Multimodal Differentiation of Forensic Fibers Using Infrared and Raman Spectroscopy, Light Microscopy, and SEM-EDS

Fibers are important trace evidence in forensic cases. As a result, it is necessary to develop reliable methods to distinguish between fibers found at the crime scene and those from the suspects. Molecular spectroscopic techniques, including infrared (IR) and Raman spectroscopy, elemental analyses with scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) and polarizing light microscope (PLM), have been successfully applied for fiber characterization. In this study, blended fibers were analyzed with the four instruments in combination. The indicative peaks of the fibers in the IR and Raman spectra were summarized, and their assignments were interpreted. Elemental analysis using SEM-EDS further confirmed the results obtained from molecular spectroscopy. Additionally, significant differences in color and structure of the fibers were observed with PLM, which could be used in the identification and discrimination of the fibers in forensic cases. The information obtained from multiple instruments can complement and confirm each other, enabling accurate identification of the composition and structure of the sample, which can play a particularly important role in forensic cases.

Trace evidence analysis cannot achieve complete recognition. This is determined by the large stock and wide distribution of trace evidence. But the actual requirement for forensic cases is to approach the same determination or completely exclude it as much as possible. The higher the probability of physical evidence appearing, the lower the possibility of achieving the same recognition or complete exclusion. In contrast, physical evidence with a lower probability of appearance is more likely to allow for the same recognition or complete exclusion.

There is a large stock of the materials with the same type, and their physical and chemical properties are similar, making it difficult to distinguish them. However, the similarity between items of the same type is merely a similarity. There will always be differences between two nonhomologous individuals and these differences can be used to distinguish them. So, it is crucial for us to find a method or a physical quantity (characteristic reflector) to characterize this difference. For organic matter samples, SEM-EDS can determine the types of elements present in the sample;^1-4 IR can identify the types of functional groups and chemical bonds;^5-7 mass spectrometry (MS) can determine the molecular weight and formula;^8-9 ultraviolet-visible (UV) spectroscopy can reveal the type and size of conjugated systems and chromophores in molecules;¹⁰ nuclear magnetic resonance carbon spectroscopy (¹³C NMR) can provide detailed structural information about the number, type, and arrangement of molecular carbon atoms.^11-12 By using various instrument methods and characterizing the characteristics of substances from multiple perspectives, it is possible to uncover more differences between different individuals, systematically exclude nonhomologous individuals, gradually approach the same identification, or completely exclude, or calculate the matching degree between on-site test materials and suspected samples.^13-14

The information obtained from multiple instruments can not only complement and confirm each other, allowing for accurate identification of the samples’ composition and structure, but also reduce the difficulty of relying on a single instrument for identification. For example, when comparing two pieces of clothing, comparable features such as color, size, style, fiber variety, spinning thread thickness, weaving method, stitch size, button type and size, dye type, suture type, lining fiber type and color, as well as defect characteristics can all be considered. The more careful the inspection, the more features are compared, and higher the rate of discrimination. Each feature acts as a reflector, confirming and complementing each other, and leading to conclusions that are more accurate and reliable.

It is the same when fiber samples are found as physical evidence. Synthetic fibers are made from polymer compounds, with commonly used types including polyester, nylon, acrylic, chlorinated fiber, vinylon, spandex, and polyolefin elastic yarn. Natural fibers are obtained from plants and animals and are important materials in the textile industry.¹⁴ It is of great importance to differentiate these numerous types of fibers for forensic scientists to discover the exact investigation scope or minimize the number of suspects in robbery or murder cases, especially in hit-and-run cases, in which fibers collected from the suspect cars are compared with the fibers from the clothes of the victims.

Microscopical analysis by Fourier transform infrared (FT-IR) spectroscopy has been the preferred method to obtain additional information about the polymeric composition of fiber.^15-21 Vibrational spectroscopy has been an important physical method for studying fibers for more than 80 years. There have been numerous studies on forensic analysis of fibers using Raman spectroscopy.²² Both IR and Raman methods have proven to be accurate and fast. However, there are still some kinds of fibers which cannot easily be discriminated by vibrational spectroscopy. Therefore, many other analytical instruments are necessary to accurately characterize the fibers, such as X-ray fluorescence spectrometry, mass spectrometry (MS), gel permeation chromatography, and others.^23-25 In this study, a combined analytical approach utilizing four instrumental techniques was employed to examine the blended fibers. Characteristic peaks in both FT-IR and Raman spectra were systematically summarized and assigned, while elemental analysis from SEM-EDS further validated the molecular spectroscopy results. Additionally, PLM revealed distinct color and structural characteristics of fibers, providing crucial evidence for forensic fiber identification. This multimodal methodology, through comprehensive cross-verification of results, enables more accurate determination of sample composition and structural features, demonstrating significant application value in forensic investigations, particularly for cases lacking surveillance evidence (for example, hit-and-run accidents).

