Artworks and Cultural Heritage
Materials: Using Multivariate Analysis
to Answer Conservation Questions
Presenting by:
BT22MME034 MORE SANGHARSH VIJAY
BT22MME049 CHINTHALA MANOHAR
BT22MME081 SAVAN KUMAR
BT22MME102 JADHAV ANKITA ARUN
contents:
• 1. ROCK ART PETROGLYPHS EXAMINED WITH REFLECTANCE NIR
SPECTROSCOPY AND PCA
• 2. ADHESIVES STUDY OF CYPRIOT POTTERY COLLECTION WITH FTIR
SPECTROSCOPY AND PCA
• 3. EGYPTIAN SARCOPHAGUS EXAMINED WITH ToF-SIMS, XANES, AND PCA
• 4. ATTRIBUTION STUDIES OF AN ITALIAN RENAISSANCE PAINTING: ESEM
IMAGING
• 5. OCHRE PIGMENTS IMAGED USING SYNCHROTRON XRF
ROCK ART PETROGLYPHS EXAMINED WITH REFLECTANCE NIR
SPECTROSCOPY AND PCA
Australia's largest and most important collection of indigenous petroglyphs
are located in Burrup Peninsula. Petroglyphs have been created through
pecking, scraping or incising through thin red brown rock patina.
Red coloration – Background and White coloration – Engraved surface
Major concern was some of the petroglyphs are located next to industrial
areas, petroglyphs could be damaged by air-borne emissions.
The BRAMMC was established and their role was to review and oversee
studies that were conducted pertaining to emissions and their effect on the
petroglyphs.
From 2004 to 2011, visible near infrared reflectance (VNIR) spectra were
collected from seven sites using a Fiber-optic reflectance spectrometer.
At each of the seven petroglyphs, six spots were measured consisting of
three in the engraved region and three in the background of the rock
surface
Spectra of engravings were mostly different from those of
background which confirms the minerology difference.
Some year to year variance is observed, but it was due to
difference in moisture content affecting brightness and
reflectivity.
It can also be appreciated at this point that the overwhelming
number of spectra(336) from eight annual measurements of six
points per petroglyph on seven petroglyphs are difficult to
observe simultaneously without the aid of more sophisticated
visualization approaches.
For that PCA projection using the Unscramblers X, Version
10.0, was performed.
The first PC (PC1) represents the majority of variance in the
data (79%) and the second PC (PC2) represents 10% of
variance within the data.
Clearly, representation of 336 near infrared (NIR) spectral measurements is a challenge for traditional methods of
spectral interpretation. Visualization of complex data can be achieved through data reduction by PCA.
PCA clearly facilitated assessment of various attributes and variables such as mineralogy according to:
whether spectra were obtained from background or engraving regions
how much influence the site has on the spectral variance
and whether changes were observed over time.
Adhesive Analysis:
ADHESIVES STUDY OF Various adhesives, including animal-based glue,
CYPRIOT POTTERY cellulose nitrate (CN), poly(vinyl acetate)
COLLECTION WITH FTIR (PVA), acrylic (ACR), and polystyrene (PS),
were identified through comparison with
SPECTROSCOPY AND reference spectra of known adhesives.
PCA
FTIR and PCA Application:
• FTIR, performed in the museum storeroom, helped
collect spectra using acetone-moistened cotton
swabs.
• PCA was applied to CN spectra, focusing on the
wavenumber range between 650 and 1800 cm⁻¹.
The first two principal components (PC1 and PC2)
showed no significant difference. However, PC3 and
PC4 distinguished the CN samples into two
distinct groups.
Differentiation of Adhesive Groups:
•These two groups correlated with the two acquisition
periods: 1972 (Group 2) and 1987 (Group 1).
•Dibutyl phthalate (DBP), a plasticizer commonly added to
CN adhesives, was found to be the key factor distinguishing
the groups. The 1972 adhesives had higher DBP content
(15%), while the 1987 adhesives had lower levels (2%).
GC-MS Validation:
•Gas chromatography–mass spectrometry (GC-MS)
confirmed the DBP plasticizer's presence and quantity in the
CN adhesives, reinforcing the PCA findings.
