Hemodialysis membrane prepared from cellulose/N‐methylmorpholine‐N‐oxide solution. III. The relationship between the drying condition of the membrane and its permeation behavior

The effects of drying condition on the performance (ultrafiltration rate, diffusive solute permeability, and sieving) of hemodialysis membranes prepared from cellulose/N‐methylmorpholine‐N‐oxide (NMMO) solution (NMMO membrane) and cellulose/cuprammonium solution (cuprammonium membrane; the referential membrane) were studied. The drying condition investigated was the glycerin concentration of the solution, which was used to substitute glycerin for the water in the membrane before the membrane was dried. A lower glycerin concentration in the solution brought about a lower reswelling degree (water content) in the dried membrane in pure water, which resulted in a drop in the performance of the as‐cast membrane. The NMMO membrane had a high water content and a high membrane performance compared with the cuprammonium membrane when both the membranes were treated under the same drying condition. The differences in the performance between both membrane series is discussed on the basis of the results of the observation of the membrane morphology by scanning electron microscopy, the observation of the crystallinity of the membranes by wide‐angle X‐ray diffraction, and the estimation of the pore structure of the membranes. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 89: 1671–1681, 2003     

Hemodialysis membrane prepared from cellulose/N‐methylmorpholine‐N‐oxide solution. I. Effect of membrane preparation conditions on its permeation characteristics

Flat hemodialysis membranes were prepared from cellulose/N‐methylmorpholine‐N‐oxide (NMMO) solutions (dope) with different cellulose concentrations (6–8 wt %) by using a phase‐inversion method. The coagulant used was NMMO aqueous solution, of which the NMMO concentration and its temperature were varied in the range of 0 to 50 wt % and 5 to 60°C, respectively. The effects of these preparation conditions on the permeation characteristics, the ultrafiltration rate (UFR) of pure water, and sieving coefficient (SC) of dextran, were investigated. The decrease in cellulose concentration of the dope and the increases in both temperature and NMMO concentration of the coagulant gave a membrane with high UFR. Concerning the SC, the increase of the cellulose concentration and the decreases in both temperature and NMMO concentration gave a good result. Consequently, the membrane having the preferable UFR and SC as a hemodialysis membrane was obtained when the 8 wt % cellulose dope was coagulated in water at 5°C. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 84: 2302–2307, 2002     

Atomic force microscopy of cellulose membranes prepared from the N‐methylmorpholine‐N‐oxide/water solvent system

Cellulose membranes were obtained by solutions of cellulose being cast into a mixture of N‐methylmorpholine‐N‐oxide (NMMO) and water under different processing conditions. Atomic force microscopy (AFM) was used to investigate the surface structures of the membranes. The AFM method provided information on both the size and shape of the pores on the surface, as well as the roughness of the skin, through a computerized analysis of AFM micrographs. The results obtained showed that the surface morphologies were intrinsically associated with the permeation properties. For the cellulose membranes, increasing the NMMO concentration and the temperature of the coagulation bath led to higher fluxes and lower bovine serum albumin rejection. These were always correlated with higher values of the roughness parameters and larger pore sizes of the membrane surfaces. When the cellulose concentration of the casting solution was 11 wt %, the membrane showed a nodular structure with interconnected cavity channels between the agglomerated nodules. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 86: 3389–3395, 2002     

Rheological aspects of fiber spinning from cellulose solutions in N‐methylmorpholine‐N‐oxide

Based on rheological experiments with a cellulose solution in N‐methylmorpholine‐N‐oxide (NMMO), it was found that the shearing stress generated in the flowing viscoelastic fluid decreases with an l/d ratio in a rheometer capillary. This reduces the elastic response and the outflow of the fluid becomes more uniform. At constant temperature, the elongational viscosity of the solidified stream of the cellulose solution in NMMO is reduced with increase of the deformation rate, which makes it possible to increase the fiber‐formation velocity within the air zone. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 79: 1860–1868, 2001     

Modified cellulose fibers prepared by the N‐methylmorpholine‐N‐oxide (NMMO) process

Cellulose fibers with modified properties have been prepared from cellulose solutions in N‐methylmorpholine‐N‐oxide (NMMO). Poly(ethylene oxide) as a hydrophilic modifier and polyethylene as a hydrophobic modifier were added to the spinning solution. Based on microscope examination and measurements of such properties of fibers as porosity, moisture absorption, water retention, and tensile strength, structural changes as well as physical and mechanical properties of the resultant fibers depending on the amount of modifier added to the spinning solution were analyzed. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 86: 907–916, 2002     

