Cap Bubbles Research Articles

Experimental investigation of the characteristics of the transition from spherical cap bubble to slug flow in a vertical pipe

Reliable and accurate prediction of the transition from spherical cap bubble to slug flow is crucial not only to the operation of industrial facilities such as the crude oil pipelines, bubble column, and nuclear reactors but also for model development for computational fluid dynamics (CFD) studies. The present paper presents a review of the transition mechanics from spherical cap bubble flow to slug flow in vertical pipes. The bubble flow was split into sub-regions, bubbles and spherical cap bubbles and the mechanisms to classify them (i.e., bubble terminal velocity and cap bubble velocity) was analysed. For now, the literature review does not present some important previous works. This paper presents an original data set of gas–silicone oil in vertical pipes to support the new findings. The experimental two-phase data classifies the flow patterns using the probability density function (PDF) and shows the important flow variables such as average void fraction, pressure gradient, slug body void fraction, liquid slug, Taylor bubble and slug unit lengths, structural velocity and frequency obtained by electrical capacitance tomography (ECT) and a wire mesh sensor (WMS).

Modeling sensitivity in prediction of interfacial area concentration in boiling flow

A comprehensive review and validation of three approaches to predict interfacial area concentration in subcooled boiling flow are presented. The three methods are geometry-based semi-empirical correlations, empirical correlations, and the interfacial area transport equation (IATE). Extensive boiling experiment databases with the measurement of local interfacial area concentration and other local two-phase parameters are employed in the validation. The benchmark shows that the most suitable semi-empirical or empirical correlations chosen to predict the interfacial area concentration are different in the low void fraction region and high void fraction region. The closure model sensitivity in the IATE evaluation is analyzed for the bubble layer thickness models, point of net vapor generation (PNVG) determination approaches, and the presence of large cap and slug bubbles. The analysis of the bubble layer thickness model in IATE reveals that the model is necessary for subcooled boiling systems and the flat bubble layer thickness model is more appropriate than the triangle bubble layer thickness model. The result suggests a necessity of measuring the Point of Net Vapor Generation (PNVG) information in experiments for IATE calculation and the necessity of improvement to the PNVG models. The benchmark shows that the one-group IATE has exceptional performance in low group-2 void fraction region; while in high void fraction flows, the multi-group approach is needed to account for the different characteristics between small and large bubbles.

Single bubble in a 3D sand fluidized bed gasifier environment: A CFD-MPPIC simulation

Biomass gasification can be performed in dense phase sand fluidized beds. The proposed model involves Computational Fluid Dynamics Multiphase Particle-In-Cell (CFD-MPPIC) simulations, allowing the evaluation of bubble dynamics and bubble drag correlations. The proposed model accounts for a single bubble injection into a unit loaded with 520-µm sand particles and 2.7 cm × 0.7 cm cylindrical wood pellets. Model development also includes the selection of an adequate number of computational cells, the sizing of the particle-in-cell population, and the cell distribution throughout the column. Predictions of the proposed CFD-MPPIC Model were compared with bubbles measured by the HESC Model (Torres Brauer et al., 2020). This data was obtained in a 0.44 m diameter Plexiglas unit, using CREC-Optiprobes and a spherical cap bubble assumption. It is anticipated that the CFD-MPPIC Model could have significant value for the design of scaled-up biomass units operated under dense phase fluidization conditions.

Pressure drop, void fraction and flow pattern of vertical air–silicone oil flows using differential pressure transducer and advanced instrumentation

New data for pressure drop, void fraction and flow pattern in a vertical riser using air–silicone oil as the system fluid are reported in this work. A differential pressure cell (DP cell) was used to measure the pressure drop. Also, void fraction data were recorded simultaneously by an electrical capacitance tomography (ECT) and wire mesh sensor (WMS). The observed flow patterns are the spherical cap bubble, slug and churn flows. However, only the slug flow without the presence of churn flow is seen within the transition line as predicted by the map. The characteristic probability density function (PDF) derived from void fraction data was used to determine the flow patterns. A comparison between present experimental results and the air–water data reported in the literature was carried out and various levels of agreement were achieved. The PDFs obtained from the DP cell signals for spherical cap bubble and slug flows significantly differ from those derived from the ECT and WMS outputs. Current void fraction and pressure gradient results were compared with the values predicted by ten empirical correlations selected from the literature. Statistical tools such as Mean Square Error (MSE), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) were applied in the comparison. The Greskovich and Cooper correlation gave the least MSE, RMSE and MAPE values of 0.0007908, 0.013 and 3.05%, respectively for slug flow.

