160
Langmuir 2005, 21, 160-165
Improvement of the Wetting and Absorption Properties of Lignocellulosic Fibers by Means of Gas Phase Ozonation Max O ¨ stenson and Paul Gatenholm* Department of Materials and Surface Chemistry, Biopolymer Technology, Chalmers University of Technology, S-412 96 Go¨ teborg, Sweden Received April 26, 2004. In Final Form: September 27, 2004 Gas phase ozonation was done on sheets made from chemical thermomechanical pulp in order to improve the wetting properties of the lignocellulosic fibers. The degree of modification was varied by letting the reaction continue for different lengths of time, ranging from 1 to 60 min. Changes in the chemistry of the fibers after ozone exposure were investigated using Fourier transform infrared (FT-IR) spectroscopy and electron spectroscopy for chemical analysis (ESCA). The evolution of a carbonyl signal and the decrease of aromatic absorption over time was observed with FT-IR spectroscopy. The carbonyl peak grew in intensity as the reaction continued throughout the whole range of treatment times. The ESCA showed that carbonyl and carboxyl functionalities were introduced after 10 min of ozone exposure and that the intensity of the peak from the aliphatic and aromatic carbons decreased. However, an ozone treatment longer than 15 min did not affect the chemical surface composition, as analyzed by ESCA. The single-fiber contact angle with water, measured using a Cahn balance, decreased with extended ozonation. Measuring the time required for the sheet to absorb a water droplet with a high speed camera showed that even a very short ozone exposure (1 min) dramatically affected the absorption behavior. The rate of absorption dramatically increased after as little as 1 min of ozone exposure. This improvement in absorption rate was most likely due to the formation of low molecular weight degradation products, acting as wetting agents, created during the ozonation.
Introduction Wettability is a vital property in absorbency products made from lignocellulosic fibers, such as kitchen towels and diapers. The absorption of liquid into fiber assemblies is dictated by two factors: the fiber surface-water interaction and capillary forces.1 The fiber surface-water interaction is governed by the ability of the surface to interact with water through, for example, acid-base, electrostatic, and hydrogen bonding interactions. The capillary action is dictated by the surface roughness and the porous structure of the fiber network.2 Lignocellulosic fibers consist of cellulose, hemicellulose, lignin, and extractives. The polysaccharides cellulose and hemicellulose are hydrophilic in nature and form contact angles with water of 20-30°.1,3 The extractives comprise a wide range of different compounds and have a great influence on the wetting even though they are present in very low amounts. This is due to their hydrophobic nature and tendency to accumulate on the fiber surfaces.4,5 Lignin is rather hydrophobic; contact angels ranging from 55 to 75° have been reported.6 The composition of the different groups depends mainly on the mode of which the fibers are liberated. A chemical pulp is more hydrophilic than a mechanical pulp, since most of the hydrophobic compounds are removed during the chemical pulping process.7 However, the lignin-rich mechanical pulp fibers form networks with much higher bulk than chemical fibers do. * Corresponding author. E-mail: chalmers.se.
Paul.Gatenholm@chem.
(1) Hodgson, K. T.; Berg, J. C. Wood Fiber Sci. 1988, 20, 3-17. (2) Adamson, A. W. Physical Chemistry of Surfaces; Wiley: New York, 1990; p 495. (3) Luner, P.; Sandell, M. J. Polym. Sci., Polym. Symp. 1969, 28, 115-142. (4) Nylund, J.; Sundberg, K.; Shen, Q.; Rosenholm, J. B. Colloids Surf., A 1998, 133, 261-268. (5) Kokkonen, P.; Korpela, A.; Sundberg, A.; Holmbom, B. Nord. Pulp Pap. Res. J. 2002, 17, 382-386. (6) Lee, S. B.; Luner, P. Tappi J. 1972, 55, 116-121. (7) Stro¨m, G.; Carlsson, G. J. Adhes. Sci. Technol. 1992, 6, 745-761.
