Mean Bubble Research Articles

Hydrodynamics of gas-liquid bubble columns under bubbling, transient, and jetting flow regimes using volume of fluid computational fluid dynamics

A volume of fluid computational fluid dynamics (VOF-CFD) model was used to investigate hydrodynamics of air-water bubble columns in the gas distributor region under bubbling, transient, and jetting flow regimes. The Reynolds-averaged Navier-Stokes (RANS) and large eddy simulation (LES) turbulence equations were coupled with the VOF-CFisms@2015D model. A novel bubble detection algorithm was developed to calculate the bubble size, where a gas fraction of 0.5 was adopted to detect bubbles in the gas-liquid mixture. The CFD results for all three flow regimes were compared to the experimental data in terms of the bubble size distribution (BSD) and mean bubble size. In the framework of VOF, the LES turbulence model captured bubble behaviors in the three flow regimes more accurately than the RANS approach by taking computational time 20% more. Increasing the surface tension from 0.025 to 0.0719 N/m resulted in the increment of the Sauter mean diameter from 4.0 to 5.4 mm. The VOF-CFD model coupled with LES yielded the initial BSD near the gas distributor, which is necessary for Eulerian CFD often applied to an entire column.

Experimental study of rheological behavior of foam flow in capillary tubes

• Rheological behavior of the pre-generated foam was studied in capillary tubes • Tube materials were: PTFE and FEP (hydrophobic) and glass tubes (hydrophilic) • Foam behaves as shear-thinning fluid in all capillary tubes • Transition foam quality in tubes depends on the material type and tube diameter • Foam's apparent viscosity is higher in hydrophobic tubes than in hydrophilic tube due to wall-slip velocity • Wall-slip velocity depends on the tube material type and diameter • Foam texture in tubes is highly dependent on foam quality and flow rate Studies of foam flow in highly permeable porous media are still limited due to foam's complex behavior and discrepancies in foam research. Specifically, it is still unclear how foam flows in capillary tubes and what the effects of material and tube diameter are. We have investigated the rheology of pre-generated foam in capillary tubes. Experiments were carried out using two types of capillary tubes: hydrophobic (PTFE – polytetrafluoroethylene; FEP – fluorinated ethylene propylene) and hydrophilic (GT – glass tubes). The foam was previously formed in the sand-pack by co-injecting a surfactant solution and nitrogen gas. We investigated the effect of material and tube size on foam rheology versus gas fraction (for a fixed flow rate) and flow rate (for a fixed gas fraction). A three-parameter Herschel-Bulkley model was used to describe foam rheology in capillary tubes. Pictures of foam flow in GT and FEP tubes were taken to determine the mean bubble area using image analysis. We estimated wall-slip velocity in capillary tubes and compared the results with the bulk-foam rheology using an analytical expression of the Herschel-Bulkley model for volumetric flow through the circular tubes. We observed shear-thinning behavior in all capillary tubes (FEP, PTFE, GT), and the Herschel-Bulkley model successfully fitted its behavior. The foam in PTFE tubes behaved as a yield-stress fluid, while yield stress was not observed in GT and FEP tubes. We also found that transition foam quality depends on the material type and tube diameter. The results show that foam's apparent viscosity is higher in hydrophobic tubes (FEP and PTFE tubes) than in hydrophilic glass tubes. This was explained by the wall-slip velocity being higher in glass tubes than in FEP and PTFE tubes because of the difference in surface roughness. The corrected flow rate without wall slip matches the flow rate calculated using the measured bulk foam viscosity better. Therefore we conclude that considering wall-slip velocity is important when studying foam flow in porous media.

Bubble size analysis in a two-phase countercurrent flow in the narrow rectangular column

The flow of bubbles in a two-phase system has great implications in chemical, petrochemical, and biochemical applications. This work enunciates the measurement of bubble size distribution and bubble aspect ratio in three-different axial zones in the countercurrent flow mode with a gas and liquid velocity range of 0.044–0.321 and 0.019–0.058 m/s, respectively. Bubble size is measured by the photographic technique. The bubble aspect ratio and bubble size distribution results reveal that the impact of gas and liquid velocities is significant on the Sauter mean bubble size. The Sauter mean bubble size increases as the gas velocity increases, whereas it decreases with the liquid velocity. The Sauter mean bubble diameter ranges from 2.65 to 6.16 mm. The distribution of bubble sizes follows the LogLogistic probability density function. In addition, a correlation is also proposed for the interpretation of bubble diameter in terms of Reynolds number and Froude number. The bubble aspect ratio changes with axial zones and gas and liquid velocities. Experiments reveal that the bubble aspect increases with liquid velocity while decreasing with gas velocity and axial zones. The bubble aspect ratio correlations are developed in terms of Eötvös and Reynolds numbers. The present results will be helpful for the process intensification of bubble-aided two-phase flow applications.