Two-phase flow pattern identification in minichannels using image correlation analysis

Abstract In the field of two-phase flow image analysis many methods require a complex experimental setup, its demanding calibration and an advanced image analysis process. Thus, in the paper a noncomplex approach to two-phase flow pattern identification involving high speed recording and simple, unfiltered images analysis was proposed. The coefficient which is the percentage measure of rate of gas distribution changes in time inside the minichannel was defined. Such coefficient was used to identify the two-phase flow patterns in water-glycerol and air flow in 3 mm circular, horizontal minichannel. It was shown that the value of a coefficient is suitable for a flow pattern map reconstruction. The correlation analysis of the flow pattern dynamics provided sufficient information of two-phase flow patterns. The following two-phase flow patterns were identified: minibubbles, bubbles, cap bubbles with slugs and slugs. The maximum fluctuation of the proposed Vv coeffcient value was equal to 36% for minibubbles flow and minimum to 3% for slugs flow and cap bubbles with slugs flow. Additionally, the method was tested for different two – phase flows (water – air) including boiling flow. Finally, it can be concluded that the proposed algorithm has the potential to be applied for two-phase flows inside minichannels.

Flow structure of bubbly to slug transition flow in a small pipe

To develop a mechanistic constitutive model to characterize the flow structure development in more than one flow regime, the detailed measurement of the flow structure is indispensable. In this study, the void fraction, interfacial area concentration, and bubble velocity profiles of upward air-water two-phase flow has been measured using the four-sensor conductivity probe in a 25.4 mm ID pipe. The test matrix has two features that distinguish it from previous studies, which is the void fraction-based flow condition selection and the wide range. The test matrix consists of 24 flow conditions with 3 different superficial liquid velocities (0.5, 1.0, 2.0 m/s) and 8 different void fractions (0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.45, 0.60). The results indicate that the flow characteristics and flow structures not only depend on the void fraction but also significantly depend on the superficial mixture velocity. The velocity effects on various flow characteristics are discussed such as void distribution, cap bubble generation, interfacial area. Besides, a velocity correlation has been developed to predict the bubble velocity radial distribution, with the mean absolute percentage error of 10.05% for the current data.

One-dimensional drift-flux model and a new approach to calculate drift velocity and gas holdup in bubble columns

In view of the practical importance of the drift-flux model and the variety of industrial applications of large-diameter columns, the model coefficients, including the distribution parameter (C0) and drift velocity (C1), have been investigated for the upward two-phase flow in large- and small-diameter columns. The distribution parameter and drift velocity reflect the non-uniformity in the flow and the local relative motion between two phases, respectively. As the formation of cap bubbles and occurrence of liquid recirculation are two of the most important characteristics in a large-diameter column, the distribution parameter and drift velocity in such a column are rather different from those in a small-diameter column. In this study, the local flow parameters measured by other researchers were collected, and C0 and C1 were calculated based on their definitions. The experimental values in small- and large-diameter columns were compared, and the performances of various existing constitutive equations for the drift-flux model were re-evaluated. Two semi-empirical correlations for the drift velocity and gas holdup were developed, and their performances were verified against experimental data for various two-phase flow conditions.

Spherical cap bubbles in fluidized beds and bubble columns: A new formula for added mass

A simple approximate formula was found on heuristic basis for the added mass of both rigid and deformable cap bodies. Its prediction agrees with the published CFD data. It can be used as a closure for numerical studies.

