Search

In liquid-liquid systems, the effects of agitation for promoting mass transfer are set forth as follows:(1) Increase in the interfacial area.(2) Diminution of diffusional resistance in continuous phase.(3) Improvement in the mixing of the inner part of droplets.The authors measured the interfacial area and the over-all reaction rate for various liquid-liquid systems in agitation, and obtained the following conclusions.(1) For a liquid-liquid system whose density difference is zero and where simple physical dissolution of dispersed phase or diffusion-controlling reaction takes place, the increase in the rate of mass transfer is proportional to the increase in the interfacial area caused by the increase in agitator speed. Therefore the mass transfer resistance in the outer-side diffusional film of droplets is constant regardless of the agitator speed.(2) For a system where density difference is zero and where diffusion-controlling reaction takes place, the increase in agitator speed is slightly effectivc for the diminution of the outer-side diffusional resistance of droplets. However, as the density difference in liquid-liquid system is very small, the purpose of agitation is chiefly to increase the inter-facial area and to diminish, only slightly, the diffusional resistance.(3) For the system where density difference is zero and the diffusional and reaction resistance are comparable in magnitude, the over-all rate of reaction is, in the range of lower agitator speed, proportional to the increase in the interfacial area caused by the intensification of agitation.At a higher agitation speed, the concentration of dissolved and unreacted reactant in the continuous phase increases gradually with the increase in agitator speed, but the increase in the interfacial area is not accompanied by the increase in the over-all rate of reaction. Finally, the continuous phase is to be in saturated condition and the over-all reaction rate is to be controlling, if the overall rate is to be independent of agitator speed. Thus in liquid-liquid reaction system, there is sometimes an upper limit to effective agitator speed.

The authors conducted experiments on the solution of solid particles in agitated liquids in a series of geometrically similar agitators. A great variety of solid particles, including several kinds of spherical crystals, whose particle size, shape and density varied over a wide range, were used as solid samples. Pure water, dilute hydrogen chloride and aqueous solutions of gelatine, polyvinyl alcohol, sucrose and glycerine of various concentrations were used as liquids.Between the agitator speeds of Nf and Na, all the data were correlated by the following dimensionless equations.For solid particles with an unknown surface shape factor, (13)and for solid particles with a known surface shape factor, (14)where(24)and(10)These equations represent the general correlation of the variables involved in determining the magnitude of the rate of mass transfer in agitated liquid-solid systems, and are used-to estimate the mass transfer rate for any liquid-solid combination in a series of geometrically similar vessels and impellers.

The pattern of the flow state on a sieve plate can be divided into three regions according to different gas velocities. At the range of lower gas velocity, Region I, the clear liquid layer exists on the floor of the plates and moreover the cellular froth layer is formed on it. In the range of intermediate gas velocity over uG=10 to 15cm/sec, Region II, the cellular froth layer grows up rapidly. In this region, the holdup of the gas phase in the froth layer first increases rapidly with the increase in gas velocity, reaches the maximum value and then decreases. At the range of higher gas velocity, Region III, the cellular structure is broken and the froth creates eddies as it flows. In this region, the holdup of gas phase is independent of gas velocity.In Region I, the liquid-mixing in its direction of flow on the plate obeys diffusion mechanism and is controlled by the mixing in the clear liquid layer. In Region III, the liquid-mixing also obeys diffusion mechanism and is presumed to be caused by the eddies in the froth layer. On the contrary the liquid-mixing in Region II does not always obey diffusion mechanism because the circulation of the cellular froths in the direction of liquid flow is one of the important factors controlling the liquid-mixing.On the other hand, gas phase is presumed to flow through the froth layer in a state close to piston flow in Region III.The conclusions mentioned above and the theoretical relations for the concentration gradient in liquid phase on a tray which were already derived by authors are confirmed by means of the experiments of desorption of ammonia or oxygen in water by air.

