The behavior of cellulose molecules in aqueous environments

Molecular motions of cellulose chains in aqueous environments were investigated by comparison with those in non-aqueous environments using molecular simulation techniques. The cellulose chains under non-aqueous conditions approached each other closely and then made tight aggregates that were formed by direct hydrogen bonding. Those in aqueous environments, such as in a bio-system, were separated from each other by water molecules and did not have direct hydrogen bonding between the cellulose chain molecules. Folded-chain structures were not found in either aqueous or non-aqueous environments that were somewhat crowded. In the aqueous system, the water molecules around the cellulose chains restricted their molecular motions and interrupted formation of direct, interchain hydrogen bonds. In the non-aqueous system, the cellulose chains approached each other closely and then made a tight cluster before the chain molecules could wind and bend. It was concluded that a very dilute solution of cellulose molecules in appropriate solvents is necessary to create folded-chain or random-coiled structures. We also confirmed that the driving force for making tight clusters of cellulose molecules in highly concentrated solutions is the energy of the hydrogen bonding created directly between the hydroxyl groups of the cellulose chains. These results strongly suggest that hydrogen bonding plays a very important role in the characteristics of cellulose molecules.     

Effects of temperature and molecular weight on dissolution of cellulose in NaOH/urea aqueous solution

Dissolution of cellulose having different viscosity-average molecular weight (M _η ) in 7 wt%NaOH/12 wt%urea aqueous solution at temperature from 60 to −12.6°C was investigated with optical microscope, viscosity measurements and wide X-ray diffraction (WXRD). The solubility (S_a) of cellulose in NaOH/urea aqueous solution strongly depended on the temperature, and molecular weight. Their S_a values increased with a decrease in temperature, and cellulose having M _η below 10.0 × 10⁴ could be dissolved completely in NaOH/urea aqueous solution pre-cooled to −12.6°C. The activation energy of dissolution (E_a,s) of the cellulose dissolution was a negative value, suggesting that the cellulose solution state had lower enthalpy than the solid cellulose. The cellulose concentration in this system increased with a decrease of M _η to achieve about 8 wt% for M _η of 3.1 × 10⁴. Moreover, cellulose having 12.7 × 10⁴ could be dissolved completely in the solvent pre-cooled to −12.6°C as its crystallinity (χ _c) decreased from 0.62 to 0.53. We could improve the solubility of cellulose in NaOH/urea aqueous system by changing M _η , χ _c and temperature. In addition, the zero-shear viscosity (η ₀ ) at 0°C for the 4 wt% cellulose solution increased rapidly with an increase of M _η , as a result of the enhancement of the aggregation and entanglement for the relatively long chains.     

Direct observation of cellulose dissolution in subcritical and supercritical water over a wide range of water densities (550–1000 kg/m<Superscript>3</Superscript>)

Direct observations of the heating of microcrystalline cellulose (230 DP) in water at temperatures up to 410 °C and at pressures up to 700 MPa were made with a batch-type microreactor. Cellulose particles were found to dissolve with water over temperatures ranging from 315 to 355 °C at high pressures. Dissolution temperatures depended on water density and decreased from about 350 °C at a water density of 560 kg/m³ to a minimum of around 320 °C at a water density of 850 kg/m³. At densities greater than 850 kg/m³, the dissolution temperatures increased and reached a value of about 347 °C at 980 kg/m³. The cellulose dissolution temperatures were independent of heating rates for values ranging from 10 to 17 °C/s. The low dependence of dissolution temperatures on the heating rates is strong evidence for simultaneous dissolution and reaction of the cellulose. Different phenomena occurred depending on water density. At low densities, particles turned transparent and seemed to dissolve into the aqueous phase from the surface. From 670 to 850 kg/m³, the cellulose particles visibly swelled just before completely collapsing and dissolving into the aqueous phase. The swelling probably increased water accessibility and particle surface area and thus lead to the lower dissolution temperatures observed. From 850 to 1000 kg/m³, the particles required longer times to dissolve and many fine brown-like particles were generated as the particles dissolved. FT-IR spectra of the residues were analyzed. Residues formed from heating cellulose at high densities still retained some cellulose character whereas those as low densities had little cellulose character, especially in the O–H stretching vibration region.     

