Next Article in Journal
Integrating Archaeological Data in Multidisciplinary Environmental Studies—Methodological Notes from High-Resolution Mapping of Ancient Features in Southern Israel
Next Article in Special Issue
The Pigments of the Painter Fleury Richard (1777–1852), a Model for Multidisciplinary Study
Previous Article in Journal
The World’s Oldest Book Printed by Movable Metal Type in Korea in 1239: The Song of Enlightenment
Previous Article in Special Issue
Evolution of Liu Kang’s Palette and Painting Practice for the Execution of Female Nude Paintings: The Analytical Investigation of a Genre
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Scientific Study of the Origin of the Painting from the Early 20th Century Leads to Pablo Picasso

by
Marica Bakovic
1,*,
Slobodanka Karapandza
2,
Sajed Mcheik
3 and
Ana Pejović-Milić
3
1
Department of Human Health and Nutritional Sciences, University of Guelph, 50 Stone Rd E, Guelph, ON N1G 2W1, Canada
2
Historical Archive of Belgrade, Palmira Toljatija 1, 11000 Belgrade, Serbia
3
Department of Physics, Toronto Metropolitan University (Formerly Ryerson University), 350 Victoria St., Toronto, ON M5B 2K3, Canada
*
Author to whom correspondence should be addressed.
Heritage 2022, 5(2), 1120-1140; https://doi.org/10.3390/heritage5020060
Submission received: 12 May 2022 / Revised: 25 May 2022 / Accepted: 26 May 2022 / Published: 28 May 2022

Abstract

:
This study applied multiple scientific approaches to establish the significance of an old work of art, Red Guitar, by examining its historical origin and the color materials used in its creation. Additionally, the study provides thus far unknown pieces of Olga Picasso’s family history to be added to her biography. Scientific approaches included digital X-ray radiography, X-ray fluorescence spectroscopy, infrared Fourier transform spectroscopy, Raman spectroscopy, and elemental thermal conductivity analysis. This combination of techniques provided a broad confirmation as to when the painting was created. The artwork includes colors (white, black, blue, yellow, green, red, and brown/red) and prevalent use of lead- and iron-based historic pigments—chrome yellow, yellow ochre, and red ochre. It also documents the use of unconventional materials, such as the colorant Pigment Red 4, and nitrocellulose. This investigation led to the conclusion that the artwork, Red Guitar, is authentic and in accordance with Picasso’s work during the first two decades of the 20th century.

1. Introduction

Complementary investigative approaches combine historic documentation with scientific analyses of pigments and other materials to collect the information required to establish an artwork’s provenance, and to separate original artworks from forgeries. The information about pigments and synthetic colorants from the late 19th to early 20th century is well documented, and the growing number of databases is a testimony to the increased significance of the scientific methods in art history, conservation, and trade. An extensive analysis of Picasso’s work was of value for the current investigation. The periods that were of particular consideration for this study include the artist’s early years, works from the Blue and Rose periods, and synthetic cubism. The studies of painting materials of six portraits from 1895–1900 [1], Blue room (1902) [2], Acrobat family (1906) [3], and four 1917 cubist paintings [4], were of the most relevance for understanding the artist’s palette and color techniques. During that period, Picasso also started experimenting with non-conventional materials and media supports, and that was of additional benefit for this study [5,6,7,8].
This study applies multidisciplinary approaches to establish the origin and significance of a painting attributed to Picasso through the examination of historical and chronological records and materials used. Furthermore, the study provides new insights about Olga Picasso’s family history, which can be added to her biography. It documents the use of pigments and non-artistic materials, which is in accordance with Picasso’s work during the first decades of the 20th century.

1.1. Art Historical Context

The artwork, Red Guitar, is connected to Nikolaj Kokhlov, the brother of Olga Picasso. The current owner attests that Picasso gifted the work to the Yugoslav Army Chief of Staff, Stevan Hazdic, as a token of appreciation for giving employment to Nikolaj Kokhlov. This statement was the starting point for collecting additional evidence concerning this artwork. The evidence enlightens the topic of familial connections, particularly during the period when Nikolaj Kokhlov was in direct contact with Picasso and Olga. Of prominence, is the information from a relative of Nikolaj Kokhlov, who offered letters, cards, and photos that demonstrate strong familial connections between Nikolaj, Olga, and Pablo Picasso. Figure 1a–d show Nikolaj Kokhlov in Yugoslav Army uniform, his wedding photo, and his wedding certificate from 1928. Moreover, also shown is a postcard from Picasso written to Nikolaj in 1923, signed in the Cyrillic alphabet. The archival documents show that there was a Picasso exhibition in Belgrade in October 1926, organized by Paul Rosenberg and Paul Epstein. Of note, is that this exhibition was right before Nikolaj’s wedding and perhaps an opportunity for interactions between family members. (Supplementary Figure S1).
Of importance is a letter from Olga to her brother in 1930, in which she offered to help him to get a job in the Yugoslav army (Figure 1e). Olga wrote: “I have received your letter however I was saddened by the news that your job transfer has been blocked. According to Derocco everything depends on Hadzic (original art owner, Supplementary Figure S2). Maybe you can talk to Madam Hadzic in my name and ask her to explain what happened and how they can help with your job transfer. Did you talk to brother and sister Petrovic and what they think about this? Maybe it will be better if I additionally write to Madam Hadzic. What do you think? Write to me in detail and figure out precisely on whom your job transfer may depend on in addition to general Hadzic. Maybe someone else. Write to me more details.”
In her letter, Olga references “Derocco,” who is Aleksandar (Sacha) Deroko (1894–1988), a Serbian architect who made the acquaintance of Picasso while on a study scholarship in Paris during the mid-to-late 1920s. Mentioning Derocco is significant because of the fact that Derocco, in his autobiographic book, describes how Nikolaj obtained the job. In his book When the plane was flying above Belgrade (Supplementary Figure S4), Derocco describes his relationship with Olga and Pablo Picasso. Derocco, Rasko (artist and diplomat), and his sister, Nadežda Petrović (artist), were direct contacts between Olga and Nikolaj, and they visited Picasso on several occasions. Derocco describes his role in helping Nikolaj through his friendship with Hadzic’s daughter. More importantly, the book references a painting changing hands from Picasso to the Hadzic family. He described an instance when Picasso sent art gifts to Hadzic’s wife, asking Derocco and Rastko Petrović to deliver them, but there is no mention that Red Guitar was a part of this particular event. He further commented that at the time, Picasso worked for Rosenberg, and was not allowed to give away his art to anybody. Thus, when leaving Rue de la Boete, Picasso asked them not to show drawings to the porter. See more in Figure S3.
One of the gifts Derocco received from Picasso is a variation on the theme of the guitar on the table. Picasso inscribed this work in French: “To Sacha Derocco for our friendship, Picasso,” with the additional inscription on the side that reads: “Paris—December 1927” (Figure S4). This lithograph given to Derocco in 1927, is based on Picasso’s painting “La table devant la fenêtre”, dated 1919. It shows that it was not uncommon for Picasso to keep and give away older artworks as gifts, and that he was still in possession of some of his earlier works in the late 1920s.
It is not surprising that written statements from Hadzic or his heirs, art exhibitions, and so on, could not be found, and this is due to several reasons; the art was a gift, and because of close connections with the king and the monarchy, the Hadzic family was living a quiet post Second World War life in communist Yugoslavia, when it was not popular to sell or export art privately, nor it was very trendy to collect it. This political atmosphere explains why the art was inherited quietly, losing its authenticity in the process, when new owners did not even believe that the art was in fact an original Picasso.
Taken together, the back of the artwork is officially stamped by the Serbian Ministry of Art and Culture, which casts no doubt over the idea that the current owner of Red Guitar is legitimate. The documentation showed the history behind the present narrative in the official statement prior to the owner’s acquisition of the artwork in 1997. The artwork is undoubtedly a gift that is directly linked to particular parties (Olga Picasso, her brother Nikolaj, Alexander Derocco, and Stevan Hadzic); it is mentioned in the letter from Olga to her brother Nikolaj, and in Aleksandar Derocco’s autobiography.

