## No full-text available

To read the full-text of this research,

you can request a copy directly from the authors.

New data are presented on drop size distribution at high dispersed phase fractions of organic-in-water mixtures, obtained with a light back scattering technique (3 Dimensional Optical Reflectance Measurement technique, 3D ORM). The 3D ORM technique, which provides fast, in-situ and on-line drop distribution measurements even at high concentrations of the dispersed phase, is validated using an endoscope attached to a high-speed video recorder. The two techniques compared favourably when used in a dispersion of oil (density (ρ) = 828 kg m−3, viscosity (µ) = 5.5 mPa s, interfacial tension (σi) = 44.7 mN m−1) in water for a range of 5–10% dispersed phase fractions. Data obtained with the ORM instrument for dispersed phase fractions up to 60% and impeller speeds 350–550 rpm showed a decrease in the maximum and the Sauter mean drop diameters with increasing impeller speed. Phase fractions did not seem to significantly affect drop size. Both techniques showed that drop size distributions could be fitted by the log-normal distribution. Copyright © 2005 Society of Chemical Industry

To read the full-text of this research,

you can request a copy directly from the authors.

... Estimating the drop size distribution (DSD) is crucial as it aects the mass and heat transfer in the liquid-liquid systems [2,3]. There is an extensive number of the experimental works [4,5,6,7,8,9,10,11] to predict the DSD in stirred tank reactors. Furthermore, the population balance equation (PBE) approach [12] has been used frequently to model breakup and coalescence, which are the main phenomena that aect the DSD. ...

... This diameter depends on the local uid dynamic parameters such as the local shear rate and the local mixing length [47]. [8,9,11,10]. Finally a conclusion ends this paper. ...

... Since DSD in some cases of stirred tanks and other systems follow log-normal distribution [8,9,11,10,47] we use ...

In the current work, we present an Eulerian-Eulerian-Lagrangian hybrid model for the simulation of liquid-liquid emulsions in agitated tanks. The main idea of such a modeling strategy is to use a combination of a Lagrangian discrete phase model (DPM) and an Eulerian-Eulerian two-fluid model (TFM) to take advantage of the benefits of those two formulations. Furthermore, droplet breakup can be efficiently evaluated for each DPM trajectory. Here, we propose a novel material property dependent breakup model, where breakup is triggered by the local shear. An implicit approach for coalescence is examined in combination with droplet breakup. The presented method is implemented by an user-defined function in the commercial CFD code ANSYS FLUENT. Finally, the models are validated for several stirred tank configurations. The resulting droplet size distribution (DSD) shows fairly good agreement with the experimentally determined DSDs.

... However, in accordance with [32], the absolute values obtained by the FBRM are significantly smaller than the values of the endoscope measurements with image analysis. The reason for this is FBRM measures a chord length distribution (CLD) with reasonable efforts to transfer the CLD into a PSD [33], which is related to the size and shape of the particles in dispersions. The IPP probe system has been tested successfully in a fluidized bed application [34], but shows disadvantages in liquid dispersions, especially when flow directions are inconsistent. ...

... For each concentration, four image sequences of 100 images have been acquired. Due to the illumination and optical properties of the system refraction increases with higher concentration, 33 whereas the overall brightness and sharpness as features of the acquired images decreases. Knowing that the LAPD can be used as an indicator for the sharpness and the standard intensity an indicator for the brightness of an image, for every image sequence first the LAPD 1,2 is generated by the convolution of the Laplace filter followed by arithmetic averaging. ...

Particles occur in almost all processes in chemical and life sciences. The particle size and shape influence the process
performance and product quality. In many processes particles have certain attributes with respect to their size and
shape distribution. The grain and drop size as well as shape during production is influenced by the flow behavior of the
particles. Monitoring and con-trolling such characteristics in multiphase systems to obtain sufficient qualities will greatly
facili-tate the achievement of reproducible and defined distributions. So far, obtaining this information inline has been
challenging, because existing instruments lack measurement precision, being un-able to process overlapping signals
from different particle phases in highly concentrated multi-phase systems. However, recent advances in photo-optics
made it possible to monitor such fea-tures (particle size distribution, aspect ratio and particle concentration) with
advanced image analysis in real-time. New analysis workflows as well as single feature extractions from the im-ages
using multiple image analysis algorithms allowed the precise real-time measurements of size, shape and concentration
of particle collectives even separated from each other in three or four phase systems. The performances, advantages
and drawbacks with other non-photo-optical methods for assessing the particle size distribution are compared and
discussed.

... The effects of drop size and size distribution on the performance of an extraction column are the most important hydrodynamic characteristics, because under steady operating conditions, drop size is related to the interfacial area by: investigator found a different distribution function suitable to describe his data [Lovick, et al. 2005]. No attempts have been done till now to study the effect of drop size distribution with mass transfer conditions on the performance of RTL contactor. ...

... This was because increasing the rpm increased drop breakup and lead to smaller drops. This was in agreement with Coulaloglou and Tavlarides [1976], Lovick, et al. [2005]. Equation (7) shows a good agreement with the experimental results as shown in the two figures. ...

The present work studied the drop size distribution in the RTL contactor using two liquid-liquid systems, xylene-acetone-water and kerosene-acetone-water. Process variables studied were: rpm (10-50 min-1), continuous phase flow rate (4-12 l/h), dispersed phase flow rate (4-12 l/h), and concentration (0.1-0.5 mole/l). It was found that Sauter mean drop diameter (d 32) decreased with increasing rpm and continuous phase flow rate, and decreasing dispersed phase flow rate and concentration. An empirical correlation, with a correlation coefficient equal to (92.6%), relating d 32 with process variables was developed having the form:

... They concluded that breakage mechanisms seem to be modified at high phase ratios, and DSD decreases with increased impeller speed and low holdup conditions. DSD in batch-stirred tanks was studied by Lovick et al. (2005) using light back scattering techniques such as 3D Optical Reflectance Measurement (3D ORM). A Standard 0.128 m diameter stirred tank with a Rushton turbine impeller, and a water-kerosene liquid system was used. ...

Hydrodynamic studies were carried out at high organic to aqueous (O/A) phase ratios in a liquid-liquid batch stirred tank. Minimum mixing speed for complete dispersion, Sauter mean diameter, and drop size distribution have been measured with Rushton turbine and pitched blade turbine for three different liquid-liquid systems. Experiments have been performed at various impeller speeds and organic-to-aqueous phase ratios. Minimum mixing speed for complete dispersion (Nmin) was measured by visual observation as well as by sample withdrawal technique. Surfactant stabilization technique has been used to measure the drop size. Minimum mixing speed was found to be strongly dependent on impeller type and organic-to-aqueous phase ratios. About 10–40% decrease in minimum mixing speed was observed when organic to aqueous phase ratio was increased from 10 to 50. Sauter mean diameter was significantly affected by impeller speed, physical properties of the liquid systems, and dispersed phase holdup. Experimental drop size distributions were fitted with the log-normal distribution. Empirical correlations have been developed to predict minimum mixing speed and Sauter mean diameter at various operating conditions. Absolute Average Relative Error for the prediction of minimum mixing speed and Sauter mean diameter using developed correlations were found to be 18.45% and 11.06%, respectively.

... 61,62 On the other hand, online sizing of the droplets in industrial processes has great technical relevance in many applications. Some techniques of the online determination of the drop size and drop size distribution are (1) electrical resistance or capacitance tomography, 63,64 (2) direct standard camera photography technique, 7,62,65,66 (3) 3D and 2D optical reflectance measurement (ORM), 67 (4) focused beam reflectance (FBRM), 16,68−74 (5) encapsulation, 75 and (6) laser Doppler anemometer (LDA). 11 Generally, these techniques can be used to measure the final steady-state drop size and drop size distribution within emulsions typically for a volume fraction less than 10%. ...

... For the analytic scalings, several models to predict the resulting DSD were proposed. Among these, the most commonly adopted distributions are: normal [275,276], log-normal [277][278][279], Rosin-Rammler [280], Weibull [281], upper limit equation [282] and power law [271,274,[283][284][285][286]. perimental [271,272,287] and numerical [152,153,274,284,285,288] works, in which turbulent flows laden with drops are considered, report a good agreement with a power law scaling. ...

Turbulent flows laden with large, deformable drops or bubbles are ubiquitous in nature and in a number of industrial processes. These flows are characterized by a physics acting at many different scales: from the macroscopic length scale of the problem down to the microscopic molecular scale of the interface. Naturally, the numerical resolution of all the scales of the problem, which span about eight to nine orders of magnitude, is not possible, with the consequence that numerical simulations of turbulent multiphase flows impose challenges and require methods able to capture the multi-scale nature of the flow. In this review, we start by describing the numerical methods commonly employed and discussing their advantages and limitations, and then we focus on the issues arising from the limited range of scales that can be possibly solved. Ultimately, the droplet size distribution, a key result of interest for turbulent multiphase flows, is used as a benchmark to compare the capabilities of the different methods and to discuss the main insights that can be drawn from these simulations. Based on this, we define a series of guidelines and best practices that we believe important in the simulation analysis and in the development of new numerical methods.

... (30) (Brauner, 2001)). This may not be true for some systems as reported elsewhere (Lovick et al., 2005;Simmons and Azzopardi, 2001). Thus, the use of more comprehensive formulation on this respect can definitely lead to better results. ...

Prediction of dispersed flow regimes in liquid-liquid pipe flow is of great importance for many industrial processes. This work proposes a mechanistic model to determine the stability bounds of dispersed liquid-liquid flow patterns that presents significant improvements compared to other classical criteria. The model accounts for turbulent break-up of dispersed phase droplets, sedimentation, dispersion, and accumulation of droplets. In addition, droplet coalescence and segregation are considered by means of attaining critical concentrations that can be associated with the phase inversion point of the liquid-liquid mixture. Modeled bounds are compared with available experimental data, showing good agreement in flows of mineral oil and water, as well as flows of crude oil and water. Moreover, the model is more accurate and descriptive than other predictions from commonly used criteria. The effect of fluid properties, flow rates, and pipe geometry are discussed. Limitations and possible refinements of the suggested model are also treated.

... The importance of the DSD raised the attention of many researchers and several models to predict the resulting DSD were proposed. Among these, the most commonly adopted distributions are as follows: normal (Brown & Pitt 1972;Chen & Middleman 1967), log-normal (Chatzi & Kiparissides 1994;Lovick et al. 2005;Perlekar, Biferale & Sbragaglia 2012), Rosin-Rammler (Karabelas 1978), Weibull (Brown & Wohletz 1995), upper limit equation (Mugele & Evans 1951) and power law (Garrett et al. 2000;Deane & Stokes 2002;Skartlien et al. 2013;Deike, Melville & Popinet 2016). Although universal agreement over several decades has not yet been demonstrated, a reasonable number of experimental (Deane & Stokes 2002) and numerical (Skartlien et al. 2013) works showed good agreement with the power-law scaling proposed by Garrett et al. (2000). ...