Physical agglomeration behavior in preparation of cellulose‐N‐methyl morpholine N‐oxide hydrate solutions by simple mixing

The different melting temperatures of N‐methyl morpholine N‐oxide (NMMO) hydrates in the cellulose–NMMO hydrate solution may be explained by the rather different crystal structures of NMMO hydrates, which are determined by the amount of the hydrates. The preparative process of cellulose–NMMO hydrate solution may result in cellulose structural change from cellulose I to cellulose II, depending on the amount of the hydrate. Mixtures of cellulose and NMMO hydrate in a blender was changed from the granules to slurry with increasing mixing time at 60–70°C, which is below the melting point of the NMMO hydrate. In the case of 15 wt % cellulose–NMMO hydrate granules, which were made by mixing for 20 min, the melting points of various NMMO hydrates were obtained as 77.8°C (n = 0.83), 70.2°C (n = 0.97), and 69.7°C (n = 1.23), respectively, depending on the hydrate number. However, the melting points of cellulose–NMMO hydrate slurry and solution were shifted lower than those of cellulose granules, while the mixing time of slurry and solution are 25 and 35 min, respectively. These melting behaviors indicate instantaneous liquefaction of the NMMO hydrate and the diffusion of the NMMO hydrate into cellulose during mixing in a blender. When cellulose was completely dissolved in NMMO hydrate, the crystal structure of cellulose showed only cellulose II structure. In the cellulose–NMMO products of granules or slurry obtained by high‐speed mixing, which is a new preparation method, they still retained the original cellulose I structure. © 2004 Wiley Periodicals, Inc. J Appl Polym Sci 93: 1687–1697, 2004     

Blend films of O‐carboxymethyl chitosan and cellulose in N‐methylmorpholine‐N‐oxide monohydrate

To introduce N‐methylmorpholine‐N‐oxide (NMMO) process to prepare antibacterial lyocell fiber, the blend films of O‐carboxymethyl chitosan (O‐CMCS) and cellulose were prepared. O‐CMCS in aqueous suspension with particles having a surface mean diameter of 2.24 μm was blended with cellulose in NMMO hydrate. The blend films with different O‐CMCS content were prepared with the blend solutions. SEM confirmed that O‐CMCS remained within the cellulose film in the particle. The mechanical properties of the blend films show little increased value when O‐CMCS was less 5%; however, it decreased sharply when O‐CMCS was over 8%. Thus, the optimum O‐CMCS content may give a good combination of antibacterial action and mechanical properties. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 102: 4601–4605, 2006     

Cellulose nanofibers prepared by the N‐methylmorpholine‐N‐oxide method

The process of electrospinning is very suitable for obtaining fibers with a diameter on a nanometer scale. Such fibers can be spun from almost all kinds of known polymers, copolymers, and polymer blends. In this work, we present cellulose nanofibers obtained by the electrospinning process from spinning dopes containing cellulose dissolved in an N‐methylmorpholine‐N‐oxide/water system. Under different electrospinning process conditions, cellulose fibers, a nonwoven fiber network, and a cellulose membrane were obtained. The fibers were examined with scanning electron microscopy. The diameters of the fibers were in the submicrometer range. © 2005 Wiley Periodicals, Inc. J Appl Polym Sci 98: 1855–1859, 2005     

Effects of a pretreatment with N‐methylmorpholine‐N‐oxide on the structures and properties of ramie

To improve the dyeing properties of ramie, the ecofriendly organic solvent N‐methylmorpholine‐N‐oxide (NMMO) was used to substitute sodium hydroxide as a ramie‐fiber swelling solvent. Through padding and baking pretreatment, ramie fabric was modified by an NMMO aqueous solution. Ultraviolet–visible spectrophotometry, Fourier transform infrared spectroscopy, X‐ray diffraction, and differential scanning calorimetry were used to investigate the effects of NMMO pretreatment on the structure of the ramie, whereas the color strength (K/S, where K is the light absorption coefficient and S is the scattering coefficient), adsorption isotherm, and dye uptake rate curve were measured to investigate the effects of NMMO pretreatment on the dyeing properties of the ramie. The results show that the ramie fiber experienced a limited and irreversible swelling because of the partial breakage of interhydrogen and intrahydrogen bonds of cellulose molecules in the amorphous area, but the crystal and chemical structure of the ramie fiber did not change obviously under the experimental conditions. The K/S value of the NMMO‐modified ramie fabrics dyed with reactive dyes increased by about 100%, and the dye uptake increased by 27.88% compared to that of the raw sample, whereas the standard affinity and diffusion coefficient value of the reactive dyes on the NMMO‐modified ramie fabric were higher than those of the raw ramie fabric. © 2010 Wiley Periodicals, Inc. J Appl Polym Sci, 2010     