Emptying of a bottle: How a robust pressure-driven oscillator coexists with complex two-phase flow dynamics

The emptying of a bottle is one of the most common two-phase flows encountered in everyday’s life fluid mechanics. We report on a detailed experimental analysis of this flow configuration covering a wide range of neck-to-bottle diameter ratios, d*, and initial filling ratios, F. The joint use of a pressure sensor at the top of the bottle and a shadowgraph technique to track the evolution of the upper liquid surface allows the average gas volume fraction in the fluid column to be computed throughout the discharge. Variations of the bottle emptying time, gas content, and characteristics of the pressure signal as a function of d* and/or F are analyzed, and scaling or evolution laws are derived. A specific focus is put on the initial transient stages following the neck opening, especially on the nature and growth of the first bubble that reaches the upper free surface. Over a wide range of d*, this bubble takes the form of a large spherical cap or Taylor bubble, the rise speed of which is controlled by the bottle diameter. Such a bubble directly emerges from the neck for large enough d*, but rather results from successive coalescence of smaller bubbles for small d*. Whatever d*, the volume of this leading bubble increases over time, since a fraction of the small bubbles swarm trailed in its wake merges with its back face. This coalescence-induced growth is quantified through a simple model.

Numerical and experimental investigation of pulse bubble aeration with high packing density hollow-fibre MBRs

A three-dimensional Computational Fluid Dynamics (CFD) model was developed to study shear stress induced by spherical cap bubbles in hollow fibre (HF) membrane modules configured with a packing density of 38 m2/m3, to predict the shear profile in a commercial hollow fibre membrane module of 265 m2/m3. The CFD model's computational effort was minimised by simulating the formation of bubble structures and their rising velocities in modules with packing densities of 1.8 and 38 m2/m3 and validated with experimental calibration of shear profiles via electro-diffusion methods (EDM). Pulse bubbles (300 mL) generated from a single sparger at 0.5 Hz produced more satellite bubbles in the wake zone of the leading bubble in high packing density (38 m2/m3) than in low packing density modules (1.8 m2/m3). The bubble rise velocity was approximately 8% lower in the 38 m2/m3 than in the 1.8 m2/m3 module. Increasing packing density reduced the shear profile from a single sparger and the dispersion of the satellite bubbles in the horizontal plane, especially in the upper part of the membrane module. For systems with multiple spargers, the interaction between pulses generated more shear than the pulses from a single sparger, and produced a more uniform shear profile in the module through asynchronous bubble release from adjacent spargers than synchronous release. A 33% increase in the “Zone of Influence”, the flow region where the upward velocity >0.2 m/s, was achieved by moving from a synchronous to an asynchronous form of aeration.

Numerical study on the single bubble rising behaviors under rolling conditions

The accurate prediction of bubbles rising characteristics is very important for the two-phase flow and heat transfer calculation, especially under the ocean conditions. The additional force will be introduced and applied on the bubbles and the bubble behaviors is different from that under the stationary conditions. In present work, two methods, the additional force and dynamic mesh of Arbitrary Lagrangian–Eulerian (ALE) method, are compared, the result shows these two methods are in accordance to calculate the effect of rolling. Then Volume of fluid (VOF) and ALE method have been used to calculate the single bubble rising characteristic under rolling conditions. Results show that sphere bubble, ellipsoidal bubble and sphere cap bubble exist during the whole rising process and the deformation is becoming larger with the increase of bubble size. The bubble tends to move laterally and periodically under rolling conditions. The period of lateral movement is in accordance with the rolling period and the amplitude of the bubble is proportional to the rolling amplitude. Also, larger bubbles have the higher rising velocity than the small ones. The bubble rising characteristics are determined by not only the rolling motion but also the surface oscillation under rolling conditions.

