- Email: [email protected]
The structure of different cellulosic fibres characterized by Raman spectroscopy
Vibrational Spectroscopy 86 (2016) 324–330
Contents lists available at ScienceDirect
Vibrational Spectroscopy journal homepage: www.elsevier.com/locate/vibspec
The structure of different cellulosic fibres characterized by Raman spectroscopy$ Ana Paula P. Alvesa , Luana P.Z. de Oliveiraa , Aloísio A.N. Castrob , Reiner Neumannc, Luiz F.C. de Oliveiraa , Howell G.M. Edwardsd, Antonio C. Sant’Anaa,* a
Departamento de Química, Instituto de Ciências Exatas, Universidade Federal de Juiz de Fora, 36036-900 Juiz de Fora, MG, Brazil Museu de Arte Murilo Mendes, Universidade Federal de Juiz de Fora, 36036-900 Juiz de Fora, MG, Brazil Setor de Caracterização Tecnológica, Centro de Tecnologia Mineral, 21941-901 Rio de Janeiro, RJ, Brazil d Centre for Astrobiology and Extremophile Research, Department of Chemical & Forensic Sciences, School of Life Sciences, University of Bradford, BD7 1DP Bradford, United Kingdom b c
A R T I C L E I N F O
Article history: Received 2 March 2016 Received in revised form 4 June 2016 Accepted 17 August 2016 Available online 18 August 2016 Keywords: Vibrational assignment Fluorescence FT-Raman Near-infrared excitation
A B S T R A C T
Different samples of cellulosic materials were analyzed by Raman spectroscopy and wood chips from Pinus elliottii, treated with acidic and alkaline aqueous solutions, were used to evaluate diagnostic signatures of the chemical structure of the cellulosic fibres. Cotton and whiskers synthesized from cotton, ancient Egyptian linen from a mummy wrapping, and five different paper sheets used in museum handling were compared. The complementarity of the Raman spectroscopic and scanning electron microscopic data facilitated the evaluation of the crystallinity, the level of organization and chain sizes of the fibres and the identification of different oxidation products. Intensity ratios measured from pairs of key bands were used to characterize the crystallinity, chain lengths and presence of oxidative decomposition in the range of the studied samples. Finally, the Raman spectra of the ancient Egyptian linen specimen indicated a potential future application of the proposed analysis for the characterization of archaeological pieces composed of linen. ã 2016 Elsevier B.V. All rights reserved.
1. Introduction Cellulose is a natural polymer which has played a major role in human history [1–4]. The large number of potential sources of cellulose from plants brings a challenge in the characterization of the botanical source of cellulosic samples especially due to the presence of contaminants in the fibre structures and the use of different processes applied historically for chemical treatment [5– 7]. More recently, due to their unique properties, cellulosic materials have been used for the production of novel composite materials in a diverse range of applications in biology, biofuels manufacture and engineering [8–13]. Sheets of paper can be produced from cellulosic raw materials such as wood, cotton and linen fibres. The assessment of the state of paper-based works in museums and libraries involves working with books, paints, prints, documents and other art objects; the typical analyses using electron microscopy, infrared absorption
$ Selected paper from for IV Encontro Brasileiro de Espectroscopia Raman (EnBraER), in Juiz de Fora, December 06–09, 2015. * Corresponding author. E-mail address: [email protected] (A.C. Sant’Ana).
http://dx.doi.org/10.1016/j.vibspec.2016.08.007 0924-2031/ã 2016 Elsevier B.V. All rights reserved.
and X-ray diffraction are fundamental but these techniques lead to the provision of only partial information about the structure of the fibres and usually require laborious preparation of the samples [14–16]. Hence, Raman spectroscopy can afford a suitable alternative for characterizing cellulosic materials for crystallinity, fibre structure and the sources of the raw materials as well as can inform changes that have been induced during the preparation or purification processes. The treatment of raw cellulosic materials using acid or alkalis (mercerization) is widespread [17–23], and Raman spectroscopy has been shown to be a valuable technique for analyzing structural changes in the fibres which arise from physical, chemical or mechanical processing [6,24–31]. In this work we report the Raman spectroscopic studies of the cellulose structure of Pinus elliottii wood after alkaline and acidic treatment and further evaluate these results for assessing the structural properties of other cellulosic materials: five commercial paper sheets, whiskers from cotton and ancient Egyptian linen samples. The vibrational spectroscopic conclusions were supported by data from scanning electron microscopy (SEM) analyses. The results permit an assessment to be made about the level of crystallinity of the cellulosic fibres, the interchain order and the presence of shortened polymer chains.
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
2. Materials and methods 2.1. Reactants and papers Sodium hydroxide and nitric acid were purchased from SigmaAldrich and used as received. The deionized water used in the solution preparation was of Milli-Q grade with an 18.2 MV cm resistivity. Cellulose whiskers prepared from cotton were kindly supplied by the Brazilian Agricultural Research Corporation (EMBRAPA) [32]. The ancient Egyptian linen specimen was obtained from mummy wrappings from Nekht-Ankh taken from a XIIth Dynasty rock tomb burial (ca. 2100 BCE), The Tomb of the Two Brothers (Nekht-Ankh and Khnum-Nakht), excavated by Sir William Flinders Petrie in 1906 in the Nile Valley, Der Rifeh, near Assiut [3]. The paper samples were provided by the Museu de Arte Murilo Mendes, Juiz de Fora, Brazil. They are used in museum conservation for the manipulation of artistic objects and consist of i) Kozo Japanese low-acidity paper produced from Broussonetia papyrifera, with a density of 8 gm2, which is used for filling or masking in the restoration of degraded cellulosic artefacts; ii) paperboard and filiset papers, freed of acidity by the presence of an alkaline additive, which are used for both support and as archival mounting boards in technical storage; iii) filter paper used as a blotter during aqueous treatment of artefacts where the principal requirement is the absence of additives; iv) silicone-coated paper, with one face protected against humidity, which is used in the transport of artistic works and in this case the Raman spectra were recorded from the non-siliconed side of the paper. 2.2. Chemical modification of Pinus elliottii wood chips Chips from Pinus elliottii wood were treated sequentially with aqueous solutions of nitric acid (volume fraction in%) followed by aqueous solutions of sodium hydroxide (mass fraction in%) using different H+/OH concentration ratios: 5/5, 20/5, 30/5, 5/10, 5/20, and 5/30. The treatment consisted of the immersion of the wood chips in an aqueous solution of nitric acid, using 10.0 mL per gram of wood chips, keeping the mixture under reflux at 85–90 C for 1 h. This was then followed by copious washing with deionized
325