In this work, we compute numerically breakage/coalescence rates and size distribution of surfactant-laden droplets in turbulent flow. We use direct numerical simulation of turbulence coupled with a two-order-parameter phase-field method to describe droplets and surfactant dynamics. We consider two different values of the surface tension (i.e. two values for the Weber number, $We$ , the ratio between inertial and surface tension forces) and four types of surfactant (i.e. four values of the elasticity number, $\unicode[STIX]{x1D6FD}_{s}$ , which defines the strength of the surfactant). Stretching, breakage and merging of droplet interfaces are controlled by the complex interplay among shear stresses, surface tension and surfactant distribution, which are deeply intertwined. Shear stresses deform the interface, changing the local curvature and thus surface tension forces, but also advect surfactant over the interface. In turn, local increases of surfactant concentration reduce surface tension, changing the interface deformability and producing tangential (Marangoni) stresses. Finally, the interface feeds back to the local shear stresses via the capillary stresses, and changes the local surfactant distribution as it deforms, breaks and merges. We find that Marangoni stresses have a major role in restoring a uniform surfactant distribution over the interface, contrasting, in particular, the action of shear stresses: this restoring effect is proportional to the elasticity number and is stronger for smaller droplets. We also find that lower surface tension (higher $We$ or higher $\unicode[STIX]{x1D6FD}_{s}$ ) increases the number of breakage events, as expected, but also the number of coalescence events, more unexpected. The increase of the number of coalescence events can be traced back to two main factors: the higher probability of inter-droplet collisions, favoured by the larger number of available droplets, and the decreased deformability of smaller droplets. Finally, we show that, for all investigated cases, the steady-state droplet size distribution is in good agreement with the $-10/3$ power-law scaling (Garrett et al. , J. Phys. Oceanogr. , vol. 30 (9), 2000, pp. 2163–2171), conforming to previous experimental observations (Deane & Stokes, Nature , vol. 418 (6900), 2002, p. 839) and numerical simulations (Skartlien et al. , J. Chem. Phys. , vol. 139 (17), 2013).

... They are divided into three main groups: sound, laser and photo based techniques. The FBRM® [1], the 2-D ORM® [2] and the PARSUM IPP 30 [3] which all give online and in-situ information together with an in house developed photo optical technique SOPAT-VR® [4,5] will be discussed in detail. They have been tested for different applications in various multiphase system (liquid-liquid, gas-liquid-liquid, gas-liquid-liquid-solid). ...

... They are divided into three main groups: sound, laser and photo based techniques. The FBRM® [1], the 2-D ORM® [2] and the PARSUM IPP 30 [3] which all give online and in-situ information together with an in house developed photo optical technique SOPAT-VR® [4,5] will be discussed in detail. They have been tested for different applications in various multiphase system (liquid-liquid, gas-liquid-liquid, gas-liquid-liquid-solid). ...

• Introduction of new automation technique for multiphase applications
• General overview of inline techniques for the particle characterization given
• ISO standards are applied to the evaluation and show the advantages of image analysis

... They found that droplet dissolution was a significant issue, which made it impossible to obtain a steady-state droplet distribution at low phase fractions, while at higher phase fractions (φ > 0.2), despite breakup, most droplets coalesce to form a single connected region with multiple smaller satellite droplets. Increasing the resolution of the Kolmogorov scale remedied droplet dissolution to some extent, and a log-normal droplet distribution was shown from transient simulations, as has been experimentally found for turbulent liquid-liquid dispersions (Pacek, Man & Nienow 1998;Lovick et al. 2005). The multiphase energy spectra could not be reproduced due to spurious currents which caused unphysical energy gain at high wavenumbers, whose magnitude was found to be close to the turbulent velocity scale u . ...

We perform direct numerical simulations (DNS) of emulsions in homogeneous isotropic turbulence using a pseudopotential lattice-Boltzmann (PP-LB) method. Improving on previous literature by minimizing droplet dissolution and spurious currents, we show that the PP-LB technique is capable of long stable simulations in certain parameter regions. Varying the dispersed-phase volume fraction $\unicode[STIX]{x1D719}$ , we demonstrate that droplet breakup extracts kinetic energy from the larger scales while injecting energy into the smaller scales, increasingly with higher $\unicode[STIX]{x1D719}$ , with approximately the Hinze scale (Hinze, AIChE J. , vol. 1 (3), 1955, pp. 289–295) separating the two effects. A generalization of the Hinze scale is proposed, which applies both to dense and dilute suspensions, including cases where there is a deviation from the $k^{-5/3}$ inertial range scaling and where coalescence becomes dominant. This is done using the Weber number spectrum $We(k)$ , constructed from the multiphase kinetic energy spectrum $E(k)$ , which indicates the critical droplet scale at which $We\approx 1$ . This scale roughly separates coalescence and breakup dynamics as it closely corresponds to the transition of the droplet size ( $d$ ) distribution into a $d^{-10/3}$ scaling (Garrett et al. , J. Phys. Oceanogr. , vol. 30 (9), 2000, pp. 2163–2171; Deane & Stokes, Nature , vol. 418 (6900), 2002, p. 839). We show the need to maintain a separation of the turbulence forcing scale and domain size to prevent the formation of large connected regions of the dispersed phase. For the first time, we show that turbulent emulsions evolve into a quasi-equilibrium cycle of alternating coalescence and breakup dominated processes. Studying the system in its state-space comprising kinetic energy $E_{k}$ , enstrophy $\unicode[STIX]{x1D714}^{2}$ and the droplet number density $N_{d}$ , we find that their dynamics resemble limit cycles with a time delay. Extreme values in the evolution of $E_{k}$ are manifested in the evolution of $\unicode[STIX]{x1D714}^{2}$ and $N_{d}$ with a delay of ${\sim}0.3{\mathcal{T}}$ and ${\sim}0.9{\mathcal{T}}$ respectively (with ${\mathcal{T}}$ the large eddy timescale). Lastly, we also show that flow topology of turbulence in an emulsion is significantly more different from single-phase turbulence than previously thought. In particular, vortex compression and axial straining mechanisms increase in the droplet phase.

... Liquid-liquid emulsions have been the subject of numerous experimental (Berkman & Calabrese 1988;Pacek, Man & Nienow 1998;Lovick et al. 2005) and computational studies ( Perlekar et al. 2012;Skartlien, Sollum & Schumann 2013;Komrakova, Eskin & Derksen 2015;Scarbolo, Bianco & Soldati 2015;Dodd & Ferrante 2016). The computational studies can be broadly categorized as forced homogeneous isotropic turbulence ( Perlekar et al. 2012;Skartlien et al. 2013;Komrakova et al. 2015), decaying homogeneous isotropic turbulence (Dodd & Ferrante 2016) and turbulent wall flows ( Scarbolo et al. 2015). ...

We simulate the flow of two immiscible and incompressible fluids separated by an interface in a homogeneous turbulent shear flow at a shear Reynolds number equal to $15\,200$ . The viscosity and density of the two fluids are equal, and various surface tensions and initial droplet diameters are considered in the present study. We show that the two-phase flow reaches a statistically stationary turbulent state sustained by a non-zero mean turbulent production rate due to the presence of the mean shear. Compared to single-phase flow, we find that the resulting steady-state conditions exhibit reduced Taylor-microscale Reynolds numbers owing to the presence of the dispersed phase, which acts as a sink of turbulent kinetic energy for the carrier fluid. At steady state, the mean power of surface tension is zero and the turbulent production rate is in balance with the turbulent dissipation rate, with their values being larger than in the reference single-phase case. The interface modifies the energy spectrum by introducing energy at small scales, with the difference from the single-phase case reducing as the Weber number increases. This is caused by both the number of droplets in the domain and the total surface area increasing monotonically with the Weber number. This reflects also in the droplet size distribution, which changes with the Weber number, with the peak of the distribution moving to smaller sizes as the Weber number increases. We show that the Hinze estimate for the maximum droplet size, obtained considering break-up in homogeneous isotropic turbulence, provides an excellent estimate notwithstanding the action of significant coalescence and the presence of a mean shear.

... Giapos et al. [3] reported about 52% increase in the mean drop sizes (d 32 ) and wider drop size distribution along with reduction in the number of impeller blades from 8 to 2 for a system of kerosene and distilled water with low dispersed phase holdup. A reduction tendency in maximum and the mean drop sizes was obtained by Lovick et al. [21] by increasing the agitation speed in mixtures of tap water and kerosene using a sixbladed Rushton turbine for up to 60% dispersed phase at the speeds of 350-550 rpm. They also observed insignificant effect of phase fractions on drop sizes which was inconsistent 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 [22]. ...

The present study investigates the effects of impeller design and dispersed phase volume ratio on mean drop sizes (d 32 )in immiscible liquid-liquid stirred vessel through experimental and modeling approaches. Various impeller designs including conventional and new impeller designs were employed to cover both radial and axial flow impellers. The microscopic method associated with image processing tools was used for the drop size analysis. The results showed the hydrofoil impeller produced the largest drop sizes while the double-curved blade turbine produced the smallest drop sizes, corresponding to about 37% difference. Increasing the dispersed phase volume ratio from 1% to 10%)increased the d 32 by approximately 20–40%. Adaptive neuro-fuzzy inference system based on fuzzy C–means (ANFIS-FCM)clustering algorithm was used to develop a model to predict drop sizes, and its validation and accuracy were examined by comparing the results to the experimental data. The results also proved the superior prediction capability of the ANFIS-FCM method over the empirical correlations for the most cases.

... They found that droplet dissolution was a significant issue, which made it impossible to obtain a steady state droplet distribution at low phase fractions, while at higher phase fractions (φ > 0.2), despite breakup, most droplets coalesce to form a single connected region with multiple smaller satellite droplets. Increasing the resolution of the Kolmogorov scale remedied droplet dissolution to some extent, and a log-normal droplet distribution was shown from transient simulations, as has been experimentally found for turbulent liquid-liquid dispersions (Pacek et al. 1998;Lovick et al. 2005). The multiphase energy spectra could not be reproduced due to spurious currents which caused unphysical energy gain at high wavenumbers, whose magnitude was found to be close to the turbulent velocity scale u . ...

We perform direct numerical simulations (DNSs) of emulsions in homogeneous, isotropic turbulence using a pseudopotential lattice-Boltzmann (PP-LB) method. Improving on previous literature by minimizing droplet dissolution and spurious currents, we show that the PP-LB technique is capable of long, stable simulations in certain parameter regions. Varying the dispersed phase volume fraction $\phi$, we demonstrate that droplet breakup extracts kinetic energy from the larger scales while injecting energy into the smaller scales, increasingly with higher $\phi$, with the Hinze scale dividing the two effects. Droplet size ($d$) distribution was found to follow the $d^{-10/3}$ scaling (Deane & Stokes 2002). We show the need to maintain a separation of the turbulence forcing scale and domain size to prevent the formation of large connected regions of the dispersed phase. For the first time, we show that turbulent emulsions evolve into a quasi-equilibrium cycle of alternating coalescence and breakup dominated processes. Studying the system in its state-space comprising kinetic energy $E_k$, enstrophy $\omega^2$ and the droplet number density $N_d$, we find that their dynamics resemble limit-cycles with a time delay. Extreme values in the evolution of $E_k$ manifest in the evolution of $\omega^2$ and $N_d$ with a delay of $\sim0.3\mathcal{T}$ and $\sim0.9\mathcal{T}$ respectively (with $\mathcal{T}$ the large eddy timescale). Lastly, we also show that flow topology of turbulence in an emulsion is significantly more different than single-phase turbulence than previously thought. In particular, vortex compression and axial straining mechanisms become dominant in the droplet phase, a consequence of the elastic behaviour of droplet interfaces.

... Liquid-liquid emulsions have been the subject of numerous experimental (Berkman & Calabrese 1988;Pacek et al. 1998;Lovick et al. 2005) and computational studies (Perlekar et al. 2012;Skartlien et al. 2013;Komrakova et al. 2015;Scarbolo et al. 2015;Dodd & Ferrante 2016). The computational studies can be broadly categorized as forced homogeneous isotropic turbulence (Perlekar et al. 2012;Skartlien et al. 2013;Komrakova et al. 2015), decaying homogeneous isotropic turbulence (Dodd & Ferrante 2016) and turbulent wall flows (Scarbolo et al. 2015). ...

