PEER-REVIEWED
COATING
Changes with aging in the surface hydrophobicity of coated paper CECILIA LIDENMARK, SVEN FORSBERG, MAGNUS NORGREN, HÅKAN EDLUND, and OLA KARLSSON
ABSTRACT: Time-dependent changes in the surface properties of coated papers were studied as the evolution
of surface hydrophobicity of laboratory and commercially coated papers. We measured the apparent contact angles on the papers during several weeks post-production. Hydrophobicity upon aging increased for all samples made from traditional coating colors on mechanical base stock and on base stock made from cotton linters. Accelerated aging by heat treatment intensified the increase of the apparent contact angles and accelerated the time-dependent behavior. A key mechanism in these changes may be the spreading of latex particles in a coating. Application: The apparent contact angle increases over time for both commercially produced lightweight coated (LWC) papers and laboratory coated papers could explain variations in printability of coated papers from the same source. The time elapsed between paper production and printing, as well as storage conditions, can change printability.
L
ightweight coated (LWC) papers are used in magazines, brochures, and other products requiring highquality printing. Because the quality of the print obtained is related to the surface quality of the paper used, it is desirable to know as much as possible about the properties of the paper and the way they interact with other components in the printing process. Coated paper consists of two components: porous base stocks covered to some degree by a porous coating color. In reality, however, many components are involved. The base stock consists of thermomechanical (TMP) and chemical pulp fibers, to which inorganic fillers, retention aids, strength polymers, and various other additives may have been added. Mechanical pulp fibers have large amounts of lignin present, as well as polysaccharides, sterols, proteins, and other components. The presence and distribution of the various compounds in different fractions of TMP were investigated thoroughly by Kleen, Kangas, and Laine [1]. It is well known that bleaching affects the colloidal stability of pitch particles from TMP pulps [2]. Broke will almost always be part of the wet end. Consequently, components from the coating will be present in the wet end. In addition, the age of the various components in the wet end will vary depending on their retention, which is known to affect their surface properties [3]. The properties of the coating color can be varied, for instance, by choosing different types of pigment systems such as clays, ground marble calcium carbonates (GCC), precipitated calcium carbonates (PCC), titanium dioxide, or mixtures of these. There are also many ways to vary the size and shape distribution of the pigments. The rheology of the coating color is altered by the addition of solution polymers, such as anionic starch, hydroxyethyl cellulose (HEC) and carboxymethyl cellulose (CMC), or synthetic emulsion thickeners such as various 40
TAPPI JOURNAL | MAY 2010
forms of hydrophilic or hydrophobic polyacrylates. These polymers also act as binders to various extents. In addition, synthetic binders are added to provide strength and reduce dusting and chalking. Stabilizing agents such as surfactants or dispersion agents are added to prevent flocculation of the pigment and binder particles. The most commonly used binders are water-based latex emulsions, in which the two main components are the polymeric (colloidal) latex particles and the water serum. Various substances (e.g., low molecular weight oligomers, monomeric acids, and surfactants) remain, however, after emulsion polymerization in the water serum. Physicochemical and colloidal properties of latex particles vary, depending on the type and composition of constituent monomers, acids, surfactants, size distribution, the amount of crosslinking, and the emulsion polymerization process. To further complicate the situation, the final coating will be affected by a number of process parameters including the type of applicator, machine speed, nip pressures, drying set up, and calendering condition. Base sheet sorption properties and the presence and amount of latex, surface active chemicals, and choice of pigments in the coating color will govern the chemistry and microstructure of the coating layer, resulting in coatings of different properties. After a coating layer has dried, a nonuniform distribution of pigment particles and binder domains are formed. Above the critical pigment volume concentration (CPVC) normal for paper coatings, the coating layer is also porous. During the 1980s, a series of papers by Aspler et al. [4–7] showed that self-sizing affects wetting and sorption of newsprint. Ness and Hodgson [8] demonstrated that TMP extractives migrate during accelerated aging and also pointed out that these extractives were not chemically bound to fiber surfaces. Kokkonen et al. [9] showed that extractives covered fibers in an evenly distributed layer and that there is a correlation between
