Micro-Raman spectroscopic study of artificially aged natural and dyed wool

JOURNAL OF RAMAN SPECTROSCOPY J. Raman Spectrosc. 2008; 39: 638–645 Published online 11 March 2008 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/jrs.1899 Micro-Raman spectroscopic study of artificially aged natural and dyed wool Brenda Doherty1 , Costanza Miliani2∗ , Ina Vanden Berghe3 , Antonio Sgamellotti1,2 and Brunetto Giovanni Brunetti1 1 Dipartimento di Chimica, Università degli Studi di Perugia, via Elce di Sotto, 8, I-60123 Perugia, Italy Istituto CNR di Scienze e Tecnologie Molecolari (ISTM), Dipartimento di Chimica, Università degli Studi di Perugia, via Elce di Sotto, 8, I-60123 Perugia, Italy 3 Koninklijk Instituut voor het Kunstpatrimonium, Jubelpark 1, B-1000 Brussels, Belgium 2 Received 24 July 2007; Accepted 6 November 2007 The objective of this study was to evaluate the use of micro-Raman spectroscopy as a non-invasive vibrational spectroscopic technique applied to the examination of wool samples, which may be applied to textile materials of cultural heritage interest. In this work, a selection of wool materials were primarily investigated in their unaged states through the utility of a natural wool reference together with selected samples dyed with different natural colorants, namely woad, weld and madder. The identification of the main modes of vibration of the wool fibre keratin was assessed in all the samples, which aided the determination of the changes within the protein structure, in particular, through the cysteine and peptide cross-linkages brought about by the addition of the dyes that can produce effects similar to degradation. The dye too was assessed importantly to enable its identification through its characteristic scattering or fluorescence emissions on a woollen matrix, as well as to ascertain whether a uniform covering across the surface of the wool was achieved or not. Regarding the artificial degradation of the samples it was possible to observe numerous modifications within the molecular structure of the wool, in particular, within the amide I, C–H bending, amide III and S-S stretchings along with the physical photo-yellowing of fibres given by the presence of lipids dispersed across the surface of the wool. The effects of ageing on the dyed samples were also investigated, indicating that many of the bands relative to the colorants were still present, yet so too were numerous vibrations from the wool that also indicated a certain level of stress and degradation to the underlying wool. Copyright  2008 John Wiley & Sons, Ltd. KEYWORDS: keratin; micro-Raman spectroscopy; dyed wool; artificial ageing INTRODUCTION The basic building blocks of wool fibre are the fibrous, sulphur-containing ˛-keratins that are part of the intermediate filament super family of proteins.1 These proteins are almost completely helical, consisting of ˛-helical segments, linked by non-helical sequenced stretches. Matrix proteins are noted for their high content of either cysteine residues or glycine and tyrosine residues and are thought to surround the intermediate filaments at a later stage in the development of the follicle interacting with them through inter-molecular disulphide bonds.2 These keratin tissues have regions of high-sulphur and low-sulphur content, which provides a Ł Correspondence to: Costanza Miliani, Istituto CNR di Scienze e Tecnologie Molecolari (ISTM), Dipartimento di Chimica, Università degli Studi di Perugia, via Elce di Sotto, 8, I-60123 Perugia, Italy. E-mail: miliani@thch.unipg.it Copyright  2008 John Wiley & Sons, Ltd. direct correlation with the amount of cystine, and hence the –S–S–disulphide bonds, present in that particular part of the tissue. Keratins, the naturally occurring proteins, owe their rigidity and strength to the sulphur–sulphur cross-linking between cysteine amino acid residues and extensive intramolecular hydrogen bonding.3 Although keratins are complex residues, they are classified as ‘hard or soft’ on the basis of their sulphur content, the soft deriving mainly from ˛-helical keratin and the soft form layers of the ˇ-sheet conformation with the former being predominantly found in mammalian keratins while the latter in reptilian keratin.4 It is noted that irreversible changes in the structures of wool fibres can be induced by physical or chemical damage by chlorination and bleaching processes, application of dyes or by induced photodegradation.5 Sunlight too, is an important cause where damage