© MICROSCOPY SOCIETY OF AMERICA 2014
Micro-Analytical Study of a Rare Papier-Mâché Sculpture Marta Manso,1,2,* Ana Bidarra,3,4 Stéphane Longelin,1 Sofia Pessanha,1 Adriana Ferreira,5 Mauro Guerra,1,7 João Coroado,3,6 and Luísa Carvalho1,7 1
Centro de Física Atómica da Universidade de Lisboa, Av. Professor Gama Pinto 2, 1649-003 Lisboa, Portugal Faculdade de Belas-Artes da Universidade de Lisboa, Largo da Academia Nacional de Belas-Artes, 1249-058 Lisboa, Portugal 3 Geobiotec/Departamento de Geociências, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal 4 Cinábrio, Conservação e Restauro, R. Almirante Cândido dos Reis, 3800-096 Aveiro, Portugal 5 Arquivo Municipal de Lisboa, Rua B Bairro da Liberdade, Lote 3-6, Piso 0, 1070-050 Lisboa, Portugal 6 Departamento de Conservação e Restauro, Instituto Politécnico de Tomar, Quinta do Contador, Estrada da Serra, 2300-313 Tomar, Portugal 7 Departamento de Física, Faculdade de ciências e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal 2
Abstract: The analysis of a Portuguese “papier-mâché” sculpture depicting Saint Anthony is presented in this case study. Several questions were addressed such as the characteristics of the support, pigments used, and artistic technique in order to establish a possible timeline for its production. Qualitative analyses of the cross-sections and of the paper support were performed by optical microscopy using reflected light. Two polychrome layers from different periods and a rag pulped support were identified on the sculpture. The use of micro X-ray fluorescence and Raman microscopy techniques enabled the differentiation of coloring materials used in both polychromies. Semi-quantitative analyses of the gilded samples were also performed by scanning electron microscopy in combination with energy-dispersive spectroscopy allowing the determination of a common Au–Ag–Cu alloy with differences in the purity of the gold. The identified coloring materials lead us to believe that the sculpture was produced in the 19th century, being overpainted in the first half of the 20th century. Key words: papier-mâché, sculpture, micro-Raman, micro-XRF, SEM-EDS
I NTRODUCTION “Papier-mâché” is a term that has been applied to threedimensional (3D) objects of recycled paper fiber whether layered in sheet form with an adhesive or cast as beaten pulp (van der Reyden & Williams, 1986; Thornton, 1993). Its origin is at least as ancient as the invention of paper itself (China during the Han Dynasty, c. 202 B.C.–220 A.D.) and the first 3D objects were produced during this period. The spread of papermaking was followed by the production of objects made with paper fiber and it was widely used in several countries as an architectural material from the 17th to the early 19th century. In Portugal, there are very few examples of paper objects either in architectural decoration or in isolated objects such as sculptures (Franco, 2012). In this last case most of the known works are integrated in larger structures made with different materials, such as clay nativity scenes, and often represents secondary characters. The studied sculpture, depicting Saint Anthony, is a full-length piece ~ 30 cm high. It is a standing image, placed in a frontal position and facing forward. In his hand Saint Anthony holds baby Jesus. The color pallet is very simple and consistent with the iconographic representation of the Saint: a Franciscan monk. The tunic is brown and decorated Received March 28, 2014; accepted August 12, 2014 *Corresponding author. [email protected]
with small fish-scale drawings, and the hood, the extremity of the sleeves and the lower part of the tunic are decorated with a vegetalist pattern. This decoration is achieved through the “estofado technique” that allows seeing the gold leaf beneath the brown when scraping the upper paint layer. The globe on left hand of baby Jesus is blue and the base mimics a blue marble effect. The sculpture is composed of several materials, but the main one is papier-mâché that was used to mold the body of the sculpture. The hands are made of wax, baby Jesus in clay, Saint Anthony’s eyes in glass, and a wood pin was used to join the head to the torso. The base is also made of wood. The lower part of the sculpture is hollow (Fig. 1). Very few technical studies have been published on papiermâché polychromated sculptures. Dumont et al. (2011) reported the techniques and materials involved in fabrication of a papier-mâché anatomical model of a horse created in the mid-19th century by Dr. Auzoux. Investigation was conducted using analytical techniques such as X-ray radiography for examination of the internal structure, gas chromatography coupled to mass spectrometry for identification of organic components in the paper pulp and as pigment binders, scanning electron microscopy with energy-dispersive X-ray spectrometry (SEM-EDS) and Raman spectroscopy for paper fillers and ground and paint layer characterization. In the study of Saint Anthony’s sculpture, we report the use of digital microscopy for identification of the type of
Micro-Analytical Study of a Rare Papier-Mâché Sculpture
57
Figure 1. Front and rear view of Saint Anthony papier-mâché sculpture before the conservation treatment.
fibers and binders used in the papier-mâché production. The sample cross-sections were also analyzed through optical microscopy (OM); this technique allowed determination of the number, thickness, adhesion, and cohesion of the different layers (Calvo, 2003), the presence of overpaints and provided a comparative approach to the optical characteristics of the different strata. For paper filler and pigment identification, micro X-ray fluorescence (μ-XRF) and Raman analysis were used as complimentary techniques. The decoration technique of the cloak was determined using μ-XRF elemental mapping. The gold leaf was analyzed by SEM-EDS. The use of SEM in combination with EDS increases sensitivity for lighter elements and the spatial resolution for spot analysis is higher. In addition, it allows examination of the microstructure with line scans or 2D mapping of element concentration (Guerra & Calligaro, 2004; Hein & Degrigny, 2008). This method has been widely used in determination of the major elements and purity of the gold alloys (Bidarra et al., 2009, 2013, 2011). Suitability of XRF for paper filler characterization was reported by Manso et al. (2007, 2008a, 2008b) in historical and modern paper elemental characterization. X-ray fluorescence with Raman analysis is often used for pigments determination on paper support folding screens (Pessanha et al., 2010, 2012), wallpapers (Pessanha et al., 2009), and maps (Castro et al., 2008a, 2008b). Application of elemental mapping has been useful in the study of pigment dissemination (Carvalho et al., 2009) and ink composition(Manso et al., 2011). The material and artistic production of the sculpture under study was carried out as an opportunity to learn more about a rather unknown and rare art form and to establish a possible timeline for its production.
MATERIALS
AND
METHODS
The number of samples obtained from the sculpture was limited owing to its state of conservation. Most of the samples were acquired from areas that were already loose from the sculpture but could be clearly identified. Only the sample of the flesh tone was collected in the boundary of a lacunae area. A total of eight samples were collected.
