Ep 32 — Cultural Heritage Conservation: Identification of Pigments, Binders, and Substrates

Series: Infrared Spectroscopy Encyclopedia: From Principles to Practice
Chapter: Part 3 · Intermediate — Industry Applications (Latter Part)
Target Audience: Cultural heritage conservators and restorers, museum researchers, analytical chemists in cultural heritage, art historians and graduate students
Prerequisites: Ep 14 (ATR), Ep 19 (Library Search), Ep 26 (μ-FTIR Imaging), Ep 31 (Trace Evidence μ-FTIR)
Reading Time: Approx. 45 minutes

Introduction: When Science Enters the Sistine Chapel

In 2003, the restoration of the Sistine Chapel ceiling frescoes in the Vatican entered its final phase. Restorers faced a thorny issue: How to distinguish Michelangelo's original work from later repaints? Visual inspection could no longer determine which brushstrokes belonged to the original 1508–1512 work and which were later additions. Ultimately, a research team from the Italian National Research Council (CNR) used macroscopic mid-IR reflection hyperspectral imaging to clearly distinguish repainted areas from original work without contacting the fresco — this common practice has since become a benchmark for non-invasive analysis of cultural heritage [1].

Similar cases have accumulated in European and American museums for over 40 years. In 1986, the Getty Conservation Institute (GCI) first introduced FTIR for pigment identification in cultural relics; in 2002, the Italian CNR-ISMN established the first infrared spectral database for cultural heritage (IRUG, Infrared and Raman Users Group); in 2022, the Kazarian group at Imperial College London published a comprehensive review in Analyst, systematically summarizing the applications of ATR-FTIR and ATR-FTIR imaging in cultural heritage [2]. The review states:

"ATR-FTIR spectroscopy and ATR-FTIR spectroscopic imaging have become indispensable tools for the non-invasive and minimally invasive analysis of cultural heritage materials, providing molecular-level information on binders, pigments, degradation products and layer structures."
—— Liu G L, Kazarian S G. Analyst 2022 [2]

In this episode, we will systematically explain the application of FTIR in cultural heritage conservation: from non-invasive identification of bone black pigments to chemical imaging of cross-sections of oil paintings, from long-range macro hyperspectral imaging to submicron analysis with O-PTIR, and finally the infrared monitoring of degradation products in cultural relics.

1. Why Does Cultural Heritage Conservation Favor Infrared Spectroscopy?

1.1 The "Rigorous" Requirements of Artifact Analysis

The most significant characteristic of artifact analysis is that samples are extremely precious and must not be destroyed [2][3]. Any analytical method must meet the following conditions:

• Non-invasive or minimally invasive: sampling should be avoided if possible; if necessary, it must be controlled at the μg level;
• In-situ or on-site capability: many artifacts cannot be moved (frescoes, sculptures), requiring portable or remote equipment;
• Rich information: able to simultaneously identify pigments, binders, substrates, and degradation products;
• Database comparison: reliable spectral libraries dedicated to cultural heritage.

ATR-FTIR is almost naturally suited to these requirements [2]:

Table 1: Comparison of common methods for cultural heritage analysis (data source: Liu & Kazarian Analyst 2022 [2])

1.2 Three Sampling Modes

The review by Liu and Kazarian classifies the applications of ATR-FTIR in cultural heritage into three categories by "degree of contact" [2]:

   Contact level        Method                    Typical scenario
   ────────────────────────────────────────────────────────────────
   Non-contact           Reflection IR / O-PTIR        Fresco and painting surfaces
   ↓                     Remote macro hyperspectral    Large-area fresco scanning
   Minimal contact       ATR (contact force N-level)   Movable small artifacts
   ↓                     μ-ATR                        Micro-sampled particles
   Micro-sampling        μ-FTIR transmission           Cross-section slices

Each mode has its unique value; we will elaborate on each below.