Numerical simulation of multi-size bubbly flow in a continuous casting mold using an inhomogeneous multiple size group model

Multi-size bubbly flow in a one-quarter-scale water model of a continuous casting mold was simulated using the inhomogeneous multiple size group (IMUSIG) model. The Sauter mean bubble diameter, gas volume fraction, interfacial area density, turbulence kinetic energy, and velocity field predicted by the IMUSIG and multiple size group (MUSIG) models were compared. The IMUSIG model separated the motions of the small and large bubbles on the basis of two velocity groups, and both models adopted 15 size groups. The results showed that the IMUSIG model better reflected the bubble size distribution near the nozzle. Small bubbles were further transported by the jet flow, which enlarged the bubble cluster range and decreased the gas volume fraction and interfacial area density in the upper part of the submerged entry nozzle, likely diminishing the coalescence rate in this region; consequently, the overestimation of Sauter mean bubble diameter outside the nozzle was corrected. The liquid-phase flow fields predicted by the IMUSIG and MUSIG models were very similar, and the IMUSIG-predicted magnitudes of velocity and turbulence kinetic energy were only slightly lower than the MUSIG predictions; this slight difference is attributable to the expanded bubble cluster range and higher interfacial area density, which weakened the upper recirculation in the IMUSIG model.

Computational modeling of high pressure bubbling fluidized bed hydrodynamics using ilmenite particles

The multiphase particle-in-cell approach was used to model the hydrodynamics of five different ilmenite particle size distributions in a high pressure cylindrical fluidized bed. The model was created using the commercial software Barracuda Virtual Reactor® and validated against previously obtained experimental data. Simulations for minimum fluidization velocity were dependent on the model input close pack factor, with a value of 0.58 resulting in the best fit with an average absolute deviation (AARD) of 3.9%. Afterwards, trends in bed expansion with fluidization ratio and pressure were well represented by the model, but consistently under predicting by an average of 4.8%. The model also predicted bubble size distribution with change in relation to fluidization ratio, pressure, and elevation. The calculated mean bubble diameter deviated by 18.7% from the experimental results, and the Mori & Wen correlation for bubble growth was adapted to the simulated data resulting in an AARD of 22%.

Influence of admixtures on the characteristics of aqueous foam produced using a synthetic surfactant

Foam characteristics, including density, stability, and texture, influence the properties of end products like foamed concrete. The present work investigates the individual and combined influence of xanthan gum and low-cost sodium-based admixtures (NaOH, NaCl, and Na2CO3) on the viscosity of sodium lauryl sulfate (SLS) foaming solution and the foam characteristics. Where xanthan gum induced its characteristic pseudo-plastic behavior to the foaming solution, the addition of sodium-based admixtures did not affect the Newtonian behavior of the SLS solution. The addition of sodium-based admixtures elevated the critical turbidity and Krafft temperature of the surfactant solution. Xanthan gum substantially reduced the liquid drainage and improved stability to a great extent. However, the increment in foam density was more gradual with increasing xanthan gum dosage, following an increase in lamella thickness. Sodium-based admixtures significantly increased the density of foam while reducing the bubble sizes. Further, the SLS solution admixed with xanthan gum and sodium-based admixture resulted in foam with properties intermediate to the foams stabilized with individual admixtures. The resulting foam had a higher density and a smaller mean bubble size than unstabilized or xanthan gum stabilized foam. In contrast, the stability was more increased compared to unstabilized or sodium-based admixture stabilized foam.