Materials and Methods

A hit-and-run traffic accident occurred in a certain area, and a tear was found on the lower left leg of the victim’s pants. Following an investigation, the police discovered a suspect car and extracted fibers from the right end of the front bumper. To determine if contact occurred, the fibers extracted from the front bumper were compared with the fibers taken from the victim’s pants.

Two small sample packages were submitted for inspection: one contained yarn, and the other contained fabric. The fiber labeled "Sample S" (suspected) on the packaging bag was extracted from the car’s bumper and was referred to as the inspection material. The fiber labeled "Sample V" (victim) on the bag referred to the fabric extracted from the tear on the victim’s pants, abbreviated as the sample. The laboratory was asked to determine whether the fiber types in the inspection material and sample were of the same kind.

A comprehensive suite of analytical instruments was utilized in this study. A Spectrum GX 2000 system (PerkinElmer) equipped with a diamond anvil cell, a deuterated triglycine sulfate (DTGS) detector, and a spectra 5.01 workstation was used for IR observation. The spectral background was subtracted for every measurement. Spectra were collected in high-resolution mode in transmission mode. Triplicate tests were performed at different sites for each sample to ensure accuracy. For Raman spectroscopy, a Renishaw inVia confocal Raman microscope system with two lasers emitting at 532 nm and a charge-coupled device (CCD) detector was used. The laser power was optimized between 1% and 10% to achieve a sufficiently strong signal. Prior to sample measurement, the Raman system was calibrated using crystal silicon with fixed peak position at 520 cm^-1. The Raman spectra were collected in WiRE software in extensive mode ranging from 200–3200. Every spectrum was obtained in 4 accumulations to enhance the signal/noise ratio. A Leica DM6000 M PLM system was employed for the fiber optical observation, and an FEI Quanta 600 SEM with a high-vacuum backscattered electron detector and Oxford X-MAX EDS was employed for elemental analysis. The working distance was set at 10 mm; the accelerating voltage at 20 kV; and the spot was set at 6.0 to optimize imaging and elemental detection.

Results and Discussion

IR spectra

The IR spectra of Sample S are illustrated in Figure 1. Both polyethylene terephthalate (PET) (1) and cellulose (2) were detected in the IR spectra of Sample S at different areas. For polyethylene terephthalate, the peak at 3056 cm^-1 corresponds to the CH stretching vibration of the benzene ring. The peaks at 1616 cm^-1, 1579 cm^-1, and 1504 cm^-1 are attributed to the stretching vibrations of the benzene ring skeleton. The peak at 1021 cm^-1 represents the in-plane angular vibration of two adjacent =CH on the para-substituted benzene ring. The peak at 731 cm^-1 is the out-of-plane angular vibration of two adjacent =CH planes on the 1,4-para-substituted benzene ring. The nonpolar 1,4-para-substitution of the benzene ring results in an out-of-plane angular vibration of the =CH group, typically detected in the range of 860-800 cm^-1. For example, in the IR spectra of polycarbonate, bisphenol A, and epoxy resins, the out-of-plane angular vibration of =CH is found around 830 cm^-1. However, in polyester molecules, both substituents are ester groups, which are polar and undergo π-π conjugation with the benzene ring, reducing the out-of-plane angular vibration frequency of CH to 731 cm^-1. The peak at 2970 cm^-1 and 2909 cm^-1 correspond to the antisymmetric and symmetric stretching vibrations of CH₂, respectively. Because of the strong electronegativity of the adjacent carboxyl group, the frequencies of these stretching vibrations are higher than the typical values (2925 cm^-1 and 2855 cm^-1). The peak at 1457 cm^-1 represents the left-right CH₂ in-plane angular vibration, while the peak at 1409 cm^-1 is because of the in-plane angular vibration of O-CH₂. Since CH₂ is connected to oxygen atoms, which have strong electronegativity, the in-plane angular vibration frequency is lower than the normal value (1465 cm^-1). The peak at 1340 cm^-1 represents the out-of-plane oscillation vibration of trans-CH₂. The peaks at 850 cm^-1 and 794 cm^-1 are in-plane oscillation vibrations of CH₂. The peak at 1721 cm^-1 corresponds to the stretching vibration of C=O, with its frequency doubling detected at 3434 cm^-1. The π-π conjugation formed between the carbonyl and benzene rings reduces the double bond properties of carbonyl groups, lowering its stretching vibration frequency compared to that of saturated fatty ester carbonyl groups (1740±10 cm^-1). The peak at 1254 cm^-1 corresponds to the antisymmetric stretching vibration of C-O-C, and 1105 cm^-1 is because of the symmetric stretching vibration of C-O-C. One of the characteristics of the IR spectrum of terephthalic acid esters is that the three peaks at 1721 cm^-1, 1254 cm^-1, and 1105 cm^-1 have similar intensities. As an aromatic ester, polyester exhibits a higher antisymmetric stretching vibration frequency for C-O-C compared to that of fatty acid esters, which are typically in the range of 1240-1160 cm^-1 for antisymmetric stretching and 1050–1000 cm^-1 for symmetric stretching. Finally, the peak at 974 cm^-1 represents the out-of-plane angular vibration of the trans C-O bond. In the IR spectrum of blended fabrics, the presence of bands around 730 cm^-1, 1105 cm^-1, and 1021 cm^-1 can generally confirm the presence of polyester in the blended fabric.