1.2. Description of the Painting

The subject of the painting is cubistic and belongs to synthetic cubism. This artwork, named Red Guitar, is a 35.7 cm × 25.9 cm oil on cardboard painting (Figure 2a). It depicts a series of clearly delineated planes of color, rendered on a bright palette that radiates uniformly across the composition. Emerging from these planes is a table surface set against a mottled red background. Upon this table rests a man’s head, guitar, and bottle, among other abstract forms. The contours of each facet are clearly delineated with bold strokes of black paint, and each facet conjures a sense of compositional flatness.
A signature reading “Picasso” appears in the artwork’s upper right side (viewer’s left), with the date reading “9-4-21” in the artwork’s lower left side (viewer’s right). The enlarged images of the date and signature are shown in Figure 2b,c. Color smeared over the signature suggests that the signature was executed before the painting was finished. The date at the bottom, on the other hand, was placed on with a dry, thick, oil layer after the artwork was completed. The penmanship of numbers “4” and “2” was affected by the uneven paint surface, and the number “9” has stains from a neighboring green paint. The Red Guitar signature and date are not notably different from Picasso’s signatures and dates from the same period, and they are deemed authentic by the National Museum of Art (Belgrade, 20 April 2011, del.br 37.14).
The geometry and color palette in Red Guitar relate to Picasso’s cubistic paintings from the same period. The following information was obtained from the “Online Picasso Project” (https://picasso.shsu.edu/ (accessed on 1 January 2022)). The watercolor, Composition 1 (10 September 1920), has identical arrangements of the red-blue-yellow-green colors with a central red area, which is reminiscent of Red Guitar. Gouache paintings from 1920 (Femme à la guitare (Fall 1920), Piano I (Fall 1920), and Le Piano (26 September 1920)) also share the central positions of red and the color combination of Red Guitar. Red Guitar also has colors and elements from an earlier painting, Fillette au cerceau (Winter 1918~1919). The extensive use of white paint in the red backdrop of the Red Guitar is similar to the white use in Compotier et bouteille sur un guéridon (January 1920) (OPP.20:141). Finally, works with the greatest similarity to Red Guitar are the two variations, both dated only a day later, namely, Guitare et compotier (10 April 1921) (OPP.21:014; Z.VI:1426) and The Table (10 April 1921) (OPP.21:257; Z.IV:260), a watercolor/oil on canvas and a crayon on cardboard, respectively, with a high level of similarity between Red Guitar and Guitare et compotier (‘Blue’).
The immediate question concerns why such a close duplicate of the documented painting exists. During the 1920s, Picasso had a habit of reproducing paintings in his various styles. The most obvious example is the cubistic painting, Verre et paquet de tabac-Fall/1921, which exists in two identical versions and one modestly modified version (OPP.21:109, OPP.21:268, and OPP.21:077; from the Online Picasso Project: https://picasso.shsu.edu/ (accessed on 1 January 2022)). The next example is The Portrait d’Olga (Spring 1921), cardboard (OPP: 21:050). The ink on paper version of this artwork has been recently found (OPP.20:492). Picasso kept this 1920s version in his collection until the 1970s; however, it was taken from his house by Pierre Le Guennec, who was also in possession of never before seen Picasso’ works from 1900 to 1932. These two examples indicate that the existence of multiple versions of guitar paintings resonates closely with the artist’s tendency to recreate his works in the early 1920s. Additional examples of duplicate works include Pierrot et arlequin (Summer 1920) (OPP.20:010 and OPP.20:272), Autoportrait de profil (March 1921) (OPP.21:082 and OPP.21:209), Chien-loup (7 March 1921) (OPP.21:074 and OPP.21:266), Femme assise ((Spring) Summer 1921) (OPP.21:230 and OPP.21:231), Trois femmes à la fontaine (Summer 1921) (OPP.21:224 and OPP.21:225) are very similar, whereas OPP.21:187 and OPP.21:227 are modestly different from the rest.

2. Materials and Methods

2.1. Digital X-ray Radiography

To allow the best sharpness and geometric detail, the X-ray radiography scans were performed using a portable digital X-ray detector (Siemens Mobilett) with a wide dynamic range. This technique has an advantage over the use of traditional X-ray radiographic films that have limited range, and thus, reduced image contrast. The digital X-ray images are formed on the principle that a low-density area absorbs fewer X-ray photons; therefore, they produce darker shades, whereas more dense areas absorb more photons and produce lighter shades on the images. After an optimal exposure of 200 s, raw images were processed for display and storage using Digital Image Management System (DIMS) with image characteristics optimized without losing the initial quality.

2.2. Fourier Transform Infrared Spectroscopy

Fourier transform infrared spectroscopy (FTIR) utilizes infrared radiation to specifically detect the presence of organic, polymeric, and inorganic compounds. The sample analyses were performed on the Thermo Scientific Nicolet 6700 FTIR spectrometer.
We analyzed the paint samples without preparation. From the leaked paint, and from the back area after the varnish and paint layer were separated, we prepared a minimum of 100 mg of sample to cover a 3 mm circle area and to produce multiple measurements. FTIR spectra were measured in two ranges: 4000 cm–1–225 cm–1 with DLaTGS detector and CIS beam splitter. The spectra resolution was 4 cm–1 and the number of scans was 256. Thermo Electron’s OMNIC software was used to collect and process the data.

2.3. Organic Elemental Analysis

Organic elemental analysis was conducted with a Thermo Fisher Flash 2000 analyzer. This technique simultaneously detects C, H, N, and S in organic materials by pyrolysis, separation by gas chromatography, and detection with a thermal conductivity detector (TCD). The samples were calibrated against the thermal standard, Acetanilide (C:71.09, H:6.71, N:10.36 and O:11.84%).

2.4. X-ray Fluorescence

A portable Bruker Tracer III SD X-ray fluorescence spectrometer (XRF, S/N: T3S2416, manufacturing date: 19 December 2012, Bruker, Billerica, MA, USA) was used to investigate the elemental composition of the pigments. The spectrometer is equipped with a micro-X-ray tube with a Rhodium anode, and Silicon Drift Detector (SDD). The beam spot size was 10 mm in diameter and the distance of the spectrometer to the painting was 1 mm. The instrument was used in a stationary mode on a tripod. It was operated at 45 keV and 30 µA using a filter (Ti, Al, Cu) without vacuum, with a 120 s exposure time to maximize the signal to noise ratio.

2.5. Raman Spectroscopy

Raman spectra were recorded with a LabRAM HR 800 Raman spectrometer equipped with an integrated Macrochamber and Confocal Microscope for high spatial resolution imaging and a benchtop spectrometer that has a 633 nm He:Ne 20 mW laser, as well as externally mounted lasers at 785 nm (NIR) and 532nm Nd:YAG. Measurements were performed using the 785 nm laser, and depending on the sample, the maximum power was adjusted using neutral density filters (0.01% to 100%). The minimum wavenumber for the 785 nm laser was 100 cm−1 and the maximum wavenumber was 3300 cm−1 with a resolution of 4 cm−1. The recorded spectra had varying acquisition parameters with the time and power determined on a sample-by-sample basis. Spectra were corrected for instrument response, and when a strong luminescent background was present, the spectra were not taken into consideration.