We simulate the flow of two immiscible and incompressible fluids separated by an interface in a homogeneous turbulent shear flow at a shear Reynolds number equal to 15200. The viscosity and density of the two fluids are equal, and various surface tensions and initial droplet diameters are considered in the present study. We show that the two-phase flow reaches a statistically stationary turbulent state sustained by a non-zero mean turbulent production rate due to the presence of the mean shear. Compared to single-phase flow, we find that the resulting steady state conditions exhibit reduced Taylor microscale Reynolds numbers owing to the presence of the dispersed phase, which acts as a sink of turbulent kinetic energy for the carrier fluid. At steady state, the mean power of surface tension is zero and the turbulent production rate is in balance with the turbulent dissipation rate, with their values being larger than in the reference single-phase case. The interface modifies the energy spectrum by introducing energy at small-scales, with the difference from the single-phase case reducing as the Weber number increases. This is caused by both the number of droplets in the domain and the total surface area increasing monotonically with the Weber number. This reflects also in the droplets size distribution which changes with the Weber number, with the peak of the distribution moving to smaller sizes as the Weber number increases. We show that the Hinze estimate for the maximum droplet size, obtained considering breakup in homogeneous isotropic turbulence, provides an excellent estimate notwithstanding the action of significant coalescence and the presence of a mean shear.

... Lovick et al. (2005) estudiaron la distribución de tamaños de gota utilizando agua y keroseno en un tanque agitado. Los autores encontraron que el tamaño de gota promedio a la altura del impulsor decrece con la velocidad del impulsor. ...

The bioconversion of the bicyclic ketone ((±)-cis-bicycle [3.2.0] hept-2-in-6-one) to their
corresponding lactones (Baeyer-Villiger reaction) can be performed using whole-cell using the strain Escherichia coli TOP10 pQR239, which synthesizes the intracellular enzyme cyclohexanone monooxygenase (CHMO). This type of biotransformation is favored with the addition of a second liquid phase (ionic liquid) using a three phase (aqueous-ionic liquid-air) partitioning bioreactor (TFPB) by increasing the specific productivity of the process. In this review, we summarize the most important aspects of the involved phenomena that occur in a BTPF. The biotransformation rate of the bicyclic
ketone taking into account both substrate inhibition and inactivation by excess of oxygen can be predicted through the development of micro kinetic model of the reaction. Hydrodynamic studies in the three-phase system during the bioconversion of the bicyclic ketone, mathematical correlations can be obtained to express the oxygen transfer coefficient and the Sauter mean diameter, d32, in terms of operation variables (agitation, aeration, and dispersed phase fraction). The development of a mathematical model that considers mass transfer phenomena, hydrodynamics, as well as the bioconversion kinetics and cell deactivation makes it possible to predict and describe the biotransformation of bicyclic ketones to lactones. Carrying out a regime analysis in the partitioning bioreactor through the determination of the characteristic times of the involved mechanisms defines the limiting step of the system. Finally through the integration of all these elements, new strategies can be developed for scaling up or down, which serve as a basis to scale this type of bioprocesses.

... The literature reports studies of the possible applications of the cited reflectance measurement techniques. Lovick et al. [17] and Yusoff [18] obtained accurate measurement for diameter of liquid droplets using the 3D ORM and FBRM technologies, respectively. Authors such as Li et al. [19] achieved good results recently using an FBRM probe to measure micron-sized bubbles. ...

Gas-liquid separation is a very common process in industrial plants, where often high Efficiency of Gas Separation (EGS) is demanded. In the upstream oil industry, gas separation is also crucial for the proper operation of Electrical Submergible Pumps (ESP). Recent studies report the excellent performance of a gravitational separator known as Inverted-Shroud separator (IS-separator). Laboratory results indicate that this kind of separator can achieve total gas separation for a wide range of operation conditions under continuous two-phase flow. The original intent is to use the IS-separator for downhole gas separation in oil production wells. However, by account of its simple design and relatively compact size, it may be suitable for using in industrial plants. In this paper, we present the IS-separator in details, including geometry characteristics and phenomenology. The equipment performance is also discussed. In addition, the challenges related to the deep understanding of the gas separation process inside the IS-separator and the possible practical solutions are outlined.

... However, it is close to the ratio of 3.6 reported by Simmons and Azzopardi (2001) also obtained in oilwater pipe flow. Other works performed in liquid-liquid dispersions in stirred tanks show d max / d 32 ratios from 1.8 to 4 ( Calabrese et al., 1986;Lovick et al., 2005;Sprow, 1967 ), which are in the range of the present findings. As aforementioned, residual water contents always below 13% of the total injected water volume fraction were measured upon exiting the oil-water separator for each experimental run. ...

Oil-water dispersed flows produced at valves or restrictions are very common in industry. For example, crude oil desalting processes normally use dispersion valves to mix the dilution water. In this case, the knowledge of dispersed droplet sizes is crucial for the design and optimum control of the process. However, little work has been devoted to characterize and model droplet sizes produced in industrial-type valves. The present work studies water droplet sizes produced by the passage of oil-water flow through a globe valve mounted in a large scale flow loop of 0.1 m internal diameter. Experiments were performed under different pressure drops across the globe valve, and different oil and water flow rates. Produced water droplet sizes were measured in situ downstream from the globe valve location using particle video microscopy. Droplet sizes were compared with theoretical scales for turbulent and viscous break-up. The effect of the volume fraction of dispersed phase on droplet size was also examined. In addition, droplet size distributions were analyzed and fitted using known statistical functions.

... Bubble size measurements were performed with the use of the three-dimensional Optical Reflectance Measurement (3D ORM) technique. Lovick et al. (2005) reported good results with this type of measurement technique in comparison with data acquired with a high-speed camera coupled with an endoscope, with which droplet sizes were acquired in an oil-water agitated mixture. In the present work this was similarly performed with an ORM IDAS (In situ Droplet Aerosol Analysis System based on ORM with spiral vertical moving selective focus), provided by SEQUIP TM . ...

The Inverted-Shroud (IS) separator is a promising solution for gas-liquid separation in the oil and gas industry. The IS separator combines high gas separation efficiency with the inexistence of moving mechanical parts. It consists of a closed-end tube (shroud) located between the production pipe (tubing) and the well casing. Experimental studies on the IS separator suggest that gas separation efficiency is affected by air entrainment inside the apparatus, where liquid in a free-surface flow impacts the internal liquid surface and carries with it gas that is entrained and disperses into bubbles due to turbulent kinetic energy dissipation. According to phenomenological models from the literature, the prediction of the energy dissipation rate correlates with the entrained bubble size, among other factors. The purpose of this paper is to study the gas entrainment phenomenon on the IS separator. First, a review of the current state-of-the-art research for a similar geometry is presented. Then, a comparison between experimental bubble size and predictions is performed. Bubble size distributions were obtained with a 3-D Optical Reflectance Measurement (ORM) probe installed in an experimental IS-separator test section. Different methods of average diameter (Sauter, mean, median and maximum size) are used to determine which best adjusts to the models' predictions. Results will improve the understanding of the IS separator and make possible the enhancement of its current modeling.

... An empirical fit was found to give reasonable agreement between the FBRM and PVM mean sizes measured for droplets with an average error less than 20%. Cull et al. (2002) and Lovick et al. (2005) employed an 3D optical reflectance measurement (ORM) technique, similar in operation to the FBRM and the 2D optical reflectance measurement in a liquid-liquid biocatalytic reactor. ...

... Particularly, the FBRM was found to undersize droplets in an emulsion. Cull et al. (2002) and Lovick et al. (2005) employed a 3D optical reflectance measurement (3D-ORM) technique, similar in operation to the FBRM and the 2D optical reflectance measurement (2D-ORM) in a liquid/liquid biocatalytic reactor successfully. ...

Despite the extensive literature dealing with both the fluid dynamic and the interface science aspects, the dispersion of immiscible liquids remains one of the most difficult and least understood mixing problems. Minor changes in the chemical composition of the system drastically affect the performance of such systems. Therefore, an improved understanding of the evolution of liquid/liquid dispersions is a key factor in operation, control and optimization of these processes. This becomes even more complex with the increasing geometrical diversity of agitated reactors. Growing markets and economies demand higher production rates. Limits in space and transportation have changed the outfit of the used mixing vessels. The height (H) of a reactor is increasing with constant tank diameter (T). H/T ratios of 2.0 or 3.0 are common today and ratios over 4.0 are expected.
The aim of this study is to develop and validate a model which will be able to predict system changes and intensification possibilities of stirred liquid/liquid breakage dominated processes. The dominance of breakage is achieved by the use of high surfactant concentrations clearly above the critical micelle concentration.
As a means of process intensification, dispersed phase fraction and reactor volume is increased in order to enhance productivity while the used power is decreased. Moreover, this is done exemplary for four different organic solvents. Based on simulation results of the used population balance equation (PBE), the process volume was increased by increasing the H/T ratio from 1.0 to 5.0 and the dispersed phase fraction from 25 to 45%. Additionally the used power was decreased by 20%. For the power decrease the impeller was changed from a single stage retreat curve impeller into a multi stage flat blade impeller. That lead to an overall process intensification in terms of an increase of the power related product by a factor of six.
All simulation results have been evaluated with lab-scale experiments, using a photo optical in-situ method, to size the drops. Selected experiments have been repeated at pilot plant scale to evaluate also the scale-up capacity of the used model. The detailed flow field was investigated using computational fluid dynamics to obtain circulation flow streams and energy dissipation rates throughout the reactors.
Furthermore, fundamental research was carried out on breaking single drops. These insights lead to a new model of a breakage rate, which is a sub-model within the PBE framework. This new model is able to predict the process intensification in terms of increasing reactor volume and reducing power input by using different impellers for the four investigated phases. The prediction of process changes by changing the dispersed phase fraction is possible until catastrophic phase inversion occurred. The model is not able to predict this dramatic process failure at the critical value of the dispersed phase fraction.
To overcome such failures a dynamic tracking of the evolving drop size distribution has been developed. The used endoscope probe allows real time recording of two dimensional images of the drops. A measurement of drop size distribution in the size range of 5 to 5000 µm is provided fully automated with computation times below 3 min per data point. Manual evaluation of the drops on the images was used to quantify the accuracy of the image algorithm software. The results showed a very good agreement between manually and automatically determined values, as long as the particle shapes stay spherical. Even the high concentrations of 45% can be investigated with the method. If the particle shape is irregular in comparison to spheres, the automated recognitions failed to measure the real particle size.

... Based on the above theories, several semi empirical correlations have been presented to predict particle size in liquid-liquid dispersions as a function of the dimensionless numbers (Lovick et al., 2005;Qi et al., 2015). For a viscous dispersed phase, Calabrese et al. (1986) based on a large number of experimental data presented the following correlation which takes into account the effect of viscous stresses to resist break up of the dispersed phase droplets in agitated tanks. ...

A correlation, based on fluid mechanics, has been investigated for the mean particle diameter of crosslinked dextran microgels (CDMs) prepared via a water/oil emulsification methodology conducted in a single-stirred vessel. To this end, non-dimensional correlations were developed to predict the mean particle size of CDMs as a function of Weber number, Reynolds number and viscosity number similar to ones introduced for liquid–liquid dispersions. Moreover, a Rosin–Rammler distribution function has been successfully applied to the microgel particle size distributions. The correlations were validated using experimentally obtained mean particle sizes for CDMs prepared at different stirring conditions. The validated correlation is especially applicable to medical and pharmaceutical applications where strict control on the mean particle size and size distribution of CDMs are extremely essential. (Figure presented.)