COATING surface coverage and water contact angle. There is no reason why these extractives should not continue towards a surface in the coating, if it is thermodynamically more favorable. Di Riso and Yan [10] suggested a correlation between the binder domains and the packing ability of the pigments, even when no binder migration occurs. The structure and chemistry of paper coatings during the drying and coalescence steps have been thoroughly investigated by Al-Turaif [11–12]. In his work, he showed that an enrichment of latex at the top of the coatings above CPVC could be correlated to the geometry of the pigment particles, where prismatic particles increase the porous structure needed for binder migration [11]. Al-Turaif also stated that the increase of latex on the coating surface, after coalescence had begun, was explained by a spreading mechanism of the latex on to the surface of the pigment particles [12]. This mechanism was governed by the dispersive contributions of the surface energy of the pigments and latex particles. As coating consolidation proceeds, the polar contribution in the surface energy reduces. Within the coating bulk, the spreading of latex was limited by the presence of platelike pigments that could cage the latex into discrete volumes. The spreading behavior of latex particles on mineral surfaces and its potential effect in coatings have been studied previously. Granier, Sartre, and Joanicot stated a relation between latex surface acidity and increased spreading on calcium carbonate [13]. Unertl (1998) showed that the surface energy is the same for single particles and film-formed styrene-butadiene latexes and that the particle spreads anisotropically in calcite [14]. Engqvist et al. showed that the spreading of latex particles with glass transition temperature (Tg) 10°C on silica surfaces is slow even at temperatures above Tg [15]. After 2 h at a temperature of +30ºC Tg, spreading of a single particle was not complete; latex particles continued to spread for several weeks at room temperature. Sample
Sheet Type
Surfactants present in a latex migrate toward an interface when film forming and cause a change in the surface chemistry when compared with the polymer bulk [16]. These types of residual surface-active chemicals are present also in the coating binder. Lafay et al. concluded that surfactants added to the coating color affected offset printing properties [17]. Both the process of migrating surface-active components and lipophilic substances and the process of latex spreading have the potential to gradually alter the surface composition of the coatings and change such properties as the hydrophobicity and liquid sorption of the coated paper. From an industrial point of view, these are important qualities, especially for a printer. The print quality and ink uptake are affected [17], as is the ink/fountain solution balance in the printing press. This latter condition is important for both the runnability of the printing press and ink emulsion, which affects the printing quality. Quality control is usually done directly after production of the paper, but weeks or months may elapse due to logistics and stockpiling before it is used for printing [18]. Therefore, the printer may find that paper reels from the same supplier behave differently in the printing press, depending on the time of their production. In addition, reels produced at the same time may vary in sorption properties due to different storage environments conditions (e.g., temperature and humidity) between production and printing. Several researchers have addressed time effects on the surface chemistry of fibers and papers, but little is known about how time affects the coating layer. Therefore, we set out to investigate whether aging has any detectable effect on the liquid sorption of coated papers. MATERIALS AND METHODS We prepared three series of samples and labeled them A–D. The samples in each series were also numbered so that every
Latex
Coating
Accelerated Aging
Drying
Production
Fiber
Monomers Tg
(°C)
g/m2
T (C°)
Time
105°C, 1h
A1
com.
TMP
Sty/But
10
11
on machine
–
–
A2
com.
TMP
Sty/But
10
11
on machine
–
Yes
A3
lab
TMP
Sty/But
10
11
175
1 min
–
A4
lab
TMP
Sty/But
10
11
175
1 min
Yes
B1
com.
TMP
Sty/But
10
12
on machine
–
–
B2
com.
TMP
Sty/But
10
12
on machine
–
Yes
B3
com.
TMP
Sty/But
10
8
on machine
–
–
B4
com.
TMP
Sty/But
10
8
on machine
–
Yes
C1
lab
chem.
Sty/But
10
11
60-70
1 min
–
D1
lab
TMP
Sty/But
5
12
175
1 min
Yes
D2
lab
TMP
Sty/But
5
12
175
1 min
Yes
Abbreviations: com. = commercially produced paper, lab = laboratory coated paper, TMP = thermomechanical fibers, chem.= cotton linters, But = butadiene, Sty = styrene.
I. Characteristics of the four different paper sample series and their preparation conditions. MAY 2010 | TAPPI JOURNAL
41
COATING sample had a specific name (e.g., B2 or C1). To avoid contamination and variations due to climate changes that could affect the measurements, all samples were handled while we were wearing cotton gloves and were kept in a room with a controlled climate of 23°C and 50% RH. Between test runs, the papers were stored in dark envelopes in the climate-controlled room. Table I presents the different series of samples. When the coating weight is 8 g/m2, the basestock is considered to be fully covered by the coating. This criterion was fulfilled in the tests we performed in our investigation. Variations between different papers of series were not regarded to substantially affect the results of our work.