A network with high bulk is good for absorption properties, as it provides a large number of voids into which liquid can flow. The wood fiber undergoes chemical modification during classical pulping operation, and the chemistry has been studied for many decades. More recently, interesting developments in selective chemical oxidation in the wet stage8 and enzymatic oxidation9 have been reported. Modifying the surface chemistry of lignocellulosic fibers without changing the bulk properties has been the goal of many scientists in the past decades. Goring broke new ground with his pioneering work on improving the bond strength of cellulose strips by corona discharge treatment.10 The first to attempt to improve the wetting properties of cellulosic materials by surface modification were Brown and Swanson, who corona treated cellophane.11 Low temperature plasmas, working with various gaseous mixtures, have also been investigated.12-14 In addition to these physical surface modifications, wet chemical methods have also been evaluated, such as amidation, sulfonation, and hemicellulose adsorption.15-17 The ozone treatement of mechanical pulps has been described in the literature.18-21 The ozone treatment of (8) Besemer, A. C.; De Nooy, A. E. J.; Van Bekkum, H. Cellulose Derivatives; ACS Symposium Series 688; American Chemical Society: Washington, DC, 1998; pp 73-82. (9) Paice, M. G.; Bourbonnais, R.; Reid, I. D. Tappi J. 1995, 78 (9), 161-169. (10) Goring, D. A. I. Pulp Pap. Mag. Can. 1967, 68, T372-T376. (11) Brown, P. F.; Swanson, J. W. Tappi J. 1971, 54, 2012-2018. (12) Benerito, R. R.; Ward, T. L.; Soignet, D. M.; Hinojosa, O. Text. Res. J. 1981, 51, 224-232. (13) Carlsson, G.; Stro¨m, G. Langmuir 1991, 7, 2492-2497. (14) Carlsson, G.; Stro¨m, G. Nord. Pulp Pap. Res. J. 1994, 72-75. (15) Bo¨rås, L.; Gatenholm, P. Holzforschung 1999, 53, 429-434. (16) Westerlind, B. S.; Berg, J. C. J. Appl. Polym. Sci. 1988, 36, 523534. (17) Henriksson, Å.; Gatenholm, P. Cellulose 2002, 9, 55-64. (18) , De Choudens, C.; Lombardo, G.; Monzie, P. Rev ATIP 1978, 32 (9), 350-359. (19) McKelvey, R. D.; Thompson, N. S.; Lyse, T. E. Cellul. Chem. Technol. 1983, 17 (4), 355-61.
10.1021/la040066y CCC: $30.25 © 2005 American Chemical Society Published on Web 12/08/2004
Improvement of Properties of Lignocellulosic Fibers Table 1. HTCTMP Chemical Bulk Composition compound sugar
lignin extractives
type
total wt %
glucose mannose xylose galactose arabinose Klason in acetone
52.0 13.0 6.2 2.5 1.4 27.2 0.2
cellulosic materials in the gas phase has not yet found many uses in wood pulp fiber applications. Surface activation by ozonation as a pretreatment for graft polymerization,22 bleaching of old documents without using any wet chemistry,23 and promoting the enzymatic degradation of lignocellulosic material24 are a few of the applications found in the literature. Ozone is today found in the bleaching stage in some totally chlorine free (TCF) and elemental chlorine free (ECF) pulp mills.25 Numerous studies of the ozone reactions with the wood constituents in aqueous media have been carried out. Observations made in these studies are applicable to the reactions taking place in the flowing gas reactor used in this work, since the events taking place concern compounds from three different phases.26,27 In the gas reactor, the three-phase system found in a bleaching stage is used, but the proportions of the phases are inverted. The aim of this work was to improve the wetting properties of lignocellulosic fibers using a gas phase