CHAPTER 1: Introduction to Analytical Strategies for Cultural Heritage

The aim of this book is to review the analytical strategies used in the past and to describe new methods proposed more recently for the study of the materials used in cultural heritage assets. Both movable (artworks and archaeological items) and immovable (mural paintings, archaeological sites, historical built heritage) cultural heritage assets with significant importance have been key examples of the roles of different professionals and how a multidisciplinary view is demanded nowadays.

First, the analytical techniques used to ascertain the elemental and molecular characterization of such materials are described in Parts 2 and 3, respectively. Subsequently, the different multianalytical strategies available to characterize materials (classical and modern) are described in Part 4 for different important examples of cultural heritage assets that people all around the world can appreciate.

This structure of the book is intended to consider first the individual analytical techniques and coupled methodologies, both at the laboratory level and in the field, and then the specific problems related to the different materials involved (metals, pigments, binders, inorganic mortars, stones, modern organic materials, etc.) that require attention when dealing with cultural heritage assets.

Taking into account that the degradation of cultural heritage materials starts at the moment the artwork is finished and exposed, each chapter of the book considers both the nature of the original materials and the degradation products that can be produced as a function of the reactivity of such materials with the surrounding environment. When possible, chemical modelling is included as a tool to ascertain the sources of impacts that degrade the materials used in the manufacture of the artwork, archaeological object or historical building.

The field of science related to cultural heritage (CH) is vast and complex, encompassing analytical and physical chemistry, biology, engineering, materials science, etc. Sampling, quality assurance, simulation/modelling and chemometrics are usually far from the capabilities of restorers and daily practitioners. Fortunately, analytical chemistry is nowadays a science that is playing a more important and critic role in assisting the work of other professionals in the field of CH.

Analytical science can now provide answers to questions about the conservation state of CH assets that are under study or rehabilitation. This is mostly done by merging the characterization of original and deterioration-related compounds with the aim of ascertaining the decay pathways, because most of these degradations are due to chemical reactions between the original compounds and chemicals present in the environment surrounding the artwork, the archaeological remains or the historical building under study.

To obtain the required information to reach these goals, we as analytical chemists have an important role in the multidisciplinary teams involved with a project of rehabilitation or conservation. We must impress upon other professionals that we are not in a project to perform analyses; unfortunately, there are many restorers who reduce the problem to executing a number of analyses to be included in a report, and this is starting to be mandatory in practically all the administrations in charge of CH preservation. This dynamic must be broken, changing the past concept of “perform standard analyses” to a more realistic “design dedicated analytical strategies to solve the particular problem”.

We do not perform chemical analyses: technicians do this. Our role as analytical chemists is first to define the extent and nature, together with the other professionals, of the “problem to be solved” in the project. If the project is complex, we must encourage defining small, partial objectives that are more easily affordable. Once this has been clearly stated, we start our specialized work as analytical chemists.

First, we must select the most suitable analytical methodologies to obtain the knowledge required to reach each partial objective. If the small problem of the partial objective is similar to others reported in the literature, we can take advantage of that past expertise developed by other team(s). However, often this do not happen and we cannot find any previous similar problem described with sufficient detail in the literature. For such cases, we need to design dedicated analytical methodologies that must be validated before starting to apply them to the CH objects. This is important, because if we have success we will contribute to the advance of knowledge if we publish our new development and the results obtained.

At this stage of the book, an important aspect should be emphasized. Even if one belongs to a private institution, the projects around CH conservation issues are partially, if not totally, funded with public money. Part of that money will certainly come to the author's institution and therefore we must publish and disseminate our developments, findings and results. These are not property of the author's institution unless the institution has provided all the expenses with its own non-public resources. Fortunately, the number of private institutions that publish their results on CH projects is increasing year by year, demonstrating their commitment to society.

Our starting point in any CH project, after the problem to be solved has been clearly identified and the partial and global objectives have been defined, is to think about the most suitable set of analytical methods to be used, that is, what will be the most appropriate analytical strategy to achieve the global objective?