H. Kramersら6) が行なったと同様な電極法によって, 回分撹拌槽内の母液中に投入した少量の塩水の混合速度を測定した。混合過程が槽内の液の対流循環と乱流拡散によって支配されることを認め, 各種形状の撹拌機についてそれらの混合性能1/TM・nと動力数Npおよび液循環係数Nqdとの闇にEq.(3) のような相関式を得た。

翼長, 翼巾, 羽根板角度, 翼段数などの装置条件を包含した撹拌槽内冷却コイルよりの伝熱相関式を実験的に求めた。著者らの得た実験式によれば従来報告されている実験値の相関は良好であった。また従来の研究者によるRe数の指数の相違 (0.5, 0.62および0.67) は, 翼による吐出流が冷却コイルに直接衝突する位置にあるか否かによることを明らかにした。なお邪魔板設置による伝熱係数の上昇は冷却コイル側ではわずかであった。

The average degree of polymerization and the molecular weight distribution in the outflowing fluid from the homogeneous continuous reactors were theoretically estimated, assuming that the fluid mixing was expressed by mean, mixing diffusivity.The effects of the fluid mixing on the degree of polymerization and the molecular weight distribution were found to depend on the polymerization mechanism. Provided that the polymerization mechanism consisted of the initiation reaction of zero order for monomer, of propagation reaction and of termination reaction by recombination, and that the concentration of active polymer was much smaller than that of dead polymer, the degree of conversion of monomer and the average degree of polymerization was increased with the decrease in fluid mixing. On the other hand, the spread of molecular weight distribution was decreased with the increase in fluid mixing at a certain mean residence time.The effects of the fluid mixing was great in the range of M=0-5 and in higher degree of reaction yield.The polymer properties obtained by means of diffusion model was approximated by means of the stirred tanks in series.

The kinetics of the catalytic hydrogenation of benzene on a commercial nickel catalyst (dp=2.65mm) was studied by a differential type of flow reactor.The conversion of benzene is 5-8% for ordinary runs, and the maximum deviation of the catalyst temperature in the bed is less than 8°C.Initial rates of reaction without cyclohexane and the rate for reactants containing some cyclohexane were measured at atmospheric pressure over the following range of conditions.Temperature: 140, 160, 180, 200 and 220 [°C]Partial pressure: pH=0.4-0.9, pB=0.1-0.6 pO=0-0.5 [atm.]Flow rate: 26-31 [gr/cm2hr.]Twelve different mechanisms were presented, and checked by comparing experimental data of initial rates with the characteristic curve of the modified rate equationAs a result, the most plausible rate-controlling step was found to be in the surface reaction between one adsorbed benzene molecule and three adsorbed hydrogen molecules on the same type of active sites.The final equation recommended for the vapor-phase hydrogenation of benzene on a nickel catalyst is as follows

通気撹拌におけるガス滞留率の正確な測定方法として従来の試みとは異なるトレーサーガス混入法を提案し, 標準平羽根ターピン邪魔板付撹拌装置の幾何学的に相似な一連および形状の異なる2, 3の標準翼につき滞留率を実測しこれと通気量, 撹拌速度, 槽径などとの相関関係を求め, また撹拌所要動力との関係も明らかにした。

2次反応を伴う2種のガスの同時吸収に関する理論的ならびに実験的研究を行なった。浸透説に基づく基礎方程式を有限差分法により解き, 各ガスの反応係数を与える線図をパラメーターの種々の組に対して提出した。さらに液柱塔および濡壁塔を用いてCO2, NH3の水中への同時吸収実験を行なった。実験データをCO2, NH3両ガスの反応係数を考慮して解析したところ, CO2の吸収速度は理論値と良く一致することが確かめられた。

Analytical studies were attempted to make clear the flow patterns of the liquid in the cylindrical mixing vessel without baffles, as the experimental results had already been obtained as shown in the previous reports.We could not well explain the details of the liquid flow patterns in the vessel by means of the so-called “Rankine's combined vortex” model (Refer to Fig. 1 and Eq.(1)). Therefore, we studied three important factors contributing to the momentum transfer in the agitated liquid; i.e., the liquid viscosity, the displacement of the liquid by the agitating impeller and the secondary circulation flow which is induced by the discharge flow from the tips of the impeller blades.In the range where Reynolds number was very small, the momentum transfer due to the viscosity effect was found to be predominant, so that the analytical results, obtained upon the assumption that the effects of the two factors other than the liquid viscosity are negligible, had a tendency to show good agreement with the experimental data on the tangential velocity distribution of the liquid in the vessel, as shown in Figs. 4 and 5.On the other hand, in the turbulent flow range where Reynolds number was very large, the secondary circulation flow had considerably important effects on the liquid flow pattern. Considering its contribution to the momentum transfer (Refer to Fig. 6), Eqs.(10) and (13) were analytically derived to express the tangential velocity distribution of the agitated liquid. The comparisons between the experimental data and the calculated results by these equations are shown in Fig. 8. They show good agreement.Furthermore, there were obtained data on the distributions of the turbulent kinematic viscosity and the static pressure in the vessel, as shown in Figs. 9 and 10, respectively.These results are considered to show the utility of the analysis.