ozone treatment. Materials and Methods Pulp. A commercial grade high temperature chemi-thermomechanical pulp (HTCTMP) made from spruce chips at O ¨ rstrand pulp mill, SCA, Sweden, was used as the substrate. The chemical composition of the pulp is given in Table 1. Sheets with a grammage of 30 g/m2 were made in a dynamic sheet former. The pulp was defibrated at 10 000 revolutions. A traversing nozzle distributed the pulp suspension at 890 cm/min on a water-covered wire rotating at a speed of 1415 rpm. The sheet was then dewatered for 30 s at a maintained drum speed of 1415 rpm. Drying was done in a STFI plate drier (Fibertech, Sweden) for 3 min at 147 °C. Gas Phase Oxidation. Prior to ozone treatment, the sheets were stored overnight in water-saturated air. The water content of the sheets was 31 wt %. The experimental setup of the ozone reactor was developed prior to this work and is described in detailed by Karlsson.28 A flow of 150 L/h of pure oxygen gas was passed through the ozone generator (Thermex-HF NG 10 from Ozone Systems Company, Germany), which created ozone from the oxygen by means of electrical discharge. The outlet from the reactor contained 7 g of ozone/m3 of total volume gas (NTP). The ozone/oxygen mixture was then passed through a temperaturecontrolled (38 °C) water seal, which humidified the gas to ∼80% relative humidity. The humid gas stream then passed through the sample chamber, in which a 2 × 15 cm sample was mounted. The temperature in the reactor was maintained at 32 °C. Before the treated samples were collected, the system was purged from (20) Petit-Conil, M.; De Choudens, C.; Espilit, T. Nordic. Pulp Pap. Res. J. 1998, 13 (1), 16-22. (21) Long, P.; Hsieh, J. S.; Baosman, A. TAPPI Pulping/Process and Product Quality Conference, 2000, Boston, MA, Nov 5-8, 2000; pp 874881. (22) Karlsson, J. Department of Polymer Technology; Chalmers University of Technology: Go¨teborg, Sweden, 1998. (23) Giuliani, A.; Luciaini, M. Cellul. Carta 1963, 14, 12-24. (24) Ishibashi, T.; Ishida, M.; Odawara, Y. European Patent Application, 1982. (25) Johansson, E. E. Nuclear Chemistry; Royal Institute of Technology: Stockholm, Sweden, 2003; p 60. (26) Bin, A. K.; Duczmal, B.; Machniewski, P. Chem. Eng. Sci. 2001, 56, 6233-6240. (27) Xie, T.; Ghiaasiaan, S. M.; Karrila, S.; McDonough, T. Chem. Eng. Sci. 2003, 58, 1417-1430. (28) Karlsson, J.; Gatenholm, P. Polymer 1997, 38, 4727-4731.
Langmuir, Vol. 21, No. 1, 2005 161 unreacted ozone by a flow of nitrogen. The samples were treated for 1, 5, 10, 15, 30, and 60 min, respectively, and were analyzed 2 days after treatment. The samples were stored in a desiccator at room temperature between treatment and analysis. FT-IR. Fourier transform infrared (FT-IR) spectroscopy was carried out to investigate the changes in chemical structure of the sample caused by the ozone exposure. A Perkin-Elmer 2000 FT-IR instrument was used with Spectrum v2.00 software from Perkin-Elmer, U.S.A. Fibers were pulled at random from the sample, ground together with KBr, and pressed into tablets. The tablets were then reground in order to fragment the fibers and pressed again. The spectra of the samples were collected using transmission mode in the range 3000-400 cm-1. Twenty scans with a resolution of 4 cm-1 were made for each spectrum. Electron Spectroscopy for Chemical Analysis. The surface chemistry was investigated using electron spectroscopy for chemical analysis (ESCA). Data were collected using a Quantum 2000 scanning ESCA microprobe from Physical