Any analytical methodology requires the fulfilment of different steps, after the definition of the problem to be solved, namely:

• homogeneity of the sample(s) under analysis;

• representativeness of the samples being analysed with respect to the whole object;

• accuracy and precision required;

• constraints around the sample to be analysed;

• selection of a suitable method;

• quality control and quality assurance of the process;

• critical analysis of the results, including their overall uncertainty;

• chemometric analysis of the data and chemical modelling.

These are not always taken into account by non-analytical chemists using modern analytical instrumentation “at the touch of a button”. This can be observed in some papers published in non-analytical chemistry journals and in some manuscripts submitted to such journals that cannot go ahead in the review process owing to the lack of descriptions of adequate analytical procedures.

In CH, there is a common concept of homogeneity: all the artworks, objects and buildings are heterogeneous. For example, painted artworks (easel paintings, wall paintings, rock art paintings, historical wallpapers, polychromed sculptures, etc.) have several layers intentionally applied by the artist during their production. Buildings have an external patina resulting from the action of the environment on the original stones, mortar, balconies, roofs, etc. Archaeological objects also have patinas due to the action of the burial (soils or sediments) on the original materials.

If the object or asset has been subjected to past restoration processes, new layers are present on top of the original ones. Moreover, if the decay due to those interactions with the (indoor and outdoor) environment, where the object or asset is located, has progressed without any preventive action being taken, new compounds can be formed as sub-efflorescences (new chemical compounds inside the pores of the bulk) or efflorescences (new chemical compounds on the surface of the bulk). The detail of the wall painting shown in Figure 1.1 reflects signs of vandalism, detachments and the effect of past restoration.

To sum up, if the CH element has suffered a natural impact (earthquakes, fires, floods, etc.) or a human impact (industrial processes, bad agricultural practices, vandalism, graffiti, stolen, etc.), new chemical compounds should be expected to be present. Natural impacts have been more and more frequently observed in recent years and will continue to increase as climate change progresses, especially for those CH elements located near rivers or the sea. However, human impacts are also increasing, especially dangerous being the consequences of bad agricultural practices where the excess of chemicals used is passed to rivers and ground waters, ultimately reaching the foundations of historical buildings and transporting soluble acids, cations and anions to the walls by capillary processes.

All of these complex processes affect the homogeneity of CH objects and their samples. Visual and photographic inspection, together with the past expertise of the professionals involved in the project, will give us a first idea of the inhomogeneity, which can be considered at the end as the sum of different homogeneities (each layer in a painting could be homogeneous but the sum of the layers gives us the inhomogeneity). This first analysis conditions the number of individual samples that should be analysed to characterize the CH element properly. Also, this first analysis must be made at the time of defining the problem to be solved by the project, because the different homogeneities will give us the number and extent of the particular objectives.

Are you sure that your sample represents the overall problem? This is the classical question in the field of analytical chemistry. However, this may not be questioned by other professionals in the field of CH because “a sample is a sample and represents a particular universe that is systematically observed”. We have observed this controversy too many times, but our responsibility as analytical chemists is to be aware about representativeness.

If we do not take representative samples, the overall conclusions that we obtain at the end of the project will not cover all the different degradation processes that are really present in the CH element as a whole. As an example, Figure 1.2 shows three public sculptures made of CorTen steel, two exposed outdoors and one displayed inside a museum, which suggest a high degree of homogeneity, guaranteeing the representativeness of the samples. However, the first analyses performed with portable instrumentation showed in all three cases the presence of different patinas as a function of orientation, which led to inhomogeneous surfaces at the millimetric level, conditioning the representativeness if sampling does not cover all the different surfaces.

To guarantee representativeness, we must define a number of actions that must be included in the overall set of quality control and quality assessment actions of the project, because the absence of representativeness affects both the accuracy and precision of the final results.

The use of the terms accuracy and precision in analytical science generally refers to quantitative data, but in the field of CH not only quantitative data are used but also qualitative information about what compounds are present in a given layer of a multilayered object under study. For quantitative data, the treatments (statistical tests) for accuracy and precision are the same as in other fields where analytical chemistry is involved and the same rules should be used to select one analytical method or another as a function of the required accuracy and precision, something that is set in the step of “defining the problem”. In this section, we must focus on the qualitative information relating to CH materials.

