On the Indirect Polyelectrolyte Titration of Cellulosic Fibers. Conditions

A. Elisabet Horvath, Tom Lindström*, and Janne Laine. STFI-Packforsk ...... Laine, J.; Stenius, P.; Carlsson, G.; Ström, G. Cellulose 1994, 1, 145. ...
4 downloads 0 Views 140KB Size
Download PDF
824

Langmuir 2006, 22, 824-830

On the Indirect Polyelectrolyte Titration of Cellulosic Fibers. Conditions for Charge Stoichiometry and Comparison with ESCA A. Elisabet Horvath, Tom Lindstro¨m,* and Janne Laine† STFI-Packforsk AB, Box 5604, SE-114 86 Stockholm, Sweden ReceiVed August 15, 2005. In Final Form: October 18, 2005 The effect of electrolyte (NaHCO3) concentration on the adsorption of poly-DADMAC (poly-diallyldimethylammonium chloride) onto cellulosic fibers with different charge profiles was investigated. Surface carboxymethylated fibers were obtained by grafting carboxymethyl cellulose (CMC) onto the fiber surface and bulk carboxymethylated fibers were obtained by reacting the fibers with monochloroacetic acid. It was shown that nonionic interactions do not exist between cellulose and poly-DADMAC, rather electrostatic interactions govern the adsorption. Charge stoichiometry prevails under electrolyte-free conditions, whereas surface charge overcompensation occurs at higher electrolyte concentrations. It was shown that charge stoichiometry prevails if the thickness of the electric double layer κ-1 was larger than the mean distance between the charges on the fiber surface, as predicted by polyelectrolyte adsorption theories, taking lateral correlation effects into account. In a second set of experiments the ESCA technique served to independently calibrate the polyelectrolyte titrations for determining the surface charge of cellulosic fibers. Various molecular masses of poly-DADMAC were adsorbed to carboxymethylated fibers having different charge profiles. The adsorption of low Mw poly-DADMAC (7.0 × 103), analyzed by polyelectrolyte titration, was about 10 times higher than that of the high Mw poly-DADMAC (9.2 × 105). Despite the difference in accessibility of these two polyelectrolytes to the fiber cell wall, ESCA surface analysis showed, as expected, only slight differences between the two polyelectrolytes. This gives strong credibility to the idea that surface charge content of cellulosic fibers can be analyzed by means of adsorption of a high-molecular-mass cationic polymer, i.e., by polyelectrolyte titration.

Introduction The adsorption of polymers is important for many applications in the industry, and there are several theories describing adsorption.1-5 For polyelectrolytes, the driving force for adsorption onto an oppositely charged surface is of an electrostatic nature.6-8 The charge density of both the polyelectrolyte and the surface affects adsorption. Electrolyte concentration and, in some cases, pH, are important factors as well. According to polyelectrolyte adsorption theories,4,6,7 adsorption should decrease with increasing electrolyte concentration when charge interaction is the primary reason for adsorption. Van de Steeg et al.9 found that polyelectrolyte adsorption could be divided into two regimes, a screening-reduced adsorption regime and a screening-enhanced adsorption regime. In the screening-reduced regime, adsorption decreases with increasing electrolyte concentration if the interactions are purely electrostatic. The screening-enhanced regime appears when a nonelectrostatic interaction is introduced. A maximum in the adsorbed amount as a function of electrolyte concentration exists in the screeningenhanced adsorption regime. Experiments have also shown that * To whom correspondence should be addressed. E-mail: [email protected].. † Present address: Helsinki University of Technology, Laboratory of Forest Products Chemistry, P.O. Box 6300, FIN-02015 HUT, Finland. (1) Flory, P. J. Principles of Polymer Chemistry, 1st ed.; Cornell University: Ithaca, NY, 1953. (2) de Gennes, P. G. Macromolecules 1981, 14, 1637. (3) van der Schee, H. A.; Lyklema, J. J. Phys. Chem. 1984, 88, 6661. (4) Fleer, G. J.; Cohen Stuart, M. A.; Scheutjens, J. M. H. M.; Cosgrove, T.; Vincent, B. Polymers at Interfaces, 1st ed.; Chapman and Hall: London, 1993. (5) Dobrynin, A. V.; Deshkovski, A.; Rubinstein, M. Macromolecules 2001, 34, 3421. (6) Hesselink, F. TH. J. Colloid Interface Sci. 1977, 60, 448. (7) Cohen Stuart, M. A.; Fleer, G. J.; Lyklema, J.; Norde, W.; Scheutjens, J. M. H. M. AdV. Colloid Interface Sci. 1991, 34, 477. (8) Andelman, D.; Joanny, J.-F. Polym. Interfaces 2000, 4, 1153. (9) van de Steeg, H. G. M.; Cohen Stuart, M. A.; de Keizer, A.; Bijsterbosch, B. H. Langmuir 1992, 8, 2538.

