Micro-Spatially Offset Raman Spectroscopy (Micro-SORS) for Research in the Conservation of Cultural Heritage

Micro-spatially offset Raman spectroscopy (micro-SORS), is a new method for the chemical analysis, identification, and assessment of decay extent of cultural heritage objects. The ability to measure the chemistry of multilayered, micrometer-scale structures is providing the answers and knowledge base from which to catalog and construct artwork histories and the decay processes of priceless cultural treasures. Before the development of micro-SORS, subsurface molecular analysis of valuable objects was performed using an invasive and destructive cross section approach, which for obvious reasons is completely unsatisfactory. The use of micro-SORS provides a breakthrough in non-invasive and non-destructive analysis of the subsurface layers of artworks and is a key development in the understanding of conservation science for cultural heritage objects. In this interview, Dr. Claudia Conti, a senior researcher at the Institute of Heritage Science (ISPC) of the Italian National Research Council (CNR), talks about her work in micro-SORS. Conti is the winner of the 2020 Craver Award presented by the Coblentz Society, to be given at the 2020 SciX conference for her Raman research. The SciX conference is scheduled for October 11-16 in Sparks, Nevada. This interview is part of a series of interviews with winners of awards presented at SciX.

You have worked with SORS related research for several years and were able to demonstrate its utility for analysis paintings (1). Specifically, you were able to penetrate the topmost highly turbid layers of paintings to measure and analyze the pure Raman spectra of paint sublayers completely obscured by the paint overlayers. What prompted you to explore this approach to surface layer analysis? Since this work, do think that SORS has lived up to your expectations, or from your perspective are there still further instrumentation or method developments that are required to make this an optimized analysis method?

After my PhD in Materials Engineering at Milan Polytechnic (2010), I searched for new innovative solutions to face critical needs in conservation of cultural heritage. My supervisor, Prof. Giuseppe Zerbi, suggested that I attend the ICORS conference in Boston (2010), where I became aware of a novel tool developed by Prof. Pavel Matousek that enables Raman sensing deep inside turbid media on a macro-scale. Right from the beginning I realized the strong impact that this method could have in conservation sciences, permitting the non-invasive molecular identification of pigments and decay and conservation products through the stratigraphy volume of paintings, frescoes, painted sculptures, and so forth. In fact, the knowledge of the subsurface composition of artifacts is critically important in conservation sciences: Beyond the surface there are a number of hidden compounds, selected by the artist or formed during the decay processes or applied over the centuries for conservation purposes. Scientists are required to provide the conservators and art historians with the elucidation of these puzzles to facilitate comprehensive knowledge of these objects to enable proper planning of the conservation actions; to reach this goal we need to develop non-invasive methods, one of the most vital research activities of the last decades in our field.

At the beginning of the collaboration with Prof. Matousek, macro-scale SORS did not live up our expectations because it did not work with our samples. Then we tried to find an alternative way to discriminate between painted layers, extending SORS to the micro-scale by combining SORS with microscopy. In fact, conventional SORS is only applicable to relatively thick layers (on the order of millimeters or greater), whereas the thickness of typical overlayers of the concealed substances in artworks is an order of magnitude or two lower. After almost six years of micro-SORS research, I would say that this method exceeded my expectations because we have found a wide range of successful applications both within as well as outside the field of cultural heritage. Based on these promising achievements and also taking into consideration existing limitations, we are now focusing on further technical and methodological developments to extend the non-invasive molecular subsurface investigations to more complex situations, for example, the detection of the diffusion of conservation treatments into decayed substrates and also the micro-SORS measurements in-situ with portable devices.  

What are the real advantages of using micro-SORS for probing through thin painted layers compared to other analytical methods, including confocal Raman microscopy and conventional Raman spectroscopy?

In-depth probing using conventional backscattering confocal Raman microscopy is applicable only to (semi)-transparent materials, since in turbid media the diffusion of the light in the sample leads to subsurface Raman signal being often overwhelmed and obscured by surface Raman signals and the deep photons cannot be discriminated by conventional confocal optics. These deeper generated Raman photons are subject to a large number of scattering events and consequently end up spread further apart from each other as they reach the surface. Cultural heritage materials are mainly turbid; transparent varnishes are an exception to this, and these can be scanned with conventional confocal z-scanning. Pigments, decay products, and other turbid compounds, as well as tissues and powders, need to be scanned with micro-SORS, which is able to acquire and discriminate the contribution of the sublayer from that of the top layer even when dealing with highly diffusive materials. This is achieved by enlarging the illumination and collection areas or by collecting the Raman signal from an area which is spatially offset from the laser illumination point. So, in other words, before the development of micro-SORS the only feasible way to study the stratigraphy of cultural heritage objects with Raman involved destructive cross-sectional analysis, which should be avoided in this context.

Other analytical methods have been developed for the knowledge of the subsurface of cultural heritage materials by non-invasive means, such as multispectral, hyperspectral and Terahertz imaging, X-ray radiography and infrared reflectography, scanning macro and confocal X-ray fluorescence, and optical coherence tomography. The specificity of micro-SORS is the direct retrieval of the molecular composition of organic, inorganic, crystalline and amorphous compounds, not fully accomplished by these alternative techniques.