... Immiscible and miscible liquid-liquid dispersions in mixing tanks are the key processes in various industries such as chemical, biotechnological, pharmaceutical, and food processing industries [1,2]. Mixing generates essential interfacial areas to support mass and heat transfer between phases [3,4]. ...

The mixing efficiencies of impellers vary according to their designs. In this work, the effects of 6-curved-blade impellers of different curvature angles and central disk sizes on the reaction in a stirred vessel were investigated and the results were compared to that of a Rushton Turbine. The impeller efficiency was defined by the ratio of reaction rate to power consumption, (rA/P). The experiments were performed at the rotational speeds of 5, 6 and 7 rps. The interaction among the experimental parameters was investigated using Response Surface Methodology. The rA/P values were found to decrease with increasing curvature angles. The lowest rA/P value was obtained for the impeller with curvature angle of 140°. The result showed that the results for Rushton turbine was relatively low compared to curved-blade impellers and increase in central disk size did not significantly affect rA/P. In conclusion, curved-blade impellers were more economically efficient than Rushton turbine.

... An empirical fit was found to give reasonable agreement between the FBRM and PVM mean sizes measured for droplets with an average error less than 20%. Cull et al. (2002) and Lovick et al. (2005) employed an 3D optical reflectance measurement (ORM) technique, similar in operation to the FBRM and the 2D optical reflectance measurement in a liquid-liquid biocatalytic reactor. ...

The design of stirred tank reactors for liquid-liquid dispersions usually requires expensive experimental investigations. Complete models for the drop size distribution as a function of power input, material and process parameters are rare and relatively inaccurate. Therefore, it is necessary to analyze drop size distributions for accurate modelling. Furthermore, industrial applications, like e.g., suspension polymerization processes, require a distinct average drop diameter and a small standard deviation of the distribution. For the controlling of such systems fast acquisition of information is needed but difficult to obtain. While many users are confronted with both requirements an adequate measurement technique for all applications is needed but has not yet been developed. The aim of this publication was to give an overview for already existing measurement techniques and compare the most important properties and the resulting data to simplify the decision process for the selection of the "right" measurement technique according to the used system. The measurement applications are divided into three main groups: sound, laser and photo based techniques. The Lasentec FBRM®, the 2D-ORM®from Messtechnik Schwartz GmbH and the FBR-sensor which all give online and in-situ information and furthermore an in-house developed endoscope technique are discussed in detail. They have been tested for various applications in a stirred tank and the results are compared. It is clearly shown that laser based methods give only qualitatively good but fast results for the tested system toluene-water. Quantitatively accurate experimental results of the endoscope technique gave a good base for testing and developing numerical models for even transient behaviour of drop size distributions.

... The Sauter mean droplet diameter (d 32 ) was measured using a 3D optical reflectance measurement (ORM) particle-size analyser (MTS, Messtechnik Schwartz) [20]. A total of 2000 droplet diameter (d 1 ) determinations were performed automatically for each measurement, enabling the calculation of d 32 according to: ...

... It was shown experimentally, for instance, by Lovick et al. 55 and Pacek et al. 56 that the DSD of a liquid-liquid dispersion formed under turbulent flow conditions is close to a log-normal distribution. The DSD for the case with g K ¼ 2:5 [lu] and capillary number Ca ¼ 0:004 is shown in Figure 13a at time instant t ¼ 117:9s K . ...

A numerical approach is developed to gain fundamental insight in liquid-liquid dispersion formation under well-controlled turbulent conditions. The approach is based on a free energy lattice Boltzmann equation method, and relies on detailed resolution of the interaction of the dispersed and continuous phase at the microscopic level, including drop breakup and coalescence. The capability of the numerical technique to perform direct numerical simulations of turbulently agitated liquid-liquid dispersions is assessed. Three-dimensional simulations are carried out in fully-periodic cubic domains with grids of size 1003-10003. The liquids are of equal density. Viscosity ratios (dispersed phase over continuous phase) are in the range 0.3 to 1.0. The dispersed phase volume fraction varies from 0.001 to 0.2. The process of dispersion formation is followed and visualized. The size of each drop in the dispersion is measured in-line with no disturbance of the flow. However, the numerical method is plagued by numerical dissolution of drops that are smaller than 10 times the lattice spacing. It is shown that to mitigate this effect it is necessary to increase the resolution of the Kolmogorov scales, such as to have a minimum drop size in the range 20-30 lattice units [lu]. Four levels of Kolmogorov length scale resolution have been considered\eta_K=1, 2.5, 5 and 10 [lu]. In addition, the numerical dissolution reduces if the concentration of the dispersed phase is increased. This article is protected by copyright. All rights reserved.

... Optical Reflectance Measurement (ORM) techniques are able to show the change in drop size distribution for various power inputs, although they are not able to measure the exact size of the drops. Nevertheless, they are successfully used in control of processes [17,18]. In this study, the Sauter mean drop diameter (d 32 ) was measured using a 3D ORM particle size analyser (Messtechnik Schwartz, Germany). ...

In this study we developed a pseudo heterogeneous mathematical model for the oxidation of bicyclic ketone bicyclo[3.2.0]hept-2-to-6-one to bicyclic lactone (1S,5R)-(-)-2-oxabiciclo[3.3.0]oct-6-en-3-ona using whole cells of E. coli strain TOP10 pQR239 in a three-phase partitioning bioreactor (TPPB). The pseudo heterogeneous TPPB model accounted for mass transfer mechanisms occurring in the air–water and water–ionic liquid phases along with bioconversion and loss of cell viability caused by ionic liquid – trioctylmethylammonium bis(trifluoromethylsulfonyl)imide ([OMA][BTA]). The development of the model was based on reactor engineering principles and, hence, experiments with and without bioconversion were carried out in order to characterize thermodynamic (Kps and Kpp), transport (ks, kp and A) and kinetic (ki and ) parameters along with the loss of cell viability (kin) parameter for the mechanisms involved. The model described adequately the bioconversion of experimental data with two different ionic liquid fractions, namely 5% (v/v) and 12.5% (v/v). A parametric sensitivity analysis of the model was conducted to obtain information on the effect of oxygen transport rate on bioconversion. The development of the TPPB model led to the following findings: (i) 5% (v/v) ionic liquid fraction caused less cell deactivation; (ii) presence of ionic liquid decreased the oxygen transport rate; and (iii) a higher oxygen transport rate lead to a higher bioconversion but also cell inactivation.

... Desnoyer et al. have measured the droplet size distribution ranging from 1 m to 2 mm of nickel dioxide solution at the density of 1020 kg/m 3 and hydrogen chloride solution at the density of 1050 kg/m 3 in the mixing stirred vessel using a laser granulometer [5]. Lovick et al. have counted the rule of droplet size distribution mainly ranging from 20.6 m to 230 m in kerosene-water two-phase flow using a light back scattering technique and a high-speed video recorder in an agitation tank with the oil phase fractions in a range of 5-10% [6]. Lovick and Angeli [7], Al-Wahaibi and Angeli [8] have carried out a study on the droplet size distribution using a stainless steel tube with a diameter of 38 mm and at the oil phase viscosity of 5.5 and 6 mPa⋅s, respectively. ...

The conventional gathering and transportation mode of heating the produced fluid of oil wells with hot water or steam may result in excessive energy consumption. In order to perform the unheated transportation, the idea of hydraulic suspension transport of the gelled crude oil is proposed based on the actual production of Daqing Oilfield, and the experimental system is established to test characteristics of oil particle distribution which have an important effect on the hydraulic suspension transportation. In the experiment, the image of gelled crude oil particle distribution was obtained in a horizontal pipe with inner diameter of 0.053 m, and then the law of particle distribution was investigated by the theoretical model. The results showed that the gelled crude oil hydraulic suspension transport could be achieved without any chemical reagent when the gelled crude oil was transformed into particles and dispersedly suspended in water. The results also showed that the gelled oil particles of 0-4 mm in size accounted for 92% or more of all particles, and the percentage of gelled crude oil particles of a size of 4 mm gradually increased with the increasing mixed flow rate.

... The effect of dispersed phase holdup on the DSD mainly depends on the physical properties, hydrodynamics and the breakage and coalescence behavior of liquid-liquid systems [23,25,31,49]. ...

Droplet size distribution (DSD) and mean droplet size of chloroaluminate ionic liquid (IL)–heptane dispersion in a stirred vessel were investigated as a model system for the composite ionic liquid/hydrocarbon system applied in the commercial isobutane alkylation process. Focused beam reflectance measurement was used to identify the phase inversion, and particle video microscope (PVM) was employed to measure the DSD of IL-continuous dispersions. The effects of agitation speed and dispersed phase holdup on the DSD and mean droplet size were investigated. Phase inversion was found at a heptane hold-up of just above 0.5. The observed DSDs could be described well by logarithmic normal distributions in spite of variable operating conditions. The DSDs appeared to be more or less independent of the dispersed phase holdup (ϕheptane ⩾ 0.3) at the same agitation speed. A semi-empirical correlation based on Shinnar theory was proposed to predict the droplet mean size in the impeller region for the IL-continuous dispersion, and the predicted value agreed well with the experimental data. This work will be a good basis to predict the dispersion behavior under process conditions.

... The effects of drop size and size distribution on the performance of an extraction column are the most important hydrodynamic characteristics, because under steady operating conditions, drop size is related to the interfacial area by: investigator found a different distribution function suitable to describe his data [Lovick, et al. 2005]. No attempts have been done till now to study the effect of drop size distribution with mass transfer conditions on the performance of RTL contactor. ...

The present work studied the drop size distribution in the RTL contactor using two liquid-liquid systems, xylene-acetone-water and kerosene-acetone-water. Process variables studied were: rpm (10-50 min -1), continuous phase flow rate (4-12 l/h), dispersed phase flow rate (4-12 l/h), and concentration (0.1-0.5 mole/l). It was found that Sauter mean drop diameter (d 32) decreased with increasing rpm and continuous phase flow rate, and decreasing dispersed phase flow rate and concentration. An empirical correlation, with a correlation coefficient equal to (92.6%), relating d 32 with process variables was developed having the form:

... It is further assumed that the extension of the laser beam is much smaller than the particle size. For instruments with a detection size limit of 1 µm it can be concluded that the laser beam spot has an approximate diameter of 0.6 to 2 µm [Spa94,Lov05,MTS07]. However, as the laser propagates into the liquid, the laser broadens (signal gets weaker, decrease in laser intensity) up to one order of magnitude within the measurement zone [Ruf00,Pon06,Kai07,Kai08]. The resolution of the instrument towards particles less or equal to the laser diameter remains unclear [Spa94,Pon06]. ...

Dissertation Zur Erlangung des akademischen Grades Doktoringenieur (Dr.-Ing.) genehmigt durch das Zentrum für Ingenieurwissenschaften der Martin-Luther-Universität Halle-Wittenberg als organisatorische Grundeinheit für Forschung und Lehre im Range einer Fakultät (75 Abs.