is caused mainly by Artificially aged natural and dyed wool short wavelength ultraviolet light by forcing natural fibres to undergo photo-bleaching (caused by blue light, i.e above 400 nm), followed by yellowing and tendering of the fibres (caused mainly by UV radiation between 290–310 nm). This is because both the peptide and cystine cross-linkages of wool are destroyed and a new range of ionic groups are formed,6 causing a reduction in the tenacity of the fibre. It has been further suggested that yellowing in wool is due to environmental effects on the lipids situated on the surface of the fibre.7 It is also recognised that although some dyes accelerate the loss of strength in the fibre on exposure to sunlight, treating the fibre with aluminium salts followed by dyeing with mordant dyes has the opposite effect.8 Studies show that significant changes occur in the molecular structure of wool when it is exposed to stimulated sunlight9 as the disulphide bonded cystine amino acid residues are oxidised with cysteic acid being the major product, there is a reduction in the ˛-helical content, while random coil or ˇ-sheet protein chain conformations are favoured. There are structural changes in the fibre as a consequence of cystine photo-oxidation and tryptophan amino acid residues in the wool fibre are degraded upon exposure to light, also causing photo-yellowing in the fibre. In materials of artistic interest in the field of cultural heritage, coloured organic materials are often used as pigments in paintings, in various media, either singly or commonly absorbed on an inorganic support (aluminium hydroxide, calcium carbonate or calcium sulphate) as lake pigments. Yet the most obvious use of natural dyestuff is the one used in textiles. For natural fibres, vegetable and animal (cellulose and protein), exist various categories of suitable dyes, namely direct dyes, vat dyes, acid dyes and mordant dyes. In natural protein fibres, there are many ways in which a dye can bind to the fibre. It may form covalent or ionic bonds to the -NH2 and/or -CO2H groups on the ends of the polymer or form similar interactions with the amino acid side chains (R). It is of paramount importance then that traditional methods for dyeing wool use elevated temperatures that can damage the fibres and require absorption enhancers that create pollution concerns. Since keratins are responsible for the major structural and mechanical properties of wool fibres, their scientific classification and identification have been of continued interest since the first attempt to fractionate them in 1935.4 In particular, Raman spectroscopy can be considered a standard method for assessing wool fibre keratin, not only in its natural state but also by determining any changes within the protein structure brought about by the action of dyes and natural or artificially enforced degradation. This article focuses on the use of this non-invasive vibrational spectroscopic technique applied to wool samples, namely natural wool, dyed wool and aged wool by estimating the keratin and its changes by means of the analysis of the intensity and position of amides as well as the modification of the cystine bonds. Copyright  2008 John Wiley & Sons, Ltd. EXPERIMENTAL Materials All wool samples, otherwise dyed or natural, unaged or aged were obtained in collaboration from the Brussels Royal Institute for Cultural Heritage (Table 1). They were selected out of a large series of wool and silk references that were made at the occasion of a European research project concerning the ‘Monitoring of Damage in Historic Tapestries’ (EVK4-CT-2001-00048). Within this project, wool and silk fibres were mordanted and dyed with different natural dye sources using laboratory recipes based on 16th–17th century recipes. The main literature references for silk dyeing were found in ‘The Plictho of Gioanventura Rosetti (Venice, 1548) and 17th century manuscripts of the Six and Kerspin Family (Holland), while wool was dyed according to historic recipes found in ‘Tbouck van Wondre’ (Brussels, 1513). Accelerated ageing of both natural and dyed wool and silk samples was completed according to EN ISO 105-B02, using a Xenotest 150S Xenon lamp with simulation of sunlight through window-glass by the use of an IR and UV filter, under controlled humidity (65% š2%) and temperature conditions (20 ° C š1%), hence creating a total illuminance at the surface of the samples of 150 klux for 200, 400, 600 and 800 h. Characterisation method, micro-Raman spectroscopy The JASCO Ventuno dispersive micro-Raman spectrophotometer with Nd : YAG laser (532 nm) with a power from 10 to 30 mW was utilised and spectra measurements were collected with continuous scans from 30 to 4700 cm
