Journal of Mechanical Engineering Research and Developments
ISSN: 1024-1752
CODEN: JERDFO
Vol. 43, No. 7, pp. 26-50
Published Year 2020
26
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-
liquid in a horizontal smooth and turbulator conduit – A review
Hyder M. Abdul Hussein†, Sabah Tarik Ahmed‡, Laith Jaafer Habeeb‡†
†University of Kufa – Faculty of Engineering ‡University of Technology – Mechanical Engineering Department
‡†University of Technology – Training and Workshop Center
*Corresponding Author Email: [email protected]
ABSTRACT: This paper presents a review of experimental investigations, the analytical formulations, and
numerical models of (gas-liquid), (gas-solid), (liquid-solid), and (liquid-liquid). Two groups of studies
turbulator and smooth in conventional passages tubes, pipe, and channels conduit are reviewed. Different
aspects such as flow pattern, pressure drop, maps, void fraction, and estimates of the kinetic energy and
momentum are of interest. The need to systematize the huge amount of manuscript published on the multiphase
flow and to understand the limitations of the techniques employed constitutes the motivation for this review.
Predicting flow pattern and transient flow conditions is developed at experiment (photographs, visual
observation, image processing, parallel wire conductance sensor and Particle Image Velocimetry visualization,
Planar Laser-Induced Fluorescence, Simultaneous two-line laser-based, high speed video-camera, Conductivity
needle Probe, and Gamma Densitometer) were used. In two-phase flows, the utmost decisive parameter is the
pressure gradient over the flow. Practically, the major thing for effective processes is a delicate prognostication
for the gradient of pressure over the flow of two-phase. An experimental study was generally obtained using
pressure transducer, and multi-tube manometer. Essentially, different kinds of patterns exist in order to identify
as a drift-flux pattern, homogeneous no-slip flow pattern as well as separated pattern. The numerical solution
has been steady state transitions to the unsteady state cases, which used code or commercial CFD software.
KEYWORDS: Multiphase flow, flow pattern, pressure drop, maps, and void fraction.
INTRODUCTION
The term multiphase flow is used to refer to any fluid flow consisting of more than one phase or component [1].
A phase is simply one of the states of matter and can be either a gas, a liquid, or a solid. Multiphase flow is the
simultaneous flow of several phases. Two-phase flow is the simplest case of multiphase flow [2]. The general
subject of multiphase fluid flow is widely used [3]. Multiphase flows in the context of fluid mechanics can be
perceived as a flow system that consists of two or more distinct phases flowing in a fluid mixture where the
level of separation between the phases is at a scale well above the molecular level [4]. Furthermore, the rate of
publication has been increasing in recent years. This work has been carried out at universities, national
laboratories, and at industrial research and design organizations in many countries of the world [5]. The
behaviors of multiphase flow under a wide range of flow conditions vertical, horizontal, and inclination angles
constitute an outstanding interdisciplinary problem with significant applications [6]. The condensation of steam
upon windows, the human body's blood flow as well as the rising of bubbles in a cold soda glass are models for
that. The constituents' nature and the relative distribution of them are the things on which these flows are
heavily dependent. The case that in which the same pure component is existing in two various phases for a flow
of two-phase is the simplest state of multiphase flow. The flow of steam - water is a model for that. Further, the
multicomponent flow is normally indicated to the flow that has various chemical substances.
Most common class for multiphase flows are two-phase flows and these include the following [7]:
1- Gas-liquid flows, which are probably the most important form of multiphase flow and is found widely in
industrial applications Such flow exists in a range of industrial plant which includes evaporators, condensers,
boilers, distillation towers, chemical reactors, air ejectors, pipelines for oil and natural gas, turbines, etc. [8].
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
27
2- Gas-solid flows, where solid particles are suspended in gases, which are of industrial importance in
pneumatic conveying, in the combustion of pulverized fuel and in fluidized beds.
3- Liquid-solid flows, which are widely encountered in hydraulic conveying of solid material. Suspensions of
solids in liquids also occur in crystallization systems.
4- Liquid-liquid flows, which include emulsion flows of oil and water in pipelines (of interest in the present
context) and flows through packed columns, pulsed columns, nuclear reactors, distillation columns, stirred
contractors and pipeline contractors in liquid-liquid solvent extraction.
Gas and liquid flow in horizontal pipes show a number of different interfacial configurations. Classification and
description of the flow distributions into different patterns are frequently very subjective [9]. This paper reviews
the multiphase flow in horizontal conventional conduit [10] smooth and ribbed walls which were done by
experimental and numerical work concerned with a system used gaseous, liquid, and solid phase for use in two-
phase. The main aim of this paper, given the significant recent to study the flow pattern, deference pressure,
flow pattern maps [11], and void fraction developments in experimental techniques and numerical solution, and
to provide broader guidance to experimentalists performing detailed, high-fidelity, numerical methods
aforementioned fluid-flow systems. Gas-liquid flows have received more attention than other forms of two-
phase flows and, in addition to the plethora of the experimental data, many predictive models have also been
developed [12]. Among the earliest studies in the gas-liquid field were by [13-21]. In the past, many studies that
are dedicated to the pipe of particle-laden as well as channel flows of various boundary conditions have been
issued. During one of the initial studies, developed for the one-dimensional flow of a gas, containing solid
particles of two different sizes was studied [22-25]. Tsuji and Morikawa gave an extremely itemized group of
tests for a gas-solid flow [26]. For the purpose of characterizing and parameterizing of turbulence modulation in
gas-solid flows Gore and Crowe they were frequently utilized [27]. By many researchers in horizontal pipes,
empirical data on the action of pressure gradient for systems of liquid-solid have been gained [28-33]. The
prediction of pressure drops and flow patterns is a complex problem and is treated mostly via correlations of
experimental data. Some of the empirical correlations claim to apply to all flow patterns for liquid-solid systems
[34-37]. At fully suspended flow, an uncomplicated model for the pressure drop has been mentioned by [38-40].
However, they based it upon an experimental relationship for the gradient of concentration. One of the first
studies in liquid-liquid flow was carried out where three flow patterns were identified by visual observations,
namely, stratified flow, bubble flow and mixed flow [41]. A number of studies have been conducted to develop
the efficiency of pumping for pipelines that crude oil transporting in it, however, these studies have been
performed in horizontal pipes [41-50]. The aims of the present review paper are: (1) to review and summarize
the recent studies focus in flow pattern, pressure drop, maps, void fraction, hold up, and estimates of the kinetic
energy and momentum for a two-phase flow with and without turbulator in conduit and (2) to present research
gaps which need to be considered for future research work in the area.
TWO-PHASE FLOW WITH TURBULATOR
Pipe and Channel that is coarser with turbulator or grooves are widely used in several applications such as
ventilation, turbine blades, heat exchangers, and refrigeration. Ribbed channels are commonly used for the
enhancement of convection heat transfer. Since the presence of ribs in the conduit generated a turbulence flow
by breaking the laminar sub-layer if compared with the smooth channel.
Gas-liquid flow
Zarnett and Charles [51] were the first to investigate the flow of air-liquid mixtures with continuous spiral ribs
having 1.57 and 2.79 of pitch-to-diameter ratios and the internal diameter is ¾ inch diameter Lucite tubes. The
main effect of the spiral rib was to move the gas phase away from the tube walls. Thereafter, Weisman [52]
performed testing of pressure drops as well as two-phase flow model through single and double helically
circular ribbed tubes both 2.54 and 5.1 cm circular tube and 3 m section with consistent wire spacing. 0.32, 0.64
and 1.27 cm heights were used for Helically ribbed of a circular cross-section. The ratios of helical twist
were1.2, 2.1, 2.3 and 2.5. It could be indicated from the results that the moment that the smallest velocity of
liquid passed, the flow of swirling annular noticed at low qualities. Consequently, over this smallest velocity of
the liquid, it is predictable that the critical heat flux would also be enhanced. Salcudean [53] investigated
experiments various obstruction shapes and sizes where pressure-loss coefficient along a 25.4 mm inside
diameter tube was measured. Then Hameed [54] examined an analytical with various obstruction shapes to
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
28
investigate the pressure drop in horizontal air-water flow. Results gave a good agreement obtained between the
theoretical and experimental results for both 25% and 40% obstruction area, noting that the 40% obstruction
area has the best agreement than the 25%.
Figure 1. Shape and location of the obstruction in the channel. a- Central, b- Horizontal segment, c- Bottom
segment, d- Peripheral e- Vertical segment f- and Top segment.
Ansari and Arzandi [55] investigated experimentally adiabatic air-water two phase flow using rectangular ducts
that were smooth and ribbed to show the effect of ribs height on the boundaries, they also presented a flow map
diagram. Three ribs of different heights (1, 2, and 4 mm) were used shown in Figure 2. The rib width (10 mm)
and pitch (50 mm) were held on the bottom wall (waterside), on the top wall (airside) and on both the top and
bottom walls. The location of the ribs in the duct did not alter the shape of the flow regimes, but the regime
boundaries were considerably changed.
Figure 2. Geometry of the ribs.
Ansari [56] carried out experimentally at room temperature and atmospheric pressure was for the purpose of
investigating the impact of thickness for rib and pitch within a test section of rectangular ribbed having a length
of 3.6 m and 100×50 mm cross-section. For nine different configurations of rib, the transition boundaries and
the diagrams for flow model were achieved with 2, 4- and 8-mm thicknesses and 50, 60, and 80 mm pitches.
Unlike rectangular channel that is non-ribbed, the configuration of sidelong rib did not permit for the happening
of any stratified flow, boundary values undergo tangible changes. Despite wavy, slug, as well as plug flows
were parallel in both conditions of flow, the presence of rib created explicitly coarser forms of model, see Figure
3. Keeping the pitch constant, while increasing the rib thickness, results in various models of flow in order to
happen and spectacular variations in boundaries forms as well as positions.
a b c
f e d
In-Line H
H P
W
e
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
29
Figure 3. Wavy flow model Samples in ribbed ducts.
Habeeb and Al-Turaihi [57] investigated the two-phase (air-water) flow around triangular–section obstacle in
rectangular channel experimentally and numerically for steady and unsteady flows. The horizontal rectangular
channel was made of transparent Perspex with dimension of (10 × 3 × 70 cm). The flow’s behavior and pressure
difference studied. Four different values of water flow rate (20, 25, 35 and 45 l/min), and four different values of
air flow rate (10, 20, 30 and 40 l/min) were used. CFD used to perform the numerical study by fluent package. It
found that the pressure difference increased as the flow rates of air or water increased, and more turbulence seen
which produced more bubbles and waves.