Combined effects of alcohol and electrolyte on mass transfer from single carbon-dioxide bubbles in vertical pipes

Mass transfer rates, kL, of single carbon-dioxide (CO2) bubbles rising through vertical pipes filled with alcohol-electrolyte mixed aqueous solutions were measured to investigate the combined effects of alcohol and electrolyte on kL. 1-octanol and 1-heptanol, which are known to adsorb to the interface and behave like surfactant, were used. Sodium chloride (NaCl) was used for the electrolyte. The surface tension, σ, decreased with increasing the NaCl concentrations, CN, while keeping the alcohol concentration, CA, constant. Three combinations of CA and CN having the same values of σ were selected for the experimental conditions to avoid the effect of σ. The pipe diameters, D, were 12.5 and 18.2 mm. A wide range of bubble diameter covered various bubble shapes, i.e. ellipsoidal, cap, semi-Taylor and Taylor bubbles. The conclusions obtained are as follows: (1) the combination of CA and CN for the same σ does not affect the aspect ratios of ellipsoidal bubbles and the lengths, L, of Taylor bubbles, (2) the Sherwood numbers, Sh, of bubbles in alcohol-NaCl mixed aqueous solutions depend on the combination of CA and CN due to the change in the Schmidt number, Sc, even at the same σ for all the tested combinations of concentrations of impurities, and (3) the modified Sherwood number, ShD, of contaminated Taylor bubbles is well correlated in terms of the bubble Reynolds number, Sc, the dimensionless group for surfactant properties and L/dT, where dT is the transition bubble diameter at the transition from the ellipsoidal-cap bubble regime to the transition regime.

One-dimensional drift-flux correlation for vertical upward two-phase flow in large size concentric and eccentric annuli

In view of the practical importance of predicting void fraction in two-phase flow, a correlation based on the drift-flux model is often utilized in many industrial fields. To use the drift-flux type correlation, two drift-flux parameters such as distribution parameter and void fraction weighted mean drift velocity should be modeled accurately. Since the drift-flux parameters are susceptible to channel size, they are modeled differently depending on whether the channel size is characterized as small (or mini), medium or large. Although drift-flux type correlations with modeled distribution parameter and drift velocity for medium size channels with various geometries have been well-developed, limited drift-flux correlations are available for large size channels. Two-phase flow characteristics in a large size channel are quite different from those in a medium size channel at low flow conditions. Large slug bubbles observed in a medium size channel do not form in a large size channel due to the presence of surface instability. As large slug bubbles disintegrate into smaller cap bubbles, the drift velocity is increased. Complex two-phase flow characteristics in a large size channel also affect the distribution parameter. This study successfully models the distribution parameter and drift velocity for two-phase flow in a vertical concentric annulus. The drift-flux correlation with modeled distribution parameter and drift velocity is validated by existing data taken in large size annuli. The applicability of the drift-flux correlation to two-phase flow in a vertical eccentric annulus is also examined by existing data.

Experimental study of viscous effects on flow pattern and bubble behavior in small diameter bubble column

An experimental study was carried out to explore viscous effects on the flow pattern and bubble behavior in a small diameter bubble column using silicone oil fluids of viscosities 5, 100, 1000, and 5000 mPa s. The flow pattern was obtained from the probability density function using the time-averaged void fraction traces of the real-time qualitative and quantitative measurements from Electrical Capacitance Tomography (ECT). This was confirmed from the high-speed camera and ECT images which also gave a vivid description of the bubble behavior. Further confirmation of the observed flow pattern was obtained using the diameter ratio, λ, where for slug flow λ > 0.6. The flow pattern was observed to vary from spherical cap bubbles to developing slug and slug flow within the gas superficial velocities considered (0.02 ≤ Ugs ≤ 0.361 m/s). As viscosity increases, the appearance of spherical cap bubbles decreases, while slug flow tendency increases. The observed flow patterns were compared with previous work. The developing slugs in liquid viscosities of 5 and 100 mPa s were observed to be deformed. The Taylor bubble obtained from 1000 mPa s is called prolate spheroid, while that obtained from 5000 mPa s is called oblate spheroid. In addition, as the superficial gas velocity increases, the length of Taylor bubbles increases, while that of slug flow decreases. The bubble behavior was further characterised using the inverse dimensionless viscosity and Eotovos number. The process of leading and trailing bubble merging to form Taylor bubbles via coalescence was also captured and explained using the high-speed camera video.