SIZE EXCLUSION CHROMATOGRAPHIC COLUMN PACKED WITH REGENERATED CELLULOSE GELS*

Microporous regenerated cellulose gel particles were prepared by mixing cellulose cuoxam with silk fibroin aspore former, and the mean pore size and pore volume of the pallicles were 525 nm and 7.27 mL g~(-1), respectively. Apreparative size-exclusion chromatography (SEC) column (550 mm×20 mm) packed with the cellulose gel particles wasused for the fractionation of two polysaccharides Dextran 07 (M_w = 7.14×10~4, d= 1.7) and Dextran 50(M_w = 50.5×10~4,d = 3.8) in water phase. The fractionation range of the stationary phase covered M_w from 3×10~3 to 1.1×10~6. The dailythroughput was 2.9 g for Dextran 07 (D07) and 4.3 g for Dextran 50 (D50) with a flow-rate of 1.5 mL min~(-1). The fractionsobtained by using the SEC were analyzed by an analytical SEC combined with laser light scattering (LLS), and thepolydispersity indices of fractions for Dextran 07 and Dextran 50 were determined to be 1.34-1.57 and 1.53-3.36,respectively. The preparative SEC is a simple, rapid, and suitable means not only for the fractionation of polysaccharides inwater but also for other polymers in organic solvents.     

Dissolving of cellulose in PEG/NaOH aqueous solution

Here, a new solvent system for cellulose is reported. The solvent is a mixed aqueous solution of 1.0 wt.% poly(ethylene glycol) (PEG) and 9.0 wt.% of NaOH. Cellulose powder was added into the mixture at room temperature at first, and freezing it at −15 °C for 12 h following a thaw of the mixture at room temperature under strong stirring. There formed a clean solution of cellulose, and the optical microscopy was used to record the dissolving process. ¹³C-NMR, FT-IR, XRD, and intrinsic viscosity measurements revealed that there forms a homogeneous solution of cellulose in the new solvent system. The maximum solubility of cellulose with average molecular weight of 1.32 × 10⁵ g mol⁻¹ in the solvent system is 13 wt.%. The cellulose solution in the new solvent system is stable, even for 30 days storage at room temperature.     

Behavior of cellulose in NaOH/Urea aqueous solution characterized by light scattering and viscometry

Cellulose was dissolved in 6 wt % NaOH/4 wt % urea aqueous solution, which was proven by a ¹³C NMR spectrum to be a direct solvent of cellulose rather than a derivative aqueous solution system. Dilute solution behavior of cellulose in a NaOH/urea aqueous solution system was examined by laser light scattering and viscometry. The Mark–Houwink equation for cellulose in 6 wt % NaOH/4 wt % urea aqueous solution at 25 °C was [η] = 2.45 × 10^?2 weight‐average molecular weight (M_w)^0.815 (mL g^?1) in the M_w region from 3.2 × 10⁴ to 12.9 × 10⁴. The persistence length (q), molar mass per unit contour length (M_L), and characteristic ratio (C_∞) of cellulose in the dilute solution were 6.0 nm, 350 nm^?1, and 20.9, respectively, which agreed with the Yamakawa–Fujii theory of the wormlike chain. The results indicated that the cellulose molecules exist as semiflexible chains in the aqueous solution and were more extended than in cadoxen. This work provided a novel, simple, and nonpollution solvent system that can be used to investigate the dilute solution properties and molecular weight of cellulose. © 2003 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 42: 347–353, 2004     

Crystalline transformation of native cellulose from cellulose I to cellulose ID polymorph by a ball-milling method with a specific amount of water

We have demonstrated for the first time that a mechano-chemical treatment of native cellulose with a specific amount of water (30 wt%) present ID the cellulose solid state caused the crystalline transformation from cellulose I into cellulose ID polymorph. X-ray diffractometry was used to show that the extent of transformation into cellulose ID increased with milling time. This specific phenomenon can be explained by considering the chain mobility ID the cellulose–water system, because ₁ ^1H measurement shows that cellulose molecules are most mobile when the water content ID around 30 wt%, and thus are favorable for molecular rearrangement under external forces.     