this is the case for the adsorption of various polyelectrolytes onto silica,7,10 latex,11 mica,12 and cellulose.13-15 Self-consistent field (SCF) lattice theories (e.g., in refs 4,16, and 17) predict that charge overcompensation can only occur if a nonionic interaction exists between the polyelectrolyte segments and the surface. Dobrynin et al.5 developed a scaling theory for polyelectrolyte adsorption and predicted that the polyelectrolyte surface excess in a 3D adsorbed layer increases at low ionic strength and decreases at higher ionic strength. Netz and co-workers18,19 discussed adsorption of polyelectrolytes onto a substrate. The authors went beyond the mean field theory and considered lateral correlation effects. An adsorption phase diagram was presented for strongly charged polyelectrolytes as a function of substrate charge density and inverse screening length. At constant substrate charge density, charge compensation occurs at low electrolyte concentrations. Increases in electrolyte concentration cause charge reversal of the substrate, followed by polyelectrolyte desorption. In essence, these authors show that there is no need for nonelectrostatic interactions in order to get a screening-enhanced adsorption. This affects the formation of, for instance, polyelectrolyte multilayers.18,20,21 (10) Shubin, V. J. Colloid Interface Sci. 1997, 191, 372. (11) Shubin, V.; Samoshina, Yu.; Menshikova, A.; Evseeva, T. Colloid Polym. Sci. 1997, 275, 655. (12) Rojas, O. J.; Ernstsson, M.; Neuman, R. D.; Claesson, P. M. Langmuir 2002, 18, 1604. (13) Tanaka, H.; Tachiki, K.; Sumimoto, M. Tappi J. 1979, 62, 41. (14) Lindstro¨m, T.; Wågberg, L. Tappi J. 1983, 66, 83. (15) van de Steeg, H. G. M.; de Keizer, A.; Cohen Stuart, M. A.; Bijsterbosch, B. H. Nord. Pulp Pap. Res. J. 1993, 8, 34. (16) Shubin, V.; Linse, P. J. Phys. Chem. 1995, 99, 1285. (17) Shubin, V.; Linse, P. Macromolecules 1997, 30, 5944. (18) Netz, R. R.; Joanny, J.-F. Macromolecules 1999, 32, 9013. (19) Netz, R. R.; Andelman, D. Phys. Rep. 2003, 380, 1. (20) Decher, G. Science 1997, 277, 1232. (21) Dubas, S. T.; Schlenoff, J. B. Langmuir 2001, 17, 7725.

10.1021/la052217i CCC: $33.50 © 2006 American Chemical Society Published on Web 12/15/2005

Polyelectrolyte Titration of Cellulosic Fibers

Charge stoichiometry during polyelectrolyte adsorption is often observed for substrate systems at the limit of infinite dilution of an electrolyte.7 Kokufuta and Takahashi22 investigated the adsorption of poly-DADMAC onto silica and found that stoichiometry between the cationic polyelectrolyte and anionic surface prevails in water but deviates in the presence of electrolyte. Cellulose is another substrate of interest. It was argued in a publication by Rojas et al.23 that regenerated cellulose has so few charged groups that the adsorption of acrylamide and 3-(2methylproionamido)propyl trimethylammonium chloride was governed by nonelectrostatic interaction as well as an electrostatic component. However, cellulose fibers are more complicated as they are highly charged, partly fibrillated, and porous. As fibers do not have a smooth surface, it is difficult to separate the polyelectrolytes adsorbing on the surface from those penetrating into the pores and the interstices between the surface fibrils. The surface of the fibers and its charges are, however, a very important factor in the papermaking industry, and it is important to be able to measure the surface charge content. Prerequisites for measuring the surface charge of cellulosic fibers are that the polyelectrolyte has a sufficiently high molecular mass, to be excluded from the fiber cell wall, and that charge stoichiometry prevails. Horn24 found that the polyelectrolyte complex is stoichiometric with respect to charge if the charge density of a polyelectrolyte is sufficiently high and the ionic strength is sufficiently low (see also refs 25-28). Several publications29-33 have also shown that adsorption of cationic polyelectrolytes onto cellulosic fibers obey stoichiometry at the limit of low polyelectrolyte concentrations in the absence of electrolytes. However, the issue has not been resolved for poly-DADMAC adsorption onto cellulose, prompting this investigation of the effects of electrolyte concentration and to deduce under which conditions there is a deviation from charge stoichiometry. Second, the indirect polyelectrolyte titration procedure is often used for cellulose substrates with a wide range of surface charge densities (e.g., in refs 34, 35). Therefore, it is of interest to study the effects of bulk and surface charge density on poly-DADMAC adsorption. The charge density has been altered by using two methods: bulk carboxymethylation36 and surface grafting of carboxymethyl cellulose.37 The final objective in this part of the study is a discussion of the adsorption by using a simple model based on the mean distance between the surface charges and the thickness of the diffuse electric double layer, which will give a semiquantitative explanation of the obtained results. This concept is indeed similar (22) Kokufuta, E.; Takahashi, K. Macromolecules 1986, 19, 351. (23) Rojas, O. J.; Ernstsson, M.; Neuman, R. D.; Claesson, P. M. J. Phys. Chem. B 2000, 104, 10032. (24) Horn, D. Prog. Colloid Polym. Sci. 1978, 65, 251. (25) Tanaka, H. Jpn. TAPPI J. 1983, 37, 75. (26) Tanaka, H. Jpn. TAPPI J. 1983, 37, 39. (27) Plunkett, M. A.; Claesson, P. M.; Ernstsson, M.; Rutland, M. W. Langmuir 2003, 19, 4673. (28) Claesson, P. M.; Dedinaite, A.; Rojas, O. J. AdV. Colloid Interface Sci. 2003, 104, 53. (29) Winter, L.; Wågberg, L.; O ¨ dberg, L.; Lindstro¨m, T. J. Colloid Interface Sci. 1986, 111, 537. (30) Wågberg, L.; Winter, L.; O ¨ dberg, L.; Lindstro¨m, T. Colloids Surf. 1987, 27, 163. (31) Wågberg, L.; O ¨ dberg, L.; Lindstro¨m, T.; Aksberg, R. J. Colloid Interface Sci. 1988, 123, 287. (32) Wågberg, L.; O ¨ dberg, L.; Glad-Nordmark, G. Nord. Pulp Pap. Res. J. 1989, 4, 71. (33) Wågberg, L.; O ¨ dberg, L. Nord. Pulp Pap. Res. J. 1989, 4, 135. (34) Laine, J.; Stenius, P. Pap. Puu 1997, 79, 257. (35) Zhang, Y.; Sjo¨gren, B.; Engstrand, P.; Htun, M. J. Wood Chem. Technol. 1994, 14, 83. (36) Walecka, J. A. Tappi J. 1956, 39, 458. (37) Laine, J.; Lindstro¨m, T.; Glad-Nordmark, G.; Risinger, G. Nord. Pulp Pap. Res. J. 2000, 15, 520.