2. Infrared Differentiation of Mineral and Synthetic Pigments

2.1 Infrared "Fingerprints" of Pigments

Pigments used in cultural heritage can be roughly divided into three categories [3][4]:

• Inorganic mineral pigments: Vermilion (HgS), lead white (2PbCO₃·Pb(OH)₂), lapis lazuli (Na₈..₁₀Al₆Si₆O₂₄S₂..₄), malachite (Cu₂CO₃(OH)₂), ochre (Fe₂O₃·nH₂O), etc.
• Organic natural pigments: Indigo (C₁₆H₁₀N₂O₂), alizarin, carminic acid, etc.
• Modern synthetic pigments: Titanium white (TiO₂, only after 1916), cadmium red (CdSe, early 20th century), phthalocyanine blue (CuPc, after 1935)

Their characteristic peaks in the infrared spectrum are entirely different, forming the chemical basis for dating identification — if titanium white is detected in a painting claimed to be from the 16th century, it must be a later repaint or forgery [3][4].

The table below lists the infrared characteristics of common pigments [3][4][5]:

Table 2: Infrared characteristic peaks of common pigments (data sources: West Fitzhugh Artists' Pigments [4]; IRUG database [5])

[Continue with the rest of the article...]

2.2 Non-invasive Infrared Reflection Identification of Bone Black Pigment

Bone Black is a black pigment used since prehistoric times, produced by calcining animal bones under oxygen-deficient conditions. Its main components are calcium phosphate (hydroxyapatite, Ca₁₀(PO₄)₆(OH)₂) + carbon [6].

Daveri et al. in 2018 in Journal of Analytical Methods in Chemistry proposed a method for non-invasive reflection infrared identification using the strong stretching vibration of phosphate PO₄³⁻ in bone black at 1090–1040 cm⁻¹ and the weak but stable 'phosphate overtone/combination band' at 2013 cm⁻¹ [6]:

"The 2013 cm⁻¹ band, although weak, is highly characteristic of bone black and is absent in other carbon-based blacks (lamp black, vine black, ivory black). This makes it a useful marker for non-invasive identification of bone black on paintings using mid-IR reflection spectroscopy."
—— Daveri A et al. J Anal Methods Chem 2018 [6]

The significance of this discovery: the 2013 cm⁻¹ lies in an atmospheric transmission window, allowing long-distance infrared reflection spectrometers to detect this peak without interference, providing a reliable "non-contact ID card" for bone black pigments on murals and oil paintings [6].

2.3 Identification Example: Layer Analysis of a 14th Century Altarpiece

Taking a 14th-century altarpiece from Siena, Italy, as an example [2], the research team performed the following work:

1. Selected 5 representative areas on the painting surface and scanned in situ with portable reflection FTIR (30 seconds per point);
2. Measured strong peaks at 1404/1045 cm⁻¹, identified as lead white ground (gesso sottile);
3. Measured double peaks at 870/818 cm⁻¹, identified as malachite pigment;
4. Measured strong peaks at 1495/1095 cm⁻¹, identified as azurite pigment;
5. Measured a weak peak at 2013 cm⁻¹, identified as bone black used in shadow areas.

The entire process was zero-sampling, only contacting the surface with a fiber optic probe at 0.5 N pressure, leaving the artifact completely undamaged [2].

III. Identification of Ancient Binders

3.1 Function and Classification of Binders

Binders (medium) are the "glue" that binds pigment particles together and adheres them to the substrate. Common ancient binders broadly include [2][7]:

Table 3: Infrared features of common binders (data from Liu & Kazarian Analyst 2022 [2]; Derrick Painting Conservation [7])

3.2 Egg Tempera vs. Oil Painting: Protein vs. Drying Oil

Infrared spectra of early Renaissance tempera and later oil paintings differ significantly [2][7]:

• Tempera: Typical biphasic system—simultaneously shows Amide I 1650 / Amide II 1540 (protein) and 1740 cm⁻¹ ester C=O (lecithin and other lipids);
• Oil painting: Predominantly 2925/2850 CH₂ + 1740 triglyceride C=O, no Amide I/II doublet;
• Aged oil painting: 1740 cm⁻¹ weakens, 1710 cm⁻¹ (carboxylic acid C=O) strengthens, a sign of oil hydrolysis and decarboxylation.