This study investigated the effect of dual impeller geometry on the droplet size in the suspension-PVC (S-PVC) polymerization process. To simulate the process, 1,2-dichloroethane was used as a dispersed phase, because it has been used to replace the toxic vinyl chloride monomer (VCM). Using a borescope method, a droplet size was measured for a biphasic liquid system, and the Sauter mean diameter increased by 46.5% as the upper paddle impeller was replaced by 20° pitched paddle. It also increased when the impeller diameter and the blade width increased. Considering this effect, a geometrical factor (F) was revised, and a calculated maximum energy dissipation rate was used for establishing the Sauter mean diameter correlation. The proposed correlation can estimate the Sauter mean diameter within ±20% error, and one can predict the normality of the polymerization under specific impeller geometry using this correlation.

Liquid-liquid mixings in stirred tanks are commonly found in many industries. In this study, we performed computational fluid dynamics (CFD) modeling and simulation to investigate the liquid-liquid mixing behavior. Furthermore, the population balance model (PBM) was used to characterize the droplet size distribution. The PBM model parameters were calibrated using the experimental data of droplet sizes at different agitation speeds. Additionally, we employed the steady-state Sauter mean droplet size to validate the developed CFD-PBM coupled model at different dispersion phase holdups. Then, the validated CFD-PBM coupled model was employed to evaluate the role of impeller structural parameters on the liquid-liquid mixing efficiency based on a user-defined mixing index. It was found that the position of impellers significantly affects the mixing efficiency, and an increase in stirring speed and the number of impellers improved the mixing efficiency.

The mixing of immiscible oil and water by a pitched blade turbine in a cylindrical vessel is studied numerically. Three-dimensional simulations combined with a hybrid front-tracking/level-set method are employed to capture the complex flow and interfacial dynamics. A large eddy simulation approach, with a Lilly–Smagorinsky model, is employed to simulate the turbulent two-phase dynamics at large Reynolds numbers $Re=1802{-}18\ 026$ . The numerical predictions are validated against previous experimental work involving single-drop breakup in a stirred vessel. For small $Re$ , the interface is deformed but does not reach the impeller hub, assuming instead the shape of a Newton's Bucket. As the rotating speed increases, the deforming interface attaches to the impeller hub which leads to the formation of long ligaments that subsequently break up into small droplets. For the largest $Re$ studied, the system dynamics becomes extremely complex wherein the creation of ligaments, their breakup and the coalescence of drops occur simultaneously. The simulation outcomes are presented in terms of spatio-temporal evolution of the interface shape and vortical structures. The results of a drop size analysis in terms of the evolution of the number of drops, and their size distribution, is also presented as a parametric function of $Re$ .

In this work, steady-state droplet size distributions in a DN300 stirred batch vessel with a Rushton turbine impeller are investigated using an insertion probe based on the telecentric transmitted light principle. High-resolution droplet size distributions are extracted from the images using a convolutional neural network for image-analysis in order to investigate the influence of impeller speed and phase fraction (up to 50 vol.-%). In addition, Sauter mean diameters were calculated and correlated with two semi-empirical approaches, while the standard approach only accomplished 5.7% accuracy, and the correlation of Laso et al. provided a relative mean error of 4.0%. In addition, the correlated exponent in the Weber number was fitted to the experimental data of this work yielding a slightly different value than the theoretical (−0.6), which allows a better representation of the low coalescence tendency of the system, which is usually neglected in standard procedures.

Stirred tanks have been widely used for liquid-liquid mixing in the chemical industry. In the present work, the influences of flexible impeller on the hydrodynamic characteristics and droplet size distribution of oil in water dispersions were investigated using experimental and numerical approaches. The one-way Fluid Structural Interaction (FSI) analysis was carried out to model the geometry of the flexible impeller used in CFD simulations. Computational fluid dynamics (CFD) was coupled with the population balance model (PBM) to simulate the liquid-liquid dispersion process in the stirred tank. The numerical model was validated by the experimental results. Compared with the rigid impeller, the flexible impeller generates a more uniform distribution of turbulent kinetic energy and turbulent dissipate rate in the stirred tank. Compared with the case of rigid impeller, the region indicating the trailing vortices behind the impeller blades was significantly reduced by the flexible impeller. It was found the flexible impeller with a smaller elastic modulus can result in narrower droplet size distributions in turbulent dispersion process. Its peak value of the number fraction frequency was in the range of 244-308μm while that was 308-388μm for the case of rigid impeller. It was found that most of the dispersed phase droplets occur in the turbulent inertial subrange regime. Related correlations between the Sauter mean diameter and local turbulent field were also obtained. The shape factor accounting for the effects of the impeller shape on the Sauter mean diameter was found to be dependent on the material property of the impeller blades, such as the Poisson's ratio and elastic modulus.

This study investigated the effect of multiple impeller designs and configurations on the Sauter mean diameter and the uniformity of droplet size in a 100 L scale stirred tank. By using a borescope installed inside the tank, droplet images of a highly turbid liquid-liquid system were captured even at high impeller speeds, and by adjusting the borescope position, it could be observed how the droplet size changed depending on the position. The area of the flow pattern produced by the impeller was taken as an impeller region, and it explained well the change in the droplet size due to the varying liquid phase volume and impeller spacing. In addition, the change of the Sauter mean diameter and the droplet size uniformity was also elucidated by the variation of the impeller diameter, blade angle, and number of impellers. All three parameters showed a decrease in the deviation between droplet sizes as they increased, but increasing the impeller diameter was the most effective in reducing the Sauter mean diameter itself.

Effects of secondary emulsification of field ASP flooding produced fluid by centrifugal pumps and electrostatic field on its oil/water separation performances were evaluated. Secondary emulsification of certain ASP flooding produced fluid with high surfactant content by centrifugal pumps drastically downgraded its oil/water separation performances, increasing the residual water-cuts after 24h static settling at 60°C respectively by 70% for original o/w emulsion and 1080% for w/o emulsion. Certain w/o emulsion produced by ASP flooding with high surfactant content was susceptible to secondary emulsification in high strength electrostatic field with up to 660% increase of residual water-cut after 24h static settling at 60°C. Mechanisms governing the effects of secondary emulsification on the oil/water separation performances of ASP flooding produced fluid were discussed. In comparison with produced fluid by water flooding and polymer flooding, the oil/water interfacial tension in ASP flooding produced fluid is much lower, making oil droplets in water and water droplets in oil significantly more sensitive to shear and high strength electrostatic field. Energy dissipation rate, surfactant content and the volume ratio of dispersed phase in the emulsion were identified to be the most important factors affecting the secondary emulsification of ASP flooding produced fluid in turbulent flow. Water in oil emulsion extracted from high water-cut o/w or w/o/w ASP flooding produced fluid was found to have significantly higher water phase surfactant content than the original crude oil emulsion and is much more susceptible to secondary emulsification in turbulent flow than the original reverse emulsion. Surfactant content, content of dual wettability micron and submicron sized particles and the strength of electrostatic field were found to be major factors affecting the secondary emulsification of ASP flooding produced w/o emulsion in AC electrostatic field. As interfacial tension between oil and water in w/o emulsion produced by ASP flooding can be 1 to 2 orders lower than that in the w/o emulsion produced by water flooding and polymer flooding, strength of electrostatic field in electrostatic treaters has to be optimized as per the surfactant content in the ASP flooding produced w/o crude oil emulsion they treat.

Novel experiments are designed to visually study the interphase mass transfer process of immiscible liquid-liquid system by using the planar laser-induced fluorescence (PLIF) method combined with the refractive index matching technique. Rhodamine B is chosen as the fluorescence tracer because it can dissolve both in the aqueous phase and the organic phase. The experimental results indicate that the mass transfer equilibrium time is negatively correlated to the agitation speed. For the Rushton disk turbine, the interphase mass transfer efficiency is enhanced as the impeller clearance increases from T/10 to 2T/5. Furthermore, the interphase mass transfer process is dampened and the equilibrium time increases continuously when the volume fraction of NaI solution increases from 2% to 50%. When the impeller eccentricity increases, the mass transfer equilibrium time increases correspondingly and its change trend in the immiscible liquid-liquid system is contrary to the trend of mixing time in single phase cases.

Stirred tanks are the common unit operations employed for liquid–liquid dispersion in many industries. In this study, computational fluid dynamic (CFD) coupled with population balance modeling (PBM) were applied to simulate mixing of water in crude oils in a stirred tank equipped with Rushton turbine. An Eulerian multiphase model and standard k-ε turbulence model were employed to simulate the flow field inside the tank. Experimental results were then used to validate simulation results. The effect of impeller speed, volume fraction of dispersed phase, and oil viscosity on droplet size distribution were investigated in this work. An increase in agitation speed was found to decrease the mean and Sauter mean diameter while increasing the homogeneity of the system. A wider droplet size distribution profile was achieved at higher oil volume fractions with almost no significant change in droplet size. Additionally, increasing the viscosity of the oil phase resulted in a gradual shift towards smaller droplets and narrower droplet size distribution. Liquid–liquid dispersions were occurred in two different turbulent regimes depending on the viscosity of continuous phase. Related correlations for each regime were obtained and the values of maximum droplet size were fitted with the dimensionless numbers.

Stirred tanks are commonly employed for liquid–liquid dispersion processes in many industries. This study investigated the effects of agitation speed, and oil type and volume fraction on the hydrodynamic characteristics and chord length distribution (CLD) of dilute oil in water dispersions in a stirred tank. Electrical resistance tomography (ERT) and a focused beam reflectance measurement (FBRM) instrument were used to assess the liquid–liquid flow and to measure droplet size inside the tank, respectively. We also simulated flow field via computational fluid dynamic and modeled droplet size distribution in a stirred tank by population balance modeling. An increase in agitation speed was found to decrease the mean and Sauter mean diameter while improving the homogeneity of the system. Wider distributions were observed at higher oil volume fractions, without a significant change in droplet size. Increasing the viscosity of the oil phase resulted in poor mixing, with a gradual shift towards smaller droplets. The effect of the dispersion process on droplet shape and deformation rate were also investigated using CLD results. Further shape analysis was performed using Python coding. Our results show that the shape of the droplets changed from sphere to spheroid with an increase in agitation speed. An increase in droplet deformation rate was also observed with an increase in the oil phase viscosity and a decrease in the interfacial tension between the two immiscible liquids.

The solubility of drugs is a crucial physicochemical property in the drug discovery or development process and for improving the bioavailability of drugs. There are various methods for evaluating the solubility of drugs including manual measurement methods, mathematical methods and smart methods. Manual measurement and mathematical methods have some defects which make the smart systems more reliable and important in this field. In this review, various instruments used for the solubility determination, along with the smart systems, have been discussed. Mechanism and applications of each method have been elaborated in detail. Moreover, unique characteristics as well as some limitations of discussed methods are also described.

Several studies have reported that the hydrodynamics are not affected by the biomass in multiphase partitioning bioreactors. This work aims to demonstrate the effect of biomass concentration (0, 1, 3 and 5 g L ⁻¹ ) on the oxygen mass transfer coefficients, the droplet size of the dispersed phase, power consumption and superficial tension in a multiphase partitioning bioreactor (ionic liquid-aqueous-air-biomass system). At a biomass concentration of 5 g L ⁻¹ , the oxygen mass transfer coefficient ( kLa ) increased by 55% (249 h ⁻¹ ) compared with the abiotic system (160 h ⁻¹ ). In the multiphasic system, the droplet size ( d32 ) decreased when the biomass concentration was increased, producing an increment in the mass transfer area of the dispersed phase. In addition, the power consumption decreased by 44 % compared to a previous report without biomass. Furthermore, the increment of biomass concentration decreased the superficial tension by up to 15 %. A biomass increment in a multiphase system not also increases product yield, but also enhances the bioconversion process. The results obtained suggest that it is obligatory to consider the effect of biomass concentration on hydrodynamic characterisation, design, scale-up and optimisation for improving the performance of biotechnological processes using multiphase bioreactors.