Cellulose contains methylene, methylene–ether bonds, and hydroxyl groups. Therefore, the IR spectrum can mainly reflect the absorption of these functional groups and chemical bonds. The broad and strong absorption around 3340 cm^-1 corresponds to the stretching vibration of the associated hydroxyl (OH) group. The strongest spectral band for natural fibers is located near 1056 cm^-1, with a deep well shape, accompanied by several weak shoulder peaks on both sides of the main peak. These bands represent the combined effects of C-O-C ether bond stretching and hydroxyl in-plane bending vibration within the fiber molecules. The peak at 2901 cm^-1 corresponds to the stretching vibration of CH and CH₂.

The 1636 cm^-1 peak is because of the bending vibration of adsorbed water. The 1429 cm^-1 peak represents the in-plane angular vibration of O-CH₂. The in-plane bending vibration of COH occurs between 1370 and 1315 cm^-1. The broad and strong absorption of approximately 667 cm^-1 is attributed to the oscillation of adsorbed water OH. The 617 cm^-1 peak represents the out-of-plane bending vibration of cellulose COH. As shown in Figure 1, Sample S is composed of polyethylene terephthalate and cellulose. The IR spectrum of Sample V also indicated that cellulose was the main component, so the spectrum was not illustrated again. However, since cellulose is the primary component of cotton, ramie, jute, hemp, banana, sisal, kapok, viscose fibers, and copper ammonia fibers, further analysis was still needed for differentiation.

Because of the widespread use of blending technology in fabrics, analyzing fiber samples using IR spectroscopy is challenging. The high resolution and small view of IR microscopy may lead to missed judgments when only a single long fiber or spot within the yarn is selected for analysis. In this case, if only a long-fiber yarn is investigated, it may be misidentified as either polyethylene terephthalate or cotton. However, the sample in this case is a blend of polyethylene terephthalate and cotton. It suggested that more yarns should be selected for analysis to avoid erroneous judgments in IR analysis.

PLM observation

As shown in Figure 2, polyethylene terephthalate and cotton were detected in Sample S with a polarized microscope. Figure 3 showed that viscose fiber was detected in Sample V.

Figure 2a presents a typical picture of polyethylene terephthalate. Polyester has the highest birefringence index among textile fibers, resulting in the most vibrant polarized light interference patterns. Under a microscope, the polyester fibers appear flat and smooth, with a uniform diameter, straight alignment and a full structure. The fiber cross-section displays a circular morphology, exhibiting parallel interference fringes aligned along the longitudinal axis with symmetrical chromatic distribution. Typically, interference fringes in red, yellow, green, and cyan are observed. In both fully extinct and semi-extinct fibers, the use of titanium dioxide (TiO₂) as an extinction agent can produce noticeable black spots under the microscope. Figure 2b presents a typical image of cotton. Under the microscope, cotton fibers appear as flat, ribbon-like strips with natural twists along their length. Mature cotton fibers exhibit approximately 40–65 twists per centimeter. At these twist points, the cellulose layer has changed direction, and under cross-polarized light, a distinct color variation can be observed. The polarization colors of cotton fibers vary with the maturity of the cotton fibers. Mature fibers generally appear orange or yellow-blue with a deeper color tone, while less-mature fibers display interference patterns in shades of sky blue or dark blue. Figure 2 confirms that polyethylene terephthalate and cotton are the main components of Sample S.