water to neutral pH and then the material was submitted to boiling with sodium hydroxide aqueous solution using also 10.0 mL per gram of wood chips, keeping the mixture under reflux at 95–100 C for 1 h. After this procedure, a repeated washing was carried out to neutral pH and then the material was dried. 2.3. Equipment and methods Raman spectra were recorded using a Bruker, RS-100, FourierTransform Raman Spectrometer with Nd-YAG laser source operating at a wavelength of 1064 nm with a germanium detector cooled by liquid nitrogen. The spectra were recorded from a coupled optical microscope and each spectrum presented is an average of those recorded from 3 different points on the sample using 50 mW laser power at source, 300 scans accumulated and 4 cm1 spectral resolution. The fragments of paper used in museum handling, the ancient Egyptian linen and cellulosic whiskers were analyzed without any previous preparation being undertaken. The signal recorded at different points of the sample was very reproducible, which indicated the homogeneity of these samples, excepting that recorded from wood chips, where the estimated variability in the signal was 5–10%. SEM micrographs were recorded using a Jeol, JSM-5310 microscope using 15 kV electron beam voltage, an ETD detector at a working distance at 15 mm. The samples were coated with a thin gold film prior to analysis. 3. Results and discussion Fig. 1 presents the Raman spectra of Pinus elliotti wood with and without treatment using nitric acid solution followed subsequently with sodium hydroxide solutions at different concentrations. The Raman spectra were analyzed by measuring the intensity ratios of pairs of bands, without baseline correction being made, but accounting for the background emission. The values of these ratios are given in Table 1 and they are described as follows: a) 1121/1096 cm1 (I1121/I1096), assigned to symmetric and asymmetric b-(1,4)-glycosidic linkage stretching modes, respectively [6,33], which are characteristic monitors of the level of
Fig. 1. Raman spectra of cellulose whiskers (a); Pinus elliottii wood chips without (b) and with treatment using H+/OH ratios: 5/5 (c), 20/5 (d), 30/5 (e), 5/10 (f), 5/20 (g), 5/30 (h); and ancient linen (i). All spectra were normalized to the intensity of the band at 2896 cm1.
326
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
Fig. 2. SEM micrographs (50 mm 40 mm) of Pinus elliottii wood without (a) and with treatment using different H+/OH ratios: 5/5 (b), 20/5 (c), 30/5 (d), 5/20 (e), 5/30 (f).
Table 1 Intensity ratios of couple Raman bands from the spectra of wood with and without chemical treatment, cellulose whiskers and ancient linen. Material
I1121/I1096
I1096/I2896
I380/I1096
I1476/I1461
Wood (Pinus elliottii) H+/OH: 5/5 H+/OH: 20/5 H+/OH: 30/5 H+/OH: 5/10 H+/OH: 5/20 H+/OH: 5/30 Whiskers Ancient Egyptian linen
0.97 0.84 0.71 0.61 0.82 0.74 0.64 0.71 0.68
0.61 0.87 0.70 0.84 0.83 0.88 0.68 0.93 0.95
0.42 0.44 0.32 0.22 0.44 0.36 0.25 0.52 0.43
0.65 1.14 0.67 0.31 1.17 0.67 0.40 1.45 1.25
organization of the fibres, as an intensity decrease of the band at 1121 cm1 occurs in amorphous cellulose due to the fission of interchain linkages; b) 1096/2896 cm1 (I1096/I2896), the former being a glycosidic linkage stretching mode and the latter a CH2 stretching mode [6], whose intensity ratio decreases with the shortening of cellulose chain lengths by disruption of b-(1,4)-glycosidic linkages; c) 380/1096 cm1 (I380/I1096), with the former being assigned to CCC ring bending modes [34], and the latter the glycosidic
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
linkage stretching mode, where this ratio can be correlated with the crystallinity of the fibres; d) 1476/1461 cm1 (I1476/I1461), both assigned to CH2 bending modes from highly crystalline and low crystalline arrangements, respectively [25], and such a ratio also permits a correlation to be made between Raman spectral intensity and fibre crystallinity. The chemical modification procedures for the cellulosic fibres leads to greater oxidation in an alkaline medium (5/10, 5/20, 5/30), inferred by the presence of a broad band at ca. 1605 cm1 assigned to unsaturation moieties, which increases with the increase in emission background along with the increase in hydroxide concentration. The absence of these features at low hydroxide concentrations (5/5) indicates that these cannot be assigned to the presence of lignin, which has been largely removed by the acid and alkali treatment [20,21,35,36]. The oxidation of the fibres was significantly smaller at intermediate acid concentration (5/5; 20/5) as confirmed by the absence of the band at 1605 cm1. However, at a higher concentration of acid or alkalis (30/5, 5/30) the enhancement of the emission background is always observed which is indicative of the oxidative process having taken probably via a different path since the band at 1605 cm1 is only present in an alkaline medium. The progressive increase in the acid and alkalis concentrations in the treatment procedures led to a decrease in both the I380/I1096 and I1476/I1461 ratios. These decreased ratios are ascribed to the loss of crystallinity with increased acid and alkalis concentrations. This behavior shows a high correlation with both the decrease of the I1121/I1096 ratio, ascribed to the loss of structural order when interchain hydrogen bonds are broken, and the decrease of the I1096/I2896 ratio associated with the breaking of glycosidic linkages that leads to the production of shortened chains. Nevertheless, the broad CH stretching feature at 2896 cm1 shifts to 2889 cm1, with a narrowing of wavenumber of the band profile, when intra and intermolecular linkages are broken which is probably due to the smaller number of distinct chemical environments around the CH2 group moieties [17,37]. Fig. 2 shows the SEM micrographs of these samples where the surface aspect of the fibres after the chemical treatment can be observed. Since the wood chips were treated without any kind of mechanical milling, the inhomogeneity of the samples led to some
327
variability in the observed Raman intensity ratios with an error estimated at 5–10%; however, this error limit is still significantly less than the changes in observed intensity ratios recorded in the Raman spectra. The Raman spectrum of cellulose whiskers, also shown in Fig. 1, presents high values of the I1121/I1096, I380/I1096 and I1476/ I1461 ratios, suggesting that this material has high crystallinity and has a highly preserved structural order. The high value of the I1096/I2896 ratio infers that the length of the cellulose chains has been maintained in the chemical modification procedure. Surprisingly, it should be mentioned that all these band ratios have similar values to those noted in the cellulose whiskers in the Raman spectrum of the 4000-years old ancient linen obtained from the wrapping of an Egyptian mummy; it may be inferred therefore that both materials have comparable structural parameters and contain cellulose in a state of high crystallinity. That ancient linen still possesses high crystallinity indicates the stability of cellulosic materials with age and that only a slow breaking down of the fibres took place in the environmental conditions operating in the tomb. In the Raman spectrum of cellulose whiskers, a band at 2330 cm1 is observed and can be assigned to the CC triple bond stretching mode [38,39], whose origin can be ascribed to one or more of the following processes: firstly, some oxidative mechanism takes place during the chemical modification conditions or, secondly, there is possibly a high tendency of the whiskers to oxidize in air due to their high surface area and the small dimension of the fibres. Both of these oxidative processes should be considered distinct from that reported using the acidic and alkaline chemical treatment. It is noteworthy that in the Raman spectrum of a wood sample the lower values of I1096/I2896 and higher values of I1121/I1096 can be ascribed to the presence of contaminants such as lignin and noncellulosic materials, whose characteristic features in the 1000– 1200 cm1 and 2800–2900 cm1 spectral regions are overlapped with cellulose bands [17]. Lignin is perhaps the most important contaminant in the cellulose spectrum of wood as recognized by the additional bands at 3070, 1600, 1660 cm1, assigned to CH, C¼C and C¼O stretching modes, respectively. Fig. 3 shows the Raman spectra of some paper sheets and a representative cotton sample. The same intensity ratios described above were used to analyze the cellulose structure of these
Fig. 3. Raman spectra of cotton (a), and paper samples: Japanese (b), paperboard (c), filter (d), siliconed (e) and filiset (f). All spectra were normalized to the intensity of the band at 2896 cm1.