Electronics, U.S.A, with monochromatic Al as the X-ray source. An area of 500 µm × 500 µm was analyzed with a takeoff angle of 45°. Calculations of peak intensities and curve fitting were done with MultiPak software from Physical Electronics. The relative amounts of differently bound carbon were calculated with a Gaussian curve fitting of the high resolution C1s peak. The different positions of CsC (C1), CsO (C2), OsCsO or CdO (C3), and OsCdO (C4) were 285.0 ( 0.2, 286.7 ( 0.2, 281.1 ( 0.2, and 289.4 ( 0.2 eV, respectively. Contact Angle. The contact angle of single fibers was measured with a microbalance using the Wilhelmy plate method. The balance used was the DCA-322 from Cahn Instruments, U.S.A. A stage speed of 20 µm/s was used, and Millipore water (surface tension of γ ) 72.8 mN/m) was the wetting liquid. Single fibers were pulled at random from the samples with tweezers and mounted with adhesive tape to a holder which was hung on the microbalance. The principles for measuring the contact angle of wood and pulp fibers can be found elsewhere.29,30 The advancing contact angle was calculated assuming that the cosine of the receding contact angle is equal to 1. This assumption is based on the fact that the receding contact angle of lignocellulosic surfaces is very low.30 The results presented are the arithmetic average of 10 individual fibers. High Speed Camera. Initial water absorption tests were done with a Motion Scope PCI 500S digital high speed camera from Redlake Imaging, U.S.A., and an EDOS 5222 automatic pipet from Eppendorf, Germany. The camera captures images of a 5 µL droplet of deionized water being applied to the top of a sample sheet and absorbed into it. The time required for the sheet to absorb the water droplet is then measured as the time that elapses from the time of initial contact between the droplet and the sheet until the drop is completely gone. The test does not yield a contact angle of the sheet but gives a measurement of the rate of water absorption. Five measurements were made for each sample. Atomic Force Microscopy. The surface morphology of the fibers was investigated using atomic force microscopy (AFM) in tapping mode. Single fibers were pulled from the sheets by pressing the sheets onto adhesive tape. The tape also fixates the fibers during the analysis, which is crucial for attaining a micrograph. The equipment used was a Dimension 3000 large sample atomic force microscope with a type G scanner, Digital Instruments, U.S.A. A standard silicon tip was used. Height and phase images for fibers ozonated for 60 min and an untreated reference sample were captured.
Results FT-IR. The fibers subjected to ozone treatment were analyzed using FT-IR spectroscopy. The results are shown in Figure 1. To be able to quantitatively compare data from the IR spectra, normalization was done over the peak at 1045 cm-1 that originates from the C-O stretch in the carbohydrate ring. This is valid as long as no significant (29) Gardner, D. J.; Generalla, N. C.; Gunnels, D. W.; Wolcott, M. P. Langmuir 1991, 7, 2498-2502. (30) Klungness, J. H. Tappi J. 1981, 64, 65-66.
162
Langmuir, Vol. 21, No. 1, 2005
O ¨ stenson and Gatenholm
Figure 3. High resolution ESCA spectra of the C1s peak of untreated HTCTMP (left) and HTCTMP ozone treated for 30 min (right). Changes in peak intensities are marked by arrows.
Figure 1. IR spectra of untreated HTCTMP and HTCTMP ozone treated for 15, 30, and 60 min, respectively.
Figure 4. Evolution of the relative peak areas during ozonation, as detected by ESCA.