Under the microscope, ordinary adhesive fibers exhibit patterns on their longitudinal surface. The cross-section is divided into two layers: the cortex and the core layer, with serrated edges along the cross-sectional boundary. The formation of this skin-core structure occurs during the fiber manufacturing process. Because of the low viscosity of the adhesive and the high concentration of acid, alkali, and salt used in production, along with a relatively fast spinning speed, the filament solidifies in the coagulation bath. In this process, the surface solvent is first released, causing the cellulose to spin. However, the solvent in the inner core cannot be released simultaneously with the outer layer and is only gradually released in later stages. This results in the pre-formed cortex shrinking, wrinkling, and forming stripes. Figure 3 shows the detection of viscose fiber in Sample V.

When the chemical composition of a fiber is identified as cellulose, it is important to focus on distinguishing distinguishing between cotton and other types of cellulose fibers, such as hemp, ramie, and flax. Under a polarizing microscope, hemp displays bright colors and smooth surfaces. The vertical interference fringes appear in colors such as red, yellow and green, bright and full, while the horizontal fringes show numerous knot patterns resembling palms or bamboo, or thick bamboo-like knots. Figure 4 illustrates the distinction between (a) jute, (b) ramie, (c) thread hemp, and (d) cotton with PLM.

Raman Spectra

Figure 5 shows the Raman spectra of Sample S, while Figure 6 presents the Raman spectra of polyethylene terephthalate. Figure 7 shows the spectra of cellulose. It can be summarized that the main peaks in Figure 4 (1729 cm^-1, 1616 cm^-1, 1291 cm^-1, 859 cm^-1, 1121 cm^-1, 1097 cm^-1, and 899 cm^-1) are the result of the overlap between Figure 5 and Figure 6, which also indicates the presence of polyethylene terephthalate and cellulose in Sample S.

From Figure 6, the main spectral peaks of polyethylene terephthalate include the stretching vibrations of the carbonyl C=O at 1726 cm^-1, stretching vibrations of benzene ring at 1613 cm^-1, bending vibrations of CH₂ at 1437 cm-¹, antisymmetric stretching vibrations of C-O at 1293 cm^-1, stretching vibrations of C-O at 1126-1060 cm^-1, swinging vibrations of CH and C=O at 857 cm^-1, and bending vibrations of para substituted benzene ring at 631 cm^-1.

From Figure 7, the main Raman shifts of cotton are 1339 cm^-1, 1122 cm^-1, 1096 cm^-1, 1054 cm^-1, 998 cm^-1, 899 cm^-1, 450 cm^-1, and 355 cm^-1, respectively. The peaks at 1096 cm^-1 (asymmetric stretching of C-O-C, glycosides), 1122 cm^-1 (symmetric stretching of C-O-C, glycosides), and 899 cm^-1 (C-O-C in-plane expansion and loop expansion) are indicative in the identification process. Unlike IR spectroscopy, Raman spectroscopy directly displays the superposition of spectra from the two fiber components. This might be because of the larger area of the sample focused on and scanned by Raman spectroscopy, or because the area focused on by Raman spectroscopy coincidentally overlaps where the two components are present.

SEM-EDS

In the production of polyethylene terephthalate, TiO₂ is an important additive. It can enhance the weather resistance, brightness, and glossiness of the product by reacting with dicarboxylic acids in raw materials. TiO₂ can effectively absorb UV rays, preventing aging and fading caused by sunlight exposure, thus extending the lifespan of the product. TiO₂ also provides several key functions and characteristics. It has excellent whiteness and covering performance, improving the whiteness and color of polyester. Additionally, TiO₂ offers antibacterial, antifungal, and antistatic properties, providing comprehensive protection for the performance of polyester products.

The application of TiO₂ in the polyester spinning process involves incorporating TiO₂ into the polyester mixture in the form of a slurry, ensuring thorough integration with the polyester raw material. During mechanical stirring, the TiO₂ slurry is evenly distributed throughout the polyester, which helps maintain the uniformity and stability of the product. Figure 8 presents the elemental analysis results of Sample S (1) and Sample V (2) by SEM-EDS. Titanium (Ti) was identified in Sample S but not in Sample V, providing further confirmation of the differentiation results obtained from other instruments and ruling out any suspicion regarding Sample V.