Usually, qualitative work is based in the interpretation of the spectral responses of one or several spectroscopic analytical techniques, so how can we define accuracy and precision? The answer is not unique, because it depends on the number and kind of spectroscopic techniques used. We are going to consider some possibilities, proposing how we can define and apply accuracy and precision. These proposals are based on the experience we have accumulated in the last 20 years working in the field of analytical chemistry for CH. Accuracy must be understood as certainty in the presence of a given element or chemical compound. Precision must be understood as the variability in the form and position of the bands/peaks characteristic of a given element or compound. Depending on the spectroscopic techniques used, we could have three possibilities.

For the case of using only one spectroscopic technique, a chemical element or compound is considered to be present if the unknown spectrum has at least two characteristic bands/peaks of the element (for techniques of elemental analysis) or compound (for techniques of molecular analysis) and those two signals appear more than five times in the spectra collected from different parts of the sample.

For the case of two spectroscopic techniques sensitive to the same chemical target, the same criteria as described above apply for both accuracy and precision. For example, as infrared (IR) and Raman vibrational spectroscopic methods are sensitive but complementary to the nature of the chemical compounds, we must extend the two previous requirements (to observe at least two bands for each compounds and in five different spectra) to a third requirement: the chemical compound should be identified by the two spectroscopic techniques. As another example, as laser-induced breakdown spectroscopy (LIBS) and X-ray fluorescence (XRF) spectrometry are sensitive to the presence of particular chemical elements in the sample, we must consider that an element is present if both techniques detect its presence.

For the case of two spectroscopic techniques sensitive to different targets, one for elements (LIBS or XRF spectrometry) and the other for chemical compounds (IR or Raman spectroscopy), the identification by the two techniques must be compatible. For example, if the technique sensitive to elements detects calcium, iron and sulfur, the technique sensitive to chemical compounds (a vibrational spectroscopic technique) must identify species that contain one or two of those elements simultaneously present in the vibrational spectra [e.g. anhydride (CaSO₄) + haematite (Fe₂O₃) or calcite (CaCO₃) + rozenite (FeSO₄)].

We must consider that not all analytical techniques are equally sensitive to the same element/compound, to the matrix effect (spectral interferences) and to the amount present in the sample. Therefore, when analysing the spectra, all these constraints in the spectral responses must be taken into account. To aid in this task, the chapters in Parts 2 and 3 describe such spectroscopic techniques that are usually employed in CH to identify the presence of chemical elements and compounds.

In the current practices in CH diagnostics, sampling is a concept that has been revised, and readapted, to the new possibilities opened up by the use of portable/handheld devices.

The use of laboratory techniques requires the physical extraction of a given amount of material. This is the classical sampling process. That amount is conditioned by the previous visual inspection of the area that must be sampled (e.g. metal with rust detachments, layered polychromy, patina on a rock or in a building, etc.). Sometimes, after analysing that first collected sample, a new sampling must be performed again because we suspect that the initially collected sample is not representative of the whole. Here, a conflict between the analytical chemist and restorers could arise because one needs more sample to be taken but the restorer is reluctant to remove material from the CH object.

The advent of portable instrumentation has helped considerably in the particular field of CH diagnostics because the physical extraction of material from an object is not required in order to perform the analysis. When we are using a portable device, a sample is really the spot analysed by a portable spectroscopic technique; the size of the spot depends on the spectroscopic technique and on its physical implementation in the given portable instrument, it being possible to have setups with spot sizes from 1 cm to 50 µm in diameter. Figure 1.3 shows two studies performed using portable Raman spectroscopy at the micrometric level, on Roman paint panels deposited in the Naples National Museum of Archaeology and on the wall of a house in Pompeii.

Fortunately, we have a portable solution for most of the spectroscopic techniques used in the characterization of CH materials, which has opened up an enormous range of possibilities for performing a project around the CH object under study. Let us consider an example. In 5 h we can collect (physically remove) fewer than 20 samples, if restorers allow us to effect that “damage” to the CH object. In 5 h we can measure more than 100 spots with a portable instrument, hence the sampling capability of portable measurements is 5–10 times higher than the classical sampling performance. Moreover, in classical sampling we must then go to the laboratory to perform the analysis, whereas with the portable instrument we perform sampling and analysis in the same event.

