“I want to express the utmost intensity of the color.”
Adolph Gottlieb1
1Adolph Gottlieb is one of the most prominent American artists of the mid-20th century. He is an early, and principal, protagonist in the story that would become known as abstract expressionism, and his work directly influenced the rise of color field painting2. Gottlieb is most widely recognized for a series of paintings that he began in 1956. Collectively referred to as “Bursts”, they consumed his attention for seventeen years, until his death in 1974.
- 3 Gruen, 1972, p. 258.
- 4 The color and the shape of the disc and the blast vary from painting to painting. There may be more (...)
2Each “Burst” is a subjective image whose meaning is intended to be deeply personal and identified by the subconscious of the viewer3. The language Gottlieb used to achieve this was limited to three elements: a monochromatic background onto which is painted a “disc”, that is often suspended above a “blast”4. The disc, spherical or elliptical in shape, is usually surrounded by a subtle halo that makes it appear to glow. The blast resembles an expanding tangled mass of gestural and more sharply defined forms.
3Gottlieb stated: “…since I eliminated almost everything from my painting except a few colors and perhaps two or three shapes, I feel a necessity for making the particular colors that I use, or the particular shapes, carry the burden of everything that I want to express, and all has to be concentrated within these few elements. Therefore, the color has to carry the burden of this effort5.”
- 6 Several “bursts” have the same dimensions as Thrust, but none are larger.
- 7 Friedman, 1962, p. 34.
4Thrust (1959) is among the largest of all the “Bursts” and was created while Gottlieb was still defining the imagery6 (fig. 1). It is also a classic example of these paintings and is executed in the colors Gottlieb found most exciting for having “within themselves a certain charge”: white, red, and black7.
Fig. 1. Adolph Gottlieb, Thrust, 1959, oil on canvas, 274.3 x 228.6 cm, George A. Hearn Fund, 1959 (Acc.# 59.164)
Before treatment (2018) photograph captured in ordinary light.
© The Metropolitan Museum of Art/E. Read, Department of Paintings Conservation. © Adolph and Esther Gottlieb Foundation/Adagp, Paris, 2023.
- 8 As per Sanford Hirsch, Executive Director of the Adolph and Esther Gottlieb Foundation, Adolph Gott (...)
5Gottlieb began Thrust by stapling cotton duck canvas onto a foldable strainer of his own fabrication8. He then re-stretched the canvas, likely after decrimping it. For the background, he laid down a thin, white, variably matte layer consisting of multidirectional strokes made with a brush and lengthwise strokes made with a roller. Several lengthwise streaks of paint were applied that fill the canvas weave more than the surrounding areas and appear glossy in specular light; it is not clear how these were achieved. As we discuss below, in cross-section these paint applications are indistinguishable from one another, suggesting a wet-in-wet technique.
- 9 Unpublished communication with Sanford Hirsch, Executive Director of the Adolph and Esther Gottlieb (...)
- 10 Friedman, 1962.
6With the painting face up on sawhorses (fig. 2 a), Gottlieb then created the black blast and the red disc. For the blast, he describes his process: “I get my paint in a rather viscous state9 so that it’s fluid, but not buttery, the way it comes out of the tube, and then I pour some of the paint on the canvas and it makes a puddle which is more or less a round puddle. It just takes that form by itself. And then I start manipulating this paint with a large spatula or a stick or a squeegee or whatever I feel will do what I want it to do, and I start flinging the paint out of this circular puddle in the direction I want it to go; and it’s as if the paint were then spun with a centrifugal motion and it flies in tangents, and it has to have the right viscosity so that it will handle and it will have the proper fluidity for this type of handling... And then, of course, if I decide it hasn’t spun out quite the way I want it to, I might drip it on a bit; but it isn’t primarily a dripping technique, which is done very differently...10”
Fig. 2 a. Adolph Gottlieb working on Ascent, New York, 1958
Courtesy of the Adolph and Esther Gottlieb Foundation archive/ Rudolph Burckhardt. © Adagp, Paris, 2023.
Fig. 2 b. Adolph Gottlieb in front of his painting Positive, New York, 1958
Courtesy of the Adolph and Esther Gottlieb Foundation archive. © Adagp, Paris, 2023.