Huang [58] used the fluctuating pressure produced by a bluff body in order to identify the flow actions of gas-
liquid two-phase for introducing a disorganized property index as well as a correlation dimension. The inner test
pipe has a diameter D of (50 mm). The front face width cross section was (w = 14 mm) of the bluff body a
truncated isosceles triangle which inserted through the pipe, as shown in Figure 4. The bluff body to the test
section ratio was (0.28), that in the bluff body wake, it could produce intense and uniform vortex shedding. A
synthetic neural system trained to assist chosen proper parameters of flow which connected with correlation
dimension in order to create a new flow pattern map of gas-liquid, that had the ability to recognize among the
slug, plug, bubble/plug transitional, bubble as well as annular flows including sensible correctness. Moreover, a
quantitative relationship having the style of ug=AD2B uC was established by the universal fitting and the pattern-
specific fitting with the coefficients of determination R2 approaching to one. In view of the simplicity and the
convenience of vortex generation and pressure measurement, the correlation dimension-based method provided
an effective and practical idea to gas-liquid two-phase flows study.
Figure 4. Sectional view of the bluff body (a) Overall structure; (b) Dimensions of the bluff body in mm (c)
Side view.
Gas-solid flow
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
30
For rib-welding wall as well as smooth wall, an equipment was planned and the identical test was done by Song
[59] for the purpose of verifying the results of CDF. Both computational by using k-ɛ model and results of the
experiment showed that wall erosion produced from the effects of the particle in the particular two-phase flow
was reduced by supplementing ribs on the wall. The wastage of the wall does not change linearly as the height
of the rib. The area of test section inlet is 300 × 125 mm, and the area of test section increases gradually along
the flow direction. Particles are coal ash with 57 µm mean particle size and arc let out influx of 10 kg/min.
When the width of rib width was tantamount to the gap among fibs and under a specific height of rib, it was
utmostly profitable to decrease erosion. In the zone that has relatively small size of the particle, the change of
the rate of wastage undergoes steeply variation, whereas the size of particle passes a criterion, the rate of
wastage undergoes smoothly variation, and for a specific flow. The wall wastage impact by the first angle of the
direction of movement among gas and particles
Bin [60] and Zhou [61] have been investigated vortex structures and particle dispersions inflows around a
circular cylinder. The first group researchers studied gas-solid two-phase flow across a single cylinder and two
tandem cylinders by using Lagrangian-Lagrangian model at high Reynolds number. The settlement, entrainment
and aggregation of solid particles moving with the large-scale coherent vortex structure in the wake of a single
cylinder and between two cylinders numerically investigated, and the effects of St Number on the distribution of
solid particles obtained. The second group researcher’s verification of vortex structures and particle dispersions
inflows around a circular cylinder by lattice Boltzmann method (LBM), with Non-equilibrium Extrapolation
Method (NEM) dealing with the computational boundaries. The Reynolds number (Re = 40-100) effectiveness
on the growth of the structures of vortex studied. Perfect deals of Strouhal number, the lift coefficient, as well as
drag coefficient, reached with prior searches. For both Stokes number and Reynolds number, it discovered that
they provide an important impact on the arrangement of particle. The small particles (St = 0.01) succeed the
fluid movement so strongly and have the ability to scatter within the core zones of the vortex frame.
Yan and Rinoshika [62] built an experiment pipeline test has a horizontal smooth acrylic tube of about 5 m
length and 80 mm inside diameter. The polyethene is particles diameters 2.3 and 3.3 mm with a density of 978
kg/m3 and 952 kg/m3 utilized as test materials. The arrangements of concentration and the time-averaged
particle velocity estimated at 0.45 kg/s and 0.40 kg/s of the flow rate of solid mass as well as the lower air
velocity in fully developed as well as the acceleration regimes. It was noticed that although the velocities of air
for utilizing fins are minimal as compared with that of non-fin, the concentration of particle for utilizing fins
presents nearly the same profile as that of non-fin. It suggests that the long fins easily accelerate the particles
and result in low-pressure drop and low air velocity. Within the low pipeline portion, the vertical particle
velocity reduces, because of the impact of fins oscillation, producing in the dispersal of particle and the decrease
of collision for particles-wall.
Figure 5. Mounted soft fins in a test pipe.
Borello [63] presented two-phase flow in a rib-roughened duct describing the turbine blade internal cooling
channel. Means of the k-ε-ζ-f elliptic relaxation model utilized for measuring of the flow field and validation
versus the data of LES executed. The well-validated T-Flows FV code as utilized in the purpose of calculations.
Particle-laden flow simulated considering actual properties of the particle. Particles are moved by carrier flow
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
31
(here modelled with a steady RANS) utilizing a one-way coupling way in a Lagrangian framework.
Mechanisms of adhesion and impact reduced utilizing the Thornton and Ning way intending at recognizing the
zone at which deposit happens. The rotation impact upon the movements of the particle as well as the flow will
be studied within RANS intending at identifying the deposit regions adjustment while the particle-laden flow
submitted to Coriolis forces and centrifugal.
Figure 6. Comparison of LES and URANS streamlines and mean velocity field in the central vertical plane.
Liquid-solid flow
Pathak [64] numerically investigated the impact of the impediment bed on the dispersal of solid particles in a
flow of two-phase. In order to model this flow, an algebraic slip mixture model utilized and in order to
determine flow turbulence, the re-normalization group k-ɛ model utilized. For the purpose of investigating the
impact of three various sizes of solid particles on turbulence modulation, calculations were presented for them.
Within the region including the wake and stagnation regions, a deformity of the self-similar mean velocity
profile recognized containing the standard layered dispersal of solid particles and that was due to the retardation.
Its length rises with the size of particle. High preferred deposition of particles characterizes stagnation region,
while a low concentration gradient exists within the wake region. As the size of the particle develops, the impact
of wake vortices on particle dispersal reductions. The strength of turbulence develops with the size of the
particle, whereas within complex flow regions for larger-sized particles, comparative turbulence modulation
recognized.
Pathak and Khan [65] made a computational investigation within a solid-liquid two-phase flow in a rectangular
duct in order to investigate the impacts of the size of the particle on flow turbulence and inter-phase slip
velocity. In the simulation, the procedure of finite volume beside renormalization k-ε style and an algebraic slip
mixture style has been utilized. In order to explain the influences of three various sizes of particles upon mean
and turbulent flow characteristics, simulations have been done for these particles. In the recirculation and
stagnation regions, the existence of retardation varies the standard layered arrangement of microparticles where
the recirculation region identified by the low value of solid particles concentration and stagnation region
identified by high value. The slip velocity among the particles and liquid phases has recognized further within
the upstream than the downstream of the retardation. Slip velocities, as well as particles arrangements variation
made due to the existence of retardation, vanish at an appointed downstream distance of the retardation and the
characteristics of flow recapture their un-disturbed cases. The size of the particle is what this settling distance
relies on. The flow turbulence improves by particles, and for partials of large size, the impact in the region of
complex flow has noticed more. Although Stokes number connected with the flow is small, slip velocity as well
as the turbulence have developed because of the disruption of the flow formed by the retardation.
Liquid–Liquid flow
Wong [66] made in the flow of an oil-water mixture within a parallel-plate (single fracture) style, a study of the
pressure gradient/phase saturation relationship. For the relationship, it noticed that it consists on the interaction
among the two flowing fluids, or pattern of flow, that conversely ruled by the flow rate of water, fracture surface
roughness, viscosity ratio, fracture aperture, and injection method. Three distinguished patterns of flow
recognized, namely, mixed, dispersed, as well as channel flows. Determinations of phase saturation, as well as
pressure gradient, propose that the publicly utilized Romm’s proportional permeability correlation viable to
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
32
channel flow in which water oil and phases were continuous in the split. In the flow that in which either the oil
or water phase was discontinuous, for dispersed or mixed flow, a Lockhart-Martinelli kind relationship formed
for the flow of gas-liquid within pipes must be correct not at high viscosity ratio and water flow rate but at low
viscosity ratio.
TWO-PHASE FLOW WITHOUT TURBULATOR
There are many common examples of two-phase flows. Some, such as fog, smog, smoke, rain, clouds, snow,
icebergs, quicksand, dust storms, and mud, occur in nature. Others, such as boiling water, tea making, egg
scrambling, and sugar stirring are frequent occurrences in kitchens and dining rooms.
Gas-liquid flow
In special, the channel permits the investigation of air/water slug flow at atmospheric pressure. Parallel to the
tests, CFD estimations provided by Höhne [99]. Under the HAWAC test facility, a specific inlet device presents
fully defined as well as changeable boundary conditions, that provide so great CFD-code effectiveness
possibilities. A picture series filmed through slug flow matched with the equivalent CFD simulation produced
by the code ANSYS CFX. The two-fluid model implemented with a specific turbulence damping method at the
free surface. An Algebraic Interfacial Area Density (AIAD) model based on the completed mixture model
presented and completed. It improved the physical modelling; discovery of the morphological model and the
identical switching of every relationship was now potential. While variations need continuity of the work, the
behavior of slug generation and propagation at the test system reproduced. Tests similar velocity and pressure
determinations designed and will permit quantitative comparisons at other superficial velocities.
Zeguai [67] done empirical research intends at explaining two-phase flow patterns of air-water for laminar flow
with accurately managed conditions in a horizontal tube. In order to create a two-phase flow in a glass tube
having an internal diameter of 3 mm with co-current water and air flows, empirical test equipment has been
introduced. For superficial velocity of the liquid and gas, the ranges of studies were from 0.78×10-3 ms-1 to
79×10-3 ms-1 and from 2.3×10-3 ms-1 to 3.54 ms-1 sequentially, unlike earlier investigations. a high-speed camera
has been utilized in order to record flow visualizations in the region of the entrance at L/D = 10 in
correspondence to the two phases mixing zone and far away downstream at L/D = 420. The effects utilized for
highlighting the patterns of flow in both regions. Various patterns of the flow namely annular, slug and bubbly
flow pattern gained inside the investigated ranges of superficial velocities. In the entry zone, extra complicated
patterns of flow conceived that develop to uncomplicated and fewer arrangement of flow with smoother,
interfaces downstream. Maps of the flow were drawn and submitted. They explain some rearrangements of the
pattern. In the paper, the physical explanation of the variety within the structure of flow among the two studied
regions and a full explanation has been provided.
McCaslin and Desjardins [68] conducted for three various groups of conditions in the flow regimes of stratified-
annular as well as annular within horizontal pipes, exploratory mathematical simulations of liquid-gas flows. In
order to select dominated parameters in a form that yields flow that were related to factual implementations of
engineering while resting computationally tractable, the accurate dimensional investigation was utilized.
Statistics of the height of the liquid film and the velocity field estimated as a circumferential location function in
the pipe, showing the presence of a viscous sublayer in the liquid film, and a viscous layer near the interface and
a log law zone in the core of gas. As the impacts of gravitational development, it has been noticed that, the dry-
out conditions’ possibility at the wall within the above sections of the pipe has been developed. In order to
inform the mechanisms that are probable for liquid film sustainment, analyzing of circumferential motion of the
phases of gas and liquid in the cross-section of pipe has been done. An uncomplicated model improved which
assists in aids in learning the secondary gas flow impact on the film circumferential motion as well as identify
the liquid annulus dynamics. Comparison for the asymmetry of film, a fraction of void and height of film
against the empirical relationships obtainable in the literature has been done.