As seen, the productivity of in situ analysis with portable instruments is much higher than that of the classical approach of collecting samples and analysing them in the laboratory, not only with respect to the higher number of samples (spots) that portable devices can analyse but also the overall time required to obtain the analytical signals for interpretation and treatment in the office.

However, portable instruments are suitable only for the analysis of surfaces owing to their low penetration capacity. Hence, if the object under analysis is layered, we will need to perform both in situ and laboratory analyses. For such cases (when in situ analysis cannot solve the overall problem), our expertise suggests performing as much as possible of the in situ analysis to detect the different areas as a function of different spectroscopic responses from the portable spectrometers; the deep analysis of that information gives us the maximum number of different homogeneous areas/surfaces that should be sampled, each sample being homogeneous in its composition.

With such working methodology, we guarantee both minimal damage to the CH object and the collection of homogeneous samples, fulfilling the constraints commented on in Sections 1.3.1–1.3.3 and optimizing the costs/results ratio to obtain the overall and complete diagnostics prior to any further intervention.

The selection of a suitable method to solve a particular problem must follow the standard procedure of any analytical methodology after the previous step of problem definition. We recommend taking into consideration also the suggestions given in Sections 1.3.1–1.3.4 that are more directed to the special field of CH research.

The chapters in Parts 2 and 3 provide information about the analytical capabilities of different techniques to aid the analytical chemist in the selection of the most appropriate method(s) for solving a given problem. The minimum requirement for in situ analysis is to work with one technique for elemental analysis and another for molecular identification. This is the case shown in Figure 1.4, where portable Raman and XRF instruments are used on the same Neolithic painted panel inside a cave.

If the knowledge of the analytical chemist is insufficient, numerous dedicated works are available in the extensive literature on analytical methods for CH. Our experience suggests that in more than 70–90% of cases, previously published studies contain the required information to select the most suitable method for our problem.

Any analytical methodology must include dedicated tasks to ensure the internal quality control of the good performance of instruments (both portable and laboratory). Validation of methods, calibration of instruments, access to high-quality databases for qualitative interpretation of spectroscopic signals, calibration of analytical signals for quantitative purposes, data treatments, estimation of accuracy and overall uncertainty are procedures that must always be implemented.

Moreover, the standard procedures for quality assessment must be included in the work (purchasing goods and consumables, collection of samples and transport, storage in suitable conditions in terms of the nature of the samples, etc.) as in any other analytical laboratory working in different fields.

Also, the procedures of internal quality control and quality assessment must be documented for all of the steps in the analytical strategy designed to solve the particular problem and the working conditions in the field and in the laboratory. This ensures traceability, which is extremely important if some unexpected problems arise in the project. It is quite common to hear in these multidisciplinary projects something like “the problem comes from the analysis”. To refute this, we must have all the documentation ready to show to everybody and to demonstrate that our results are accurate and well dimensioned for the needs of the project.

One of the important problems we have detected in different projects concerning CH materials is the a priori assumption of the presence of some expected compounds. This “knowledge” can condition the interpretation of the results and it should never happen. We must interpret the information as if we did not know about the presence of any element or compounds in the samples being analysed.

Moreover, we have encountered past conservation interventions that took the decision to use certain compounds because the expertise of the restorers suggested a given, but not real, composition in the artworks. Also, to interpret why a particular restoration compound is present over the original paint layers sometimes is not easy, unless you consider that maybe that past restoration was performed without a previous analysis of the object to be restored. Unfortunately, the number of past interventions made without previous analysis is high because that was the standard practice of some private and public restoration teams.