7For the disc and halo, he describes: “I might put a disc, let us say, of red, and I’ll put it on with a spatula, just a big, flat disc. This is rather thick paint, even impasto. Then I might take a rag, soak it with turpentine, work around the edge, and develop something that is like a halo, so that I get this impasto-like color opposed to the very transparent, stained sort of color11.”
8For the black blast of Thrust, Gottlieb used a squeegee, likely followed by a brush. For the red disc, he used a palette knife. For the halo, he used a rag soaked in turpentine. The resulting surfaces have a variable gloss related to his materials and his techniques; overall, the blast is matte, the disc is glossy, and the halo is matte (fig. 2 b).
- 12 Bell et al., 1997.
- 13 McGonigle, Ciullo, 1996.
9As mentioned above, Gottlieb began by applying a white layer onto the canvas, likely in more than one application, suggested by the image taken under UV illumination (fig. 3) that shows distinct areas that fluoresce green and purple. These applications were most likely done wet-in-wet as it is not possible to distinguish more than one layer by optical microscopy or scanning electron microscopy coupled with energy dispersive spectroscopy (SEM-EDS) in the sample cross-section examined (fig. 4 a-c). Raman spectroscopy analysis of the white paint in this sample showed features characteristic of rutile at ca. 447 and 609 cm-1 and of barium sulfate at ca. 988 cm.-1 12. The presence of barium sulfate and rutile particles was confirmed by micro X-ray diffraction (µXRD) and by SEM-EDS analysis, which also revealed abundant talc particles and few canvas fibers in the white paint (fig. 4 d). Talc was also identified by Fourier transform infrared spectroscopy (FTIR) by its absorptions at ca. 3677, 1024 and 666 cm-1 in samples of white paint with varying fluorescence and gloss indicating a commercial formulation of the paints13. In addition to these components, particles of lithopone, a pigment produced through the co-precipitation of zinc sulfide and barium sulfate, were detected by SEM-EDS in the white layer regardless of green or purple fluorescence (fig. 4 d).
Fig. 3. Before treatment (2018) visible light photograph under ultraviolet illumination
© The Metropolitan Museum of Art/E. Read, Department of Paintings Conservation.
© Adolph and Esther Gottlieb Foundation/Adagp, Paris, 2023.
Fig. 4 a-d. Photomicrographs of a sample cross-section taken from a white paint area in Thrust taken under visible illumination
(a); ultraviolet illumination (b); and using backscattered electrons in the SEM (c), respectively; d. Detail of the image shown in (c) indicating particles of barite, lithopone, and talc in a very fine-grained rutile matrix.
10The relevant elemental distribution maps obtained by macro X-ray fluorescence (MA-XRF) of two areas of the painting include white, black, and red paints and are presented in figure 5. The technique confirmed that titanium, barium, and zinc, along with relatively very small amounts of lead, are present in the white paint (fig. 5, the barium distribution map is not shown in this figure); lead was below the detection limit of SEM-EDS performed on most of the cross-section shown in figure 5, but was found in trace amounts associated with some lithopone particles.
Fig. 5. On the left side of the figure, images of the painting taken under visible (top) and UV (bottom) illumination
On the right of these images are the elemental distribution maps acquired by MA-XRF in the two areas of Thrust indicated by red (Area 1) and blue (Area 2) rectangles (see image in the top left). The maps correspond, from left to right, to: zinc (Zn), calcium (Ca), iron (Fe), lead (Pb), manganese (Mn), potassium (K), phosphorus (P) and titanium (Ti).
© Adolph and Esther Gottlieb Foundation/Adagp, Paris, 2023.
11In the area of the red disc, calcium, iron, manganese, potassium, and phosphorus were observed by MA-XRF (fig. 5, Maps Area 1). FTIR analysis of a particle of the organic red pigment in the red paint identified it as an anthraquinone-based colorant. Visual examination and the detection of phosphorus and calcium by MA-XRF indicate that a black pigment, likely a bone or ivory black, is also present in the red paint. It is noteworthy that iron and manganese are not observed in the border of the disc.