The velocity distribution of liquid in a cylindrical mixing vessel with flat-plate-baffles like those commonly used was measured by the method similar to the one adopted for the agitator without baffles.1)Some of the experimental results are shown in Figs. 3, 4, 5, 6, 7, 8 and 9.Variations of the liquid velocity distributions caused by the baffle-plates inserted in the agitated vessel are shown in Fig. 10. Obviously, the insertion of baffle-plates reduces the circulating flow round the impeller axis (the circumferential component υt of liquid velocity) and promotes the circulation flow in the vertical direction (caused by the discharge flow from the tip of the impeller).The discharging performance of various impellers is represented by the ratio NPB/Nq1, which is a dimensionless factor corresponding to the relative power required for a unit quantity discharge. The ratios for various impellers are listed in Table 3 together with those in the non-baffled condition. It is to be noted that, in spite of a considerable increase in Nq1, the circulation efficiency of agitators is lowered by the insertion of baffle-plates.Furthermore, the power consumption in the neighbourhood of the impeller (NPimp) was calculated and compared with that consumed in the outer region of the vessel (ΔNP) as shown in Table 4.It may be concluded from above that the improvement in the circulating capacity can be accomplished to a certain extent by a proper design of baffle-plates.

The stirrer speeds at which the contents of agitation vessels of various sizes come to similar states of mixing, are proportional to -2/3 power of the vessel diameters.Moreover, in these corresponding speeds, the extraction velocities of solutes between two immiscible liquids and also the dissolution velocities of solid particles are equal in all ranges of stirrer speeds.In the geometrically similar systems, powers absorbed per unit volume of agitated liquids are proportional to N3D2.From these results, it can be concluded that the agitation intensities can be measured by the power per unit volume of liquids, say H.P./m3.

Energy balance and energy consumption in an agitating system, consisting of cylindrical mixing vessel and an impeller, are analyzed in this report, as the flow patterns of agitated liquid were made clear in the previous studies.In the first place, based upon the results of the calculation of the energy balance on the cylindrical boundary surface, where the impeller is placed as shown in Fig. 1, it is made clear that the agitating energy loss is greatest in the liquid near the impeller (refer to Table 1).Then, the mechanisms of the energy consumption in the agitated liquid are examined.According to the analytical consideration, the energy consumption in the liquid inside the impeller is caused mainly by the head loss due to abrupt change in the liquid velocity and that outside the impeller, by the energy dissipation due to the turbulent viscosity, under non-baffled condition. However, under fully-baffled condition, the head loss due to the abrupt change in the liquid velocity prevails everywhere in the agitated liquid (refer to Table 2).Furthermore, the general considerations are made on the energy consumption in a simplified model, which shows the flow patterns in the vessel as the combination of so-called compound vortex flow and vertical circulation flow, as given in Fig. 5.In short, it is emphasized in this report that both the liquid-circulating capacity and the turbulencegenerating ability of an impeller are important factors in the performance of an agitator.

通気撹拌における液側容量係数と撹拌所要動力との比を「気-液混合効果の動力効率」とよび, これを比較の基準として通気撹拌装置の撹拌翼形状, 翼高さ, 邪魔板などの最適条件の決定を試みた。気-液系では槽内全体にわたる循環流より液の部分的乱れすなわち主として速度一, 圧力頭からなる全ヘッドをより大にする場合, 上記の動力効率がよい. これには邪魔板の受ける抵抗を幾分弱くし, 羽根板枚数および幅が小さく翼長が槽内径の3分の1程度に小にし, また翼の回転抵抗をある程度小にして高速回転させるのがよい。