3. Results and Discussion

3.1. Visual and X-ray Analysis of the Red Guitar Painting

The initial indication that the Red Guitar is not a copy of Guitare et compotier (‘Blue’) was obtained by a visual comparison of the two paintings (Figure 3a,b). There was a difference in the way the contour of the bottle is conveyed in the two versions. In the Red Guitar composition, the edge of the vessel is not defined. Instead, the shoulder-like top of the green vessel was rounded and narrowed, and a thick layer of black color indicated this change (Figure 3a,b). In the ‘Blue’ composition (see Guitare et compotier OPP.21:014 in “Online Picasso Project”: https://picasso.shsu.edu/ (accessed on 1 January 2022).), the righthand edge of the vessel is defined, despite the fact that it falls over the stylized, jagged area of the bottle behind it. In addition, two vessel-like drawings, clearly visible in Red Guitar (inside the white rectangle in Figure 3a), were painted over into a jagged black shadow.
Further evidence concerning artistic changes arose from the X-ray image analysis of the Red Guitar (Supplementary Figure S5 and Figure 3c). The bottom area of the X-ray image did not show significant compositional changes (Supplementary Figure S5). The top area (Figure 3c) revealed drawings that are absent or modified in the final composition: (1) a big rectangle right and above the head; (2) the area inside the head that is redrawn several times; (3) the black jagged shadow that was initially curved; (4) the small blue rectangle was initially a circle with extra lines around it; and (5) additional, less defined drawings, below the signature and above the bottle, were hidden in the final work. Those artistic changes demonstrate that Red Guitar is an original artwork, not a copy of the ‘Blue’ painting.
To strengthen this conclusion, and provide further evidence, a detailed pigment analysis of Red Guitar was performed and compared with the pigments used by Picasso in other artworks. Pigment analysis was obtained using non-invasive X-ray fluorescence (XRF) spectroscopy, and microanalysis of color samples using FTIR and Raman spectroscopy. The results of the pigment analysis, along with their elemental compositions and chemical compounds, is summarized in Table A1.

3.2. White Paint

A strong XRF signal of Zn in the main composition, and all other areas, identified zinc oxide (zinc white) as the primary white pigment (Figure 4a). The other white pigments are based on Ca and Pb, in the form of calcium carbonate (calcite) and lead carbonate (lead white). This composition suggested the use of artistic tube paint, not the enamel paint (Ripolin), which is a high-quality zinc white that does not contain Ca and Pb components [9]. Traces of Co and Fe/Mn were also present, and they are known to be impurities of Zinc whites. Sr is a known impurity of calcium carbonate [10]. Traces of Ti could be from the cardboard support, but it could also derive from a natural impurity of white pigments, as is proposed for Picasso’s cubist painting, Man with a Fruit Bowl (1917) [4].
The ground layer pigments were analyzed from the leaked paint using FTIR and Raman spectroscopy. Calcium carbonate was identified from the characteristic infrared signals at 1379 cm−1, 872 cm−1, and 712 cm−1 (Figure 4b), and the Raman signals at 1088 cm−1, 714 cm−1, 284 cm−1, and 158 cm−1. (Figure 4c). Traces of iron oxide (burnt sienna) were identified at 116 cm−1 (Figure 4c) using the refence spectra from Pigments Checker [10]. Other pigments were not detected in the leaked white paint. Although the FTIR spectrum of the leaked white paint was overwhelmed with calcium carbonate (Figure 4b), it additionally showed signals of an oleo resinous binder, with characteristic sharp peaks at 2850 cm−1 and 2919 cm−1, and a shoulder at 2954 cm−1. These signals were completely identical to a resinous binder from Acrobat family (1906) [3]. The binder spectra also include the oil carbonyl group at 1730 cm−1, and carboxylate bands at 1578 cm−1, 1584 cm−1, and 1535 cm−1, from metal soaps formed from the reaction of zinc white and oil components (Figure 4b).

3.3. Black Paint

The XRF elemental analysis of the main black composition revealed the presence of Zn, Ca, Ba, Fe, and Sr (Figure 4d). The signals for Ca and Sr were stronger in the black paint than in the white paint or any other paint. A very strong signal for Ca in the black pigment is usually attributed to bone (ivory) black. The strong appearance of Sr indicates more evidence for bone black, as Sr is a common component of this type of pigment [10]. The strong signal for Fe could be from an additional black pigment, black iron oxide (magnetite), but also from Prussian blue (a cyanide–iron complex that was a widespread pigment at that time). The mix of bone black and Prussian blue was also detected in Acrobat family (1906) and in Picasso’s early paintings [1,2,3]. Traces of Cu in the black paint could be from adjacent blue paint or from the support.
The strong presence of Ba is specific to this paint. Barium was not detected in any other color, and was perhaps used as a filler in this black pigment. The lines, signature, date, and dark areas in other colors did not contain Ba, and consequently, were executed with a different black paint. The analysis in Figure 3 documented that the jagged black composition does not completely match the initial drawing. Use of only one type of black pigment in this area means that the compositional changes were made during the process of art creation. If something had been subsequently changed, the XRF analysis could indicate differences in paint composition in the altered and original areas [10]; however, due to a limited depth of XRF detection, it is not entirely impossible that the subsurface layer consists of a different composition from the surface layer.

3.4. Blue Paint

The XRF elemental analysis identified three blue pigments, based on Co, Cu, and Fe, in a mixture with large amounts of Ca and Zn whites, and traces of Ti, Pb, and Sr in the white paint (Figure 5a). The blue area appears fragmented and discolored; however, a uniform blue color under UV light (Supplementary Figure S6) indicated that a fluorescent blue pigment was in the mixture. The pigments of Fe (Prussian blue) and Cu (copper carbonate), do not absorb and appear black under the UV light [11]. Cobalt pigments are dark blue under UV light, but the amount of Co seems insufficient to be the sole absorbent.
The only remaining pigment that is blue under the UV light is ultramarine [11]. The faded blue appearance may suggest “ultramarine disease”, a catalytic process of the degradation of the oil binder when ultramarine acts as a catalyst of degradation [12]. Ultramarine was clearly detected in the dark yellow paint by its specific Raman absorption at 550 cm−1 (Figure 6a). In Acrobat family (1906), Picasso mixed Prussian blue and ultramarine to produce different blue shades, and similar patches of faded blue paints were identified [3]. Additional analyzes of the blue, as well as black, orange-red, and light green paints, were not performed as sampling would be destructive to the painting.

3.5. Yellow Paints

There are two different shades of yellow, bright yellow and dark yellow. They are both dominated by pigments based on Fe, Pb, and Cr. Ca and Zn whites, and their accompanying trace elements, were present in both shades (Figure 5c,d). The presence of Pb and Cr is from lead chromate (PbCrO4, chrome yellow), frequently used in earlier works by Picasso, but also by van Gogh and other 19th and early 20th century artists.
Figure 6a,b shows images and Raman spectra taken from the leaked dark yellow paint. The pigment of Fe is iron oxide–hydroxide (FeOOH, yellow ochre), with characteristic Raman bands at 246 cm−1, 301 cm−1, 390 cm−1, 480 cm−1, and 550 cm−1 (Figure 6a). Yellow ochre, however, has a weak signal at 550 cm−1; therefore, the observed strong signal at 550 cm−1 is attributed to the presence of ultramarine. This suggested that the green shades on the dark yellow surface were probably created by a combination of yellow ochre and ultramarine. Chrome yellow (lead chromate, PbCrO4) was identified with a characteristic signal for CrO4 at 844 cm−1 and additional signals at 389 cm−1, 362 cm−1, 337 cm−1, and 325 cm−1 (Figure 6b). The leaked paint also contained calcium carbonate, and the green pigment from the neighboring dark-green paint, is described in Figure 7.

3.6. Green Paint

There are two different green colors in Red Guitar. The analysis of the centrally located light green identified Fe, Pb, and Cr, indicating that the light green color was created by mixing Prussian blue and chrome yellow (Figure 5b). This combination of blue and yellow pigments to obtain green shades was common procedure of that time, and it was used in Picasso’s early works [1]. A combination of Prussian blue with a yellow pigment was also used in one of the green colors in Acrobat family (1906) [3].
The bottom dark green elements are the same as those in the light green, and they include Fe, Pb, and Cr pigments (Figure 7a); however, the overall look of the paint suggested that the area was repainted. The green stains were also evident in the adjacent colors, and especially noticeable in the neighboring dark red area and on the arts’ date (Figure 2c), demonstrating that this intervention was performed after the painting was completed.
The green pigment identified in the leaked green paint is copper phthalocyanine (PG7, phthalo green). This pigment was discovered in the blue version (phthalo blue) in 1935, and the green version was available from 1938. This is the only pigment that deviated from the date that the painting was made. The appearance of the green stains over the date strongly attests that this was a later intervention. It is not known, however, who performed this intervention. Interestingly, the green color issues were present in Picasso’s work. In a collage from 1913–14, a green area was retouched multiple times, most likely by the artist himself [13], and the preparation of the second green color in Acrobat family was suspected to include a yellow dye, but this was not confirmed [3]. Phthalo blue was used in the restoration of La famille Soler (1912) [14], and one green pigment used in the Antibes collection was not identified, but presumably it was not phthalo green [15].

