Appl Phys A DOI 10.1007/s00339-010-5645-9
Artificial versus natural ageing of paper. Water role in degradation mechanisms Tomasz Łojewski · Paweł Mi´skowiec · Marcin Molenda · ´ Anna Lubanska · Joanna Łojewska
Received: 30 July 2009 / Accepted: 9 March 2010 © Springer-Verlag 2010
Abstract The cellulose–water system was traced down by the in situ transmission FTIR spectroscopy upon water desorption/sorption cycles at various conditions (dry air, humid air, water vapour) and temperatures. The results provide the information on the changes of the hydrogen bonds network during water desorption from paper and the changes in water sorption capacity of paper in the cycles.
1 Introduction Artificial ageing, commonly applied in science and technology of new materials, cannot also be avoided in the studies of the paper degradation usually performed to assess the paper condition or the methods of conservation performed on it. However, it brings a risk of inadequacy of the results which may or may not reflect the phenomena that occur during natural ageing. For this reason, paper life expectancy derived from the kinetic experiments seems useful but not easy to achieve a measure. To put it more precisely, the time after which paper looses the assumed properties defined by an end-user can be calculated by extrapolation of the reaction rate from the artificial ageing conditions to the storing conditions [1]. This is usually done converting the kinetic data into the Arrhenius plots. The whole operation is allowed,
T. Łojewski () · P. Mi´skowiec · M. Molenda · J. Łojewska Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Kraków, Poland e-mail:
[email protected] A. Luba´nska Institute of Catalysis and Surface Chemistry, Polish Academy of Sciences, Niezapominajek 8, 30-239 Kraków, Poland
provided that the degradation kinetics is known in the relatively broad range of the ageing parameters such as temperature, pressure and moisture content [2, 3]. This explains the current effort of several laboratories to resolve paper degradation kinetics which possibly could include a vast array of phenomena that occur during this process. The subsequent approaches to the paper degradation kinetics thus included additional factors influencing it. The model describing pure hydrolysis by Ekenstam [4, 5] opens the list and is followed by the improved models covering paper acidity and its initial degradation stage [5–7] and crystallinity [8, 9]. No matter how advanced will a model be available, the practical remedy for artificial ageing seems to approach the conditions of the natural ageing as close as possible. Since paper degradation is slow at lower temperature, none of the available standards recommended for artificial ageing describes the procedure that would meet the above requirement [10]. For example, according to the ASTM standard [11] followed in our laboratory, the artificial ageing tests are performed at 59% relative humidity in air at 90◦ C. As it has been proved for cellulose, new reactions start to evolve and predominate in increasing temperature ranges [12, 13]. The ranges can be roughly identified as follows: 20–100◦ C—water desorption, recrystallization, glycosidic bonds cleavage 100–150◦ C—functional groups formations 150–250◦ C—new array of degradation products 250◦ C—pyrolysis, dehydration This suggests that artificial ageing should be performed at least below 100◦ C to prevent intensive oxidation, dehydration or alkoxy elimination in cellulose chains as possible reaction routes [14] or finally avoid mere water desorption. In fact, water plays manifold functions in cellulosic materials starting from the mechanical and structural proper-
T. Łojewski et al.
ties to the chemical ones as a reaction agent, media or even as a communicator between cellulose molecules. The picture complicates additionally when we take into account that water subtle structure and thus properties (boiling point, viscosity, etc.) owed to the hydrogen bonding network are very much sensitive to the chemical environment (local potential). Thus talking of the cellulose–water system, we have to take into account such changes. Water in cellulose can be roughly recognized as: free water located in voids, whose macroscopic properties are similar to bulk water, and bound water firmed to cellulose chains by intermolecular hydrogen bonds [15, 16]. One of the criteria that allows one to distinguish types of water is freezability [16], and another is the type of hydrogen bonds [17]. According to the termogravimetrical and calorimetrical results obtained by Hakateyama et al. [16], bound water is connected to amorphous cellulose and may appear both in a freezable and nonfreezable form whose amount depends on the total water content in cellulose. From the perspective of the hydrogen bonding system in freezable water the vast majority of them comes from intermolecular interactions between water molecules alone, and conversely in the nonfreezable form the contribution of the cellulose–water interaction should be substantially higher. While the hydrogen bond network for crystalline and amorphous cellulose seems to be well resolved [18–22], the information on the types of hydrogen bonds in cellulose–water systems is rather scars [15, 17, 23]. The main question stated in this article is to which extent artificial ageing changes the environment and water structure in paper material and whether we can talk of the same system at all. Our goal is thus to provide an experimental basis for the discussion of the artificial ageing conditions in terms of water structure and functions as observed by in situ infrared spectroscopy (in situ FTIR).