A new simplified design method is presented for the design of fixed bed catalytic reactors.The basic partial differential equations, (1)′ and (2)′ were reduced into the corresponding simpler forms, (15) and (16), or (17) and (18), by means of introducing the following assumptions besides the ordinary assumptions: 5) The radial profile of temperature (t) at any axial position is geometrically similar to that in the hypothetical model reactor, although the temperature level itself may be different from the one in the other. 6) The radial mass transport at the central axis of the catalytic reactor tube is negligibly small.The calculation is to be conducted as follows: 1) In order to obtain the temperature and conversion profiles at the central axis, Eqs.(17) and (18) ought to be integrated simultaneously by stepwise numerical method. It is to be noted, however, that the values of the factor φ are not constant, but vary with the axial distance from the inlet. 2) The temperature at any radial position and the mean temperature at any cross section of the reactor tube can be calculated by Eqs.(21) and (22), respectively, using the value of tcobtained in the first step. 3) The mean conversion curve is given by Eq.(29) or (29)′, using the value of tm, obtained in step2.Diagrams for φ and ψ as functions of the axial distance should be prepared in order to carry out the procedureed mentioned above. In this work the authors obtained the formulae representing these values, φ and ψ, for three kinds of reaction rate models and showed them in Figs. 2, 3, and 4.Temperature and conversion distribution curves computed by the proposed method are in good agreement with the experimental data reported by Smith et al. and Wilson.(Figs. 5, 6, and 7)It is also shown in Fig. 8 that the key assumption 5) is reasonable.

The Optimum construtions for gas-liquid agitation vessel have been determined as follows:(1) The optimum submersion length of baffle plates is equal to half the liquid depth. Therefore, it is erroneous to use the baffle plates extending to the bottom of vessel as usually be found in Penicillin broth tanks.(2) The optimum sparger loop diameter is equal to 80% of the paddle length.(3) The optimum baffle plate width is equal to 18% or 12.5% of the tank diameter respectively, according to the number of baffle plates is 2 or 4.(4) There is a certain minimum agitator speed needed, to absorbe gases effectively.(5) Several types of agitators were compared each other.

Experimental studies were made on the limiting flow rate in a countercurrent liquid-1iquid extraction column equipped with stirrers.Two types of operating regions were recognized to be caused by flow rates and stirrer speeds as shown in Fig. 3: the regular and the transition-region.The regular-region was caused in direct proportion by higher stirrer speeds and higher flow rates of continuous phase. The maximum flow rate of dispersed phase was observed when flooding phenomena occurred.The transition-region was caused in inverse proportion by lower stirrer speeds and lower flow rates of continuous phase. The maximum flow rate in this case was observed when inversion of two phases (from the state of oil in water to the state of water in oil) occurred.Experimental data were correlated by plotting the dimensionless groups of (Ud/Uc) vs (Ucμc/σ) (NDND2ρc/μc)0.94 (NDND2ρd/μd)0.10 (NN2D3ρc/σ)-0.53 (NN2D/g)1.51 (Δρ/ρc)-1.90 for flooding velocities, and (Ud /Uc) vs (Ucμc/σ) (NDND2ρc/μc)0.97 (NN2D3ρc/σ)-0.90 (NN2D/g)0.76 (Δρ/ρc)-0.60 for phase inversion velocities as shown in Fig. 9.

An approximate method is presented for estimating the effects of heat evolution which is accompanied with chemical reaction on heat transfer in packed beds.It is assumed that the radial temperature profile at any axial position of the catalyst bed is geometrically similar to that in a model reactor, although the temperature level itself may be different from one another.In order to evaluate the effect of reaction heat on heat transfer coefficients, the reaction coefficient for heat transfer, O, is defined by Eq.(28) for model reactor type 1. The value of O for real reactors may be derived from the value of the model reactor and found to be approximately the same.As a result, it may be concluded that chemical reaction increases the overall heat transfer coefficients in packed beds. This is mainly due to the change of radial temperature profile which originated in the heat of reaction.

The following relations have been derived for the local and plate efficiencies in a plate column on the assumption that the back-mixing diffusivity in liquid flow on a tray and flow velocities of both phases are constant:whereBy means of the above equations, the effects of the liquid mixing, holding time and mass transfer coefficient may be predicted, and these correlations explain the results which have been obtained empirically or qualitatively by many investigators.

