Wall Drawing #736: Revealing Sol LeWitt’s Ink Mural Technique Using a Multi-Analytical Approach

Abstract

Sol LeWitt, a pioneer of conceptual art, created during his career over 1350 wall drawings, including the Wall Drawing #736 (1993) at the Center for Contemporary Art Luigi Pecci in Prato (Prato, Italy). The painting, executed by Andrea Marescalchi and Antony Sansotta under LeWitt’s instructions, features a grid of coloured rectangles obtained by overlapping different layers of inks. During a 2021 restoration by the Wall Paintings and Stuccoes Department of the Opificio delle Pietre Dure (Firenze, Italy), an in-depth investigation of the composition and the materials used by LeWitt’s assistants in producing Wall Drawing #736 was performed. A multi-analytical approach entailing Raman spectroscopy, high-performance liquid chromatography coupled to diode array and high-resolution mass spectrometry, gas chromatography–mass spectrometry (GC–MS), and pyrolysis coupled with GC–MS was applied. Our results revealed the use of animal glue, shellac resin, paraffin wax, linseed oil, and various organic pigments. The binder in the preparation layer was identified as poly(vinyl acetate), while poly(n-butyl methacrylate) was determined as a fixative. This research provided valuable insights into LeWitt’s techniques. The acquired knowledge on the paint technique is highly relevant in supporting conservators in restoration and consolidating the many wall drawings produced exploiting the same technique all over the world.

1. Introduction

The assessment of the material nature of works of art has become a standard procedure, essential for guiding conservation practices and enriching the historical and artistic understanding of artists’ techniques. In this context, the characterisation of the materials employed in the creation of wall paintings by the American artist Sol LeWitt (1928–2007) stands as a unique case study.

Originating as faint graphite marks on the wall, Sol LeWitt’s wall drawings transitioned into coloured pencils in 1969 at the Dwan Gallery in New York. Subsequently, in 1982, the introduction of coloured inks marked a pivotal evolution in Sol LeWitt’s painting technique, yet these were supplanted by acrylic paints in the mid-1990s (signifying the artist’s commitment to innovation and the evolution of materials within his practice).

Over the span of approximately 38 years, Sol LeWitt produced ca. 1350 wall drawings, adding up to 3500 when accounting for installations dispersed across over 1200 locations. Among this rich body of work, the series “Wall Drawings” stands out as one of the most distinctive and emblematic of the artist, revered as one of the progenitors of the multifaceted movement within contemporary arts known as conceptual art.

Indeed, from the 1980s onward, a distinctive operative practice emerged, that we also observed in Wall Drawing #736, located in the Center for Contemporary Art Luigi Pecci in Prato (Italy). Since the late 1960s LeWitt began to delegate the execution of his wall drawings to trusted collaborators, who followed precise instructions that the artist himself provided. In particular, Anthony Sansotta and Andrea Marescalchi contributed to the production of most of the Italian artworks. This shift is crucial to understand the peculiar nature of LeWitt’s wall drawings: these are therefore “institutionalized” art projects with legal and contract value, described in special “diagrams”, certified as authentic and documented, including the names of the first performers and the place of the first installation. Using the same diagram, the work could be replicated and exist even in different places, even at the same time, so that the work consists both of a conceptual project created by the artist and of its physical execution by a performer. To define such a relationship between idea and execution, Sol LeWitt often used musical score as a metaphor: while the score written by the composer is unique and fixed, its executions on behalf of many performers is never the same.

At first, these creations were explicitly conceived as ephemeral, or, according to the artist, “permanent installations until they are destroyed” [1], essentially because they were initially intended for temporary exhibitions, and their “destruction” was somehow anticipated. When purchased by a collector, the owner could readily recreate the artwork if needed, for instance if relocating or repainting the walls, while remaining faithful to the project created by the artist and outlined in the accompanying diagram.

According to the artist, maintaining the exact correspondence of the appearance of the artwork to the primary idea is of paramount importance, as any “alteration” could betray the original intention. LeWitt articulated this concept in 1984, stating that “If a wall drawing is done and redone every certain amount of years, it’ll be more exactly the same as when it was first done and conceived. It’s a way for preserving the work... so you don’t have deterioration” [2,3,4]. As a consequence, in many of LeWitt’s works, including Wall Drawing #736, a disconnection occurs between two concepts traditionally related in conservation: authenticity and authorship of an artwork as materially created by the artist. At the same time, LeWitt introduces an innovative concept of artwork that can potentially be remade indefinitely without incurring a forgery.