3.7. Red Paints

Three red colors were characterized: bright red, orange-red, and brown-red. In addition to Zn and Ca white pigments, the bright red has Fe, Pb, and traces of Cr (Figure 8a).
As with other samples taken from the leaked paint, FTIR and Raman spectroscopy only identified calcium carbonate as a white pigment. Raman spectroscopy (Figure 8b) identified Pigment Red 4 (PR4), an early synthetic mono-azo colorant, also called Chlorinated Para Red, among other names. The reference spectrum (CI 12085) of PR4 from the SOP spectral library (https://soprano.kikirpa.be/index.php?lib=sop&id=PR4_B_785_kikirpa (accessed on 1 January 2022)) is completely identical to red paint sample 1. The red paint samples, 2 and 3 (Figure 8b,d), show additional signals for iron oxide (red ochre), with specific bands at 226 cm−1, 291 cm−1, and 410 cm−1, that were clearly separated from the PR4 signals. Thus, it can be concluded that PR4 and red ochre were the main pigments used in the preparation of the bright red color.
The centrally located orange-red differed from other red colors due to the dominance of Pb, with some amounts of Cr- and Fe-based pigments. (Figure 9a). The amounts of Zn and Ca white pigments was lowest in this color. Due to the dominant presence of Pb, the main pigment in this color is probably red (orange) lead (Pb3O4, minium). A significant amount of Cr, however, also indicated the presence of chrome yellow. An orange pigment, which is a combination of lead carbonate, lead chromate, and lead oxide (Pb (OH)2 PbCrO4 + PbO.PbCrO4) that has been used in France since the middle of the 19th century [16], cannot be excluded. The Fe based pigment could not be assigned, and it could be red ochre or yellow ochre, as identified in other areas of the painting.
The brown-red color in the lower part of the painting was the least complex. This color contained Ca and Zn based white pigments, and the red pigment was exclusively Fe-based. Detection of traces of Mn showed that the Fe pigment is a natural earth pigment (Figure 9b). Raman analysis identified ferric oxide in the form of hematite, with signals at 225 cm−1, 292 cm−1, 409 cm−1, 495 cm−1, 609 cm−1, and 1302 cm−1 (Figure 9c). Additional signals in 1300–1600 cm−1 range suggested the presence of a black pigment. The presence of the bone black pigment was excluded due to the absence of the signal for the phosphate group at 960 cm−1.

3.8. Cardboard Analysis

Infrared spectra detected hemicellulose and lignin, which remain after cellulose production from wood pulp [17].
As shown in Figure 10a, the art support was made of a highly lignified cellulose with characteristic FTIR signals for hemicellulose at 1726 cm−1 and 1594 cm−1, and lignin at 1506 cm−1 and 1252 cm−1. Similar signals for hemicellulose and lignin are present in the spectra of French paper cellulose, but not in the German paper cellulose produced in the late 19th and early 20th century (Figure 10b,c) [18]. German paper was produced from a greater amount of bleached cellulose, a technological process that removes hemicellulose and lignin [18]. In the cellulose spectrum for Acrobat family (1906), painted on cardboard, the hemicellulose signal was masked by fillers; however, the signal for lignin is strong (Figure 10c), which agrees with other analyses [3]. On the other hand, the cellulose spectra of early 20th century cardboard from Italy (Figure 3b) did not have a significant amount of lignin, thus showing a different production process [19].
The main inorganic components in the Red Guitar cardboard include Zn, Fe (Mn), Cu, and Ca, with traces of Sr, Pb, and Cr (Figure 10d). Unexpectedly high amounts of Zn, with traces of Pb and Cr, were attributed to pigment migration into the cardboard. The European papers from the late 19th and early 20th centuries similarly have Ca-, Fe- (Mn), and Cu-based materials, with Zn and Ti varying in level, depending on the country of origin and date of manufacturing [18]. The Acrobat family cardboard also contains Zn, Ca, and Fe (Mn) as the main elements, whereas Cu and Sr are present as trace elements [3]. Thus, the old European papers and cardboards shared industrial ingredients, but highly lignified cellulose was common for the old French papers, Picasso’s Acrobat family cardboard, as well as the cardboard from the Red Guitar support, was built.

3.9. Composition of the Back Area

The smooth and shiny surface of an ivory color without visible traces of paintbrush of the artwork’s back, is reminiscent of the industrial house paint, Ripolin (Figure 11a). This type of house paint was popular among artists because of its good quality and lower price compared with artists’ paints. They were used by Picasso, Kandinsky, and Picabia; this is widely documented. Contrary to visual expectations, the infrared spectra identified nitrocellulose as the major component in the varnish and paint layers.

3.9.1. Varnish Layer

Nitrocellulose is identified by characteristic FTIR signals for the NO2 group at 1641 cm−1 and 1275 cm−1, CO group at 1062 cm−1, and the O-NO2 group at 834 cm−1 (Figure 11b). The spectrum has additional signals for the carbonyl group at 1720 cm−1, aliphatic deformations at 1455 cm−1 and 1372 cm−1, sharp bands for CH2 at 2920 cm−1, and CH at 2851cm−1, which together, completely correspond to the natural resin copal [11]. The results cannot show whether nitrocellulose was mixed with the resin at the time of application or whether it has already been applied; however, there is evidence that Picasso used copal, most likely Siccatif de Harlem, and mixed it with oil paints, as evident in Still life (1922) [2].
Raman spectra of the varnish layer (Figure 12a) identified the nitrocellulose plasticizer camphor with characteristic bands at 1728 cm−1, 1451 cm−1, and 852 cm−1, and diethyl phthalate with bands at 1600 cm−1, 1581 cm−1, 1167 cm−1, and 1042 cm−1. Importantly, this chemical composition fully corresponds to a type of nitrocellulose known as “French ivory” that was widely used at the end of 19th and early 20th century [16]. Nitrocellulose degradation is an autocatalytic process caused by aging. In nitrocellulose artifacts from the 19th and early 20th century, in France and England, the percentage of nitrogen in degraded artifacts is 3–6%, and in undegraded artifacts it is 6.7–9.5%, with a maximum of 11%, which corresponds to the presence of two nitrate groups in nitrocellulose [20]. Nitrocellulose from the back varnish layer of Red Guitar contained 3.24% of nitrogen, showing that it had seriously degraded.

3.9.2. Back Paint Layer

The elemental analysis of the back paint established Ti, Zn, and Fe as major elements, with minor contributions from the Ca and Pb white pigments, and traces of Mn, Cu and Sr. Although Ti, Fe (Mn), and Cu did not vary, Zn, Pb, Ca, and Sr varied in different areas of the painting (Figure 11c). This could suggest that Pb, Ca and Sr were part of the zinc white paint that was applied separately from titanium white (with Fe/Mn and Cu impurities). Raman analysis of the lower layer identified titanium dioxide in two crystalline forms, as anatase with characteristic band at 145 cm−1 and rutile with bands at 234 cm−1, 441 cm−1, and 609 cm−1 (Figure 12b). The spectra also detected traces of nitrocellulose (1280 cm−1, 923 cm−1, and 849 cm−1), camphor (1731 cm−1 and 1453 cm−1) and diethyl phthalate (1598 cm−1, 1580 cm−1, 1166 cm−1, and 1041 cm−1) in the paint particles. Lead white and calcium carbonate were not detected, which indicates that their amount was low, or that they were mainly in the upper layer. The main Raman signal for zinc white at 436 cm−1 overlaps with rutile and it cannot be identified; therefore, it could not be determined if zinc white is present in the lower layer, varnish, or in both layers.