When dealing with the analysis of results, we must be aware of the requirements of representativeness, homogeneity, accuracy and precision commented on above, together with the requirement to document all the processes related to the analysis of the qualitative and quantitative data. To evaluate the overall uncertainty, it is suggested that the standard procedures recommended by the European Association for Quality Assurance in Higher Education are followed.¹ 

Chemometrics must be considered as another analytical tool, especially if chemical information is obtained from several analytical techniques or if sampling has been done so extensively that some relationships can be contained in the data (not necessarily quantitative values). The searching of that information is the aim of chemometrics. We can detect what (a chemical element or compound) is correlated with what, or how many variables really affect the experimental data, or which data belong to the same family (clusters) of data or how samples are grouped.

To apply chemometrics, a minimum number of data/spectra is required. All the chemometric methods are statistical methods with defined minimum degrees of freedom to obtain representative results. For example, if we have five samples with 12 concentration values of elements, nothing can be done; for such situations, we need more than 20 samples to have at least six degrees of freedom in the statistical computations, The results of chemometric analysis can be used to define, for example, (a) what the different sources of impacts suffered by the CH object or asset are, and/or how they interact with the original materials, (b) the past trade routes of commerce, (c) how many quarries were used to construct a historical building, etc. That is, the analytical information gives data to characterize the object and/or to diagnose its conservation state, but also contains information on the past history of the object, on its production process, on the different environmental impacts in certain years, etc.

With chemometrics we go further than the simple characterization, but if we incorporate chemical modelling we can go even further. Most of the degradations described for CH objects can be explained by the chemical reactions between the original materials and the different chemicals contained in the environment that surround the object. By chemical modelling we can define a known chemical that reacts with an original compound (experimentally measured) to produce a new decayed compound (that has been detected in the analytical process). Thus, we have the tool to assign possible candidates that have been interacting (chemically) with the original materials to produce the decayed compounds.

Usually, those candidates cannot be measured at the time of developing the project because they were acting in the past, but we can define their necessary presence to explain the transformations experimentally observed from the original compounds to the decayed ones. In other cases, when the impacts are still acting on the CH objects, we can assume the possible sources that emit such necessary compounds that explain the degradations observed and then define monitoring campaigns around the CH element under study to clarify if the theoretical sources are effectively the ones emitting the chemicals that cause the damage observed.

This book does not contain all the methods and techniques that are available to be used in studies on CH objects, assets and sites. Rather, we have selected the most useful ones and the most suitable to conform with the suggested rules concerning CH that are coming in the near future.

Part 2 of the book describes non-destructive methods (and techniques) for the elemental characterization of CH materials. First, the different X-ray fluorescence techniques (energy-dispersive XRF, wavelength-dispersive XRF, total XRF) are described to characterize the elemental composition of artworks and materials of CH, including point-by-point and imaging techniques, with laptop and portable setups. Subsequently, laser-induced breakdown spectroscopy (LIBS) is described as it is gaining more and more importance in the field of CH; the advantages of this technique with regard to other more conventional techniques, the problems related to the method of conducting the analysis in the field and the possibility of quantifying the lighter elements (C, O, N, P and S, among others) are discussed. Finally, synchrotron radiation (SR) sources for characterization at the microscopic level are described, because the increasing number of contributions that have appeared in this century, exploiting the advantages of the brilliant radiation sources, are making an important contribution to the applicability of SR-based methods for CH studies.

Part 3 of the book deals with the molecular characterization of CH materials. First, the most classical spectroscopic technique is described, because Fourier transform infrared (FTIR) spectroscopy in transmission, attenuated total reflectance (ATR) and diffuse reflectance (DRIFT) modes is still widely used, and also the most recent reflectance near-infrared (NIR) and fibre-optic reflectance spectroscopy (FORS); all of these infrared techniques are used for the molecular characterization of CH materials and are reviewed in this first chapter of Part 3. Raman spectroscopy is then described for both FT-Raman and dispersive Raman modes, as they are the most commonly used vibrational spectroscopic techniques to characterize CH materials; confocal micro-Raman spectroscopy, Raman imaging and portable Raman spectroscopy are also considered in that chapter, together with the most modern approaches, spatially offset Raman spectroscopy (SORS) and surface-enhanced (SERS and SOERS) spectroscopy, discussing the future applications of different Raman configurations.