1 ; exposure times from 10 to 150 s, accumulations: 7–15; lens power of x50 (equivalent spot size of 2 µm); laser power from 1 to 100% and confocal pin hole diameter of 200 or 3000 µm (correspondent spatial resolution of 8–120 µm). Slit width of 150 µm and slit height of 2.5 mm of which the spectral resolution was maintained about 2 cm
1 pixel
1 . All spectra were calibrated with polystyrene. Table 1. Wool samples under study Wool sample Natural wool reference, undyed Indigoid, woad dyed wool Flavonoid, weld dyed wool Anthraquinone, madder dyed wool UNAGED AGED AGED AGED AGED 200 h 400 h 600 h 800 h X X X X X X – X – X X – X – X X – X – X J. Raman Spectrosc. 2008; 39: 638–645 DOI: 10.1002/jrs 639 640 B. Doherty et al. RESULTS AND DISCUSSION Unaged wool The wavenumbers and approximate assignments of the vibrational modes for the Raman spectrum obtained in this study for unaged, natural wool is shown in Table 2. The resulting Raman spectrum is provided in Fig. 1 representing the (CH) region, 3600–2500 cm
1 , the 1800–1200 cm
1 region, which contains the amide I, II and III features10 and information about the skeletal backbone, the (CC) skeletal 1200–1000 cm
1 region and the 1000–400 cm
1 region including the sulphur–sulphur bridging modes of the cystine residues. (CH) region, 3600–2500 cm
1 In each of the wool samples under study, similar vibrational modes in the 3600–2700 cm
1 region were reported. Broad yet well-defined peaks were present for all samples in the 3260 cm
1 region and these have been assigned to N–H stretching vibrations.11 Weak features observed in the 2765–2725 and 3061–3059 cm
1 regions, and were assigned to groups of aliphatic and olefinic C–H stretching vibrations, respectively.12 It was also suggested that the 2930 and 2875 cm
1 bands form a Fermi resonance doublet through the interaction of the overtone of the CH3 symmetric stretching.3 Table 2. Mode assignments for natural wool Washed, unaged wool 3289 mbr 3061 s 2932 vs 2878 ssh 2728 w 1653 s 1621 msh 1587 w 1550 w 1450 s 1317 mbr 1207 m 1174 w 1124 w 1031 w 1002 s 958 wsh 932 w 901 w 852 w 831 w 758 w 644 w 621 w 511 wbr Approximate assignments (NH) symmetric stretch (CH) olifinic ⊲CH3 ⊳ symmetric ⊲CH2 ⊳ symmetric (CH) aliphatic (CO) amide I: ˛ sheet (CC) olifinic  (CC) olifinic υ(NH); (CN) υ⊲CH2 ⊳ scissoring υ⊲CH2 ⊳ deformation (CC) (CC) (CC) skeletal, trans conformation (CC) skeletal, cis conformation (CC) aromatic ⊲CH3 ⊳; υ (CC) ⊲CH3 ⊳ terminal, (CC) ˛-helix ⊲CH2 ⊳ υ(CCH) aromatic tyrosine υ(CCH) aliphatic tyrosine ⊲CH2 ⊳ in-phase (CS) (CS) (S–S) gauche–gauche–gauche Copyright  2008 John Wiley & Sons, Ltd. Figure 1. Micro-Raman spectrum of natural wool. 1800–1200 cm
1 region This region, in particular, contains important information of the keratins, concerning the nature of the amide I and II modes, (CONH) and υ⊲NH2 ⊳, respectively. A strong band is reported at 1652 cm
1 for the samples and was assigned to the (C–O) stretching amide I vibration band, indicating that mammalian keratins exist predominantly in the ˛-helix conformation. The amide II mode appeared in the 1550 cm
1 region with weak intensity; this is due to N–H in-plane bending with contributions from the C–N stretching vibrations. It is possible to assign the band in the region 1621 cm
1 to tyrosine.13 The C–H deformations and the amide III bands are located in the 1500–1200 cm
1 region. Two bands are noted attributed to methylene at 1450 and 1335 cm
1 with the former assigned to the (CH2 ) deformation band (scissoring). 1200–1000 cm
1 The characteristic (CC) and (CN) bands from the keratotic and lipoidal components of the tissues may be noted in this range. Studies of phospholipid systems14 indicate that it is possible to assign the skeletal C–C stretching modes. The 1150–1030 cm
1 skeletal C–C stretching mode gives information on the intra-molecular trans : gauche conformational changes that may occur within the hydrophobic acyl chain matrix15 of the bilayers. Other C–C stretching vibrations include bands at 1156–1154 cm
1 , all resulting from sidechain amino acids, e.g. the υ(OH) assignment at 1156 cm
1 probably arises from tyrosine.16 Clearly, in this region, the most intense band was at 1003–1002 cm