Core design and performance of the FASt TEst reactor (FASTER)

Characteristics of the slug flow, mean void fraction, liquid slug and Taylor bubble lengths; and structure frequency were all extracted from the void fraction time series. Average void fractions at nine axial positions have been measured on a 6 m long (and 34 mm diameter) pipe test section by using conductance probes.For the flow conditions used, bubbly (with spherical cap bubble), slug/plug and churn/semi-annular flow patterns were observed. These observations were also confirmed by using statistical numbers.Time series analysis of the mean void fraction and its corresponding PDFs show that the effect of the mixer on the flow becomes, practically, negligible from a distance of 95 pipe diameter.It was found that, generally, the structure velocity can be well predicted from other available correlations in the literature.The theoretical model of Brauner and Ullmann was used to estimate the liquid in slug void fraction. It was inferred that this model is able to predict well the liquid slug void fraction; thus, it can be considered as one of the useful methods for predicting such parameter and to ascertain slug flow regime.The total pressure gradient was found to decrease with increasing gas superficial velocity.

Experimental study of the characteristics of an upward two-phase slug flow in a vertical pipe

Characteristics of the slug flow, mean void fraction, liquid slug and Taylor bubble lengths; and structure frequency were all extracted from the void fraction time series. Average void fractions at nine axial positions have been measured on a 6 m long (and 34 mm diameter) pipe test section by using conductance probes.For the flow conditions used, bubbly (with spherical cap bubble), slug/plug and churn/semi-annular flow patterns were observed. These observations were also confirmed by using statistical numbers.Time series analysis of the mean void fraction and its corresponding PDFs show that the effect of the mixer on the flow becomes, practically, negligible from a distance of 95 pipe diameter.It was found that, generally, the structure velocity can be well predicted from other available correlations in the literature.The theoretical model of Brauner and Ullmann was used to estimate the liquid in slug void fraction. It was inferred that this model is able to predict well the liquid slug void fraction; thus, it can be considered as one of the useful methods for predicting such parameter and to ascertain slug flow regime.The total pressure gradient was found to decrease with increasing gas superficial velocity.

Local flow regime and bubble size distribution in the slender particle-containing scrubbing-cooling chamber of an entrained-flow gasifier

A dual-tip conductivity probe was used to measure local radial gas holdups and bubble chord lengths to study the local flow regime and bubble size distribution in a scrubbing-cooling chamber cold model apparatus containing slender particles. The results showed that the cross-section averaging gas holdups with different fiber volume fractions and aspect ratios could be estimated within errors of ±10% using the modified Kataoka-Ishii bubbly flow semi-empirical correlation at superficial gas velocities ranging from 0.074 m/s to 0.37 m/s. The local flow regime map showed that an obvious bubbly flow formed near the inner wall of the liquid bath due to the effect of wall shear stress, while a cap-bubbly flow formed in other annulus regions due to the emergence of cap bubbles. Chord length distributions were transformed into bubble size distributions by decomposing the measured chord length distributions and estimating the bubble shape factors. Bubble size was categorized into two types—small spherical and large non-spherical bubbles—depending on if the equivalent diameter of the bubbles was smaller or larger than 2 mm. Liquid turbulence, shear stress between the fluid and the wall and liquid backmixing were enhanced by increasing the superficial gas velocity, which affected the size distributions of rising and descending bubbles as well as that of bubble populations. The extent of turbulence suppression and “fiber separation walls” varied with the fiber volume fractions, which affected the bubble size distribution in different regions. The fiber number density characterizing the variation of the fiber aspect ratio with different fiber diameter and length was used to study the bubble size distribution. The bubble size distribution at different aspect ratios was affected by interactions among fibers, bubbles, the fluid and the wall.