OM The mounting and observation of the samples followed the proceedings applied to the analysis of easel painting or polychrome sculpture (Khandekar, 2003). The samples for cross-section analysis were mounted in polyester resin (BYLAPOX 3085 A and B (2:1)) and were polished using a Struers Planopol-V machine (Ballerup, Denmark). Observation and photography of the samples (100 × magnification) were with an OM, with polarized and transmitted light (Zeiss Stemi 2000-C, Germany) and external artificial light system Zeiss KL 1500 LCD (Germany). Images were acquired with an AxioCam MRcS camera (Zeiss, Germany) and analyzed with Axio Versus 40 V4.4 software from Carl Zeiss Vision GmbH (Germany).
Digital Microscopy The identification of fiber content in the artifact was accomplished by qualitative means, first, by coloration of the pulp and second by the distinction of the fiber constituents based on their morphology. In the first type of qualitative analysis, the nature of pulp is identified by colored reactions produced between the
58
Marta Manso et al.
sample and a chemical reagent, such as Hertzberg stain. The resulting color allows identification of the raw materials and, in some cases, it is useful to determine the manufacturing process and the degree of cooking and bleaching that the pulp may have been submitted (TAPPI T401, 1992–1993). In a morphological analysis, the identification is based on observation of structural features of individual cells as dimension and form distinguishable when observed under magnification of a microscope (Freitas, 1965). Two samples were taken from margin areas, avoiding the interference of coloring material and prepared according to the Method T401—Fiber Analysis of Paper and Paperboard, described on TAPPI Test Methods 1992–1993. Each sample wetted with a drop of deionized water and defibered with the help of two dissecting needles, under the 1.6 × magnification of a stereo microscope SMT 4 (Askania Mikroskop Tecknik, Rathenow, Germany). The presence of water in this process facilitates the risk of breaking down the existing bonds and the risk of fragmenting the fibers during the defiberation. Observation of the samples was carried out with digital microscopy using an AM4013-FVW Dino-Lite Pro USB digital microscope (AnMo Electronics Corporation) with resolution of 1.3 Mp and magnification up to 225 × photomicrographs were registered with visible light at 50 and 220 × magnification.
Micro Energy-Dispersive X-ray Fluorescence (µ-XRF) Bruker’s M4 Tornado µ-XRF spectrometer (Bruker, Berlin, Germany) was used. The sample chamber is a large vacuum tight rectangular box. Inside the chamber is an X–Y–Z-stage supporting the sample, which will be excited from top. The correct sample height is adjusted by an autofocus system. OM allow a sample view inside the instrument that allows it to be positioned exactly. The excitation of fluorescence radiation is performed by an X-ray tube. The tube is an Rh micro-focus side window tube powered by an air cooled low power HV-generator. An X-Ray optic poly-capillary was used, offering a small spot size down to 25 µm combined with high excitation intensity. Detection of fluorescence radiation was performed by an energy-dispersive silicon drift detector with 30 mm2 sensitive area and energy resolution of 142 eV for Mn-Kα. The X-ray generator was operated at 50 kV and 600 µA. An Al filter of 100 μm was used. Analyses were carried out under 20 mbar vacuum conditions. Spectra acquisition and evaluation were carried out using Esprit software from Bruker.
Confocal Raman Microscopy Raman analyses were undertaken using a Horiba-Jobin Yvon XploRA confocal spectrometer (Villeneuve d’Ascq, France), operated at a wavelength of 785 nm and maximum incident power of 0.2 mW. Using a 100 × magnification objective with a pinhole of 500 µm and an entrance slit of 100 µm, the scattered light collected by the objective was
dispersed onto the air-cooled CCD array of an Andor iDus detector (Belfast, Northern Ireland) with 1,200 lines/mm grating. Raman microscopy was performed at a range of 100–3,200 cm − 1. Spectra deconvolution was performed using LabSpec (V5.78; Horiba-Jobin Yvon Villeneuve d’Ascq, France). Identification of pigments was made according to Bell et al. (1997), Burgio & Clark (2001), and Castro et al. (2005), Spectral ID, and our own reference spectra.
SEM-EDS The SEM-EDS analyses were performed in a Hitachi SU-70 UHR Schottky FESEM (Tokyo, Japan) and Bruker Quantax 400 EDS system (Berlin, Germany) with an AXS XFlash silicon drift detector (Bruker, Berlin, Germany) using a 15 kV accelerating voltage and current of 32 µA. Element analysis was taken from an area of 1 µm2, selected regarding its homogeneity and lack of voids, with spectrum acquisition times of 60 s. The areas were scanned using a 7,000 × magnification and elemental and semi-quantitative results were achieved after three measurements. The semi-quantitative results were based on a peak-to-background ZAF evaluation method (P/B-ZAF), ZAF being a matrix correction mainly based on analytical expression for atomic number (Z) dependent X-ray yield, self-absorption (A), and secondary fluorescence enhancement (F), provided by the Esprit software. The semi-quantitative results were normalized to 100%. The samples were coated with carbon.
RESULTS The two samples acquired an overall brownish pink tone (Figs. 2a, 2b) after staining with the Hertzberg solution (Aitken et al., 1988). This tone is attributed to rag pulp that can easily be distinguished from the blue or violet-blue color of chemical wood pulp and grass and from vivid yellow of mechanical pulp. Observation of paper samples under the digital microscope revealed the presence of fibers with different features. By comparing the identified fibers with reference to known samples of cotton or linen/hemp fibers, the simultaneous presence of cotton and linen/hemp fibers was verified. Images revealed a twisted or convoluted structure characteristic of cotton fibers (Fig. 2a). The long and uniform fibers observed in Fig. 2b) are characteristic of bast fibers such as linen, hemp, ramie, and jute that are different than the twisted structure of cotton. The uniformity and length of observed fibers associated with bends distributed randomly along the fiber length and to striations identified them as linen or hemp fibers (Collings & Milner, 1978; Coté, 1980; Sisko & Pfäffli, 1995). Unfortunately, it is not possible to distinguish the simple striations, characteristic of linen fibers, from the V striation from hemp cells with the magnification used. Elemental analysis using µ-XRF revealed mainly Ca with traces of Fe, Cu, Zn, and Pb (Fig. 2c). Raman microanalysis only detected calcite (CaCO3) confirmed by the presence of the characteristic bands at 153, 282, 711, and 1,086 cm − 1.
Micro-Analytical Study of a Rare Papier-Mâché Sculpture
59
Figure 2. Photomicrographs of stained fibers under visible light at 220 × magnification. a: cotton fibers. b: Red arrows point to bends in bast fibers. c: Micro X-ray fluorescence spectrum of paper pulp.
Figure 3. Micro X-ray fluorescence and Raman spectra of pigments used to obtain a pink hue in original and overpaint layers.