Figure 1.5 shows details of analyses performed on a Norman wall painting using portable DRIFT and micro-Raman spectroscopy to characterize the panels with two complementary techniques based on vibrational spectroscopy. In this project, any instrument for elemental analysis was used in the first screening that aimed to define the different areas showing homogeneity for a further physical sampling.

Finally, the molecular characterization of some organic materials used in CH is achieved with separation techniques because spectroscopic determination has some limitations; Chapter 7 describes the fundamentals of separation methods and reviews coupled chromatographic setups (GC-MS, Py-GC-MS, HPLC-MS, etc.), developed to characterize organic compounds (binders, dyes, lakes, etc.) used mainly in artworks.

Apart from the descriptions of individual techniques, in Parts 2 and 3 the concept of multianalytical approaches is introduced and a few cases are discussed as representative examples of combining the information coming from applying different analytical techniques to the same item (artwork, archaeological remains or a historical building). Analytical strategies based on a multianalytical approach are the most often used nowadays. In fact, very few problems can be now solved using only one analytical method. In the future, if restorers gain confidence about the possibilities opened up by these new analytical strategies, they will require us to use as many different techniques as possible to describe the problem properly.

Part 4 of the book considers the multianalytical approach to CH materials and their degradation, describing some significant thematic cases for which simple and complex analytical strategies, based on different multianalytical approaches, were considered by different authors at institutions all around the world. To sum up, the use of portable instruments in the most recent studies described in the literature is symptomatic and depicts the way that future work will be designed and performed.

As an example, Figure 1.6 shows the details of an analytical strategy designed to perform the characterization and diagnosis of the conservation state of late mediaeval wall paintings using exclusively portable instrumentation. One instrument was selected for detection of chemical elements (portable XRF) and two complementary vibrational spectroscopy instruments (portable DRIFT and Raman) for the detection of inorganic compounds and organic molecules. Here, portable micro-Raman spectroscopy could not be used because scaffolding was required and we were not able to control the vibrations, preventing the possibility of stable focusing at the scale of 50 µm.

The first chapter of Part 4 focuses on sculptures and public art exposed to the open air in cities with important impacts due to the stressors present in the urban atmosphere. Metals, stones and wood are the main materials used in such artworks and the characterization of such original materials and their degradation products requires the use of several analytical techniques, preferably with portable devices, as is highlighted in most of the studies reviewed in the chapter.

The spectroscopic study of pigments and binders in works of art is then considered through several examples developed in the frame of museum activities, where extensive efforts are applied to the conservation of paintings. Polychromed artworks exposed in museums have in common multilayered structures, being really complex systems that must be studied using several analytical techniques to diagnose their state of conservation (i.e. original materials and their degradation products) for the subsequent design of a conservation action plan, based on scientific data.

The third chapter of Part 4 reviews the special environmental constraints around rock art diagnosis, dating and conservation. Rock art combines layered artworks on a geological setting that was active before and after the painter executed the pictographs. This field of CH requires the use of several portable instruments to reduce the micro-sampling to the minimum necessary to perform the final characterization of the artworks and their conservation diagnostics. The developments performed to study past and current impacts of the shelter where the pictographs are located promoted the discovery of a new scientific method to date the age of the paints more accurately.

The fourth chapter in Part 4 focuses on archaeological materials, a difficult area of study owing to the poorly developed knowledge about the interaction of the soils and sediments in burials with the remaining bones, ceramics, textiles and metallic objects excavated from the site. The selection of suitable analytical techniques is essential for characterizing the materials without compromising their integrity for further conservation processes. The overall problem cannot be completely solved if we do not take into account the environmental characteristics of the burial and the impact (chemical reactivity) that different ions in the soils or sediments can have on the conservation state of the archaeological remains due to the formation of new degraded compounds.

An important chapter on polychromed enamels, ceramics, glasses and their degradation is then presented, for non-buried objects. Most of these wares are in museums but others are exposed directly to the urban atmosphere, such as the stained glass windows in castles, churches and cathedrals. These silica-based products have common chemical properties allowing the use of very similar combined analytical techniques to characterize the original major and minor compounds and their degradation products. Such degradations come from compounds inside the artworks and also from their reactivity with the surrounding environment, and these can be ascertained with the proper analytical strategies.