1 , and was assigned to the C–C stretching vibration of the aromatic ring in the phenylalanine side chain, due to the breathing vibration of the monosubstituted ring in phenylalanine.3 1000–400 cm
1 region Within this region, the spectral features of keratin given by the vibrations of the sulphur–sulphur bonds are observed. J. Raman Spectrosc. 2008; 39: 638–645 DOI: 10.1002/jrs Artificially aged natural and dyed wool The approximate relative intensities of the C–S and S–S stretching modes give a good indication of the relative sulphur content and the structural conformation of the –S–S–disulphide linkages providing information about the tertiary and the quaternary structure in keratins.4 In protein spectra, the C–S vibrational band originates from methionine, cysteine and cystine.3 The C–S stretching modes in the spectra of alkylthiols appear in the 730–620 cm
1 region. For cysteine and cystine residues, the –C–S–stretching vibration also depends on the conformation of a side chain with respect to the main chain of the protein. Samples showed two medium weak intensity bands between 644 and 621 cm
1 assigned to the stretching vibrations of C–S bonds. The stretching vibrations of the S–S bond are in the 550–490 cm
1 region. In this study, natural, untreated wool fibre keratin gave weak bands in the 510 and 490 cm
1 region for the S–S stretching mode. This range of bands is commonly found in natural proteins, reflecting that the gauche–gauche–gauche conformation of the C–C–S–S–C–C moiety is the most stable form, thus indicating that naturally occurring proteins and peptides prefer the lowest energy conformation of disulphide bonds.17 (SS) stretching vibrations in this region, given by two weak bands found near 831 and 852 cm
1 were assigned to tyrosine side-chain vibrations involving CCH aliphatic and aromatic deformations, respectively.13 The 830 and 850 cm
1 doublet are considered to be the Raman-active tyrosine bands useful for determining the environment between the ring-breathing vibration and an overtone of an out-of-plane ring-breathing vibration of the para-substituted benzene moiety.18 The relative intensity ratio of this doublet depends on the sitting of tyrosine in the skeletal framework of a protein. In this sample, the 850 cm
1 band was of greater intensity than that at 830 cm
1 , indicating an exposed tyrosine chain.19 Beyond the amide bands, two further bands in the 960–930 cm
1 region have been reported as being characteristic of the ˛-helical keratin conformation.20 Figure 2. Spectra of aged wool between 200 and 800 h exposure. intensity, yet slightly shifted in position. The 3061–3059 cm
1 regions, assigned to groups of aliphatic stretching vibrations, also noted an overall decrease in the intensity. 1800–1200 cm
1 region The strong band, reported in the 1652 cm
1 region for the samples assigned to the (C–O) stretching amide I vibration band, shows a gradual decrease in intensity from 200 h ageing to 800 h ageing. The C–H deformations and the amide III characteristic vibrations in the 1500–1200 cm
1 regions observe changes due to ageing of the fibre. The strongest band, at 1450 cm
1 , assigned to a methylene (CH2 ) deformation band (scissoring) sees its intensity gradually decrease. The methylene band at 1335 cm
1 , observes an overall shift in position, to 1344 cm
1 for 800 h ageing. (CC) skeletal region, 1200–1000 cm
1 The micro-Raman spectra of aged (200–800 h) wool (Fig. 2) were recorded in order to observe the degradation of the wool fibre. Obviously, the richest and most published region of protein Raman spectra is between 1800 cm
1 and the Raleigh line, that includes the main features of amide I, C–H bending, amide III and S–S stretching. The usefulness of the C–H stretching region (2900 cm
1 ) is also demonstrated and all are considered and compared for eventual degradation. It is noted that peaks in the lower range of the Raman shift exhibit the greatest variation with exposure time. There is the appearance of a new band at 1040 cm
1 given by the oxidation in the cystine residues and formation of S–O of cysteic acid. The sharp band at 1004 cm
1 , assigned to the C–C stretching vibration of the aromatic ring in the phenylalanine side chain, due to the breathing vibration of the monosubstituted ring in phenylalanine gradually decreases in intensity. (CH) region, 3600–2500 cm
1 1000–400 cm
1 region The aged wool sample gave the same vibrational modes in the 3600–2700 cm
1 region. Strong peaks were still present for all samples in the 3260 cm
