Umeå University's logo

The adsorption qualities of GaO(4)Al(12)(OH)(24)(H(2)O)(12)(7+), a polycation with ε-Keggin structure, and its stability in contact with anionic cellulosic materials, was investigated under different concentration and ionic strength conditions. The cellulosic materials employed were two different fully bleached fibre materials, carboxyl methyl cellulose (CMC), and a spin-coated cellulose model surface. As analytical techniques, pH-measurements, potentiometric titrations, ICP-OES, QCM-D, equilibrium calculations and Extended X-ray Absorption Fine Structure (EXAFS) were used. The adsorption is substantial and the addition of GaO(4)Al(12)(OH)(24)(H(2)O)(12)(7+) to a fibre suspension results in a rapid decrease in pH, followed by a small and slow increase in pH. This behaviour can be explained as due to a rapid and strong (log β>2) equilibrium adsorption of intact GaO(4)Al(12)(OH)(24)(H(2)O)(12)(7+) ions, followed by a slow, and minor, 3-8%, decomposition into different monomers. Alternative layer by layer adsorption of this ion, and CMC, on a spin-coated cellulose model surface constitutes further evidence for the strong interactions between the anionic cellulose materials and GaO(4)Al(12)(OH)(24)(H(2)O)(12)(7+). It is shown that the adsorption observed could not be described as due to an unspecific Donnan adsorption behaviour, neither of GaO(4)Al(12)(OH)(24)(H(2)O)(12)(7+) nor Ga and Al monomers, and specific surface complex formation is therefore discussed and applied. The (≡COO)(7)GaO(4)Al(12)(OH)(24)(H(2)O)(12) species found to explain the pH- and metal adsorption data should be considered strictly as a stoichiometric entity.

A compilation of the applied experimental conditions when studying metal ion adsorption onto kraft fibres, and the resulting conclusion, revealed that the ionic strength conditions used during the experiments were an important dividing factor. At low ionic strengths, the conclusion has regularly been that the Donnan ion-exchange model could correctly predict the adsorption while, at higher ionic strengths, it has often been concluded that the formation of specific metal-ion fibre complexes must be assumed. To study this apparent influence from the presence of monovalent sodium ions, Cu K-edge EXAFS spectra of Cu²⁺ ions adsorbed to kraft fibres were collected in media of "0" to 100 mM NaCl. Combined with previous data, these measurements confirmed that at very low ionic strength, the importance of specific interactions between the chemically modified cellulose fibres and the Cu(II) ions significantly decreased. For a detailed description of the adsorption phenomenon, both types of interactions must be considered simultaneously. For most technical and engineering applications, however, the Donnan model can be used at low ionic strength conditions, i.e. I ≤ 10 mM. At higher ionic strengths, though, the inclusion of specific complexes in the model is necessary for correctly describing the adsorption of di- and trivalent cations with strong complex forming properties.

Aiming for a better understanding of the interactions between water suspended cellulose fibres and metal ions, this study was focused on characterising the interactions between Ca2+, Cu2+ and two different fibre materials—a fully bleached softwood Kraft pulp, and a chemically modified fully bleached softwood Kraft fibre material. The study was conducted as a function of pH (2–7), and both in the absence and presence of an excess of Na+ ions, 0–100 mM Na(Cl). For both fibre materials, adsorption data collected in the absence of Na+ were fully explained by the unspecific Donnan ion-exchange model. However, in an excess of Na(Cl), data clearly indicated that higher amounts of divalent metal ions adsorbed, than predicted by the Donnan model. Therefore, to model these data, specific metal ion–fibre surface complexes were assumed to form, in addition to the Donnan ion-exchange. A neutral surface species involving two surface carboxylate groups and one metal ion was, for both metal ions, found to yield a good description of data at all ionic strengths. In the case of Cu2+, the existence of this complex was corroborated by Cu K-edge EXAFS data, suggesting that copper ions interacts directly with carboxyl groups present int the fibres. EXAFS data also indicate that one Cu2+ interacts with two carboxyls.

Graphical abstract

In combination with unspecific Donnan interactions, specific interactions between divalent metal ions (Ca2+ and Cu2+) and cellulose fibre carboxylic groups were evidenced from both macroscopic adsorption data and spectroscopic measurements.

Pulp materials ranging in charge from zero to 280 mmol/g were synthesised and characterised with respect to ion exchange properties. Both the Constant Capacitance and the Basic Stern Models were found applicable to simulate the acid/base properties. For all materials, the Donnan distribution model gave a good description for the metal binding properties.

The washing of dissolved lignin residues from fibers is significantly favored at low ionic strengths. This is shown to affect the standard procedures for COD analysis of pulp, and evaluations of washing efficiency. In the paper, strategies to take advantage of the effect, in storage towers and in washing equipment, are discussed.

To evaluate the capacity of different thermodynamic models to mimic the build-up of charge on fully bleached softwood fibres as a function of -log[H+] and ionic strength (0.01-0.6 M NaCl), the model of an Ideal Monoprotic Acid, the Constant Capacitance Model (CCM), the Basic Stern Model (BSM), and the Diffuse Layer Model (DLM) were compared. The Ideal Monoprotic Acid, IMA, model consistently yielded a too steep increase in the build up of charge with -log[H+] and thereby indicated the need for applying a surface potential model. The "best" of these, with respect to describing the experimental data, was found to be the Constant Capacitance Model, which utilises an individual equilibrium constant and an individual capacitance value at each ionic strength modelled. However, the Basic Stern Model, which uses only two variables over the entire ionic strength interval, was found to describe data almost as well. Therefore, in modelling applications where a variation in the salt concentration is significant, the Basic Stern Model must be judged as being superior to the Constant Capacitance Model. As for the IMA model, the Diffuse Layer Model, finally, was found to yield a too steep increase in the fibre charge with rising -log[H]. Nevertheless, it was found to model the variation of the apparent acidity with ionic strength, relatively well.

Aqueous gallium(III) citrate complexes have been studied in the 10(-2) M concentration range with extended X-ray absorption fine structure (EXAFS) and FTIR techniques. From EXAFS data, one mononuclear and one oligomeric species were identified at different Ga(III) to citrate ratios. The first shell of the mononuclear complex was found to be distorted, with average Ga-O bond lengths of 1.95 and 2.06 Angstrom, in agreement with the solid-state structure of Ga(Cit)(2)(3-) (Cit = citrate). Also the oligomeric species was found to have a distorted first shell, with average Ga-O bond lengths of 1.95 and 2.04 Angstrom. This complex was found to contain two Ga-Ga distances at 3.03 and 3.56 Angstrom, typical for edge and corner sharing GaO6 octahedra, respectively. The gallium(III) and aluminum(III) citrate systems were compared by means of FTIR, and were found to be analogous. The IR results suggest that the bond lengths derived from EXAFS for the 1:2 gallium(III) citrate complex also provide a good estimate of the corresponding distances in the mononuclear 1: 1 complex. Direct coordination of citrate to the metal ions in the oligomeric gallium(III) citrate complex was indicated from both EXAFS and IR results, and this complex is stoichiometrically analogous to the Al-3(H(-1)Cit)(3)(OH)(H2O)(4-) complex, which has been structurally determined. However, while the formation of the aluminum trimer has been shown to be slow, the gallium trimer was significantly more labile with a rate of formation indicated to be in the order of seconds or faster.