Accessibility and size of <Emphasis Type="Italic">Valonia</Emphasis> cellulose microfibril studied by combined deuteration/rehydrogenation and FTIR technique

We report an FTIR method to measure the accessibility and the size of cellulose microfibrils from the cell wall of Valonia ventricosa. This method is similar to the conventional deuteration technique for measuring the accessibility of cellulosic materials; however, the difference in our method is that the hydroxyl groups O2H, O3H, and O6H in the crystalline region were initially completely deuterated. The sample was then rehydrogenated by soaking in water at 25 °C, so that the OD groups on the surface were rehydrogenated. The ratio of OH to OD absorbance was used to calculate the number of surface vs. core cellulose chains in a microfibril. The obtained experimental ratio of 0.934 was consistent with the value calculated for a previously published 33 × 38 chain Valonia model (Sugiyama et al. 1984). The rehydrogenation process was further investigated by immersing the sample in water at elevated temperatures. At temperatures above 120 °C, rehydrogenation was more efficient, and the efficiency plots vs. rehydrogenation temperature showed two inflection. These points may correspond to the temperature where the cleavage of inter-chain hydrogen bonds and/or crystalline-phase transition would have been occurred.     

Dilute solution properties of cellulose in LiOH/urea aqueous system

Cellulose was dissolved rapidly in 4.6 wt % LiOH/15 wt % urea aqueous solution and precooled to –10 °C to create a colorless transparent solution. ¹³C‐NMR spectrum proved that it is a direct solvent for cellulose rather than a derivative aqueous solution system. The result from transmission electron microscope showed a good dispersion of the cellulose molecules in the dilute solution at molecular level. Weight‐average molecular weight (M_w), root mean square radius of gyration (〈s²〉_z^1/2), and intrinsic viscosity ([η]) of cellulose in LiOH/urea aqueous solution were examined with laser light scattering and viscometry. The Mark–Houwink equation for cellulose in 4.6 wt % LiOH/15 wt % urea aqueous solution was established to be [η] = 3.72 × 10^?2 M in the M_w region from 2.7 × 10⁴ to 4.12 × 10⁵. The persistence length (q), molar mass per unit contour length (M_L), and characteristic ratio (C_∞) of cellulose in the dilute solution were given as 6.1 nm, 358 nm^?1, and 20.8, respectively. The experimental data of the molecular parameters of cellulose agreed with the Yamakawa–Fujii theory of the worm‐like chain, indicating that the LiOH/urea aqueous solution was a desirable solvent system of cellulose. The results revealed that the cellulose exists as semistiff‐chains in the LiOH/urea aqueous solution. The cellulose solution was stable during measurement and storage stage. This work provided a new colorless, easy‐to‐prepare, and nontoxic solvent system that can be used with facilities to investigate the chain conformation and molecular weight of cellulose. © 2006 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 44: 3093–3101, 2006     

The role of salt on cellulose dissolution in ethylene diamine/salt solvent systems

Ethylene diamine (EDA)/salt solvent systems can dissolve cellulose without any pretreatment. A comparison of the electrical conductivity of different salts in EDA was made at 25 °C, and conductivity decreased in the order of KSCN>KI>NaSCN at the same molar concentration. Among the salts tested, potassium thiocyanate (KSCN) was capable of dissolving both high molecular weight (DP>1000) and low molecular weight (DP = 210) cellulose, and this was confirmed by polarized light microscopy. ³⁹K and ¹⁴N NMR experiments were conducted at 70 °C as a function of cellobiose concentration with EDA/KSCN as the solvent. The results showed that the K⁺ ion interacts with cellobiose more than the SCN⁻ ion does. Recovered cellulose was studied by infrared spectroscopy (FTIR) and wide angle X-ray diffraction (WAXD). Changes in the FTIR absorption bands at 1,430 and 1,317 cm⁻¹ were associated with a change in the conformation of the C-6CH₂OH group. The changes in positions and/or intensities of absorption bands at 2,900, 1,163, and 8,97cm⁻¹ were related to the breaking of hydrogen bonds in cellulose. X-ray diffraction studies revealed that cellulose, recovered by precipitating cellulose solutions with water, underwent a polymorphic transformation from cellulose I to cellulose II.     

