Conventional mixing theories consider convective mixing patterns to be the same as the temporal concentration distributions. However, these theories are not able to explain sufficiently the fact that, regardless of the initial mixing pattern, that all mixing patterns always converge to the unique inherent patterns depending only on the velocity field, In this paper, a new theory is proposed in which convective mixing patterns are regarded as dynamic patterns generated by the aggregate motion of liquid particle groups, which are interconnected from the initial time of mixing. This new theory can explain the fact that the inherent convective mixing pattern always appears naturally from inside the flow field, even if no initial pattern is given to the flow from outside.

Heat transfer coefficients of a non-baffled agitation vessel wall were measured on addition of large particles under various conditions. In the low Reynolds number region, the heat transfer from the vessel wall was suppressed due to particle deposition. Conversely, in the high Reynolds number region, heat transfer was enhanced by the particle dispersion in the vessel. This heat transfer behavior was found to be affected by the particle diameter d_p, volume fraction ϕ_v and density ρ_p. These parameters were qualitatively assessed from the results of the visualization experiments for the particle dispersion in the vessel.

In both low and high Reynolds number regions, correlation equations were proposed for the Nusselt numbers, each of which showed a slope similar to that of single-phase agitation. It was confirmed that these equations closely express the measured values.

The effects of stirring speed n, opening ratio of a perforated partition plate A_r, and vessel diameter D_T on the exchange rate of vertical vessels with three multiple impellers Q were experimentally investigated. The impeller diameter D_i was one-half of the vessel diameter D_T. The exchange rate Q was proportional to the stirring speed of impellers n, the opening ratio of the perforated partition plate A_r, and the third power of the impeller diameter D_i. When the vessel diameter D_T was changed from 0.10 m to 0.17 m, the relation between the modified dimensionless exchange rate Q/(nD_i³A_r) and the agitated Reynolds number N_Re remained unchanged.

Based on the turbulent diffusion model, a new correlation method to estimate the mixing time in agitated vessels was proposed. The dimensionless numbers X₁ and X₃ are functions of power number N_p, rotational speed n, impeller Reynolds number Re, mixing time t_m, d/D and H/D (D, vessel diameter; d, impeller diameter; H, liquid depth), and α is an experimental constant representing the aspect ratio.

The value of α was 0.25 for general-purpose impellers and 0.15 for large-scale impellers. By using the correlation method with the dimensionless numbers X₁ and X₃, it was clarified that the mixing times in the transition region and the turbulent region (Re≧100) can be accurately correlated regardless of the baffle condition and the eccentric position of impeller.

Exhaust gas from solid waste incinerator contains acid gases such as hydrogen chloride and sulfur oxides, and harmful substances such as dioxins and mercury at above regulation levels. A dry process in which slaked lime and activated carbon are injected continuously upstream of a baghouse is usually applied in a simple system to remove these harmful substances. However, this process has low removal efficiency, consuming large quantities of slaked lime and activated carbon but leaving unreacted substances. The Turbo Chemical Baghouse reported here is expected to remove harmful substances with high efficiency through a reactive and absorptive layer at the filter surface. We researched the removal performance of the Turbo Chemical Baghouse, and we found that acid gases, dioxins and mercury were removed with high efficiency.

Adsorption behaviors of metal ions were studied on a crosslinked polyvinylamine resin, which was prepared by suspension polymerization of crosslinked polyvinylamide resin followed by hydrolysis. Precious metal ions present as anionic chloride complexes in hydrochloric acid were selectively adsorbed from among many base metals. As a synthetic resin, the adsorbent has a large number of functional groups (primary amino groups) per unit mass and can be obtained at relatively low cost. Maximum adsorption capacity of Hym-A for Pd(II) calculated by the Langmuir equation was 1.68 mmol g-1, which was 1.87 times that of a polyamine-based chelate resin. For comparison, the adsorption of the hazardous elements arsenic and selenium was also investigated. Hym-A adsorbed As(V) over a wider pH range than the chelating resin bearing secondary amines as functional groups. The maximum adsorption capacities of Hym-A for As(V), Se(IV) and Se(VI) were respectively 3.97, 2.42 and 2.31 times larger than those of a commercial resin. The adsorbent can thus be used for recovery of precious metals and removal of harmful elements with a larger throughput.

Carbon-supported platinum catalysts rapidly evolve hydrogen from a liquid film of tetralin wetted under nucleate boiling conditions. In order to evaluate the catalyst layer, heat recuperation rate was defined as the percentage of the useful hydrogen recoverable from the simultaneous endothermic processes of reaction and evaporation. The ratio of reaction heat over electricity consumption for heating in total, i.e., heat efficiency rate, was adopted as the standard for evaluating the reactor. Both rates were controlled by catalyst-reactivity factors for hydrogen evolution, which were investigated through multiple approaches. Catalysis without platinum was attempted from economic and chemical-bond viewpoints.