Langmuir, Vol. 22, No. 2, 2006 825 Table 1. Properties of Poly-DADMAC molecular mass (Da) 7.0 × 10 2.2 × 104 1.4 × 105 9.2 × 105 3

Mw/Mn 1.4 2.1 2.8 6.3

to the above-discussed correlation effects in polyelectrolyte adsorption theory. In a second part of this paper, the polyelectrolyte adsorption method was validated by using an independent surface-sensitive method (ESCA). Electron spectroscopy for chemical analysis (ESCA) provides information on the relative amount of different elements on the surface of materials. Photoelectrons interact with the atoms of the material, such that the intensity of escaping electrons decreases exponentially as the depth of analysis increases. The analysis depth has been reported to be between 6 and 12 nm for polymeric materials.38 The amount of carboxylic groups on a cellulosic fiber surface can be monitored with ESCA by high-resolution analysis of the carbon (C 1s) peak. However, the carboxyl carbon band is typically rather small for cellulosic fibers,39 meaning that the determination of carboxyl groups with the detailed carbon analysis is relatively uncertain. An additional possibility is offered by replacing exchangeable protons in the fibers by other ions or by adding a cationic polyelectrolyte that interacts with anionic groups and can be detected by ESCA. Two sets of experiments were performed to verify whether the polyelectrolyte titration method could be used to determine surface charge properties of cellulosic fibers by comparing it with ESCA. The effect of the poly-DADMAC molecular mass was first compared with the two techniques. The adsorption of high-molecular-mass poly-DADMAC on carboxymethylated pulps with different charge profiles was analyzed in a second set by using both polyelectrolyte titration and ESCA. Materials and Methods The pulp used in all experiments was a never-dried unbeaten elementary chlorine-free (ECF)-bleached softwood (spruce) kraft pulp (M-real, Husum, Sweden). A Cellecofilter with 100 µm screening slots was used to remove the fines (20-25%) prior to experiments. A fractionated poly-DADMAC (Ciba, Yorkshire, U.K.), see Table 1, was used as the polyelectrolyte. The charge density was determined by direct polyelectrolyte titration24,40 to be 5.9 × 10-3 eq/g (theoretical value is 6.19 × 10-3 eq/g) by using a commercially available potassium polyvinylsulfate (Waco). The charge density of this polymer is 6.16 × 10-3 eq/g, and the molecular mass was 3 × 105 Da. The Mw and molecular mass distribution was determined on TSK gel columns (Tosoh Corp. Japan) by using appropriate data software (PL caliber GPC/SEC software version 7.01, Polymer Laboratories). Molecular mass standards were poly(ethylene oxide)s (Tosoh Corp. Japan). A more detailed description (apart from the software used) of the account of this method has also been published41 from this lab. Sodium hydrogen carbonate (NaHCO3) from Merck KGaA, Darmstadt, was used without any further purifications as the electrolyte. Pretreatment of Pulp. An excess of 10-2 M HCl was used to wash the pulp (pulp consistency ∼1%) in order to remove the metal ions. The pH was adjusted to 2 and kept at this pH for 30 min. The (38) Ashley, J. C.; Williams, M. W. Radiat. Res. 1980, 81, 364. (39) Laine, J.; Stenius, P.; Carlsson, G.; Stro¨m, G. Cellulose 1994, 1, 145. (40) Terayama, H. J. Polym. Sci. 1952, 8, 243. (41) Swerin, A.; Wågberg, L. Nord. Pulp Pap. Res. J. 1994, 9, 18.