1 region, assigned to N–H stretching vibrations even though we noted a gradual decrease in the intensities. The features reported in the 2765–2725 cm
1 assigned to olefinic C–H stretching vibrations retained its Interestingly, there is a new peak at 977 cm
1 formed, which increases in intensity as ageing develops. In literature this peak is not successfully assigned and remains unsure; however, it should be due to the products of photooxidation of the wool cysteine residues. It is known that dithiocarbamates have a strong band in this region along Aged wool Copyright  2008 John Wiley & Sons, Ltd. J. Raman Spectrosc. 2008; 39: 638–645 DOI: 10.1002/jrs 641 642 B. Doherty et al. with monothiolic acids and ionic dithiolates.16 The bands in the 960–930 cm
1 region, reported as being characteristic of the ˛-helical keratin conformation, detect its intensity decrease or completely disappear and are not observed with the onset of ageing, indicative of an increase in the disordered content of the fibre. A change in the tyrosine Fermi doublet (831 : 852 cm
1 ), indicates a sensitivity of the strength of the H-bonding to the phenolic hydroxyl group of tyrosine. The change in intensity indicates that the residues are strongly H-bonded or buried within the hydrophobic region being protected from further oxidation.21 The bands attributed to cystine, C–S at 644 and 621 cm
1 observe an overall decrease respective bands with ageing up to 800 h. The cystine S–S band located at 511 cm
1 instead detects its intensity diminish gradually and completely disappears after 400 h ageing. Table 3. Putative assignments of lipids bounded to fibres Wool fibre lipid content of the dye also needs to be approached, as the skeleton (chromophore) determines the lightfastness, whereas the substituent groups (auxochromes) can normally alter this attribution of lightfastness. The physical state of the dye is considered, as the more finely dispersed the dye within the fibre, the more rapidly it will fade. Fibres with large aggregates of dye are more lightfast, since a smaller surface area of the dye is exposed to air and light. Furthermore, the lightfastness usually increases with increasing dye concentration, the main cause being an increase in average size of the sub-microscopic particles, which the dye forms in the fibre. Ultimately, the fastness of the mordant dye depends on the mordant and dyeing method, as different metal dye complexes are formed that may differ in stability to light and because the metal may have a positive or negative catalytic effect on the photochemical degradation of the dye. Considering the micro-Raman spectroscopic identification for the many vibrational modes of wool, it is also plausible that the same technique can be utilised to identify colorants that have been used to dye-wool samples as well as to evaluate any conformational changes within the wool fibre that are induced by the dyeing process. The crucial range to observe when examining the binding of dyes with wool fibre keratin is the 500–1800 cm
1 . Micro-Raman investigations on aged wool at 400 h gave rise to spectra, which were not assignable to the wool fibre itself (Fig. 3). Instead it is more likely that these bands may be characteristic of a weak lipid content (Table 3),22 thought to be covalently bound, to the high cysteine-containing proteinaceous outer surfaces of keratin fibres and responsible for the yellowing process of the fibres.23 Dyed wool Many studies have been made of the photodegradation of the numerous classes of dyes such as indigoids, flavonoids and anthraquinones dyes on fibres.24 However, the mechanisms are not yet fully clear since there are many factors controlling the photo-fading. It seems pertinent, therefore, to derive these most important factors determined not only by light, temperature and humidity, but by the internal factors25 such as the chemical structure and physical characteristics of the fibre itself since the process of fading on protein fibres is attributed to a reductive nature. The chemical structure Washed, unaged wool 2927 sh 2879 vs (doublet) 2847 s (doublet) 2728 w br 1462 m (doublet) 1439 m (doublet) 1294 w 1229 w 1131 w 1061 w 856 w 714 m Approximate assignments ⊲CH3 ⊳ symmetric stretch (chain end) ⊲CH2 ⊳ stretch ⊲CH2 ⊳  (CH) aliphatic υ⊲CH2 ⊳ scissoring (CH) deformation Acyl chains Acyl chains Acyl chains Acyl chains Diagnostic of phospholipid  NC ⊲CH3 ⊳3 anti-symmetric Indigoid dyed wool Figure 3. Spectrum of probable surface lipid