Preparation and properties of the single‐walled carbon nanotube/cellulose nanocomposites using N‐methylmorpholine‐N‐oxide monohydrate

Single‐walled carbon nanotube (SWNT)/cellulose nanocomposite films were prepared using N‐methylmorpholine‐N‐oxide (NMMO) monohydrate as a dispersing agent for the acid‐treated SWNTs (A‐SWNTs) as well as a cellulose solvent. The A‐SWNTs were dispersed in both NMMO monohydrate and the nanocomposite film (as confirmed by scanning electron microscopy) because of the strong hydrogen bonds of the A‐SWNTs with NMMO and cellulose. The mechanical properties, thermal properties, and electric conductivity of the nanocomposite films were improved by adding a small amount of the A‐SWNTs to the cellulose. For example, by adding 1 wt % of the A‐SWNTs to the cellulose, tensile strain at break point, Young's modulus, and toughness increased ～ 5.4, ～ 2.2, and ～ 6 times, respectively, the degradation temperature increased to 9°C as compared with those of the pure cellulose film, and the electric conductivities at ? (the wt % of A‐SWNTs in the composite) = 1 and 9 were 4.97 × 10^?4 and 3.74 × 10^?2 S/cm, respectively. Thus, the A‐SWNT/cellulose nanocomposites are a promising material and can be used for many applications, such as toughened Lyocell fibers, transparent electrodes, and soforth. © 2010 Wiley Periodicals, Inc. J Appl Polym Sci, 2010     

Gas permeation performance of cellulose hollow fiber membranes made from the cellulose/N‐methylmorpholine‐N‐oxide/H2O system

Cellulose hollow fiber membranes (CHFM) were prepared using a spinning solution containing N‐methylmorpholine‐N‐oxide as solvent and water as a nonsolvent additive. Water was also used as both the internal and external coagulant. It was demonstrated that the phase separation mechanism of this system was delayed demixing. The CHFM was revealed to be homogeneously dense structure after desiccation. The gas permeation properties of CO₂, N₂, CH₄, and H₂ through CHFM were investigated as a function of membrane water content and operation pressure. The water content of CHFM had crucial influence on gas permeation performance, and the permeation rates of all gases increased sharply with the increase of membrane water content. The permeation rate of CO₂ increased with the increase of operation pressure, which has no significant effect on N₂, H₂, and CH₄. At the end of this article a detailed comparison of gas permeation performance and mechanism between the CHFM and cellulose acetate flat membrane was given. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 91: 1873–1880, 2004     

New solvent for polyrotaxane. III. Dissolution of a poly(ethylene glycol)/cyclodextrin polyrotaxane in a calcium thiocyanate aqueous solution or N‐methylmorpholine‐N‐oxide monohydrate

Calcium thiocyanate [Ca(SCN)₂] aqueous solutions above 40 wt % and N‐methylmorpholine N‐oxide (NMMO) monohydrate, which are known to dissolve cellulose, were found to be good solvents for a polyrotaxane comprising α‐cyclodextrin and poly(ethylene glycol). The polyrotaxane could be dissolved up to 12 and 10 wt % in a 40 wt % Ca(SCN)₂ aqueous solution and NMMO, respectively. These are the first instances of a neutral aqueous solution and a cyclic amine oxide, respectively, that readily dissolve the polyrotaxane. These new good solvents, as well as other solvents of the polyrotaxane, except for dimethyl sulfoxide, are identical to those of cellulose, indicating that the dissolution mechanism of the polyrotaxane is dominated by intra‐ and intermolecular hydrogen bonding of the molecule similar to that of cellulose dissolution. The concentrated polyrotaxane solution in a 40 wt % Ca(SCN)₂ aqueous solution showed apparent thixotropy and spontaneous gelation of the solution caused by a gradual increase in its viscosity. © 2007 Wiley Periodicals, Inc. J Appl Polym Sci 2007     

Dry jet‐wet spinning of bagasse/N‐methylmorpholine‐N‐oxide hydrate solution and physical properties of lyocell fibres