Thus, to remain faithful to the artist’s intentions, we found imperative to adopt a new approach in the conservation of Wall Drawing #736 and in the investigation of the materials. Here, the study of the technique and the materials was configured as a fundamental conservative action: in this case it was more imperative to preserve the process by which the artist had wanted Wall Drawing #736 to be created, than to safeguard the original materials. This entailed not only assessing of the characteristics of the inks, allowing for the peculiar chromatic effect planned by LeWitt, but also collecting the indications of the artist and all the documentation of the work to be able to re-enact it correctly in the future, using materials as similar as possible to the original ones. This is the current approach used for ephemeral or performative works like installations nowadays [5,6].

This approach follows the indications suggested by Yale University Art Gallery, where an archive and a study and conservation center dedicated exclusively to Sol LeWitt’s wall drawings was set up in 2013 [7].

During the restoration of Wall Drawing #736, we had the opportunity to contact Andrea Marescalchi’s heirs, who granted us access to the materials in his archive. Here, fundamental information was gathered for an unprecedented approach to conservation: LeWitt and Marescalchi methodology was indeed adopted by the Opificio delle Pietre Dure‘s restorers in the conservation intervention, and the original Pelikan inks conserved in the Marescalchi Archive were used to retouch the only lacuna.

In this framework, a full diagnostic campaign was carried out on selected samples collected from the wall painting and on the inks conserved in the Marescalchi archives.

Besides the application of non-invasive preliminary analyses and possibly portable XRF and in situ vibrational spectroscopies, non-destructive techniques as FTIR and Raman can be applied to obtain information on pigments and organic components of the inks and the paint samples [8,9]. Moreover, chromatographic techniques coupled with mass spectrometry (MS) are widely recognised as the most powerful micro-destructive techniques for characterizing the entire range of organic materials found in works of art and archaeological materials. Advanced gas chromatography–mass spectrometry-based techniques (GC–MS) and analytical pyrolysis–GC–MS allow the simultaneous screening of natural and synthetic binding media, additives and synthetic pigments [10,11,12,13]. Pyrolysis-based techniques are particularly useful when dealing with synthetic binders, commonly employed in the production of modern and contemporary wall paintings [14,15,16]. Additionally, detailed information can be obtained about individual classes of binding media and finishing materials via sample pre-treatment prior to GC–MS [17]. Instead, a comprehensive profile of natural, artificial and synthetic organic pigments along with their ageing products can be achieved by the application of liquid chromatography with spectrophotometric or mass spectrometric detection [18,19,20,21], provided that the compounds of interest are soluble in organic solvents.

Exploiting the full range of possible analytical techniques available for the characterization of paint materials in micro-samples, we adopted a multi analytical approach, entailing both non-destructive (Raman spectroscopy) and micro-destructive techniques (high-performance liquid chromatography coupled with diode array detector and tandem mass spectrometry (HPLC–DAD–ESI-Q-ToF), gas chromatography–MS (GC–MS), and pyrolysis coupled with gas chromatography–MS (Py–GC–MS)). The multi-analytical approach was applied both on samples collected from the Wall Drawing #736 and the inks used for its production, provided us as archive materials. The results enabled us to determine a complex composition characterised by animal glue, shellac resin, and synthetic organic pigments, with differences depending on the investigated colour. Further investigations of the materials used by Marescalchi and Sansotta for Wall Drawing #736 were performed by Py–GC–MS and Raman spectroscopy.

Eight microsamples were collected from the wall drawing, representing the entire colour scheme of the mural in accordance with the archives of the Center for Contemporary Art Luigi Pecci (Prato). This strategy allowed us to assess the presence of a fixative on the Wall Drawing #736 and to identify the binder employed in the white preparation layer, reconstructing the artistic criteria used for the creation of the artwork.

The purpose of the research was to understand the nature of the materials constituting Wall Drawing #736 to support present and future conservative interventions on this artwork and on other paintings made with this technique. Indeed, Sol LeWitt modus operandi demands an in-depth knowledge of the artist’s materials and methods to preserve this peculiar category of works and presents a unique case study in this regard. This study builds on the information collected during a 2011 interview with Marescalchi.

The idea of “remaking” the drawing infinite times, as foreseen by Sol LeWitt, can only be pursued if the original materials can be reproduced and made available accordingly to the authors’ instructions, or opportunely substituted with idoneous new materials. Since the production of commercial inks is often discontinued by the manufacturers, the analysis of the available archive artists’ materials, besides that of the actual work of art, is pivotal in this case study. The results of the multi-analytical approach entailing spectroscopic, chromatographic, and mass spectrometric techniques applied to artists’ materials (inks) and micro-samples collected from Wall Drawing #736 were compared with information gathered interviewing Marescalchi in 2011.