A Comparative Simulation: Difference between Chemical Grafting and Physical Doping of Cellulose by Using Polysilsesquioxane

In order to study the effect of different modification methods on polysilsesquioxane (POSS) modified cellulose, a molecular dynamics method was used to establish a pure cellulose model and a series of modified models modified by polysilsesquioxane in different ways. And their thermodynamic properties were calculated. The results showed that the performance of cellulose models was better than that of unmodified model, and the modified effect was the best when two cellulose chains were grafted onto polysilsesquioxane by chemical bond (M2 model). Compared with pure cellulose model, the cohesive energy density and solubility parameters of M2 model are increased by 9%, and the values of tensile modulus, bulk modulus, shear modulus and Cauchy pressure increased by 38.6%, 29.5%, 41.1% and 29.5%, respectively. In addition, the free volume fraction and mean square displacement of each model were calculated and analyzed in this work. Compared with the pure cellulose model, the molecular chain entanglement of cellulose was increased due to the existence of the chemical bonds in the M2 model, which made the cellulose molecular chains occupy more free volume, so that the system had a smaller free volume fraction, inhibited the chain movement of cellulose chains, and thus improved the thermal stability of cellulose.     

Flow birefringence of short-chain molecules: Cellulose tricarbanilates in benzophenone

Measurements of flow birefringence of cellulose tricarbanilates were carried out on nine fractions (0.27 × 10⁵ < M ≤ 12 × 10⁵) in a temperature range of 55–110°C, with benzophenone as a matching solvent (dn/dc = 0). The ratio of Maxwell constant to intrinsic viscosity, which has been found to be independent of molecular weight for the limiting case of Gaussian molecules, is successfully interpreted as a function of molecular weight in terms of the recent theory of Gotlib and Svetlov (based on the wormlike chain model of Kratky and Porod). From the measurements at 55°C a number of 36.6 monomer units per random link is deduced. This is in accord with results of small-angle x-ray scattering. For the extinction angle curves a clear transition is observed from rodlike to statistical molecules when the molecular weight is increased. At high molecular weights the master curves obtained for anionic polystyrenes and cellulose tricarbanilates coincide. Implications of this observation on the kinetic stiffness of the cellulose tricarbanilate chain are discussed. The intrinsic viscosity-molecular weight relationship is considered. From a comparison with the results of the theory of Eizner and Ptitsyn it is concluded that the cellulose tricarbanilate chain must be highly solvated in benzophenone.     

Hydration mechanism of polysaccharides: A comparative study

Water sorption was studied at 20 °C on films composed of different natural polymers. Three polysaccharides were investigated: chitosan, cellulose, and alginate. The major differences between these polymers, from a structural point of view, lay in the substitution of an OH group by an NH₂ function for chitosan and by an ionic COO^?Na⁺ group for alginate. An analysis of the experimental water sorption isotherms, expressed as the number of water molecules sorbed per repeating unit in the amorphous phase, associated with an analysis of the enthalpy profile related to the water sorption allowed us to propose a water sorption mechanism in two steps for all the polymers: water sorption on polymer‐specific sites in the first step and water clustering around the first sorbed water molecules in the second step. It was determined that two water molecules interacted with the polymer chains for cellulose and chitosan, whereas four water molecules were bonded to alginate chains. The specific sorption sites were identified as OH groups for cellulose, OH and NH₂ groups for chitosan, and ionic and OH groups for alginate. A systematic reduction of the half‐sorption time was observed in the activity range corresponding to this first sorption step, and it was explained by a water plasticization effect. On the other hand, an increase in the half‐sorption time was observed in the second sorption step, at a high activity (>0.8), for chitosan and alginate. A modelization associating the Guggenheim–Anderson–de Boer model and the clustering theory, applied to our systems, allowed us to relate the occurrence of this last phenomenon to the formation of water clusters containing more than two water molecules. © 2004 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 43: 48–58, 2005     

