Geoffrey F. Hewitt

Flow structure in horizontal oil-water flow

The flow structure occurring during the cocurrent flow of oil (1.6 mPa s viscosity and 801 kg/m(3) density) and water was investigated using two 1-in. nominal bore horizontal test sections made from stainless steel and acrylic resin respectively. Two methods were used for the flow pattern identification, namely high speed video recording and determination of the local phase fractions with a high frequency impedance probe, while the continuous phase in dispersed flows was recognised with a conductivity needle probe.Measurements were made for mixture velocities varying from 0.2 to 3.9 m/s and input water volume fractions from 6% to 86%, Over this range of conditions, many different flow patterns were observed, ranging from stratified to fully mixed. Annular flow did not appear. In general, the mixed flow pattern appeared in the steel pipe at lower mixture velocities than in the acrylic pipe, where, also, oil was the continuous phase for a wider range of conditions. The visual observations were consistent with, the measurements using the high frequency impedance probe. In certain ranges of conditions the distribution of the phases differed dramatically between the stainless steel and the acrylic pipes. The average in-situ velocity ratios of the two phases in the acrylic pipe calculated from the phase distribution measurements were in general lower than unity

Pressure gradient in horizontal liquid–liquid flows

Abstract Pressure gradients were measured during the cocurrent flow of a low viscosity oil (1.6 mPa s viscosity and 801 kg/m3 density) and water in two 1-inch nominal bore horizontal test sections made from stainless steel and acrylic resin, respectively. Measurements were made for mixture velocities from 0.3 to 3.9 m/s and for water volume fractions from 0 to 100%. The main finding is the large difference between the results for the respective tube materials which cannot be explained only in terms of the difference in tube roughness. It is postulated that the different wettability characteristics of the two pipe materials are also responsible for this disparity. Furthermore, it was found that at high mixture velocities, where dispersed flow patterns prevail, there is a peak in pressure gradient during phase inversion and an apparent drag reduction effect when oil is the continuous phase.

Drop size distributions in horizontal oil-water dispersed flows

Data on drop size distributions formed during oil-water flows is presented in this paper; the data was obtained using a video recording technique which employed an endoscope. The experiments were performed with both the water and the oil (1.6 mPa s viscosity and 801 kg m(-3) density) as continuous phases, in two 1-in nominal bore horizontal test sections, made from stainless-steel and acrylic resin, respectively. Continuous phase velocities from 1.1 to 1.7 m/s and dispersed phase Volume fractions from 3.4 to 9% were used. The experimental drop size distributions were satisfactorily represented by the Rosin-Rammler distribution (Eq. (14)), with the values of the parameter delta ranging from 2.1 to 2.8. The results showed that the drop size distributions were strongly influenced by the pipe material, with the drops being smaller in the steel than in the acrylic pipe for the same flow conditions. They were also influenced by the nature and the velocity of the continuous phase. Both the maximum and the Sauter mean diameters were found to depend on the (-1.8) power of the continuous phase velocity. None of the theoretical correlations for the maximum drop size could represent accurately the experimental data, while the often used Hinze (1955) equation underpredicted the experimental results in all cases

Prediction of the slug-to-churn flow transition in vertical two-phase flow

Abstract An assessment is made of the various viewpoints on the slug-to-churn flow transition in vertical upward flow in the light of recent experimental results obtained at Harwell Laboratory. It is found that the flooding model of McQuillan & Whalley and the bubble entrainment model of Barnea & Brauner give satisfactory results at low and high liquid flow rates, respectively. An improved model for flooding, which takes account of the effect of the falling film, has been proposed. It is shown that this new model is in good agreement with experimental results at both low and high liquid flow rates.

Flooding and churn flow in vertical pipes

Abstract Experimental studies are reported on the flooding phenomenon and on the closely associated churn flow regime. Flooding experiments were carried out with air-water flow in a 32 mm dia vertical pipe with various forms of liquid outlet, namely a porous wall, a tapered outlet and a square-edged outlet. For the first time, downflow (penetration) measurements were made with the porous wall outlet and showed significant differences between penetration rates beyond flooding and the flooding rate. This contrasts with other types of injectors, where the mechanisms of flooding are somewhat different. Measurement of pressure drop and holdup in the churn flow regime were made both with and without a co-existing falling film below the liquid injector. These showed that the falling-film and churn flow regions were essentially decoupled. Analysis of the data for churn flow showed that the minimum pressure gradient does not, for this data, correspond to the condition of zero wall shear stress as had been suggested by some earlier analytical studies. Interfacial shear stresses in churn flow were compared with those used in current reactor safety codes (TRAC and RELAP) and it was found that, for the churn flow region, the relationships used in the RELAP code were more appropriate.

