manuscript lab



Initially, microscopy referred to magnified images obtained with the help of wide illumination in the visible range of the electromagnetic radiation. Today this type of microscopy, called light or optical microscopy, covers a broad spectrum of specific techniques designed to enhance contrast and to improve photomicrography. In the field of Cultural Heritage, we often employ stereomicroscopy with reflected light and moderate magnification. In contrast, optically anisotropic specimens are best studied with polarization microscopy. Recent technological developments led to the appearance of digital microscopy, which directly captures the images.

Electron-scanning microscopy is fundamentally different from optical spectroscopy. Here, the inter-action of a beam of fast electrons with the sample surface reflects the density of the material under investigation. On the one hand, materials with different densities produce contrasting images; on the other, the penetration depth of the electron beam leads to a three-dimensional presentation of the sample surface. The working scales and magnification of electron microscopy range from 200 times to approximately 106.

Selected Application

Fiber analysis - dating and provenance studies for paper

In addition to visually examining paper (watermarks, surface structure), it is possible to investigate the physico-chemical composition of the material. Apart from fibrous materials, paper contains a multitude of different components, such as fillers, pigments, sizing, and wet strength agents, in addition to other chemical additives influencing the properties and the quality of paper. The paper surface, too, often carries coatings containing a number of components that may include several layers. All these features can assist in dating and provenance studies of paper.

Micro-invasive investigations focus on precursor fibers that can be determined with the help of microscopy. A varying magnification from 50 to 600 times with both plain and polarized light is required for analysis. During investigation, a small sample is immersed in distilled water and boiled for about 20 minutes. The water is then removed and the sample is drained off, de-fibered into a fine suspension of individual fibers, and placed on a slide. Fibers are first observed without any reagent and then treated with Herzberg or "C" stain (Isenberg 1967). Attention is paid to stain coloring and the morphology of the fibers and of other cells and elements of the pulp. Both the width and length of fibers have to be measured to support the identification in particular cases.


UV and IR reflectography is commonly used to examine Cultural Heritage artifacts. Most common is the use of so-called “long wave UV”, or UVA (400 – 320 nm). Some materials absorb UV radiation and re-emit it in the visible spectrum as UV-induced visible fluorescence. UV reflectography exploits the sensitivity of photographic emulsion to these wavelengths by capturing both visible fluorescence and any reflected (not absorbed) UV radiation using photographic filters that block specific wavelengths of radiation. Most interesting for our purposes is that UV radiation causes visible fluorescence of proteinaceous materials, such as parchment or animal glue.

Selected Applications

Ink Typology

Many materials also exhibit different visual appearances under specific wavelengths of IR radiation, absorbing, transmitting, or reflecting the radiation. Materials that absorb IR radiation appear dark, those that transmit it are transparent, and those that reflect it appear white, though the appearance of a material can change with different IR wavelengths. This property is used for the typological classification of inks. Plant inks become transparent at about 800 nm, iron gall inks at between 1000 nm and 1200 nm. Writing materials that contain elemental carbon (such as carbon and graphite) will absorb the infrared light across the entire mid-infrared range and will appear as black lines.

Optical response of three different types of inks: iron gall ink (top row), plant ink (middle row), carbon ink (bottom row). Reproduced from: R. Mrusek, R. Fuchs, D. Oltrogge (1995), ‘Spektrale Fenster zur Vergangenheit’, in Naturwissenschaften, 82, p. 68–79.

Since the 19th century, paper analysis includes the inspection of watermarks, which are the imprints of metal-wire figures placed on molds or sieves during the paper production. Dating according to the watermarks is based on the "degree of kinship".

Watermarks can be correlated as

  • identical (i.e., congruent)
  • similar (depicting the same wire figure, but with some damage or displacement)
  • the same type of motif

Studies of the watermarks related to European paper mills resulted in watermarks catalogues. It was estimated that in the 16th century a mill working all year would renew the mold pairs annually. A mold renewal was usually accompanied by small but detectable changes in dimensions. Furthermore, it is assumed that the paper produced was consumed within one to four years (Piccard 1961). Therefore, a successive chain line of the watermarks allows rather precise dating of the paper.