Ferric oxide-catalyzed polymerization of methyl methacrylate

The polymerization of methyl methacrylate was carried out in water at various concentrations of sodium bisulfite, ferric oxide, and methyl methacrylate at 30, 40, and 50°C. The effect of ferric oxide on the rate of polymerization was studied at 50°C. Rates of polymerization increased in the presence of ferric oxide. For example, the rate of polymerization increased from 3.4 × 10^?5 mole/l.-sec to 11.8 × 10^?5 mole/l.-sec when the ferric oxide concentration was varied from 0 to 15 g/l. water. The molecular weight of the polymer decreased from an average of 1.4 × 10⁶ in the absence of ferric oxide to 2.8 × 10⁵ when the ferric oxide was present. The variation of molecular weight of the polymers with temperature and conversion was studied. At a fixed conversion of 80%, the average molecular weight decreased from 3.4 × 10⁵ at 30°C to 2.2 × 10⁵ at 50°C. The average molecular weight was also found to increase with increasing monomer and initiator concentrations. It increased from 8.1 × 10⁴ to 5.3 × 10⁵ and from 3.4 × 10⁵ to 8.9 × 10⁵ as the initiator and monomer concentrations increased from 0.01 to 0.05 mole/l. and from 0.235 to 0.705 mole/l., respectively. The apparent energy of activation for the polymerization was found to be 15.6 and 9.7 kcal/mole in absence and in presence of ferric oxide, respectively.     

Synchrotron X-ray structures of cellulose I<Subscript>β</Subscript> and regenerated cellulose II at ambient temperature and 100 K

Synchrotron X-ray data have been collected to 1.4 Å resolution at the NE-CAT beam-line at the Advanced Photon Source from fibers of cellulose I_β and regenerated cellulose II (Fortisan) at ambient temperature and at 100 K in order to understand the effects of low temperature on cellulose more thoroughly. Crystal structures have been determined at each temperature. The unit cell of regenerated cellulose II contracted, with decreasing temperature, by 0.25%, 0.22% and 0.1% along the a, b, and c axes, respectively, whereas that of cellulose I_β contracted only in the direction of the a axis, by 0.9%. The value of 4.6×10^?5 K^?1 for the thermal expansion coefficient of cellulose I_β in the a axis direction can be explained by simple harmonic molecular oscillations and the lack of hydrogen-bonding in this direction. The molecular conformations of each allomorph are essential unchanged by cooling to 100 K. The room temperature crystal structure of regenerated cellulose II is essentially identical to the crystal structure of mercerized cellulose II.     

Molecular conformations at the cellulose–water interface

¹³C-NMR chemical shifts were measured for C-4 and C-6 in a collection of eight crystalline glucoses and glucosides. The influence of the hydroxymethyl conformation was greater at C-4 than at C-6, with mean chemical shifts for gauche–trans molecules displaced 3.1 ppm (C-4) and 2.5 ppm (C-6) relative to gauche–gauche molecules. This information was used to interpret ¹³C-NMR spectra of crystalline celluloses. Chemical shifts for C-4 in the crystallite cores of celluloses I and II differed by just 0.2 ppm, but the corresponding chemical shifts for well-ordered crystallite surfaces differed by 3.0 ppm. The separation between crystallite-surface signals was attributed to different hydroxymethyl conformations at the cellulose–water interface, i.e., gauche–gauche and gauche–trans on crystallites of cellulose I and cellulose II, respectively. A broad C-4 signal in the spectrum of cellulose II indicated gauche–gauche conformations in disordered cellulose. Chemical shifts for C-6 were consistent with these conformations.     