Oliveira [69] characterized elongated bubbles at approximately atmospheric conditions in a horizontal pipe
analytically for the flow of air-water. The flow rates range used covered regimes at the transition from the
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
33
elongated bubble to slug. In order to discover the interface of liquid-gas with the helping of a collection of photo
gates for synchronization passage of bubble with the acquisition of image, ensemble-averaged digital image
treatment techniques utilized. Quantitative data of tail, as well as front bubbles portions examined for various
velocities of mixture and the results, proved visible notes usually described during the literature. Near to
transition, a linear inclination of the nose of the bubble to proceed towards the centerline location of the pipe, for
frequently larger amounts of the Froude numbers, quantified as well as recognized. The hydraulic jumps are
noticed to be subject of Froude number and bubble tail shapes quantified, whilst the liquid volume fraction
governed the thicknesses of the liquid film. Variations on the properties of bubbles were clearly connected to
changes in the velocities of bubble and appear to display an opposition among inertia and viscous impacts.
Bae [70] conducted experiments using PIV visualization in a rectangular channel and a parallel wire
conductance sensor, that has 50 mm height and 40 mm width. During the tests, the condition of flow included
the gas Reynolds number Reg range of 14,000 to 70,000 and the liquid Reynolds number Rel range of 450 to
3540. The results showed that the kinds of perceived wave according to the conditions of flow within the
rectangular channel were comparable to those in a horizontal pipe. Complicated trends according to the
gathering of Rel and Reg have been displayed from the interfacial wave slope, frequency and height, which
impacts the wave breakup as well as coalescence. Particularly, the wave slope, as well as wave height, contain
adverse trends about the standard of Reg=34,000. The interfacial drag force significantly impacts the disturbance
wave slope and height for states in which Reg ≥34,000. On the other hand, for Reg < 34,000, the wave
development contains a significant impact on the parameters of the wave. Lastly, new empirical correlations for
the interfacial wave slope, frequency and height introduced for utilization to the improvement of a droplet
entrainment model in a horizontal stratified flow.
It was essential to estimate the impact of surface tension that change by large quantity (air/water = 72 dyne/cm
and gas/oil = 35 dyne/cm) on flow regime map for the purpose of improving flow regime maps generality,
which was done by Bageri [71]. Consequently, the aims of this study are the evaluation of the surface tension
influence upon the flow regime map and address its utilization. Development of a number of flow regime maps
gas-liquid flow area in horizontal pipes has been done. Moreover, some dimensionless groups and parameters of
the mappings are applied as coordinates in order to express the flow regime maps. The combination of
dimensionless number and parameters are utilized for the purpose of generalizing the applicability of the flow
regime maps. Consequently, many of empirical data is needed to create these maps and normally it is not
achievable to estimate the influence of every single parameter (viscosity, density, surface tension, geometry or
pipe size) upon the flow regime map. Evaluating the impact of surface tension flow regimes horizontal pipe for
the water system is the purpose of this research. For the purpose of evaluating the surface tension impact, the
tests were carried out with two-phase water-air system. Decreasing the surface tension with the support of
surfactant has been utilized in order to introduce the surface tension impact. In order to create a flow pattern
map, the test data was used and based on the change in the boundaries of various flow patterns, the surface
tension influence was estimated. For evaluating the impact of the surface on the boundaries of various flow
patterns, four various concentrations (0.01, 0.05 0.1, 0.5%) of surfactant have been chosen.
Silva [72] aimed to compare pressure gradients of natural gas and heavy oil mixture in a horizontal pipe for
various flow patterns utilizing the application Lockhart and Martinelli, ANSYS CFX 13.0, and Beggs and Brill
relationships. The study examined the outcomes for stratified, plug as well as bubbly flows. The results
displayed that Beggs and Brill over prophesied values of the pressure gradient. It additionally noted a great
correspondence among Lockhart and Martinelli and numerical relationship for flows of bubbly and plug,
including 5.78 and 19.55 per cent root-mean-square deviations (RMSD), sequentially. The results of numerical
displayed a weak correspondence as for cases of the stratified flow, including an RMSD higher than 90 per cent.
The high per cent variation for this flow regime was because of the increase in the gas input content. They
proposed the use of free surface and turbulence models and also various values of drag coefficient in the
numerical setup for estimating the high gas velocity influences and, therefore, develop the correspondence.
Eyo and Lao [73] used gas-liquid flows in annulus channels are considerably encountered in the underbalanced
drilling process when the gasified drilling fluid. For gain, a larger knowing of the behaviors of flow in the
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
34
channels, therefore to secure an effective drilling process reached, an exact description of two-phase flow
regimes in such channels is critically essential. Test investigations concerning gas-liquid flows in a concentric
and fully unusual horizontal annulus were described in this paper. The experiment part settings of the flow loop
contain 10.8 m of length, with 0.060 m and 0.0768 m inner and outer diameters of pipe sequentially. At
atmospheric pressure, water and air formed the liquid and gas phases and the gas and liquid superficial
velocities range studied through this research was 0.14 – 24 m/s and 0.15 – 2.78 m/s sequentially. The observed
flow regimes in both annulus setups by high-speed camera imaging were annular, wavy annular, wavy slug,
slug, churn, elongated bubble and dispersed bubble. A complete explanation of the flow regimes including
various characteristics is displayed jointly with high-quality images.
For gaining more penetrations around the properties of the examined flow regimes, the local liquid holdup time
series jointly with its probability density function (PDF) is utilized. The investigating of the impacts of the
annulus eccentricity upon the detected flow regimes are also has been done. It is noticed that the structure as
well as the shape of the annular, wavy annular as well as elongated bubble flow regimes impacts by the annulus
eccentricity. It is further noted that the annulus eccentricity creates the transformation from an elongated bubble
to dispersed bubble for happening at higher superficial velocities of liquid. It is further observed that in the fully
eccentric annulus creates transitions between various flow regimes to happen at lower gas and higher liquid
superficial velocities as compared with that of concentric ones. By basing on liquid and gas Froude numbers
with integrating the results of analysis from this study with over 1000 data points observed in literature, an
enhanced flow regime map is introduced.
Baghernejad [74] used experiment technique to obtaining flow patterns of two-phase and associated transition
boundaries, for measuring flow patterns below various rotational speeds of pipe, an individual test set-up was
created. In order to enable direct noticing of flow pattern, a Plexiglas pipe, having an in diameter of 25.4 mm
and length of 4000 mm was utilized in tests. An electromotor coupled beside gearbox was utilized to turning of
the pipe that let various speeds of rotation. The set-up also carries inclination up to ±25°. The utilized liquid and
gas phases were water and air, sequentially. For the purpose of drawing flow pattern maps at six various
rotational speeds of 0, 50, 100, 200, 300 and 400 rpm in both 10° inclined as well as horizontal pipe, above 3800
tests were reached. findings of the test were examined with earlier study for a horizontal fixed pipe state for
validating the results. From results, it has been noticed that the rotation of pipe has an important impact on the
transition between boundaries and flow pattern map. Within horizontal pipe state, it was noticed that the
stratified smooth flow regime reduces as the pipe rotation rises and at high revolution speeds it vanishes.
Further, by rising pipe rotational speed, the annular regime increases. During a 10° inclined pipe, the stratified
wavy region seems as the pipe rotation speeds up. It was recognized that for both horizontal as well as inclined
pipe, pressure drop extremely develops as the pipe rotational speed improvements. Moreover, the influence of
inclination of pipe on the pressure drop is decreased as the rotational speed rises.
Gas-solid flow
Kussin and Sommerfeld [75] determinations in a detailed way within a developed particle-laden horizontal
channel flow (6 m in length, 35 mm in height, the length is about 170 channel heights) are displayed utilizing
phase-Doppler anemometry for simultaneous measurement of particle velocity and air. Spherical glass beads
particles including mean diameters in the range of 60 µm–1 mm was utilized. The conveying velocity could be
changed among around 10 m/s and 25 m/s, and the particle mass loading could approach values of around 2 (the
mass loading is described as the ratio of particle to gas phase mass flow rates), basing upon the size of the
particle. With the wall plates exchanging, the wall roughness degree could be transformed, for the first time. The
examination of the impact of these parameters and the seven effects of inter-particle collisions on the profiles of
particle mean and fluctuating velocities and the normalized concentration in the developed flow have been done.
It was explained that roughness of the wall reduces the particle mean velocity and improves fluctuating
velocities because of irregular bouncing of the wall and an improvement in wall impact frequency, i.e. decrease
in mean free path. Thereby, the bigger particles are principally further regularly distributed across the channel,
and gravitational settling is decreased. By developing mass loading because of inter-particle impacts and the
loss of momentum included, both elements of the particle velocity variation were decreased. Furthermore, the
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
35
impact of the particles upon turbulent variations as well as the air flow was investigated on the principle of
profiles in the developed flow and turbulence spectra defined for the stream wise velocity component. The
impact of the roughness of the wall was investigated in addition to the influence of mass loading as well as the
size of the particle upon modulation of turbulence. It was obviously noted that due to two-way coupling, within
an intensive turbulence dispersion, developing roughness of wall also occurs.
Figure 7. Installation of the PDA on the computer-controlled traversing system.
Laín [76] analyzed the influences of the size of particle, mass loading ratio of the particle as well as the
roughness of wall at the developing of secondary flow under turbulent conditions in a circular cross-section
horizontal pipe. The calculations based on the Euler-Lagrange suggestion estimating for collisions of inter-
particle as well as the roughness of wall (i.e., four-way coupling). The secondary flow consisting of two cells of
recirculation with an upward flow near the vertical (symmetry) axis and a downward flow waste to the walls if
inter-particle collisions ignored in the state of inertial particles. Moreover, during collisions of inter-particle
estimated for, the pattern depends on the concentration profile of particle besides approximately low roughness
walls (smooth), two recirculation cells located, except with rough walls four recirculation cells created. By
growing the mass loading ratio, a shift between two and four recirculation cells in the secondary flow could be
recognized for smaller particles.
Mallouppas and Wachem [77] scrutinized the Large Eddy Simulation (LES) method for emulation the
performance in a turbulent channel flow for interacting particles. For investigating the significance of the single
physical phenomena happening inflows of particle-laden, a set of simulations that were completely (four-way),
two-way and one way coupled completed. Further, hard sphere as well as the soft sphere forms, that explain the
communication among interfering particles, compared with each other and the disadvantages and benefits of
every algorithm presented. Various styles for describing the sub-grid scale stresses with LES compared.