Two polychrome layers, an original and an overpaint, which often present the same color palette, were perceived on the sculpture. μ-XRF and Raman techniques enabled the differentiation of the coloring materials used in both polychromies. Both pink layers from the flesh tone had in common a high amount of Zn and traces of Hg. Furthermore, Pb was only found in the overpaint layer. Raman characteristic bands of vermilion (HgS) (256, 285, 347 cm − 1), and vermillion with lead white (Pb3(CO3)2(OH)2) (1,050 cm − 1) were identified respectively in the original and in the overpaint layers (Fig. 3). Elemental mapping obtained using μ-XRF revealed similar gilding methods in both polychromies (Fig. 4) with the same stratigraphic sequence: brown color, gold leaf, red aluminosilicate, and iron-based layer (bole), and finally a white calcium-based ground layer. Mainly Mn and Fe were detected in the original brown color, and vermilion (255, 287, 344 cm − 1), calcite (715, 1,088 cm − 1), and carbon black (disordered carbon) (1,323, 1,596 cm − 1) were identified on the overpaint (Fig. 5). OM and SEM images of the gilded samples revealed that the gilding technique from both layers share important features such as the extreme thinness of the gold leaf and the presence of a ground and bole layers, typical of a traditional gilding, with good adhesion and cohesion between them (Fig. 6). The main
differences were obtained in the EDS analysis, which can be observed in Table 1. Raman analysis revealed the presence of calcite and gypsum (CaSO4∙2H2O) in the ground layer of both polychromies (Fig. 7). The material used in Saint Antony’s hands was identified by Raman microscopy as paraffin (Fig. 7).
DISCUSSION The presence of different cotton and linen or hemp fibers on paper substrate indicates the use of recycled paper as a raw material. It is known that rag pulp made of mixtures of linen and hemp were frequently present in paper specimens made in Europe between 1400 and 1800, so it is natural to find recycled paper made of these papers in the following centuries. It is also referenced that cotton or cotton fabrics were not common enough to generate substantial raw material for papermaking and therefore cotton fibers were rare in papers before 1800 (Barrett, 1989), but the invention of the cotton gin in the late 18th century increased the use of this fiber. Calcite was identified within fibers as the filler, which was applied to the paper mainly to improve sheet formation by filling the voids between the fibers (Krogerus, 1999). The presence of traces of Fe, Cu, Zn, and Pb can be explained
60
Marta Manso et al.
Figure 4. Photograph and X-ray fluorescence elemental mapping obtained at 100 × for Ca, Fe, Au, and Hg in the recent gilding.
Figure 5. Micro X-ray fluorescence and Raman spectra of pigments used to obtain the brown hue.
taking into account that cellulose fibers very rapidly accumulate dissolved metals in water (Manso et al., 2008a). The two polychrome layers identified on the sculpture are most likely from different periods. They are both gilded using the estofado technique and present the same color palette consisting mainly of brown from clothing and pink from the flesh tones. Both pink layers were achieved admixing vermillion and a zinc-based pigment, most likely zinc white (ZnO), although the latter was not identified by Raman microscopy. The presence of lead, found only in the pink overpaint, is probably from the mixture of lead and zinc white and allowed the distinction of the two layers. In both layers, the estofado technique was used in the decoration of Saint Anthony’s cloak where gold leaf was applied on a red iron-based bole, likely red ocher. Brown
coloring material was then applied over the gold leaf and carefully scraped in order for the gold to be visible and to create different patterns, like the texture of the cloak or the leaf decoration in the lower part of the vest. However, brown colors used in both layers are distinct. The original brown color consisted of iron and manganese-based pigment, likely umber (iron and manganese oxides), while the brown overpaint was obtained admixing vermilion, carbon black, and calcium carbonate pigments. The gold from both layers share different features in the percentage of elements of the alloy. Purest 23-carat gold was used in the recent layer, while the original gold was 22.30 carat. The significant difference is in the relative silver percentage, varying between 2.55% (recent) and 5.27% (original).
Micro-Analytical Study of a Rare Papier-Mâché Sculpture
Figure 6. Detail of original polychrome layer as seen in a crosssection at 100 × magnification. a: brown color; (b) gold leaf; (c) bole; and (d) ground layer. Table 1. (wt%). Gold Alloy Original Recent
Gold Leaf Semi-Quantification Obtained by SEM-EDS Au
Ag
Cu
Carat
92.93 ± 11.33 95.84 ± 12.75
5.27 ± 0.68 2.55 ± 0.42
1.80 ± 0.31 1.61 ± 0.32
22.30 22.98
Deviations were obtained using 3 Sigma. SEM-EDS, scanning electron microscopy-energy-dispersive spectroscopy.
61
These conclusions are based on the identification in both polychromies of zinc white and vermillion: the first pigment was only introduced commercially after 1834, although there are references to its use since 1780 and artificial vermillion was used since the early medieval period until the 20th century—the introduction of cadmium red in 1910 led to a fast decrease of its use (Eastaugh et al., 2008). Through XRF elemental mapping it was possible to establish that both the original and overpaint layers were applied with the same decorative technique—estofado. The short period of time that occurred from the creation of the sculpture to its posterior intervention, as well as the employment of the same decorative techniques, lead us to believe that the artwork suffered not from a fashion adjustment, but from the necessity to return to its original appearance. The reason for this procedure is unclear, but some scenarios are possible, such as an accident or the decay of the materials applied to its construction. It is interesting that the recent work was achieved using the same techniques as the original, although with different pigments contemporary to the period of the intervention. The fact that the gold used in the recent gilding is of high quality also points to careful work.
ACKNOWLEDGMENT M. Manso, M. Guerra, A. Bidarra, and S. Pessanha acknowledge the support of the Portuguese Foundation for Science and Technology for the grants SFRH/BPD/70031/2010, SFRH/ BPD/92455/2013, SFRH/BD/38593/2007, and SFRH/BPD/ 94234/2013, respectively. Fundação Nun’Álvares (Gouveia, Portugal) and Dra. Joana Lourenço Pereira (Departamento dos Bens Culturais da Diocese da Guarda-Portugal).
REFERENCES
Figure 7. Raman spectra of the ground layer and of Saint Antony’s hand material.
The ground layer of both polychromies was achieved by the use of an admixture of calcium carbonate and sulfate.
CONCLUSIONS The studied sculpture provided an opportunity to learn more about papier-mâché artworks. Several questions were addressed such as the study of the support and pigments in order to establish a possible timeline for its production. The identified coloring materials lead us to believe that the sculpture was produced in the 19th century, being overpainted during the end of the same century or the beginning of the 20th century.