Commercial coating papers The coating colors were approximately a 30/70 mix of kaolin and GCC with a styrene butadiene latex as a binder and starch as a co-binder and rheological modifier. Precise coating formulations of the commercial papers are not necessary to support our conclusions because the results from paper aging, the focal point of our study, are dependent on process parameters previously described.
Series A (1–4) Samples A1 and A2 were LWC paper coated on TMP basestock obtained fresh from a commercial machine equipped with a blade coater. The coating weight was 11g/m2. Coating color from the same batch was collected at the time and used to coat Samples A3 and A4 using a laboratory coater (DT Paper Science, Turku, Finland). The laboratory-coated paper was dried at maximum IR-effect on the coater (175°C). The coating weight was 11 g/m2 for all laboratory samples. Laboratorymade samples were calendered (15 bars, 60°C, three nips) after drying. Samples A2 and A4 were heat treated at 105°C for 1 h prior to testing.
Series B (1–4) Samples B1 and B2 were coated TMP basestock obtained fresh from the same machine as the samples in Series A. However, the binder formulation and type of the clay pigment was altered between the two samples. The coating weight was 12 g/m2. Samples B3 and B4 were also fresh coated TMP basestock, but from a different machine equipped with a film coater and another coating formulation. The coating weight was 8 g/m2. Samples B2 and B4 were aged at 105°C for 1 h prior to testing.
Imerys Minerals AB, Sundsvall, Sweden); 10 parts of carboxylated styrene-butadiene latex; and 0.5 parts of CMC. Sample C1 was calendered under 15 bars, 60°C. The coating weight was 11 g/m2 for C1.
Series D The coating color consisted of 70 parts of calcium carbonate (Carbital® 90), 30 parts of non-plastic (NP) clay; 10 parts carboxylated styrene-butadiene latex; and 0.5 parts Finnfix® 10 (CPKelco Oy, Äänekoski, Finland). After 30 min of stirring, the color was divided into two separate batches, D1 and D2, and D1 had no further treatment. To D2, 0.3% (based on dry color content) of sodium dodecyl sulfate (Calbiochem, EMD4Biochemicals, Gibbstown, NJ, USA) was added followed by 30 min of extra stirring. The dry content of both batches was determined to be 67%, and the colors were refrigerated for 48 h before further use. Samples D1 and D2 were coated on TMP basestock, 39 g/m2 with the same laboratory coater as Series A. Subsequently, the papers were calendered under 60°C, 15 bars, and three nips. The samples subjected to accelerated aging were heat treated at 105°C for a predetermined time. Before each test, samples were acclimatized for 1 h to the ambient environment in the climate-controlled room.
Dynamic angle tester (DAT) All contact angle measurements were performed on a dynamic angle tester (Fibro DAT AB, Stockholm, Sweden). This instrument uses image analysis to monitor the spreading of a liquid on a paper surface as a function of time [19]. The measurement involved two major processes: (1) the spreading of the liquid drop on the paper surface and (2) the sorption of the liquid into the paper. The angle was determined at the moment when the spreading reached an equilibrium state and the sorption was still negligible. In our study, the spreading of a water droplet was monitored during 10 s after impact with the coated papers. The contact angle between the drop and the surface was then evaluated as a function of time. Each value presented is an average of eight approved measurements with a margin of error of ±5%, unless otherwise stated. When evaluating the effect of aging of the samples, the values of the contact angles are those obtained after 1 s. The papers tested with the DAT were cut into strips at 45° angles to minimize effects from anisotropy in the sheets and stored in the laboratory where the tests were performed. Each strip was only tested once to avoid errors from the distorted surface after contact with water.
Series C (1) Basestock of precalendered cotton linters was used and coated on a bench coater (Dow, Horgen, Switzerland) in this series. Sample C1 was coated with a laboratory-prepared coating color and dried at 70°C for 2 min. Any extractives present in mechanical wood fibers were avoided by using the mostly cellulosic cotton linters. The laboratory coating color consisted of 70 parts of calcium carbonate (Carbital® 90, Imerys Minerals AB, Sundsvall, Sweden); 30 parts of clay (ND3090, 42
TAPPI JOURNAL | MAY 2010
Gloss tester The gloss tests were performed on Zehntner ZLR 1050 Laboratory Glossmeter (Zehntner GMBH, Hoelstein, Switzerland) according to ISO 8254-1 standard in a controlled climate at 23ºC and 50% RH.