🔗 Further reading: The Amide I/II vibrations of protein in binders are consistent with the amide functional group page on ftir.fun https://ftir.fun/ir/group/amide. Amide I is mainly C=O stretching (1600–1690 cm⁻¹), Amide II is N–H bending + C–N stretching (1480–1575 cm⁻¹); together they confirm the presence of protein.

3.3 Identification of Beeswax Seal

Ancient Egyptian mummies, Hellenistic encaustic paintings, and Byzantine icons often used beeswax as a binder or protective layer [2][7]. Beeswax is a mixture of long-chain alkanes (C₂₅–C₃₁) and long-chain fatty acid esters, with distinctive infrared features:

• Strong 2918/2850 cm⁻¹ CH₂ stretching;
• 1735 cm⁻¹ ester C=O;
• 1470 cm⁻¹ CH₂ bending;
• 730 cm⁻¹ doublet—CH₂ in-plane rocking of orthorhombic crystalline alkanes, the "fingerprint" of beeswax.

💡 Industry experience: Whether the 730 cm⁻¹ doublet splits into 730/720 double lines is an important basis for judging whether beeswax has undergone heating and recrystallization—ancient encaustic painting required heating to melt the beeswax, and changes in wax crystallinity leave traces in the infrared spectrum [7].

IV. Cross-Sectional μ-FTIR Chemical Imaging of Oil Paintings

4.1 Why Cross-Sections?

An oil painting from bottom to top typically consists of [2][7][8]:

1. Support: Canvas, wood panel;
2. Ground layer (gesso): Lead white + animal glue, or gypsum + animal glue;
3. Underdrawing: Black charcoal or ink;
4. Paint layer(s): Pigment + binder, possibly multiple layers;
5. Varnish layer: Natural or synthetic resin.

Measuring only surface reflection infrared gives only the outermost layer information. To understand layer structure, chemical composition of each layer, and distribution of degradation products, cross-section analysis is necessary [2][8].

4.2 μ-FTIR Cross-Section Imaging Workflow

Liu and Kazarian reviewed the standard workflow for cross-sectional μ-FTIR imaging [2]:

1. Sampling: Using a scalpel, take a 0.5×0.5 mm sample from the edge or inconspicuous area of the painting;
2. Embedding: Embed in polyester or epoxy resin (non-interfering in infrared);
3. Sectioning: Cut 1–5 μm thin sections with a diamond knife microtome, place on BaF₂ or KBr micro-windows;
4. FPA Focal Plane Array Imaging: 128×128 or 64×64 FPA detector, each pixel 1–5 μm, acquire a complete data cube in one measurement;
5. Chemical Imaging: Select characteristic peak integrals to draw two-dimensional distribution maps of each component.

4.3 Case Study: Layer Structure of a 16th Century Venetian School Painting

A cross-section of a 16th-century oil painting attributed to the Titian school was analyzed by the Courtauld Institute and Kazarian's group at Imperial College using FPA-ATR imaging [2][9]:

• 2925/2850 cm⁻¹ (CH₂ stretching) → drying oil binder distribution;
• 1404/1045 cm⁻¹ (lead white) → ground layer;
• 1495/1095 cm⁻¹ (azurite) → blue paint layer;
• 1735 cm⁻¹ (beeswax) → surface seal.

The resulting chemical imaging map clearly shows: the underlying lead white layer is about 50 μm thick, the middle azurite + drying oil layer is about 30 μm thick, and the surface beeswax layer is 5 μm thick. This layered structure conforms to the typical craftsmanship of 16th-century Venetian painting, providing key evidence for dating the artwork [2].

   Surface                                    Bottom
   ┌─────────────────────────────────────┐
   │ ▓▓▓▓▓▓▓▓▓▓  Beeswax seal (1735)        │ 5 μm
   ├─────────────────────────────────────┤
   │ ░░░░░░░░░░░  Drying oil + azurite (1495,1095)│ 30 μm
   ├─────────────────────────────────────┤
   │ ▒▒▒▒▒▒▒▒▒▒▒  Lead white ground (1404,1045)   │ 50 μm
   ├─────────────────────────────────────┤
   │ ▓▓▓▓▓▓▓▓▓▓▓  Canvas fibers                 │
   └─────────────────────────────────────┘

