Washington and Lee University (W&L) and the North Carolina Museum of Art (NCMA) have had an ongoing collaboration for several years in which W&L undergraduates and faculty staff work with museum professionals at NCMA. This work benefits both institutions---W&L is able to provide NCMA with useful data about their collection, and W&L students and faculty staff get the chance to work with international experienced art conservators and an outstanding collection of cultural heritage material. In this paper, the portable X-ray fluorescence spectroscopy and infrared imaging (IR) of three works by Francescuccio Ghissi are described in the context of an NCMA exhibition that run from September 2016 to March 2017 to then move on to the Portland Art Museum from March till July 2017. This exhibit unifies the extant works of a dispersed altarpiece and displays a reconstruction project that has recreated a lost panel missing from the altarpiece.
This paper uses a specific example from Washington and Lee University’s ongoing collaborations with the North Carolina Museum of Art to illustrate the benefits of linking W&L’s portable instrumentation (pXRF, multispectral imaging, fibre optic reflectance spectroscopy FORS, IR imaging) with NCMA’s collection of art. Travelling to museums with non-destructive instrumentation works best when conservators have formulated specific questions, such as whether or not degraded smalt is present in a painting 1, what pigment distribution looks like in a painting with significant overpainting 2, are the pigments in objects consistent with their attributed date or anachronistic 3, are pigments subject to degradation present in locations where their degradation makes them difficult to assess 4-6. In this project, non-destructive methods supported a painting reconstruction project.
Francescuccio Ghissi was the most prominent pupil of well-known 14th century Italian painter, Allegretto Nuzi 7. In 2016, the NCMA opened an exhibition that reunites eight of the nine scattered components of an altarpiece attributed to Ghissi (Figure 1): The Art Institute of Chicago 1937.1006 7 8, The Metropolitan Museum of Art: 69.280.1-3 9, Portland Art Museum: 61.32, and NCMA: 60.17.18-20. One panel, the last portion of the altarpiece, has been missing for over a hundred years and there is no record of its subject. An exciting aspect of the exhibition is that it features Charlotte Caspers’ hypothetical reconstruction of the missing panel, created in collaboration with Dr. David Steel, the NCMA’s Curator of
European Art. Caspers wanted as much insight as possible into the materials used to construct the other panels, and understanding the pigments was critical to that investigation. Thus, portable X-ray
fluorescence spectroscopy (pXRF ) was used to analyse multiple spots in the NCMA Ghissi paintings, and InGaAs IR imaging was also performed, along with IR imaging with a modified Lumix camera.
Creation of the Collaboration
Uffelman met Caspers at a conference in 2007 and learned that she had recently completed training as an expert in painting reconstruction. The relationship grew when Caspers graciously agreed to meet Uffelman’s students during iterations of Netherlands course work 10-11. When Uffelman and Caspers met in the spring of 2012, Caspers suggested W&L’s pXRF and other portable instrumentation could assist her with a project she was about to undertake at NCMA, in which she was to create a missing panel from an altarpiece by Francescuccio Ghissi. Caspers electronically introduced Uffelman to Brown; Uffelman knew Brown professionally from his collaborative work with Mass 12. Brown arranged for Uffelman, Soley, and Marsh-Soloway to spend a day at NCMA in September of 2012 to perform pXRF analyses and IR imaging of the three NCMA Ghissi paintings. Soley and Marsh-Soloway were W&L senior Art History majors who had taken Uffelman’s courses on the technical examination of 17th- century Dutch painting, and Soley had also worked with Uffelman on IR imaging of a painting in Portugal, while Marsh-Soloway had performed SEM studies at RCE with Dr. Bill Wei in Amsterdam on antique silver. A W&L Lenfest faculty summer grant provided travel funding to NCMA.
The NCMA Ghissi panels were imaged with a Goodrich InGaAs IR camera (Figures 2 and 3) and with an XNite Lumix LX5 UV+VIS+IR camera whose IR blocking filter was removed (MaxMax). [715 nm or 1000 nm blocking filters (MaxMax) were used to eliminate electromagnetic radiation of shorter wavelengths than those values. This is a system Dr. Greg Smith graciously allowed Uffelman to duplicate from the Indianapolis Museum of Art.] Hints of the underdrawings reported using a vidicon system in 1995 were detected by the InGaAs system on all three paintings. Both cameras revealed many retouchings (e.g., Figure 3) from a 1995 conservation treatment and emphasized the need for careful work with UV illumination for positioning the pXRF (Figure 4).
pXRF analyses of the Ghissi panels were straightforward.