2 Experimental 2.1 Samples The samples were pure cellulose (C) from bleached softwood craft pulp (JNO Pine 90, Botnia, Finland) and model paper (P1) composed almost of pure cellulose (99.5% related to dry mass of sample) from bleached softwood sulphite cellulose (TNO, The Netherlands) [24]. The paper sheets were stored at 23 ± 1◦ C and 50 ± 2% relative humidity, RH (TAPPI room). The water content in the samples stored at such conditions amounts to about 7.2 wt%.
2.2 Techniques and procedures 2.2.1 FTIR The transmission FTIR in situ measurements were performed in Excalibur 3000 (Diglab) spectrometer with DTGS detector. The spectrometer was equipped with a glass cuvette with CaF2 windows, an electrical furnace and a gas supplying/evacuating system. The apparatus scheme is presented in our previous paper [25]. Overall, 128 scans were collected per each spectrum. To be able to perform the transmission measurements, the thin plates of the samples were obtained according to the method described in [25, 26]. The samples were transferred from the TAPPI room to the cuvette. Prior to the water sorption/desorption experiments, the air was evacuated, the dynamic vacuum was maintained for 2 minutes, and then the appropriate gas mixture was fed to the cuvette. The experiments were performed at three gas mixtures: dry air (AD), humid air (AW, air with water vapour added), and in pure water vapour (WV). The partial pressure of water vapour was kept at 2.3 kPa measured at room temperature. Since the volume of the cuvette is high (250 cm3 ) and the samples small (around 100 mg), the contribution of the water included in paper to the total water partial pressure is negligibly low. For the details discussing the conditions on closed vessels during paper ageing at elevated temperature, a reader should refer to [27]. Each spectrum was collected at the equilibrium state after a chosen temperature of the sample was attained and stabilised. It has been proved that the position of the maxima of the spectra collected at elevated temperature ( AW > AD for both samples, which is logical. This indicates that adsorption of water is at equilibrium: on increasing water concentration in the gas phase, the amount of adsorbed water increases. It should be noted that the 2.3 kPa is much too low amount for the occurrence of the freezable water in paper (see Fig. 6 for comparison [27].
Fig. 4 Desorption of water from paper (P1) at 20◦ C at vacuum by in situ transmission FTIR. a Differential spectra of the OH bands with prolonged time of evacuation, b standardised absorbance spectra of the bound water bending vibrations for the initial and final state after the readsorption
This time again the remarkable differences in the quantitative results obtained at various conditions can be noted. The general tendency is that at other than pure water vapour (WV) conditions the readsorption of water is irreversible (Figs. 5a–d). Upon water desorption the amount of the bound water in the samples decreases to reach a constant value of around 0.3 for both kinds of the samples regardless of the conditions used. This residual value comes from the carbonyl groups stretching vibrations that form in the samples on bleaching during paper production and is characteristic of paper initial stage of degradation (see Fig. 2) [25]. The desorption extent decreases in the order depending on the conditions: WV(40%) > AW(28%) > AD(20%) and is the highest at WV conditions because at water vapour paper additionally saturates with water. It can be inferred from the highest values of the initial standardised absorptions values in the FTIR spectra presented in Figs. 5e and f for C and P1 samples, respectively. After the readsorption on the reverse temperature ramp, the final state is different from the initial state for the samples treated at AD and
Artificial versus natural ageing of paper. Water role in degradation mechanisms
Fig. 5 Quantitative results of desorption of water from cellulose (C) paper (P1) samples upon increasing temperature for various conditions obtained from the in situ FTIR transmission spectra in the range 1780–
1500 cm−1 . Dry air (AD): a sample C; b sample P1. Humid air (AW): c sample C; d sample P1. Water vapour (WV): e sample C; f sample P1; !—desorption, "—readsorption
AW atmosphere: the sorption capacity of the samples evidently decreases. In WV, the desorption/adsorption curves nearly coincide for both studied samples. The swapping of the desorption curve with the readsorption curve (Figs. 5e and f) can be due to the relatively high experimental er-
ror of the FTIR measurements in water vapour atmosphere where the rotational spectrum of water vapour interfere with the vibrational spectra of bound water in cellulose. It has to be pointed out though that for the corresponding conditions for the P1 and C samples, the results are very