826 Langmuir, Vol. 22, No. 2, 2006 pulp was then washed with deionized water until the conductivity of the filtrate was below 5 µS/cm. This state is defined as the hydrogen form of the pulp. All adsorption isotherms were carried out with the pulp in its sodium form. To transfer the pulp to its sodium form, the pulp was washed with an excess of 10-3 M NaHCO3 (pulp consistency ∼1%) and soaked in this electrolyte for 10 min. The pH was then adjusted to 9 with NaOH and held constant for 30 min. The pulp was subsequently washed with deionized water until the conductivity of the filtrate was below 5 µS/cm. This is defined as the sodium form of the pulp. The exchange to the sodium form has been checked by comparing the sodium content with the number of carboxylic groups on the pulp. By using the outlined procedure, the two determinations give equal numbers for a number of bleached kraft pulps. Bulk Carboxymethylation. Bulk carboxymethylation was carried out according to a method developed by Walecka.36 The term bulk carboxymethylation is used to describe a method where carboxymethylation occurs uniformly throughout the fiber cell wall, i.e., there is no topochemical specificity. Solvent exchange from water to ethanol was carried out. Various amounts of monochloroacetic acid were dissolved in several 2-propanol solutions, depending on the desired degree of substitution (DS). The pulp was then impregnated with the various solutions for 30 min. Meanwhile, a solution of NaOH dissolved in methanol was added to 2-propanol and then heated close to boiling. The impregnated pulps were added to the NaOH/2-propanol solutions, and refluxed for 1 h. After being filtered and washed with a sequence of deionized water, 10-1 M acetic acid, and deionized water again, the pulps were finally placed in a 4% NaHCO3 (consistency 1%) solution for 1 h, then filtered, and finally washed with deionized water. Three levels of DS were achieved: 0.009, 0.045, and 0.076. Surface Carboxymethylation (CMC Grafting). Surface carboxymethylation was carried out by using carboxymethyl cellulose (CMC) according to a method developed by Laine et al.37 The term surface carboxymethylation is used for a method where the carboxymethylation has topochemical specificity. The pulp, in its sodium form, was placed in a solution of 5 × 10-2 M CaCl2 and 10-2 M NaHCO3. CMC (Finnfix WRH, Noviant, Finland, DS ) 0.52, Mw ) 1 × 106 Da) was added to the pulp, which was subsequently diluted to a pulp consistency of 2.5% and then heated at 120 °C for 2 h in a pressurized vessel. This was followed by washing with deionized water until the conductivity of the filtrate was below 5 µS/cm. The pulp was subsequently transferred to its hydrogen form, followed by a 2 h leaching period of the pulp into its sodium form in order to remove excess CMC, which had not been attached to the fibers. Three samples were prepared, 4 mg/g, 8 mg/g, and 14 mg/g grafted CMC. Charge Determination. The total charge of the pulps was determined by conductometric titration, according to Katz et al.42 Polyelectrolyte adsorption followed a procedure by Winter et al.29 After being adjusted to the desired electrolyte concentration, adsorption was carried out with pulp in its sodium form at a pulp concentration of 5 g/ L. All adsorption measurements were carried out with the pulps in their Na form. The pH was 8.0 ( 0.1 in all adsorption experiments. An appropriate excess of poly-DADMAC was added, and the suspension was shaken until adsorption equilibrium was reached (30 min). The fibers were separated from the solution by filtration and dried in order to record the dry weight. The filtrate was saved and titrated by using the polyelectrolyte titration24,40 procedure in order to determine the amount of polyelectrolyte adsorbed. Sheet Preparation for ESCA Measurements. Adsorption of poly-DADMAC onto the fibers was also determined by making handsheets (grammage 60 g/m2) and subjecting them to ESCA analysis (only Mw 7.0 × 103 and 9.2 × 105). Four different amounts of poly-DADMAC were added to obtain an adsorption isotherm (in 10-5 M NaHCO3.) ESCA measurements were carried out at Helsinki University of Technology by using an AXIS-HS spectrometer from Kratos Analytical. Monochromatized Al KR radiation was used to excite the electrons. Analyses were made on three different locations (42) Katz, S.; Beatson, R. P.; Scallan, A. M. SVen. Papperstidn. 1984, 87, R48.

HorVath et al.

Figure 1. Adsorption isotherms at different electrolyte concentrations for poly-DADMAC, molecular mass 9.2 × 105 Da, adsorbed onto bleached softwood kraft pulp.

Figure 2. Charge ratio as a function of electrolyte concentration during adsorption of poly-DADMAC, molecular mass 9.2 × 105 Da, onto bleached softwood kraft pulp. in each sample. The amount of adsorbed poly-DADMAC was determined by monitoring the nitrogen (N 1s) content on the fiber surfaces. By using a high-molecular-mass poly-DADMAC (Mw ) 9.2 × 105, Mw/Mn ) 6.3), the determination of the fiber surface charge by polyelectrolyte titration was verified with the ESCA technique. Different toposelectively modified carboxymethylated pulps were used. The added amount of poly-DADMAC in the ESCA analyses corresponded to an equilibrium concentration of 8.50 ( 3 mg/ L in the adsorption isotherms. Water Retention Value. The water retention value (WRV), a measure for pulp swelling,43 was determined according to the proposed SCAN method (SCAN-C 102 XE; 15 min, 3000 g).