Atomic force microscopy Atomic force microscopy images were obtained in a con-
COATING
1. Development of the apparent contact angle of water on a commercially coated paper (Sample A1) over a 10-s period. The test was performed on three different occasions. The time span is indicated by the number of days counting from the production date.
2. Results from aging a commercially made (A1) and the corresponding laboratory-made (A3) coated paper over a period of 58 days. The results from the same samples first subjected to accelerated aging (A2 and A4) are represented by the unfilled symbols.
trolled climate at a temperature of 23ºC and humidity of 50% RH with a Dimension 3100 (Veeco Instruments Inc., Plainview, NY, USA) in contact mode using standard silicon nitride tips with a nominal tip radius of 20 nm. RESULTS AND DISCUSSION
Contact angle measurements on paper surfaces It is virtually impossible to measure a “true” contact angle since the spreading of a droplet is influenced by the microroughness and surface chemical properties of the paper under investigation. However, contact angle measurements have become an accepted method of monitoring the liquid-paper surface interactions [19]. We used the term “apparent contact angle” in the following text for clarity to represent values of the contact angle obtained with the DAT technique. Figure 1 shows the development of the apparent contact angles for water droplets on Sample A1 over a period of 10 s at three different times after production of the paper. The apparent contact angles continued to change with time, since the surface was adsorbing the water. Note how the standard deviation of the results increased during the 10-s run. The time span between the different tests is indicated by the age of the samples. The values of the apparent contact angle increased with the age of a sample. Figure 2 shows data from Series A for a commercial coated paper (A1) and a corresponding laboratory-coated paper (A3). Additionally, data are included for the corresponding samples subjected to accelerated aging, i.e., commercial paper (A2) and laboratory paper (A4). Samples were subjected to analysis on five different occasions over a period of 58 days. Each individual symbol specifies the apparent contact angle 1 s after applying the water drop. There is correlation between the commercial and laboratory-made samples; both show an increase in the apparent contact angle of about 10º. The samples subjected to accelerated aging show a higher initial apparent contact angle and a less-pronounced increase over the same period.
3. Changes in apparent contact angle over a period of 140 days for two different commercially produced papers (B1 and B3). The results from the corresponding samples subjected to accelerated aging are represented by the unfilled symbols (B2 and B4).
Figure 3 present results of the aging test of Series A repeated once with Series B (made from another latex). This time, samples were also taken from a paper machine with a film coater and a different coating formulation in order to find out whether similar behaviour could be detected with different papers. For Samples B1 and B2, produced on the same machine as A1 but with a different coating formulation, the initial apparent contact angle was the same as in the previous test, but the change was less pronounced. There is a time-dependent phenomenon in the commercial paper coatings (Figs. 1 and 2). This behavior prevails with different paper grades from different machines and with various coating colors. It is possible to induce the observed aging process by heat treating the paper. One possible explanation for the aging behavior of the coated paper may be that extractives not only migrate to the MAY 2010 | TAPPI JOURNAL
43
COATING
4. Development of the coating hydrophobicity on a basestock of nearly pure cellulose (Sample C1).
fiber surface, but also to the surface of the coated paper, thereby increasing its hydrophobicity. To test this theory, a laboratory sample, C1, was made from cotton linters and a commercial coating color. In addition, C2 was made from the same type of linters and a color without any latex in the formulation. The high sorption of the paper when using cotton linters as a basestock causes high porosity in the coating layer and therefore a faster water sorption. This affects the apparent contact angle measurements, even during the first second of a measurement, and lowers the value of the apparent contact angle. Because our main interest was the variation in apparent contact angle over time, not the absolute value of the contact angle, this was acceptable. The sorption of water into C1 was higher compared with the samples in Series A and B that were made from commercial TMP basestock. Figure 4 shows the initial lower apparent contact angle (55°) with commercial samples starting at 65°– 75°. We observed a large increase of the apparent contact angle for this sample, probably partly caused by the lower energy input during drying and calendering and therefore reduced spreading of the latex and/or surface-active agents. Nevertheless, the change in hydrophobicity could not be caused by extractives, because extractives were absent in the cotton linter base paper. Instead, the increase in the apparent contact angle originated from the coating color. A previous study [15] of latex spreading onto an inorganic mineral surface concluded that the process is time-dependent and can continue for weeks. This may mean that latex particles in a coating continue to spread over time after coated paper manufacturing, changing the surface pore structure and hydrophobicity of the coating layer. Another possibility is that surfactants, or oligomers, originating from the latex emulsion used as the coating binder and from the makeup of the coating color could migrate to the surface and change the coated paper’s surface energy. In Series D, two laboratory prepared coatings (D1 and D2) were applied to a TMP basestock. The colors were produced in the same batch, then divided. Extra surfactant was added to 44
TAPPI JOURNAL | MAY 2010
5. Evolution of apparent contact angle for laboratory coated TMP papers of the same color, but D2 has an extra dose of surfactants added.