Investigation of the gas-liquid-particle multi-phase hydrodynamics of Wemco flotation cells

Computational Fluid Dynamics (CFD) modelling of a 1.3 m3 flotation cell fitted with a Wemco 56 mechanism has been performed, backed up by experimental measurements. The flow field, gas dispersion and solids concentration in the flotation cell were studied. Both the PIV measurements and CFD modelling of the cell revealed that the rotor generated a jet stream towards the wall, which induced two strong vortices, one below the rotor level and the other extending from the rotor region to near the free surface. The strong downward flow along the wall of the tank turns horizontally passing underneath the false bottom and up through the draft tube. Inside the draft tube, an average axial velocity of about 0.7 m s−1 with a flow rate of 0.05 m3 s−1 was predicted, in good agreement with experiment. The void fraction was found to vary significantly with height in the cell. The maximum value of void fraction reached 0.55 near the froth interface in the air–water system. In the presence of solids, the air bubbles were able to penetrate further downwards to the bottom of the tank and the void fraction reached a maximum value of 0.35 near the froth interface. The Sauter mean bubble diameter was found to be in the range of 0.3–0.9 mm in the air–water system. The solids concentration was practically uniform throughout the tank except for the region near the surface. CFD simulation results for velocity, bubble size, void fraction and solids concentration are analysed in detail. CFD simulation was then carried out for an industrial-scale Wemco 225 cell (92 m3). Predictions are similar to those for the smaller Wemco 56, though there are differences in detail. Voidage predicted by CFD model of the Wemco 225 cell agrees with plant measurements.

Frother Characterization Using a Novel Bubble Size Measurement Technique

Bubble size measurement is a vital part of flotation system analysis and diagnostics. This work evaluates a commercial camera probe as a novel method for in situ bubble size measurement. This device is compared to the conventional Anglo Platinum Bubble Sizer (Stone ThreeTM). It was found that, in laboratory applications, the in situ bubble size analysis technology appears to be a more user-friendly and reliable option for determining bubble size in flotation, whereas the Anglo Platinum Bubble Sizer is more applicable for full scale industrial work. This probe was then used to conduct a rigorous comparison of the behavior of different frother chemistries at a variety of background solution ionic strength conditions. The critical coalescence concentrations and the minimum Sauter mean bubble diameters were determined. Five frothers were compared in terms of their ability to reduce bubble size and sensitivity to salinity. In order to adjust plant recipe and control strategy accordingly, it is recommended that the plant would need to use less frother during periods of the high salinity of process water to achieve the minimum Sauter mean bubble size.

A turbulent bubble breakup model with application to LES

We present a new turbulent breakup model that considers a linear combination of the critical energy obtained from the deformation energy and the force criteria. The relative velocity between the bubbles and the liquid, which affects the eddy/bubble collision rate, is introduced for the first time in a turbulent breakup model and is found to play a significant role. To extend use to a wide range of turbulent flows, a full turbulent spectrum is employed in the derivation of mean turbulent eddy velocity. A collision efficiency caused by the contact angle between bubbles and eddies is introduced and an effective eddy velocity is proposed when eddy size is larger than the bubble size. In principle, the model considers all the sizes of eddies in the turbulent flow, removing the need to limit the integration range when computing the contribution of eddy sizes to the breakup rate. Two modeling constants are calibrated with experiments, and then the model is validated against other experiments. Good agreement is found for both breakup rate and bubble size distribution in different turbulent flows. When the model is tested on large eddy simulation (LES) of bubbly homogeneous isotropic turbulence, it is found that the void fraction and the pressure are strongly negatively correlated and the bubble relative velocity is affected by pressure gradients, affecting the local bubbly flow. The predicted mean bubble size distribution is almost the same as Reynolds-averaged Navier–Stokes (RANS) when the dissipation rates are low. For high dissipation rates the bubble relative velocity increases as the pressure gradients increase, playing an important role in determining the size distribution, and supporting the rationale of adding bubble relative velocity to the turbulent breakup model.

A new, optimized Doppler optical probe for phase detection, bubble velocity and size measurements: Investigation of a bubble column operated in the heterogeneous regime

A new technique dedicated to bubble velocity and size measurements in complex bubbly flows is proposed. This sensor combines the phase detection capability of a conical optical fiber with velocity measurements from the Doppler signal induced by an interface approaching the fiber extremity. It provides the translation velocity of bubbles projected along the probe axis with an uncertainty at most 14%. This sensor has been successfully exploited in a 0.4 m diameter bubble column. In the heterogeneous regime, the mean bubble velocity profiles scaled by the value on the axis are shown to be self-similar in the quasi fully developed region of the column. Besides, the standard deviation of bubble velocity scaled by the mean velocity increases with Vsg in the homogeneous regime, and it remains stable, close to 0.55, in the heterogeneous regime. Such a sensor opens the way to velocity measurements conditioned by the local void fraction.