component. Copyright  2008 John Wiley & Sons, Ltd. A straightforward example may be seen in the case of wool dyed with woad (Isatis tinctoria), one of the earliest known sources of indigo. Woad is a biennial or perennial herb, indigenous, although not necessarily native, to northern Europe. The colouring matters of woad are indigotin and indirubin. Indigotin is a symmetrical dye molecule that has good lightfastness properties that is relatively independent of substituent groups. These factors, combined with the physical state of the dye, may explain its superior lightfastness compared to other natural dyes. The precursor to produce indigo is the ester, isatan B (indoxyl-5ketogluconate), the carbohydrate moiety is hydrolysed from J. Raman Spectrosc. 2008; 39: 638–645 DOI: 10.1002/jrs Artificially aged natural and dyed wool the indoxyl group and two of the resulting indoxyl molecules combine oxidatively forming the indigo product, which precipitates from the solution.26 Raman scattering from indigo in its powdered form is relatively easy to achieve, as is the scattering from the woad colorant on dyed wool (Fig. 4). It is noticed that the characteristic bands from the wool substrate are overlapped with those provenant of the colorant (Table 4) with the exception of a few, namely at 1607, 1300 and 864 cm
1 suggesting that there is a good coverage achieved by the colorant. The first two bands correspond to a change in the amide I of the wool fibre, or to the constituents of the secondary keratin structure of tyrosine and tryptophan, respectively.27 Instead, the observed peak at 864 cm
1 may be an indication of the lipid content on the surface of the fibre, that in comparison with natural wool aged to 400 h, it can be suggested that the dyeing process actually causes a certain level of degradation to the wool fibres. Reference literature also supports this theory as traditional methods for dyeing wool use elevated temperatures and products that can damage the fibres and that the dyeing and diffusion properties of the wool and fibres are governed by wool lipids, and thus an understanding of how the dyeing process changes these structures is key to its optimisation.28 Wool dyed with woad and artificially aged (800 h) has been investigated to determine the level of photodegradation. The resulting micro-Raman spectrum (Fig. 5) gives many of the characteristic bands of the indigoid colorant, suggesting that woad has remained well dispersed across the surface, retaining it high lightfast qualities.30 It has been suggested that the degradation of woad occurs slowly as the dye forms large aggregates inside the fibre, which reduces the surface area of the dye accessible to oxygen, light and moisture.25 Yet it is noticed that bands of the underlying textile are also observed with greater intensities than those revealed with unaged woad dyed wool indicative of an Figure 4. Collective spectra of (a) woad dyed wool, (b) indigo powder standard, (c) undyed, unaged wool. Copyright  2008 John Wiley & Sons, Ltd. Table 4. Overview of the micro-Raman band positions of woad dyed wool Woad dyed Indigo powder Assignment of vibration wool (cm
1 ) standard (cm
1 ) (Ref. 29) (cm
1 ) 1702 m 1635 w 1607 m 1583 m 1486 w 1465 w 1367 m 1301 m 1252 m 1147 m 936 vw 864 vw 762 vw 600 w 1694 m 1627 w – 1580 m 1488 w 1466 w 1364 m 1306 w 1249 m 1147 w 941 vw – 757 vw 600 w 548 w 545 w (C O) (CC), υ(CH) – (CC), (C C), (C O) (CC), υ(CH) (CC), υ(CH) υ(NH), υ(CH) – υ(CH), υ(C O) υ(CC) (CH) (CN) υ(CH), υ(N–C–C) υ(C O), υ(CH), υ(C–NH–C), υ(C O) υ(C C–CO–C) increase in the disordered content of the fibre. The (CH) region of the wool is very intense circa 3000 cm
1 and there is also a certain overlapping of bands in the 1600 cm
1 region where the amide I of the protein keratin appears more intense. Signs of degradation from the wool fibres are evident from the bands at 1311, 1040 and 977 cm
1 relative to the tryptophan content, the S–O of cysteic acid and products of photo-oxidation of wool cysteine residues, respectively. Furthermore, a change in the intensity at 823 and 853 cm