T. Łojewski et al.
Fig. 6 Evolution of water content in paper closed in a vial (23◦ C, 50% RH) upon increasing temperature. RH estimated from the mass balance of water inside the vial
similar, which is the validation of the quantitative analysis. The irreversible effects of water sorption observed in the samples in the presence of air can be explained by the formation of hydrogen bonds in cellulose of higher strength than between cellulose and water molecules. Indeed the strongest bands of low intensity appear in the spectra at around 2934 cm−1 . We might also expect the bands emerging at higher frequency (3230–3310 cm−1 ), but this time the growing and decreasing effect might have compensated on the differential spectra. Their possible presence in the differential spectra can only be indirectly inferred from the asymmetric and steep shape of the decreasing part of the band at 3517 cm−1 . On the other hand, the stabilisation of weak hydrogen bonds could be due to steric or geometrical reconstruction of the pore shapes in cellulose which prevent water molecules from reaching the cellulose chains and which we do not perceive with FTIR. Such an effect was noted by Kato et al., who studied the decrease in the void fraction filled with free water in paper during thermal treatment. An interesting observation also is that for the AD and AW conditions, the sorption trend changes around 60◦ C. A similar observation was made by Yano et al. [31], who attributed the change in the trend of Young modulus of cellulose at 60◦ C upon thermal treatment of amorphous cellulose to the creation of new hydrogen bond. It is difficult to verify the formation of particular strong H-bonds in our samples because of the reasons stated above, but what we can say is that indeed above 50◦ C the spectra change the shape and the maxima appear at higher frequency than those measured at 40◦ C (see Figs. 3a and b). Discussing the problem of water forms and irreversible changes in paper, we would have to approach the conditions of artificial ageing used typically in our and other laboratories [13]. The question that emerges immediately is what kind of water is present in paper under such conditions. It is,
however, not easy to answer since a structural in situ analysis of water under harsh conditions (high humidity and temperature) brings serious technical problems connected with experimental set up and signal collecting. For this reason, to approach the problem we used the NIR probe dedicated to the assessment of water content adsorbed in the materials. The results are presented in Fig. 6. The content of water in the gas phase inside the vessel calculated from the mass balance of water using the NIR results is presented in the same Fig. 6 as the relative humidity values. According to the results upon increasing temperature at highly humid atmosphere, the amount of water included in paper (distinguishable by the NIR probe) only slightly decreases by around 10 wt% of the initial value, which secures the reaction media for the hydrolysis. It is not quite clear though whether this kind of adsorbed water comes from the same form of bound water in paper seen by FTIR. The question concerning water form present in paper during such ageing conditions remains open.
4 Conclusions The ageing tests performed on paper material at elevated temperature and moisture content do not reflect the phenomena that occur during natural ageing of paper. The conclusion is based on the determination of the changes in the hydrogen bonding system upon sorption/desorption experiments at various conditions (temperature, gas phase composition) by the in situ transmission FTIR. It has been demonstrated that in the presence of air the new hydrogen bonds are created in cellulose. Under such conditions, the sorption process is irreversible after exposure to high temperature (100◦ C), which causes the decrease in water sorption capacity in paper (cellulose). Quite different behaviour was noted in pure water vapour where the sorption/desorption cycles are. This is accounted for by the plasticising effect of water in cellulose, which prevents the formation of the intermolecular H-bonds in cellulose. The results show that the conditions used usually for the ageing tests cause irreversible changes in cellulose that are different from those occurring at room temperature during natural ageing. For this reason and in opposition to natural changes, such tests reflect artificial changes in paper rather than accelerated ones. In view of the results, to be able to accelerate the overall process of ageing caused mainly by hydrolysis the oxygen free pure water vapour at high temperature should be applied. Such a kind of accelerated ageing can be useful to assess both the kinetics and life expectancy of paper hydrolytic degradation, which in this way approach the value expected at natural conditions. Acknowledgements The authors are grateful for the financial support provided by the Polish Ministry of Science and Higher Education SPB K/PMN/000018 and SPB K/PMN/000023 as financial to COST/D42 action.