An experimental study was performed to discuss the effects of curvature angles and central disk sizes of 6-curved-blade impellers on mean drop size in an agitated vessel. A system with 1% oil in water in the presence of a surfactant solution was used. The effects of impeller speed on drop size were also investigated. Laser diffraction technique and RSM method was employed to measure and analyze data, respectively. Decreased curvature angles from 180° to 140° reduced the drop size up to 9%, 10%, and 10% at agitation speed of 5, 6 and 7 rps, respectively. Moreover, decrease in central disk size from 3/4D to 1/4D reduced the drop size up to 16%, 18% and 22% at the agitation speed of 5, 6 and 7 rps, correspondingly. Two mathematical models were suggested and the most significant parameters of each experimental design were identified through the Analysis of Variance.

Evolutions of drop/particle size and size distribution in liquid-liquid dispersions and suspension polymerizations of methyl methacrylate (MMA) were monitored by using an online optical reflectance measurement (ORM), and effects of operating parameters such as the agitation rate, concentration of poly(vinyl alcohol) (PVA) dispersant, and initial concentration of poly(methyl methacrylate) (PMMA) in MMA monomer on the Sauter mean diameter (d32) and size distribution of drop/particle were investigated. According to the variations of d32 of drops/particles with time, four characteristic particle formation stages can be identified for suspension polymerization process. The factors that lead to increase the rate of drop break up, such as increasing of concentration of PVA and decreasing of viscosity of dispersed phase, would postpone the particle growth stage. The d32 and size distribution breadth of drops/particles were significant increased when the liquid-liquid dispersions or suspension polymerizations were conducted at low PVA concentrations or MMA/PMMA solutions with high PMMA contents were used as the dispersed phase, in consistent with the scanning electron micrograph observation on final PMMA particles. It is clear that ORM can be effectively applied in online monitoring of size and size distribution of drops/particles in the liquid-liquid dispersions and suspension polymerizations.

The average droplet size and size distribution of 1-chlorobutane (ClBu)Water emulsions, as a model system of vinyl chloride (VCM)/water emulsions, were studied by acoustic spectroscopy on-line with an agitated laboratory reactor. The emulsions were stabilised with poly (vinyl alcohol-co-vinyl acetate) copolymers obtained by partial hydrolysis of poly (vinyl acetate). The effect of agitation speed, stirring time, concentration and hydrolysis degree of the copolymers, were investigated. A correlation between particle size and the ClBu/water interfacial tension could further be established.

To improve the dispersity of the emulsion phase in an emulsion liquid membrane (ELM) system, a modified rotating disc contactor (MRDC) was developed. Then, the Sauter mean diameter (d32) of emulsion drops and their drop size distribution were measured by photographic method and analyzed by an image processing program in MATLAB. The effects of rotating speed, flow ratio, total flow, stirring paddle width and surfactant concentration on the drop size and its distribution were studied. The results show that, with the increase in the rotating speed and the paddle width, the degree of turbulence was enhanced which led to the reduction in the drop size. Meanwhile, membrane breakage increased with the turbulent fluctuation, which resulted in the leakage of the internal phase. Fortunately, the membrane breakage could be prevented by a suitable increase in the surfactant concentration. In addition, the drop size decreased with the increase in the surfactant concentration. Besides, the increase in the emulsion phase flow obviously increased the drop size, whereas the increase in the continuous phase flow induced the entrainment of small drops. An empirical correlation for the prediction of the d32 was established with an average absolute relative error (AARE) of 4.1%. The drop size distribution based on the drop volume was accurately fitted with a normal distribution, and its probability density function parameters (α and β) were well predicted by the dimensionless correlations, with the AAREs of 4.2% and 5.9%, respectively. This journal is

The present work describes the dispersion process of a viscous fluid in water in a cylindrical vessel agitated at Re = 24,000. The formation of viscous filaments and other amorphous structures in turbulent conditions produced in the early stage, before oil drops saturate the continuous phase, is shown. The oil-phase evolution is followed with high-speed video recordings and compared with the flow pattern produce in the liquid bulk. The effects of four kinds of perturbations are identified: intermittences in the radial velocity, turbulent fluctuations, rotation, and the stretching. As a consequence, the viscous-phase experiences instabilities that include random deformation, elongation, hairpin filaments formation, folding, and fragmentation. In the final part of this study, a mechanism describing the drop size reduction has been proposed.

Laser-backscattering instruments, such as the focused beam reflectance measurement or threefold dynamical optical reflectance measurement, are promising tools to aid crystallization process development, allowing an in situ, real-time, and nondestructive measurement of particle size distributions. Besides the instrument principles, in detail geometrical and optical models are discussed which deconvolute the recorded chord length distribution. Emphasis is thereby laid on the influence of the suspension density on instrument recordings. The application of laser-backscattering devices for determination of kinetic constants is discussed and future directions and perspectives are given.

Experimental investigation was conducted on a liquid–liquid dispersion in a stirred vessel in which the effects of dispersed phase viscosity were studied. Different grades of silicone oils were used to create oil-in-water dispersion by using Rushton turbine as an impeller, and drop sizes were measured by laser diffraction technique. Dispersion with higher uniformity of drop sizes was produced at low viscosity and high impeller speed. The dispersed phase viscosity influenced the equilibrium Sauter mean diameter, d32 by contributing to drops stabilization. The decrement of d32 with an increase in impeller speed is larger for high dispersed phase viscosities. It shows the influence of number and size of drop fragments formed after drop breakup on the mean drop size. Correlations relating d32 and dispersed phase viscosity were proposed with an accuracy of more than 90% between the predicted and experimental values.

The effect of hydrodynamic conditions and physical properties on the characteristics of an emulsion dispersion in a stirred tank was studied in the presence of mass transfer, in an attempt to identify the parameters for Sauter mean diameter data correlation in emulsion liquid membranes. The organic phase of the water-in-oil emulsion contained the cadmium extractant Cyanex 302 and an emulsion stabilizer, and the dispersion continuous phase was an aqueous solution of phosphoric acid containing cadmium. Variations in emulsion viscosity and interfacial tension were achieved by changing the extractant and surfactant concentrations. The dispersion drop size distribution was measured from photographs taken with a high-speed video camera system attached to an endoscope. It was found that the emulsion physical properties, emulsion swelling, and operating conditions were interrelated in a complex way. However, emulsion swelling seemed to lump together the effect of some of the parameters on globule size, which made possible the development of a simple correlation for the Sauter mean diameter in terms of We-0.6, emulsion swelling percentage, and empirical correction factors for emulsion viscosity and dispersed-phase holdup. A modified Rosin−Rammler probability distribution function fitted well the globule size distribution data.

The square cross-section mixing tank is used in many industrial mixer-settler installations but its performance is poorly documented in comparison to the more traditional cylindrical mixing tank. As part of a programme to establish the characteristics of the square cross-section tank for liquid-liquid dispersions, measurements have been made of power consumption, hold-up and drop size. These measurements show the simple assumption that the square tank is equivalent to a baffled cylinder is not justified; in this paper the drop size characteristics of the square tank are described and compared with published data for the cylinder. Drop size and drop size distributions of organic drops dispersed in an aqueous continuum have been measured by a capillary withdrawal technique and some photographic studies are also reported. The measurements were made for dispersions produced in a 152 mm square cross-section tank agitated by a 51 mm Rushton turbine impeller. The variables of hold-up, impeller speed and fluid properties have been investigated and the results interpreted in terms of established theory for coalescing and non-coalescing systems. (A)

The drop size distribution is one of the most important characteristics of liquid–liquid dispersions. Several shapes of the distribution have been proposed by previous investigators. This work investigates the drop size distribution in very dilute (0.03% by volume) liquid–liquid dispersions over a wide range of rotational speeds, using different impellers with varying diameters and off-bottom clearances. Four impellers were used: one radial flow impeller (Rushton turbine); and three axial flow impellers [pitched blade turbine, HE3 turbine and fluidfoil (A310) turbine]. Drop sizes were measured using a phase Doppler particle analyzer (PDPA) in both the bulk and impeller regions in an agitated tank. It was found that the drop size distribution changes with an increase in the rotational speed. Typically, four types of drop size distribution evolve with increasing rotational speed: long tail, double peak, skew and skew-normal distribution. A new scaling parameter; the product of average power input per unit mass and ND2 is proposed to define the regions where the four types of drop size distribution occur.

Drop size distributions produced for the kerosene water system in a 30 cm dia. stirred vessel dispersed phase hold-up values of 0·05,0·10 and 0·20 and stirrer speeds of 250, 300, 350 and 400 rpm are examined. Evidence is presented to suggest that this system does not coalesce. The data in the form of a normal distribution are in close agreement with previous work. A detaild examination, however, reveals a bimodal form of distribution. No effect of impeller speed is evident although the dispersed phase hold-up is found to influence the shape of the bimodal distribution.RésuméLes auteurs examinent les distributions de la taille des gouttes pour le système kérosèneeau dans un bac agité de 30 cm de diamètre, pour des valeurs de la retenue de phases dispersées de 0·05, 0·10 et 0·20 et pour des vitesses de l'agritateur de 250, 300, 350 et 400 rpm.It apparait que ce système ne fusionne pas. Les données sous forme d'une distribution normale correspondent étroitement à une étude antérieure. Un examen détaillé permet cependant de reconnaître une forme bimodale de distribution. La vitesse du rotor ne semble pas avoir d'effet bien que la retenue des phases dispersées influencer la forme de la distribution bimodale.ZusammenfassungEs werden die Tropfengrössenverteilungen untersucht, die für das Kerosin-Wassersystem in einem 30 cm Durchmesser Rüchrkessel bie Holdup-Werten der dispergierten Phase von 0,05, 0,10 und 0,20 und Rührergeschwindigkeiten von 250, 300, 350 und 400 Upm gebildet werden. Es wird Beweismaterial dargelegt, das darauf hindeutet, dass dieses System nicht koalesziert. Die Messwerte, in der Form einer normalen Verteilung, stimmen gut mit früheren Arbeiten überein. Eine detaillierte Untersuchung zeigt jedoch eine bimodale Form der Verteilung. Die Rührergeschwindigkeit scheint ohne Einfluss zu sein, doch wird festgestellt, dass der Holdup der dispergierten Phase die Form der bimodalen Verteilung beeinflusst.

Experiments were carried out to determine the effect of the dispersed phase concentration,.φ, on the Sauter mean drop diameter of xylene in water in a mechanically agitated vessel. In the course of the experiments, the rheology of the dispersions was also measured. It was found that the viscosity of the dispersion exhibited strong viscous non-Newtonian characteristics for dispersed phase concentrations greater than approximately 50% by volume. The increase in the apparent viscosity of the dispersion with increasing dispersed phase concentration changed the flow condition in the tank from turbulent regime to transitional and laminar flow regime. The Sauter mean drop diameter measured as a function of the dispersed phase was successfully interpreted in terms of a turbulent eddy-drop breakage mechanism for φ < 50% and by a boundary layer drop breakage model for φ > 50% by volume.

Drop size measurements were made in the break-up zone at the tip of three 6 bladed disc turbines of different geometries in a 0·30 m dia. vessl. Three systems kerosene, methyl iso-butyl ketone (MIBK) and n-butanol at a volumetric fractional hold-up of 0·05 in water were examined. Power input and circulation time characteristics were determined and a new dimensionless group (ϵ−tc/T) is proposed to account for the effect of geometric parameters in the correlation of the drop size.