You have applied micro-SORS to artifacts of historical significance originating from Italy and dating from the medieval to contemporary periods (2,3). What discoveries you have made on rare artifacts using this Raman technique?

First of all I would highlight that our research produced a methodological discovery, going beyond the specific results achieved on the analyzed artworks. We were able to “see” through the surface in a non-invasive way and get information about the presence of hidden pigments, decay products, and preparation layers spread over different substrates such as paper, plaster, and stone. To name just a few examples of discoveries in artworks, we have analyzed prestigious Renaissance terracotta sculpture of the “Sacred Mounts” (UNESCO World Heritage sites) which have been repainted several times over the centuries, thus exhibiting a complex stratigraphy made of a number of superimposed micrometric layers.

Conventional confocal Raman microscopy on the intact fragments identified only the pigments on the sample surface and was prevented by the interfering surface signal from seeing the composition of individual layers below the surface; in contrast, a micro-SORS analysis on the same intact fragment allowed the presence of pigments as separate (deeper) layers to be ascertained. It is also important to be aware of the method limitations highlighted by the parallel cross-sectional analysis carried out to validate the micro-SORS results: A number of compounds may not be “visible” due to their weak scattering cross section or fluorescence, or due to their depth being deeper than the accessible depth of the technique. However, micro-SORS has demonstrated the possibility of recovering the pigment information from below the surface in a non-destructive way and thus reconstructing, even partially, the sample stratigraphy.

A second example concerns a publicity card from the 19^th century that we placed directly under the Raman microscope objective without any sampling. We performed  mapping of selected areas of the card and in all the micro-SORS sequences the Raman signal of a pigment, white lead, which was not visible to the naked eye because it is covered by the external pigments, emerged from the subsurface. This outcome indicates the presence of a layer, on top of the card paper and below the external layer, applied for the preparation of the painting. Moreover, in a red part of the card, above the most internal white lead layer and below the external red layer made of cinnabar, we observed the presence of another, intermediate layer made of Prussian blue. This is important information for the understanding of artist’s technique and for conservation purposes and we achieved this in a non-invasive way.

You were able to expand the use of micro-SORS from cultural heritage objects to a variety of other sample types where detailed non-invasive analysis of diffusely scattering turbid layers were involved, including polymers, seeds, and paper (4). What was most surprising to you relating to technique or sampling requirements as you moved from cultural heritage objects to other materials? Are you able to image or create images of surfaces using micro-SORS?

One of the most exciting experiments outside cultural heritage was the investigation by micro-SORS of the structure of the wheat seed, research carried out in collaboration with the Norwegian Institute of Food. The identification of diseases at an early stage is very important for the grain producers, therefore the use of rapid and non-destructive methods for characterization of inner grain components is essential. The seed has an external micrometric envelope made of ferulic acid and an internal kernel containing starch. Micro-SORS experiments made it possible to unequivocally discriminate the Raman spectrum of the envelope from that of the starch, without touching the seed.

As for the second question, we carried out an exciting experiment concerned the reconstruction of images hidden by a turbid overlayer. We simulated in a laboratory environment real situations encountered in cultural heritage that deal, for example, with hidden paintings vandalized with graffiti or covered by superimposed painted layers or whitewash. One of the mock-up specimens was prepared by painting the letter “F” using phthalocyanine blue over a whitewashed background; two pieces of paper colored with pink ink were used as the top layer. Conventional mapping showed only the Raman signal distribution of the pink ink, whereas with the chemical image obtained with micro-SORS mapping we were able to “read” through this turbid overlayer  and reconstruct the graphic symbol of the letter “F”.

In one research application of micro-SORS, you analyzed medieval polychrome sculptures from the Parma baptistery and Ferrara cathedral (5). What specifically was learned from these objects?

In a sculpture of the Ferrara cathedral portal, painted with a red hue, we discovered that, under the external red layer made of cinnabar, a second hidden red layer is present and made of minium (a bright orange red pigment made of red lead), indicating that the external cinnabar layer is most likely a repainting. In fact, it is well known that the portals of European cathedrals could have been subject to more than one decorative phase, probably whenever the color started to look shabby due to the outdoor environment and natural degradation, and with micro-SORS we had direct analytical evidence of this historical information. Moreover, in the same sample we have found the presence of two salts, related to decay processes; micro-SORS sequences and the related spectral subtractions allowed us to ascertain that they are mainly diffused on the surface and within the cinnabar layer. This is an outstanding outcome since it informs us about the migration and the re-crystallization processes of soluble salts within the layers, with a consequent effect on the mechanical stress of the entire system.

What were some of the key challenges you encountered during your research? How did you overcome them?