Artificial versus natural ageing of paper. Water role in degradation mechanisms
References 1. A. Bara´nski, A. Konieczna-Molenda, J.M. Łagan, L.M. Proniewicz, Restaurator 4, 36–45 (2003) 2. A. Bara´nski, P. Begin, J.M. Łagan, T. Łojewski, T. Sawoszczuk, J. Pulp Pap. Sci. 32, 238–244 (2006) 3. A. Bara´nski, J.M. Łagan, T. Łojewski, e-Preserv. Sci. 3, 1–4 (2006) 4. A. Ekenstam, Berrichte Dtsch. Chem. Gesselschaft 69, 553–559 (1936) 5. A. Bara´nski, Restaurator 23, 77–88 (2002) 6. P.L. Bégin, E. Kami´nska, Restaurator 23, 89–105 (2002) 7. X. Zou, T. Uesaka, N. Gurnagul, Cellulose 3, 269–279 (1996) 8. A.M. Emsley, R.J. Heywood, M. Ali, C.M. Eley, Cellulose 4, 1–5 (1997) 9. P. Calvini, A. Gorassini, A.L. Merlani, Cellulose 15, 193–203 (2008) 10. H.J. Porck, Rate of Paper Degradation. The Predictive Value of Artificial Aging Tests (ECPA, Amsterdam, 2000) 11. American Society for Testing Materials ASTM D6819-02 (2007) 12. A.J. Michel, H.G. Higgins, Infrared Spectroscopy in Australian Forest Products Research (CSIRO Forestry Forest Products, Melbourne, 2002) 13. M. Strliˇc, Aging Stabilisation of Paper (National and University Library, Ljubljana, 2005) 14. D.N. Hon, N. Shiraishi, Wood and Cellulosic Chemistry (Dekker, New York, 2000) 15. K.L. Kato, R.E. Cameron, J. Appl. Polym. Sci. 74, 1465–1477 (1999)
16. H. Hatakeyama, T. Hatakeyama, Thermochim. Acta 308, 3–22 (1998) 17. Y. Maréchal, H. Chanzy, J. Mol. Struct. 523, 183–196 (2000) 18. R.G. Zhbankov, S.P. Firsov, D.K. Buslov, N.A. Nikonenko, M.K. Marchewka, H. Ratajczak, J. Mol. Struct. 614, 117–125 (2002) 19. B. Hinterstoisser, L. Salmen, Cellulose 6, 251–263 (1999) 20. B. Hinterstoisser, M. Akerholm, L. Salmen, Carbohydr. Res. 334, 27–37 (2001) 21. D.A. Sukhov, Fibre Chem. 35, 277–282 (2003) 22. N.E.D. Regulus, E. Hunter, J. Appl. Polym. Sci. 23, 249–259 (1979) 23. H. Hatakeyama, C. Nagasaki, T. Yurugi, Carbohydr. Res. 48, 149– 158 (1976) 24. J. Havermans, Restaurator 132, 209–233 (1995) 25. J. Łojewska, P. Mi´skowiec, T. Łojewski, L.M. Proniewicz, Polym. Degrad. Stab. 88, 512–520 (2005) 26. J. Łojewska, A. Luba´nska, T. Łojewski, P. Mi´skowiec, L.M. Proniewicz, e-Preserv. Sci. 2, 1–12 (2005) 27. T. Sawoszczuk, A. Bara´nski, J.M. Łagan, T. Łojewski, K. Zi˛eba, J. Cult. Herit. 9, 401–411 (2008) 28. A. Kochanowski, R. Dziembaj, M. Molenda, A. Izak, E. Bortel, J. Therm. Anal. Calorim. 88, 499–502 (2007) 29. S. Park, R.A. Venditti, H. Jameel, J.J. Pawlak, Carbohydr. Polym. 66, 97–103 (2006) 30. R.A. Young, R. Rowel, Cellulose: Structure, Modification and Hydrolysis (Wiley-Interscience, New York, 1986) 31. S. Yano, H. Hatakeyama, T. Hatakeyama, J. Appl. Polym. Sci. 20, 3221–3231 (1976)