1 indicates a change in the sensitivity of the strength of the H-bonding to the phenolic hydroxyl group of tyrosine. Figure 5. Aged woad dyed Wool (800 h). J. Raman Spectrosc. 2008; 39: 638–645 DOI: 10.1002/jrs 643 644 B. Doherty et al. Flavonoid dyed wool Weld, originating from the leaves and stem of Reseda luteola L., is reported to be the oldest European dyestuff for dyeing wool and silk yellow.29 Flavonoids are responsible for the colour of weld. The flavonoid content of each weld plant depends on its geographical proveniences and ranges from 1 to 4% of constituting materials.31 The main colouring components are luteolin (5,7,30 ,40 -tetrahydroxyflavone) and apigenin (5,7,40 trihydroxyflavone).32 The luteolin/apigenin ratio is reported to be approximately 9/1.33 Present in the plant as glycosides, they are hydrolysed to the parent flavonoid during the dyeing process. The lightfastness of these components is dependant on the number and position of substituents (auxochromes).34 Wool dyed with weld does not easily yield characteristic Raman scattering, instead a subtle fluorescence emission from the colorant is observed (Fig. 6). The flavonoid does not, however, completely cover the signals from the keratin fibres underneath as the C–H stretching vibrations close to 3000 cm
1 are moderately visible. The (CO) of the I amide is clear at 1650 cm
1 as is the band at 1450 cm
1 assigned to a methylene (CH2 ) deformation band. The band at 1007 cm
1 , shifted from 1004 cm
1 for virgin wool, is assigned to the C–C stretching vibration of the aromatic ring in the phenylalanine side chain. Each of the aforementioned bands are equally observed in the case of aged weld dyed wool (800 h), if not in slightly higher intensities, also the fluorescence emission slightly changes its maximum. It can be mentioned that weld is more or less retained on the surface of the fibre and the ageing process does not appear to degrade the colorant nor significantly weaken the underlying fibre. This type of degradation suggests that a small portion of the dye remains molecularly dispersed within the fibre, yet some of it forms aggregates. The size and form of aggregates determine the extent of the exposed air-dye interface, and this in turn determines the rate of photodegradation.35 Anthraquinone dyed wool To follow this class of dyes, madder is utilised as a representative example of their behaviour on wool. Generally, the anthraquinone dyes have a good lightfastness; for the anthraquinones, the fastness decreases as the number of hydroxyl groups increases. Madder is one of the oldest and most popular red dyestuffs found in nature worldwide. Extracted from dried roots of Rubia tinctorum L., madder has been used since antiquity for dyeing textiles (in particular, in Europe, the Middle East and India where the plant was indigenous). The colouring matter of Rubia tinctorum L. roots is based on anthraquinone dyes and can vary with the age of the plant. The main colouring components are alizarin (1,2-hydroxyanthraquinone) and purpurin (1,2,4-hydroxyanthraquinone), but a number of other anthraquinones (at least 19) such as ruberythric acid (the glycoside of alizarin), pseudopurpurin, xanthopurpurin, rubiadin and munjistin is also present.25 When wool is dyed with madder, the resulting microRaman spectrum presents a high fluorescence emission most likely from the dye itself that completely covers any scattering from the wool substrate (Fig. 7). This is sufficient in suggesting that the dye is well fixed to the surface before the ageing process. Following 800 h ageing, the spectrum still exhibits a fluorescence emission, yet this has changed significantly showing two maxima instead of only one, perhaps as an indication of its degradation. It is noted that the lightfastness of alizarin, in particular, may be partly due to dye-metal complex formation. It is presumed that metal ions quench the excited states; therefore, their presence in a system usually increases the stability in light.36 CONCLUSIONS Figure 6. Spectra of (a) unaged weld dyed wool (b) aged weld dyed wool (800 h). Copyright  2008 John Wiley & Sons, Ltd. This study has seen the utility of micro-Raman spectroscopy in examining wool fibres. The identification of such, in its natural state, has then been compared to artificially aged wool and many differences have been noted that include the main features of amide I, C–H bending, amide III and S–S stretching. The usefulness of the C–H stretching region (2900 cm