Eventually, the rough walls of the channel’s emulations considering to compared to simulations among smooth
walls. Discussion of the simulations results was conducted including the aid of the data of test of Kussin J. and
Sommerfeld M., at Reynolds number 42,000 depending upon the whole height of the channel. A three-
dimensional domain of 0.175 m × 0.035 m × 0.035 m were used for the simulations in which the direction of
gravity was perpendicular to the flow. It has been shown from the results of the simulation that collisions of
inter-particle as well as rough walls, still for very dilute flows, contain an essential impact into particles
redistributing over the channel. Introducing of a new model of roughness has been done and it takes into account
the reality that within the model of a soft sphere, a collision was completely resolved and it is explained that the
fresh model was in so great approval with the obtainable data of the experiment.
van Wachem [78] described a total system for prognosticating the behavior of interacting non-spherical
particles within a turbulent flow among high Stokes numbers. For describing the collisions between non-
spherical particles and a rough wall, a model of particle-rough wall interaction established forward too. For
simulation, the behavior of horizontal turbulent channel flow among 5 variously formed particles: a sphere, two
kinds of ellipsoids, a disc, and a fiber, the structure was linked with a DNS-LES program. By using the
correlations that obtained utilizing true DNS, the torque, lift as well as drag forces on the particles have been
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
36
measured. By following their longest axis perpendicular to the local flow direction, the results of simulation
present that non-spherical particles direct to locally maximize the drag force. Making resolved direct numerical
simulations of an ellipsoid in flow produced a more explaining for this phenomenon. These simulations showed
that if an axis of the non-spherical particle not aligned with the flow, the high-pressure zone on the acute sides
of a non-spherical particle occurs in a torque.
If the axis of the particle was perpendicular to the local direction of the flow, this torque was just zero. Further,
if the longest axis was aligned perpendicular to the flow, the particle was most stabilized. Non-spherical
particles holding a higher average velocity balanced to spherical particles having the similar equivalent diameter
caused because of the arrangement of the longest axis of a non-spherical particle perpendicular to the local flow.
It additionally explained that disc-shaped particles flow in a steadier trajectory balanced to elongated particles,
like fibers as well as elongated ellipsoids. This linked to the measure of the pressure gradient on the non-
spherical particles acute side. Eventually, it explained that the wall roughness impact influences non-spherical
particles variously as compared with spherical particles. Especially, a non-spherical particle collision with a
rough wall causes a notable measure of rotational energy, whereas an identical collision with a spherical particle
produces in mostly a variation in translational movement. The size, shape, sphericity and the definition of the
angle of attack of the five particles considered is shown in table below.
Zhou [79] studied the distributions for particle in a fully developed horizontal channel of a turbulent gas-solid
boundary layer, and with particle image velocimetry (PIV) with main upstream speeds of Uh = 2.1 m/s, 5.2 m/s,
and 8.74 m/s, the gas phase velocity field was gained. The particle phase was carborundum powders including
diameters of 38 μm and 60 μm and ghost particles were used that were titanium dioxide nanoparticles. It noticed
that within a logarithmic region, the peak of the particle distribution happens. The peak move toward the wall
according to an expansion of particle size. The peak moves toward the center of the channel as the velocity
improvements. The sweeps, as well as gravity, contain impact toward the motion of particle motion to the wall;
collisions of particle–wall, ejections, and the Magnus lift force contain impact toward the motion of particle to
the center of the channel. The presence of ejections, as well as sweeps created by quasi-flow vortices, has
approved within the examination of shear strain distribution and the velocity field within the gas boundary layer.
Meantime, it observed that the peak of shear strain happened near the wall and that wall-normal inconstancy and
shear strain through high flow rate was significantly larger than at low flow rate; these properties were uniform
with the distribution of the particles within the boundary layer.
Abdelhamid [80] conducted a detailed experimental investigation within ejectors for studying the (air-air)
single-phase and the (air-solid) two-phase flows. A try produced via presenting modifications of design for the
mixing part of the ejector for developing the quantity of solids that transported. The duct of mixing has
increased by a tail part. Mixing ducts of three various geometries besides ejector tail section has designed,
fabricated as well as experimentally examined. The impacts of tail part forms, as well as the mixing duct on the
performance of ejector, have been studied. Moreover, the impacts of the solid particles mass flow rate, as well
as air motive pressure against the produced vacuum pressure as well as the static pressure distribution, have
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
37
been investigated too. The results collected explained that the mixing duct geometry of convergent-constant-
divergent provides suitable results showing of the ejector as well as greater vacuum pressure. Inlet particle
volume fraction, particle phase material density, as well as particle diameter, were three major parameters
dominating the flow physics of dispersed gas-particle flows.
Liquid-solid flow
Toda [81] numerically analyzing of cared out the limit-deposit velocity at which in the bottom of a pipe, a
deposit of solid starts to build-up on the basis of a two-dimensional flow model and a compare is formed with
the results of the experiment. The model describes fully the reality that with an improvement in the solids,
concentration can be achieved, the limit-deposit velocity raises and it reduces progressively at greater
concentrations next to attaining a maximum point at a specific concentration. It is explained that the shear stress
effecting on the surface of the bed of particles is the governing factor for moving the solids at weaker solid
concentrations, while the drag force of fluid in the bed presents a highly significant performance at greater
concentrations.
Transmission of Solid-liquid slurry has been studied [40]. Within a horizontal slurry flow, novel data of test on
pressure drop was obtained. For the prognostication of flow properties in solid-liquid flow, a theoretical model
was produced. The theory is dependent on a model of a two-layer. The solid-fluid characteristics are; the flow
rate of slurry, the geometry of conduit as well as input concentration are defined, it permits the prognostication
of flow patterns as well as pressure drop. The prognostications model is examined to the novel experimental
data and to some generally applied relationships.
The experiment or theoretically investigation of the flow of solid-liquid mixture at low velocities was done by
Takahash [82]. The creation of dunes within the pipe is a standard flow property that within their motion it
creates a pressure vacillation. As the mixture mean velocity developed, the dunes velocity improved. As the
concentration of solid developed up to 10%, the fluctuation of pressure developed somewhat. For investigating
the periodicity, the Power Spectral Density function (PSD) of the fluctuation of pressure was collected too. The
presence of a dominant frequency has been shown from the results. This dominant frequency was linked to the
dunes motion and it improved according to the dune’s velocity improvement. Depending on a supposed PSD,
the fluctuation of pressure was simulated numerically. The results of pressure fluctuation that simulated were in
great dealing with the tests.
Doron and Barnea [83] studied various flow patterns which could be encountered in a solid-liquid pipeline, that
affect the hydrodynamic characteristics of the flow. A method for displaying their range of existence by means
of flow maps was presented. The maps were drawn from the results of a phenomenological model. The effect of
the various operational conditions on the flow pattern transitions was obtained easily using the maps.
Figure 8. Superficial velocities flow pattern map, ρs = 1240 kg/m3, D = 50 mm, dp = 3 mm, ----- three-layer
model, - -, Turian et al. (1987) correlation, ---- Turian & Yuan (1977) correlation, ○ experimental data (Doron
& Barnea 1995b).
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
38
Figure 9. Mixture velocity-delivered concentration flow pattern map, ρs = 1240kg/m 3, D = 50ram, dp = 3 mm.
For the purpose of obtaining the numerical solution in a sand-water slurry flow, a simplified 3D Algebraic Slip
Mixture (ASM) model was submitted by [84]. The turbulent model of RNG k–ɛ was applied beside the ASM
model in order for the investigation to achieve the precise numerical solution in a flow of fully developed
turbulent. For solving the dominant equations, a control volume finite variation process was applied, and to
discretize the entire computational domain, an unstructured (block-structured) non-uniform grid was selected.
The comparison of the mean pressure gradients from the numerical solutions with the experimental data was
carried out. When the slurry mean velocity is greater than the corresponding critical deposition velocity, the
solutions were observed to be in great correspondence. Furthermore, several major properties of slurry flow
have been displayed from the numerical studies, like slurry density, volume fraction distributions, slurry mean
skin friction coefficient distributions, slurry mean velocity distributions and slip velocity magnitude within a
fully developed region, that have never been presented in the tests.
Tanaka [85] executed an analysis in a solid-liquid two-phase flow on the interaction between Karman vortices
and concentrated particles (hereafter named the Cluster). Utilizing the moving of a cylinder within a shoal
region container, the Karman vortices have gained. The particle density was approximately similarly as that of
liquid. For distinguishing the Cluster velocity from the vector area of the around flow, we have utilized high-
resolution PIV. It decided that the Clusters improve the flow along the vortices and the steady rigid body
rotation of the Karman vortices from the data that collected. Moreover, the Clusters remain their rotational
movement and do not support the flow along the vortices totally.
Figure 10. The towing tank.
Kaushal [86] conducted experiments within a horizontal pipe which has a diameter of 54.9 mm on two sizes of
glass beads of which mean diameter and geometric standard deviation are 440 µm & 1.2 and 125 µm & 1.15,
sequentially, and a mixture of the two sizes in similar fraction by mass. Flow velocity was up to 5 m/s and for
each velocity, the overall concentration up to 50% by volume. Concentration profiles, as well as pressure drop,
estimated. The profiles presented crossing isokinetic sampling examinations in vertical planes, 45˚ inclined as
well as the horizontal having the pipe axis. Samples of slurry from the mixture, obtained in the vertical plane
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
39
investigated for concentration profiles of every particle batch forming the mixture. It was discovered that the
pressure drops reduced for the mixture at large concentrations except for 5 m/s and a special variety of
concentration profiles was recognized for particles of 440 µm showing a sliding bed regime, while in the
horizontal plane, the profiles stay relatively fixed irrespective of the kind of slurry, overall concentration as well
as flow velocity.
Goharzadeh and Rodgers [87] an experimental investigation on solid particle conveys through a horizontal pipe
of gas-liquid slug flow with two kinds of tests carried. By utilizing high-speed photography, the impact of the
length of the slug on transportation of solid particle is described. The division of velocity through the slug body
is estimated by utilizing mixed Particle Image Velocimetry (PIV) with Refractive Index Matching (RIM) and
fluorescent tracers (two-phase oil-air loop). A penetration is given toward the physical mechanism of solid
particle transportation due to slug flow by mixing these empirical analyses. It was recognized that mobility of
the solid particle is largely affected by the slug body. It was discovered that the physical mechanism of solid
particle transportation was discontinuous. The inactive region (inwards of solid particle transportation) upstream
of the slug nose was quantified as a function of the size of solid particle and composition of gas-liquid. An
important drop in the quantity of velocity instantly upstream of the slug nose and subsequently the critical
velocity for solid particle lifting is touched additional upstream and all of that have been noticed from the
calculated velocity distributions.
Capecelatro and Desjardins [88] performed calculations of liquid-solid slurries within horizontal pipes for
investigating the dynamics of complex multiphase flow joined with conditions of operating over and under the
velocity of critical deposition. Under a flow of fully developed turbulent, a high-fidelity simulation frame was
connected beside a Lagrangian particle-tracking solver to estimate polydispersed settling particles. The two
phases completely linked by momentum exchange terms as well as volume fraction, and a two-step filtering
method applied to relieve each reliance of the liquid-phase mesh size on the diameter of particle, allowing the
arrest of an extended domain of spatial turbulent scales. An employment of completely stable immersed
boundary system for estimating the geometry of pipe on a regular Cartesian mesh has been done. Two states
simulated, every with a geometry of pipe and particle size distribution meeting an empirical investigation of
Roco & Balakrishnam, which estimates a mean volumetric solid concentration of 8.4%, matching to only above
16 million particles. The first state estimates a Reynolds number depending on 85,000 of the liquid bulk flows,
occurring within a heterogeneous suspension of particles in all places of the cross-section of pipe.