Flow development in vertical annular flow

Abstract Air–water upward annular flow experiments were conducted in a 10.8 m long tube of 31.8 mm internal diameter. Local values of the pressure gradient, film thickness, wall shear stress, film flow rate, disturbance wave velocity and frequency were measured over a range of gas and liquid flow rates. The results show that rapid changes in the film flow parameters occur within the first 50 tube diameters or so from the inlet. Entrained fraction and the film thickness appear to be the slowest to respond and these may take 100 to 300 tube diameters to develop fully.

Hydrodynamics and heat transfer of wavy thin film flow

Abstract The hydrodynamics and heat transfer of thin film flow with a wavy interface has been studied using computational fluid dynamics techniques. The velocity and temperature fields are obtained for periodic laminar flow with an assumed interface shape. The effect of sinusoidal and solitary waves on the heat transfer across the film is investigated. It is shown that the overall heat transfer coefficient is determined mainly by conduction through the film, rather than by the recirculation, if any, under the waves. However, the presence of interfacial waves still enhances the heat transfer coefficient due, mainly, to the effective thinning of the film. Copyright

INVERSE HEAT CONDUCTION PROBLEM OF ESTIMATING TIME-VARYING HEAT TRANSFER COEFFICIENT

An inverse heat conduction problem is solved by using Alifanovs iterative regularization method to estimate the time-varying heat transfer coefficient of forced-convective flow boiling over the outer surface of a heater tube. The estimation is based on transient temperature measurements taken by a thermocouple on the inner surface of the circular tube on which the flow boiling occurs. The inverse problem is solved as an optimization problem in which a squared residue functional is minimized with the conjugate gradient method. A sensitivity problem is solved to determine the step size in the direction of descent, and an adjoint problem is solved to determine the gradient of the functional. No prior information is used in the functional form of the heat transfer coefficient variation with time. Numerical experiments are performed to test the proposed mathematical model of the measurement of flow boiling heat transfer coefficient. Effects of the time scale of the heat transfer coefficient variation, the measurement errors, and the data acquisition rates on the inverse solution are investigated. We found that for a given experimental configuration there is a physical limitation on the time scale of the heat transfer coefficient variation that can be estimated by the inverse analysis.

Flooding in vertical two-phase flow

Abstract The phenomenon of flooding limits the stability of a liquid film falling along the walls of a tube in which a gas is flowing upward. This paper describes a series of experiments to observe and photograph the flooding process. Flooding in a vertical perspex tube was observed by viewing axially along the tube from the top, and flooding on the walls of a stainless steel rod supported centrally within a perspex tube was observed by viewing normally through the tube walls. The resulting observations show that flooding occurs as a result of the formation and motion of a large disturbance wave on the surface of the liquid film. A simple theoretical treatment of flooding is presented, based on the observations of the flooding disturbance. The resulting equations enable the calculation of the gas velocity and the liquid wave height at flooding, and the predictions are in good agreement with experimental observations reported here and elsewhere.

New insights into the relationship between internal phase level of emulsion templates and gas–liquid permeability of interconnected macroporous polymers

Interconnected macroporous polymers can be made by polymerisation of emulsion templates consisting of an aqueous phase and a monomer phase (typically styrene and divinylbenzene) in which the aqueous (internal) phase is in the form of drops and the monomer phase is the continuous phase between the drops. Until recently it was thought that interconnected macroporous polymers could only be produced from the polymerisation of high internal phase emulsion (HIPE) templates with an internal phase level exceeding 74 vol%. Improvement of the poor mechanical performance, characteristic of such macroporous polymers, was achieved simply by increasing the material density of the macroporous polymer. However, this required a reduction in the internal phase volume of the emulsion template. Polymerisation of the continuous organic phase of emulsion templates with an internal phase volume ranging from 84 vol% to 70 vol% resulted in the production of poly(merised)HIPEs, polymerisation of medium internal phase emulsions with internal phase volume ranging from 70 vol% to 30 vol% in polyMIPEs and polymerisation of a low internal phase emulsion with an internal phase volume of 25 vol% in a polyLIPE. The resulting macroporous polymers were characterised in terms of mechanical and structural properties as well as gas and mercury permeability. Compression tests show that mechanical properties improved as the material density was increased. Gas and mercury permeability measurements show that as the internal phase volume of the emulsion template is reduced, the permeability of the resultant macroporous polymer is also reduced. However, surprisingly even macroporous polymers produced from low internal phase emulsion templates (25 vol%) were permeable with a gas permeability of 2.6 × 10−14m2 indicating that polyLIPEs are still interconnected macroporous polymers. Reconstruction modelling of the transport properties of porous materials shows that the permeability of a porous material with similar structures to that of the macroporous polymers increases exponentially with the porosity.