Results and Discussion Effect of Electrolyte Concentration on Poly-DADMAC Adsorption. The adsorption of high-molecular-mass polyDADMAC (9.2 × 105 Da) onto bleached softwood kraft pulp was carried out at various electrolyte (NaHCO3) concentrations. Typical adsorption isotherms are displayed in Figure 1. The charge ratio (adsorbed amount of cationic charges/total number of anionic charges on the fiber) was calculated from each adsorption isotherm, and the results are displayed in Figure 2. Similar results has also been reported by Tanaka and coworkers.13 Adding electrolyte increased the adsorption of polyDADMAC up to 10-1 M NaHCO3, after which the adsorption decreased to zero. A bleached kraft pulp has comparatively few charges and has limited polyelectrolyte swelling. Corresponding water retention values for this pulp are 133, 135, 136, 129, and 124 for this pulp in deionized water, 10-4, 10-3, 10-2, and 10-1 M NaHCO3, respectively. The ionic conditions near a charged surface can be described as a diffuse electric double layer, the “thickness” of which equals (43) Scallan, A. M. Tappi J. 1983, 66, 73.

Polyelectrolyte Titration of Cellulosic Fibers

Langmuir, Vol. 22, No. 2, 2006 827 Table 2. Total Charge Density and Surface Charge Density for Bulk Carboxymethylated (DS ) 0-0.076) and Carboxymethyl Cellulose Surface-Grafted Fibers (0-14 mg/g Grafted CMC)a

Figure 3. Schematic of polyelectrolyte adsorption at (a) low and (b) high electrolyte concentration.

the Debye length (κ-1), which depends on the properties of the liquid rather than on the surface charge density. The distance between the fiber charges, d, may be much less than the thickness of the electric double layer (d , κ-1) in a solution with low electrolyte concentration so that the surface charges act as a mean field of charges instead of a series of point charges (see Figure 3). The same is, of course, valid for the charges on the polyelectrolyte. The interaction of the polyelectrolyte and the fiber surface can, therefore, be considered as an interaction between two mean fields of charge, and 1:1 stoichiometry will prevail. When the electrolyte concentration increases, κ-1 will decrease, and eventually, d will exceed κ-1. The polyelectrolyte coils due to a decreased intrachain repulsion between the charges along the backbone, which are effectively screened by the electrolyte. The charges on the polyelectrolyte and on the fiber surface instead behave as discrete charges, causing the adsorption to deviate from 1:1 stoichiometry. Under such conditions, the geometric fit between the polyelectrolyte charges and the surface charges is expected to be critical. This is comparable with the theory of polyelectrolyte adsorption discussed by Netz and Joanny18 and Netz and Andelman,19 and will be discussed in further detail below. It has earlier been shown that the adsorption is stoichiometric (i.e., the number of released counterions on the fibers ) the number of adsorbed cationic charges) at the limit of zero electrolyte concentration.44,45 The charge ratio approaches a constant value at low electrolyte concentrations, provided the poly-DADMAC has a sufficiently high molecular mass to be excluded from the fiber cell wall. As seen in Figure 2, the adsorption increases with increasing electrolyte concentration up to 10-1 M NaHCO3, after which it sharply decreases to zero at high electrolyte concentrations. The initial increase in adsorption is due to at least two factors: the swelling of the adsorbed layer,21,46 resulting in a deviation from charge stoichiometry, and a possible penetration of the coiled polyelectrolytes into the swollen fiber cell wall. The relative importance of the two factors is difficult to assess at this moment, but the swelling of the adsorbed layer is certainly a dominating factor at low and intermediate electrolyte concentrations, as will be discussed further. The lack of adsorption at high electrolyte concentrations suggests that only electrostatic interactions govern the adsorption process. Effects of Bulk and Surface Charge Density on PolyDADMAC Adsorption. Fiber charge density varies for different papermaking pulps. To change the charge density in a controlled manner, untreated pulp was bulk and surface carboxymethylated. In the bulk carboxymethylation procedure, hydroxyl groups were reacted in a topochemically nonselective manner with monochloroacetic acid in 2-propanol. Table 2 shows the total charge density (as determined by conductometric titrations) and surface charge (44) Lindstro¨m, T. Fundamentals of Papermaking, Transactions of the 9th Fundamental Research Symposium held at Cambridge; Baker, C. F., Punton, V. W., Ed.; Mechanical Engineering Publications Ltd.: London, 1989; p. 309. (45) Wågberg, L.; O ¨ dberg, L.; Glad-Nordmark, G. STFI Medelande A-989; Swedish Pulp and Paper Research Institute: Stockholm, Sweden, 1991. (46) Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153.

b

pulp treatment

total charge (µeq/g)

surface charge (µeq/g)

charge ratio

DS 0.000 DS 0.009 DS 0.045 DS 0.076 0 mg/g CMC 4 mg/g CMC 14 mg/g CMC

37 89 314 506 34 45 74

1.7 4.1 16 34 1.6 10 21

0.05 0.05 0.05 0.07 0.05 0.22 0.29

surface selectivityb 0.05 0.05 0.07 0.75 0.49

a All the pulps in the table have never been dried during preparation. ∆ surface charge/∆ total charge

Figure 4. Charge ratio for bulk carboxymethylated pulp as a function of electrolyte concentration during adsorption. The polyelectrolyte used was poly-DADMAC with molecular mass 9.2 × 105 Da.