328
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
Table 2 Intensity ratios of couple Raman bands from the spectra of cotton and paper sheets. Material
I1121/I1096
I1096/I2896
I380/I1096
I1476/I1461
Cotton Japanese Aged Japanese Paperboard Filter Filiset Siliconed
0.78 0.78 0.70 0.77 0.84 0.81 0.86
0.60 0.66 0.94 0.66 0.64 0.45 0.49
0.39 0.36 0.30 0.32 0.34 0.29 0.34
1.50 0.87 0.94 1.02 1.09 1.05 1.09
samples and the data are presented in Table 2. The SEM micrographs of these samples are shown in Fig. 4. Cotton and Japanese paper have thin, cylindrical fibres, while the other cellulose samples have flat and wide fibres whose morphologies are typically those observed in materials that originate from wood sources (see Fig. 2). Cotton and Japanese have similar values for the I1121/I1096 ratio, while filter, filiset and siliconed papers have an analogous value of this ratio near the values obtained for a 5/5 wood sample, reinforcing the assumption that these materials are cellulosic fibres from wood. The low values of the I1096/I2896 ratio found in the majority of the paper sheet samples in comparison with that of the 5/5 wood sample indicates that the process of preparation of the cellulose involved
Fig. 4. SEM micrographs (100 mm 80 mm) of cotton (a), and paper samples: Japanese (b), paperboard (c), filter (d), filiset (e) and silicone (f).
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
329
Fig. 5. Raman spectra of Japanese paper recently purchased (a) and aged (b). Both spectra normalized to the intensity of the band at 1096 cm1.
in the production of the papers could lead to a significantly higher level of breakage of the chains. The lower values of the I380/I1096 and I1476/I1461 ratios found in all paper samples in comparison with cotton and the 5/5 wood samples show that the formers have a lower crystallinity, whilst the ratios from cotton strongly infer that it has the highest crystallinity observed in all the cellulose samples analyzed in this study. Paperboard and filiset samples possess the lowest values of I1121/I1096, I380/I1096 and I1476/I1461 ratios which can be ascribed to the low integrity and low structural order of the fibres correlated with their frayed appearance in the SEM micrographs. Japanese paper gives a different spectral pattern when aged after 3 years in air storage conditions and this spectrum can be compared with that from other more recently purchased material in Fig. 5. The high intensity of the band at 2330 cm1, assigned to the CC triple bond stretching mode, observed in the aged sample suggests that an oxidative process has taken place. The values of the I1121/I1096 ratio decreases while I1096/I2896 increases considerably. The first indicates some interchain linkages are broken, while the latter allows to suppose CH2 moiety can be the oxidation site. These ratios from aged Japanese paper are similar to those observed in the cellulose whiskers spectrum, shown high level of chain interactions. However, the lower values of the I380/I1096 and I1476/I1461 ratios infer this sample has lower crystallinity than cellulose whiskers. In this respect, Kozo fibres are demonstrated to be susceptible to an oxidative mechanism similar to that observed in the cellulose whiskers that have been synthesized from cotton and this can perhaps be ascribed to the presence of a similar softfibre organization.
surprising but can be attributed to the stable atmospheric conditions operating in the sealed tomb; in addition the mummy was enclosed in an unopened sarcophagus and the mummified remains were generally well-preserved as expected from Egyptian Dynastic Middle Kingdom funerary procedures. Hence, the linens and consequently the cellulose fibres had not been subjected to acid or alkaline attack and had not been degraded biologically by contact with decomposing tissue. In contrast, in an earlier study [41] the difference in linen preservation was noted for another archaeological sample, namely a specimen of linen cloth, which had been in contact with decomposed human skeletal remains near the strongly alkaline and corrosive environment of the Dead Sea at Kasr-el-Yahud. In an archaeological excavation, the burial ground of people slaughtered in battle approximately 1500 years ago revealed small samples of badly degraded linen which were extremely friable: we can now attribute the poor survival of the cellulose in these samples to the strongly alkaline and arid depositional environment, the presence of putrefying bacteria and chemical byproducts of tissue decomposition, all of which will serve to hasten the chain fission of the cellulose structures and aid the formation of short fibres noted in these samples. Unfortunately, a sample of the Kasr-elYahud linen was not available for this current study, but it would be reasonable to propose that the current data would serve as a good preliminary forensic protocol for the nondestructive characterization of cellulosic materials from archaeological excavations and forensic crime scenes.
4. Conclusions
The authors would like to thank CNPq, CAPES and FAPEMIG (CEX-APQ-01752/13, PRONEM APQ-01283-14) Brazilian Research Funding Agencies, Dr. Humberto de Mello Brandão and Dr. Michelle Munk from EMBRAPA – Natural Center for Research on Dairy Cattle for the supply of cellulose whiskers.