2. Materials and Methods

2.1. Contemporary Inks from the Marescalchi Archive

Andrea Marescalchi (Rome, 1954–Florence, 2015), known as “Bobo”, was an Italian artist, collaborator of Sol LeWitt since the 1980s, and executor, usually together with Sansotta, of many of his wall drawings in Italy. Since 2017, the “Associazione Culturale Archivio Andrea Marescalchi”, or Marescalchi archive, has been dealing with promoting the archival activity and the diffusion of the work of Andrea Marescalchi. Based on the information found in the archive, drawing inks (Pelikan D.I.) were used in Sol LeWitt’s mural drawings. Bottles of the original inks used to draw a series of paintings in Tuscany and in other parts of Italy in the decades 1980s–1990s are presently conserved in the Marescalchi archive. 1 mL of each ink was collected from their original bottles to perform preliminary analyses on their composition. Each ink was then cast onto a microscope glass slide, spreading it with a spatula to obtain a homogenous layer. The ink mock-ups were allowed to dry for two weeks before sample collection. The dried samples were then subjected to both non-destructive and micro-destructive analysis, as described in Section 2.4.

2.2. Technical Information from the 2011 Interview with Andrea Marescalchi

In 2011 we had the opportunity to interview Andrea Marescalchi, regarding the wall drawings of the Giuliano Gori collection in Celle, near Pistoia, made in the mid-1980s with the same materials used for the one in Prato. Marescalchi, on that occasion, provided us with highly valuable technical information. Here we summarise the guidelines followed in the production of the artwork as described in the interview [22].

After preparing the wall with semi-washable white acrylic paint and having obtained an orange peel-like surface similar to watercolour paper, the drawing was transferred with graphite from a diagram to the wall; then a mask was made for each section using paper rolls for calculators and adhesive tape. For Pecci’s work, to obtain precise boundary lines and prevent the ink from “smudging” between one section and the next, the tape was sealed with an acrylic emulsion applied by brush. In the section to be coloured, two more white layers were applied prior to the coloured paint. Once the tape was removed, the boundary line was extremely well defined. Following the artist’s scheme, each specific listed colour was applied, after dilution in water in precise proportions, in the exact sequence indicated (colours were superimposed and not mixed), so that the different sequences of primary colours created an extraordinary variety of tones. In contrast, the black ink was initially diluted in water (1:60, v/v) to obtain a grey ink that was added to the red, yellow, and blue ones before their application, as detailed in the guidelines reported in the Marescalchi archive (1987 formulations, Figure 1).

Each colour was applied in three overlapping coats, using pads made from seamless cotton jersey t-shirts to avoid any visible marks. The first layer was applied in three layers using circular movements with a technique that Marescalchi defines as “wiped”, the second also in three layers by dabbing, and the third once again by ”wiping” as the first one. As the ink dried rapidly, it was necessary to work quickly.

Finally, a matte varnish produced by the company Lascaux, probably 575UV Matte [23], was applied as a fixative by airbrush.

2.3. Technical Information from the 2011 Interview with Andrea Marescalchi

The Wall Drawing # 736 (338 × 1908 cm) consists of a grid of coloured rectangles that covers the upper part of the walls of a circular room. The production technique involved the use of drawing inks, overlapped in different layers of colours, above a white preparation layer, as described in the previous section. The sequence of the colour layers was specified in the diagram written by LeWitt by acronyms (R = red, B = blue, Y = yellow, G = grey), where the sequence of letters indicates the expected order of application of the inks. Moreover, a preliminary examination with UV lamp have highlighted fluorescent regions, further investigated with a Dinolite microscope on the edges of the wall painting, which suggested the presence of a fixative finishing. Sansotta reported the use of a matte varnish by Lescaux, applied as a spray. Figure 2a shows a scheme of the stratigraphic section of Wall Drawing #736. To investigate the organic materials constituting the artwork, eight representative samples of Wall Drawing #736 were collected and analysed:

• Samples 1, 2, and 8 were fragments detached from the wall;
• Samples 3, 4, 5, 6, and 7 were selectively collected to fully characterise the materials used in the wall drawing.

The sampling points are shown in Figure 2b, while a detailed description of each sample is provided in

Table 1. Samples and description.

Label	Description	Sample Fragment	Label	Description	Sample Fragment
1	BRB: collected from the floor below one of the blue areas: fragment with layers of blue and red inks on a white base preparation.		2	RYG: collected from the floor below one of the red areas: fragment with layers of red, yellow, and grey inks on a white base preparation.
Label	Description	Sampling spot	Label	Description	Sampling spot
3	White preparation layer without fixative.		4	White preparation layer with fixative.
5	YY: selective surface sampling of yellow ink. Probable traces of black ink and fixative.		6	GG: selective surface sampling of grey ink (black ink diluted 1:60 in water). Probable traces of fixative.
7	RRR: selective surface sampling of yellow ink. Probable traces of black ink and fixative.		8	RGG: fragment with layers of red and black inks on a white base preparation.

.