The first major challenge emerged after the development of the first micro-SORS setup, called defocusing. This is the most basic and easily deployable micro-SORS variant but it is not as effective as a full micro-SORS set-up because there is not a real separation between laser illumination and Raman collection areas. We realized that a more sophisticated variant could be necessary to investigate complex and heterogeneous materials as encountered in cultural heritage. With limited funds we focused our research for more than one year on searching for the best technical solution; these research efforts ultimately lead to the successful modification of a conventional micro-Raman, able to separate the laser and Raman paths. In this way we fully extended the macro-SORS to the microscale.

One of the major obstacle in Raman spectroscopy is the fluorescence emission from the sample, especially when dealing with cultural heritage materials. We have faced a critical situation when the top layer is fluorescent and we needed to recover chemical information from the subsurface. It was tricky because it is well known that fluorescence signal can be much more intense than Raman, and furthermore, in our case the target compound was covered by a turbid overlayer. The challenge was very high but we applied the analytical capability of micro-SORS in the most advanced variant and it worked; at the imaged position the Raman spectrum was overwhelmed by the fluorescence of the top layer, whereas at a certain value of spatial offset the signal of the subsurface kept increasing, emerging clearly from the fluorescence background.

What would you consider to be the most meaningful contributions of your research work? What discovery has most impressed you?

Recently, we obtained an outstanding result: We analyzed with micro-SORS two 16^th century panel paintings by Marco d’Oggiono, using a modified portable instrument directly in-situ. This is one of key goals of our research since the beginning, namely the development of a portable device able to carry out micro-SORS without any restrictions on the dimensions of the artwork or its location. In fact, cultural heritage objects cannot be easily moved and transferred to our laboratory for analytical purposes; we have to find a way to analyze them where they are (museum, churches, historical buildings, and so forth). In this context portable micro-SORS can be considered a very important development, providing a further step toward establishing a more powerful multi-analytical approach, essential when dealing with complex materials as encountered in the cultural heritage field. Portable micro-SORS could be relevant also outside the art field, as in forensic science and solar cells analysis. Further optimizations of our current portable device are still required to improve its stability and the sensitivity; this work is in progress.

Would you share with our readers your philosophy of how to work in a research environment?

From my experience I learned that curiosity triggers the research. This may seem trivial, but in practical terms I noticed that it is not easy to keep our minds open to new inputs, accepting the risk of going beyond our comfort zone and far from what we already know. Making curiosity prevail over habit is one of the most prolific exercises we have to perform every day; how does it help me? To find researchers with whom to share passion for knowledge and innovation, someone to learn from, and create a research team available to overcome limitations and barriers. I will never forget when my PhD supervisor, Professor Zerbi, pushed me to broaden my horizons presenting my analytical results at one of the most important spectroscopy conferences in the United States when I still was a PhD student. I felt inadequate for that, but I went to the conference, and I discovered that curiosity and boldness were an integral part of my work. The power of our micro-SORS research is the fruitful and unceasing collaboration between CNR Italian and UK teams, providing the necessary incentive to face the numerous challenges of the research.

References

(1) C. Conti, C. Colombo, M. Realini, G. Zerbi, and P. Matousek, Subsurface Raman analysis of thin painted layers, Appl. Spectrosc. 68(6), 686–691 (2014).

(2) C. Conti, C. Colombo, M. Realini, and P. Matousek, Subsurface analysis of painted sculptures and plasters using micrometer‐scale spatially offset Raman spectroscopy (micro-SORS), J. Raman Spectrosc. 46(5), 476–482 (2015).

(3) A. Botteon, C. Colombo, M. Realini, S. Bracci, D. Magrini, P. Matousek, and C. Conti, Exploring street art paintings by microspatially offset Raman spectroscopy, J. Raman Spectrosc.49(10), 1652–1659 (2018).

(4) C. Conti, M. Realini, C. Colombo, K. Sowoidnich, N.K. Afseth, M. Bertasa, A. Botteon, and P. Matousek, Noninvasive analysis of thin turbid layers using microscale spatially offset Raman spectroscopy, Anal. Chem., 87(11), 5810–5815 (2015).

(5) C. Conti, A. Botteon, C. Colombo, D. Pinna, M. Realini, and P. Matousek, Advances in Raman spectroscopy for the non-destructive subsurface analysis

Claudia Conti

Claudia Conti is a permanent researcher at CNR-ISPC (Institute of Heritage Science), unit of Milan. Conti obtained a degree in geological sciences at the University of Perugia in the field of archaeometric problems, and a PhD in materials engineering at the Politecnico di Milano working on the analysis of chemical stability of calcium oxalate in conservation and in historical films. Conti has established expertise in the area of advanced applications of vibrational spectroscopy to material analysis, particularly in the field of cultural heritage. In 2012, she established a major collaboration with the Rutherford Appleton Laboratory (UK) to explore a novel tool, Spatially Offset Raman Spectroscopy (SORS) that enables Raman sensing deep inside turbid media on macro-scale. This collaboration lead to the successful transformation of macro-SORS conceptually to the area of cultural heritage and the demonstration of a new SORS variant, micro-SORS, capable of resolving micrometer thick layers of paint for the first time. Outside cultural heritage, the most relevant impact of her research on micro-SORS concerns the biomedical area, polymer, paper and food fields.