AITKEN, Y., CADEL, F. & VOILLOT, C. (1988). Constituants fibreux des pâtes papiers et cartons: Pratique de l’analyse, 1st ed. Grenoble, France: Centre Technique du Papier et Ecole Française de Papeterie et des Industries Graphiques. BARRETT, T.D. (1989). Early European papers/contemporary conservation papers—a report on research undertaken from fall 1984 through fall 1987. Paper Conservator 13, 1–189. BELL, I.M., CLARK, R.J.H. & GIBBS, P.J. (1997). Raman spectroscopic library of natural and synthetic pigments (Pre-1850 AD). Spectroc Acta Part A 53, 2159–2179. BIDARRA, A., ANTUNES, P., DESTERRO, T., COROADO, J. & ROCHA, F. (2013). Technical analysis of a renaissance limestone altarpiece. In The Renaissance Workshop: The Materials and Techniques of Renaissance Art, Saunders, D., Spring, M. & Meek, A. (Eds.), pp. 162–164. London: Archetype Publications. BIDARRA, A., COROADO, J. & ROCHA, F. (2009). Gold leaf analysis of three Porto Baroque altarpieces. Archeo Science, Revue d’Archéométrie 33, 417–422. BIDARRA, A., COROADO, J. & ROCHA, F. (2011). Retábulos Barrocos: definição de uma metodologia para o estudo da folha de ouro. In AuCORRE—Conservação e técnicas de análise para o estudo e salvaguarda do património metálico. Available at http://aucorre. org/project/pdf/301530.pdf (accessed 15 December 2013).
62
Marta Manso et al.
BURGIO, L. & CLARK, R.J.H. (2001). Library of FT-Raman spectra of pigments, minerals, pigment media and varnishes, and supplement to existing library of Raman spectra of pigments with visible excitation. Spectrochim Acta Part A 57, 1491–1521. CALVO, A. (2003). Conservación y Restauración. Materiales técnicas y procedimientos. De la A a la Z, 3rd ed. Barcelona, Spain: Ediciones del Serbal. CARVALHO, M.L., MANSO, M., PESSANHA, S. & GUILHERME, A. (2009). Quantification of mercury in XVIIITH century books. J Cult Herit 10, 435–438. CASTRO, K., PESSANHA, S., PROIETTI, N., PRINCI, E., CAPITANI, D., CARVALHO, M.L. & MADARIAGA, J.M. (2008a). Non-invasive and non-destructive NMR, RAMAN and XRF analysis of a blaeu’s coloured map from the XVII century. Anal Bioanal Chem 391, 433–441. CASTRO, K., PROIETTI, N., PRINCI, E., PESSANHA, S., CARVALHO, M.L., VICINI, S., CAPITANI, D. & MADARIAGA, J.M. (2008b). Multianalytical spectroscopic analysis of a coloured dutch map from the XVIII century. Analytica Chimica Acta 623, 187–194. CASTRO, K., PEREZ-ALONSO, M., RODRIGUEZ-LASO, M.D., FERNANDEZ, L.A. & MADARIAGA, J.M. (2005). On-line FT-Raman and dispersive Raman spectra database of artists’ materials (e-VISART database). Anal Bioanal Chem 382, 248–258. COLLINGS, T. & MILNER, D. (1978). The identification of oriental paper-making fibres. Paper Conservator 3, 51–79. COTÉ, W.A. (1980). Papermaking Fibers, A Photomicrographic Atlas. New York: Syracuse University Press. DUMONT, B., DUPONT, A.L., PAPILLON, M.C. & JEANNEL, G.F. (2011). Technical study and conservation treatment of a horse model by Dr Auzoux. Stud Conserv 56, 58–74. EASTAUGH, N., WALSH, V., CHAPLIN, T. & SIDDALL, R. (2008). Pigment Compendium—A Dictionary and Optical Microscopy of Historic Pigments. Oxford, UK: Elsevier Science & Technology. FRANCO, A. & HENRIQUES, A. (Eds.) 2012). Do Mar e da Terra. Presépios naturalistas. Estudo e Reabilitação. Catálogo de exposição. Lisboa: DGPC/MNAA. (Col. «Sala do Tecto Pintado»). FREITAS, M. (1965). Primeira contribuição para a análise micrográfica de pastas e papéis. Revista da Junta das Investigações do Ultramar 13, 545–570. GUERRA, M.F. & CALLIGARO, T. (2004). Gold traces to trace gold. J Archaeol Sci 31, 1199–1208. HEIN, A. & DEGRIGNY, C. (2008). The application of non-destructive technologies for the damage assessment of metal objects. In Metals and Museums in the Mediterranean, Protecting, Preserving and Interpreting, Argyropoulos V. (Ed.), pp. 125–140. Greece: The PROMET Consortium.
KHANDEKAR, N. (2003). Preparation of cross sections from easel paintings. Rev Conserv 4, 52–64. KROGERUS, B. (1999). Fillers and pigments. In Papermaking Chemistry, Neimo, L. (Ed.) 117–149. Helsinki, Finland: TAPPI press. MANSO, M. & CARVALHO, M.L. (2007). Elemental identification of document paper by X-ray fluorescence spectrometry. J Anal At Spectrom 22, 164–170. MANSO, M., CARVALHO, M.L., QUERALT, I., VICINI, S. & PRINCI, E. (2011). Investigation on the composition of historical and modern italian papers by EDXRF, XRD and SEM-EDS. App Spectrosc 65, 52–59. MANSO, M., COSTA, M. & CARVALHO, M.L. (2008a). X-ray fluorescence spectrometry on paper characterization: Case study on XVIII and XIX century documents. Spectroch Acta B 63, 1320–1323. MANSO, M., COSTA, M. & CARVALHO, M.L. (2008b). Comparison of elemental content on modern and ancient papers by EDXRF. Appl Phys A 90, 43–48. PESSANHA, S., GAC, A., MADEIRA, T.I., BRUNEEL, J.L. & CARVALHO, M.L. (2012). Evaluation of the intervention of a folding screen belonging to the momoyama period by Raman spectroscopy using different wavelengths. J Raman Spectrosc 43, 1699–1706. PESSANHA, S., CARVALHO, M.L., CABAÇO, M.I., VALADAS, S., BRUNEEL, J.L., BESNARD, M. & RIBEIRO, M.I. (2010). Characterization of two pairs of 16th century nambam folding screens by Raman, EDXRF and FTIR spectroscopies. J Raman Spectrosc 41, 1220–1226. PESSANHA, S., GUILHERME, A., BITTENCOURT, K., CABAÇO, M.I., BRUNEEL, J.L., BESNARD, M. & CARVALHO, M.L. (2009). Study of a XVIII century hand-painted Chinese wallpaper by multianalytical non-destructive techniques. Spectrochimica Acta Part B 61, 922–928. SISKO, M. & PFÄFFLI, I. (1995). Fiber Atlas, Identification of Papermaking Fibers. Germany: Springer Publisher. TAPPI T401 (1992–1993). Fiber Analysis of Paper and Paper Board. TAPPI Standards, Technical Association of the Pulp and Paper Industry, Atlanta. THORNTON, J. (1993). The history, technology, and conservation of architectural papier-mâché. J Am Inst Conserv 32, 165–176. VAN DER REYDEN, D. & WILLIAMS, D.C. (1986). The Technology and Conservation Treatment of a 19th Century “Papier-Mâché” Chair. In Preprints of the American Institute for Conservation, 14th Annual Meeting, Chicago, pp. 125–142.