2.4. Apparatus

2.4.1. Raman Spectroscopy

Raman analyses were carried out using a Renishaw RM 2000 instrument coupled to a Leica DLML optical microscope, equipped with a 50× NPLAN objective. The laser sources were a He–Ne and a diode laser with wavelengths of 633 nm and 785, respectively. The laser power output on the sample was adjusted between 1 and 5 mW. The spectrometer consists of a single grating monochromator (1000/1200 lines mm⁻¹), coupled to a CCD. The spectral resolution was 0.5 cm⁻¹ @ 633 nm and 2 cm⁻¹ @ 785 nm; the spectral range analysed was between 100 and 3200 cm⁻¹. The spectral calibration of the instrument was performed on the 520.5 cm⁻¹ band of a pure silicon crystal.

2.4.2. Gas Chromatography–Mass Spectrometry (GC–MS)

To determine binder and additives of the Marescalchi archive inks (black, blue, yellow, and red), they were investigated by gas chromatography–mass spectrometry (GC–MS) after hydrolysis, extraction, purification, and derivatisation steps for the identification of proteinaceous, lipid, polysaccharide, resin, and wax components [10]. A 6890 Gas Chromatograph coupled with a 5975 Single Quadrupole Mass Selective Mass Spectrometer (Agilent Technologies, Santa Clara, CA, USA) was used, and further details on analytical conditions are reported in [10].

2.4.3. Pyrolysis Coupled to Gas Chromatography–Mass Spectrometry (Py–GC–MS and Py(HMDS)–GC–MS)

Samples from the ink mock-ups and Wall Drawings #736 were also analysed by analytical pyrolysis combined with gas chromatography–mass spectrometry (Py–GC–MS). A multi-shot pyrolyser EGA/PY-3030D (Frontier Lab) coupled to an 8890 gas chromatograph, combined with a 5977B mass selective single quadrupole mass spectrometer detector (Agilent Technologies) was used. For the MS unit apparatus, the following parameters were applied: electron impact ionization (EI 70 eV) in positive mode; ion source temperature set at 230 °C; scan range of 35–600 m/z; interface temperature set at 280 °C. To characterise the different classes of materials, the following two different approaches were used: (a) for the identification of synthetic polymers and dyes, pyrolysis was carried out at 600 °C; (b) for evaluating the presence of shellac resin and/or gum Arabic or other natural organic binders and additives, pyrolysis was carried out at 550 °C with in situ silylation using 3 µL of hexamethyldisilazane (HMDS) as derivatizing agent. The pyrolysis products were separated by a HP-5MS capillary column (95% dimethyl-5% diphenyl-polysiloxane; 30 m × 0.25 mm, film thickness 0.25 µm; Agilent Technologies, USA). Helium (He, purity 99.9995%) was used as gas carrier, with a constant flow of 1.2 mL/min. The GC injector was operated in split mode at 280 °C and with a 20:1 ratio. The GC oven temperature programs were as follows: (a) for synthetic polymers and dyes analysis: 40 °C for 5 min, 10 °C/min up to 310 °C for 20 min; (b) for shellac resin and gum Arabic analysis: 35 °C for 10 min, 10 °C/min up to 310 °C for 20 min.

2.4.4. High-Performance Liquid Chromatography (HPLC–DAD–ESI-Q-ToF)

Prior to the liquid chromatography analysis, 1 mg of yellow and red ink was collected from the mock-up and solubilized in 300 μL of DMSO (99.8% purity; J.T. Baker, CA, USA). The DMSO sample pretreatment was performed in accordance with [24]. Injection volumes were 10 µL for HPLC–DAD and 2 µL for HPLC–ESI-Q-ToF experiments. The HPLC–DAD system consisted of a PU-2089 quaternary pump equipped with a degasser, an AS-950 autosampler and an MD-2010 spectrophotometric diode array detector (DAD) (Jasco International Co., Hachioji, Japan). The HPLC–ESI-Q-ToF system consisted of an HPLC 1200 Infinity, coupled to a Jet Stream ESI-Q-ToF 6530 Infinity detector and equipped with an Agilent Infinity autosampler (Agilent Technologies, Palo Alto, CA, USA). For both systems, the chromatographic separation was performed on a Poroshell 120 EC-C18 column (3.0 × 75 mm, particle size 2.7 μm, Agilent Technologies), equipped with a Zorbax pre-column (4.6 × 12.5 mm, particle size 5 μm, Agilent Technologies). The chromatographic system was thermostated at 30 °C. The mobile phases were A) formic acid (FA 0.1% v/v) in water and B) formic acid (FA 0.1% v/v) in acetonitrile. The flow rate was 0.4 mL min⁻¹ and the program was 15% B for 2.6 min, then to 50% B in 13.0 min, to 70% B in 5.2 min, to 100% B in 0.5 min and then hold for 6.7 min; re-equilibration took 11 min. Further details can be found in [25,26].