Knowledge of the production site of the paper does not automatically determine the origin of a document. Here, the quality of a paper plays a decisive role. Expensive, high-quality paper would travel farther, since its price could easily cover the costs of transport and duties. Indeed, there is some evidence that much low-quality paper was consumed locally, i.e., close to the paper mill.

Since the 1860s, paper dating has been conducted by simple visual inspection of the watermarks. From the 1940s till the mid-1980s, various x-ray imaging methods, such as electron or beta radiog-raphy, and soft x-ray methods were employed to aid in giving visibility to the watermarks and the paper structure (Dietz and Delft 2009). Backlight techniques offer a simpler solution, though back-light imaging appears to provide only ca. 80% of the contrast offered by the x-ray techniques.

The backlight method is executed in three steps:

  1. take a digital picture of the paper with backlight off;
  2. take a digital picture of the paper with backlight on. The image shows the paper structure and the drawing
  3. by subtracting one of the images from the other, and by applying image enhancement techniques, we obtain a new picture that mainly shows the paper structure and watermarks the same type of motif

(Dietz et al. 2012)

Databases of watermarks:

Briquets dictionnaire des filigranes

Gerhard Piccard


Multispectral (MSI) and hyperspectral imaging (HSI)

Multispectral (MSI) and hyperspectral imaging (HSI) have been developed to recover obscured writing and other information from damaged, deteriorated, or erased manuscripts (palimpsests). The MSI system collects high-resolution images at 19 wavelengths (365 nm to 1050 nm). In contrast, the HSI set-up collects images from 80 wavelengths in the same range and automatically extracts reflectance curves that can be also used to monitor the state of the object and to identify classes of inks (e.g., iron gall or carbon-based inks). Both systems support the differentiation of UV reflectance and UV fluorescence data as well as the differentiation of UV fluorescence data into blue, green, orange, and red components. Fluorescence is the phenomenon by which parchment and other organic materials absorb short wavelength radiation (e.g., UV and blue light) and emit radiation at a longer wavelength. Experience has shown that the collection of both UV fluorescence images for different colors and UV reflectance images enables the recovery of obscured writing and reveals distinctive features of the ink and its support.

Visible spectrophotometry

Visible spectrophotometry measures the reflectance of the surface as a function of the wavelength. It is a convenient tool for the analysis of colored materials (e.g. to classify pigments and dyes). The method uses only the visible range of the electromagnetic spectrum (380 nm - 730 nm); therefore, it is not possible to detect black and white materials.

Limitations: pigment mixtures; copper green pigments

X-ray emission spectrometry

Elemental analysis by X-ray emission techniques relies on the study of characteristic patterns of X-ray emissions from atoms irradiated with high-energy X-rays or particles: X-ray fluorescence (XRF), particle-induced X-ray emission (PIXE), and energy-dispersive X-ray spectroscopy (EDX). When the external excitation beam interacts with an atom within the sample, an electron is ejected from the atom's inner shell, creating a vacancy. In the next step, another electron from an outer shell fills the vacancy. The energy of the emitted X-ray fluorescence is characteristic for a certain element, whereas the signal intensity allows one to determine the amount of the element in the sample. It is noteworthy that each technique has its applicability limits and different penetration depths, so that excitation by electrons (EDX), conventionally used with electron microscopy, is limited to the study of surfaces (but capable of detecting light elements), whereas excitation by X-rays (XRF) has greater penetration power and allows one to detect elements with Z > 13, that is, elements heavier than aluminum. XRF is one of the most suitable methods to extract qualitative and semi-quantitative information on the elemental composition of various materials. However, it should be noted that XRF is not well suited for the study of organic materials. This technique benefits from the availability of a variety of transportable instruments ranging from single-spot measurements to high-resolution scanning equipment.