© MICROSCOPY SOCIETY OF AMERICA 2014
Micro-Analytical Study of a Rare Papier-Mâché Sculpture Marta Manso,1,2,* Ana Bidarra,3,4 Stéphane Longelin,1 Sofia Pessanha,1 Adriana Ferreira,5 Mauro Guerra,1,7 João Coroado,3,6 and Luísa Carvalho1,7 1
Centro de Física Atómica da Universidade de Lisboa, Av. Professor Gama Pinto 2, 1649-003 Lisboa, Portugal Faculdade de Belas-Artes da Universidade de Lisboa, Largo da Academia Nacional de Belas-Artes, 1249-058 Lisboa, Portugal 3 Geobiotec/Departamento de Geociências, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal 4 Cinábrio, Conservação e Restauro, R. Almirante Cândido dos Reis, 3800-096 Aveiro, Portugal 5 Arquivo Municipal de Lisboa, Rua B Bairro da Liberdade, Lote 3-6, Piso 0, 1070-050 Lisboa, Portugal 6 Departamento de Conservação e Restauro, Instituto Politécnico de Tomar, Quinta do Contador, Estrada da Serra, 2300-313 Tomar, Portugal 7 Departamento de Física, Faculdade de ciências e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal 2
Abstract: The analysis of a Portuguese “papier-mâché” sculpture depicting Saint Anthony is presented in this case study. Several questions were addressed such as the characteristics of the support, pigments used, and artistic technique in order to establish a possible timeline for its production. Qualitative analyses of the cross-sections and of the paper support were performed by optical microscopy using reflected light. Two polychrome layers from different periods and a rag pulped support were identified on the sculpture. The use of micro X-ray fluorescence and Raman microscopy techniques enabled the differentiation of coloring materials used in both polychromies. Semi-quantitative analyses of the gilded samples were also performed by scanning electron microscopy in combination with energy-dispersive spectroscopy allowing the determination of a common Au–Ag–Cu alloy with differences in the purity of the gold. The identified coloring materials lead us to believe that the sculpture was produced in the 19th century, being overpainted in the first half of the 20th century. Key words: papier-mâché, sculpture, micro-Raman, micro-XRF, SEM-EDS
I NTRODUCTION “Papier-mâché” is a term that has been applied to threedimensional (3D) objects of recycled paper fiber whether layered in sheet form with an adhesive or cast as beaten pulp (van der Reyden & Williams, 1986; Thornton, 1993). Its origin is at least as ancient as the invention of paper itself (China during the Han Dynasty, c. 202 B.C.–220 A.D.) and the first 3D objects were produced during this period. The spread of papermaking was followed by the production of objects made with paper fiber and it was widely used in several countries as an architectural material from the 17th to the early 19th century. In Portugal, there are very few examples of paper objects either in architectural decoration or in isolated objects such as sculptures (Franco, 2012). In this last case most of the known works are integrated in larger structures made with different materials, such as clay nativity scenes, and often represents secondary characters. The studied sculpture, depicting Saint Anthony, is a full-length piece ~ 30 cm high. It is a standing image, placed in a frontal position and facing forward. In his hand Saint Anthony holds baby Jesus. The color pallet is very simple and consistent with the iconographic representation of the Saint: a Franciscan monk. The tunic is brown and decorated Received March 28, 2014; accepted August 12, 2014 *Corresponding author. [email protected]
with small fish-scale drawings, and the hood, the extremity of the sleeves and the lower part of the tunic are decorated with a vegetalist pattern. This decoration is achieved through the “estofado technique” that allows seeing the gold leaf beneath the brown when scraping the upper paint layer. The globe on left hand of baby Jesus is blue and the base mimics a blue marble effect. The sculpture is composed of several materials, but the main one is papier-mâché that was used to mold the body of the sculpture. The hands are made of wax, baby Jesus in clay, Saint Anthony’s eyes in glass, and a wood pin was used to join the head to the torso. The base is also made of wood. The lower part of the sculpture is hollow (Fig. 1). Very few technical studies have been published on papiermâché polychromated sculptures. Dumont et al. (2011) reported the techniques and materials involved in fabrication of a papier-mâché anatomical model of a horse created in the mid-19th century by Dr. Auzoux. Investigation was conducted using analytical techniques such as X-ray radiography for examination of the internal structure, gas chromatography coupled to mass spectrometry for identification of organic components in the paper pulp and as pigment binders, scanning electron microscopy with energy-dispersive X-ray spectrometry (SEM-EDS) and Raman spectroscopy for paper fillers and ground and paint layer characterization. In the study of Saint Anthony’s sculpture, we report the use of digital microscopy for identification of the type of
Micro-Analytical Study of a Rare Papier-Mâché Sculpture
57
Figure 1. Front and rear view of Saint Anthony papier-mâché sculpture before the conservation treatment.
fibers and binders used in the papier-mâché production. The sample cross-sections were also analyzed through optical microscopy (OM); this technique allowed determination of the number, thickness, adhesion, and cohesion of the different layers (Calvo, 2003), the presence of overpaints and provided a comparative approach to the optical characteristics of the different strata. For paper filler and pigment identification, micro X-ray fluorescence (μ-XRF) and Raman analysis were used as complimentary techniques. The decoration technique of the cloak was determined using μ-XRF elemental mapping. The gold leaf was analyzed by SEM-EDS. The use of SEM in combination with EDS increases sensitivity for lighter elements and the spatial resolution for spot analysis is higher. In addition, it allows examination of the microstructure with line scans or 2D mapping of element concentration (Guerra & Calligaro, 2004; Hein & Degrigny, 2008). This method has been widely used in determination of the major elements and purity of the gold alloys (Bidarra et al., 2009, 2013, 2011). Suitability of XRF for paper filler characterization was reported by Manso et al. (2007, 2008a, 2008b) in historical and modern paper elemental characterization. X-ray fluorescence with Raman analysis is often used for pigments determination on paper support folding screens (Pessanha et al., 2010, 2012), wallpapers (Pessanha et al., 2009), and maps (Castro et al., 2008a, 2008b). Application of elemental mapping has been useful in the study of pigment dissemination (Carvalho et al., 2009) and ink composition(Manso et al., 2011). The material and artistic production of the sculpture under study was carried out as an opportunity to learn more about a rather unknown and rare art form and to establish a possible timeline for its production.
MATERIALS
AND
METHODS
The number of samples obtained from the sculpture was limited owing to its state of conservation. Most of the samples were acquired from areas that were already loose from the sculpture but could be clearly identified. Only the sample of the flesh tone was collected in the boundary of a lacunae area. A total of eight samples were collected.