The distribution of drop diameters produced in the vicinity of a turbine impeller has been determined in a dilute iso-octane + salt water emulsion, using an electronic particle counter. The experimental distribution data have been fitted to the Schwarz—Bezemer equation with good agreement. The results are compared with theoretical predictions based on the existence of local isotropy near the impeller.RésuméOn a déterminé la répartition de la roue motrice des diamètres des gouttelettes produites dans le voisinage de la zone motrice d'une turbine, dans une émulsion diluée d'iso-octane + eau salée, en utilisant un compteur électronique de particules. Les données expérimentales de la distribution s'accordent avec l'équation de Schwarz—Bezemer. Les résultats sont comparés avec les prévisions théoriques basées sur l'existence d'une isotropic locale près de la roue motrice.ZusammenfassungDie Verteilung der Tropfendurchmesser im Bereich eines Turbinenpropellers wurde in einer verdünnten Iso-Oktan- und Salzwasser-Emulsion unter Anwendung eines elektronischen Teilchenzählers bestimmt. Die versuchsweisen Verteilungsdaten entsprechen der Schwarz—Bezemer Gleichung gut. Die Ergebnisse werden mit theoretischen Vorhersagen aufgrund lokaler Isotropie im Bereich eines Propellers verglichen.

The scanning laser microscope (Par-tec) has been used to measure drop size distributions of oil-water mixtures in both batch and online operations and the performance of the instrument has been assessed. Measurements have been made to determine the influence of the shear produced by pipes and fittings on the dispersion characteristics of oil-water mixtures. The influence of oil composition has also been investigated.
The Par-tec has been found to give reproducible mean chord data for oil-in-water and water-in-oil dispersions in both a batch mixing process and on-line at various dispersed phase concentrations and for both clear and optically dense oil systems. However, there were some important limitations noted. Firstly, the drop size measured by the instrument is not a drop diameter as measured by most other particle sizing instruments but a drop chord length; secondly, the drop size was found to vary with the focal length of the instrument.
Changes in drop size due to changes in process conditions can easily detected using the Par-tec instrument. In this work the effect of horizontal pipe length and number of bends was investigated. It was found that the drop size increased with increasing horizontal pipe length due to drop coalescence. The drop size was also found to increase as the number of bends in the pipe was increased in a 1.5" I.D. U-pipe fitting, indicating that the energy dissipated in the system is not enough to outweigh the effect of coalescence due to the pipe length.
The sequence of pipe fitting was found to influence the nature of dispersion. The value of the mean chord at the exit of a needle valve followed by a small U-pipe fitting was not the same as the mean chord for the U-pipe followed by the needle valve. The main advantages of this instrument are its flexibility (it can be easily inserted into process streams in most pipes and vessels) and its ability to obtain data in two phase systems with high dispersed phase concentrations and/or an optically dense continuous phase.
Introduction
The detailed design of equipment for the removal of water from oil (or indeed any kind of liquid-liquid separation) requires knowledge of the dispersion properties such as the droplet size distribution, dispersed phase concentration, phase continuity and the interfacial tension of the system. An understanding of the nature of heterogeneous flow through pipes and fittings would enable more accurate predictions of conditions at a separator inlet to be made and hence the equipment could be designed in a much more scientific way. However in order to make significant improvements in performance, it is necessary not only to have information about the drop size distribution at the separator inlet but also to determine the drop size profile along vessels of different design. At present this data is usually produced from simulations such as those produced by computational fluid dynamics (CFD) software. While CFD is a very useful tool in the study of separation processes, it is limited in its ability to simulate multi-phase flow, and it is important that data produced is validated by experiment.
The scanning laser microscope (Par-tec) is a relatively new particle/droplet sizing instrument which works on the measurement of reflected light, thus it is not subject to the same restrictions of turbidity as other laser techniques which require a transmitted light signal.
This instrument offers a potential means of direct drop size measurement in the field. This would not only provide important data for equipment design but would also be an invaluable tool for field trails. In fact similar instruments have already been used during the offshore trial of novel separation equipment, Schmoll and Cowie.
The major aims of this work were:
P. 785

Two laser-based, optical techniques for drop-size measurement, which have been employed to obtain drop sizes in liquid-liquid pipe flow, are described. The measurement systems used are a laser diffraction technique (Malvern 2600 instrument) and a laser backscatter technique (Par-Tec 300C). The instruments are checked through measurement of a batch of glass beads suspended in water in a stirred cell. An image analysis technique and a phase doppler anemometer provide independent measurements of the size of the beads. The pipe, which is 0.063 m in diameter is installed on a liquid-liquid flow rig. The liquids used are kerosene (continuous phase) and aqueous potassium carbonate solution (dispersed phase). Special test sections are used to deploy the instruments. There are significant differences between the results from the two instruments.

Experimental and theoretical work has recently shown that classical drop size correlations have significant limitations. In particular, that work indicated a slow drift towards smaller drops when agitation is maintained, as well as smaller drops and faster break-up when scaling up at constant power per unit volume. Moreover, the exponent on Weber number fell below −0.6. It was considered that the phenomenon of turbulent intermittency was the mechanism causing the limitations. Here, these ideas are explored farther using equations for stable drop size and drop break-up in intermittent turbulence, the latter being modelled by a multifractal spectrum. These equations are then successfully applied to new drop size measurements for two geometrically similar stirred tanks having different scales, giving further support for the need to consider the phenomenon of intermittency when modelling mixing processes in stirred tanks in the turbulent regime.

The present paper is concerned with the conditions of flow in tanks containing stirred fluids. An attempt is made to apply the theoretical concepts of local isotropy to explain the behaviour of liquid in liquid dispersions, subjected to turbulent agitation. Relations describing quantitatively the influence of turbulence on both break-up and coalescence of individual droplets are derived and are compared with experimental evidence. A special type of dispersion is described in which droplet size is controlled by the prevention of coalescence due to turbulence. The dependence of droplet size on energy dissipation per unit mass, as predicted by the theory of local isotropy, is put to an experimental test using geometrically similar vessels of different sizes.

A method for the rapid and accurate determination of the bivariate distributions of drop volume and a tracer dye concentration in a continuous flow agitated dispersion is described. The method involves extracting a sample of the dispersion from a vessel, protecting the sampled drops with a surfactant, and pumping the sample at a constant flow rate through a capillary smaller than the smallest drop of interest; light transmission gives dye concentration, and time-of-passage of a drop gives drop volume simultaneously. The method yields information about the rate processes of breakage and coalescence in an agitated dispersion.

Theoretical formulations previously put forward to account for the effect of multiple scattering in the measurement of gas-liquid interfacial areas, at high AL values, are critically assessed. It is shown that inconsistencies arising from these can be satisfactorily resolved on the basis of the theory of Al Taweel et al. (1984) and experimental data available in the literature for the evaluation of the different parameters associated with the theory. An equation is proposed for the calculation of large gas-liquid interfacial areas for AL ≤ 280 and excellent agreement between predicted and experimental values is demonstrated.On a effectué une étude critique de formulations théoriques proposées antérieurement pour tenir compte de l'effet de la dispersion multiple lors des mesures de surfaces interfaciales gaz-liquide aux grandes valeurs de AL. On montre que les incohérences de ces formulations peuvent être résolues de manière satisfaisante en utilisant la théorie d'Al Taweel et coll. (1984) et les données expérimentales publiées pour l'évaluation des différents paramètres associés à cette théorie. On propose une équation pour le calcul des grandes surfaces interfaciales gaz-liquide pour AL < 280 et on montre un excellent accord entre les valeurs prédites et les valeurs expérimentales.

The splitting of globules is an important phenomenon during the final stages of disintegration processes. Three basic types of deformation of globules and six types of flow patterns causing them are distinguished.
The forces controlling deformation and breakup comprise two dimensionless groups: a Weber group NWe and a viscosity group NVi. Breakup occurs when NWe exceeds a critical value (NWe)crit. Three cases are studied in greater detail: (a) Taylor's experiments on the breakup of a drop in simple types of viscous flow, (b) breakup of a drop in an air stream, (c) emulsification in a turbulent flow.
It is shown that (NWe)crit depends on the type of deformation and on the flow pattern around the globule. For case (a) (NWe)crit shows a minimum value ∼ 0.5 at a certain value of (NVi) and seems to increase indefinitely with either decreasing or increasing ratio between the viscosites of the two phases. For case (b) (NWe)crit varies between 13 and ∞, depending on NVi and on the way in which the relative air velocity varies with time, the lowest value refers to the true shock case and Nvi→0. For case (c) (NWe)crit, which determines the maximum drop size in the emulsion, amounts to ∼1, and the corresponding values of NVi appear to be small. A formula is derived for the maximum drop size.

Photographic measurements of transient drop-size distributions from stirred liquid-liquid systems of low dispersed phase fraction in a batch vessel confirm consistency with a similarity behavior from population balance established earlier by Narsimhan et al. (1980). Moreover, breakage functions are presented in generalized dimensionless form accounting for dependence on physical properties of the system and power input through stirring. This information is essential for predicting drop-size distributions in stirred liquid-liquid systems. Experimental measurements of steady-state drip-size distributions obtained from stirred, continuous flow systems with known inlet drop sizes compare very favorably with theoretical predictions based on population balance analysis using the breakage functions obtained from batch experiments.

Drop size and dispersed-phase holdup were measured for liquid dispersions in covered, unbaffled agitated vessels with no gas-liquid interface and hence no vortex for batch and continous flow. Vessel diameters of 0.245 and 0.372 m, turbine impellers of diameters 0.0762 and 0.127 m in two locations, four organic- and three water-dispersed systems of a wide range of properties, and a wide range of operating conditions were studied. The data for average drop diameter and dispersed-phase holdup, from which specific interfacial area may be computed, were successfully correlated through modification of the theories of maximum drop size in an isotropically turbulent fluid.

A mathematical model is proposed that takes into account interaction between drops or bubbles in a swarm as well as the effect of particle size distribution. The model is used to solve the equations for unsteady state mass transfer with and without chemical reaction when the drops or bubbles are suspended in a nonextraordinarily purified agitated fluid. Steady state diffusion to a family of moving drops with clean interface and without interaction and chemical reaction has been also analyzed. The model demonstrates the sort of error that may arise when one applies uniform drop size assumptions to drop populations. It is shown quantitatively that this error is usually small when one replaces the variable particle size by the mean. By using variables that can be determined and predicted, the equations presented permit the estimation of diffusion rate per unit area of interface, as well as the average concentration and the total average rate of mass transfer in the disperser under pseudo steady state conditions.

The drop size distributions resulting from the agitation of immiscible liquids were measured over a wide range of parameters. The average drop size is correlated by \documentclass{article}\pagestyle{empty}\begin{document}$ \overline D _{32}/L $\end{document} = 0.053 N We −0.60 . The distribution function for volume fraction is normal and depends only upon \documentclass{article}\pagestyle{empty}\begin{document}$ \overline D _{32} $\end{document} . The results are valid for very dilute solutions wherein coalescence of droplets plays no role in the dispersion mechanism.

A mathematical model simulating light transmission through liquid-liquid dispersions has been developed. Numerical solutions of the model for various drop size distributions show that the fraction of parallel light which passes through a dispersion is a unique function of a dimensionless group, here named the Transmission Number, regardless of the drop size distribution. The results, which were verified by actual light transmission experiments, show that the interfacial area of a liquid-liquid dispersion can be calculated from a light reading provided that the light detector receives only parallel light. Application of the results to other two-phase dispersions is indicated.