Conclusions

The characteristic vibrational peaks of two forensic fiber samples were systematically analyzed using both IR and Raman spectroscopy, with preliminary assignments made to their respective functional groups. These vibrational spectroscopic methods provided valuable information and differentiated the major fiber categories (polyethylene terephthalate and cellulose) successfully. However, PLM provided clear discrimination not only between the primary fiber types, but also enabled identification of specific cellulose variants (cotton versus viscose). SEM-EDS detected the element Ti in the sample, which further confirmed the existence of polyethylene terephthalate in IR and Raman, and the observation of black spots in PLM. This multimodal analytical approach achieved comprehensive discrimination, underscoring the critical need for complementary techniques in forensic fiber analysis to ensure accurate identification and minimize evidentiary misinterpretation.

References

1. Tambuzzi, S.; Gentile, G.; Boracchi, M.; Zoha, R. A Narrative Review of Peculiar Forensic-Pathological Applications of Scanning Electron Microscopy with Energy Dispersive X-ray Analysis (SEM/EDX). Minerva Forensic Med. 2022, 142 (4), 58–65. https://doi.org/10.23736/S2784-8922.22.01820-9.
2. Tambuzzi, S.; Gentile, G.; Primavera, R.; Muccino, E; Zoja, R. Pilot Application of SEM/EDX Analysis on Suspected Cigarette Burns in a Forensic Autopsy Case of Child Abuse. Am. J. Forensic Med. Pathol. 2024, 45 (2), 135–143. https://doi.org/10.1097/PAF.0000000000000922.
3. Pathak, J. S.; Saxena, A. K. Significance of Using Scanning Electron Microscopy with Energy Dispersive X-ray Spectrometry (SEM-EDX) for Analysis of Evidence Material in Forensic Sciences. Asian J. Res. Rev. Phys. 2024, 8 (2), 1–11. DOI: 10.9734/ajr2p/2024/v8i2159
4. Nguyen-Hong, M.; Payré, B.; Aries, S.; Dedouit, F.; Telmon, N., Savall, F.; Guilbeau-Frugier, C. Comparison of SEM-EDX and ICP-MS for the Analysis of Gunshot Residues after Burial: An Experimental Study. Forensic Imaging 2024, 39, 200604. https://doi.org/10.1016/j.fri.2024.200604.
5. Alkhuder, K. Attenuated Total Reflection-Fourier Transform Infrared Spectroscopy: A Universal Analytical Technique with Promising Applications in Forensic Analyses. Int. J. Legal Med. 2022, 136 (6), 1717–1736. https://doi.org/10.1007/s00414-022-02882-2.
6. Mistek-Morabito, E.; Lednev, I. K. Discrimination of Menstrual and Peripheral Blood Traces Using Attenuated Total Reflection Fourier Transform-Infrared (ATR FT-IR) Spectroscopy and Chemometrics for Forensic Purposes. Anal. Bioanal. Chem. 2021, 413 (9), 2513–2522. https://doi.org/10.1007/s00216-021-03206-w.
7. Sharma, S.; Chophi, R.; Kaur, C.; Singh, R. Chemometric Analysis on ATR-FT-IR Spectra of Spray Paint Samples for Forensic Purposes. J. Forensic Sci. 2021, 66 (6), 2190–2200. https://doi.org/10.1111/1556-4029.14806.
8. Sisco, E.; Forbes, T. P. Forensic Applications of DART-MS: A Review of Recent Literature. Forensic Chem. 2021, 22, 100294. https://doi.org/10.1016/j.forc.2020.100294.
9. Ferrari Júnior, E.; Caldas, E. D. Determination of New Psychoactive Substances and Other Drugs in Postmortem Blood and Urine by UHPLC-MS/MS: Method Validation and Analysis of Forensic Samples. Forensic Toxicol. 2022, 40 (1), 88–101. https://doi.org/10.1007/s11419-021-00600-y.
10. Steiner, A.; Lurie, I. Applicability of Liquid and Supercritical Fluid Chromatographic Separation Techniques with Diode Array Ultraviolet Detection for Forensic Analysis. Forensic Chem. 2021, 26, 100359. https://doi.org/10.1016/j.forc.2021.100359.