3. Results and Discussion

3.1. Marescalchi Archive Inks

The combination of spectroscopic and mass spectrometric techniques provided us with the exhaustive characterization of the binder, dyes/pigments, and additives in the inks. To identify the formulation, comparison with standards and reference materials or matching data from the literature were used [10,11,27,28]. The results are summarised in

Table 2. Results of the characterisation of Marescalchi archive inks (archive inks).

Archive Ink	Pigments	Binder/Additives
Black	Carbon black	Animal glue, shellac resin, paraffin wax, 1,6-hexanediol
Blue	PB 15:3	Animal glue, shellac resin
Yellow	PY3, PY13	Animal glue, shellac resin, siccative oil
Red	PR149, PR11, unknown naphthol AS pigment	Animal glue (?), shellac resin, siccative oil

, and the pyrolytic markers (obtained after pyrolysis with and without HMDS) used for the identification of artistic materials are summarised in

Table 3. Summary of the results obtained by Py–GC–MS and Py(HMDS) –GC–MS analysis of archive inks (G, B, Y, and R) and samples from Wall Drawing #736. For each identified material, the following information are reported: pyrolytic markers, retention time (t_r, min), main m/z in the MS spectrum, and samples in which the marker was identified, marked with x (1 = BRB; 2 = RYG; 3 = white preparative layer; 4 = white preparative layer + fixative; 5 =YY; 6 = GG; 7 = RRR; 8 = RGG; G = black archive ink; B = blue archive ink; Y = yellow archive ink; and R = red archive ink). * was identified only with Py(HMDS)–GC–MS.

Material

Pyrolysis Products

tr (min)

m/z

Samples

G

B

Y

R

3

4

5

6

7

1

2

3

8

PB 15:3

1,2-dicyanobenzene

12.9

128, 101, 75

x

Benzonitrile

15.7

103, 76, 50

x

O-cyanobenzoic acid

16.5

147, 76, 104

x

PY13

2,4-dimethylaniline

14.5

121, 106, 120

x

N-(2,4-dimethylphenyl) formamide

14.6

147, 118, 132

x

x

N-(2,4-dimethylphenyl) acetamide

16.8

121, 106, 163

x

x

3,3′-dichlorobiphenyl-4-amine

19.8

237, 239, 167

x

PR149

m-xylene

11.5

91, 106, 77

x

Benzo[c,d]indol-2(1H)-one

17.1

169, 144, 140

x

Naphthol AS pigment

Toluene

9.9

91, 65, 39

x

Phenol

12.8

94, 66, 39, 55

x

p-aminotoluene

13.7

106, 77, 51

x

2-naphthol

16.7

144, 115

x

Animal glue

Pyrrole

10.6

67, 39, 52

x

Toluene

10.9

91, 65, 39

x

3-methyl-1H-pyrrole

12.1

53, 80

x

Hexahydro-pyrrole [1,2-a] pyrazine-1, 4-dione

18.8

111, 83, 154, 70, 91, 48, 126

x

Shellac resin

Butolic acid-TMS *

33.0

373, 357, 275, 215, 185, 157, 87

x

PVAc (polyvinyl acetate)

Acetone

5.4

43, 58

x

x

x

x

x

Acetic acid

7.2

43, 60

x

x

x

x

x

Benzene

7.9

78, 51, 39

x

x

x

x

x

Toluene

9.9

91, 65, 39, 51

x

x

x

x

x

Ethylbenzene

11.4

91, 106, 51, 77

x

x

x

x

x

Styrene

1.,8

104, 78, 51, 63

x

x

x

x

x

Indene

13.5

115, 89, 63, 58, 39

x

x

x

x

x

1-methyl indene

14.4

130, 115, 51, 64, 77, 39, 102

x

x

x

x

x

3-methyl indene

14.5

130, 115, 51, 64, 77, 39, 102

x

x

x

x

x

Naphthalene

14.7

128, 102, 51, 39

x

x

x

x

x

2-methyl naphthalene

15.5

142, 115, 63, 89

x

x

x

x

x

1-methyl naphthalene

15.6

142, 115, 63, 89

x

x

x

x

x

Acenaphthene

15.9

154, 76, 115, 128, 51, 63, 102

x

x

x

x

x

PBMA (poly (n-butyl methacrylate))

1-butene

4.9

41, 56

x

x

x

x

x

x

x

Benzene

7.9

78, 63

x

x

x

x

x

x

x

Methyl methacrylate

8.8

41, 69, 100, 85, 59

x

x

x

x

x

x

x

Methacrylic acid

10.8

39, 86, 58

x

x

x

x

x

x

x

n-butyl methacrylate

12.7

69, 87, 41, 56, 29, 99, 113

x

x

x

x

x

x

x

PEA (poly (ethyl acrylate))

Ethanol

5.2

31, 45

x

Ethyl acrylate

8.5

55, 45, 99, 73

x

Ethyl methacrylate

10.7

69, 41, 99, 84, 114

x

Diethyl glutarate

15.1

143, 115, 87, 55, 73, 101

x

Dimer

15.5

98, 154, 53, 81, 109

x

Trimer

18.2

134, 208, 255, 79, 181, 106, 152, 55, 125, 226

x

.