Strong calcium signals were present in all spectra, and sulphur signals were often distinguishable from lead M lines in areas where no lead white or less lead white was used---all of this is consistent with the Italian use of gypsum in gesso layers for panel preparation. The lead white was confined to specific colours which were tinted with white. Caspers noted that there is a light-blue underpainting underneath St. John’s drapery, this would be lead white and azurite. Also, the ‘shot coloured’ drapery of Aristodemus is underpainted with lead-tin yellow. There is no lead present in the red samples for the red drapery of the figure with turban, just Mercury. Copper was detected in areas of green and blue. Iron was detected throughout the spectra, but was, not surprisingly, strongest in areas that had an ochre appearance. Interestingly, no manganese was detected at any appreciable level in any of the brown background areas. Mercury from vermilion was responsible for red areas. One bold yellow colour was clearly attributable to lead-tin yellow.
For Caspers and Brown, the work provided confirmation of pigment hypotheses based on optically examining the paintings. Sightings of K, Ca, S, Fe, Cu, Sn, Hg, and Pb were useful, and the possible assignment of Al in lake areas and the possible assignment of P in a black area were suggestive. The absence of Mn in any abundance in the brown areas was relevant to pigment choices to be made in the reconstruction. Charlotte Caspers’ reconstruction has been the reference for the virtual rejuvenation of the Ghissi Altarpiece by the Mathematics Department at Duke University using novel colour mapping techniques and a crack removal algorithm—the virtual altarpiece, approximating what it might have looked like when first completed in Ghissi’s studio, has been displayed on a large monitor in the exhibition. Charlotte Caspers’ panel was virtually aged using colour mapping techniques and an algorithm developed to add age cracks and has been displayed with the altarpiece.
Uffelman learned considerably more about the art history of 14th- century Italian painting and learned about Dr. Ingrid Daubechies’ work in mathematical brush stroke analysis at Duke 13, since Daubechies also participated in the Ghissi project. Uffelman could contextualize Daubechies’ work with efforts his
students learned in The Netherlands from Eric Postma 14. He saw that NCMA not only possessed a first class collection of old master paintings, which he already knew, but that the conservators were tremendously collaborative, friendly, and helpful.
For Soley, and Marsh-Soloway, visiting NCMA enabled them to further practice their knowledge gained from Uffelman’s two courses on the technical examination of 17th- century Dutch painting. Carefully comparing IR images taken between 715-1050 nm versus 1000-1050 nm versus 900-1700 nm was valuable. This was especially true for Soley, who had used 715 -1050 nm and 1000-1050 nm to image a painting in Portugal, but could not use the InGaAs camera in Portugal due to US ITAR export restrictions. [n.b., InGaAs camera users MUST familiarise themselves with US ITAR regulations to avoid serious legal trouble with US law!] In the pXRF analyses, the contrast of gypsum used in Italian gesso versus the chalk used with glue in northern Europe was obvious to them. They practiced their pigment identification skills and reviewed the sum peak and escape peak pitfalls possible in spectrum analysis, as well as peak interference issues (e.g., the lead M lines versus the sulphur K lines). Soley, currently performing PhD work in art history at Columbia adds, “When encountering an artwork (in my research or otherwise), I am now in the habit of approaching it as a material object rather than merely an image, and thus take advantage of the physical indicators that reveal a plethora of information about the artwork, artist, historical context, etc. This work expanded my undergraduate research on the Viseu Altarpiece with Dr. Andrea Lepage, where we used IR imaging to gain a clearer picture of the panel in question.” The NCMA visit allowed Marsh-Soloway to build upon expertise gained with Bill Wei at RCE. Both experiences also drove her Senior Art History thesis, Colliding Cultures at Chora Church. This thesis examined the hidden images of both Christian and Muslim iconography. Marsh-Soloway remains passionate about Art History. The experiences helped build a life-long curiosity to look beneath the painting and never stop questioning. Today she works for the Coca-Cola Scholars Foundation which provides scholarships for high school seniors to pursue their college careers in all fields.
NCMA Collaboration Expands
Since 2012, W&L has made yearly research visits to NCMA, with both institutions sharing the travel costs. In addition to pXRF and IR imaging, recent efforts have utilized W&L’s new fibre optic reflectance spectrometer (FORS) and multispectral imaging system based on Dr. John Delaney’s design at the National Gallery of Art (Washington, DC) 15. In addition to working with the full NCMA conservation staff, Uffelman and his students have been able to learn Dr. Warren Warren’s cutting edge conservation science methods via collaboration with Tana Villafana 16. All of this work feeds not only into Uffelman’s courses on the technical examination of 17th century Dutch painting, but also into General Chemistry courses for W&L science majors, and, crucially, into NSF Chemistry in Art Workshop material taught to university faculties from around the US.
We gratefully acknowledge the NSF for funding W&L’s instruments reported here (NSF MRI grants CHE-0959625, and CHE-1337481). W&L University Lenfest funds permit travel to NCMA. Useful conversations with Dr. David Steel, Dr. Greg Smith, Dr. John Delaney, and Dr. Kate Dooley are gratefully acknowledged. Related research at NCMA not reported here with Tana Villafana, G. Perry Hurt, and Noelle Ocon is gratefully appreciated and will be the subject of future writing.
For the full list of references and images please download the pdf version of this articlein News in Conservation, Issue 58, April 2017