4. Significance and Conclusions

We did not anticipate that examining this work would become a historical journey into Olga Picasso’s past. The autobiography of Aleksandar Derocco led to information about Olga’s brother, who lived in Belgrade in the 1920s. When Nikolaj’s relative was found in Belgrade, the circumstances related to this work and why the painting was in Serbia became more obvious. The historical materials from the Nikolaj Kokhlov collection and the correspondence with Olga and their broader family, are of additional importance for future research on Olga Picasso. Her biography lacks information about her brother, Nikolaj, that can now be amended.
Without the information from Dr. Mullen’s ‘Online Picasso Project’, it would be impossible to systematically study the artwork produced during the first two decades of the 20th century. This database was central for the identification of paintings that were stylistically not only from the same period, but also showing other duplicate works and two versions of Red Guitar. The visual and X-ray analysis exposed the process of the creation of Red Guitar, strongly suggesting that Red Guitar is an original artwork and not a copy of Guitare et compotier.

4.1. Use of Picasso’s Palette of Historic Pigments

The palette of white, black, blue, yellow, green, red, and red-brown is the same color palette that Picasso used in his early work and maintained throughout his career. Red Guitar is dominated by zinc, lead, and iron-based pigments. Chrome yellow, yellow ochre, red ochre, hematite, ultramarine, and the colorant PR4, were identified by Raman spectroscopy. Zinc white, lead white, carbon black, Prussian blue, and red lead (minium) were strongly implicated in the elemental analysis. According to this color palette, the painting could be dated to the beginning of the 20th century.
Zinc white is the primary white pigment in Red Guitar, and in addition to the presence in the ground layer, it has been applied alone or in a mixture with calcium carbonate in the surface layer, resulting in a ‘milky effect’ around the main composition. Zinc white appears to be the primary white pigment in several paintings from the Picasso’s Blue period—Blue room (1901), Poenies (1901), The Child with a Dove (1901) [2]—and it was used in the ground layer of Man with a Fruit Bowl (1917) [4]. Another application involved mixing a high amount of zinc white with other colors, which was highly visible in blue and yellow areas; again, this is consistent with the color preparation documented for the Blue room [2] where zinc white was the basis for the preparation of blue and other colors. The surface application of zinc white appears in some early Picasso’s portraits, and Sessa and colleagues concluded that the same trend was also recognized for Van Gogh paintings [1].
Red guitar has at least one black pigment identified as bone (ivory) black. Bone black was more frequently used in earlier Picasso paintings, including early portraits [1] such as Blue room [2] and Science and charity [21]. Chrome yellow was popular in the 19th century, but because it darkens upon exposure to light, it has been replaced with cadmium yellow. Although Picasso predominantly used cadmium yellow in his later works [6,22,23], the exclusive use of chrome yellow was documented in his earlier paintings [1,2,21]. Iron-containing earth pigments, such as red, yellow, and brown iron oxides, have been used since antiquity and they were constantly present in Picasso’s palette.

4.2. Use of Industrial, Non-Artist Materials

Picasso was innovative in working with different supports, binders, and paints during his career [4,5,6,7,8]. This work clearly identified the same tendencies in the use of industrial products, the red colorant PR4, and nitrocellulose. Natural and industrial organic dyes (printing colors, textile colors) were known to be used in early 20th century paintings, but they are not well-characterized, even in Picasso’s work. They are difficult to analyze, and most of the time, they could only be implicated by the presence of metals, such as barium, which are used to precipitate those compounds into more durable ‘lakes’. It was also not uncommon for synthetic dyes to be added to artistic pigments, to change their color hue, or because they were much cheaper alternatives. In the early 20th century, Winsor and Newton, for example, used Pigment Red 3 (PR3) in artists’ paints, which was not officially acknowledged [24]. The earliest record of the use of PR3 in paintings is by German author, Kirschner (1909), and in the paintings on glass by C. Menze (1913) [25]. Here, the painting investigated is the first known case in which PR4 was identified in an artistic work, which is a significant discovery that is important for the future classification of the historic use of PR4. PR4 was synthesized in 1906 [25] by coupling beta-naphthol with 2-chloro-4-nitroaniline. PR4 had major applications as industrial paint, but its use in artists’ colors, pencils, and wax crayons is also known [26].
Nitrocellulose is the oldest semi-plastic material that became very popular in the early 20th century. The oldest product was collodion, found in 1848, which was used for dressing wounds and preparing photographic plates. Wider applications of nitrocellulose began with the use of camphor as a plasticizer and the invention of celluloid in 1872. Celluloid found very wide applications in household products, coatings, varnishes, and in the production of the first celluloid tapes for the film industry. The ivory-like French fans from the late 19th and early 20th centuries were made from celluloid, with camphor as a plasticizer and diethyl phthalate as an additive to increase the strength of the material [16]. This type of celluloid was widely known as “French ivory” because of the specific type of celluloid production developed in France, and it was applied as a cheap substitute for real ivory in home and decorative products [16]. Zinc oxide was the main pigment used to produce the ivory look of the nitrocellulose products [27]. This historical description and the identification of identical nitrocellulose components, camphor and diethyl phthalate, was quite remarkable, and demonstrated that the Red Guitar nitrocellulose resembles the “French Ivory” celluloid.
Titanium dioxide was manufactured in the form of anatase until the 1940s, when pure rutile was introduced [28]. Ilmenite ore, which is an ore of iron and titanium (Fe, Mg, Mn, Ti) O3, was used to produce anatase. Before 1920, anatase was impure and of an ivory color since it contained a large percentage of rutile and iron due to inadequate industrial processing [11]. In 1920, France implemented a new procedure known as “the Blumenfeld process” for the manufacturing of pure anatase, but the purity remained problematic [11]. The very high presence of rutile and Fe/Mn in a mixture with zinc white in the back paint of Red Guitar implicated an early product, dated between 1908 and 1920, when many attempts were made to produce purer anatase [11].
In the 1920s and 1940s, most titanium white products were composites of 15–30% titanium dioxide with 85–70% barium sulphate or calcium sulphate/phosphate [28]. Sulphates and phosphates were not detected in Red Guitar, thus showing that the titanium white used was not a composite. The ratio between Zn and Ti in the Red Guitar back paint is close to an early product that was on the market before 1920, known as Kronox Extra A, which had 30% zinc white and 70% Kronox Extra X. Kronox Extra X had low amounts of CaPO4 (25%) and BaSO4 (10%) [28] that could have escaped detection.
Based on the paint leaks, Red Guitar was completed after the paint on the back of the artwork was probed. Perhaps the idea to combine nitrocellulose with copal varnish and titanium dioxide/zinc white was an artistic experiment to test the properties of nitrocellulose, or maybe it was an attempt to produce an ivory glossy paint that would mimic Ripolin. There is plenty of evidence that Picasso was experimenting with various materials, and that he combined traditional oil paints with copal to make them look similar to Ripolin [6,23,29]. This work documents the earliest known use of nitrocellulose in art, famously used by David A. Siqueiros in the early 1930s [30] and Sidney Nolan in the late 1940s [31].
To conclude, the provenance of the artwork contains convincing, circumstantial, evidence to show how the artwork was gifted to the Hadzic family in appreciation for the help given to Olga Picasso’s brother, which is how it found its home in Serbia. The artistic analysis offered strong support for the use of a color palette that is in accordance with Picasso’s artistic preferences. The scientific analyses of the paints, and supporting materials, provided validation of the use of historic pigments and materials, which is in accordance with the period of the late 19th and early 20th century, and France as the place where the art had been produced; therefore, the multidisciplinary approach to the analysis of Red Guitar gathered sufficient evidence and arguments to provide proof that this painting is likely to be Picasso’s original.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/heritage5020060/s1. Figure S1: (a) The archival documents from the prominent culture journals “Letopis Matice Srpske” and “Srpski Narodni Glasnik” published during the 1926 art season in Belgrade, Yugoslavia. (b) A flyer of the art exhibition organized by Paul Rosenberg and Paul Epstein in October 1926, which exhibited the work of Picasso. Figure S2: The historic information about the first art owner. Figure S3: In his autobiography book “When the plane was flying above Belgrade” (published in 1987 by Narodna Knjiga, Yugoslavia) on pages 178–179, Derocco describes Rue de la Boetle, similarly to what he said in the text ‘At Picasso’ in Figure 3. Figure S4: (a) A gift from Picasso to Sacha (Alexander) Derocco, in a memory of their friendship, December 1927; (b) One of Picasso’s letter to Derocco written in Serbian, the exact date is not visible. Figure S5: The X-ray image of Red Guitar. More details are in the main text. Figure S6: The UV fluorescence image of Red Guitar.