(as determined by polyelectrolyte titration by using highmolecular-mass poly-DADMAC (Mw ) 9.2 × 105 Da)) for the carboxymethylated pulps. The nonselective bulk carboxymethylation is shown by the fact that the surface charge ratio is around 0.05 regardless of the degree of substitution, i.e., 5% of the charges are located on the fiber surface. It is possible to graft carboxymethyl cellulose (CMC) onto the fiber surface by using a procedure developed in this laboratory.37 Cellulosic fibers are contacted with a solution of CMC at high temperature and high electrolyte concentration, irreversibly cocrystallizing the CMC backbone onto the cellulosic fiber surfaces. It is not possible to remove the grafted CMC, even after extensive washing of the pulp into its sodium form with deionized water. The grafting becomes surface selective if the CMC has a sufficiently high molecular mass, as shown in Table 2. The decrease in surface selectivity at high grafting levels is presumably due to extensive surface swelling,47 allowing the CMC to penetrate the outermost surface layers of the fibers. Hence, there are two sets of fiber suspensions with varying bulk and surface charge densities. One set of fibers has a uniform distribution of charges, and one set has been carboxymethylated at the surface. Figure 4 shows how the adsorption, expressed in terms of the charge ratio, varies with the electrolyte concentration for bulk carboxymethylated pulp. The charge ratio is constant and independent of the degree of substitution (DS) for pulps with a low DS. The pulp with the highest degree of substitution exhibits an increase in the charge ratio, most likely dependent on the excessive swelling of the highly substituted pulp, which increases the accessibility of the poly-DADMAC to the cell wall of the fiber gel. Indeed, an examination of the degree of swelling in (47) Horvath, A. E. Licentiate Thesis, Department of Fibre and Polymer Technology, Royal Institute of Technology, Stockholm, Sweden, 2003.

828 Langmuir, Vol. 22, No. 2, 2006

HorVath et al.

Figure 5. Charge ratio of CMC grafted pulp at different electrolyte concentrations by using poly-DADMAC with molecular mass 9.2 × 105 Da.

the series with different DS revealed that the WRV of the pulps in this series increased from 133, 139, 175, to 218 g water/100 g pulp. A significant observation is that the increase in polyelectrolyte adsorption is shifted toward higher electrolyte concentrations with increasing DS. This is not surprising because the distance between the charges (d) decreases at higher DS. Hence, a higher electrolyte concentration is needed to fulfill the limiting condition of κ-1 > d. In the second pulp treatment, surface carboxymethylation increases the surface charge by grafting a high-molecular-mass CMC onto the fiber surface. The results from these adsorption experiments are shown in Figure 5. Opposite to the bulk carboxymethylation experiments, the surface charge ratio increases with the amount of grafted CMC, as expected. To obtain the same relative increase in charge ratio, higher electrolyte concentrations are needed to fulfill the limiting condition of κ-1 < d for fibers with a higher surface charge density. Critical Electrolyte Concentration. A critical electrolyte concentration (CEC) was defined in order to test the conditions when the adsorption stoichiometry deviates from 1:1. The CEC was defined as the electrolyte concentration where the apparent surface charge increased by 20%, which was arbitrarily chosen. The Debye-Hu¨ckel shielding length, κ-1, was then calculated from the corresponding CEC. The distance between the charges was calculated according to eq 1:

x 3

d)

VS ) σS × N A

x 3

WRVNa × 5% σS × N A

(1)

where VS is the “surface volume”, σS the surface charge density, NA Avogadro’s number, and WRVNa the water retention value for untreated pulp in its sodium form. To calculate the distance between the charges on the fiber surface, the concept of a surface volume (VS) was introduced. The wet surface area of cellulosic fibers is an ill-defined property because of the fuzzy nature of cellulosic surfaces, whereas a surface volume may be more strictly defined and calculated. Calculation for the bulk carboxymethylated fibers showed that approximately 5% (see Table 2) of the total charges are located in this “surface volume”. Assuming uniform charge distribution and swelling across the cell wall of the fiber, the surface volume can be calculated as 5% × WRVNa, because the WRV is a good measure of fiber swelling. It can be seen from Figure 6 that d ≈ κ-1 (the dashed line represent where d ) κ-1) for all the different pulp treatments. Hence, the assumption that adsorption stoichiometry prevails if κ-1 g d is a simple, but powerful concept for interpretation of the effects of electrolyte concentration on polyelectrolyte adsorption.

Figure 6. Surface charge distance (d) as a function of the diffuse electric double-layer thickness (κ-1), calculated from the critical electrolyte concentration (CEC) (see eq 1). ] denotes surface carboxymethylations and b denotes bulk carboxymethylations.