It has been demonstrated here that Raman spectroscopy can be used as a powerful analytical technique for the investigation of structural properties of cellulosic materials. Three relative intensity ratios: I1121/I1096, I380/I1096 and I1476/I1461 carry important information about the crystallinity and interchain order, whilst the I1096/I2896 ratio can be related to both the presence of shortened polymer chains and the lesser contribution of the scattering from CH2 moieties. The production of standard source samples obtained from Pinus elliottii wood treated with acid and alkaline solutions, which showed different levels of structural order and oxidation was essential for the comparative analysis of the cellulose component of paper sheets which are commonly used in the handling of art objects in museum conservation. An interesting conclusion which emerged from this study was the excellent state of preservation of the cellulosic structure of the archaeological specimen of mummy wrapping from the Tomb of the Two Brothers [40] which dates from about 2000 years BCE as demonstrated from the Raman spectrum. This may seem
Acknowledgements
References [1] H. Edwards, N. Nikhassan, D. Farwell, P. Garside, P. Wyeth, J. Raman Spectrosc. 37 (2006) 1193–1200. [2] M. Oriola, G. Campo, C. Ruiz-Recasens, N. Pedragosa, M. Strlic, Stud. Conserv. 60 (2015) S193–S199. [3] H.G.M. Edwards, M.J. Falk, Appl. Spectrosc. 51 (1997) 1134–1138. [4] I. Bjurhager, H. Halonen, E.L. Lindfors, T. Iversen, G. Almkvist, E.K. Gamstedt, L. A. Berglund, Biomacromolecules 13 (2012) 2521–2527. [5] H. Edwards, L. de Oliveira, M. Nesbitt, Analyst 128 (2003) 82–87. [6] H. Edwards, D. Farwell, D. Webster, Spectrochim. Acta A 53 (1997) 2383–2392. [7] J. Blackwel, P.D. Vasko, J.L. Koenig, J. Appl. Phys. 41 (1970) 4375. [8] S. Rosa, N. Rehman, M. de Miranda, S. Nachtigall, C. Bica, Carbohydr. Polym. 87 (2012) 1131–1138. [9] J.P.S. Morais, M.D. Rosa, M.D.M. de Souza, L.D. Nascimento, D.M. do Nascimento, A.R. Cassales, Carbohydr. Polym. 91 (2013) 229–235. [10] P. Kumar, D.M. Barrett, M.J. Delwiche, P. Stroeve, Ind. Eng. Chem. Res. 48 (2009) 3713–3729.
330
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
[11] S.J. Eichhorn, A. Dufresne, M. Aranguren, N.E. Marcovich, J.R. Capadona, S.J. Rowan, C. Weder, W. Thielemans, M. Roman, S. Renneckar, W. Gindl, S. Veigel, J. Keckes, H. Yano, K. Abe, M. Nogi, A.N. Nakagaito, A. Mangalam, J. Simonsen, A.S. Benight, A. Bismarck, L.A. Berglund, T. Peijs, J. Mater. Sci. 45 (2010) 1–33. [12] R.J. Moon, A. Martini, J. Nairn, J. Simonsen, J. Youngblood, Chem. Soc. Rev. 40 (2011) 3941–3994. [13] E.W. Nery, L.T. Kubota, Anal. Bioanal. Chem. 405 (2013) 7573–7595. [14] M. Manso, M. Carvalho, Spectrochim. Acta B 64 (2009) 482–490. [15] T. Lojewski, P. Miskowiec, M. Missori, A. Lubanska, L. Proniewicz, J. Lojewska, Carbohydr. Polym. 82 (2010) 370–375. [16] M. Bronzato, P. Calvini, C. Federici, S. Bogialli, G. Favaro, M. Meneghetti, M. Mba, M. Brustolon, A. Zoleo, J. Chem. Eur. 19 (2013) 9569–9577. [17] A. Jahn, M. Schroder, M. Futing, K. Schenzel, W. Diepenbrock, Spectrochim. Acta A 58 (2002) 2271–2279. [18] K. Schenzel, S. Fischer, Cellulose 8 (2001) 49–57. [19] D. Stewart, H. Wilson, P. Hendra, I. Morrison, J. Agric. Food Chem. 43 (1995) 2219–2225. [20] V. Torgashov, E. Gert, O. Zubets, F. Kaputskii, Russ. J. Bioorg. Chem. 36 (2010) 838–846. [21] J. Sun, X. Sun, H. Zhao, R. Sun, Polym. Degrad. Stab. 84 (2004) 331–339. [22] X.F. Sun, R.C. Sun, J. Tomkinson, M.S. Baird, Polym. Degrad. Stab. 83 (2004) 47– 57. [23] J. Sufiyama, J. Persson, H. Chanzy, Macromolecules 24 (1991) 2461–2466. [24] A. Ibrahim, P. Oldham, T. Conners, T. Schultz, J. Microchem. 56 (1997) 393–402.
[25] [26] [27] [28] [29] [30] [31] [32]
[33] [34] [35] [36] [37] [38] [39] [40] [41]
K. Schenzel, S. Fischer, E. Brendler, Cellulose 12 (2005) 223–231. H. Edwards, D. Farwell, A. Williams, Spectrochim. Acta A 50 (1994) 807–811. Y. Liu, S. Kokot, T. Sambi, Analyst 123 (1998) 633–636. J. Wiley, R. Atalla, Carbohydr. Res. 160 (1987) 113–129. J. Cael, K. Gardner, J. Koenig, J. Blackwell, J. Chem. Phys. 62 (1975) 1145–1153. U. Agarwal, S. Ralph, Appl. Spectrosc. 51 (1997) 1648–1655. U. Agarwal, S. Ralph, R. Reiner, R. Moore, C. Baez, Food Sci. Technol. 48 (2014) 1213–1227. M. Munk, H.M. Brandao, S. Nowak, L. Mouton, J.C. Gern, A.S. Guimaraes, C. Yepremian, A. Coute, N.R.B. Raposo, J.M. Marconcini, R. Brayner, Ecotoxicol. Environ. Saf. 122 (2015) 399–405. N. Gierlinger, S. Luss, C. Konig, J. Konnerth, M. Eder, P. Fratzl, J. Exp. Bot. 61 (2010) 587–595. U. Agarwal, R. Reiner, S. Ralph, Cellulose 17 (2010) 721–733. O. Brendel, P.P.M. Iannetta, D. Stewart, Phytochem. Anal. 11 (2000) 7–10. E.W. Crampton, L.A. Maynard, J. Nutr. 15 (1938) 383–395. F. Kolpak, J. Blackwell, Macromolecules 9 (1976) 273–278. N. Sheppard, D.M. Simpson, Q. Rev. Chem. Soc. 6 (1952) 1–33. M. Vandeven, H. Vanlangen, W. Verwer, Y.K. Levine, Chem. Phys. Lipids 34 (1984) 185–200. R. David, Mysteries of the Mummies: The Story of the Manchester University Investigation, Book Club Associates, London, 1978. H.G.M. Edwards, E. Ellis, D.W. Farwell, R.C. Janaway, J. Raman Spectrosc. 27 (1996) 663–669.