Figure 2. Relative peak height at 1735 cm-1 as a function of ozone treatment.
carbohydrate degradation takes place.31 Several studies of the kinetics of ozone-induced cellulose degradation show that carbohydrate degradation is reduced in the presence of lignin.32-35 The normalization is valid for short ozone exposure times, but some carbohydrate degradation might have taken place for the highest treatment dosage (60 min of ozone exposure), making the normalization procedure less certain. No significant changes in the spectra are found for treatment times below 5 min. From 5 min of treatment and onward, some changes are noted. Peaks (31) Marchessault, R. H.; Liang, C. Y. J. Polym. Sci. 1962, 59, 357378. (32) Kang, G.; Zhang, Y.; Ni, Y.; van Heiningen, A. R. P. J. Wood Chem. Technol. 1995, 15, 413-430. (33) Pan, G. Y.; Chen, C.-L.; Gratzl, J. S.; Chang, H.-m. Res. Chem. Intermed. 1995, 25, 205-222. (34) Wang, Y.; Hollingsworth, R. I.; Kasper, D. L. Carbohydr. Res. 1999, 319, 141-147. (35) Zhang, Y.; Kang, G.; Ni, Y.; Van Heiningen, A. R. P. J. Pulp Pap. Sci. 1997, 23, J23-J27.
Figure 5. Oxygen-to-carbon ratio as a function of ozone treatment.
owing to the formation of carbonyl groups develop in the IR spectra, at ∼1630 and ∼1735 cm-1. At the same time, the aromatic absorption (1450, 1500, and 1590 cm-1) de-
Improvement of Properties of Lignocellulosic Fibers
Langmuir, Vol. 21, No. 1, 2005 163
Figure 6. Tensiograms of untreated HTCTMP (a), HTCTMP ozone treated for 10 min (b), and HTCTMP ozone treated for 60 min (c).
creases. These results are in agreement with previous findings.36 Plotting the relative peak height at 1735 cm-1, originating from the carbonyl group, versus the time of ozone exposure shows that the degree of modification increases with the reaction time, as can be seen in Figure 2. ESCA. The results of surface analysis show that oxygen is incorporated into the surface during treatment and that aliphatic and/or aromatic carbons are being consumed during ozonation. This is clearly seen in the changes in the relative peak areas (see Figure 3). The C1 peak decreases, whereas the C3 and C4 relative peak areas increase. This was expected, since the possible oxidation reactions taking place at the lignocellulosic fiber surface involve the rupture and incorporation of oxygen into the aromatic ring in lignin.37,38 The reaction of ozone with polysaccharides leads to chain scission, but very few functional groups are introduced.39 The changes of the surface chemistry caused by ozone treatment are mainly the result of the oxidation of lignin at the surface. The changes in the C1 and C3 + C4 peak areas during the ozone treatment can be seen in Figure 4. The amount of oxygen that is incorporated increases with the reaction time up to 15 min. After 15 min, prolonged ozone exposure does not yield any further increase in the oxygen-to-carbon relative ratio, as seen in Figure 5. Dynamic Contact Angle. The contact angle of single fibers with water was measured with the Wilhelmy plate technique. Wetting force tensiograms for untreated fibers and fibers treated for 10 and 60 min are shown in Figure 6. The lower tracks in the figures are the measured forces as the fibers come into contact with water and relate to the advancing contact angle. The higher tracks are the (36) Olkkonen, C.; Tylli, H.; Forsskåhl, I.; Fuhrmaa, A.; Hausalo, T.; Tamminen, T.; Hjortling, B.; Janson, J. Holzforschung 2000, 54, 397406. (37) Månsson, P.; O ¨ ster, R. Nord. Pulp Pap. Res. J. 1988, 75-81. (38) Nakano, J.; Ishizu, A.; Hosoya, S.; Kaneko, H.; Matsumoto, Y. 1982 TAPPI Research and Development Division Conference; TAPPI: Asherville, NC, 1982; pp 61-70. (39) Chirat, C.; Lachenal, D. Holzforschung 1994, 48, 133-139.