Sugarcane bagasse, a cheap cellulosic waste material, was investigated as a raw material for producing lyocell fibers at a reduced cost. In this study, bagasse was dissolved in N‐methylmorpholine‐N‐oxide (NMMO) 0.9 hydrate, and fibers were prepared by the dry jet‐wet spinning method with coagulation in an aqueous NMMO solution. The effects of NMMO in 0 to 50% concentrations on the physical properties of fibers were investigated. The coagulating bath contained water/NMMO (10%) solution produced fiber with the highest drawability and highest physical properties. The cross‐section morphology of these fibers reveals fibrillation due to the high degree of crystallinity and high molecular orientation. In the higher NMMO concentrated baths (30 to 50%), the prepared fibers were hollow inside, which could be useful to make highly absorbent materials. The lyocell fibers prepared from bagasse have a tensile strength of 510 MPa, initial modulus of 30 GPa, and dynamic modulus of approximately 41 GPa. These properties are very comparable with those of commercial lyocell fibers. © 2010 Wiley Periodicals, Inc. J Appl Polym Sci, 2011     

Preparation and characterization of dense cellulose film for membrane application

Regenerated cellulose films were prepared with environmentally friendly process by utilized N‐methylmorpholine‐N‐oxide (NMMO)‐Cellulose system. To prepare a dense cellulose film for membrane application, some parameter process which influence porous forming such as cellulose DP, cellulose concentration, addition NMMO in coagulation bath, coagulation bath temperature, and drying condition were investigated. We resumed that the porosity and pore size of cellulose membrane decrease with lower cellulose DP, higher cellulose concentration, addition of NMMO in coagulation bath, applying room temperature in coagulation bath and drying, and applying vacuum on drying process resulted in membranes with porosity in range of 24–41% and pore size 13.4–20.2 nm. The main factor for controlling porosity and pore size of dense cellulose membrane was coagulation process condition especially addition of NMMO into coagulation bath. © 2011 Wiley Periodicals, Inc. J Appl Polym Sci, 2012     

The Online Measurement of Lyocell Fibers and Investigation of Elongational Viscosity of Cellulose N‐methylmorpholine‐N‐oxide Monohydrate Solutions

In this paper, the development of diameter and surface temperature of Lyocell fibers was measured online. The diameter and tensile force on the spin line in the coagulation bath were traced. The velocity, velocity gradient and the tensile stress profiles development of the fibers in the air gap were studied. The apparent elongational viscosity of cellulose N‐methylmorpholine‐N‐oxide monohydrate (NMMO‐MH) solutions was studied by steady‐state melt spinning theory. The decrease of the fiber diameter was mainly taking place near the spinneret, and the decrease of the diameter became more dramatic with increasing taking‐up speed. The surface temperature of the fibers was also dropping faster with increasing taking‐up speed for the heat transfer coefficient increased. The diameter of the Lyocell fibers almost did not change before and after it entered the coagulation bath. The tensile force on the spin line increases with increasing taking‐up speed and coagulation bath length. The velocity and the tensile stress increase slowly near the spinneret, and then accelerate. The apparent elongational viscosity of cellulose NMMO‐MH solutions decreases with increasing temperature at the same elongation rate and decreases with increasing elongation rate at the same temperature. The fiber of the Lyocell process was not really solidified in the air gap and a gel or rubbery state was formed.     

Effect of various additives on pore size of polysulfone membrane by phase‐inversion process

The phase‐inversion process was used to prepare integrally skinned asymmetric polysulfone (PSf) membranes with different pore sizes. Membranes were prepared from a casting solution of PSf; N‐methyl‐2‐pyrrolidone (NMP) as solvent; and 1,4‐dioxane, diethylene glycol dimethyl ether (DGDE), acetone, and γ‐butyrolactone (GBL) as additives by immersing them in water as a coagulant. The effect of the additives on membrane performance and structure was investigated. The low miscibility of 1,4‐dioxane, DGDE, and acetone with the coagulant resulted in reduced membrane pore size. However, by using GBL as additive pore size of the membrane was slightly increased because of its higher miscibility with the coagulant than NMP. Changing the amount of additives in the casting solution could control the molecular‐weight cutoff values of asymmetric membranes. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 89: 2562–2566, 2003     

Rheological characterization of cellulose solutions in N‐methyl morpholine N‐oxide monohydrate