OM The mounting and observation of the samples followed the proceedings applied to the analysis of easel painting or polychrome sculpture (Khandekar, 2003). The samples for cross-section analysis were mounted in polyester resin (BYLAPOX 3085 A and B (2:1)) and were polished using a Struers Planopol-V machine (Ballerup, Denmark). Observation and photography of the samples (100 × magnification) were with an OM, with polarized and transmitted light (Zeiss Stemi 2000-C, Germany) and external artificial light system Zeiss KL 1500 LCD (Germany). Images were acquired with an AxioCam MRcS camera (Zeiss, Germany) and analyzed with Axio Versus 40 V4.4 software from Carl Zeiss Vision GmbH (Germany).
Digital Microscopy The identification of fiber content in the artifact was accomplished by qualitative means, first, by coloration of the pulp and second by the distinction of the fiber constituents based on their morphology. In the first type of qualitative analysis, the nature of pulp is identified by colored reactions produced between the
58
Marta Manso et al.
sample and a chemical reagent, such as Hertzberg stain. The resulting color allows identification of the raw materials and, in some cases, it is useful to determine the manufacturing process and the degree of cooking and bleaching that the pulp may have been submitted (TAPPI T401, 1992–1993). In a morphological analysis, the identification is based on observation of structural features of individual cells as dimension and form distinguishable when observed under magnification of a microscope (Freitas, 1965). Two samples were taken from margin areas, avoiding the interference of coloring material and prepared according to the Method T401—Fiber Analysis of Paper and Paperboard, described on TAPPI Test Methods 1992–1993. Each sample wetted with a drop of deionized water and defibered with the help of two dissecting needles, under the 1.6 × magnification of a stereo microscope SMT 4 (Askania Mikroskop Tecknik, Rathenow, Germany). The presence of water in this process facilitates the risk of breaking down the existing bonds and the risk of fragmenting the fibers during the defiberation. Observation of the samples was carried out with digital microscopy using an AM4013-FVW Dino-Lite Pro USB digital microscope (AnMo Electronics Corporation) with resolution of 1.3 Mp and magnification up to 225 × photomicrographs were registered with visible light at 50 and 220 × magnification.
Micro Energy-Dispersive X-ray Fluorescence (µ-XRF) Bruker’s M4 Tornado µ-XRF spectrometer (Bruker, Berlin, Germany) was used. The sample chamber is a large vacuum tight rectangular box. Inside the chamber is an X–Y–Z-stage supporting the sample, which will be excited from top. The correct sample height is adjusted by an autofocus system. OM allow a sample view inside the instrument that allows it to be positioned exactly. The excitation of fluorescence radiation is performed by an X-ray tube. The tube is an Rh micro-focus side window tube powered by an air cooled low power HV-generator. An X-Ray optic poly-capillary was used, offering a small spot size down to 25 µm combined with high excitation intensity. Detection of fluorescence radiation was performed by an energy-dispersive silicon drift detector with 30 mm2 sensitive area and energy resolution of 142 eV for Mn-Kα. The X-ray generator was operated at 50 kV and 600 µA. An Al filter of 100 μm was used. Analyses were carried out under 20 mbar vacuum conditions. Spectra acquisition and evaluation were carried out using Esprit software from Bruker.
Confocal Raman Microscopy Raman analyses were undertaken using a Horiba-Jobin Yvon XploRA confocal spectrometer (Villeneuve d’Ascq, France), operated at a wavelength of 785 nm and maximum incident power of 0.2 mW. Using a 100 × magnification objective with a pinhole of 500 µm and an entrance slit of 100 µm, the scattered light collected by the objective was
dispersed onto the air-cooled CCD array of an Andor iDus detector (Belfast, Northern Ireland) with 1,200 lines/mm grating. Raman microscopy was performed at a range of 100–3,200 cm − 1. Spectra deconvolution was performed using LabSpec (V5.78; Horiba-Jobin Yvon Villeneuve d’Ascq, France). Identification of pigments was made according to Bell et al. (1997), Burgio & Clark (2001), and Castro et al. (2005), Spectral ID, and our own reference spectra.
SEM-EDS The SEM-EDS analyses were performed in a Hitachi SU-70 UHR Schottky FESEM (Tokyo, Japan) and Bruker Quantax 400 EDS system (Berlin, Germany) with an AXS XFlash silicon drift detector (Bruker, Berlin, Germany) using a 15 kV accelerating voltage and current of 32 µA. Element analysis was taken from an area of 1 µm2, selected regarding its homogeneity and lack of voids, with spectrum acquisition times of 60 s. The areas were scanned using a 7,000 × magnification and elemental and semi-quantitative results were achieved after three measurements. The semi-quantitative results were based on a peak-to-background ZAF evaluation method (P/B-ZAF), ZAF being a matrix correction mainly based on analytical expression for atomic number (Z) dependent X-ray yield, self-absorption (A), and secondary fluorescence enhancement (F), provided by the Esprit software. The semi-quantitative results were normalized to 100%. The samples were coated with carbon.
RESULTS The two samples acquired an overall brownish pink tone (Figs. 2a, 2b) after staining with the Hertzberg solution (Aitken et al., 1988). This tone is attributed to rag pulp that can easily be distinguished from the blue or violet-blue color of chemical wood pulp and grass and from vivid yellow of mechanical pulp. Observation of paper samples under the digital microscope revealed the presence of fibers with different features. By comparing the identified fibers with reference to known samples of cotton or linen/hemp fibers, the simultaneous presence of cotton and linen/hemp fibers was verified. Images revealed a twisted or convoluted structure characteristic of cotton fibers (Fig. 2a). The long and uniform fibers observed in Fig. 2b) are characteristic of bast fibers such as linen, hemp, ramie, and jute that are different than the twisted structure of cotton. The uniformity and length of observed fibers associated with bends distributed randomly along the fiber length and to striations identified them as linen or hemp fibers (Collings & Milner, 1978; Coté, 1980; Sisko & Pfäffli, 1995). Unfortunately, it is not possible to distinguish the simple striations, characteristic of linen fibers, from the V striation from hemp cells with the magnification used. Elemental analysis using µ-XRF revealed mainly Ca with traces of Fe, Cu, Zn, and Pb (Fig. 2c). Raman microanalysis only detected calcite (CaCO3) confirmed by the presence of the characteristic bands at 153, 282, 711, and 1,086 cm − 1.
Micro-Analytical Study of a Rare Papier-Mâché Sculpture
59
Figure 2. Photomicrographs of stained fibers under visible light at 220 × magnification. a: cotton fibers. b: Red arrows point to bends in bast fibers. c: Micro X-ray fluorescence spectrum of paper pulp.
Figure 3. Micro X-ray fluorescence and Raman spectra of pigments used to obtain a pink hue in original and overpaint layers.