Indeed, it should be pointed out that this concept is similar to predictions based on lateral correlation effects in polyelectrolyte adsorption theory,18,19 where the concept of mesh size for adsorbed polyelectrolytes was introduced. The adsorbed polyelectrolyte was assumed to form a disordered surface pattern with chains crossing each other, characterized by a certain mesh size, ξs, corresponding to the average distance between the crossings. Charge overcompensation was predicted when ξs becomes of the order of κ-1. If the mesh size is associated with the charge distance, the present model is identical to the model of Netz and co-workers. Molecular Mass. The results presented so far have only considered the adsorption of high-molecular-mass (Mw ) 9.2 × 105) poly-DADMAC, which is excluded from the gel phase. It was, therefore, of interest to investigate how effective the exclusion was by using various molecular masses of polyDADMAC at different electrolyte concentrations. Figure 7a displays the charge ratio dependence on poly-DADMAC molecular mass and electrolyte concentration for the ungrafted kraft pulp (surface charge ) 1.6 µeq/g). The results for pulp that had been grafted with 14 mg/g CMC (surface charge ) 21 µeq/ g) are seen in Figure 7b. It is obvious that the charge ratio (adsorption) increases for lower molecular masses of polyDADMAC because of cell wall penetration. The charge ratio levels off to a finite value for higher-molecular-mass polyDADMAC, defining the surface charge at the limit of zero electrolyte concentration. This leveling-off value is less defined for a pulp with a high surface charge because surface charging will also open up the surface pore structure, therefore, increasing the polyelectrolyte accessibility. The effect of molecular mass on the charge ratio has also been investigated by Swerin and Wågberg.41 These authors found that the charge ratio increased from very small values up to between 0.4 to nearly 1.0 in deionized water for a molecular probe (poly3.6-ionene) with a molecular mass around 8.0 × 103. This is much higher than the values displayed in Figure 7a. Apart from the difference that another molecular probe was used in these experiments, the most likely explanation is that the pulp used in their investigation was unbleached kraft pulp with a much higher charge density. A higher bulk charge density will most likely make cell wall penetration easier, explaining the apparent discrepancy between the two data sets. Comparison Between the Surface Nitrogen Content as Determined by ESCA and Polyelectrolyte Titrations. It was also of interest to compare the surface nitrogen content of fibers with adsorbed poly-DADMAC with polyelectrolyte titrations. For this purpose, the bulk carboxymethylated pulp with the highest charge density was chosen (DS ) 0.076, total charge 506 µeq/g).

Polyelectrolyte Titration of Cellulosic Fibers

Langmuir, Vol. 22, No. 2, 2006 829

Figure 7. Charge ratio for fractionated poly-DADMAC as a function of electrolyte concentration during adsorption on untreated (a) and CMC-grafted pulp (b).

Figure 8. Adsorbed amount of polyelectrolyte charges onto a carboxymethylated bleached kraft pulp as a function of polyDADMAC molecular mass as measured by polyelectrolyte titration.

The adsorbed amount of polyelectrolyte as a function of molecular mass in 10-5 M NaHCO3 is given in Figure 8. As expected, the adsorbed amount of poly-DADMAC decreases as the molecular mass, i.e., the size of the polyelectrolyte, increases, and its accessibility to the charges in the cell wall decreases. The adsorbed amount of poly-DADMAC is about the same for polyelectrolytes with Mw 1.4 × 105 and 9.2 × 105. Hence, the accessibility of those polyelectrolytes to the fiber surface charges was about the same, suggesting that only surface regions could be reached when the adsorption was performed by using a sufficiently high Mw poly-DADMAC (Mw > 1.4 × 105). It is well established that kraft pulp fibers contain most of their pores in the range up to 5-10 nm,48,49 while the radius of gyration of the poly-DADMAC with a molecular mass of 2 × 105 has been reported to be about 33 nm.50 On the basis of these results, poly-DADMAC with a molecular mass of 9.2 × 105 was chosen as a polyelectrolyte probe for surface charge analysis by using polyelectrolyte titration. Pore sizes obtained from solute exclusion studies may not be entirely comparable to pore sizes excluding cationic polyelectrolytes, where an electrostatic driving force promotes cell wall penetration. Alince and van de Ven51 found, for instance, that sphere-like polyethyleneimine of 26 nm was fully accessible to pulps similar to those used in this study. The (48) Scallan, A. M. Fibre-Water Interactions in Paper-Making, Transactions of the 6th Fundamental Research Symposium held at Oxford; TAPPI TS-CPPA BPBIF, Ed.; Beccles and Colchester: London, 1977; p 2-1. (49) Bristow, J. A.; Kolseth, P. Paper Structure and Properties; International Fibre Science and Technology Series/8; Marcell Dekker Inc., NY, 1986; Chapter 4. (50) Burkhardt, C. W.; McCarthy, K. J.; Parazak, D. P. J. Polym. Sci. 1987, 25, 209. (51) Alince, B.; van de Ven, T. G. M. Fundamentals of Papermaking, Transactions of the 11th Fundamental Research Symposium held at Cambridge; Cambridge, U.K., September 1997.