Best correspondence with the results of test has been shown from statistics on the particle phase velocity as well
as concentration for this state. In regards of a lower Reynolds number of 42,660 in second case has been done
which leads to the creation of a fixed bed of particles. Corresponding to a rigid bed at the bottom of the pipe, an
extremely collisional shear flow just above the bed, and a dilute suspension of particles far away from the bed
are three distinguished areas recognized within the second state. Computational results indicated division within
the size of the particle over the vertical direction, by the smallest particles placed at the top, growing
monotonically till the surface of the bed, at which the biggest particles were placed. The velocity, as well as
concentration covariance of every phase, introduced, providing more penetration on the dynamics of multiphase.
For every state, the statistics on the single mechanisms that provide to each particle movement, namely forces
due to drag, the surrounding fluid viscous stresses and pressure gradient, and collisions, have been provided. It
recognized that for the preponderance of the pipe cross-section, the drag force controls for every state, that
stabled by gravity in the vertical direction and by collisions of inter-particle in the stream wise direction. For
investigation of closures from Reynolds average modeling of multiphase flows, simulation results are utilized
too.
Titus and Aidil [89] studied simulation of a computational fluid dynamics (CFD) which approves the Eulerian-
Eulerian inhomogeneous two-fluid form in ANSYS CFX-15 which was utilized to investigate the impact of (90
µm to 270 µm) size of particle and (10% to 40%) in situ particle volume fraction on the radial arrangement of
the concentration of particle and velocity and frictional pressure loss. The robustness of different turbulence
models like eddy viscosity transport, the k-epsilon (k-ɛ), SSG Reynolds stress, k-omega (k-ω), shear stress
transport was examined in prognosticating data of test data of particle concentration profiles. The style of k-
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
40
epsilon exactly met the experimental data strongly as compared with the other models of turbulence. As the size
of the particle developed at fixed particle volume fraction, a reduction in frictional pressure loss has been shown
from the results. Moreover, the radial distribution of concentration of the particle developed with developing the
size of the particle, where at the bottom of the pipe, a large concentration of particles happened, for a constant
particle volume fraction. Particularly for the large particle volume fraction of 40%, particles of size 90 µm were
approximately buoyant. The CFD study explains that if the perception of pipeline wear, particle attrition, or
agglomeration is to be forward, observations of the variation of these parameters with pipe position is very
important.
Leporini [90] noticed an ingrained issue for both the production of natural gas and oil that was sand particles
precipitation within the pipeline, that could guide to difficulties like the descent of production, immoderate
pressure drops, erosion of pipeline, and defeat of equipment. Within various operations of flow like sand
multiphase mixtures, the description of sedimentation, as well as transport of sand particles, is necessary to
prognosticate entrainment processes within carrying pipelines of gas and oil as well as the transport velocity of
sand. Despite, it appears that no model subsists having the ability to carefully describe the deposition as well as
transport of sand within the multiphase pipeline. Actually, during the previous contract, some investigators
attempted to develop the modeling of liquid-solid flow to gas-liquid-solid flow, however, no respectable
outcomes have been achieved, mainly in slug flow condition because of the complication of the phenomenon.
More and more experimental data are required for developing and validating a mathematical model suitably
expressed for sand critical deposition velocity estimation within the flow of gas-liquid. A preparatory
experimental study within a horizontal pipe for three-phase flows (air-water-sand) submits in this paper and the
sand-liquid model's application submits in literature. Through the experimental study, important considerations
were presented from which some inferences were formed. The establishing of various sand flow regimes were
done using physical investigation and data analysis: stationary bed, moving dunes as well as fully dispersed
solid flow. The determining of critical deposition velocities at various concentrations of sand were done. It was
achieved that the critical deposition velocity, as well as properties of sand transport, are completely based on the
concentration of sand as well as on the flow regime of gas-liquid.
(a) (b) (c)
(d) (e)
Figure 11. Various kind of observed sand flow regimes: (a) Suspension, (b) Moving bed, (c) and (d) Moving
dunes, (e) Stationary bed (images of the bottom of the pipe) - the flow direction from the right to left.
Liquid-liquid flow
The different results. as well as notions undergone in gas-liquid of two-phase flows, cannot easily be turned to
liquid-liquid operations; a basis for creating a general two-fired flow pattern map was formed from an effort that
made by [91]. The variation from a steady layered arrangement to another bounding flow patterns has
investigated the representation of the relations between the states for the actuality of characteristics and
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
41
instability stander, that result from searching the dominated equations stability as well as well-posedness.
Basing on mechanistic models, the transitive boundaries among the other flow patterns met in operations of
liquid-liquid are achieved. For enormous domains of physical and geometry characteristics, a parametric study
presented, as met liquid-liquid operations, involved too. Sensible correspondence has been shown in
comparisons of the introduced transitional standard with (limited) obtainable data liquid-liquid operations. The
meeting of the highly viscous core flows data, on the one hand, and the common standards to the excess data of
gas-liquid, on the other hand, is satisfying.
Both experimentally and theoretically studying of the transition of Oil/water flow-pattern within horizontal
pipes have been done [92]. A new state-of-the-art oil/water test facility was designed, constructed, and operated.
A transparent test section (5.013 cm inside diameter × 15.54 m long) can be inclined at any angle, to study both
upward and downward flow simultaneously. Mineral oil and water were the working fluids (µo/µw= 29.6, ρo/ρw
=0.85. and σ = 36 dynes/cm at 25.6°C). Stability analyses exhibit that the stratified/unstratified transition should
be treated with the total two-fluid model. While the ST&MI flow pattern is prognosticating by inviscid KH
theory prognosticated, the viscous Kelvin-Helmholtz (KH) analysis are used for prognosticating stratified flow.
The prognosticated drop sizes from the Hinze and Levich models are adjusted to estimate for the impact of the
concentration of dispersed phase, for the dispersed flow pattern. The water fraction is the dominating parameter
for the coalescence phenomena. The execution of the model is premium and matches with published data fully.
Figure 12. Experimental flow-pattern map. Figure 13. Experimental flow-pattern map (mixture
velocity).
Angeli and Hewitt [93] used two methods for the identification of flow pattern, namely definition of the local
phase fractions with a high-frequency impedance probe and high-speed video recording, while the recognition
of the continuous phase in dispersed flows has been done with a conductivity needle probe. Determinations are
done for input water volume fractions from 6% to 86% and mixture velocities ranging from 0.2 to 3.9 m/s.
Several various flow patterns, changing from stratified to fully mixed recognized over this field of conditions.
Annular flow did not seem. Generally, at lower mixture velocities within the steel pipe, the mixed flow pattern
seemed as compared with that within the acrylic pipe, where, also, the continuous phase for a wider range of
conditions was oil. By utilizing the high-frequency impedance probe, the visible noticing was regular with the
determinations. The arrangement of the phases changed dramatically between the acrylic pipes and the stainless
steel, within specific conditions limits. The average in-situ velocity ratios of the two phases in the acrylic pipe
estimated from the phase distribution determinations were, in common, lower than unity.
Fairuzov [94] used a 16-in pipeline carrying light crude oil during the tests. In order to control the input water
volume fraction, the line was joined to a network of freshwater. The oil flow rate controlled by a gate valve
placed at the inlet of the pipeline. Estimating the transversal water fraction profile was used for determining the
transition from stratified flow to dispersed flow. A specific device, the multi-point sampling probe, was
designed and placed within the pipeline for this objective. Mobile sampling tubes are present with the probe that
support getting samples together at six points over the pipe diameter. For minimizing the impact of the probe on
the determined water fraction profile, the retraction rate of every sample was regulated using a needle valve
according to the velocity of the mixture. A standard system for measuring the water fraction in crude oils was
utilized for analyzing the samples for water content in a laboratory. A flow pattern map was formed, based on
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
42
the data collected. The empirical stratified/unstratified transition boundary was matched with two theoretical
standards received in the linear stability analysis of stratified two-phase liquid-liquid flow. The outcomes of this
investigation can be helpful for the validation of unltifield multidimensional models of two-phase flow and for
the design and operation of pipelines carrying crude oil.
Shi [95] have been experimentally investigated in a horizontal pipe that has a diameter of 10 cm, two-phase oil-
water flows for studying the surfactant impact on distributions of oil-water. At input water cut of 20 per cent and
below, results explained that the water layer velocity is lower than mixed layer velocity up to an input mixture
velocity of 1.6 m/s. Nevertheless, at 40 per cent input water cut and higher, the water layer velocity is lower
than the mixed layer velocity up to an input mixture velocity of just 0.8 m/s. At the medium input water cuts
between 40 and 60 per cent, oil and water are very easier to be mixed. The degree of oil-water mixing flow
improves by the addition of surfactant. By improvement of the concentration of surfactant, the water layer
vanishes, mixing of water and oil at lower mixture velocity begin, and the homogeneous flow pattern was seen
at a very lower velocity of input mixture. Further, the mixed layer involves a very higher fraction of the pipe.
These show that at lower input superficial mixture velocity besides surfactants in oil-water flows, corrosion
could be decreased.
Oil–water two-phase flow experiments were conducted [96] within a length of 15 m, a diameter of 8.28 cm,
inclinable steel pipe utilizing mineral oil (830 kg/m3 density and 7.5 mPa s viscosity) and brine (1060 kg/m3
density and 0.8 mPa s viscosity ). Steady-state data on flow patterns, two-phase pressure gradient and holdup
were achieved across -5°, -2°, -1.5°, 0°, 1°, 2° and 5° of the entire range of flow rates for pipe inclinations. By
noticing recorded movies and via analysis of the relative deviation from the homogeneous performance, the
classification of boundaries of flow patterns and their characterization was obtained. A stratified wavy flow
pattern including no mixing at the interface was recognized in an upward and a downward flow. For exact
determination of the absolute in situ volumetric fractions (holdup) of every phase for whole flow patterns, two
gamma-ray densitometers provided. Comprehensive results of two-phase pressure gradient and hold up as a
function of inclinations, flow pattern and the superficial velocities were described. The new experimental data
were matched with outcomes of a flow pattern subordinate prognostication model, which utilizes the
homogeneous model for dispersed flow and the area-averaged steady-state two-fluid model for stratified. For
whole flow patterns noticed, prognostication rigor for pressure gradients and oil/water holdups were displayed
as a function of pipe inclination. There is a possibility for perfection for in specific dual-continuous flow
patterns.