Investigations of foaming, interfacial and structural properties of dispersions, batters and cakes formed by industrial yolk-contaminated egg white protein

The aim of this study was to describe the foaming properties of egg white protein systems in the presence of egg yolk in varying percentages (0–1, wt%) from an industrial “contamination” view, and to investigate the relationship among their dispersions, batters and cakes in detail. Foaming properties of dispersions, likely the most vital factors determining the quality of final products, were highly correlated with the specific density of batters and specific volume of cakes. To better understand the problem, their effects on interfacial (apparent viscosity, surface tension, size and zeta potential) and structural (optical micrographs, mean bubble sizes and cake slices) properties were also investigated. Results revealed that the interfacial properties of sample dispersions exhibited high correlations with their foaming and structural properties. Rheological characteristics of batters showed higher G′ could be reached during heating with less egg yolk (<0.2, wt%), whereas higher concentrations of egg yolk (0.5–1, wt%) had negative effect on overrun, specific volume and cross-sections. Our results indicated that the quality of cakes can be reflected by the interfacial properties of protein dispersions and the structural properties of cake batters. In addition, our study has supplied a fundamental approach for food industrial applications (such as baked goods) based on egg white to achieve optimal properties of cakes.

Numerical simulation of multi-size bubbly flow in a continuous casting mold using population balance model

A population balance model based on the multiple-size-group (MUSIG) approach was developed to investigate the multi-size bubbly flow in a slab continuous casting mold. Firstly, effects of the bubble size group number, the initial bubble diameter, and the calibration factors for breakage and coalescence in MUSIG model on the multi-size bubbly flow were discussed in detail. Then, effects of both the water and gas flow rates on the bubble size distribution, the gas volume fraction, and the turbulence intensity of liquid phase were investigated. A relatively high water flow rate intensified the turbulence in the upper recirculation region of the mold and also eased the turbulence near the submerged entry nozzle (SEN) by promoting the breakage and horizontal transportation of bubbles; thus, an appropriate water flow rate is essential for better stability with a lower turbulence intensity near the SEN to prevent exposed slag eye formation and slag entrainment. With increase in the gas flow rate, the mean bubble diameter and gas volume fraction increased significantly, the turbulence near the SEN was also significantly increased, but this effect was much weaker in the region away from the SEN.

Structure and dynamics of a laminar separation bubble near a wingtip

The three-dimensional flow topology of a laminar separation bubble forming on the suction surface of a semispan wing with an aspect ratio of $2.5$ and NACA 0018 airfoil section is characterised experimentally using surface pressure measurements and particle image velocimetry at a chord Reynolds number of $125\ 000$ . In the inboard region of the wing, the separation bubble is essentially two-dimensional, and the transition process in the separated shear layer leads to periodic vortex shedding, which dominates the bubble dynamics, similar to two-dimensional separation bubbles. However, progressive spanwise changes in the mean structure and vortex dynamics occur near the wingtip, leading to an open separation and eventual suppression of the bubble. In the immediate proximity of the wingtip, the boundary layer remains attached, no vortex shedding occurs and the flow remains laminar, terminating separation bubble formation. Despite variations in the mean separation bubble topology and vortex dynamics along the span, the fundamental shedding characteristics remain nearly invariant across the portion of the wing where vortex shedding occurs, and the flow appears to lock onto a common instability mode across the span, leading to minimal changes in the mean bubble characteristics despite notable changes in the effective angle of attack along the span. A comparison with available surface flow visualisations from previous studies indicates that the observed changes to the mean bubble footprint along the span of the wing are similar across different geometries and flow characteristics, suggesting similarities in the three-dimensional bubble topology and dynamics on finite wings.

Development and validation of a comprehensive 1-D model to simulate gas hold-up and gas–liquid transfer in deep air–water bubble columns

This study proposed to develop a model coupling hydrodynamics and mass transfer in order to gain an understanding of measurements taken on air–water bubble columns with low gas hold-up. Three experimental datasets with various operating conditions (water quality, liquid height, air flow range) were chosen. The model analyzes and interprets the significant impact of the local hydrostatic pressure and the effects of contamination on hydrodynamic and mass transfer parameters. The oxygen concentration in gas significantly depletes with the distance from diffusers, which explains the difference between the calculated mean of the local <kLa> and global KLa coefficients. This difference is highly significant for a high bubble column and/or systems with a low mean bubble size. The impact of water quality on mass transfer can be characterized by the contamination angle using comprehensive 1-D modeling and highlights a differentiated impact on the hydrodynamic or mass transfer parameters.