Author Contributions

Writing—M.B.; review and editing, A.P.-M.; data collection and curation, M.B., S.K., S.M. and A.P.-M. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Additional data is available upon request.

Acknowledgments

This work would not have been possible without help from Leonid S. Brown and David Mutch in translating Olga Picasso’s letters and Peter Tremaine and Grzegorz Szymanski for the expertise and use of their Raman spectroscopy facility at the Electrochemical Technology Centre, the University of Guelph. We also like to thank the ANALEST-Department of Chemistry, the University of Toronto, for performing FTIR and organic elemental analysis, and the Ontario Veterinary College, the University of Guelph, for performing X-ray analysis.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

Table A1. Red Guitar: Colors along with the elements and components identified in the painting.
Table A1. Red Guitar: Colors along with the elements and components identified in the painting.
ColorElements aPainting Materials b
WhiteZn, Ca, Fe (Mn), Co, Ti, Pb, SrZinc white; Calcite
BlackZn, Ca, Ba, Fe, Sr, Cu, PbBone (ivory) black, Fe-pigments, Barium sulfate
BlueZn, Ca, Co, Cu, Fe (Mn), Pb, Ti, SrZinc white, calcite, Co and Cu blue pigments
Yellow:
Yellow I (bright)Pb, Cr, Fe, Zn, Ca, Ti, SrChrome yellow; Yellow ochre, Zinc white, Calcite
Yellow II (dark)Pb, Cr, Fe, Zn, Ca, Ti, SrChrome yellow *; Yellow ochre, *
Zinc white, Calcite *, Ultramarine *
Green:
Green I (bright)Fe, Pb, Cr, Zn, Ca, Cu, Ti, SrPrussian blue, Chrome yellow, Zinc white, Calcite
Green II (dark)Fe, Pb, Cr, Zn, Ca, Cu, Ti, SrChrome yellow, Yellow ochre, Zinc white, Calcite #*
Drying oil #, Phthalocyanine green (retouch) *
Red:
Red I (bright)Fe, Pb, Cr, Zn, Ca, TiRed Iron *, Chrome Yellow, Zinc white *, Calcite *
Red II (orange)Pb, Cr, Zn, Ca, Fe, Cu, Ti, SrRed Lead, Chrome yellow, Lead white
Zinc white, Chalk, Yellow ochre (traces)
Red III (brown-red)Fe (Mn), Zn, Ca, Pb, Ti, SrIron oxide (hematite)*, Zinc white, Calcite #*, Drying oil
Preparation layer: Calcite #*, Drying oil #
Back:Ti, Zn, Fe (Mn), Ca, Pb, Cu SrTitanium white, Zinc white, Iron oxide
Varnish Nitrocellulose #*, Copal # Camphor *, Diethyl phthalate *
Ivory white Titanium dioxide (anatase and rutile) *
Nitrocellulose #*
Support:Zn, Ca, Fe (Mn), Cu, Cr, Ti, Sr, PbCellulose # (hemicellulose, lignin)#
a XRF: the main elements (Bold); the minor elements (Italics); b FTIR spectroscopy (#) and Raman spectroscopy (*).