Vielma [97] studied the characteristics of oil-water flow such as pressure drop, flow patterns, droplet size and
phase fraction, as function of flow patterns in horizontal pipes. The tests were conducted in a 5.08 cm. Various
data were acquired from horizontal test section utilizing water and mineral oil (0.85 g/cm3 of density and 15 cp
of viscosity) beside ranging from0.025 m/s to 1.75 m/s for superficial velocities. For measure droplets,
identifying flow patterns, a high-speed video camera was utilized and to gain phase distributions, ten
conductivity probes were utilized. For fully dispersed flows, three probabilistic distributions have been
experimented. Over the pipe diameter, Sauter Mean Diameter (SMD) analysis was conducted. For flow pattern
of dispersed oil in water and water (D o/w & w), an experimental relationship to prognosticate the SMD profile
of droplets over cross section of the pipe was improved. The experimental relationship provided agreeable
outcomes. Comparisons of model explained that none of them could certainly describe the data of test. The
novel data can guide to better modeling and design of dispersed systems and the new data on droplet sizes can
have an important influence on the design of separator. Furthermore, the description of production logs in
horizontal wells heavily relies on the performance of flow.
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
43
Figure 14. Experimental Observations Compared with
Trallero's Theoretical Transition Boundaries.
Figure 15. Experimental Pressure Gradients.
Grassi [98] presented experimental flow-pattern maps as well as pressure drops connected to oil and water flow
within slightly inclined as well as horizontal pipe, where the selected liquids characterized by about 800:1 of an
oil-to-water viscosity ratio at 20 °C. Different theoretical models have examined, with special notice to core-
annular flow two-fluid model and oil-in-water dispersion homogeneous no-slip model for the prognostication of
connected pressure drops, and flow-pattern map transition criteria including the regimes encountered in the
empirical experiments. The theoretical prognostications have then matched to the experimental outcomes. A
satisfying correspondence has gained particularly as interests pressure drop comparisons. Boundaries
superimposed on the identical flow-pattern maps, the ‘free’ parameters have provided based on observations and
experimental results, and the last correspondence was great in the prognostication of both the transition to oil-in-
water dispersion and the core-annular flow region of presence, as regards the prognosticated transition. No
inference can be represented on transition rules including stratified flow, which just hardly has noticed in the
completed tests.
Wang and Gang [100] had been made an experimental study of high viscosity mineral oil-water flow through a
horizontal pipe loop. Results indicate that phase inversion for oil phase with high viscosity occurs much earlier
than low viscosity oil, and phase inversion tends to be delayed, with the increment in experimental temperature.
The influence of mixture velocities on the inversion process could be neglected in the range of mixture
velocities that we studied. Phase inversion for oil phase with high viscosity occurs much earlier than low
viscosity oil, and inversion point tends to be delayed with the increment in experimental temperature. Inversion
does not take place at the same time and at all locations in the pipe loop. It starts at a certain location and grows
as it flows downstream.
Oliemans [101] verified the oil/water model approach in detail by using pipe flow data for an oil/brine system
with a viscosity ratio of 9.5. Two very simple modeling approaches have considered the two-fluid model for
stratified oil/water flow with complete phase separation and the homogeneous model for dispersed oil/water
pipe flow conditions. That is not the case for the transitions between stratified-flow and stratified-flow with
mixing. The current flow pattern maps over-estimates the stratified flow region. A transition to stratified-flow
with mixing occurs already at much lower phase velocities than currently predicted. The calculation errors for
the two-fluid model version for stratified flow were quite large, in particular for the oil/water pressure losses.
Typically, average errors of 20-40% and maximum errors of about 100% are quite common. Although the
average errors for the water hold-ups were lower (about 10%), here the calculations were also rather inaccurate
with maximum errors of 50%. When the model used in an inverse mode to compute from measured water hold-
up and oil/water pressure loss the oil and water superficial velocities, the errors were unacceptably high: average
errors of 45% with maximum errors of 200-300% clearly make these results not suitable to determine
production rates in oil wells in a reliable way.
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
44
Flow pattern transition prediction models were presented [102] for oil-water flow in horizontal pipes. The
transition between stratified and unstratified flow is predicted using Kelvin–Helmholtz (KH) stability analysis
for long waves. New, simplified, and more practical physical mechanisms/mechanistic models were proposed
for the prediction of the transition boundaries to semi dispersed and to fully dispersed flow. The proposed flow
pattern classification significantly simplifies the flow pattern map for liquid-liquid flow and agrees well with the
experimental data.
Desamala [103] presented complete investigation on the prognostication of radial distribution of velocity,
volume fraction and pressure of a pair of immiscible liquids within a horizontal pipeline as well as flow pattern
maps by computational fluid dynamics (CFD) simulation utilizing ANSYS FLUENT 6.3. The fluid pair for
research that has been used was moderately viscous oil and water. For prognosticating different patterns of flow
by co-axial flow, unsteady flow assuming, constant liquid characteristics and immiscible liquid pair, volume of
fluid (VOF) system has been applied. Chosen 47 037 number of quadrilateral mesh components for the entire
geometry, from the grid independent research. Simulation strongly prognosticates about all the flow patterns
(slug, plug, stratified wavy, annular and stratified mixed), excepting dispersion of water in oil as well as the
dispersion of oil in water. The results that simulated are certified with test results of flow pattern map and oil
volume fraction. Radial distribution velocity profile, volume fraction and pressure explain the reality of flow
pattern of the annular, stratified mixed and stratified wavy. These profiles support to advancing the
phenomenological relationships of interfacial properties in two-phase flow.
Shi [104] employed the CFD package FLUENT to horizontal oil-water flow simulation in some various regimes
of flow (core annular flow, oil plugs/bubbles in water and dispersed flow) with matched density and medium
viscosity ratio (µo / µw =18.8). The oil-water flow modelling performed by the volume of fluid (VOF)
multiphase flow process and the SST k-ω scheme with turbulence damping were applied to simulate. For a core-
annular flow (CAF) case, the impacts of wall contact angles, as well as the turbulence schemes on the results of
simulation, were studied. The comparison between the results of simulation and experimental counterparts were
made. Satisfying a good correspondence in the prognostication of flow patterns for oil plugs/bubbles in water
were achieved and CAF.
The results of simulation also showed several particular flow properties of CAF with approximately low-
viscosity oil (oil viscosity one order higher than the water viscosity in the existing research matched to the
largely investigated CAF with oil viscosity being two to three orders higher than the water viscosity). Various
from the high-viscosity oil CAF velocity profiles where there is an intense difference in the velocity gradient at
the phase interface with velocity over the oil core being roughly flat, there is no intense difference in the
velocity gradient at the phase interface for CAF with approximately low-viscosity oil.
Santos [105] carried out mixtures of paraffin oil in water including various oil phase concentrations which were
studied in an empirical and a simulation investigation. As regards to the empirical investigation three procedures
were utilized to obtain data on the distribution of oil phase within the pipe cross-section: Electrical Impedance
Tomography (EIT), sample collection in various radial positions for oil quantification, and photographs of the
flow. Concerning the EIT system, a relative protocol was utilized in the acquiring because it permits for further
accurate reconstruction of the images in the stratified flows state. For simulations carried principally in 2D, the
COMSOL Multiphasic modeling software was utilized, for describing the phenomenon in the oil/water interface
in joining with the k-ε turbulence model, the level set model was using.
The very high computational time required for simulations of 3D carried which did not show further accurate
too. In a present pilot rig, the experimental tests were achieved. superficial oil inlet velocity varied between 0.17
and 1.27m/s superficial water inlet velocity in the interval between 0.30 and 0.45 m/s, during the course of the
investigation. The dominance variable was a pressure drop across the test section, and a great suitable was
reached between simulations in a 2D geometry and tests, 9.7% was the highest variation between the empirical
and simulated pressure drops.
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
45
CONCLUSIONS
This review paper presented a number of previously employed experimental and numerical two-phase flow
systems (gas-liquid, gas-solid, liquid-solid, and liquid-liquid) reported in the literature. From these studied,
some conclusions are reported:
1. The previous studies are divided into two main groups: the first where the researches did an actual project that
needs to be studied, the second where assumed case study which achieved to understand the behavior and action
of different fluids.
2. Most previous studies depend on the experimental method using of visualization optical methods such as
direct photography, laser-induced fluorescence, laser Doppler velocimetry or phase Doppler anemometry,
particle image or tracking velocimetry a digital camera, and this method lacks precision to determine the
difference of pressure and distribution of velocity through the tubes or ducts.
3. Computational methods are promising in the view of the availability of fast computing tools. Investigations
have been very limited to the basic process of bubble bursting and droplet generation. However, the CFD tools
have great potential to model the basic two-phase processes taking place at the interface with techniques like
Level set methods, VOF, Volume tracking and front tracking methods etc.
4. There is a wide area of research in the field of two-phase flow especially in liquid-solid and liquid-liquid,
which are important areas of industry and applied research.
There is a further need for a review of experimental and numerical works done in this field, in order to assist
researchers in setting up their numerical simulations and finding gaps in it. The target of the present review is:
1. To highlight the developed measurement techniques for two-phase systems and provide more experimental
studies under various conditions, which leads to more accurate conclusions in terms of measuring the pressure
difference and velocity of the fluids.
2. To report the employ numerical simulations of two-phase flows to confrontation challenges and problems.
Also, the key findings of numerical research in points of using the models, discretization, and case setting.
3. To more achieving research that focuses on the behavior of liquid-liquid two-phase flow.
4. To discuss the different types of turbulator and study their effect on flow pattern and change pressure
difference in the closed conduit.
REFERENCES
[1] E.B. Christopher, “Fundamentals of multiphase flows,” Cambridge Univ. Press UK, 2005.
[2] G.B. Wallis, One-dimensional two-phase flow. McGraw-Hill, 1969.
[3] G.B. Wallis, “Annular two-phase flow: GF Hewitt and NS Hall Taylor, Pergamon Press, $20, 310 pp.”
Pergamon, 1972.
[4] G.H. Yeoh, and J. Tu, “Basic Theory and Conceptual Framework of Multiphase Flows,” Handb.
Multiph. Flow Sci. Technol., Pp. 1–47, 2017.
[5] G.F. Hewitt, J.M. Delhaye, and N. Zuber, “Multiphase science and technology”, vol. 2. Springer Science
& Business Media, 2013.
[6] Z.K. Gao, N.D. Jin, and W.X. Wang, “Nonlinear analysis of gas-water/oil-water two-phase flow in
complex networks”, Springer Science & Business Media, 2013.
[7] G. Falcone, G. Hewitt, and C. Alimonti, “Multiphase flow metering: principles and applications”, vol.
54. Elsevier, 2009.
[8] S.I. Pai, “Two-phase flows”, vol. 3. Springer-Verlag, 2013.
[9] C. Rogero, “Experimental investigation of developing plug and slug flows” Technische Universität
München, 2009.
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
46
[10] S.G. Kandlikar and W.J. Grande, “Evolution of microchannel flow passages--thermohydraulic
performance and fabrication technology,” Heat Transf. Eng., vol. 24, No. 1, Pp. 3–17, 2003.