Contents lists available at ScienceDirect
Vibrational Spectroscopy journal homepage: www.elsevier.com/locate/vibspec
The structure of different cellulosic fibres characterized by Raman spectroscopy$ Ana Paula P. Alvesa , Luana P.Z. de Oliveiraa , Aloísio A.N. Castrob , Reiner Neumannc, Luiz F.C. de Oliveiraa , Howell G.M. Edwardsd, Antonio C. Sant’Anaa,* a
Departamento de Química, Instituto de Ciências Exatas, Universidade Federal de Juiz de Fora, 36036-900 Juiz de Fora, MG, Brazil Museu de Arte Murilo Mendes, Universidade Federal de Juiz de Fora, 36036-900 Juiz de Fora, MG, Brazil Setor de Caracterização Tecnológica, Centro de Tecnologia Mineral, 21941-901 Rio de Janeiro, RJ, Brazil d Centre for Astrobiology and Extremophile Research, Department of Chemical & Forensic Sciences, School of Life Sciences, University of Bradford, BD7 1DP Bradford, United Kingdom b c
A R T I C L E I N F O
Article history: Received 2 March 2016 Received in revised form 4 June 2016 Accepted 17 August 2016 Available online 18 August 2016 Keywords: Vibrational assignment Fluorescence FT-Raman Near-infrared excitation
A B S T R A C T
Different samples of cellulosic materials were analyzed by Raman spectroscopy and wood chips from Pinus elliottii, treated with acidic and alkaline aqueous solutions, were used to evaluate diagnostic signatures of the chemical structure of the cellulosic fibres. Cotton and whiskers synthesized from cotton, ancient Egyptian linen from a mummy wrapping, and five different paper sheets used in museum handling were compared. The complementarity of the Raman spectroscopic and scanning electron microscopic data facilitated the evaluation of the crystallinity, the level of organization and chain sizes of the fibres and the identification of different oxidation products. Intensity ratios measured from pairs of key bands were used to characterize the crystallinity, chain lengths and presence of oxidative decomposition in the range of the studied samples. Finally, the Raman spectra of the ancient Egyptian linen specimen indicated a potential future application of the proposed analysis for the characterization of archaeological pieces composed of linen. ã 2016 Elsevier B.V. All rights reserved.
1. Introduction Cellulose is a natural polymer which has played a major role in human history [1–4]. The large number of potential sources of cellulose from plants brings a challenge in the characterization of the botanical source of cellulosic samples especially due to the presence of contaminants in the fibre structures and the use of different processes applied historically for chemical treatment [5– 7]. More recently, due to their unique properties, cellulosic materials have been used for the production of novel composite materials in a diverse range of applications in biology, biofuels manufacture and engineering [8–13]. Sheets of paper can be produced from cellulosic raw materials such as wood, cotton and linen fibres. The assessment of the state of paper-based works in museums and libraries involves working with books, paints, prints, documents and other art objects; the typical analyses using electron microscopy, infrared absorption
$ Selected paper from for IV Encontro Brasileiro de Espectroscopia Raman (EnBraER), in Juiz de Fora, December 06–09, 2015. * Corresponding author. E-mail address: [email protected] (A.C. Sant’Ana).
http://dx.doi.org/10.1016/j.vibspec.2016.08.007 0924-2031/ã 2016 Elsevier B.V. All rights reserved.
and X-ray diffraction are fundamental but these techniques lead to the provision of only partial information about the structure of the fibres and usually require laborious preparation of the samples [14–16]. Hence, Raman spectroscopy can afford a suitable alternative for characterizing cellulosic materials for crystallinity, fibre structure and the sources of the raw materials as well as can inform changes that have been induced during the preparation or purification processes. The treatment of raw cellulosic materials using acid or alkalis (mercerization) is widespread [17–23], and Raman spectroscopy has been shown to be a valuable technique for analyzing structural changes in the fibres which arise from physical, chemical or mechanical processing [6,24–31]. In this work we report the Raman spectroscopic studies of the cellulose structure of Pinus elliottii wood after alkaline and acidic treatment and further evaluate these results for assessing the structural properties of other cellulosic materials: five commercial paper sheets, whiskers from cotton and ancient Egyptian linen samples. The vibrational spectroscopic conclusions were supported by data from scanning electron microscopy (SEM) analyses. The results permit an assessment to be made about the level of crystallinity of the cellulosic fibres, the interchain order and the presence of shortened polymer chains.
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
2. Materials and methods 2.1. Reactants and papers Sodium hydroxide and nitric acid were purchased from SigmaAldrich and used as received. The deionized water used in the solution preparation was of Milli-Q grade with an 18.2 MV cm resistivity. Cellulose whiskers prepared from cotton were kindly supplied by the Brazilian Agricultural Research Corporation (EMBRAPA) [32]. The ancient Egyptian linen specimen was obtained from mummy wrappings from Nekht-Ankh taken from a XIIth Dynasty rock tomb burial (ca. 2100 BCE), The Tomb of the Two Brothers (Nekht-Ankh and Khnum-Nakht), excavated by Sir William Flinders Petrie in 1906 in the Nile Valley, Der Rifeh, near Assiut [3]. The paper samples were provided by the Museu de Arte Murilo Mendes, Juiz de Fora, Brazil. They are used in museum conservation for the manipulation of artistic objects and consist of i) Kozo Japanese low-acidity paper produced from Broussonetia papyrifera, with a density of 8 gm2, which is used for filling or masking in the restoration of degraded cellulosic artefacts; ii) paperboard and filiset papers, freed of acidity by the presence of an alkaline additive, which are used for both support and as archival mounting boards in technical storage; iii) filter paper used as a blotter during aqueous treatment of artefacts where the principal requirement is the absence of additives; iv) silicone-coated paper, with one face protected against humidity, which is used in the transport of artistic works and in this case the Raman spectra were recorded from the non-siliconed side of the paper. 2.2. Chemical modification of Pinus elliottii wood chips Chips from Pinus elliottii wood were treated sequentially with aqueous solutions of nitric acid (volume fraction in%) followed by aqueous solutions of sodium hydroxide (mass fraction in%) using different H+/OH concentration ratios: 5/5, 20/5, 30/5, 5/10, 5/20, and 5/30. The treatment consisted of the immersion of the wood chips in an aqueous solution of nitric acid, using 10.0 mL per gram of wood chips, keeping the mixture under reflux at 85–90 C for 1 h. This was then followed by copious washing with deionized
325
water to neutral pH and then the material was submitted to boiling with sodium hydroxide aqueous solution using also 10.0 mL per gram of wood chips, keeping the mixture under reflux at 95–100 C for 1 h. After this procedure, a repeated washing was carried out to neutral pH and then the material was dried. 2.3. Equipment and methods Raman spectra were recorded using a Bruker, RS-100, FourierTransform Raman Spectrometer with Nd-YAG laser source operating at a wavelength of 1064 nm with a germanium detector cooled by liquid nitrogen. The spectra were recorded from a coupled optical microscope and each spectrum presented is an average of those recorded from 3 different points on the sample using 50 mW laser power at source, 300 scans accumulated and 4 cm1 spectral resolution. The fragments of paper used in museum handling, the ancient Egyptian linen and cellulosic whiskers were analyzed without any previous preparation being undertaken. The signal recorded at different points of the sample was very reproducible, which indicated the homogeneity of these samples, excepting that recorded from wood chips, where the estimated variability in the signal was 5–10%. SEM micrographs were recorded using a Jeol, JSM-5310 microscope using 15 kV electron beam voltage, an ETD detector at a working distance at 15 mm. The samples were coated with a thin gold film prior to analysis. 3. Results and discussion Fig. 1 presents the Raman spectra of Pinus elliotti wood with and without treatment using nitric acid solution followed subsequently with sodium hydroxide solutions at different concentrations. The Raman spectra were analyzed by measuring the intensity ratios of pairs of bands, without baseline correction being made, but accounting for the background emission. The values of these ratios are given in Table 1 and they are described as follows: a) 1121/1096 cm1 (I1121/I1096), assigned to symmetric and asymmetric b-(1,4)-glycosidic linkage stretching modes, respectively [6,33], which are characteristic monitors of the level of
Fig. 1. Raman spectra of cellulose whiskers (a); Pinus elliottii wood chips without (b) and with treatment using H+/OH ratios: 5/5 (c), 20/5 (d), 30/5 (e), 5/10 (f), 5/20 (g), 5/30 (h); and ancient linen (i). All spectra were normalized to the intensity of the band at 2896 cm1.