•The lower DBP levels in the 1987 samples suggest higher
conservation priority, as less plasticizer makes the adhesive
more brittle and prone to deterioration.
XANES (Green Pigment Analysis on Tjeseb's Coffin)
3. EGYPTIAN SARCOPHAGUS • Challenge: Standard techniques couldn’t identify the
EXAMINED WITH XANES, green pigment on the coffin.
AND PCA • Solution: Used X-ray Absorption Near-Edge Structure
(XANES), a non-destructive method, to examine the
green pigment.
• Findings:
• The green pigment was copper-based.
• By comparing it to copper minerals, the pigment was
identified as likely being atacamite (cu hydroxy
chloride).
• Method: Combined PCA (Principal Component
Analysis) with XANES to organize and compare data,
helping identify the pigment more easily.
• Importance: XANES and PCA together allowed
precise identification of the pigment without
damaging the artifact.
Multivariate Compositional Imaging
What It Is: A method to map the chemical composition of a surface by analyzing data at each
point.
How It Works: Each point has a spectrum (data set), creating a detailed chemical map.
Uses:
•Helps find patterns of material composition on artifacts like paintings and historical items.
•Non-invasive: Does not damage the artifact during analysis.
Benefit: Shows the distribution of materials across a surface, helping conservators understand
how to preserve the artifact.
Summary:
XANES+ PCA= Accurately identifies materials like pigments.
Compositional Imaging = Maps the material distribution for better conservation without harm.
Attribution Studies of an Italian
Renaissance Painting: ESEM Imaging
•Key Discovery: The presence of a gray-brown imprimatura layer between the gesso priming and oil paint layers. This
type of preparatory layer was uncommon in early 16th-century .
•Technique Used:
•ESEM-EDS (Environmental Scanning Electron Microscopy with Energy Dispersive Spectroscopy): Nondestructive
imaging that provided detailed elemental maps of the painting’s cross-sections.
•PCA (Principal Component Analysis): Applied to analyze the data and produce color-coded images showing different
material compositions.
•Cluster Analysis: Identified similar compositional areas within the layers.
•Findings:
•Azurite pigment (a blue copper-containing
pigment) was identified.
•Lead white pigment and different greenish
particles high in calcium were found in both the
paint and imprimatura layers.
•The ground layer was rich in gypsum and
anhydrite.
•Some particles, identified as dolomite and silica,
were also observed in the imprimatura layer.
•Verification: Raman spectroscopy confirmed
the presence of the materials identified by ESEM-
EDS.
OCHRE PIGMENTS IMAGED USING
SYNCHROTRON XRF
Ochre Pigments: Natural pigments that date back to prehistoric times, providing colors like
browns, reds, yellows, orange, pink, and violet. They are still used today, making up 20% of
the global pigment market.
Composition: Ochres contain iron oxides (hematite for red, goethite for yellow) and minerals
like kaolinite, quartz, calcium compounds, and many other minerals and heavy metals.
Forgeries in Indigenous Art: There is growing interest in analyzing ochres used in
Indigenous Australian artworks due to concerns over forged artworks in the market, which has
seen a rapid increase in value.
Analytical Techniques:
•X-ray Fluorescence (XRF)-PCA Imaging: A non-destructive
method used for analyzing ochre pigments without taking physical
samples.
•Synchrotron XRF Imaging: Applied to three ochre samples (two
XRFX yellow, one red) painted on canvas. It provided detailed elemental
maps for elements like Fe, Zn, Ti, Mn, and Ca.
Results:
•Yellow ochres primarily contained goethite, while red oxide was
mainly hematite.
•High Ti levels were detected in the white background(canvas) due
to (TiO2) pigments.
•RGB images using Fe, Zn, and Mn showed distinctions between
the ochres. The two yellow ochre pigment samples are not
distinguishable from each other.
Advanced Analysis:
Principal Component Analysis (PCA) and
Minimum Noise Fraction (MNF) transforms
helped to distinguish between the two yellow
ochres in the images.
Unmixing Algorithms: These could analyze
mixed pigments, making this technique useful for
more complex paint compositions.
Future Potential: Synchrotron XRF imaging can
visualize pigment distribution across paintings, and
portable XRF instruments allow examination of
artwork without moving it from its display or storage.