11. Castaing-Cordier, T.; Ladroue, V.; Besacier, F.; Bulete, A.; Jacquemin, D.; Giraudeau, P.; Farjon, J. High-Field and Benchtop NMR Spectroscopy for the Characterization of New Psychoactive Substances. Forensic Sci. Int. 2021, 321, 110718. https://doi.org/10.1016/j.forsciint.2021.110718.
12. Arantes, L. C.; Silveira, D.; Magno Sousa de Azerêdo, G.; Enrique Estrada Semprum, O.; Lima de Oliveira, A.; Benedito, L.; Lião, L.; Oliveira, G. Enhancing Forensic Laboratories Through University Collaboration: Obtaining Conclusive Reports and Reference Materials via NMR. Magn. Reson. Chem. 2025, 63 (2), 155–166. https://doi.org/10.1002/mrc.5497.
13. Džodić, J.; Milenković, D.; Marković, M.; Marković, Z.; Dimić, D. Application of Quantum-Chemical Methods in the Forensic Prediction of Psychedelic Drugs’ Spectra (IR, NMR, UV-VIS, and MS): A Case Study of LSD and Its Analogs. Appl. Sci. 2023, 13 (5), 2984. https://doi.org/10.3390/app13052984.
14. Ali, Z.; Talpur, F. N.; Afridi, H. I.; Ahmed, F.; Brohi, N. A.; Abbasi, H. Analytical Approaches and Advancement in the Analysis of Natural and Synthetic Fiber: A Comprehensive Review. Spectrochim. Acta, Part A 2025, 326, 125164. https://doi.org/10.1016/j.saa.2024.125164.
15. Liu, Y. Vibrational Spectroscopic Investigation of Australian Cotton Cellulose Fibres. Part 1. A Fourier Transform Raman Study. Analyst 1998, 123 (4), 633–636. https://doi.org/10.1039/A707064K.
16. Frank, R. S.; Sobol, S. P. Fibres and Their Examination in Forensic Science. In Forensic Science Progress; Springer: Berlin, 1990; pp 41–125. https://doi.org/10.1007/978-3-642-75186-8_3.
17. Tungol, M. W.; Bartick, E. G.; Montaser, A. Forensic Examination of Synthetic Textile Fibers by Microscopic Infrared Spectrometry. In Practical Guide to Infrared Microspectroscopy; Marcel Dekker: New York, 1995.
18. Kirkbride, K. P. Infrared Microspectroscopy of Fibres. In Forensic Examination of Fibers; Robertson, J.; Roux, C., Eds.; CRC Press: Boca Raton, FL, 2017; pp 245–288.
19. Tungol, M. W.; Bartick, E. G.; Montaser, A. The Development of a Spectral Database for the Identification of Fibers by Infrared Microscopy. Appl. Spectrosc. 1990, 44 (4), 543–549.
20. Tungol, M. W.; Bartick, E. G.; Montaser, A. Spectral Data Base for the Identification of Fibers by Infrared Microscopy. Spectrochim. Acta, Part B 1991, 46 (11), 1535–1544. https://doi.org/10.1016/0584-8547(91)80203-F.
21. Aljannahi, A.; Alblooshi, R. A.; Alremeithi, R. H.; Karamitsos, I.; Ahli, N. A.; Askar, A. M.; Albastaki, I. M.; Ahli, M. M.; Modak, S. Forensic Analysis of Textile Synthetic Fibers Using a FT-IR Spectroscopy Approach. Molecules 2022, 27 (13), 4281. https://doi.org/10.3390/molecules27134281.
22. Zapata, F.; Ortega-Ojeda, F. E.; García-Ruiz, C. Forensic Examination of Textile Fibres Using Raman Imaging and Multivariate Analysis. Spectrochim. Acta, Part A 2022, 268, 120695. https://doi.org/10.1016/j.saa.2021.120695.
23. Takahara, H.; Matsuda, W.; Kusakabe, Y.; Ikeda, S.; Kuraoka, M.; Komatsu, H.; Nishiwaki, Y. An Evaluation of Total Reflection X-ray Fluorescence as a Tool for Forensic Discrimination of Single Polyester Fibers. Anal. Sci. 2021, 37, 1123–1129. https://doi.org/10.2116/analsci.20P348.
24. Zughaibi, T. A.; Steiner, R. R. Forensic Analysis of Polymeric Carpet Fibers Using Direct Analysis in Real Time Coupled to an AccuTOF Mass Spectrometer. Polymers 2021, 13 (16), 2687. https://doi.org/10.3390/polym13162687.
25. Matsushita, R.; Watanabe, S.; Iwai, T.; Nakanishi, T.; Takatsu, M.; Honda, S.; Funaki, K.; Ishikawa, T.; Seto, Y. Forensic Discrimination of Polyester Fibers Using Gel Permeation Chromatography. Forensic Chem. 2022, 30, 100428. https://doi.org/10.1016/j.forc.2022.100428.