- 14 With characteristic broad bands at ca. 1325 and 1580 cm-1.
- 15 The PO43- mode expected at ca. 960 cm-1 for bone black (Bell et al., 1997) was below detection in t (...)
- 16 With characteristic absorptions at ca. 1023 and 2014 cm-1.
- 17 For the red and black areas, it is not possible to determine by MA-XRF whether some titanium, zinc, (...)
12In the area of the black blast, calcium, iron, manganese, potassium, and phosphorus were observed by MA-XRF (fig. 5, Maps Area 2). Raman analysis of a black paint sample showed the presence of a carbon-based black pigment14. MA-XRF and SEM-EDS showed the presence of phosphorus and calcium indicating that it is a bone or ivory black15. FTIR confirmed the presence of bone or ivory black16, while hydroxyapatite was identified by µXRD. The detection of Fe and Mn by MA-XRF and SEM-EDS indicate that an earth pigment, possibly an umber, is present in the black paint mixture17.
- 18 With absorptions at ca. 2921, 2855, 1741 (sh), 1703, 1466, and 1411 cm-1.
- 19 Kokkori et al., 2015.
- 20 For traditional artists oil paints, fatty acid ratios of palmitic (P), stearic (S), and azelaic (A) (...)
13The binder of the paints was identified as oil by FTIR18 and further characterized by pyrolysis gas chromatography/mass spectrometry (Py-GC/MS), with and without derivatization. Samples from four white areas of varying gloss and UV fluorescence, a sample from the red, and two samples from black areas did not contain any polyol or other cross-linking molecules, ruling out the presence of an alkyd paint19. The ratios of palmitic/stearic acids (P/S) for all of the paints analyzed are nearly 1.0 while the ratios of azelaic/palmitic acids (A/P) range between 0.6 and 2.3 suggesting that the binder is linseed oil; the low ratios of suberic and sebacic acids to azelaic acid (0.1 and 0.4 respectively) indicate that the oil was not pre-polymerized20. The fatty acid profile excludes the use of significant amounts of sunflower, safflower, rapeseed, castor, tung, or fish oils, less expensive oils that were sometimes added to linseed oil, and no marker compounds for other common additives, such as pine resin, dammar, copal, or wax, were detected.
14The Metropolitan Museum of Art purchased Thrust from French & Company (New York) in 1959. Within several months of being created, the painting entered the museum’s collection and has been maintained in climate-controlled environments ever since.
15In many ways, the painting is in pristine condition. The folding strainer, while marginally flexible in the long direction when handled, is structurally sound and has maintained an even tension to the canvas, which is planar. The canvas has discolored slightly but remains strong and flexible. And the paint film is wholly intact; apart from several scratches and marks that appear to be artifacts of technique, the only blemish to the surface is a subtle three-centimeter scuff that is likely the result of handling. It’s a gem, except for one disfiguring issue: a white haze that has overtaken the black and the red passages.
16First observed in 1981 on the black blast as subtle traces of “a kind of efflorescence”, the 1986 condition report documents “an enormous increase of this disfiguring [haze]21” (fig. 6 a-b). The haze presented on the blast as an overall layer, most prominent in areas corresponding to strainer bars. It was also populated with a series of round spots, up to ~8 mm in diameter where it had accumulated more than the surrounding areas. At the right side of the blast, there were several zones within tidelines where the haze had a mottled effect.
Fig. 6 a-b. Before treatment photographs showing the progression of the haze from 1986 (a) to 2018 (b)
The efflorescence accumulated dramatically in the black paint except for areas associated with the small round spots that appear pale in the 1986 image, because they were treated at that time, and the larger areas within the tidelines, which at present may not be explained.
© The Metropolitan Museum of Art/C. Hale (a)/E. Read (b). © Adolph and Esther Gottlieb Foundation/Adagp, Paris, 2023.
- 22 The surface accumulation was not apparent on the 6-8 cm border around the perimeter of the disc.