固体の液体への溶解速度に關するHixson及びBaum兩氏の實驗値(Ind. Eng. Chem., 33., 478 (1941))を整理し直して次の如き結果を得た.(1) Hixson氏等は粒子の大きさ,密度及び重力の因子を考慮に入れていないから同氏等の得た一般式は誤である.(2) 兩相間の密度差が大きくなる程境膜低抗の抵下に對する攪拌の効果は大となる.(3) 從つて固-液系の攪拌に於ける攪拌の主要目的は或速度以上ではもつぱら擴散境膜抵抗の低下にあるが液-液系の攪拌に於ては攪拌目的は主として反應表面積の増加にあり,これに比べれば擴散境膜低抗抵下の目的は小である.

高粘性流体や非ニュートン流体の混合には槽内全体に上下循環流を強制せしめる翼が有効である。混合能力の秀れた翼として案内円筒付半楕円板翼, 案内円筒付スクリュー翼, リボンとスクリェーの組み合せ翼, および2枚羽根対称取り付けのリボン翼が見い出された。リボンの軸部にスクリューを取り付けることによって混合能力の向上が得られるが, スクリューによる送り出し液量がリボンのそれより小なる際には混合能力は低下する。

The optimum conditions for the baffle plate insertion have been determined as follows:(1) The optimum distance of baffle plates from the center line of a cylindrical vessel is equal to half the paddle length.(2) The optimum Submersion length of baffle plates is also equal to half the paddle length.(3) The optimum baffle width is equal to 10% of the vessel diameter.(4) The optimum paddle' length, on the baffled conditions mentioned above, is equal to 40% of the vessel diameter.

向流連続式の液々抽出裝置としてはスプレー塔,充填塔等が使用され又研究されているが何れも効率はあまりよくない。本報告は液々抽出を効率よく,しかも向流連続的に行わしめるために接触塔の内部を隔離板によつて多数の接触室に分割し,各室毎に攪拌翼を裝置して両液相の接触を良好ならしめた抽出裝置の研究である。本裝置の目的は攪拌作用によつて分散相を適度に微粒化して抽出表面積を増大せしめ,あわせて両相界面の拡散抵抗を低下せしめんとするにある。なお構造的には分散相ならびに分散媒の通路を分離することにより,各室内の分散状態を均等ならしめると共に分散と合一を頻繁に反覆せしめ得て回分式多重段向流抽出の繁雜なる繰作が自働的に行いうる。又,抽出塔内の溶剤のホールドアップは2液の流速比とは無関係に,任意に調節することができ,適当な設計をすれば40%に達せしめることが可能である。本抽出裝置につき著者は日本特許第182, 796號(昭和24年12月6日)を得た。

Variations of the concentration of the continuous phase in the pulsed plate column were measured in the experiments on the extraction of acetic acid from methylisobutyl ketone by means of water.In this column, the longitudinal mixing was very intensive. The variation of the concentration was approximated on the assumption that the backmixing diffusivity was constant throughout the column.Capacity coefficient eliminated from the effect of the longitudinal mixing and the backmixing diffusivity were calculated in the present work.It was concluded that the decrease of extraction efficiency in the lower range of operating condition was caused by the longitudinal mixing.

Longitudinal mixing in continuous flow reactors with multi-stage impellers (see Fig. 2) was studies on·the basis of the residence time curve.In such a vessel, the liquid flow is assumed to be a combined flow of piston flow and perfect mixing. Defining the fraction of piston flow (φ), the conversion (z) was related with it as in Eqs. (2), (3) and (4).Values of the fraction (φ) were studied by means of tracing the residence time curves, and they were correlated as presented in Eq. (8) and Fig.7 for all vessels as shown in Fig.2.The correlation between the fraction of piston flow (φ) and the mean mixing diffusivity (E) is shown in Fig.10. E was correlated as in Eq. (14) and Fig.11.Fig.12 indicates that good agreement was attained between the values Zcalc., obtained from the above equations, and zobs., observed in our experiments on the hydrolysis reaction of acetic anhydride.

6枚羽根平板タービンによる液中への空気分散の消費動力を翼長, 液粘度および空気流還を変えて測定した。実験結果はレイノルズ数, フルード数および通気係敬を用いてEq.(6) に示すごとく相関された。既に指摘されているように, 通気係数に相関に当って重要な項であるが, さらにフルード数を考慮せねばならないこと, およびレイノルズ教は消費動力にはほんの僅かな影響しか与えないことを明かにした。