3.1.1. Chemical Investigation of Binders and Additives

Py–GC–MS analysis enabled us to identify in the black ink pyrrole, hexahydro-pyrrole [1,2-a] pyrazine-1,4-dione, toluene, and 3-methyl-1H-pyrrole, characteristic pyrolytic markers of the collagen contained in animal glue (Figure 3a) [28]. The latter is historically recognised as the binder par excellence in China or India ink formulations [29,30], i.e., black inks obtained by mixing pine or oil lamp soot (carbon black) with pre-dissolved glue and additives to obtain sticks. In addition, 1,6-hexanediol was detected as an additive in the black ink. Moreover, our investigation enabled us to detect butolic acid-TMS (pyrolytic marker of shellac resin) by Py(HMDS)–GC–MS in the blue ink [31] and by HPLC–ESI-Q-ToF in red and yellow inks [26,32]. Shellac resin is a natural resin widely used over the centuries as adhesive and coating material but it was also exploited in inks for increasing their viscosity [26]. To confirm these results, and to further investigate the ink composition, a combined GC–MS analytical protocol was applied, allowing us to determine the proteinaceous material contained in the archive inks in terms of individual aminoacidic content. GC–MS analysis allowed us to detect hydroxyproline, the characteristic amino acidic marker of animal glue [33,34], in all of the Marescalchi archive inks (black, blue, yellow, and red). Data elaboration on the complete aminoacidic profile (

Table 4. Semiquantitative results obtained for amino acids (AA) profile of the proteinaceous component of Marescalchi archive inks (black, blue, and yellow), compared with an average amino acid profile of reference protein materials. The results achieved for the red ink are not shown since values lower than the LOQ for proteinaceous materials were obtained.

Material	%Ala	%Gly	%Val	%Leu	%Ile	%Ser	%Pro	%Phe	%Asp	%Glu	%Hyp
Blue ink	3.9	3.0	5.9	6.6	9.7	12.7	4.8	11.4	9.7	11.8	20.5
Black ink	5.4	12.4	3.6	5.5	5.0	6.3	32.3	5.8	5.0	6.9	11.7
Yellow ink	16.0	8.8	4.9	5.1	7.7	11.1	3.9	8.7	7.7	9.7	16.5
Casein	5.0	3.0	7.6	11.9	6.6	5.8	11.5	5.9	8.5	22.2	0.0
Egg	7.7	4.8	7.7	11.0	6.7	10.3	5.7	6.4	12.6	15.0	0.0
Animal glue	12.3	29.4	3.9	4.7	2.5	3.8	12.4	2.8	6.6	9.9	7.7

), compared to a set of reference materials [10], enabled us to confirm animal glue in the black, blue, and yellow inks, despite a relatively low glycine content compared to that reported in the literature for animal glue. For the red ink, a total aminoacidic content lower than the limit of quantitation was obtained; thus, it was not possible to quantitatively evaluate the aminoacidic profile. No polysaccharide markers were identified, proving the absence of plant gums in the archive inks. Instead, the examination of the lipid content allowed us to confirm the presence of shellac resin in all the formulations and encouraged us to further investigate the presence of drying oils. In detail, shellac resin was confirmed by the GC–MS detection of butolic acid (18.7 min), laccishellolic acid (21.4 min), derivatives of aleuritic acid (23.0 and 24.1 min), and aleuritic acid (25.3 min). Jalaric and laccijalaric acid were not observed, suggesting that the resin had undergone ageing to a certain extent, or that it was pretreated in a specific way prior to its introduction in the ink formulation. Except for laccishellolic acid, no Cannizzaro-type disproportionation products related to the typical sesquiterpenoids of shellac resin were detected; however, this is expected because a soft saponification was performed, whereas stronger conditions are necessary to identify them by GC–MS [31].