Figure 7. Advancing contact angle as a function of ozone treatment time.
forces measured as the water retracts over the fiber surface and relate to the receding contact angle. Figure 6a shows a large difference between the advancing and receding angles, which is called hysteresis. The hysteresis between the advancing force and the receding force is a result of the surface roughness and chemical heterogeneity of real solid surfaces.40 The hysteresis is much lower after 10 min of ozone treatment because the advancing contact angle has decreased (see Figure 6b). The fiber that was ozone treated for 60 min shows no hysteresis, as shown in Figure 6c. The fiber advancing contact angle is zero, and the fiber wets completely. The arithmetically calcu(40) Johnson, R. E., Jr.; Dettre, R. H. J. Phys. Chem. 1964, 68, 174450.
164
O ¨ stenson and Gatenholm
Langmuir, Vol. 21, No. 1, 2005
Figure 8. Time required for complete absorption of the droplet as a function of ozone treatment time.
lated advancing contact angle averages are shown in Figure 7. The advancing contact angle decreases with longer ozone treatment. Most of the change in the advancing contact angle is noted after 10 min of exposure. Absorption. Measurements of the time needed for the sample sheets to absorb a 5 µL droplet of water were carried out using a high speed camera (HSC). Figure 8 shows the time required for complete absorption of the 5 µL droplet for samples treated to various degrees with ozone. A dramatic increase in the rate of absorption is evident after as little as 1 min of ozone exposure. The time required to absorb a droplet of water for the sheet treated for 1 min is reduced by a full 75% as compared to the untreated sample. The time required for complete absorption decreases with prolonged ozone exposure, but the decrease is only slight. Surface Topography. Atomic force microscopy was used to investigate changes in surface topography caused by the treatment. When scanning wood pulp fibers, great variations in surface topography are always observed when comparing images, even images from the same fiber. The most noticeable change in a comparison of images of the sample treated for 60 min with the untreated reference sample was the fact that no clear image could be made of the treated sample. Figure 9 shows the height and phase images of the untreated sample and the sample treated for 60 min. The untreated surface has a fibrillar structure, which is clearly seen in both the height and phase images. The height image of the treated samples shows a surface that is swollen and has some cracks. The dark lines in the phase image are indicative of loose particles being picked up from the sample surface and dragged around by the AFM tip. Additional Experiment. The pulp fibers that were ozone treated for 60 min showed no hysteresis when tested using the Wilhelmy plate method. Since AFM had indicated the presence of loose degradation products on the fiber surfaces, the nature of the excellent wetting properties of this sample was further investigated. Repeated runs, using DCA, were done using the same single fiber during all the runs. Three runs were done on a single fiber, and the probing liquid was changed between
Figure 9. AFM height images (left) and phase images (right) of untreated HTCTMP (A) and HTCTMP ozone treated for 60 min (B).
the different runs. The advancing contact angle in the first run was zero. It became apparent that the improved wettability was only temporary (see Figure 10). Almost all of the improvement was lost after the first run, as can be seen by the appearance of a contact angle hysteresis in Figure 10b. For the third run, the hysteresis increased even more, as seen in Figure 10c. The fiber that was ozone treated for 60 min had the same wetting properties (advancing contact angle of 60°) after three wetting cycles as an untreated fiber. The loose oxidization products dramatically improved the wetting properties, but the effect was not permanent. The wetting of the fiber during DCA testing caused the wetting agents created at the fiber surface during the ozone treatment to dissolve. Discussion The high speed camera shows that, after as little as 1 min of exposure to ozone, the absorption rate has increased dramatically. It was not possible using ESCA to identify any changes in the surface chemistry after such a short treatment time. One possible reason is a loss by desorption of degradation products, which might be low molecular weight, volatile organics in the high vacuum chamber of the ESCA instrument. In addition, no change in the advancing contact angle could be detected for the samples treated for short times. Even though a thermodynamic analysis of the penetration of a drop into a thin porous medium has been performed,41 no satisfactory theory has been developed describing such a process. More is known regarding the absorption of liquid from a limited reservoir into a porous network.42 However, successful attempts have been made to couple the contact angle of single fibers to the Lucas-Washburn equation, eq 1.43
r2 2γ cos θ dh ) ( Fgh sin β dt 8ηh r
(
)
(1)
(41) Marmur, A. J. Colloid Interface Sci. 1988, 123 (1), 161-169. (42) Danino, D.; Marmur, A. J. Colloid Interface Sci. 1994, 166 (1), 245-250.