Selected Application

The fingerprint model for the differentiation of iron gall inks

The development and use of the fingerprint model based on the qualitative and quantitative detection of inorganic components of iron gall inks allows their reliable classification. In short, X-ray incident radiation causes ink components, such as iron, copper, zinc, etc., to emit characteristic X-rays that are recorded as spectra by an XRF spectrometer. Evaluation of the spectra yields a finger-print of the ink under investigation.

For a reliable quantitative analysis of the XRF data, absorption processes have to be taken into account. Because the main components of the ink and paper are light elements that are not detected by energy dispersive XRF, an absolute quantification is not possible. However, the determination of a fingerprint representing the amount of a trace element in relation to the main component, iron, enables the characterization of different iron gall inks. In the fingerprint model, the ink-paper system is regarded as a three-layer system with a top layer of pure iron gall ink, a diffusion layer with a linearly decreasing amount of ink and a bottom layer consisting of pure paper. For this system of layers, a fundamental parameter approach for the primary fluorescence is taken. The primary intensity can be expressed as the sum of the contributions from the ink and from the paper. Ultimately, this leads to the fingerprint value, which depends on three parameters: the transmission of the entire system, the penetration depth of the ink into the paper, and the ratio of the absorption coefficients. For a specific minor constituent "i" (such as Mn, Cu, Zn) a fingerprint value "Wi" can be specified. This fingerprint value represents the amount of a minor constituent relative to the main component, iron (Malzer et al. 2004, Hahn et al. 2004, Hahn 2004).

Two investigated manuscripts which belong to 'Faust I' were written with one iron gall ink. The ink in the 'Faust II' manuscript shows a different elemental composition. It is noteworthy that the ink used to add some corrections to a 'Faust I' text passage seems to be the same ink as used for the writing of the 'Faust II' passage.

IR and Raman spectroscopy

IR and Raman spectroscopy techniques deliver information on chemical composition and, therefore, are routinely applied to screen unknown assays. Both techniques use the fact that the bonds of atoms in a molecule interact with light at characteristic frequencies in the infrared spectrum. Therefore, their detection in an IR or Raman spectrum reveals the chemical identity of the materials under investigation. The type of interaction measured by infrared spectroscopy differs from and complements that detected by Raman. Historically, IR spectroscopy has been commonly used for the investigation of organic materials. It is therefore a well-established method for classifying binding media. To perform a conventional measurement (so-called transmission mode), a thin or powdered sample is placed in the beam pass and the amount of transmitted light detected as a function of wavelength or frequency, producing an infrared spectrum. Hence, this method required samples to be extracted from an object. To reduce the sample size, special diamond cells were developed. Rapid technological progress in this field led to the appearance of non-destructive methods such as attenuated total reflection Fourier transform infra-red (ATR-FTIR) spectroscopy to study surfaces, fiber-optic FTIR in reflection, and synchrotron-based FTIR spectroscopy. The miniaturization of infrared sources and detectors brought a new generation of portable FTIR spectrometers, for example a hand-held ExoScan.

Selected Application

Raman spectroscopy for identifying inorganic pigments

Raman spectroscopy, named for the Indian physicist Sir Chandrasekhara Venkata Raman, is a tech-nique that relies on the scattering of monochromatic light in the visible, near-infrared, and near-ultraviolet range, used in particular to provide vibrational information specific to different chemical bonds and molecules. It has proved to be a specifically powerful tool for identifying inorganic materials. In the field of Cultural Heritage, it is now routinely used to identify pigments, whose spectra are tabulated.

Raman spectrum of anatase. The main ingredient of the commercial pigment titanium white produced in 1921 - 1945.

Databases of Raman spectra:

Raman Spectroscopic Library of Natural and Synthetic Pigments (University College London)

The RRUFF Project

The Infrared and Raman Users Group (IRUG)