📷 Figure 1: Schematic cross-sectional chemical imaging of a 16th-century Venetian oil painting
Source: Adapted from Liu & Kazarian Analyst 2022, 147, 1777 [2]

V. Infrared Identification of Degradation Products in Heritage Objects

5.1 Heritage Objects Also "Age"

During long-term preservation, heritage objects undergo various chemical degradations [2][10][11]:

• Oil paintings: Hydrolysis and decarboxylation of oils → enhancement of carboxylic acid C=O at 1710 cm⁻¹; formation of metal soaps (lead/tin soaps) → appearance of carboxylate COO⁻ asymmetric stretching at 1510–1540 cm⁻¹;
• Wall paintings: Sulfate attack → appearance of SO₄²⁻ at 1100 cm⁻¹; nitrate attack → appearance of NO₃⁻ at 1380 cm⁻¹;
• Glass: Water attack → enhancement of Si–OH at 950 cm⁻¹; surface silica-rich layer → enhancement of Si–O–Si at 1100 cm⁻¹;
• Paper: Cellulose oxidation → appearance of C=O at 1730 cm⁻¹; hydrolysis → weakening of C–O at 1050 cm⁻¹;
• Dyed silk: Photodegradation of dyes → weakening of aromatic C=C at 1500–1600 cm⁻¹.

5.2 Metal Soaps — The Mystery of "Blistering" in Oil Paintings

Many 15th–17th century oil paintings exhibit surface protrusions, translucency, and blistering, long puzzling conservators. Since the 1990s, FTIR imaging studies have revealed the mystery [2][10]:

• Lead white, tin white, and other metal-containing pigments react with free fatty acids produced by the degradation of drying oils, forming metal soaps:
  • Lead stearate: 1510, 1540 cm⁻¹;
  • Zinc stearate: 1540 cm⁻¹;
• Metal soaps migrate, aggregate, and crystallize within the paint layers, leading to translucency and protrusions;
• FPA-FTIR imaging can directly observe the two-dimensional distribution of metal soap aggregates, which exactly corresponds to the "blistering" areas under visible light.

"Lead soaps formation is the principal chemical process underlying the deterioration of Old Master paintings. ATR-FTIR imaging provides the most direct evidence for their distribution within paint layers."
—— Liu & Kazarian Analyst 2022 [2]

5.3 Identification of Siloxanes in Heritage Glass

🔗 Further reading: The main component of heritage glass (silicate glass) is a SiO₂ network, with characteristic infrared absorptions at 1000–1100 cm⁻¹ (Si–O–Si asymmetric stretching), 780 cm⁻¹ (Si–O–Si symmetric stretching), and 450 cm⁻¹ (Si–O bending). See ftir.fun siloxane functional group page. When glass is attacked by moisture, Si–OH enhances at 950 cm⁻¹ and a Si–O–Si enriched layer appears at 1100 cm⁻¹, which are key indicators for diagnosing "sick glass" [11].

VI. Non-invasive Analysis: Reflection IR and O-PTIR

6.1 Advantages and Pitfalls of Reflection IR

External reflection IR is a truly "zero-contact" infrared method — light is reflected from the sample surface, and the reflected light spectrum is analyzed [1][2][3].

Advantages:

• Completely non-contact with the heritage object;
• Can operate at a distance (meter scale);
• Capable of hyperspectral imaging over large areas.

Pitfalls:

• Reflection spectra are simultaneously influenced by absorption (Kramers-Kronig) and specular reflection, causing severe spectral distortion;
• Near strong absorption peaks, derivative-like shapes appear;
• Must be corrected using Kramers-Kronig transform or KK algorithm to match transmission libraries.

6.2 O-PTIR: Sub-micron Non-contact Infrared

O-PTIR (Optical Photothermal Infrared) is a next-generation infrared technology commercialized in 2018 and emerging in the 2020s [12]. Its principle is:

1. Pulsed infrared pump light irradiates the sample, causing local temperature rise upon molecular absorption;
2. Visible probe light (e.g., 532 nm green light) irradiates the same spot; the temperature-induced refractive index change modulates the probe light intensity;
3. Demodulating the probe light intensity signal yields an "infrared spectrum" corresponding to the infrared wavenumber.

