Search

Machine learning (ML) was applied to optimize the etching profile for a line and space pattern sample in plasma etching. To investigate the effect of different initial-learning datasets on the optimization of the etching profile, high-, medium-, and low-quality datasets were prepared. The high-quality dataset was composed of etching results relatively close to a target etching profile. The low-quality dataset was composed of etching results relatively far from the target etching profile. The medium-quality dataset was intermediate between the high- and low-quality datasets. For the ML, the kernel ridge regression method was used. After six learning cycles, better etching results were obtained from the medium- and low-quality datasets than from the whole initial-learning dataset. However, the etching results from the high-quality dataset did not exceed those from the whole initial-learning dataset. These results indicate that an initial-learning dataset that has etching results far from the target profile can be useful for optimizing etching profiles.

Curdlan dissolved in alkaline solution forms a unique gel consisting of liquid crystalline gel (LCG) and amorphous gel (AG) in alternating layers by a dialysis into aqueous calcium chloride. The unique structure has been investigated by measuring the birefringence of the gel Δn, the ratio q of the thickness of LCG layer δ to the gel radius R, and the calcium content in the gel CCa in a wide range of molecular weights of fractionated Curdlan, as well as unfractionated Curdlan as a control. With increasing molecular weight of Curdlan, Δn increased and q = δ/R decreased, and both became saturated at high molecular weight. Δn and q for unfractionated Curdlan were smaller and larger, respectively, than those for fractionated Curdlan. CCa was constant irrespective of molecular weight and its distribution, which means that the abundance of calcium ions per glucose unit in the gel does not depend on the degree of orientation of mesogens. These results suggest that the amorphous phase appears when the size of the Curdlan molecules is larger than the average intermolecular distance, resulting from the random coil to triple helix transformation of Curdlan molecules associated with lowering hydroxide anion concentration in the dialysis process.

The processes of gelation and liquid crystalline formation in the dialysis of Curdlan solution have been observed under crossed nicols, and the calcium concentration and pH of the inner solution were traced. The results showed that the gelation and the liquid crystalline formation occurred simultaneously to form liquid crystalline gel (LCG), but the birefringence of the LCG increased even after the gelation, suggesting further ordering of the Curdlan molecules. On the basis of the calcium ion diffusion, a simple theory for the time development of the thickness of the LCG layer was developed. The experimental and theoretical results agree very well until an amorphous gel (AG) ring appears. The whole process was expressed by a master curve by reducing time and distance data for different radius dialysis tubes by those at the final state; a scaling behavior with respect to the dialysis tube radius was found. The experimental analysis for the calcium concentrations and the pH indicates that forming Curdlan LCG with high ordering of Curdlan molecules consists of two steps:  the diffusion of calcium ions inducing the ordering of Curdlan molecules and yielding cross-links simultaneously, and the local relaxation of the Curdlan molecules increasing the ordering degree further.

Curdlan dissolved in aqueous sodium hydroxide was dialyzed to aqueous calcium chloride to form a gel. Transparent and turbid concentric layers observed in the gel cross section perpendicular to the long axis of the dialysis tube were identified as liquid crystalline gels with refractive index gradient and amorphous gels, respectively. The thickness of each layer was proportional to the diameter of the dialysis tube, and the gelation proceeded in proportion to the root of time. The unique pattern formation was attributed to the change of curdlan conformation and calcium-induced cross-linking resulting from a diffusion of calcium cations and hydroxide anions through the dialysis tube. It is suggested that the orderedness of the curdlan molecules decreases by the increase of the curvature of the concentric liquid crystal layers as the layer comes toward the center of the dialysis tube.

The validity of the correction for the interaction parameter g of the Flory−Huggins free energy (Langmuir 2003, 19, 5240) was examined by taking the contributions of spatial variation in the volume fraction into account. Developing the statistical-mechanical view of the invariance conjecture and the Landau expansion with the nonanalytic coefficients for the polymer solutions, we showed that the chain-length dependent correction term H in the interaction parameter g was reasonably introduced. We derived the explicit function form of H as a function of chain-length and temperature to be consistent with the critical power laws for both the concentration difference and the diameter in coexisting phases. It was successfully applied to the systems polystyrene in methylcyclohexane and poly(methyl methacrylate) in 3-octanone.

A semiempirical equation of Gibbs free energy of mixing ΔG for the system polystyrene in methylcyclohexane has been determined based on the Flory−Huggins equation by using a series of coexistence curve data. The functional form of the interaction parameter g with respect to chain length of the polymer was determined so as to express the critical universality. The resultant equation of ΔG was used to prepare microcapsules of controlled size distribution by means of coacervation in the spinodal region.