D2 to determine if its presence could cause an increase in apparent contact angle. If that were the case, D2 should have had a higher increase in contact angle than D1 when aged thermally. The samples were DAT-tested continuously from 0 h (no heat treatment performed) to 24 h of accelerated aging at 105°C. Figure 5 presents results from the test. The initial apparent contact angle for D2 and D1 are compared, indicating that the presence of surfactants at interfaces reduces the apparent contact angle of the coating. One possible reason is that excess surfactants reduce surface tension, resulting in a lower contact angle. However, the heat-induced aging of the two separate samples gave similar end results. The presence of excess surfactants did not exclusively cause the increase in coated paper hydrophobicity we observed. Another explanation of the observed increase in coating hydrophobicity could be a change in the microstructure of the coating surface over time, since one of the parameters affecting the contact angle value is roughness of the surface. Changes in surface roughness can be monitored as changes in coated paper microgloss. Therefore, we took gloss measurements of the samples of the D-series using five random points in the machine direction on each of the samples produced during the accelerated aging. The same samples were also subjected to AFM-surface analysis. Four pieces were randomly cut from each sample and measured at two arbitrarily chosen points of 20 µm2 x 20 µm2. The respective standard deviation was considered in the average root mean square roughness (rms) and maximum variation in z-direction. Table II presents the gloss and roughness data for Sample D2. The data from gloss measurements spread over a wide range within each sample set, so it is difficult to draw any clear conclusions from Table II. There is a possible small reduction over 24 h, but because of the statistical variations within the data, this cannot be certain. A statistical t-test was performed on the data for the three parameters and no variation between D2 at 0 h and at 24 h could be found. Gloss values for Sample
COATING Sample
Gloss
Roughness
%
rms (nm)
Z-span (nm)
D2 0 h
48,3 ± 5,9
156 ±20
1205 ± 137
D2 3 h
46,6 ± 3,6
D2 6 h
48,0 ± 1,9
D2 24 h
43,6 ± 6,2
153 ± 20
1125 ± 128
II. Gloss and surface roughness results for TMP samples in Series D.
D1 were in the same range as for Sample D2. It is also possible that the gloss measurement performed was at a macroscopic scale and did not define values at a microscopic-enough level to correlate with surface roughness. SUMMARY We measured apparent contact angles on coated papers and showed that they increase with time. We saw the same effect when the paper samples were heat treated. There may be many reasons for this aging effect. One reason could be production condition—such as heating during paper drying—but the results are consistent for paper made using two different paper machines, as well as for laboratory-coated papers. Migration of lipophilic substances from the TMP is not the exclusive reason, although it could still be a factor, but the effect occurs on cotton linter as well. Surface-active substances (i.e., surfactants) originating from the latex emulsion could change
the contact angle with time, but the addition of extra surfactant to the coating color in purpose did not increase the effect on samples under accelerated aging. Changes in surface roughness could be yet another cause, but macrogloss and AFM measurements could not detect any changes in surface roughness. The results point to the prolonged spreading of latex particles with age as the most likely cause of the changes in the apparent contact angle over time. However, the results do not exclude other mechanisms. Comparing Figs. 3 and 4, the values of contact angles are different. The drop is large compared with the coating layer, thus hydrophobicity of the base paper may affect the apparent contact angles. The effect was also seen with nearly pure cellulose cotton linters, which only showed small changes with aging. Excess surfactant could change the surface tension on the liquid drop and skew the results, but coatings with added surfactant showed the same trend as the other coatings. Air pollutants could contaminate the surface, but samples were placed in envelopes between testing and had relatively small exposure to pollutants in the form of airborne particles. CONCLUSIONS When the apparent contact angles of water on several types of coated paper were measured as a function of time from the production date of the papers, the hydrophobicity of the coated papers increased over a time span of several weeks and the
ABOUT THE AUTHORS
In a previous study, we found that single latex particles spreads very slowly on mineral surfaces. We asked ourselves if this could have any practical consequences. We also had previous experiences of sending commercial LWC papers to printers on other continents. Sometimes, from the same printer and for the same product, we got compliments for excellent printability and at other times we got comments that the sheet is way too sorbent. We set out to find out if there are any changes in absorbency with time. The challenge was to do this with relatively simple equipment on a complicated product, such as coated paper. We observed changes, and the results point to the synthetic latex binder as a possible cause of the changes. The next step is to find out if these changes have any effect in real-world printing. It would be interesting to monitor changes in latex spreading in paper coatings over time under ambient conditions. From the industrial point of view it would be of interest to do a systematic scale printing of reels, from the same source, but stored for long periods under various conditions. This would give an answer to the magnitude of printing problems created by storage and handling times.