[11] R. Ibarra, C.N. Markides, and O.K. Matar, “A review of liquid-liquid flow patterns in horizontal and
slightly inclined pipes,” Multiph. Sci. Technol., Vol. 26, No. 3, 2014.
[12] P. Angeli, “Liquid-liquid dispersed flows in horizontal pipes.,” 1996.
[13] L.M.K. Boelter and R.H. Kepner, “Pressure drop accompanying two-component flow through pipes,”
Ind. Eng. Chem., vol. 31, No. 4, Pp. 426–434, 1939.
[14] R.W. Lockhart and R.C. Martinelli, “Proposed correlation of data for isothermal two-phase, two-
component flow in pipes,” Chem. Eng. Prog., Vol. 45, No. 1, Pp. 39–48, 1949.
[15] O. Baker, “Design of pipelines for the simultaneous flow of oil and gas,” in Fall Meeting of the
Petroleum Branch of AIME, 1953.
[16] R.M. Sharp, “Two-phase pressure losses in valves and fittings.” Georgia Institute of Technology, 1956.
[17] M. Wicks III and A.E. Dukler, “Entrainment and pressure drop in concurrent gas‐liquid flow: I. Air‐
water in horizontal flow,” AIChE J., Vol. 6, No. 3, Pp. 463–468, 1960.
[18] E.S. Kordyban and T. Ranov, “Experimental study of the mechanism of two‐phase flow in horizontal
tubes,” in Multiphase Flow Symposium, ASME, 1963.
[19] D.S. Scott, “Properties of Cocurrent Gas-Liquid Flow,” Adv. Chem. Eng., Vol. 4, No. C, Pp. 199–277,
1964.
[20] M.G. Hubbard, “The characterization of flow regimes for horizontal two-phase flow,” Proc. Heat Transf.
Fluid Mech. Inst., Pp. 100–121, 1966.
[21] Y. Taitel and A.E. Dukler, “A model for predicting flow regime transitions in horizontal and near
horizontal gas‐liquid flow,” AIChE J., Vol. 22, No. 1, Pp. 47–55, 1976.
[22] S.L. Soo, “Fully developed turbulent pipe flow of a gas-solid suspension,” Ind. Eng. Chem. Fundam.,
Vol. 1, No. 1, Pp. 33–37, 1962.
[23] F.E. Marble, “Mechanism of particle collision in the one-dimensional dynamics of gas-particle
mixtures,” Phys. Fluids, Vol. 7, No. 8, Pp. 1270–1282, 1964.
[24] E. Kouk, A. Meeting, and H.E. Mccarthy, “Turbulent Flow of Gas = Solids Suspensions,” Vol. 7, No. 3,
Pp. 471–483, 1968.
[25] T.J. Kramer, and C.A. Depew, “Analysis of Mean Flow Characteristics of Gas-Solids Suspensions,” J.
Basic Eng., Vol. 94, No. 4, Pp. 731, 2010.
[26] Y. Tsuji and Y. Morikawa, “LDV measurements of an air-solid two-phase flow in a horizontal pipe,” J.
Fluid Mech., Vol. 120, Pp. 385–409, 1982.
[27] R.A. Gore and C.T. Crowe, “Effect of particle size on modulating turbulent intensity,” Int. J. Multiph.
Flow, Vol. 15, No. 2, Pp. 279–285, 1989.
[28] R. Durand, “Basic Relationships of the Trnsportation of solids in Pipes-Experimental Research,” Intern.
Assoc. Hydr. Res., 5th Congr. Minneapolis, 1953.
[29] I. Zandi and G. Govatos, “Heterogeneous flow of solids in pipelines,” J. Hydraul. Div., Vol. 93, No. 3,
Pp. 145–159, 1967.
[30] H.A. Babcock, “Advances in Solid-Liquid Flow in Pipes and its Application,” Ed. by Zandi, I, Pp. 125,
1971.
[31] A.J. Carleton, R.J. French, J.G. James, B.A. Broad, and M. Streat, “Hydraulic transport of large particles
using conventional and high concentration conveying,” in Proc. Hydrotransport, Vol. 5, 1978.
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
47
[32] R.P. Chhabra and J.F. Richardson, “Hydraulic transport of coarse gravel particles in a smooth horizontal
pipe,” Chem. Eng. Res. Des., Vol. 61, Pp. 313–317, 1983.
[33] K. Noda, H. Takahashi, and T. Kawashima, “Relation between behavior of particles and pressure loss in
horizontal pipes,” in Proceedings 9th International Conference on the Hydraulic Transport of Solids in
Pipes D, 1984, Vol. 4, Pp. 191–205.
[34] D.M. Newitt, “Hydraulic conveying of solids in horizontal pipes,” Trans. Instn Chem. Engrs, 1955.
[35] C.A. Shook and S.M. Daniel, “Flow of suspensions of solids in pipelines: Part I. Flow with a stable
stationary deposit,” Can. J. Chem. Eng., Vol. 43, No. 2, Pp. 56–61, 1965.
[36] J.W. Hayden and T.E. Stelson, “Hydraulic conveyance of solids in pipes,” in Advances in Solid–Liquid
Flow in Pipes and its Application, Elsevier, 1971, Pp. 149–163.
[37] R.M. Turian and T. Yuan, “Flow of slurries in pipelines,” AIChE J., Vol. 23, No. 3, Pp. 232–243, 1977.
[38] Y. Televantos, C. Shook, M. Streat, and A. Carleton, “Flow of slurries of coarse particles at high solids
concentrations,” Can. J. Chem. Eng., Vol. 57, No. 3, Pp. 255–262, 1979.
[39] G.A. Wani, B.P. Mani, D.S. Rao, and M.K. Sarkar, “Studies on hold-up and pressure gradient in
hydraulic conveying of settling slurries through horizontal pipes.,” J. Pipelines, Vol. 3, No. 3, Pp. 215–
222, 1983.
[40] P. Doron, D. Granica, and D. Barnea, “Slurry flow in horizontal pipes—experimental and modeling,” Int.
J. Multiph. Flow., Vol. 13, No. 4, Pp. 535–547, 1987.
[41] T.W.F. Russell, G.W. Hodgson, and G.W. Govier, “Horizontal pipeline flow of mixtures of oil and
water,” Can. J. Chem. Eng., Vol. 37, No. 1, Pp. 9–17, 1959.
[42] M.E. Charles, G.W. t Govier, and G.W. Hodgson, “The horizontal pipeline flow of equal density oil‐
water mixtures,” Can. J. Chem. Eng., Vol. 39, No. 1, Pp. 27–36, 1961.
[43] M.E. t Charles and L.V. Lilleleht, “Correlation of pressure gradients for the stratified laminar‐turbulent
pipeline flow of two immiscible liquids,” Can. J. Chem. Eng., Vol. 44, No. 1, Pp. 47–49, 1966.
[44] J. Ward and J.G. Knudsen, “Turbulent flow of unstable liquid‐liquid dispersions: Drop sizes and velocity
distributions,” AIChE J., Vol. 13, No. 2, Pp. 356–365, 1967.
[45] D. Hasson, V. Mann, and A. Nir, “Annular flow of two immiscible liquids I. Mechanisms,” Can. J.
Chem. Eng., Vol. 48, No. 5, Pp. 514–520, 1970.
[46] A.J. Karabelas, “Droplet size spectra generated in turbulent pipe flow of dilute liquid/liquid dispersions,”
AIChE J., Vol. 24, No. 2, Pp. 170–180, 1978.
[47] K.D. Oglesby, “An experimental study on the effects of oil viscosity, mixture velocity and water fraction
on horizontal oil-water flow”, University of Tulsa, Fluid Flow Projects, 1979.
[48] H. Mukherjee, J.P. Brill, and H.D. Beggs, “Experimental study of oil-water flow in inclined pipes,” J.
Energy Resour. Technol., Vol. 103, No. 1, Pp. 56–66, 1981.
[49] S. Arirachakaran, K.D. Oglesby, M.S. Malinowsky, O. Shoham, and J.P. Brill, “An analysis of oil/water
flow phenomena in horizontal pipes,” in SPE Production Operations Symposium, 1989.
[50] A. Valle, and H. Kvandal, “Pressure drop and dispersion characteristics of separated oil/water flow,” in
Proceedings of the International Symposium Two-phase flow modeling and experimentation, Vol. 2, Pp.
583–592, 1995.
[51] G. D. Zarnett and M. E. Charles, “Cocurrent gas‐liquid flow in horizontal tubes with internal spiral ribs,”
Can. J. Chem. Eng., Vol. 47, No. 3, Pp. 238–241, 1969.
[52] J. Weisman, J. Lan, and P. Disimile, “Two-phase (air-water) flow patterns and pressure drop in the
presence of helical wire ribs,” Int. J. Multiph. Flow, Vol. 20, No. 5, Pp. 885–899, 1994.
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
48
[53] M. Salcudean and S.A.L. Udean, “Effect of Flow-Obstruction Geometry,” Vol. 9, No. I, 1983.
[54] B.M. Hameed, H.A. Kadim, and A.N. Salim, “Effect of Flow-Obstruction Geometry on Pressure Drops
In Horizontal Air-Water Two-Phase Flow,” Vol. 2, No. 3, 2009.
[55] M.R. Ansari and B. Arzandi, “Two-phase gas-liquid flow regimes for smooth and ribbed rectangular
ducts,” Int. J. Multiph. Flow, Vol. 38, No. 1, Pp. 118–125, 2012.
[56] M.R. Ansari, R. Gheisari, and M. Azadi, “Flow pattern change in horizontal rectangular laterally ribbed
ducts through alteration of the ribs thickness and pitch,” Int. J. Multiph. Flow, Vol. 54, Pp. 11–21, 2013.
[57] L. Jaafer Habeeb and R. S. Al-Turaihi, “Experimental Study and CFD Simulation of Two-Phase Flow
around Multi-Shape Obstacles in Enlarging Channel,” Am. J. Mech. Eng., Vol. 1, No. 8, Pp. 470–486,
2013.
[58] S. Huang, J. Yin, Z. Sun, S. Li, and T. Zhou, “Characterization of Gas – Liquid Two-Phase Flow by
Correlation Dimension of Vortex-Induced Pressure Fluctuation,” Vol. 5, 2017.
[59] X. Q. Song, J. Z. Lin, J. F. Zhao, and T. Y. Shen, “Research on reducing erosion by adding ribs on the
wall in particulate two-phase flows,” Wear, Vol. 193, No. 1, Pp. 1–7, 1996.
[60] C. Bin, W. Cong, W. Zhiwei, and G. Liejin, “Investigation of gas-solid two-phase flow across circular
cylinders with discrete vortex method,” Appl. Therm. Eng., Vol. 29, No. 8–9, Pp. 1457–1466, 2009.
[61] H. Zhou, G. Mo, and K. Cen, “Numerical investigation of dispersed gas-solid two-phase flow around a
circular cylinder using lattice Boltzmann method,” Comput. Fluids, Vol. 52, No. 1, Pp. 130–138, 2011.