    Pulsed IR (pump)            Visible probe (detection)
        ╲                      ╱
         ╲                    ╱
          →→→→→ Sample ←←←←←←
                    ↑
               Local heating → refractive index change
                    ↓
               Probe light intensity modulated

Key advantages of O-PTIR [12]:

• Spatial resolution ~450 nm, an order of magnitude higher than the diffraction limit of conventional μ-FTIR (~10 μm);
• Non-contact, can operate remotely via objective lens;
• Simultaneously acquires visible images and IR spectra;
• Virtually no spectral distortion, directly matchable to transmission libraries.

6.3 Marchetti et al.'s Study of 16th-Century Heritage Objects

A 2022 paper by Marchetti et al. in Science Advances demonstrated the power of O-PTIR for analyzing 16th-century heritage glass-metal composite objects [12]:

"We demonstrate, for the first time, the use of O-PTIR for the noninvasive characterization of heritage glass-metal composite objects at sub-micron spatial resolution, revealing corrosion products and interfaces that are not resolvable by conventional FTIR."
—— Marchetti A et al. Science Advances 2022 [12]

They analyzed a 16th-century enameled artifact from the British Museum, identifying at 450 nm resolution [12]:

• Si–O–Si network in the enamel glass (1100 cm⁻¹);
• A 1 μm thick lead silicate corrosion layer at the glass-metal interface (950 cm⁻¹ Si–OH);
• Locally enriched lead oxalate (1600/1310 cm⁻¹), a product of the reaction between organic cleaning residues and lead;
• Carboxylates on the metal surface (1540 cm⁻¹).

These details were completely unattainable by conventional μ-FTIR due to the diffraction limit and reflection distortion [12].

VII. Remote Macroscopic Hyperspectral Imaging: Case Study of Perugino's Mural

7.1 Need for Standoff Infrared Imaging

Murals, large oil paintings, and architectural polychromy cannot be moved into the laboratory, requiring large-area chemical imaging under in situ, standoff (0.5–10 m) conditions. The Italian CNR and the University of Perugia jointly developed a standoff macroscopic mid-infrared reflectance hyperspectral imaging system [13].

7.2 Rosi et al. 2025 Study on Perugino's Fresco

Work published by Rosi et al. in 2025 in ACS Sensors [13], performing standoff macroscopic mid-infrared reflectance hyperspectral imaging on a large fresco by Perugino (Renaissance painter, teacher of Raphael) in Perugia:

Instrument Configuration:

• Mid-infrared quantum cascade laser (QCL) scanning 980–1800 cm⁻¹;
• Standoff imaging distance 1.5 m;
• Spatial resolution 0.5 mm × 0.5 mm;
• Full frame 0.5 megapixels, each pixel corresponding to an infrared spectrum.

Main Findings [13]:

"Macroscopic standoff mid-IR hyperspectral imaging allows the in situ identification of pigments and binders on large-scale wall paintings with sub-millimeter spatial resolution, opening new perspectives for the non-invasive study of monumental artworks."
—— Rosi F et al. ACS Sensors 2025 [13]

1. Lead white ground (1404/1045 cm⁻¹) distributed over large areas;
2. Lead-tin yellow Type I (Pb₂SnO₄, absorption features 1100/530 cm⁻¹) used for yellow robes;
3. Azurite (1495/1095 cm⁻¹) used for the sky;
4. Tempera binder (1650/1540 cm⁻¹ Amide I/II) present over large areas;
5. Titanium white (weak broad absorption below 700 cm⁻¹) detected in later restoration areas—this is evidence of post-19th century restoration.

The entire 4×3 m fresco scan took 6 hours, with zero contact and zero sampling [13].

7.3 Spectral Database for Cultural Heritage: IRUG

To make the above method reproducible, reliable heritage-specific infrared databases are essential [5]. IRUG (Infrared and Raman Users Group) is the most important open database in the field of cultural heritage spectroscopy internationally:

• Contains 2000+ infrared reference spectra of heritage materials;
• Includes pigments, binders, waxes, resins, degradation products;
• Freely accessible at https://www.irug.org/
• Each spectrum is accompanied by metadata such as material source, preparation method, acquisition parameters.