The final chapter of Part 4 deals with the characterization of organic residues from archaeological findings. Information on palaeoenvironments and palaeodiets, together with the daily practices of our ancestors, is contained in the organic remains in archaeological materials. The identification of such complex molecules (sometimes the original compounds are not present but their degradation metabolites) requires the use of well-consolidated analytical procedures based on chromatography and mass spectrometry but also the use of recently introduced multianalytical approaches. The information provided by these chemical studies regarding the diet, subsistence practices, daily activities, ritual practices and technologies of past civilizations and the reconstruction of possible trade routes is reviewed through several case studies.

In summary, this book covers first the individual spectroscopic and chromatographic analytical techniques suitable to perform studies not only for the characterization of artworks and immovable cultural heritage assets but also to diagnose their conservation state. From the very beginning in the book, the concept of analytical strategies based on the combined use of individual techniques is introduced through simple examples to characterize original and degradation products present in materials belonging to CH. This concept is further extended in the last part of the book, where the influence of the environment on the conservation state is the common point in the chapters dealing with different types of artworks and archaeological remains, giving some interesting examples of the methodologies based on chemometrics and chemical modelling to identify the sources of impacts, providing the clues needed by restorers to design the most appropriate conservation treatments.

Practically all of the analytical techniques that do not have portable devices have been applied in the field of CH, some focused on the detection and quantification of chemical elements and others on identifying the chemical compounds (inorganic and organic) present in samples. For those techniques, samples must be sent to the laboratory to perform the analysis. Nowadays, this is a major drawback because restorers and conservators are really afraid of the need to collect samples, even at the microscale level. Bearing this in mind, this section covers some developments conducted with these techniques.

For the detection of chemical elements and for their quantitative determination, inductively coupled plasma (ICP) spectrometry is the most commonly used technique, either using atomic emission spectrometric (AES) or mass spectrometric (MS) detection, after acid digestion or alkaline fusion of samples. To avoid the destruction of samples, laser ablation (LA) has been introduced in the past decade. As samples are sent to the laboratory, other non-destructive techniques are used before the destruction of samples to perform the ICP-AES or ICP-MS analysis.

For the characterization of crystalline compounds, X-ray diffraction (XRD) had been the technique of choice for many years. However, when the physical collection of samples to perform the analysis in the laboratory started to be questioned, the applicability of this technique in the CH field decreased. In recent years, some new portable devices to perform XRD analysis in the field have been published, but there are still no commercially available instruments, diminishing the current possibilities. However, when such commercial portable instruments come onto the market, new developments will start, as happened with other analytical techniques in the transition between instruments available only in the laboratory to the introduction of portable versions of the same techniques.

The same has occurred with another microscopic technique, scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM/EDS), which required the sample to be introduced into a compartment with high-vacuum capability.

We have not included specific chapters on these less commonly used analytical techniques in the last decade. We recommend the interested reader to check a book that gave a good overview of the use of these techniques, not only in the field of CH, up to the mid-1990s.² 

However, in all of the chapters in Part 4, some studies using the techniques described in this book together with LA-ICP-MS, XRD and/or SEM-EDS are included and reviewed. These are candidate techniques to be used in the design of the most suitable analytical strategy to solve a problem, when physical sampling is allowed by restorers; in such cases, non-destructive laboratory techniques must be used first and then these destructive techniques can be applied. In the following, some other studies that used ICP techniques, XRD and/or SEM-EDS with other techniques are reviewed.

One of the earliest studies using several analytical techniques before ICP-AES and ICP-MS analysis was a provenance study of obsidians collected in Jerf el Ahmar village, dated from 9500 to 8700 BCE (Middle Euphrates Valley, Syria),³  where the elementary compositions of 44 obsidian artefacts and of 19 samples from potential obsidian volcanic sources were determined after the use of particle-induced X-ray emission (PIXE) and scanning electron microscopy-energy dispersive X-ray (SEM-EDX) characterization. Later, Bellot-Gurlet et al. compared the advantages of the non-destructive PIXE technique with the well-established micro-destructive ICP-AES and ICP-MS methods, confirming the advantages of PIXE for provenance studies of obsidians.⁴ 