Fluid structure behaviour in gas-oil two-phase flow in a moderately large diameter vertical pipe

Intermittent flows in vertical pipes occur in many industrial settings including power generation and downstream oil-and gas production. This type of flows include cap bubble, slug and churn flow regimes. These regimes are of interest as downstream processes and control may heavily depend on the intermittency of the inflow. There are a number of correlations that predicts the features in such flows in vertical pipes. Most of the correlations were developed for air and water fluid pair for slug flow regime in vertical pipes with 25 to 50 mm inner diameter. In this paper, an attempt has been made to assess the suitability of several of these correlations specific to slug flow regime for a fluid pair that is different to air-water system. In this work, air-silicone oil flow development was experimentally investigated in a vertical pipe with an inner diameter of 68 mm. A Wire Mesh Sensor (WMS) and an Electrical Capacitance Tomography (ECT) sensor were installed in series at four locations (15D, 30D, 45D and 65D) downstream of the mixing section. The flow was visually observed using a high speed camera. The void fraction time series obtained from the WMS and the ECT were used to establish the flow characteristics such as slug length, slug frequency, void fraction in liquid slugs and Taylor bubble velocity. A comparison showed that the void fraction measurements using ECT and WMS are in good agreement. Axial measurements shows that the flow development beyond 45D is minimal. Change in physical properties of the liquid phase is responsible for the deviation associated with the existing slug flow models, particularly those developed to predict the gas holdup in liquid slugs.

Interrogating flow development and phase distribution in vertical and horizontal pipes using advanced instrumentation

The characterization of two-phase flow in both vertical and horizontal pipes is important in oil/gas transportation. As the pipe orientation may change from well to well, it is also important to understand the characteristic responses expected of a two-phase mixture to a change in pipe inclination. This study concerns the changes in the void fraction, structure frequency and structure velocity that occur within an air–silicone oil mixture as a function of the pipe orientation. Experimental data were obtained from the combined use of electrical capacitance tomography (ECT) and wire mesh sensors (WMS), which allow the 3D visualization of the flow patterns. The reported experiments were performed on a 67 mm diameter pipe, in a flow loop, in which a pipe section may be inclined at angles of between −5 or 90° to the horizontal. For this study, the inclined pipe was either set at an angle of 0 or 90 to the horizontal, which correspond to a horizontal or vertical pipe setting, respectively. The results of flow development using the PDF of void fraction obtained at 3 measurement locations; ECT 1, ECT 2 and WMS at 4.4, 4.489 and 4.92 m, respectively, showed that the flow is fully developed and statistically stable for the vertical two-phase flow. While on the other hand, the flow is in rapid development for the horizontal two-phase flow scenario at same liquid and gas superficial velocities. The processed data reveal the differences in flow distribution produced by the pipe inclination. Within the vertical pipe configuration spherical cap bubbles, slug and churn two phase flow patterns were observed, whilst plug, slug and stratified wavy two-phase flows were identified in the horizontal configuration of the pipe. It is concluded that a plot of mixture superficial velocity against average void fraction may be used to provide a qualitative assessment of the flow patterns observed, regardless of pipe inclination. The two-phase flow variables that are concluded to significantly influence the distribution coefficient, C0, and drift velocity, VD, are the pipe orientation and flow patterns.

Insights on pulsed bubble control of membrane fouling: Effect of bubble size and frequency

A three-dimensional Computational Fluid Dynamics (CFD) model was developed to study the shear stress induced by spherical cap bubbles on industrial scale hollow fibres. Simulations and experimental data were compared for water and a 0.5 g/L xanthan gum solution that mimicked the rheological behaviour of activated sludge. The model was built on a pilot-scale membrane tank configured with five 1.5 m long hollow fibre membranes. 58–290 ml spherical cap bubbles were sparged into the system at frequencies from 0.2 to 1 Hz. The Volume of Fluid (VOF) method coupled with the Realizable k–ε turbulent model was used to simulate the transient behaviour of cap bubbles rising in a Newtonian and a viscous shear-thinning liquid. CFD prediction of root mean square (RMS) shear stress induced by pulsed bubbles exhibited an inverse relationship with experimental observations of membrane fouling. At a constant flux of 25 L/m2/h (LMH), fouling decreased from 5.1 to 3.9 kPa/min as simulated RMS shear stress increased from 0.04 to 0.25 Pa. Shear fluctuations, which are considered to be beneficial for the cleaning of membranes, were less pronounced for a non-Newtonian fluid compared to in water. Increasing pulse bubble size from 115 ml to 290 ml led to a 80% increase of RMS shear stress, and therefore improved the fouling control efficiency.