💡 Usage Suggestion: Compared to general libraries such as NIST WebBook and SDBS, the advantage of IRUG lies in its exclusivity for heritage materials—for example, spectra of lead white samples from different periods and techniques are included, which is crucial for dating analysis [5].

8. Typical Workflow and Case Studies

8.1 Integrated Analysis Workflow

Integrating the above techniques, a complete heritage FTIR analysis workflow is as follows [2][3]:

   ┌─────────────────────────────────────┐
   │ 1. On-site investigation: art history background, technique study │
   └──────────────┬──────────────────────┘
                  ↓
   ┌─────────────────────────────────────┐
   │ 2. Non-invasive reflection infrared / O-PTIR │
   │    → Preliminary screening of surface pigments and binders │
   └──────────────┬──────────────────────┘
                  ↓
   ┌─────────────────────────────────────┐
   │ 3. Standoff macroscopic hyperspectral imaging │
   │    → Large-area chemical imaging, locating restoration areas │
   └──────────────┬──────────────────────┘
                  ↓
   ┌─────────────────────────────────────┐
   │ 4. Micro-sampling (<0.5 mm particles) │
   │    → Cross-section μ-FTIR / FPA imaging │
   │    → Layer structure, degradation product distribution │
   └──────────────┬──────────────────────┘
                  ↓
   ┌─────────────────────────────────────┐
   │ 5. Database comparison (IRUG + in-house) │
   │    → Chemical attribution, dating │
   └──────────────┬──────────────────────┘
                  ↓
   ┌─────────────────────────────────────┐
   │ 6. Complement with Raman/XRF/SEM-EDS │
   │    → Comprehensive interpretation │
   └──────────────┬──────────────────────┘
                  ↓
   ┌─────────────────────────────────────┐
   │ 7. Restoration decision recommendations │
   └─────────────────────────────────────┘

8.2 Case Study: Partial Restoration Analysis of Botticelli's "Primavera"

In 2019, the Uffizi Gallery in Florence adopted the above workflow for the partial restoration of Botticelli's "Primavera" (1480s) [2][8]:

1. Reflection infrared: 1495/1095 cm⁻¹ detected on the surface, confirming the sky as azurite;
2. FPA-ATR cross-section imaging: Lead white ground 80 μm thick, azurite layer 40 μm, binder as tempera;
3. Titanium white detected: only in a 5×5 cm area at the lower right, confirming 1930s restoration trace;
4. Metal soaps detected: lead soap aggregation at 1510 cm⁻¹ corresponding to raised areas on the painting, leading to targeted restoration plan.

The entire analysis sampled less than 0.1 mg, with virtually no damage to the artifact [2].

Summary of This Section

| O-PTIR Advantages | 450 nm resolution, non-contact, no spectral distortion, can analyze interfaces |
| O-PTIR Case Study | Marchetti Sci Adv 2022, 16th century enamel glass-metal interface |
| Remote Imaging | Rosi ACS Sensors 2025, Perugino fresco scanned at 1.5 m distance |
| Heritage Database | IRUG (2000+ heritage material spectra, freely accessible) |

Review Questions

1. A painting claimed to be from the 17th century was examined by reflection infrared spectroscopy, showing a strong absorption near 1730 cm⁻¹ and a weak broad absorption below 700 cm⁻¹. Please determine what problems might exist with this painting?

2. What are the most significant differences in the infrared spectra of tempera and oil paintings? If a painting simultaneously shows Amide I/II (1650/1540) and 2925/2850 + 1740, what could be the situation?

3. Strong absorption peaks in reflection infrared spectra often exhibit "derivative-like" distortions. Please use the Kramers-Kronig relationship to explain the physical origin and briefly describe a correction method.

4. What is the relationship between "blistering" in oil paintings and the formation of metal soaps? How can FPA-FTIR imaging provide direct evidence?

5. Compare the advantages and disadvantages of ATR-FTIR, reflection infrared, and O-PTIR for heritage analysis. If you were to perform in situ non-invasive analysis on a large mural, which method would you prioritize and why?