Graft Copolymerization on to Cellulose Using Binary Mixture of Monomers

The graft copolymerization of acrylonitrile (AN) and ethyl acrylate (EA) comonomers onto cellulose has been carried out using ceric ammonium nitrate (CAN) as an initiator in the presence of nitric acid at 35±0.1°C. The addition of ethyl acrylate as comonomer has shown a significant effect on overall and individual graft copolymerization of acrylonitrile on cellulose. The graft yield (%G_Y) and other grafting parameters viz. true grafting (%G_T), graft conversion (%C_G), cellulose number (Ng) and frequency of grafting (G_F) were evaluated on varying the concentration of comonomers from 6.0–30.0×10^?1 mol dm^?3 and ceric (IV) ions concentration from 2.5–25×10^?3 mol dm^?3 at constant feed composition (f_AN 0.6) and constant concentration of nitric acid (7.5×10^?2 mol dm^?3) in the reaction mixture. The graft yield (%G_Y) and other grafting parameters were optimal at 15×10^?1 mol dm^?3 concentration of comonomers and at 10×10^?3 mol dm^?3 concentration of ceric ammonium nitrate. The graft yield (%G_Y) and composition of grafted chains (F_AN) was optimal at a feed composition (f_AN) of 0.6. The energy of activation (Ea) for graft copolymerization has been found to be 16 kJ mol^?1. The molecular weight (Mw) and molecular weight distribution (Mw/Mn) of grafted chains was determined by GPC and found to be optimum at 15×10^?1 mol dm^?3 concentration of comonomer in the reaction mixture. The composition of grafted chains (F_AN) determined by IR method was used to calculate the reactivity ratios of monomers, which has been found to be 0.62 (r₁) and 1.52 (r₂), respectively for acrylonitrile (AN) and ethyl acrylate (EA) monomers used for graft copolymerization. The energy of activation for decomposition of cellulose and grafted cellulose was determining by using different models based on constant and different rate (β) of heating. Considering experimental observations, the reaction steps for graft copolymerization were proposed.     

Molecular dynamics simulation of cellulose-coated oil-in-water emulsions

The behaviors of cellulose chains and cellulose mini-crystal in oil-in-water emulsions were studied by molecular dynamics simulations to investigate the coating states and the structural features of cellulose in these emulsions. In oil-in-water emulsion, dispersed cellulose chains gradually assemble during the progress of the simulation, eventually surrounding the octane droplet. In case of a cellulose mini-crystal, the cellulose chain at the corner of the crystal first contacts with the octane droplet through its hydrophobic surface. The other cellulose chains along the crystal plane then gradually move toward the octane molecules. In both emulsions, the cellulose was found to interact with both water and octane surfaces with specific conformations that allow the CH groups of the glucose rings to contact with octane molecules, while the OH groups of these rings contact with water molecules to form hydrogen bonds. The cellulose chains on the octane droplet also contact with each other through lateral hydrogen bonding between chains. These interactions stabilize the emulsion formed by cellulose molecules as surfactants.     

Aging of Transformer Insulation Paper

Abstract

As a transformer ages, the chemical and physical properties of the cellulose insulation materials in the transformer change, and the paper loses its strength and becomes brittle. The average molecular weight of the cellulose chains decrease with age, and degradation products are formed, including water, carbon monoxide, carbon dioxide, and furans. The molecular weight changes in the cellulose have been studied by several methods, but the GPC method for determination of the molecular weight has been shown to offer advantages over other methods because it yields the total molecular weight distribution. The tensile strength of the cellulose insulation in transformers also changes with age as a result of the changing molecular weight of the cellulose. Work carried out in our laboratory on the aging of cellulose insulation is reviewed in this paper. Our studies have included investigations of insulation materials from retired transformers as well as accelerated aging of insulation paper in transformer oil in the temperature range of 129–166°C under vacuum. In the study the relationships between the molecular weight of the cellulose and the furan degradation products and tensile strength have been delineated, and they have been correlated with information on the kinetics of degradation of the insulation paper.     

Molecular mechanism of the viscosity of aqueous glucose solutions

Experimental relations are obtained for the viscosity of aqueous glucose solutions in the temperature range of 10–80°C and concentration range 0.01–2.5%. It is found that the concentration dependence of fluidity is linear when the concentration is higher than a certain value and varies at different temperatures. The existence of such a dependence indicates that the mobilities of solvent and solute molecules are independent of the concentration of solutions. This assumption is used to construct a theoretical model, in which the structure of an aqueous glucose solution is presented as a combination of two weakly interacting networks formed by hydrogen bonds between water molecules and between glucose molecules. Theoretical relations are obtained using this model of network solution structure for the concentration and temperature dependence of solution viscosity. Experimental data are used to calculate the activation energies for water (U _w = 3.0 × 10^–20 J) and glucose molecules (U _g = 2.8 × 10^–20 J). It is shown that the viscosity of a solution in such a network structure is governed by the Brownian motion of solitons along the chains of hydrogen bonds. The weak interaction between networks results in the contributions to solution fluidity made by the motion of solitons in both of them being almost independent.