References

  1. Sessa, C.; de Garnica, R.J.; Rosi, F.; Fontana, R.; Garcia, J.F. A study of Picasso’s painting materials and techniques in six of his early portraits. J. Am. Inst. Conserv. 2016, 55, 198–216. [Google Scholar] [CrossRef]
  2. Favero, P.A.; Mass, J.; Delaney, J.K.; Woll, A.R.; Hull, A.M.; Dooley, K.A.; Finnefrock, A.C. Reflectance imaging spectroscopy and synchrotron radiation X-ray fluorescence mapping used in a technical study of The Blue Room by Pablo Picasso. Herit. Sci. 2017, 5, 13. [Google Scholar] [CrossRef] [Green Version]
  3. Buti, D.; Pullano, M.; Papa, E.; Nygårds, E.; Ludvigsen, L.; Wadum, J. Picasso’s Acrobat Family in focus: An investigation of materials and techniques of an iconic work in the collection of the Gothenburg Museum of Art. SN Appl. Sci. 2020, 2, 1411. [Google Scholar] [CrossRef]
  4. Fuster-López, L.; Izzo, F.C.; Andersen, C.K.; Murray, A.; Vila, A.; Picollo, M.; Stefani, L.; Jiménez, R.; Aguado-Guardiola, E. Picasso’s 1917 paint materials and their influence on the condition of four paintings. SN Appl. Sci. 2020, 2, 2159. [Google Scholar] [CrossRef]
  5. Gautier, G.; Bezur, A.; Muir, K.; Casadio, F.; Fiedler, I. Chemical fingerprinting of ready-mixed house paints of relevance to artistic production in the first half of the twentieth century. Part I: Inorganic and organic pigments. App. Spectrosc. 2009, 63, 597–603. [Google Scholar] [CrossRef]
  6. Muir, K.; Langley, A.; Bezur, A.; Casadio, F.; Delaney, J.; Gautier, G. Scientifically investigating Picasso’s suspected use of ripolin house paints in Still life, 1922 and The Red armchair, 1931. J. Am. Inst. Conserv. 2013, 52, 156–172. [Google Scholar] [CrossRef]
  7. Standeven, H.A.L. Oil-Based House Paints from 1900 to 1960: An Examination of Their History and Development, with Particular Reference to Ripolin Enamels. J. Am. Inst. Conserv. 2013, 52, 127–139. [Google Scholar] [CrossRef]
  8. Langley, A.; Muir, K.; Sutherland, K. Scenes from the life of Picasso’s Still Life (1922): History, materials, and conservation. SN Appl. Sci. 2020, 2, 1384. [Google Scholar] [CrossRef]
  9. Casadio, F.; Rose, V. High-resolution fluorescence mapping of impurities in historical zinc oxide pigments: Hard X-ray nanoprobe applications to the paints of Pablo Picasso. Appl. Phys. A 2013, 111, 1–8. [Google Scholar] [CrossRef]
  10. Larsen, R.; Coluzzi, N.; Cosentino, A. Free XRF spectroscopy database of pigments Checker. Int. J. Conserv. Sci. 2016, 7, 659–669. [Google Scholar]
  11. Conservation and Art Materials Encyclopedia. Available online: http://cameo.mfa.org/wiki/Main_Page (accessed on 10 May 2022).
  12. Schnetz, K.; Gambardella, A.A.; van Elsas, R.; Rosier, J.; Steenwinkel, E.E.; Wallert, A.; Iedema, P.D.; Keune, K. Evidence for the catalytic properties of ultramarine pigment. J. Cult. Her. 2020, 45, 25–32. [Google Scholar] [CrossRef]
  13. Townsend, J.H.; Fox, C.; Sacher, B. A Picasso paper collage of 1913–14: Assessment of fragility and sensitivity to light. SN Appl. Sci. 2021, 3, 669. [Google Scholar] [CrossRef]
  14. Defeyt, C.; Walter, P.; Rousselière, H.; Vandenabeele, P.; Vekemans, B.; Samain, L.; Strivay, D. New insights on Picasso’s Blue period painting La famille Soler. Stud. Conserv. 2018, 63, 24–35. [Google Scholar] [CrossRef]
  15. Casadio, F.; Miliani, C.; Rosi, F.; Romani, A.; Anselmi, C.; Brunetti, B.; Sgamellotti, A.; Andral, J.-L.; Gautier, G. Scientific investigation of an important corpus of Picasso paintings in Antibes: New insights into technique, condition, and chronological sequence. J. Am. Inst. Conserv. 2013, 52, 184–204. [Google Scholar] [CrossRef]
  16. Paris, C.; Coupri, C. Fourier transform Raman spectroscopic study of the first cellulose-based artificial materials in heritage. J. Raman Spectrosc. 2005, 36, 77–82. [Google Scholar] [CrossRef]
  17. Phinichka, N.; Kaenthong, S. Regenerated cellulose from high alpha cellulose pulp of steam-exploded sugarcane bagasse. J. Mater. Res. Technol. 2018, 7, 55–65. [Google Scholar] [CrossRef]
  18. Doncea, S.M.; Ion, R.M.; Fierascui, R.C.; Bacalum, E.; Bunaciu, A.A.; Aboul-Enein, H.Y. Spectral methods for historical paper analysis: Composition and age approximation. Instrum. Sci. Technol. 2009, 38, 96–106. [Google Scholar] [CrossRef]
  19. Casoli, A.; Fornaciari, S. An analytical study on an early twentieth-century Italian photographs collection by means of microscopic and spectroscopic techniques. Microchem. J. 2014, 116, 24–30. [Google Scholar] [CrossRef]
  20. Quye, A.; Littlejohn, D.; Pethrick, R.A.; Stewart, R.A. Investigation of inherent degradation in cellulose nitrate museum artefacts. Polym. Degrad. Stab. 2011, 96, 1369–1376. [Google Scholar] [CrossRef]
  21. Jiménez, R. A Fresh look: Technical analysis and treatment of Picasso’s Science and Charity at the Museu Picasso. SN Appl. Sci. 2021, 3, 507. [Google Scholar] [CrossRef]
  22. Delaney, J.K.; Zeibel, J.G.; Thoury, M.; Littleton, R.; Palmer, M.; Morales, K.M.; de la Rie, E.R.; Hoenigswald, A. Visible and infrared imaging spectroscopy of Picasso’s Harlequin Musician: Mapping and identification of artist materials in situ. Appl. Spectrosc. 2010, 64, 584–594. [Google Scholar] [CrossRef] [PubMed]
  23. Hanspach-Bernal, E.; Bezur, A. Mixed media: An example of Pablo Picasso’s combination of non-artist’s paints with tube colors from the Menil collection. J. Am. Inst. Conserv. 2013, 52, 173–183. [Google Scholar] [CrossRef]
  24. Pozzi, F.; Lombardi, J.R.; Leona, M. Winsor & Newton original handbooks: A surface-enhanced Raman scattering (SERS) and Raman spectral database of dyes from modern watercolor pigments. Herit. Sci. 2013, 1, 23. [Google Scholar] [CrossRef] [Green Version]
  25. Steger, S.; Stege, H.; Bretz, S.; Hahn, O. A complementary spectroscopic approach for the non-invasive in-situ identification of synthetic organic pigments in modern reverse paintings on glass (1913–1946). J. Cult. Herit. 2019, 38, 20–28. [Google Scholar] [CrossRef]
  26. Berrie, B.H.; Lomax, S.Q. Azo pigments: Their history, synthesis, properties and use in artists’ materials. Stud. Hist. Art 1997, 57, 8–33. [Google Scholar]
  27. Neves, A.; Ramos, A.; Callapez, M.E.; Friedel, R.; Réfrégiers, M.; Thoury, M.; Melo, M.J. Novel markers to early detect degradation on cellulose nitrate-based heritage at the submicrometer level using synchrotron UV–VIS multispectral luminescence. Sci. Rep. 2021, 11, 20208. [Google Scholar] [CrossRef]
  28. Van Driel, B.A.; Berg, K.J.V.D.; Gerretzen, J.; Dik, J. The white of the 20th century: An explorative survey into Dutch modern art collections. Herit. Sci. 2018, 6, 16. [Google Scholar] [CrossRef] [Green Version]
  29. Muir, K.; Gautier, G.; Casadio, F.; Vila, A. Interdisciplinary investigation of early house paints: Picasso, Picabia and their “Ripolin” paintings. In Proceedings of the ICOM Committee for Conservation 16th Triennial Meeting, Lisbon, Portugal, 19–23 September 2011. [Google Scholar]
  30. McGlinchey, C.; Aviram, A.; Zetina, S.; Arroyo, E.; Sil, J.L.R.; Pesqueira, M.E.E.; David, A. Siqueiros: His modification of oil and cellulose nitrate-based paint and his advocacy for innovation, 1931–1949. J. Am. Inst. Conserv. 2013, 52, 278–289. [Google Scholar] [CrossRef]
  31. Dredge, P. Sidney Nolan’s adventures in paint—An analytical study of the artist’s use of commercial paints in the 1940s and ’50s. AICCM Bull. 2013, 34, 15–23. [Google Scholar] [CrossRef]
Figure 1. From the Khokhlov Private Collection: (a) Nikolaj Khokhlov, brother of Olga Picasso in Yugoslav Army uniform; (b) Nikolaj’s wedding photo; and (c) the wedding certificate date, Belgrade, 12 July 1928. The certificate states: “Nikolaj Khokhlov (East Orthodox, born 2 May 1897 in Uman, Kyiv) married Jozefina Kolacek (Catholic, born 9th of October 1902 in Belgrade).” The names of parents, witnesses, and the priest are given. Dates when they previously visited this (Catholic) church were 22, 26, and 29 May 1927. There is a comment stating that they were forgiven and free to marry even though they are from different religions. At the bottom, are the signature and date of when this copy of the certificate was produced, on 11 November 1939. The purpose of this copy of Nikolaj’s wedding certificate is not known. (d) A postcard from 1923 written by Picasso to Nikolaj Khokhlov, signed ‘Picasso’ in Cyrillic. (e) A postcard of Olga to her brother Nikolaj, dated 10 August 1930 (see the main text).
Figure 1. From the Khokhlov Private Collection: (a) Nikolaj Khokhlov, brother of Olga Picasso in Yugoslav Army uniform; (b) Nikolaj’s wedding photo; and (c) the wedding certificate date, Belgrade, 12 July 1928. The certificate states: “Nikolaj Khokhlov (East Orthodox, born 2 May 1897 in Uman, Kyiv) married Jozefina Kolacek (Catholic, born 9th of October 1902 in Belgrade).” The names of parents, witnesses, and the priest are given. Dates when they previously visited this (Catholic) church were 22, 26, and 29 May 1927. There is a comment stating that they were forgiven and free to marry even though they are from different religions. At the bottom, are the signature and date of when this copy of the certificate was produced, on 11 November 1939. The purpose of this copy of Nikolaj’s wedding certificate is not known. (d) A postcard from 1923 written by Picasso to Nikolaj Khokhlov, signed ‘Picasso’ in Cyrillic. (e) A postcard of Olga to her brother Nikolaj, dated 10 August 1930 (see the main text).