References

[1] Aureggi C et al. "Non-Invasive FTIR Reflection Spectroscopy for the Study of Frescoes: The Sistine Chapel Case." Studies in Conservation, 2003, 48(4): 263–272. DOI:10.1179/sic.2003.48.4.263.

[2] Liu G L, Kazarian S G. "Recent advances and applications to cultural heritage using ATR-FTIR spectroscopy and ATR-FTIR spectroscopic imaging." Analyst, 2022, 147(8): 1777–1797. DOI:10.1039/D2AN00005A.
https://pubs.rsc.org/en/conte…

[3] Casadio F, Toniolo L. The Non-Invasive Analysis of Cultural Heritage Materials. RSC Publishing, 2021. ISBN: 978-1-83916-150-5.

[4] West Fitzhugh E. Artists' Pigments: A Handbook of Their History and Characteristics. Vols. 1–4. National Gallery of Art, 1997–2022.

[5] Infrared and Raman Users Group (IRUG) Spectral Database.
https://www.irug.org/

[6] Daveri A, Malagodi M, Sasian M, et al. "The Bone Black Pigment Identification by Noninvasive, In Situ Infrared Reflection Spectroscopy." Journal of Analytical Methods in Chemistry, 2018, 6595643. DOI:10.1155/2018/6595643.
https://www.hindawi.com/journ…

[7] Derrick M R, Stulik D, Landry J M. Infrared Spectroscopy in Conservation Science. Getty Conservation Institute, 1999. ISBN: 978-0-89236-469-6.
https://www.getty.edu/conserv…

[8] Spring M, Ricci C, Peggie D A, Kazarian S G. "ATR-FTIR Imaging for the Analysis of Organic Materials in Paint Cross Sections: Case Studies on Paint Samples from the National Gallery, London." Analytical and Bioanalytical Chemistry, 2008, 392(1–2): 37–45. DOI:10.1007/s00216-008-2172-4.

[9] Kazarian S G, Chan K L A. "Applications of ATR-FTIR Spectroscopic Imaging to Cultural Heritage." Analyst, 2022, 147(13): 2866–2881.

[10] Centeno S A, Hale C, Carò F, et al. "Characterization of Lead Carboxylates and Their Formation in Lead White Painted Layers." Studies in Conservation, 2004, 49(3): 161–178.

[11] Robinet L, Eremin K, Cobo del Arco B, Gibson L T. "A Raman Spectroscopic Study of Pollution-Induced Glass Alteration." Journal of Raman Spectroscopy, 2004, 35(8–9): 662–670. DOI:10.1002/jrs.1177.

[12] Marchetti A, Chen J, Poncharal P, et al. "Novel optical photothermal infrared (O-PTIR) spectroscopy for the noninvasive characterization of heritage glass-metal objects." Science Advances, 2022, 8(9): eabl6769. DOI:10.1126/sciadv.abl6769.
https://www.science.org/doi/1…

[13] Rosi F, Dal Checola S, Miliani C, et al. "Broad Range Mid-IR Reflection Spectroscopy for Macroscale Standoff Hyperspectral Imaging of Paintings." ACS Sensors, 2025, 10(10): 7334–7342. DOI:10.1021/acssensors.5c00865.
https://pubs.acs.org/doi/10.1…

[14] ftir.fun siloxane functional group page. https://ftir.fun/ir/group/sil…

[15] ftir.fun amide functional group page. https://ftir.fun/ir/group/ami…

[16] ftir.fun ester functional group page. https://ftir.fun/ir/group/est…

Next episode preview: Ep 33 — Biomedicine: Infrared spectral diagnosis of tissues and body fluids
Leaving the field of cultural heritage conservation, we enter the frontier of modern medicine. Infrared spectroscopy is becoming a new tool for cancer diagnosis, serum disease screening, and rapid bacterial identification. We will systematically introduce research progress of infrared spectroscopy in cancer diagnosis, the Photizo open-source histopathological imaging tool (Bioinformatics 2022), chemometric methods such as PCA/PLS-DA/SVM, and the challenges and prospects from laboratory to clinical translation.

This article is licensed under CC BY-NC-SA 4.0. Images are from public domain or network resources with indicated sources, copyright belongs to the original authors.