Two polychrome layers, an original and an overpaint, which often present the same color palette, were perceived on the sculpture. μ-XRF and Raman techniques enabled the differentiation of the coloring materials used in both polychromies. Both pink layers from the flesh tone had in common a high amount of Zn and traces of Hg. Furthermore, Pb was only found in the overpaint layer. Raman characteristic bands of vermilion (HgS) (256, 285, 347 cm − 1), and vermillion with lead white (Pb3(CO3)2(OH)2) (1,050 cm − 1) were identified respectively in the original and in the overpaint layers (Fig. 3). Elemental mapping obtained using μ-XRF revealed similar gilding methods in both polychromies (Fig. 4) with the same stratigraphic sequence: brown color, gold leaf, red aluminosilicate, and iron-based layer (bole), and finally a white calcium-based ground layer. Mainly Mn and Fe were detected in the original brown color, and vermilion (255, 287, 344 cm − 1), calcite (715, 1,088 cm − 1), and carbon black (disordered carbon) (1,323, 1,596 cm − 1) were identified on the overpaint (Fig. 5). OM and SEM images of the gilded samples revealed that the gilding technique from both layers share important features such as the extreme thinness of the gold leaf and the presence of a ground and bole layers, typical of a traditional gilding, with good adhesion and cohesion between them (Fig. 6). The main
differences were obtained in the EDS analysis, which can be observed in Table 1. Raman analysis revealed the presence of calcite and gypsum (CaSO4∙2H2O) in the ground layer of both polychromies (Fig. 7). The material used in Saint Antony’s hands was identified by Raman microscopy as paraffin (Fig. 7).
DISCUSSION The presence of different cotton and linen or hemp fibers on paper substrate indicates the use of recycled paper as a raw material. It is known that rag pulp made of mixtures of linen and hemp were frequently present in paper specimens made in Europe between 1400 and 1800, so it is natural to find recycled paper made of these papers in the following centuries. It is also referenced that cotton or cotton fabrics were not common enough to generate substantial raw material for papermaking and therefore cotton fibers were rare in papers before 1800 (Barrett, 1989), but the invention of the cotton gin in the late 18th century increased the use of this fiber. Calcite was identified within fibers as the filler, which was applied to the paper mainly to improve sheet formation by filling the voids between the fibers (Krogerus, 1999). The presence of traces of Fe, Cu, Zn, and Pb can be explained
60
Marta Manso et al.
Figure 4. Photograph and X-ray fluorescence elemental mapping obtained at 100 × for Ca, Fe, Au, and Hg in the recent gilding.
Figure 5. Micro X-ray fluorescence and Raman spectra of pigments used to obtain the brown hue.
taking into account that cellulose fibers very rapidly accumulate dissolved metals in water (Manso et al., 2008a). The two polychrome layers identified on the sculpture are most likely from different periods. They are both gilded using the estofado technique and present the same color palette consisting mainly of brown from clothing and pink from the flesh tones. Both pink layers were achieved admixing vermillion and a zinc-based pigment, most likely zinc white (ZnO), although the latter was not identified by Raman microscopy. The presence of lead, found only in the pink overpaint, is probably from the mixture of lead and zinc white and allowed the distinction of the two layers. In both layers, the estofado technique was used in the decoration of Saint Anthony’s cloak where gold leaf was applied on a red iron-based bole, likely red ocher. Brown
coloring material was then applied over the gold leaf and carefully scraped in order for the gold to be visible and to create different patterns, like the texture of the cloak or the leaf decoration in the lower part of the vest. However, brown colors used in both layers are distinct. The original brown color consisted of iron and manganese-based pigment, likely umber (iron and manganese oxides), while the brown overpaint was obtained admixing vermilion, carbon black, and calcium carbonate pigments. The gold from both layers share different features in the percentage of elements of the alloy. Purest 23-carat gold was used in the recent layer, while the original gold was 22.30 carat. The significant difference is in the relative silver percentage, varying between 2.55% (recent) and 5.27% (original).
Micro-Analytical Study of a Rare Papier-Mâché Sculpture
Figure 6. Detail of original polychrome layer as seen in a crosssection at 100 × magnification. a: brown color; (b) gold leaf; (c) bole; and (d) ground layer. Table 1. (wt%). Gold Alloy Original Recent
Gold Leaf Semi-Quantification Obtained by SEM-EDS Au
Ag
Cu
Carat
92.93 ± 11.33 95.84 ± 12.75
5.27 ± 0.68 2.55 ± 0.42
1.80 ± 0.31 1.61 ± 0.32
22.30 22.98
Deviations were obtained using 3 Sigma. SEM-EDS, scanning electron microscopy-energy-dispersive spectroscopy.
61
These conclusions are based on the identification in both polychromies of zinc white and vermillion: the first pigment was only introduced commercially after 1834, although there are references to its use since 1780 and artificial vermillion was used since the early medieval period until the 20th century—the introduction of cadmium red in 1910 led to a fast decrease of its use (Eastaugh et al., 2008). Through XRF elemental mapping it was possible to establish that both the original and overpaint layers were applied with the same decorative technique—estofado. The short period of time that occurred from the creation of the sculpture to its posterior intervention, as well as the employment of the same decorative techniques, lead us to believe that the artwork suffered not from a fashion adjustment, but from the necessity to return to its original appearance. The reason for this procedure is unclear, but some scenarios are possible, such as an accident or the decay of the materials applied to its construction. It is interesting that the recent work was achieved using the same techniques as the original, although with different pigments contemporary to the period of the intervention. The fact that the gold used in the recent gilding is of high quality also points to careful work.
ACKNOWLEDGMENT M. Manso, M. Guerra, A. Bidarra, and S. Pessanha acknowledge the support of the Portuguese Foundation for Science and Technology for the grants SFRH/BPD/70031/2010, SFRH/ BPD/92455/2013, SFRH/BD/38593/2007, and SFRH/BPD/ 94234/2013, respectively. Fundação Nun’Álvares (Gouveia, Portugal) and Dra. Joana Lourenço Pereira (Departamento dos Bens Culturais da Diocese da Guarda-Portugal).
REFERENCES
Figure 7. Raman spectra of the ground layer and of Saint Antony’s hand material.
The ground layer of both polychromies was achieved by the use of an admixture of calcium carbonate and sulfate.
CONCLUSIONS The studied sculpture provided an opportunity to learn more about papier-mâché artworks. Several questions were addressed such as the study of the support and pigments in order to establish a possible timeline for its production. The identified coloring materials lead us to believe that the sculpture was produced in the 19th century, being overpainted during the end of the same century or the beginning of the 20th century.