The literature available on pipeline flow behavior of emulsions is reviewed critically. New results concerning the laminar and turbulent flow behaviors of unstable (without any added surfactant) and surfactant-stabilized water-in-oil emulsions are presented. The unstable emulsions exhibit drag reduction behavior in turbulent flow; the measured friction factors fall well below the values expected on the basis of the laminar flow properties. Unstable water-in-oil emulsions exhibit much stronger drag reduction activity than the unstable oil-in-water emulsions. The drag reduction activity diminishes (in some cases vanishes completely) upon the addition of a surfactant to the system.

The existingm odels of drop breakage in stirred dispersions grossly overpredict the maximum drop size when surface active agents are present inspite of using the lowered value of interfacial tension. It is shown that the difference in the values of dynamic and static interfacial tension, aids the turbulent stresses in drop breakage. When the difference is zero, e.g. for pure liquids and for high concentration of surfactants, the influence of the addition of surfactant is merely to reduce the interfacial tension and can be accounted for by existingm odels. A modified model has been developed, where the drop breakage is assumed to be represented by a Voigt element. The deforming stresses are due to turbulence and the difference between dynamic and static interfacial tensions. The resisting stresses arise due to interfacial tension and the viscous flow inside the drop. The model yields the existing expressions for dmax as special cases. The model has been found to be satisfactory when tested against experimental results using the styrene-water-teepol system.

Studies on the dynamics of phase inversion available offer limited information due to the difficulty of following the transient mean and drop-size distribution. A new technique developed provides such data. A stereo microscope with a very shallow depth of field attached to a video camera gives sharp images of droplets in intensely-agitated, immiscible liquid dispersions by using a Strobotach pulsing at the camera framing rate. Droplets from 40 μm upward at concentrations up to 70% by volume dispersed phase can be measured accurately. Droplets of continuous organic phase in aqueous drops can be seen. The pictures can be analyzed semiautomatically using a computer and in-house software to give, using a variety of discretizations, cumulative and frequency distributions to any base and any mean size. Means and distributions are a function of time for phase inversions generated in three ways. The technique gives a powerful tool for understanding fast, complex dispersion processes.

A substantial effort has been made by numerous investigators to describe droplet breakage and coalescence in turbulent dispersions. An attempt is made here to improve these models based on existing frameworks and recent advances described in the literature. Two-step mechanisms are considered for both the breakage and coalescence models. The drop breakage function is structured as the product of the drop-eddy collision frequency and breakage efficiency which reflect the energetics of turbulent liquid-liquid dispersions. The coalescence function retains the former structure of the product of drop-drop collision frequency and coalescence efficiency. The coalescence efficiency model has been modified to account for the effects of film drainage for drops with partially mobile interfaces. These models overcome several inconsistencies observed in previous efforts and are applicable for dense dispersions (about ϕ[0.10–0.30]). For the daughter drops produced by breakage, a probability density is proposed based on the energy requirements for the formation of daughter drops.

The extent to which dispersed-phase viscosity influences equilibrium mean drop size and drop size distribution at constant interfacial tension is determined for dilute suspensions by dispersing silicone oils of various viscosity grades in water. A mechanistic model for mean drop size is developed which predicts the moderate-viscosity data and whose parameters correlate the high-viscosity results. Trends in the mean size data coincide with those for the drop size distribution, which broadens considerably as viscosity increases and suggests a dependency on breakage mechanism.

It has been customary to relate the Sauter mean drop size, d32, to Weber number, We, by the relationship . This relationship comes from the assumptions that d32 is a constant fraction of dmax, where dmax is the maximum stable drop size and that dmax can be predicted from theoretical considerations if Kolmogoroffs theory of isotropic turbulence is used for estimating the disruptive forces. Here it is shown, firstly on theoretical grounds, that the assumption d32∝dmax is not justified; and secondly that experimental results from 18 different runs neither support d32=Admax where A is system and agitation conditions independent nor that the exponent on We is −0.6. Further, though cumulative volume size distributions indicate self-similarity, as suggested previously, the number probability density distributions show strong bi-modality at low speed and low dispersed-phase concentration which lessens with increasing concentration and becomes uni-modal at high speeds. This study suggests that in spite of the great deal of work which has already been done, more is required in which the relationship between mean drop size and drop size distributions and agitation conditions over a wider range of concentrations are further investigated.

Increase of dispersed phase hold-up, φ, in a turbulent stirred dispersion is expected to result in increased value of dmax because of the suppression of turbulence intensity. Experiments in stirred vessels however show that with increasing φ, the dmax first increases and then decreases. This unexpected trend can be explained by exploring two mechanisms of drop breakage at the impeller, other than the hitherto recognized mechanism involving turbulent inertial stresses. These new mechanisms involve elongation flow breakage in the accelerating flow along the impeller length and the shear mechanism operating in the boundary layer at the impeller. In a stirred dispersion all the three mechanisms operate simultaneously, each with its own dmax value. The experimentally obtained value corresponds to the smallest of the dmax values obtained through the three mechanisms. The unusual trends of the variation of dmax with increasing φ, can be predicted reasonably well by considering all the three mechanisms together.

A unified approach for predicting the transition to dispersed flow patterns in gas–liquid and liquid–liquid systems is suggested. It is based on the revised models for predicting the maximal drop size in a turbulent field which account for the holdup of the dispersed phase. Examining the range of applicability of the various models for transition to dispersed flow indicates that it is determined by the Eötvös number, EoD=ΔρgD2/8σ. Comparisons with available experimental data for gas–liquid and oil–water systems show that these models are capable of predicting the effects of fluids' physical properties, tube diameter and inclination. The models suggest a non-monotomic effect of the tube diameter on the critical fluids' flow rates, which implies that the up-scaling of data should be approached with care.

Previous correlations have focused on the relationship between the mean drop size and the physical properties of the fluids (interfacial tension, density and viscosity), the volume fraction of the dispersed phase and the average power input per unit mass of fluids. This investigation relates the mean drop size to both the maximum turbulence energy dissipation rate and the turbulent flow in an agitated tank. Four impellers with varying diameters and clearances were used. The maximum turbulence energy dissipation rate was first estimated in the continuous phase (water) using a laser Doppler anemometer (LDA). Then the dispersed phase (silicone oil) was introduced. Drop sizes were measured using a phase Doppler particle analyzer (PDPA). Examination of the smallest drops showed that use of the Kolmogoroff length scale () to estimate the minimum drop size in liquid–liquid dispersions is not accurate. In some cases, more than 30% of the droplets have diameters smaller than the Kolmogoroff length. It was found that the mean drop size is better correlated to the maximum turbulence energy dissipation rate than to either the average power input per unit mass of fluids, or the tip speed of the impeller. Scale-up of liquid–liquid dispersions can be improved when both the energy dissipation and the flow are considered.

An experimental investigation has been carried out in order to analyse the drop size distributions of a liquid–liquid dispersion in a stirred vessel at high phase ratio. Two liquid–liquid systems have been investigated: one at low and one at high coalescence rate. A sampling technique has been developed in order to measure the drop size distributions in the mixer with the help of a laser granulometer. A statistical approach has been attempted to derive the most probable drop size distribution in the mixer and the results have been compared with the experimental primary distributions. The comparison suggests that the energy dissipation cannot be considered as uniformly distributed in the mixer. The mean diameter of the distribution has been correlated to the global mechanical input power and to the volume phase fraction (phase ratio) for both systems in the frame of the classical Hinze–Kolmogorov theory. The results show that for each volume fraction studied, the mean diameter of the dispersion is a decreasing power law of the Weber number with an exponent equal to −0.6 at low phase ratio. However, it appears that for both systems studied this exponent is a decreasing function of the phase ratio. This result reveals the existence of a more complex breakup mechanism with high phase ratio than that assumed in the theory which has to be discriminated from dampening effect of the dispersed phase upon the turbulent energy of the bulk phase. The study of the secondary distributions mean diameter seems to be in good agreement with the numerical predictions of Stone (Annu. Rev. Fluid Mech. 26 (1994) 65). The ratio between the mean diameter of the primary distribution to the satellite drop mean diameter is a growing function of the viscosity ratio.

The dispersion of immiscible liquids is of great importance in many chemical engineering processes. In a turbulent dispersion, the breakup and coalescence of drops usually take place continuously. These rate processes determine the drop size distribution, which is an equilibrium property, and the interfacial area is available for mass transfer. The coalescence between drops in an isotropic turbulent dispersion is examined in this investigation. With this result, a new model without any adjustable or empirical parameters is developed to predict the minimum drop size free from coalescence both in surfactant systems and pure systems. The drop size distribution and coalescence efficiency are also discussed.

Thesis (Ph. D.)--University of Rochester. College of Engineering and Applied Science. Dept. of Chemical Engineering. Bibliography: leaves 72-75.

The maintenance of constant interfacial area per unit volume is a key parameter for the successful scale-up of two-liquid phase bioconversion processes. To date, however, there is little published information on the hydrodynamics of such systems and a suitable basis for scale-up has yet to be defined and verified. Here we report power input and hydrodynamic data for a whole-cell bioconversion process using resting cells of Rhodococcus R312 to catalyse the hydration of a poorly water-soluble substrate 1,3-dicyanobenzene (1,3-DCB). Experiments were performed in geometrically similar 3-L and 75-L reactors, each fitted with a three-stage Rushton turbine impeller system. The two-phase system used comprised of 20% v/v toluene dispersed in 0.1 M aqueous phosphate buffer containing up to 10 gww×L–1 of resuspended biocatalyst and 20 g×L–1 1,3-DCB. The power input to the 3-L reactor was first determined using an air-bearing technique for both single-phase and two-phase mixing. In both cases, the power number attained a constant value of 11 at Re>10,000, while the measured power inputs were in the range 0.15–3.25 kW×m–3. Drop size distributions and Sauter mean drop diameters (d
32) were subsequently measured on-line in both reactors, using an in-situ light-backscattering technique, for scale-up on the basis of either constant power input per unit volume or constant tip speed. At both scales d
32 decreased with increasing agitation rate, while the drop size distributions obtained were log-normal. All the measured d
32 values were in the range of 30–50 µm, with the lowest values being obtained in systems with biocatalyst present. In all cases, constant power input per unit volume was found to be the most suitable basis for scale-up. This gave virtually identical d
32 values and drop size distributions at both scales. A number of correlations were also identified that would allow reasonable prediction of d
32 values for various agitation rates at each scale. While the results obtained are for a particular phase system, the scale-down methodology presented here would allow the rapid evaluation of other bioconversion processes in the 3-L reactor with a 25-fold reduction in scale. In this way, potential problems that might be encountered at the larger scale, such as the carryover of antifoam from the fermentation stage, could be quickly and efficiently identified.

Apparatus for measuring particle dimensions in fluids

- Messtechnik Schwartz

Messtechnik Schwartz GMBH, Apparatus for measuring particle dimensions in fluids. UK Patent 2 325 973 (2000).

In-situ particle sizing in highly concentrated oil-in-water emulsions

- Rimpler

Rimpler S and Daniels R, In-situ particle sizing in highly concentrated oil-in-water emulsions. Pharm Tech Europe: 8: 72–80 (1996).

Drop breakup in stirred liquid-liquid contactor

- De Brown
- K Pitt

Brown DE and Pitt K, Drop breakup in stirred liquid-liquid
contactor, in Proc Chemeca 1970 Conf, Melbourne and Sydney,
pp 83–97 (1970).

for drops in turbulent dispersions

for drops in turbulent dispersions. AIChE J 40:395–406
(1994).