17By 1996, the haze had become a prominent feature of the black paint, rendering it gray in color. The painting was placed in storage until 2018, when it was viewed in consideration for display. Except for areas that were treated in 1986 (discussed below) and the areas on the right side within tidelines, the haze had developed dramatically (fig. 6 a-b); it became crust-like in appearance and now covered the center22 of the red disc, as well (fig. 1). The red and the black paint lost their intensity, and the contrasts between them and the white background were muted. It is worth pointing out that the surface accumulation developed into a distracting haze where manganese and iron are observed in the XRF maps (fig. 5). In the red disc, the areas that contain manganese and iron have a different appearance in the UV image (fig. 3).
18The surface accumulation had a variable thickness resulting in an uneven graying effect throughout the black and the red paint; artifacts from how the paint had been manipulated with the squeegee, brush, palette knife, or rag had become exaggerated.
19Figure 7 a-c shows photomicrographs of a sample cross-section taken from a black paint area where the surface deposit is clearly visible at the top of the stratigraphy (fig. 7 c, layer 3). Conversely, the accumulation is not visible in the photomicrographs of the cross-section taken from the white paint area (fig. 4 a-c), but could be detected and characterized by chemical analysis as discussed below.
Fig. 7 a-d. a-c. Photomicrographs of a sample cross-section removed from a black paint area: taken under visible illumination
(a); ultraviolet illumination (b); using backscattered electrons in the SEM and acquired in the area indicated by a rectangle in b (c), respectively. In c, (1) is the white paint layer, (2) is the black paint layer, and (3) is the surface accumulation; d. Raman spectrum acquired in a sample taken from the accumulation in the white paint. In addition to features due to the efflorescence discussed in the text, bands at ca. 447 and 609 cm-1, characteristic of rutile, and at ca. 988 cm-1, the main peak of barium sulfate, are present in this spectrum. Rutile, barium sulfate and lithopone were detected in the white paint layer. The band for barium sulfate is shown out of range for the sake of clarity. λ0= 785 nm.
- 23 Robinet, Corbeil, 2003.
- 24 Robinet, Corbeil, 2003.
20Sample scrapings of the surface accumulation on the red, black, and white paints were analyzed by Raman spectroscopy, FTIR and µXRD. In all of these samples, Raman bands were observed at ca. 1061, 1128, 1293, 1437 and 1458 cm-1 (the last one relatively broader), wavenumbers that are consistent with strong or very strong peaks for palmitic acid, stearic acid, and the corresponding carboxylates (soaps) of zinc and lead (fig. 7 d)23. In our case, weaker peaks that would allow differentiating these species were below the detection limit of Raman spectroscopy, however FTIR analysis confirmed the presence of stearic and palmitic fatty acids as well as zinc and lead carboxylates. In addition to the Raman bands mentioned above, in a scraping taken from the surface of the black paint, a peak at ca. 1097 cm-1 was detected while a band at 1095 cm-1 was present in spectra acquired in a scraping from the surface of a white paint area. A very strong Raman band at 1093 cm-1 has been reported for zinc oleate in a sample of the pure compound, along with very strong features at ca. 1061, 1295, 1438 and 1458 cm-1 that overlap bands of palmitic acid, stearic acid, and the corresponding carboxylates of zinc and lead; while a medium intensity peak at 1097 cm-1 is present in Raman spectra of pure lead palmitate24. Therefore, we can conclude that both saturated fatty acids, most likely palmitic and stearic, along with lead and zinc soaps are present in and at the surface of the red, black, and white paints in Thrust.
- 25 Nyquist et al., 1997, p. 74.
- 26 Robinet, Corbeil, 2003.
- 27 Doherty et al., 2007.
- 28 Robinet, Corbeil, 2003.
21Only in the FTIR spectrum acquired from a scraping of the accumulation on the black paint, a weak sharp peak at ca. 1318 cm-1 along with a broad feature around 1625 cm-1 were observed. These are frequencies reported for calcium oxalate25 but they overlap bands of palmitic and stearic acids at ca. 1311 and 1313 cm-1, respectively, and at ca. 1318 cm-1 for lead stearate, all of which are CH2 deformation modes26. The Raman spectrum of calcium oxalate monohydrate (whewellite) is characterized by a doublet at ca. 1461 and 1485 cm-1 while that of the calcium oxalate dihydrate (weddellite) has a strong peak at ca. 1474 cm-1 27. The whewellite peak at 1461 cm-1 overlaps bands assigned to palmitic acid (1463 cm-1), zinc palmitate (1463 cm-1), zinc stearate (1458 cm-1), and lead stearate (1460 cm-1)28 and no bands at 1485 cm-1 were observed in the Raman spectra acquired on the samples analyzed, however it is not possible to rule out a contribution from weddellite at ca. 1475 cm-1 where a broad band is present in the Raman spectrum (fig. 7 d).