1 ) was also demonstrated and all were considered and compared for eventual degradation. The degradation of wool was documented that ultimately gives rise to photo-yellowing (yellow discoloration) and phototendering (strength loss) of the fabric.37 In addition, it has been investigated that if the fabric is coloured, depending J. Raman Spectrosc. 2008; 39: 638–645 DOI: 10.1002/jrs Artificially aged natural and dyed wool in the fluorescence emission suggests that the colorant may have degraded to a certain extent. REFERENCES Figure 7. Resulting spectra of (a) unaged madder dyed wool (b) aged madder dyed wool (800 h). on the applied dyestuff/s and mordant even more obvious fading and strength loss can occur.38 The analysis of three classes of natural dyes was undertaken namely, indigoids, flavonoids and anthraquinones in order to observe their effects when dyed on wool. It was also important to observe the effects of artificial ageing on both the wool substrate and eventual modifications with the colorant. It was seen that the identification of the indigoid was possible that covered any significant scattering from the wool underneath. On ageing, many of the bands of woad were still present, yet so too were numerous vibrations from the wool that also indicated a certain level of stress and degradation to the wool substrate. The flavonoid, representative, weld gave a fluorescence emission significant from the colorant along with bands of the wool suggesting that this dye is more inhomogeneous in its covering of the entire wool surface. When the fluorescence emissions shift slightly on ageing, the intensities of the bands of wool appear more significant, indicating changes within dye–wool complex. Instead the anthraquinone class, represented by madder, gave a fluorescence emission that completely covered any scattering from the wool before and after ageing. The change Copyright  2008 John Wiley & Sons, Ltd. 1. Weber K, Geisler N. EMBO J. 1982; 1: 1155. 2. Plowman JE. J. Chromatogr. B 2003; 787: 63. 3. Edwards HGM, Hunt DE, Sibley MG. Spectrochim. Acta, Part A 1998; 54: 745. 4. Akhtar W, Edwards HGM. Spectrochim. Acta, Part A 1997; 53: 81. 5. Collins S, Davidson RS. J. Photochem. Photobiol., A 1994; 77: 277. 6. Church JS, Millington KR. Biospectroscopy 1996; 2: 249. 7. Wojciechowska E, Pielesz A. Text. Res. J. 1992; 62: 580. 8. Miller IJ, Smith GJ. J. Soc. Dyers Colour. 1995; 111: 103. 9. Simpson WS, Page CT. The Effect of Light on Wool and the Inhibition of Light Tendering. Report no.6, Wool Research Organisation of New Zealand 1979. 10. Akhtar W, Edwards HGM, Farwell DW, Nutbrown M. Spectrochim. Acta, Part A 1997; 53: 1021. 11. Lin-Vien D, Colthup NB, Fateley WG, Grasselli JG. The Handbook of Infrared and Raman Characteristic Frequencies of Organic Molecules. Academic Press: San Diego, 1991. 12. Lawson EE, Edwards HGM, Johnson AF. Spectrochim. Acta, Part A 1995; 51: 2051. 13. Pezolet M, Pigeon-Gosselin M, Coulombre L. Biochem. Biophys. Acta 1976; 453: 502. 14. Lippert JL, Peticolas WL. Proc. Natl. Acad. Sci. U.S.A. 1971; 68: 1572. 15. Yellin N, Levin IW. Biochem. Biophys. Acta 1977; 489: 177. 16. Yu NT, Liu CS, O’Shoe DC. J. Mol. Biol. 1972; 70: 117. 17. Pielesz A, Wlochowicz A. Spectrochim. Acta, Part A 2001; 57: 2637. 18. Siamwiza MN, Lord RC, Chen MC, Takamatsu T, Harada I, Matsura H, Shimanouchi T. Biochemistry 1975; 14: 4870. 19. Chen MC, Lord RC. Biochemistry 1976; 15: 1889. 20. Frushour BG, Koenig JL. Biopolymers 1975; 14: 379. 21. Church JS, Corino GL, Woodhead AL. Biopolymers 1997; 42: 7. 22. Krafft C, Neudert L, Simat T, Salzer R. Spectrochim. Acta, Part A 2005; 61: 1529. 23. Negri AP, Cornell HJ, Rivett DE. Aust. J. Agric. Res. 1991; 42(8): 1285. 24. Katsuda N, Omura T, Takagishi T. Dyes Pigments 1997; 34: 147. 25. Cristea D, Vilarem G. Dyes Pigments 2006; 70: 238. 26. Stoker KG, Cooke DT, Hill DJ. Plant Growth Regul. 1998; 25: 181. 27. Pielesz A, Freeman HS, Weselucha-Birczynska A, Wysocki M, Wlochowicz A. J. Mol. Struct. 2003; 651: 405. 28. Martı̀ M, Barsukov L, Fonollosa J, Parra JL. Langmuir 2004; 20: 3068. 29. Hofenk de Graaff JH. In The Colourful Past. Abegg-Stiftung, Riggisberg and Archetype Publications: Riggisberg, London, 2004. 30. Vandenabeele P, Bode S, Alonso A, Moens L. Spectrochim. Acta, Part A 2005; 61: 2349. 31. Kaiser R. Angew. Bot. 1993; 67: 128. 32. Schweppe H. In Handbuch der Naturfarbstoffe. Vorkommen Verwendung Nachweis: Ecomed Velagsgesellschaft Landsberg/Lech, 1992. 33. Wouters J, Rosario-Chirinos N. J. Am. Inst. Cult. Herit. 1992; 31: 237. 34. Grupta D. Coulourage 1999; 46(8): 41. 35. Cox-Crews P. Stud. Conserv. 1987; 32(2): 65. 36. Giles CH, Baxter G. Text. Res. J. 1961; 31(10): 831. 37. Timar-Balazsy A, Eastop D. Chemical Principles of Textile Conservation. Butterworth-Heineman: Oxford, 1998. 38. Bruce BL. Stud. Conserv. 1992; 37: 371. J. Raman Spectrosc. 2008; 39: 638–645 DOI: 10.1002/jrs 645