As for the investigation of drying oils, different ratios between representative free fatty acids were evaluated and compared with the literature data and in-house databases to obtain information on the composition of the medium in the inks [35]. The details are as follows:

• A/P (azelaic acid/palmitic acid) is commonly evaluated to investigate the occurrence of drying oils (A/P > 1), since the formation of dicarboxylic acids (peaking with azelaic acid) production occurs during the curing and ageing of drying oil. As for the black and blue inks, A/P < 1 values were observed. The lipids observed in their chromatograms cannot be associated with a siccative oil, and their presence can thus be associated to free acids of shellac resin, given that Py–GC–MS analysis already excluded the presence of egg, another common source of lipids in binding media. Conversely, A/P > 1 was observed in the red and yellow inks, consistent with a molecular profile typical of drying oils;
• P/S (palmitic acid/stearic acid) is a chemical parameter commonly used to differentiate drying oils based on the different relative abundance of the two main saturated monocarboxylic acids. The quantitative analysis performed for the red and yellow inks provided us with a P/S ratio values of 1.6 and 1.1, respectively, compatible with linseed oil, whose presence cannot be fully ascertained since mixtures cannot be excluded [35];
• % ∑D percentage (dicarboxylic acids sum/total free acids) and O/S (oleic acid/stearic acid) are chemical parameters useful to evaluate the degree of ageing/curing, since oleic acid is highly reactive towards oxidation, and as stated above dicarboxylic acids are auto-oxidation products. The values achieved (O/S = 0.7, for both red and yellow inks) suggest a partial oxidation of the lipid binder, consistent with the fact that the archive inks are quite contemporary (80s–90s) and were in liquid form when cast on the glass slides as mock-ups, which were further sampled and analysed just after few weeks; notably, we determined a higher ageing degree for the red ink (% ∑D = 50) than for the yellow one (% ∑D = 36).

In summary, the molecular profile of fatty acids and dicarboxylic acids observed in the red and yellow inks provided us with the identification of partially oxidised siccative oil. Concerning the additives, GC–MS analysis (Figure 3b) highlights the presence of paraffin wax in the black ink.

3.1.2. Investigation of Dyes and Pigments

The detection in the Raman spectrum (Figure 4a) of the G (1596 cm⁻¹) and D (1325 cm⁻¹) bands, typical of carbonaceous materials [36], enabled us to characterise the black archive ink as a carbon black-based. By combining the identification of carbon black and animal glue (3.1.1), we confirm the use of China ink, validating the information found in the Marescalchi archive. Neither Py–GC–MS nor HPLC–DAD–ESI-Q-ToF provided us with further information on the pigments, as carbon black is not detectable by either of them.

As for the blue ink, Raman spectroscopy provided us with the identification of Phthalo Blue 15:3 (PB 15:3), a phthalocyanine pigment widely exploited by several 20^th century artists [37]. Consistently, the pyrolytic markers of phthalo pigments were detected by Py–GC–MS. However, the discrimination between different possible blue phthalocyanine pigments by analytical pyrolysis is quite challenging, since their pyrograms all present 1,2-dicyanobenzene and o-cyanobenzoic acid as main pyrolytic products [11,38,39]. Nevertheless, the detection of secondary markers as benzonitrile (Figure 4b) allows us to narrow the field down to PB 15 and PB 15:3, which only differ in their crystal structure. Phthalocyanine-based pigments are insoluble in almost all organic solvents, making them undetectable by techniques based on liquid chromatography.

The adopted multi-analytical approach applied to the yellow ink allowed us to highlight a mixture of synthetic organic pigments, comprising Pigment Yellow 3 (PY3), and Pigment Yellow 13 (PY13). PY3, a monoazo pigment firstly synthesised in 1910, was identified by HPLC–DAD–ESI-Q-ToF analysis (Figure 4c: t_R = 27.9 min, λ_max = 347, 407 nm, [M-H]⁻ = 393.016 m/z) [40], while PY13 was identified by Py–GC–MS (Figure 4b) and Raman spectroscopy only (SOPRANO database of Raman spectra obtained on synthetic organic pigments [41] and the related publication [42]). According to pyrolysis fragmentation pathways proposed for diarilyde pigments by Russel et al. [38], PY13 can be identified based on the detection of several dimethylaniline and dimethylphenyl compounds (Table 3). In this case, Py–GC–MS proved suitable for the specific identification of diarilyde pigments which differ in the substituents on the terminal benzenes.

Finally, in the red ink, a perylene pigment (PR149) [43] and an unknown naphthol AS pigment were determined (Figure 4b). According to Russel et al. [38], the presence of PR149 could be ascertained by the detection of 1,3-dimethyl benzene by Py–GC–MS. However, also benzo[cd]indol-2(1H)-one was detected, and a fragmentation pathway analogous to naphthalimide formation reported by Gherardi et al. [11] can be hypothesised, further supporting the identification of a perylene pigment. In addition, the pyrogram acquired for the red ink allowed us to determine 2-naphthol, which may have originated by the pyrolysis of a naphthol AS pigment [39]. Both Raman and HPLC–DAD results confirm this hypothesis: the first by detecting Pigment Red 11, and the second by allowing us to identify the characteristic UV–Vis spectrum acquired by PDA of a compound belonging to this class (Figure 4d). Unfortunately, HPLC–MS analysis did not allow us to confirm the presence of PR11 and identify the unknown naphthol AS pigment determined by HPLC–DAD, given the complexity of the chromatogram.