326
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
Fig. 2. SEM micrographs (50 mm 40 mm) of Pinus elliottii wood without (a) and with treatment using different H+/OH ratios: 5/5 (b), 20/5 (c), 30/5 (d), 5/20 (e), 5/30 (f).
Table 1 Intensity ratios of couple Raman bands from the spectra of wood with and without chemical treatment, cellulose whiskers and ancient linen. Material
I1121/I1096
I1096/I2896
I380/I1096
I1476/I1461
Wood (Pinus elliottii) H+/OH: 5/5 H+/OH: 20/5 H+/OH: 30/5 H+/OH: 5/10 H+/OH: 5/20 H+/OH: 5/30 Whiskers Ancient Egyptian linen
0.97 0.84 0.71 0.61 0.82 0.74 0.64 0.71 0.68
0.61 0.87 0.70 0.84 0.83 0.88 0.68 0.93 0.95
0.42 0.44 0.32 0.22 0.44 0.36 0.25 0.52 0.43
0.65 1.14 0.67 0.31 1.17 0.67 0.40 1.45 1.25
organization of the fibres, as an intensity decrease of the band at 1121 cm1 occurs in amorphous cellulose due to the fission of interchain linkages; b) 1096/2896 cm1 (I1096/I2896), the former being a glycosidic linkage stretching mode and the latter a CH2 stretching mode [6], whose intensity ratio decreases with the shortening of cellulose chain lengths by disruption of b-(1,4)-glycosidic linkages; c) 380/1096 cm1 (I380/I1096), with the former being assigned to CCC ring bending modes [34], and the latter the glycosidic
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
linkage stretching mode, where this ratio can be correlated with the crystallinity of the fibres; d) 1476/1461 cm1 (I1476/I1461), both assigned to CH2 bending modes from highly crystalline and low crystalline arrangements, respectively [25], and such a ratio also permits a correlation to be made between Raman spectral intensity and fibre crystallinity. The chemical modification procedures for the cellulosic fibres leads to greater oxidation in an alkaline medium (5/10, 5/20, 5/30), inferred by the presence of a broad band at ca. 1605 cm1 assigned to unsaturation moieties, which increases with the increase in emission background along with the increase in hydroxide concentration. The absence of these features at low hydroxide concentrations (5/5) indicates that these cannot be assigned to the presence of lignin, which has been largely removed by the acid and alkali treatment [20,21,35,36]. The oxidation of the fibres was significantly smaller at intermediate acid concentration (5/5; 20/5) as confirmed by the absence of the band at 1605 cm1. However, at a higher concentration of acid or alkalis (30/5, 5/30) the enhancement of the emission background is always observed which is indicative of the oxidative process having taken probably via a different path since the band at 1605 cm1 is only present in an alkaline medium. The progressive increase in the acid and alkalis concentrations in the treatment procedures led to a decrease in both the I380/I1096 and I1476/I1461 ratios. These decreased ratios are ascribed to the loss of crystallinity with increased acid and alkalis concentrations. This behavior shows a high correlation with both the decrease of the I1121/I1096 ratio, ascribed to the loss of structural order when interchain hydrogen bonds are broken, and the decrease of the I1096/I2896 ratio associated with the breaking of glycosidic linkages that leads to the production of shortened chains. Nevertheless, the broad CH stretching feature at 2896 cm1 shifts to 2889 cm1, with a narrowing of wavenumber of the band profile, when intra and intermolecular linkages are broken which is probably due to the smaller number of distinct chemical environments around the CH2 group moieties [17,37]. Fig. 2 shows the SEM micrographs of these samples where the surface aspect of the fibres after the chemical treatment can be observed. Since the wood chips were treated without any kind of mechanical milling, the inhomogeneity of the samples led to some
327
variability in the observed Raman intensity ratios with an error estimated at 5–10%; however, this error limit is still significantly less than the changes in observed intensity ratios recorded in the Raman spectra. The Raman spectrum of cellulose whiskers, also shown in Fig. 1, presents high values of the I1121/I1096, I380/I1096 and I1476/ I1461 ratios, suggesting that this material has high crystallinity and has a highly preserved structural order. The high value of the I1096/I2896 ratio infers that the length of the cellulose chains has been maintained in the chemical modification procedure. Surprisingly, it should be mentioned that all these band ratios have similar values to those noted in the cellulose whiskers in the Raman spectrum of the 4000-years old ancient linen obtained from the wrapping of an Egyptian mummy; it may be inferred therefore that both materials have comparable structural parameters and contain cellulose in a state of high crystallinity. That ancient linen still possesses high crystallinity indicates the stability of cellulosic materials with age and that only a slow breaking down of the fibres took place in the environmental conditions operating in the tomb. In the Raman spectrum of cellulose whiskers, a band at 2330 cm1 is observed and can be assigned to the CC triple bond stretching mode [38,39], whose origin can be ascribed to one or more of the following processes: firstly, some oxidative mechanism takes place during the chemical modification conditions or, secondly, there is possibly a high tendency of the whiskers to oxidize in air due to their high surface area and the small dimension of the fibres. Both of these oxidative processes should be considered distinct from that reported using the acidic and alkaline chemical treatment. It is noteworthy that in the Raman spectrum of a wood sample the lower values of I1096/I2896 and higher values of I1121/I1096 can be ascribed to the presence of contaminants such as lignin and noncellulosic materials, whose characteristic features in the 1000– 1200 cm1 and 2800–2900 cm1 spectral regions are overlapped with cellulose bands [17]. Lignin is perhaps the most important contaminant in the cellulose spectrum of wood as recognized by the additional bands at 3070, 1600, 1660 cm1, assigned to CH, C¼C and C¼O stretching modes, respectively. Fig. 3 shows the Raman spectra of some paper sheets and a representative cotton sample. The same intensity ratios described above were used to analyze the cellulose structure of these
Fig. 3. Raman spectra of cotton (a), and paper samples: Japanese (b), paperboard (c), filter (d), siliconed (e) and filiset (f). All spectra were normalized to the intensity of the band at 2896 cm1.