Ensemble prediction of mean bubble size in a continuous casting mold using data driven modeling techniques

Gas injection into a slab continuous casting mold is a common practice as it enables long casting sequences by reducing SEN clogging. However, too much gas injection can generate lots of bubbles, and bubbles formed inside the mold are also responsible for the occurrence of various quality defects such as deterioration of steel cleanliness, reoxidation of liquid steel, and occurrence of sliver and blister defects, etc. Mean bubble size is a key parameter and controlling it can resolve/reduce these quality issues. Operating parameters such as gas flow rate and liquid flow rate are the major factors affecting the mean bubble size. Two-phase water modeling​ experiments were performed to generate bubbles in the mold, and the mean bubble diameter was captured using various gas and liquid flow rates. Clear images of bubbles were recorded using a high-speed high-resolution camera along with the application of shadowgraphy. Bubble images were processed using an image processing software, ImageJ to obtain bubble characteristics, and Sauter mean diameter was calculated for each operating condition. Advance machine learning techniques such as Multilinear Regression (MLR), Decision Tree (DT), Support Vector Machine (SVM), and Artificial Neural Network (ANN) were used on the experimental data to predict the combined effect of these operating parameters on the mean bubble diameter. All four ML techniques were compared considering the values of cross-validated adjusted R2, and a performance metric is presented to compare the suitability of ML techniques in this case.

Observational prospects for phase transitions at LISA: Fisher matrix analysis

A first order phase transition at the electroweak scalewould lead to the production of gravitational waves that may be observable at upcoming space-based gravitational wave (GW) detectors such as LISA (Laser Interferometer Space Antenna).As the Standard Model has no phase transition, LISA can be used to search for new physicsby searching for a stochastic gravitational wave background.In this work we investigate LISA's sensitivity to thethermodynamic parameters encoded in the stochastic background produced by a phase transition,using the sound shell model to characterise the gravitational wave power spectrum, and the Fisher matrix to estimate uncertainties.We explore a parameter space with transition strengths 0.01 <α< 0.5 and phase boundary speeds 0.4 <v w< 0.9,for transitions nucleating at T n = 100 GeV, with mean bubble spacings 0.1 and 0.01 of the Hubble length,and sound speed c/√(3).We show that the power spectrum in the sound shell model can be well approximated by afour-parameter double broken power law,and find that the peak power and frequency can be measured to approximately 10% accuracy for signal-to-noise ratios (SNRs) above 20.Determinations of the underlying thermodynamic parameters are complicated by degeneracies, but in all casesthe phase boundary speed will be the best constrained parameter. Turning to the principal components of the Fisher matrix, a signal-to-noise ratio above 20produces a relative uncertainty less than 3% in the two highest-order principal components,indicating good prospects for combinations of parameters.The highest-order principal component is dominated by the wall speed. These estimates of parameter sensitivity provide a preliminary accuracy target for theoretical calculations of thermodynamic parameters.

Predictions of terminal rising velocity, shape and drag coefficient for particle-laden bubbles

The motion of particle-laden bubbles is common in the gas–liquid-solid three-phase process like flotation. In this study, the motion of particle-laden bubbles in the stagnant deionized water was obtained by high-speed dynamic imaging. The experimental ranges of the mean bubble diameter and particle diameter were 2.76–3.95 mm and 58.2–196.35 μm, respectively. Results show that the terminal rising velocity decreases with the increase in the diameter of covering particles and the decrease in bubble diameter. For the coverage below 50%, the terminal rising velocity decreases sharply as coverage increases; while for 50% to nearly 100%, decreases slowly. The sharp decrease is mainly because particles immobilize the bubble surface and increases the drag, which is similar to the effect of contaminants in liquid on bubble motion. When the coverage exceeding 50%, the interface mobility of particle-laden bubbles is completely retarded. The slow decrease is attributed to the decreasing net buoyancy caused by decreasing density difference between water and the particle-laden bubble. Based on the impact of particle coverage on bubble behavior, the measured data were compared with predictions by the available models of the terminal rising velocity and drag coefficient for the bare bubble in the contaminated liquid. Comparison results show that the terminal rising velocity model given by Zheng et al. (2020) and the drag coefficient model given by Wang et al. (2019) can be extended to particle-laden bubbles with a mean error of 4.8% and 0.6%, respectively. The aspect ratio of particle-laden bubbles is negatively correlated to the terminal rising velocity like that of bare bubbles. A new aspect ratio prediction model for particle-laden bubbles related to the ratio between coverage and the Eo number was introduced to facilitate direct prediction of the terminal rising velocity. The comprehensive prediction models of particle-laden bubbles behaviors given in this study are helpful for the design and optimization of equipment that contains bubble-particle aggregates.