3.2. Samples from Wall Drawing #736

The organic materials present in the samples collected from the Wall Drawing #736 were investigated by Py–GC–MS and Raman spectroscopy. Py–GC–MS analysis (Figure 5) allowed us to highlight the presence of polyvinyl acetate (PVAc) in the white preparation (sample 3) and of poly (n-butyl methacrylate) (PBMA) and poly (ethyl acrylate) (PEA) in the white preparation containing the fixative (Sample 4). Moreover, the Py–GC–MS results achieved for the coloured samples (1, 2, 5, 6, 7, and 8) enabled us to determine the following information:

• PBMA in all the samples collected selectively from the surface (samples 5, 6, and 7);
• PBMA + PVAc in the samples obtained as fragments (degradation of the wall drawing), thus composed by both the coloured ink’s layers and the white preparation (Samples 1, 2, and 8).

Therefore, except for the white preparation (Sample 3), all the samples analyzed contained an acrylic binder (PBMA). This suggests that a PBMA-based fixative was applied in Wall Drawing #736, in accordance with the reports by Sansotta that a fixative was used. Moreover, it is possible to hypothesise that the preparative white layer contained PVAc as a binder, in accordance with the results obtained for most samples (1, 2, 3, and 8). Instead, the identification of PEA in sample 4 only, collected outside the painting area, is difficult to rationalise, and could be associated with a repainting of the wall with a different white colour, carried out in that specific area. Except for Sample 5 (YY), which featured the pyrolytic markers of the same yellow pigment (PY13) highlighted in the archive ink, neither signals related to the pyrolysis products of synthetic organic pigments nor of organic binders were observed in any of the other samples collected from the wall. This could be due both to the dilution to which the inks were subjected before their application and to the reduced quantity of pigments and binders compared to the inorganic content of the samples. However, the application of Raman spectroscopy enabled us to enhance our understanding regarding the pigment composition of Wall Drawing #736. Indeed, in sample 1 (BRB) the presence of PB 15:3 was revealed, while the investigation of samples 2 (RYG) and 8 (RGG) also enabled the detection of PR11. Thus, the results achieved with Raman spectroscopy allowed us to outline a matching dye/pigment composition with the Marescalchi archive inks. Moreover, rutile was identified in samples 1, 2, 3, 4, and 8, and calcite in sample 3. The detection of rutile could be linked to the white pigments in the preparation layer, while calcite could have been used as a filler.

4. Conclusions

Our research led to a deep understanding of this special ink-on-wall technique developed by LeWitt and his collaborators. Four inks conserved in the Marescalchi archive were fully characterised by a multi-analytical approach. Animal glue and shellac resin were determined as binder and additives, suggesting a molecular profile similar to the so-called China inks. In addition, a siccative oil (possibly linseed oil) was identified in the red and yellow archive inks, while paraffin wax was determined in the black ink only. Concerning the pigment component, the black ink turned out to be carbon black-based; in the blue ink, Phthalo blue 15:3 (PB15:3) was determined, and complex mixtures of SOPs were detected both in the yellow (PY3, and PY13) and red (PR11, PR149, unknown naphthol AS pigment) inks. The materials used for painting the wall were also accessed: polyvinyl acetate (PVAc) and poly (ethyl acrylate) (PEA) were identified as the binder of the white preparative layer before and after a restoration campaign at the Center for Contemporary Art Luigi Pecci, while poly (n-butyl methacrylate) (PBMA) was determined as the fixative in the entire Wall Drawing #736. Despite the high dilution of the inks in the final wall painting and the presence of a finishing varnish, the composition of the samples matched that of the archive materials for the dyes and/or pigment components. Finally, rutile and calcite were identified in the preparation layer of the wall drawing investigated, suggesting their presence as white pigment and filler, respectively.

By comparing archive inks with samples from Wall Drawing #736, we did not identify specific degradation products related to dyes or pigments, and the pyrolytic profiles of binders and fixative components did not highlight features related to ageing. This finding is consistent with the conservation conditions of the artwork, as the room housing the wall drawing is sheltered from direct sunlight, and a relatively short time has passed since the creation of the artwork.

The in-depth study of the ink’s composition conducted in this work is a first step towards improving our understanding of the continuous technological improvements in the production of paints and inks and how this has influenced Sol LeWitt’s choice of artistic materials. Indeed, while the composition highlighted for Wall Drawing #736 matches that of the materials used in street art murals during the second half of the 20th century [15], that of the original inks proved to be an interesting combination of historical binders (animal glue and shellac resin) with synthetic dyes and pigments. Most importantly, the acquired knowledge on the paint technique and palette of the artist is highly relevant for supporting conservators in restoration and consolidation interventions; gathering detailed physical–chemical information on the unique technique used by Sol LeWitt will permit the implementation of the relationship between project and realization, as envisioned by the artist.