Heritage 05 00060 g001
Figure 2. (a) The Red Guitar painting investigated in this work. The enlarged (b) artist’s signature and (c) date.
Figure 2. (a) The Red Guitar painting investigated in this work. The enlarged (b) artist’s signature and (c) date.
Heritage 05 00060 g002
Figure 3. (a) The white dashed lines and the white rectangle point to the main visible changes in Red Guitare. (b) The same adjustments in black and white. (c) The X-ray image of the top area: yellow lines surround the drawing’s elements that are absent or modified in the final composition.
Figure 3. (a) The white dashed lines and the white rectangle point to the main visible changes in Red Guitare. (b) The same adjustments in black and white. (c) The X-ray image of the top area: yellow lines surround the drawing’s elements that are absent or modified in the final composition.
Heritage 05 00060 g003
Figure 4. The XRF elemental analysis of the main (a) white and (d) black areas. (b) FTIR and (c) Raman spectra identify calcium carbonate, with traces of an oleo resinous binder (FTIR) and iron oxide (burnt sienna) (Raman) in the leaked white paint.
Figure 4. The XRF elemental analysis of the main (a) white and (d) black areas. (b) FTIR and (c) Raman spectra identify calcium carbonate, with traces of an oleo resinous binder (FTIR) and iron oxide (burnt sienna) (Raman) in the leaked white paint.
Heritage 05 00060 g004
Figure 5. The XRF elemental analysis of the main (a) blue, (b) bright green, (c) bright yellow, and (d) dark yellow colors. The spectra shown are semi-logarithmic plots of counts vs. energy. The tick marks on the y-axis of the plot indicate that a log scale was used. Various spots were investigated to represent each color.
Figure 5. The XRF elemental analysis of the main (a) blue, (b) bright green, (c) bright yellow, and (d) dark yellow colors. The spectra shown are semi-logarithmic plots of counts vs. energy. The tick marks on the y-axis of the plot indicate that a log scale was used. Various spots were investigated to represent each color.
Heritage 05 00060 g005
Figure 6. (a) Confocal images and Raman spectra taken from the leaked dark yellow paint. (1) The scanned area and its spectrum and (2) the reference spectra for yellow ochre and (3) ultramarine. Characteristic peaks with red numbers are from yellow ochre, ultramarine, and chrome yellow. The peaks with black numbers are from calcium carbonate. (b) The scanned area, (1) its spectrum, and (2) the reference spectrum for chrome yellow. Characteristic peak positions in red are for chrome yellow and ultramarine, and black for calcium carbonate.
Figure 6. (a) Confocal images and Raman spectra taken from the leaked dark yellow paint. (1) The scanned area and its spectrum and (2) the reference spectra for yellow ochre and (3) ultramarine. Characteristic peaks with red numbers are from yellow ochre, ultramarine, and chrome yellow. The peaks with black numbers are from calcium carbonate. (b) The scanned area, (1) its spectrum, and (2) the reference spectrum for chrome yellow. Characteristic peak positions in red are for chrome yellow and ultramarine, and black for calcium carbonate.
Heritage 05 00060 g006
Figure 7. (a) The XRF elemental analysis of the main dark green area. (b) Confocal image and Raman spectra of the leaked green paint identify (2) copper phthalocyanine green and (3) calcium carbonate. The reference spectra of copper phthalocyanine (1).
Figure 7. (a) The XRF elemental analysis of the main dark green area. (b) Confocal image and Raman spectra of the leaked green paint identify (2) copper phthalocyanine green and (3) calcium carbonate. The reference spectra of copper phthalocyanine (1).
Heritage 05 00060 g007
Figure 8. (a) The XRF elemental analysis of the bright red area. (b) Raman spectra of the leaked red paints identify Pigment Red 4 (PR4) in three separate samples. The spectrum of sample 1 is identical to the reference spectrum (not shown) of PR4. The paint samples 2 and 3 show additional signals (labeled with (*)); (c) the representative confocal image of sample 3; (d) the focused area of spectrum 2 showing red ochre signals at 226 cm−1, 291 cm−1, and 410 cm−1.
Figure 8. (a) The XRF elemental analysis of the bright red area. (b) Raman spectra of the leaked red paints identify Pigment Red 4 (PR4) in three separate samples. The spectrum of sample 1 is identical to the reference spectrum (not shown) of PR4. The paint samples 2 and 3 show additional signals (labeled with (*)); (c) the representative confocal image of sample 3; (d) the focused area of spectrum 2 showing red ochre signals at 226 cm−1, 291 cm−1, and 410 cm−1.
Heritage 05 00060 g008
Figure 9. The XRF elemental analysis of (a) orange-red and (b) brown-red paints. (c) Raman spectra (samples 1, 2, and 3) of the leaked paint from the brown-red area identify hematite (225 cm−1, 292 cm−1, 409 cm−1, 495 cm−1, 609 cm−1, and 1302 cm−1); the inset confocal image is for Raman sample 1.
Figure 9. The XRF elemental analysis of (a) orange-red and (b) brown-red paints. (c) Raman spectra (samples 1, 2, and 3) of the leaked paint from the brown-red area identify hematite (225 cm−1, 292 cm−1, 409 cm−1, 495 cm−1, 609 cm−1, and 1302 cm−1); the inset confocal image is for Raman sample 1.
Heritage 05 00060 g009
Figure 10. (a) The FTIR ‘footprint’ region of the Red Guitar cardboard cellulose and pure cellulose. The cardboard bands for hemicellulose (1726 cm−1 and 15,524 cm−1) and lignin (1506 cm−1 and 1252 cm−1) are not part of pure cellulose; (b,c) the FTIR ‘footprint’ regions of the old paper and cardboard cellulose according to References [3,18,19]; (d) the XRF analysis of the Red Guitar cardboard.
Figure 10. (a) The FTIR ‘footprint’ region of the Red Guitar cardboard cellulose and pure cellulose. The cardboard bands for hemicellulose (1726 cm−1 and 15,524 cm−1) and lignin (1506 cm−1 and 1252 cm−1) are not part of pure cellulose; (b,c) the FTIR ‘footprint’ regions of the old paper and cardboard cellulose according to References [3,18,19]; (d) the XRF analysis of the Red Guitar cardboard.
Heritage 05 00060 g010
Figure 11. Characterization of the paint on the back of the artwork’s. (a) Visual image; (b) the FTIR spectrum of the upper layer (black) identifies nitrocellulose and copal; the copal and nitrocellulose reference spectra are red [11]; (c) the XRF analysis of the back’s paint was obtained in several different areas; a high level of Zn at one spot is from a white stain on the surface.
Figure 11. Characterization of the paint on the back of the artwork’s. (a) Visual image; (b) the FTIR spectrum of the upper layer (black) identifies nitrocellulose and copal; the copal and nitrocellulose reference spectra are red [11]; (c) the XRF analysis of the back’s paint was obtained in several different areas; a high level of Zn at one spot is from a white stain on the surface.
Heritage 05 00060 g011
Figure 12. Raman spectra of the paint on the back of the artwork. (a) Varnish layer contains nitrocellulose (black), camphor (green) and diethyl phthalate (blue). (b) Lower paint layer contains a titanium dioxide mixture of anatase and rutile, with traces of the varnish components identified in (a).
Figure 12. Raman spectra of the paint on the back of the artwork. (a) Varnish layer contains nitrocellulose (black), camphor (green) and diethyl phthalate (blue). (b) Lower paint layer contains a titanium dioxide mixture of anatase and rutile, with traces of the varnish components identified in (a).
Heritage 05 00060 g012
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Bakovic, M.; Karapandza, S.; Mcheik, S.; Pejović-Milić, A. Scientific Study of the Origin of the Painting from the Early 20th Century Leads to Pablo Picasso. Heritage 2022, 5, 1120-1140. https://doi.org/10.3390/heritage5020060

AMA Style

Bakovic M, Karapandza S, Mcheik S, Pejović-Milić A. Scientific Study of the Origin of the Painting from the Early 20th Century Leads to Pablo Picasso. Heritage. 2022; 5(2):1120-1140. https://doi.org/10.3390/heritage5020060

Chicago/Turabian Style

Bakovic, Marica, Slobodanka Karapandza, Sajed Mcheik, and Ana Pejović-Milić. 2022. "Scientific Study of the Origin of the Painting from the Early 20th Century Leads to Pablo Picasso" Heritage 5, no. 2: 1120-1140. https://doi.org/10.3390/heritage5020060

APA Style

Bakovic, M., Karapandza, S., Mcheik, S., & Pejović-Milić, A. (2022). Scientific Study of the Origin of the Painting from the Early 20th Century Leads to Pablo Picasso. Heritage, 5(2), 1120-1140. https://doi.org/10.3390/heritage5020060

Article Metrics

Back to TopTop