Study of chalcopyrite flotation in the presence of illite using a design of experiments approach

AbstractThe interaction between chalcopyrite and illite particles was analysed using ζ-potential measurements and flotation tests. Statistically designed tests were used to examine the factors controlling flotation (frother concentration, dispersant concentration, froth height, airflow rate and amount of clay). Furthermore, the significance levels of the impacts of these factors on responses (chalcopyrite grade/recovery, pyrite grade/recovery, dynamic froth stability and mean bubble diameter) were determined. Chalcopyrite and pyrite ζ-potentials were measured in the presence of illite. The addition of 15% illite to the chalcopyrite, especially between pH 11 and 12, shifted the ζ-potential values closer to that of pure illite, indicating complete surface coating of chalcopyrite with illite. In the flotation experiments, better results were obtained in terms of chalcopyrite grade at a low airflow rate and a high froth height. With increasing froth height there was a decline in the gangue mineral recovery as the residence time of the froth increased. The most significant factor increasing pyrite recovery was the amount of clay. Although illite is considered to be the least problematic clay mineral for flotation, as reported in the literature, an illite content of as low as 5% in the ore decreased chalcopyrite grade by 3.83%. While K and Na contents of 4% were detected after flotation without the addition of illite, their abundance increased to 5.7% after the addition of illite.

Bubble behaviour in a horizontal high-speed solid-body rotating flow

We study experimentally the behaviour of a bubble injected into a horizontal liquid solid-body rotating flow, in a range of rotational velocities where the bubble is close to the axis of rotation. We first study the stretching of the bubble as a function of its size and of the rotation of the cell. We show that the bubble aspect ratio can be predicted as a function of the bubble Weber number by the model of Rosenthal (J. Fluid Mech., vol. 12, 1962, 358–366) provided an appropriate correction due to the impact of buoyancy is included. We next deduce the drag and lift coefficients from the mean bubble position. For large bubbles straddling the axis of rotation, we show that the drag coefficient $C_D$ is solely dependent on the Rossby number $Ro$, with $C_D \approx 1.5/Ro$. In the same limit of large bubbles, we show that the lift coefficient $C_L$ is controlled by the shear Reynolds number $Re_{shear}$ at the scale of the bubble. For $Re_{shear}$ larger than 3000 we observe a sharp transition, wherein large fluctuations in the bubble aspect ratio and mean position occur, and can lead to the break-up of the bubble. We interpret this regime as a resonance between the periodic forcing of the rotating cell and the eigenmodes of the stretched bubble.

Bubble Size in a Flotation Column with Oscillatory Air Supply in the Presence of Frothers

ABSTRACT Oscillatory air supply has demonstrated advantages over conventional steady air supply in froth flotation with diffused aeration, but the interaction between oscillatory air supply and frother in determining bubble size remained unclear. In the present work, we measured the size of air bubbles in a laboratory scale flotation column with common frothers such as MIBC, Dowfroth 250, and polypropylene glycol with an average molecular weight of 425 (PPG 425), with conventional steady air supply and with oscillatory air supply. The oscillatory air supply was converted from steady airflow using a fast-switching solenoid valve with the main operating parameters being switching frequency and on/off time ratio. In the absence of frother, the Sauter mean bubble diameter (d 32) remained the same with oscillatory air supply replacing steady air supply, but the d 32 decreased by 13–35% in the presence of frother. It was also found that the critical coalescence concentration (CCC) of each frother were reduced by 11–69% using oscillatory air supply to replace steady air supply. The degree of reduction in d 32 or the CCC appeared to be dependent on the oscillatory airflow condition (i.e., frequency and on/off time ratio) and frother type. The present work also confirms the recent findings by others that the CCC is not solely a material property of frothers.