data in Figure 8 gives, however, confidence that the highmolecular-mass poly-DADMAC is excluded from the cell wall. The results obtained by polyelectrolyte titration were also compared with the surface nitrogen content obtained by the ESCA technique. Different amounts of low- (7.0 × 103) or highmolecular-mass (9.2 × 105) poly-DADMAC were added to a carboxymethylated pulp with a total charge of 506 µeq/g. After adsorption equilibrium, the fibers were separated from the solution by filtration. The remaining amount of poly-DADMAC in solution (nonadsorbed) was determined by polyelectrolyte titration. Handsheets were made from the fiber fraction and subjected to ESCA analysis. Figure 9 shows the adsorption of both low- and high-molecular-mass poly-DADMAC as a function of the equilibrium concentration when analyzed by polyelectrolyte titration and ESCA. The adsorption of low-molecular-mass poly-DADMAC, analyzed by polyelectrolyte titration, was about 10 times higher than that of the high-molecular-mass poly-DADMAC (Figure 9a). This is expected because of the higher accessibility of the low-molecular-mass poly-DADMAC to the internal void spaces of the fiber, whereas the high Mw poly-DADMAC can only be adsorbed onto the outer fiber surface. The surface content of nitrogen was then determined by ESCA and is given in Figure 9b. Figure 9b shows that the surface content of nitrogen is somewhat higher for the high-molecular-mass poly-DADMAC, but may be considered to be on a similar level. If the charge stoichiometry would have been perfect, the two nitrogen levels should have been equal. This indicates that the high-molecularmass poly-DADMAC is adsorbed in a somewhat more extended conformation. The conformation of a highly charged polyelectrolyte is generally assumed to be flat at low additions, and the thickness of a monolayer of adsorbed polyelectrolyte has been shown to be independent of the molecular mass.12 It has also been shown by Wågberg et al.32 that the stoichiometry of adsorption of both low and high Mw cationic polyelectrolytes on carboxymethylated pulps is high (above 90%). Because a reasonable agreement is obtained in the ESCA analysis between the two polyelectrolytes, the thickness of the adsorbed layer is about the same for both polyelectrolytes. In essence, these results give further credence to the hypothesis that the surface charge of cellulose can be determined by polyelectrolyte titration by using high-molecularmass poly-DADMAC. Effect of Fiber Charge Profile on Poly-DADMAC Adsorption. In another set of experiments, different bulk and surface carboxymethylated fibers were used to simulate fibers with different charge profiles. Pulps with various charge profiles (see Table 2) were chosen for ESCA analysis. In these experiments,

830 Langmuir, Vol. 22, No. 2, 2006

HorVath et al.

Figure 9. Adsorption isotherms for a bulk carboxymethylated pulp with a total charge of 506 µeq/g. Two different molecular masses (Mw) of poly-DADMAC were used: 7.0 × 103 and 9.2 × 105 Da.

(∼weeks). Hence, some caution should be exercised when setting the time for adsorption equilibrium.

Conclusions

Figure 10. Relationship between ESCA analysis and polyelectrolyte titration for determining the surface charge of surface and bulk carboxymethylated pulps with different charge profiles. The Mw of the poly-DADMAC was 9.2 × 105.

the added amount of poly-DADMAC with a molecular mass of 9.2 × 105 corresponded to an equilibrium concentration of 8.50 ( 3 mg/L in the adsorption isotherms. As can be seen from Figure 10, the adsorbed amount of high Mw poly-DADMAC analyzed by ESCA (nitrogen content) and polyelectrolyte titration are in reasonable agreement despite the wide difference in charge profiles of the pulps used. Hence, the surface charge results obtained by polyelectrolyte titration correlate well with the ESCA analyses. Altogether, this gives strong credibility to the idea that the surface charge content of cellulosic fibers can be analyzed by means of adsorption of high Mw poly-DADMAC. Recently, van de Ven52 suggested that high-molecular-mass cationic polyelectrolytes might penetrate into the cell wall of cellulosic fibers through reptation. It was shown in Figure 9 that there is no correlation between the results from ESCA and polyelectrolyte titration if the molecular mass of the polyelectrolyte is low, i.e., if the cationic polyelectrolyte is allowed to penetrate in the cell wall. Furthermore, there would not have been a good correlation between the two methods for the modified fibers if the high-molecular-mass poly-DADMAC had adsorbed to a large extent in the cell wall (Figure 10). However, it should be made clear that the equilibrium time in these experiments was 30 min. There are indications in our laboratory that reptation may occur at longer equilibrium times (52) van de Ven, T. G. M. Nord. Pulp Pap. Res. J. 2000, 15, 494.

This study has investigated the influence of electrolyte concentration on the adsorption of poly-DADMAC onto cellulosic fibers having different charge profiles. It was found that a deviation from stoichiometry occurred already at low electrolyte concentrations, where the distance between the charges becomes greater or equal to the thickness of the diffuse electric double layer. Adsorption increases until the screening of charges by the electrolyte becomes predominant, where after the adsorption decreases to zero at high electrolyte concentrations. The adsorption is also dependent on the fiber charge density. As the charge density increases, the distance between the charges decreases. This requires a higher electrolyte concentration before the distance between the charges exceeds the thickness of the double layer, which causes a deviation from stoichiometry. These results are in agreement with the theoretical predictions by Netz and coworkers18,19 regarding screening-enhanced adsorption of polyelectrolytes. A low-molecular-mass polyelectrolyte is more sensitive to changes in electrolyte concentration because the polyelectrolyte will reach more charges in the fiber pores with increasing electrolyte concentration, whereas the size of the highmolecular-mass polyelectrolyte will not decrease enough to access the pores of the cellulosic fibers. In this study, the surface-sensitive ESCA technique was also used as an independent calibration procedure for the polyelectrolyte titration method to determine the fiber surface charge. Cationic polyelectrolytes (poly-DADMAC) with different molecular mass and carboxymethylated fibers with different charge profiles were used for this purpose. Good agreement of the results was found between the two methods. It can be concluded that the polyelectrolyte titration method is a powerful technique for the determination of fiber surface charge. The most critical requirements appear to be that the adsorbed polyelectrolyte has a sufficiently high molecular mass and that the titration is carried out at low ionic strength. Furthermore, adequate analytical procedures are needed to obtain adsorption isotherms from which the surface charge density of fibers can be determined. LA052217I