[62] F. Yan and A. Rinoshika, “Characteristics of particle velocity and concentration in a horizontal self-
excited gas-solid two-phase pipe flow of using soft fins,” Int. J. Multiph. Flow, Vol. 41, Pp. 68–76, 2012.
[63] D. Borello, F. Rispoli, A. Salvagni, and P. Venturini, “Study of the deposit of solid particles on the walls
of a ribbed duct under stationary and rotating conditions,” Pp. 8–11, 2015.
[64] M. Pathak, “Computational investigations of solid-liquid particle interaction in a two-phase flow around
a ducted obstruction,” J. Hydraul. Res., Vol. 49, No. 1, Pp. 96–104, 2011.
[65] M. Pathak and M.K. Khan, “Inter-phase slip velocity and turbulence characteristics of micro particles in
an obstructed two-phase flow,” Environ. Fluid Mech., Vol. 13, No. 4, Pp. 371–388, 2013.
[66] R.C.K. Wong, X. Pan, and B.B. Maini, “Correlation between pressure gradient and phase saturation for
oil-water flow in smooth- and rough-walled parallel-plate models,” Water Resour. Res., Vol. 44, No. 2,
Pp. 1–16, 2008.
[67] S. Zeguai, S. Chikh, and L. Tadrist, “Experimental study of two-phase flow pattern evolution in a
horizontal circular tube of small diameter in laminar flow conditions,” Int. J. Multiph. Flow, Vol. 55, Pp.
99–110, 2013.
[68] J.O. McCaslin and O. Desjardins, “Numerical investigation of gravitational effects in horizontal annular
liquid-gas flow,” Int. J. Multiph. Flow, Vol. 67, Pp. 88–105, 2014.
[69] W.R. De Oliveira, I.B. De Paula, F.J.W.A. Martins, P.S.C. Farias, and L.F.A. Azevedo, “Bubble
characterization in horizontal air-water intermittent flow,” Int. J. Multiph. Flow, Vol. 69, Pp. 18–30,
2015.
[70] B. Bae, T. Ahn, J. Jeong, K. Kim, and B. Yun, “Characteristics of an interfacial wave in a horizontal air-
water stratified flow,” Int. J. Multiph. Flow, Vol. 97, Pp. 197–205, 2017.
[71] A.S. Bageri, R. Gajbhiye, and A. Abdulraheem, “Evaluating the Effect of Surface Tension on Two-Phase
Flow in Horizontal Pipes,” SPE Kingdom Saudi Arab. Annu. Tech. Symp. Exhib., 2017.
[72] L.D.S. Silva, J.L.G. Marinho, and L.M.T.M. Oliveira, “Pressure Gradient Prediction for Different Flow
Patterns of Heavy Oil and Natural Gas in a Horizontal Pipe Using Cfd Techniques and Empirical
Correlations,” Brazilian J. Pet. Gas, Vol. 11, No. 3, Pp. 175–186, 2017.
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
49
[73] E.N. Eyo and L. Lao, “Gas-liquid flow regimes in horizontal annulus,” J. Pet. Sci. Eng., Vol. 175, Pp.
573–586, 2019.
[74] Y. Baghernejad, E. Hajidavalloo, S. Mohsen, and H. Zadeh, “International Journal of Multiphase Flow
Effect of pipe rotation on flow pattern and pressure drop of horizontal two-phase flow,” Int. J. Multiph.
Flow, Vol. 111, Pp. 101–111, 2019.
[75] J. Kussin and M. Sommerfeld, “Experimental studies on particle behaviour and turbulence modification
in horizontal channel flow with different wall roughness,” Exp. Fluids, Vol. 33, No. 1, Pp. 143–159,
2002.
[76] S. Laín, M. Sommerfeld, and B. Quintero, “Numerical simulation of secondary flow in pneumatic
conveying of solid particles in a horizontal circular pipe,” Brazilian J. Chemcial Eng., Vol. 26, No. 3, Pp.
583–594, 2009.
[77] G. Mallouppas and B. van Wachem, “Large Eddy Simulations of turbulent particle-laden channel flow,”
Int. J. Multiph. Flow, Vol. 54, Pp. 65–75, 2013.
[78] B. van Wachem, M. Zastawny, F. Zhao, and G. Mallouppas, “Modelling of gas-solid turbulent channel
flow with non-spherical particles with large Stokes numbers,” Int. J. Multiph. Flow, Vol. 68, Pp. 80–92,
2015.
[79] Y. Zhou, Z. Xue, X. Wu, and J. Fan, “Experimental investigation of the particle distribution of gas-solid
turbulent flow in the boundary layer,” Powder Technol., Vol. 307, Pp. 119–128, 2017.
[80] A.A. AbdEl-hamid, N.H. Mahmoud, M.H. Hamed, and A.A. Hussien, “Gas-solid flow through the
mixing duct and tail section of ejectors: Experimental studies,” Powder Technol., Vol. 328, Pp. 148–155,
2018.
[81] M. Toda, H. Konno, and S. Saito, “Simulation of limit-deposit velocity in horizontal liquid-solid flow,”
J. Chem. Eng. Japan, Vol. 13, No. 6, Pp. 439–444, 1980.
[82] H. Takahashi, T. Masuyama, and K. Noda, “Unstable flow of a solid-liquid mixture in a horizontal pipe,”
Int. J. Multiph. Flow, Vol. 15, No. 5, Pp. 831–841, 1989.
[83] P. Doron and D. Barnea, “Flow pattern maps for solid-liquid flow in pipes,” Int. J. Multiph. flow, Vol.
22, No. 2, Pp. 273–283, 1996.
[84] J. Ling, P.V Skudarnov, C.X. Lin, and M.A. Ebadian, “Numerical investigations of liquid – solid slurry
flows in a fully developed turbulent flow region,” Vol. 24, Pp. 389–398, 2003.
[85] Y. Tanaka, G. Oba, and Y. Hagiwara, “Experimental study on the interaction between large scale
vortices and particles in liquid-solid two-phase flow,” Int. J. Multiph. Flow, Vol. 29, No. 3, Pp. 361–373,
2003.
[86] D.R. Kaushal, K. Sato, T. Toyota, K. Funatsu, and Y. Tomita, “Effect of particle size distribution on
pressure drop and concentration profile in pipeline flow of highly concentrated slurry,” Int. J. Multiph.
Flow, Vol. 31, No. 7, Pp. 809–823, 2005.
[87] A. Goharzadeh and P. Rodgers, “Experimental characterization of solid particle transport by slug flow
using Particle Image Velocimetry,” in journal of physics: conference series, Vol. 147, No. 1, Pp. 12069,
2009.
[88] J. Capecelatro and O. Desjardins, “Eulerian-Lagrangian modeling of turbulent liquid-solid slurries in
horizontal pipes,” Int. J. Multiph. Flow, Vol. 55, Pp. 64–79, 2013.
[89] T.N. Ofei and A.Y. Ismail, “Eulerian-Eulerian simulation of particle-liquid slurry flow in horizontal
pipe,” J. Pet. Eng., Vol. 2016.
[90] M. Leporini, B. Marchetti, F. Corvaro, F. Polonara, and A. Terenzi, “Sand transport in multiphase fl ow
mixtures in a horizontal pipeline: An experimental investigation,” Petroleum, Pp. 1–10, 2018.
Two-phase flow for gas-liquid, gas-solid, liquid-solid, and liquid-liquid in a horizontal smooth and turbulator conduit – A review
50
[91] N. Brauner and D. Moalem, “N. Brauner And D. Moalem,” Vol. 18, No. I, Pp. 123–140, 1992.
[92] J.L. Trallero, S.A. Intevep, C. Sarica, J.P. Brill, and U. Tulsa, “A Study of Oil / Water Flow Patterns in
Horizontal Pipes,” no. August, Pp. 165–172, 1997.
[93] P. Angeli and G.F. Hewitt, “Flow structure in horizontal oil-water flow,” Int. J. Multiph. Flow, Vol. 26,
No. 7, Pp. 1117–1140, 2000.
[94] Y.V. Fairuzov, P. Arenas-Medina, J. Verdejo-Fierro, and R. Gonzalez-Islas, “Flow pattern transitions in
horizontal pipelines carrying oil-water mixtures: full-scale experiments,” J. energy Resour. Technol.,
Vol. 122, No. 4, Pp. 169–176, 2000.
[95] H. Shi, J. Cai, and W.P. Jepson, “Oil-water two-phase flows in large-diameter pipelines,” Trans. Soc.
Mech. Eng. J. ENERGY Resour. Technol., Vol. 123, No. 4, Pp. 270–276, 2001.
[96] O.M.H. Rodriguez and R.V.A. Oliemans, “Experimental study on oil – water flow in horizontal and
slightly inclined pipes,” Vol. 32, Pp. 323–343, 2006.
[97] M. Vielma, S. Atmaca, C. Sarica, and H. Zhang, “SPE 109591 Characterization of Oil / Water Flows in
Horizontal Pipes,” 2007.
[98] B. Grassi, D. Strazza, and P. Poesio, “Experimental validation of theoretical models in two-phase high-
viscosity ratio liquid-liquid flows in horizontal and slightly inclined pipes,” Int. J. Multiph. Flow, Vol.
34, No. 10, Pp. 950–965, 2008.
[99] T. Hohne, “Experiments and Numerical Simulations of Horizontal Two-Phase Flow Regimes,” CFD
Miner. Process Ind., no. December, Pp. 1–6, 2009.
[100] W. Wang and J. Gong, “Experiment research of phase inversion in mineral oil-water two-phase flow in
horizontal pipe,” J. Energy Resour. Technol., Vol. 131, No. 4, Pp. 43001, 2009.
[101] R. Oliemans, “Oil-water liquid flow rate determined from measured pressure drop and water hold-up in
horizontal pipes,” J. Brazilian Soc. Mech. Sci. Eng., Vol. 33, No. SPE1, Pp. 260–264, 2011.
[102] C.F. Torres, R.S. Mohan, and L.E. Gomez, “Oil – Water Flow Pattern Transition Prediction in Horizontal
Pipes,” Pp. 1–11, 2016.
[103] A.B. Desamala, V. Vijayan, A. Dasari, and A.K. Dasmahapatra, “Prediction of oil-water flow patterns,
radial distribution of volume fraction, pressure and velocity during separated flows in horizontal pipe *,”
J. Hydrodyn., Vol. 28, No. 4, Pp. 658–668, 2016.
[104] J. Shi, M. Gourma, and H. Yeung, “Author’ s Accepted Manuscript different flow regimes Reference: ”
J. Pet. Sci. Eng., 2017.
[105] D.S. Santos, P.M. Faia, F.A.P. Garcia, and M.G. Rasteiro, “Journal of Petroleum Science and
Engineering Oil / water stratified flow in a horizontal pipe: Simulated and experimental studies using
EIT,” J. Pet. Sci. Eng., Vol. 174, Pp. 1179–1193, 2019.