22µXRD detected barium sulfate and rutile in surface scrapings from the white paint, and hydroxyapatite in the scraping from the black paint, all compounds due to the presence of pigment particles in the samples. The diffraction pattern collected from the mixture of the red paint and its surface deposit, although of low intensity, indicates the presence of succinic, palmitic and/or stearic acids. No other diffraction peaks that would help confirm the presence of additional crystalline phases were obtained, perhaps due to an overall low sample crystallinity and small sample sizes.
- 29 Ordonez, Twilley, 1997; Catalano et al., 2018 and 2019; Osmond, 2019; Di Tullio et al., 2020; Baij (...)
- 30 See endnote 9.
- 31 Baij et al., 2020.
23The composition of the surface accumulation in Thrust suggests that its formation involves the migration of reactive species (ions, long chain fatty acids), saponification reactions and, possibly, the migration of soaps, however it is not possible to assert what triggered the processes. Factors that would increase the mobility of species in paints include environmental conditions (e.g. high temperature and relative humidity), the composition of the paints (type of pigment and binder, the pigment to oil ratio), past conservation treatments involving organic solvents, and the artist’s working practice29. Other than the local interventions with solvents described in the next section, the painting has not been treated; however, as mentioned above, Gottlieb tailored his proportions of linseed oil, resins, turpentine, mineral spirits, and possibly other solvents30. High proportions of linseed oil in the paints will supply larger amounts of long chain saturated fatty acids, and organic solvents are known to increase the rate of fatty acid migration and metal soap formation31.
24Thrust was first treated in 1986 in an attempt to reduce the gray appearance of the efflorescence on the black paint, which was so sensitive that dusting with a sable brush picked up pigment. It was decided at that time to leave the overall accumulation of efflorescence intact, however the prominent hazy spots (fig. 6 a-b) were reduced by dotting them with a sable brush damp with ethanol without applying pressure. With the spots no longer visible, the overall haze became far less distracting.
- 32 Absorene and makeup sponges were also tested to lesser effect.
- 33 See endnote 9.
25In order to treat the dramatic increase in the surface accumulation that occurred since 1986, cleaning tests were undertaken in 2022 with dry sable and bristle brushes. The paint film appeared to have become less sensitive to mechanical cleaning than it had been in 1986. While brushing was possible in tested areas, reducing the hazy appearance required a stiff bristle brush with an appreciable amount of pressure32. Given the scale of the affected area, the variable gloss, and the fact that Gottlieb modified his paint33, it was considered that the sensitivity of the paint could be unpredictable. Additionally, even after treatment the mobile species may continue to migrate through the paint and accumulate at the surface, necessitating further remediation throughout the life of the painting. The compounded effect of repeated interventions with a stiff bristle brush was deemed too abrasive for the paint film.
- 34 Palmitic acid: 61.8°C-63.0°C; stearic acid: 68.8°C-72.0°C; lead palmitate: 130°C; lead stearate 115 (...)
- 35 The authors are unaware of any studies published on the use of this method and its short- and long- (...)
26Treatment with a hot air pencil (fig. 8 and 9) was considered as a potential alternative. This technique takes advantage of the physical properties of the compounds in the efflorescence which, depending on its composition, may begin to melt at +/- 63°C34. When the temperature is raised to the melting points of its constituents, the haze may disappear, revealing the more saturated appearance of the artist’s paint film that lies below35.
Fig. 8. During treatment (2022) photograph of the hot air pencil in use
© The Metropolitan Museum of Art/C. Hale. © Adolph and Esther Gottlieb Foundation/Adagp, Paris, 2023.