Improvement of Properties of Lignocellulosic Fibers
Langmuir, Vol. 21, No. 1, 2005 165
Figure 10. Tensiograms of HTCTMP ozone treated for 60 min: first run (a); second run (b); third run (c).
The tremendous increase in absorption rate as a result of the ozone treatment, as measured by the HSC, cannot be explained solely by a decrease in the contact angle of the single fiber. A lowering of the surface tension of the probing liquid may be the reason for the unexpectedly high absorption rate. One possible explanation of the lowered surface tension could be the formation of low molecular weight degradation products during the oxidation reaction, which could possibly act as wetting agents. An explanation of this kind was suggested by Stro¨m and Carlsson for oxygen plasma treated pulp fibers.7 The amount of wetting agent at the surface is more than enough to enhance the wetting when it is evaluated by means of a small single droplet applied to the top of the surface of a sheet. When evaluating the advancing contact angle of single fibers, 10 min of ozone exposure was needed to achieve a decrease in contact angle. After 10 min of treatment, the changes in chemical composition were recorded by ESCA. It is plausible that the amount of low molecular weight degradation products that act as wetting agents increases with the ozonation time. This makes it evident that the relationship between the size of the sample and the volume of the probing liquid is of great importance. With the high speed camera, a 5 µL droplet is placed on the tissue surface, whereas the single fiber is lowered into a volume of 5 mL of water in the Wilhelmy method. This makes the high speed camera test more sensitive to low amounts of compounds at the surface that alter the wetting behavior of the sample. The concentration of a potential wetting agent in the 5 µL droplet is much higher than that in the Wilhelmy plate probing liquid. The time of absorption, as measured by the high speed camera, decreases for higher ozone dosages, but the effect of prolonged exposure was small (Figure 8). The single-fiber measurements showed that more than 10 min of ozone treatment was needed to achieve a decrease in the advancing contact angle. Further treatment led to even lower contact angles continuously up to 60 min of exposure (Figure 7). (43) Berg, J. C. Non-Wovens. An Advanced Tutorial; Tappi: Atlanta, GA, 1989; pp 219-232.
Conclusions This study investigated the effect of a novel gas phase ozonation method on the wetting properties of lignocellulosic fibers. It was possible to oxidize the lignocellulosic material in the gas phase to different degrees by varying the treatment time. Changes in the chemistry taking place during the ozonation were clearly seen with FT-IR spectroscopy and ESCA. Two observations were made using the surface-sensitive method, ESCA. First, during the first minutes of treatment, no alterations of the surface chemistry could be detected. Second, after 15 min of ozone exposure, the surface had become fully oxidized. Extending the treatment time failed to incorporate more oxygen into the surface. However, it was clear from FT-IR spectroscopy that the oxidization process was not completed after 15 min. This shows that ozone penetrates and reacts deeper into the fiber than the analysis depth of ESCA, which is typically 10 nm for this type of material. The wetting properties of the pulp fibers were greatly enhanced after ozone treatment. The effect on initial absorption rate, measured as the time needed for a water droplet to absorb into a sample sheet, could be seen clearly after as little as 1 min of ozone treatment. To detect a significant decrease in the contact angle of single fibers, 10 min of ozone exposure was required. The sample treated for 60 min showed no hysteresis between the advancing and receding wetting forces, wetting completely. The improved wetting properties were most likely the result of the presence of degradation products that acted as wetting agents at the fiber surface. These wetting agents dissolved from the fiber surface upon wetting, after which the improvement in wetting properties was lost. The very low treatment time needed to increase the absorption rate of the fiber network is also an attractive aspect from the point of view of production. LA040066Y