328
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
Table 2 Intensity ratios of couple Raman bands from the spectra of cotton and paper sheets. Material
I1121/I1096
I1096/I2896
I380/I1096
I1476/I1461
Cotton Japanese Aged Japanese Paperboard Filter Filiset Siliconed
0.78 0.78 0.70 0.77 0.84 0.81 0.86
0.60 0.66 0.94 0.66 0.64 0.45 0.49
0.39 0.36 0.30 0.32 0.34 0.29 0.34
1.50 0.87 0.94 1.02 1.09 1.05 1.09
samples and the data are presented in Table 2. The SEM micrographs of these samples are shown in Fig. 4. Cotton and Japanese paper have thin, cylindrical fibres, while the other cellulose samples have flat and wide fibres whose morphologies are typically those observed in materials that originate from wood sources (see Fig. 2). Cotton and Japanese have similar values for the I1121/I1096 ratio, while filter, filiset and siliconed papers have an analogous value of this ratio near the values obtained for a 5/5 wood sample, reinforcing the assumption that these materials are cellulosic fibres from wood. The low values of the I1096/I2896 ratio found in the majority of the paper sheet samples in comparison with that of the 5/5 wood sample indicates that the process of preparation of the cellulose involved
Fig. 4. SEM micrographs (100 mm 80 mm) of cotton (a), and paper samples: Japanese (b), paperboard (c), filter (d), filiset (e) and silicone (f).
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
329
Fig. 5. Raman spectra of Japanese paper recently purchased (a) and aged (b). Both spectra normalized to the intensity of the band at 1096 cm1.
in the production of the papers could lead to a significantly higher level of breakage of the chains. The lower values of the I380/I1096 and I1476/I1461 ratios found in all paper samples in comparison with cotton and the 5/5 wood samples show that the formers have a lower crystallinity, whilst the ratios from cotton strongly infer that it has the highest crystallinity observed in all the cellulose samples analyzed in this study. Paperboard and filiset samples possess the lowest values of I1121/I1096, I380/I1096 and I1476/I1461 ratios which can be ascribed to the low integrity and low structural order of the fibres correlated with their frayed appearance in the SEM micrographs. Japanese paper gives a different spectral pattern when aged after 3 years in air storage conditions and this spectrum can be compared with that from other more recently purchased material in Fig. 5. The high intensity of the band at 2330 cm1, assigned to the CC triple bond stretching mode, observed in the aged sample suggests that an oxidative process has taken place. The values of the I1121/I1096 ratio decreases while I1096/I2896 increases considerably. The first indicates some interchain linkages are broken, while the latter allows to suppose CH2 moiety can be the oxidation site. These ratios from aged Japanese paper are similar to those observed in the cellulose whiskers spectrum, shown high level of chain interactions. However, the lower values of the I380/I1096 and I1476/I1461 ratios infer this sample has lower crystallinity than cellulose whiskers. In this respect, Kozo fibres are demonstrated to be susceptible to an oxidative mechanism similar to that observed in the cellulose whiskers that have been synthesized from cotton and this can perhaps be ascribed to the presence of a similar softfibre organization.
surprising but can be attributed to the stable atmospheric conditions operating in the sealed tomb; in addition the mummy was enclosed in an unopened sarcophagus and the mummified remains were generally well-preserved as expected from Egyptian Dynastic Middle Kingdom funerary procedures. Hence, the linens and consequently the cellulose fibres had not been subjected to acid or alkaline attack and had not been degraded biologically by contact with decomposing tissue. In contrast, in an earlier study [41] the difference in linen preservation was noted for another archaeological sample, namely a specimen of linen cloth, which had been in contact with decomposed human skeletal remains near the strongly alkaline and corrosive environment of the Dead Sea at Kasr-el-Yahud. In an archaeological excavation, the burial ground of people slaughtered in battle approximately 1500 years ago revealed small samples of badly degraded linen which were extremely friable: we can now attribute the poor survival of the cellulose in these samples to the strongly alkaline and arid depositional environment, the presence of putrefying bacteria and chemical byproducts of tissue decomposition, all of which will serve to hasten the chain fission of the cellulose structures and aid the formation of short fibres noted in these samples. Unfortunately, a sample of the Kasr-elYahud linen was not available for this current study, but it would be reasonable to propose that the current data would serve as a good preliminary forensic protocol for the nondestructive characterization of cellulosic materials from archaeological excavations and forensic crime scenes.
4. Conclusions
The authors would like to thank CNPq, CAPES and FAPEMIG (CEX-APQ-01752/13, PRONEM APQ-01283-14) Brazilian Research Funding Agencies, Dr. Humberto de Mello Brandão and Dr. Michelle Munk from EMBRAPA – Natural Center for Research on Dairy Cattle for the supply of cellulose whiskers.