For the characterization of sensitive microsamples from rock-art paintings in shelters and caves, Resano et al.⁵  used LA-ICP-MS, SEM-EDX and Raman spectroscopy to characterize several red-coloured paintings of post-Palaeolithic schematic style found in 10 different shelters in the vicinity of the Vero River (Huesca, Spain) using only 900 ng to afford five replicates per sample.⁵  LA-ICP-MS in combination with XRD, wavelength-dispersive (WD)-XRF, SEM-EDX, colour analysis and optical microscopy was used by Gil et al.⁶  to rediscover the earth pigments of the Alentejo region (southern Portugal). A similar approach was used by Giannossa et al.⁷  to analyse ceramic samples using optical microscopy (OM), SEM-EDS, XRD and LA-ICP-MS supported by a statistical multivariate treatment of the compositional data for the ceramic bodies in order to identify the technology and raw materials used in thin-walled pottery from Herculaneum and Pompeii in Italy.

More recently, LA-ICP-MS, together with petrographic analysis and WD-XRF laboratory instruments, were used by Germinario et al.⁸  to study the provenance quarry of Euganean trachyte used in Roman public infrastructures (building roads, bridges, forum squares and aqueducts) in north-eastern Italy during the Roman Empire. The provenance determinations allowed the commercial and economic aspects involving the supply of trachyte for public works and the management and competition among nearby Roman quarries to supply materials to the most important cities to be defined.

This book has been designed to be used as a reference work at the master's level or in the first year of a PhD project in chemistry relating to CH. When the pre-doctoral student has surpassed the knowledge level covered here, more specialized literature will be consulted and used.

The available specialized literature in the field is enormous. Here some guidelines are given to follow up in your pre-doctoral formation. When possible, go to the primary sources of information (original research papers) and forget the use of documents deposited on the Internet, because not all are scientifically accepted works.

A research work is more credible if it has been published in a research journal with a high impact factor, because such journals only accept for publication innovative studies, performed on a high number of samples (remember, when using portable instruments each spot analysed is in fact a sample) and with a variety of analytical techniques implemented in an innovative analytical strategy to solve a problem.

For a first look at the most recent innovative developments, which can be understood only when the reader has acquired important basic knowledge, numerous specialized books (new or updated from earlier editions) are published nearly every year. Starting from those published in this decade, the book written by Artoli must be considered.⁹  In the recent book edited by Mazzeo,¹⁰  a survey of the most significant developments from around 2010 until a few years ago is presented, although the content does not provide an exhaustive description of all the developments in the field.

Other interesting books more devoted to particular aspects and/or techniques for CH are available. As an example, the most interesting developments of analytical techniques at the microscopic level¹¹  were described, although the book was more oriented to the elemental characterization of art objects mainly with X-ray-based analytical techniques. Another interesting book focused on the corrosion of metallic objects has also been published,¹²  presenting highly specialized chapters in the field of metals in archaeological sites, using more electrochemical than spectroscopic techniques. In the same year, another edited book appeared that was directed at restorers and introduced the possibilities given by analytical chemistry to characterize CH materials.¹³ 

The variability of materials used in the production processes of artworks and objects used throughout past years and centuries is enormous and the changes in availability of production techniques in the different civilizations or more modern schools must be taken into account when dealing with a diagnostic project. Moreover, traditional materials have been replaced by modern ones, thus increasing the range of possible materials present in the field of CH.

The consequence of such variability is clear: it is practically impossible to find the same problem in artworks produced even by the same artist. Some compounds will remain as they were when the artwork or craft was executed by the artist or artisan (for example, some inorganic pigments) but others may have undergone a different behaviour, as a function of the changing environment around the CH asset. As a consequence, the object under study will display different chemical compositions when analysed several decades or centuries after their manufacture.

This is a great opportunity for the analytical chemist to discover something new in the CH asset under study. Also, this is great challenge for chemists but also for analytical science. Do not forget that we must work in a multidisciplinary context, receiving and giving, i.e. exchanging, with other professionals with whom collaboration is essential in the current practice around CH conservation and preservation.