Fig. 9. During treatment (2022) photographs showing the efflorescence before (right side of each photograph) and after (left side of each photograph) heating
© The Metropolitan Museum of Art/E. Read/S. Digney-Peer. © Adolph and Esther Gottlieb Foundation/Adagp, Paris, 2023.
27The technique requires balancing numerous variables, such as nozzle size, temperature, distance to the area of interest, duration of heat exposure, fan speed, and hand motion. Temperature measuring strips were used to establish a working method that allowed the melting point to be reached in a controlled manner; too much heat risked affecting the gloss of the paint and triggering additional physical and chemical changes36. A narrow nozzle size of 3.0 mm was selected as it was more controllable than wider nozzles.
- 37 It is possible that the remaining haze has a relatively larger proportion of compounds with melting (...)
28Heat treatment alone did not remove as much of the haze as heat treatment following brushing with a bristle brush. However, it achieved a resounding improvement. The process was straightforward at the left side of the blast; there was a consistent shift in saturation while the accumulation was heated, leaving a healthy-looking paint film with a matte surface upon cooling. Moving towards the right, the heat response of the efflorescence occurred more quickly with variable results; to keep the right side in balance with the left side, the duration of heat exposure and the distance of the pencil to the area of interest were constantly adjusted. The variable response of the surface accumulation to the hot air pencil treatment may be due to its heterogeneous composition37.
29Observations during testing also suggested that the paint responded differently on the right side of the blast, where it appeared to be more sensitive to heat. Heat exposure was tightly controlled to prevent the gloss from being affected. Such variability may be attributed to uneven mixing of Gottlieb’s materials38.
30The overall effect of the treatment of the black blast is that of a paint film with a faintly mottled surface, which was noticeable through the haze before treatment, with a variable gloss related to Gottlieb’s materials and techniques.
31The treatment was straightforward for the red disc, where the surface accumulation reacted to the heat consistently from one side to the other. Fewer residues appeared to remain after heating on the red disc than the black blast. This area did not exhibit the same sensitivity to brushing as the black paint, so residues that remained after heating were slightly reduced by locally brushing with a sable brush.
32As no visible sign of an accumulation was detectable with the unaided eye or with optical magnification on the white paint, these areas were not treated.
33Gottlieb stated: “I want to express the utmost intensity of the color, bring out the quality, make it expressive. At the same time, I would also like to bring out a certain immaterial character that it can have, so that it exists as sensation and a feeling and that it will carry nuances which are not necessarily inherent in the color, which are brought out by juxtaposition39.”
34The graying effect of the accumulated saturated fatty acids, lead and zinc soaps, and possibly calcium oxalate suppressed the intensity of the red disc and the black blast and undermined the nuances created by their juxtaposition. As the purpose of the painting is for the viewer’s subconscious to ascribe its own meaning to it, even slight color imbalances compromise the result that Gottlieb labored to produce and jeopardize the painting’s reason for being40.
35While a residual amount of the surface accumulation remained on the blast post-treatment, the intensity of the colors and the relationship between them have been re-established. Perhaps most notably, the halo, which had lost its relationship to the discolored disc, now functions to dramatic effect, making it appear to glow. For a painting whose only two shapes are flat forms, the combination of these elements, liberated from the gray veil, creates an animated and atmospheric space, charged with an intensity that is won by Gottlieb’s palette choices.
36The methodology used is described online in the Supplementary information file on OpenEdition Journal:
https://journals.openedition.org/techne/16944
We would like to thank Sanford Hirsch, Executive Director, Adolph and Esther Gottlieb Foundation, for being an invaluable resource throughout the treatment of Thrust and the writing of this article. Also James Squires, Chief Conservator, and Pam Skiles, Senior Paintings Conservator, Clyfford Still Museum, for sharing their experience with using a hot air pencil to treat efflorescence. We thank our colleagues at The Metropolitan Museum of Art in the Department of Paintings Conservation under the direction of Michael Gallagher, and the Department of Scientific Research under the direction of Marco Leona, for their support throughout this work; we are especially grateful to Lucy Belloli, Paintings Conservator (retired) and Charlotte Hale, Walter Burke Conservator, for their examination and treatment of Thrust in 1986.
Flux RSS