Lidenmark
Forsberg
Norgren
Lidenmark is a Ph.D. student, Forsberg is tech lic., Norgren is professor, Edlund is associate professor, and Karlsson is professor at Karlsson Edlund FSCN (Fiber Science and Communication Network), Mid Sweden University, SE-851 70 Sundsvall, Sweden. Email Lidenmark at
[email protected].
MAY 2010 | TAPPI JOURNAL
45
COATING behavior was accentuated with increasing temperature. The phenomenon was found in commercial papers coated with different coating colors, as well as in laboratory-coated papers. The effect could not be correlated to the chemistry of the base paper or the presence of excess surfactants in the coating. Therefore, our conclusion is that the observed behavior originates from the coating color. It is a strong possibility that the prolonged spreading of latex particles into pigments is the main reason for this behavior upon coated paper aging. TJ ACKNOWLEDGEMENTS The authors gratefully acknowledge the Center for Amphiphilic Polymers (CAP) Lund, Sweden, and the European Commission (Objective 1) for financial support. LITERATURE CITED 1. Kleen, M., Kangas, H. and Laine, C., Nordic Pulp Paper Res. J. 18(4): 361(2003). 2. Thornton, J.W., “Dissolved and colloidal substances in the production of wood-containing paper,” Tech.D. thesis, Åbo Akademi, Turku, Finland, 1993. 3. Rundlöf, M., Htun, M., Höglund, H., et al., J. Pulp Paper Sci. 26(9): 308(2000). 4. Lyne, M.B. and Aspler, J.S., Tappi J. 5(12): 98(1982). 5. Aspler, J.S., Davis, S. and Lyne, M.B., Tappi J. 67(9): 128(1984). 6. Aspler, J.S. and Lyne, M.B., Tappi J. 67(10): 96(1984). 7. Lepoutre, P., Inoue, M., and Aspler, J.S., Tappi J. 68(12): 86(1985). 8. Ness, J. and Hodgson, K.T., Nordic Pulp Paper Res. J. 14(2): (1999). 9. Kokkonen, P., Fardim, O., and Holmbom, B., Nordic Pulp Paper Res. J. 19(3): 318(2004). 10. Di Risio, S. and Yan, N., Colloids Surf. A 289(1–3): 65(2006). 11. Al-Turaif, H., TAPPI J. 5(8): 24(2006). 12. Al-Turaif, H. and Lepoutre, P., Prog. Org. Coat. 38(1): 43(2000). 13. Granier, V., Sartre, A., and Joanicot, M., Langmuir 11(6): 2179(1995). 14. Unertl, W.N., Langmuir 14(8): 2201(1998). 15. Engqvist, C., Forsberg, S., Norgren, M., et al., Colloids Surf. A 302(1–3): 197(2007). 16. Aramendia, E., Mallegol, J., Jeynes, C., et al., Langmuir 19(8): 3,212(2003). 17. Lafaye, J.F., Gervason, G., Maume, J.P., et al., Tappi J. 70(8): 43(1987). 18. Edbom, M., personal communication. 19. Wågberg, L. and Westerlind, C., Nordic Pulp Paper Res. J. 15(5): 598(2000).
46
TAPPI JOURNAL | MAY 2010