The dynamic processes of the volume phase transition of pH-sensitive hydrogel wall membranes of microcapsules are much different from those of the conventional bulk gels because of a large proton reservoir inside the microcapsule wall membrane and the extremely thin unsustained gel membrane. The time course of the size S of poly(l-lysine-alt-terephthalic acid) microcapsules has been measured in both the swelling and shrinking processes by pH jump. A lag phase and an S-shaped increase in S are observed in the swelling process, whereas a rapid decrease in S is observed in the shrinking process. To explain the characteristic time courses, a simple model is proposed on assumption of linear dependence of the rate of change in the size and in the charge density of the microcapsule membrane against deviations from the respective equilibrium values. The unique time courses of the microcapsule sizes observed are expressed by the theoretical equations. By a fitting of the observed time courses of S to the theoretical equations, two time constants are obtained, each related to the ratio of the elastic force of the gel polymer network to the frictional force and the ionization of the membrane.

In the interfacial polymerization of poly(urea−urethane) membranes that result in the formation of microcapsules, the mechanism of microencapsulation can be complex. To clarify this process, a series of organic solvents with different degrees of compatibility between the solvent and the polymer, consisting of diethyl phthalate (DEP), dibutyl phthalate (DBP), and dioctyl phthalate (DOP), was used to study the polymerization of triisocyanate in both film and microcapsule formation. The physical properties of both the films and of the microcapsules were then determined by using a combination of differential scanning calorimetry, optical microscopy, and atomic force microscopy. The glass transition temperatures of films with limiting amounts of DEP, DBP, and DOP and without the organic solvent were, respectively, 98, 115, 123, and 124 °C. The triisocyanate + DOP system had a lower critical solution temperature of about 5 °C, with a two-phase domain structure for the membrane synthesized with DOP above the critical solution temperature. The shape of the microcapsules was found to be spherical, biconcave, and biconcave with a hump for compositions with weight fraction of DOP, ws, being 0, 0.3−0.5, and 0.78, respectively. By taking into account the compatibility between the solvent and the polymer, the combined results suggest that chemical cross-linking, swelling, and phase separation can be coupled to control the mechanism in microencapsulation of poly(urea−urethane) membranes to form microcapsules. A homogeneous elastic membrane is formed with organic solvents that are compatible with the wall-forming monomers, whereas the membrane becomes brittle and heterogeneous with less compatible solvents due to phase separations.

The thickness δ of microcapsules consisting of a dioctyl phthalate core and an outer poly(urea-urethane) membrane has been determined by means of single-particle light scattering and the freeze-fracture method in combination with electron microscopy as a function of the dioctyl phthalate/wall-forming triisocyanate monomer volume ratio φ. The thickness δ increased nonlinearly with decreasing φ. The swelling ratio of the membrane was determined to be 1.2 in the range φ < 3, which was explained by the relatively low plasticizing effect of dioctyl phthalate.

We have developed a light-scattering apparatus using a capillary tubing sample cell that is particularly suitable for investigating the structure of a single microcapsule. This approach is capable of determining the individual structures of polydisperse particles because the scattering function of each individual microcapsule can be measured independently. The apparatus was used to determine the outer diameter d and the membrane thickness δ of poly(l-lysine-alt-terephthalic acid) microcapsules as a function of pH. The values of both d and δ were found to be the lowest at pH 4 and increased as the pH value was shifted to either side of pH 4. The change in δ was roughly proportional to the change in d, suggesting an isotropic volumetric phase transition of the hydrogel membrane of the microcapsule. Diffusion coefficients of anionic and cationic electrolyte ions in the microcapsule membrane changed in parallel with δ.

Curdlan beads consisting of liquid crystalline gel (LCG) and amorphous gel (AG) in alternating layers in a wide range of diameters were newly prepared by interfacial insolubilization reactions using calcium chloride as the setting reagent. The thickness of the liquid crystalline layer was proportional to the diameter of the gel bead, and the proportional constant agreed with that determined for the cylindrical gel prepared by a dialysis method. The proportional constant initially increased with increasing calcium concentration of the dispersing medium and saturated at a high concentration limit. These results suggest that the mechanisms for forming the alternating LCG/AG structures prepared with different boundary conditions are the same. The LCG/AG structure could be controlled by calcium concentration.

To obtain optically active polysilane aggregates consisting of an appropriate number of polysilane molecules, polysilane aggregates were prepared into poly(ureaurethane) microcapsules and their bisignate circular dichroism signals around 320 nm were observed at three different temperatures. The size of the aggregates and the chiroptical properties of polysilanes could be fixed by the preparation temperature. The memorized chiroptical property was eliminated only when the aggregation was disintegrated. This methodology is suitable to control the chiroptical properties of polysilanes.