It has been demonstrated here that Raman spectroscopy can be used as a powerful analytical technique for the investigation of structural properties of cellulosic materials. Three relative intensity ratios: I1121/I1096, I380/I1096 and I1476/I1461 carry important information about the crystallinity and interchain order, whilst the I1096/I2896 ratio can be related to both the presence of shortened polymer chains and the lesser contribution of the scattering from CH2 moieties. The production of standard source samples obtained from Pinus elliottii wood treated with acid and alkaline solutions, which showed different levels of structural order and oxidation was essential for the comparative analysis of the cellulose component of paper sheets which are commonly used in the handling of art objects in museum conservation. An interesting conclusion which emerged from this study was the excellent state of preservation of the cellulosic structure of the archaeological specimen of mummy wrapping from the Tomb of the Two Brothers [40] which dates from about 2000 years BCE as demonstrated from the Raman spectrum. This may seem
Acknowledgements
References [1] H. Edwards, N. Nikhassan, D. Farwell, P. Garside, P. Wyeth, J. Raman Spectrosc. 37 (2006) 1193–1200. [2] M. Oriola, G. Campo, C. Ruiz-Recasens, N. Pedragosa, M. Strlic, Stud. Conserv. 60 (2015) S193–S199. [3] H.G.M. Edwards, M.J. Falk, Appl. Spectrosc. 51 (1997) 1134–1138. [4] I. Bjurhager, H. Halonen, E.L. Lindfors, T. Iversen, G. Almkvist, E.K. Gamstedt, L. A. Berglund, Biomacromolecules 13 (2012) 2521–2527. [5] H. Edwards, L. de Oliveira, M. Nesbitt, Analyst 128 (2003) 82–87. [6] H. Edwards, D. Farwell, D. Webster, Spectrochim. Acta A 53 (1997) 2383–2392. [7] J. Blackwel, P.D. Vasko, J.L. Koenig, J. Appl. Phys. 41 (1970) 4375. [8] S. Rosa, N. Rehman, M. de Miranda, S. Nachtigall, C. Bica, Carbohydr. Polym. 87 (2012) 1131–1138. [9] J.P.S. Morais, M.D. Rosa, M.D.M. de Souza, L.D. Nascimento, D.M. do Nascimento, A.R. Cassales, Carbohydr. Polym. 91 (2013) 229–235. [10] P. Kumar, D.M. Barrett, M.J. Delwiche, P. Stroeve, Ind. Eng. Chem. Res. 48 (2009) 3713–3729.
330
A.P.P. Alves et al. / Vibrational Spectroscopy 86 (2016) 324–330
[11] S.J. Eichhorn, A. Dufresne, M. Aranguren, N.E. Marcovich, J.R. Capadona, S.J. Rowan, C. Weder, W. Thielemans, M. Roman, S. Renneckar, W. Gindl, S. Veigel, J. Keckes, H. Yano, K. Abe, M. Nogi, A.N. Nakagaito, A. Mangalam, J. Simonsen, A.S. Benight, A. Bismarck, L.A. Berglund, T. Peijs, J. Mater. Sci. 45 (2010) 1–33. [12] R.J. Moon, A. Martini, J. Nairn, J. Simonsen, J. Youngblood, Chem. Soc. Rev. 40 (2011) 3941–3994. [13] E.W. Nery, L.T. Kubota, Anal. Bioanal. Chem. 405 (2013) 7573–7595. [14] M. Manso, M. Carvalho, Spectrochim. Acta B 64 (2009) 482–490. [15] T. Lojewski, P. Miskowiec, M. Missori, A. Lubanska, L. Proniewicz, J. Lojewska, Carbohydr. Polym. 82 (2010) 370–375. [16] M. Bronzato, P. Calvini, C. Federici, S. Bogialli, G. Favaro, M. Meneghetti, M. Mba, M. Brustolon, A. Zoleo, J. Chem. Eur. 19 (2013) 9569–9577. [17] A. Jahn, M. Schroder, M. Futing, K. Schenzel, W. Diepenbrock, Spectrochim. Acta A 58 (2002) 2271–2279. [18] K. Schenzel, S. Fischer, Cellulose 8 (2001) 49–57. [19] D. Stewart, H. Wilson, P. Hendra, I. Morrison, J. Agric. Food Chem. 43 (1995) 2219–2225. [20] V. Torgashov, E. Gert, O. Zubets, F. Kaputskii, Russ. J. Bioorg. Chem. 36 (2010) 838–846. [21] J. Sun, X. Sun, H. Zhao, R. Sun, Polym. Degrad. Stab. 84 (2004) 331–339. [22] X.F. Sun, R.C. Sun, J. Tomkinson, M.S. Baird, Polym. Degrad. Stab. 83 (2004) 47– 57. [23] J. Sufiyama, J. Persson, H. Chanzy, Macromolecules 24 (1991) 2461–2466. [24] A. Ibrahim, P. Oldham, T. Conners, T. Schultz, J. Microchem. 56 (1997) 393–402.
[25] [26] [27] [28] [29] [30] [31] [32]
[33] [34] [35] [36] [37] [38] [39] [40] [41]
K. Schenzel, S. Fischer, E. Brendler, Cellulose 12 (2005) 223–231. H. Edwards, D. Farwell, A. Williams, Spectrochim. Acta A 50 (1994) 807–811. Y. Liu, S. Kokot, T. Sambi, Analyst 123 (1998) 633–636. J. Wiley, R. Atalla, Carbohydr. Res. 160 (1987) 113–129. J. Cael, K. Gardner, J. Koenig, J. Blackwell, J. Chem. Phys. 62 (1975) 1145–1153. U. Agarwal, S. Ralph, Appl. Spectrosc. 51 (1997) 1648–1655. U. Agarwal, S. Ralph, R. Reiner, R. Moore, C. Baez, Food Sci. Technol. 48 (2014) 1213–1227. M. Munk, H.M. Brandao, S. Nowak, L. Mouton, J.C. Gern, A.S. Guimaraes, C. Yepremian, A. Coute, N.R.B. Raposo, J.M. Marconcini, R. Brayner, Ecotoxicol. Environ. Saf. 122 (2015) 399–405. N. Gierlinger, S. Luss, C. Konig, J. Konnerth, M. Eder, P. Fratzl, J. Exp. Bot. 61 (2010) 587–595. U. Agarwal, R. Reiner, S. Ralph, Cellulose 17 (2010) 721–733. O. Brendel, P.P.M. Iannetta, D. Stewart, Phytochem. Anal. 11 (2000) 7–10. E.W. Crampton, L.A. Maynard, J. Nutr. 15 (1938) 383–395. F. Kolpak, J. Blackwell, Macromolecules 9 (1976) 273–278. N. Sheppard, D.M. Simpson, Q. Rev. Chem. Soc. 6 (1952) 1–33. M. Vandeven, H. Vanlangen, W. Verwer, Y.K. Levine, Chem. Phys. Lipids 34 (1984) 185–200. R. David, Mysteries of the Mummies: The Story of the Manchester University Investigation, Book Club Associates, London, 1978. H.G.M. Edwards, E. Ellis, D.W. Farwell, R.C. Janaway, J. Raman Spectrosc. 27 (1996) 663–669.