AITKEN, Y., CADEL, F. & VOILLOT, C. (1988). Constituants fibreux des pâtes papiers et cartons: Pratique de l’analyse, 1st ed. Grenoble, France: Centre Technique du Papier et Ecole Française de Papeterie et des Industries Graphiques. BARRETT, T.D. (1989). Early European papers/contemporary conservation papers—a report on research undertaken from fall 1984 through fall 1987. Paper Conservator 13, 1–189. BELL, I.M., CLARK, R.J.H. & GIBBS, P.J. (1997). Raman spectroscopic library of natural and synthetic pigments (Pre-1850 AD). Spectroc Acta Part A 53, 2159–2179. BIDARRA, A., ANTUNES, P., DESTERRO, T., COROADO, J. & ROCHA, F. (2013). Technical analysis of a renaissance limestone altarpiece. In The Renaissance Workshop: The Materials and Techniques of Renaissance Art, Saunders, D., Spring, M. & Meek, A. (Eds.), pp. 162–164. London: Archetype Publications. BIDARRA, A., COROADO, J. & ROCHA, F. (2009). Gold leaf analysis of three Porto Baroque altarpieces. Archeo Science, Revue d’Archéométrie 33, 417–422. BIDARRA, A., COROADO, J. & ROCHA, F. (2011). Retábulos Barrocos: definição de uma metodologia para o estudo da folha de ouro. In AuCORRE—Conservação e técnicas de análise para o estudo e salvaguarda do património metálico. Available at http://aucorre. org/project/pdf/301530.pdf (accessed 15 December 2013).
62
Marta Manso et al.
BURGIO, L. & CLARK, R.J.H. (2001). Library of FT-Raman spectra of pigments, minerals, pigment media and varnishes, and supplement to existing library of Raman spectra of pigments with visible excitation. Spectrochim Acta Part A 57, 1491–1521. CALVO, A. (2003). Conservación y Restauración. Materiales técnicas y procedimientos. De la A a la Z, 3rd ed. Barcelona, Spain: Ediciones del Serbal. CARVALHO, M.L., MANSO, M., PESSANHA, S. & GUILHERME, A. (2009). Quantification of mercury in XVIIITH century books. J Cult Herit 10, 435–438. CASTRO, K., PESSANHA, S., PROIETTI, N., PRINCI, E., CAPITANI, D., CARVALHO, M.L. & MADARIAGA, J.M. (2008a). Non-invasive and non-destructive NMR, RAMAN and XRF analysis of a blaeu’s coloured map from the XVII century. Anal Bioanal Chem 391, 433–441. CASTRO, K., PROIETTI, N., PRINCI, E., PESSANHA, S., CARVALHO, M.L., VICINI, S., CAPITANI, D. & MADARIAGA, J.M. (2008b). Multianalytical spectroscopic analysis of a coloured dutch map from the XVIII century. Analytica Chimica Acta 623, 187–194. CASTRO, K., PEREZ-ALONSO, M., RODRIGUEZ-LASO, M.D., FERNANDEZ, L.A. & MADARIAGA, J.M. (2005). On-line FT-Raman and dispersive Raman spectra database of artists’ materials (e-VISART database). Anal Bioanal Chem 382, 248–258. COLLINGS, T. & MILNER, D. (1978). The identification of oriental paper-making fibres. Paper Conservator 3, 51–79. COTÉ, W.A. (1980). Papermaking Fibers, A Photomicrographic Atlas. New York: Syracuse University Press. DUMONT, B., DUPONT, A.L., PAPILLON, M.C. & JEANNEL, G.F. (2011). Technical study and conservation treatment of a horse model by Dr Auzoux. Stud Conserv 56, 58–74. EASTAUGH, N., WALSH, V., CHAPLIN, T. & SIDDALL, R. (2008). Pigment Compendium—A Dictionary and Optical Microscopy of Historic Pigments. Oxford, UK: Elsevier Science & Technology. FRANCO, A. & HENRIQUES, A. (Eds.) 2012). Do Mar e da Terra. Presépios naturalistas. Estudo e Reabilitação. Catálogo de exposição. Lisboa: DGPC/MNAA. (Col. «Sala do Tecto Pintado»). FREITAS, M. (1965). Primeira contribuição para a análise micrográfica de pastas e papéis. Revista da Junta das Investigações do Ultramar 13, 545–570. GUERRA, M.F. & CALLIGARO, T. (2004). Gold traces to trace gold. J Archaeol Sci 31, 1199–1208. HEIN, A. & DEGRIGNY, C. (2008). The application of non-destructive technologies for the damage assessment of metal objects. In Metals and Museums in the Mediterranean, Protecting, Preserving and Interpreting, Argyropoulos V. (Ed.), pp. 125–140. Greece: The PROMET Consortium.
KHANDEKAR, N. (2003). Preparation of cross sections from easel paintings. Rev Conserv 4, 52–64. KROGERUS, B. (1999). Fillers and pigments. In Papermaking Chemistry, Neimo, L. (Ed.) 117–149. Helsinki, Finland: TAPPI press. MANSO, M. & CARVALHO, M.L. (2007). Elemental identification of document paper by X-ray fluorescence spectrometry. J Anal At Spectrom 22, 164–170. MANSO, M., CARVALHO, M.L., QUERALT, I., VICINI, S. & PRINCI, E. (2011). Investigation on the composition of historical and modern italian papers by EDXRF, XRD and SEM-EDS. App Spectrosc 65, 52–59. MANSO, M., COSTA, M. & CARVALHO, M.L. (2008a). X-ray fluorescence spectrometry on paper characterization: Case study on XVIII and XIX century documents. Spectroch Acta B 63, 1320–1323. MANSO, M., COSTA, M. & CARVALHO, M.L. (2008b). Comparison of elemental content on modern and ancient papers by EDXRF. Appl Phys A 90, 43–48. PESSANHA, S., GAC, A., MADEIRA, T.I., BRUNEEL, J.L. & CARVALHO, M.L. (2012). Evaluation of the intervention of a folding screen belonging to the momoyama period by Raman spectroscopy using different wavelengths. J Raman Spectrosc 43, 1699–1706. PESSANHA, S., CARVALHO, M.L., CABAÇO, M.I., VALADAS, S., BRUNEEL, J.L., BESNARD, M. & RIBEIRO, M.I. (2010). Characterization of two pairs of 16th century nambam folding screens by Raman, EDXRF and FTIR spectroscopies. J Raman Spectrosc 41, 1220–1226. PESSANHA, S., GUILHERME, A., BITTENCOURT, K., CABAÇO, M.I., BRUNEEL, J.L., BESNARD, M. & CARVALHO, M.L. (2009). Study of a XVIII century hand-painted Chinese wallpaper by multianalytical non-destructive techniques. Spectrochimica Acta Part B 61, 922–928. SISKO, M. & PFÄFFLI, I. (1995). Fiber Atlas, Identification of Papermaking Fibers. Germany: Springer Publisher. TAPPI T401 (1992–1993). Fiber Analysis of Paper and Paper Board. TAPPI Standards, Technical Association of the Pulp and Paper Industry, Atlanta. THORNTON, J. (1993). The history, technology, and conservation of architectural papier-mâché. J Am Inst Conserv 32, 165–176. VAN DER REYDEN, D. & WILLIAMS, D.C. (1986). The Technology and Conservation Treatment of a 19th Century “Papier-Mâché” Chair. In Preprints of the American Institute for Conservation, 14th Annual Meeting, Chicago, pp. 125–142.