Three different modes of rheological properties were measured on 11 and 13 wt % solutions of cellulose in N‐methyl morpholine N‐oxide (NMMO) monohydrate, in which concentration range lyocell fibers of much reduced fibrillation are preferably produced. The dynamic rheological responses revealed that the Cox–Merz rule did not hold for these cellulose solutions. Both cellulose solutions showed a shear thinning behavior over the shear rate measured at 85, 95, 105, and 115°C. However, 13 wt % solution gave rise to yield behavior at 85ºC. The power law index ranged from 0.36 to 0.58. First normal stress difference (N₁) was increased with lowering temperature and with increasing concentration as expected. Plotting N₁ vs shear stress (τ_ω) gave almost a master curve independent of temperature and concentration, whose slope was about 0.93 for both cellulose solutions over the shear rate range observed (τ_ω > 500 Pa). In addition, the cellulose solutions gave high values of recoverable shear strain (S_R), ranging from 60 to 100. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 86: 216–222, 2002     

pH‐ and temperature‐sensitive microfiltration membranes from blends of poly(vinylidene fluoride)‐graft‐poly(4‐vinylpyridine) and poly(N‐isopropylacrylamide)

The copolymer poly(vinylidene fluoride)‐graft‐poly(4‐vinylpyridine) (PVDF‐g‐P4VP) was prepared through the graft copolymerization of poly(vinylidene fluoride) with 4‐vinylpyridine. Through the blending of the PVDF‐g‐P4VP copolymer with poly(N‐isopropylacrylamide) (PNIPAm) in an N‐methyl‐2‐pyrrolidone solution, PVDF‐g‐P4VP/PNIPAm membranes were fabricated by phase inversion in aqueous media. Elemental analyses indicated that the blend concentration of PNIPAm in the blend membranes increased with an increase in the blend ratio used in the casting solution. Scanning electron microscopy revealed that the membrane surface tended to corrugate at a low PNIPAm concentration and transformed into a smooth morphology at a high PNIPAm concentration. The surface morphology and pore size distribution of the microfiltration membranes could be regulated by the blend concentration of the casting solution, temperature, pH, and ionic strength of the coagulation bath. X‐ray photoelectron spectroscopy revealed a significant enrichment of PNIPAm on the membrane surface. The flux of aqueous solutions through the blend membranes exhibited a pH‐ and temperature‐dependent behavior. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 100: 4089–4097, 2006     

Microporous membranes prepared from blends of polysulfone and sulfonated poly(2,6‐dimethyl‐1,4‐phenylene oxide)

Membranes were prepared from solutions containing Udel‐type polysulfone (PSf) and sulfonated poly(2,6‐dimethyl‐1,4‐phenylene oxide) (SPPO). Polymer solutions in 1‐methyl‐2‐pyrrolidone were cast on a nonwoven textile and precipitated in a water bath. The permeabilities and selectivities of the prepared membranes depended on the concentrations of both polymers in the casting solution. The higher the concentration of PSf, the lower were the permeabilities to water and average pore sizes of the membranes. On the other hand, a very small amount of SPPO in the casting solution (about 1–4 wt % relative to the casting solution weight) brought about a considerable increase in water permeabilities and had a small influence on the average pore sizes. The effects were most pronounced if SPPO with a degree of sulfonation of 20–40% was used. The considerable increase in water permeabilities was explained by separation of the PSf and SPPO phases during precipitation in water and by the concentration of hydrophilic SPPO on the surface of the membrane and its pores. The determinations of the oriented concentration potentials proved the presence of a negative surface charge in the membranes. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 81: 134–142, 2001     

Synthesis and characterization of dense and porous cellulose films

Whereas cellulose‐derived polymers are routinely used as membrane materials, the cellulose polymer itself is not directly used to synthesize dense/porous films for membrane applications. Recently, N‐methylmorpholine N‐oxide (NMMO) and dimethylacetamide (DMAc)/lithium chloride (LiCl) have been successfully employed for dissolving unmodified cellulose. This provides a strong rationale for reexamining the possibility of cellulose membrane fabrication using these solvents. By judiciously selecting solvents, casting conditions, and solvent exchange steps, we successfully synthesized dense/asymmetric‐porous cellulose films. The pore size and porosity of the porous films decreased systematically with increasing cellulose concentration. SEM analysis of the cross sections revealed an asymmetric skinned structure with monotonically increasing pore size away from the skin. The measured pore diameters were in the range 1.8–4.8 μm. Mechanical testing indicated that the dense films possessed tensile properties comparable to those of cellulose acetate (CA) films. Though nitrogen permeability values were comparable for cellulose and CA dense films, cellulose film permeability depended upon the type of drying protocol employed. Overall, these results demonstrate that processability need not be a constraint in the use of cellulose polymer for membrane fabrication. In selected applications, cellulose membranes could become a cost‐effective, environmentally friendly alternative to other more commonly employed membrane polymers. © 2007 Wiley Periodicals, Inc. J Appl Polym Sci, 2007