Investigation of periodic structures in gas-liquid flow

Alamu, Mhunir Bayonle (2010) Investigation of periodic structures in gas-liquid flow. PhD thesis, University of Nottingham.

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (39MB) | Preview

Abstract

Three different experimental campaigns had been launched to investigate the periodicity of the two-phase flow structures like the void waves in bubbly flow, liquid slugs and Taylor bubbles in slug flow, huge waves in chum flow, disturbance waves in annular flow and wisps in wispy-annular flow. The three experiments and data analysis had been considered in increasing order of complexity. Time varying data were acquired in all the three campaigns. At the simplest level of analysis statistical measures were extracted from the time series, presented, discussed and conclusions drawn.

In the first campaign, time varying void fraction has been measured using a stack of five flush mounted ring-type conductance probes fitted with a 0.005 m internal diameter pipe, 3m long. The test section was located 390 pipe diameters from the two-phase mixer. Three liquids of different physical properties with viscosity values of 1, 10 and 12 mPa s were mixed with air to establish two-phase flow. Pressure drop measurement was carried out using two pressure taps separated by a distance of 100 pipe diameters. Upstream tap was positioned 290 pipe diameters from the mixer. Structure frequency was determined from Power Spectrum Density (PSD) of auto-covariance function; structure velocity was estimated by cross-correlating two signals separated 0.002 m apart. It is shown that flow structure becomes more periodic as liquid viscosity increases. Air/more viscous liquid two-phase flow exhibits higher structure frequency than air/water two-phase flow maintained at similar conditions. On the other hand, structure velocity decreases with increase in liquid viscosity due to lower liquid phase momentum. Analyzing the time series by considering the variations in amplitude and frequency space to generate Probability Density Function (PDF) provides additional detail. The plot of PDF against void fraction for air/water and air/12 mPa s viscous liquid at same liquid and gas superficial velocities of 0.64 m/s and 0.27 m/s shows a marked shift between the void fractions as liquid viscosity increases. For air/12 mPa s two-phase flow, the PDF is characterized by a single, serrated, taller, narrower peak, with void fraction varies from 0.18 to 0.40, fingerprinting bubbly flow. In the case of air/water, the PDF displays a twin peak distribution typical of slug flow, with an average void fraction of 0.80 and a long tail that extends to zero void fraction. Next, activity in the second campaign is summarized.

In the second campaign, dynamic drop concentration, Sauter Mean Diameter (SMD) and Mass Median Diameter (MMD) were measured in annular two-phase flow for the first time employing a new generation instrumentation based laser diffraction technique. The measurement was conducted with air/water flow on a 0.019 in internal diameter vertical special test section. The gas superficial velocity was varied systematically from 13 m/s to 43 m/s at fixed liquid superficial velocities of 0.05 and 0.15 m/s. Additional tests were carried out with the gas velocity fixed at 14 m/s for liquid superficial velocities range between 0.03 and 0.20 m/s.

Conductance probes were used to log the void fraction and the film hold-up. Spraytec was used to measure the drop concentration and the characteristic drop diameters (SMD & MMD). Pressure drop was investigated and measured between two pressure taps separated 1.55 m apart in the vertical direction. All instruments were linked and synchronized to achieve simultaneous data acquisitions such that one-to-one correspondence was established between the time varying film hold-up, drop concentration and the system pressure drop. The time averaged values extracted from the dynamic signals were compared with data from previous measurements. They were found to be in good agreement. Sauter Mean Diameter (SMD) and entrained fraction display an obvious similar signature characterized by inflection points at the transition to co-current annular flow when gas superficial velocity is 21 m/s. Mass Median Diameter (MMD) on the other hand detected the transition to mist annular flow which occurs at a gas superficial velocity of 30 m/s. Additional detail is revealed when the variation in the time series of the Mass Median Diameter (MMD) is considered in amplitude and frequency space to generate Probability Density Function (PDF). The PDF of the MMD changed from multiple maxima/peaks to single maximum/peak signifying transition from heterogeneous/ multi-modal to homogeneous/mono-modal drop size distribution at a gas superficial velocity of 30 m/s. This change has been linked to transition from co-current to mist annular flow. Beyond the transition, momentum of drops produced from thinner film start decreasing as the drop size decreases as gas superficial velocity increases. However, an exact opposite trend is observed in the case of drop momentum produced from thicker film. The entrained drops within this regime exhibit higher mass density, momentum and drop concentration/ entrained fraction as a result of greater interaction with the gas core.

A relationship between pressure drop and entrained fraction is observed to follow a power law having minimum at a point where pressure drop is minimum marking transition to mist flow at gas superficial velocity of 30 m/s and entrained fraction of 0.20. Visualization studies of the high speed videos recorded during the experiment show that at a fixed gas superficial gas velocity of 14 m/s, below liquid superficial velocity of 1.2 m/s, wisps are seen in region classified as chum flow regime by existing flow pattern transition models. The observed wisps have been linked to the incomplete atomization of liquid film principally caused by dominance of gravity over drag force. Huge waves dominate with high amplitude and high standard deviation the film hold-up.

At another level of analysis, fluctuation of film hold-up and drop concentration with time is considered. Both show evidence of periodicity, film thickness being more periodic than drop concentration. Thus, frequency analysis is carried out to establish the relationship between drop and wave frequency. Surprisingly, wave frequency is seen to be higher than drop frequency most cases examined. This is due to the rate of drop coalescence and turbulent diffusion in the dispersed phase. Traditional Strouhal number-Lockhart-Martinelli parameter provides a good correlation for the wave frequency. However, correlating drop frequency using same approach proves inadequate. Second campaign, therefore, concludes that the subtle changes in Sauter Mean Diameter (SMD), entrained fraction and the Mass Median Diameter (MMD) make the evidences of flow pattern transition within annular flow more compelling.

In the third and the last campaign, flow structures were investigated in the most complex flow geometry used. The test section is made up of a 0.005 m internal diameter vertical dividing junction; located 1.3 m vertically downstream of the two-phase mixer constitutes the test section. The junction splits and mal-distributes the incoming two-phase flow into the vertical run and the horizontal side arm. The fraction of gas and liquid taken off at the horizontal side arm determines the measured flow splits.

Effect of liquid viscosity on partial separation of phases between the outlets of the junction has been investigated by testing with air/water and air/viscous liquids. The liquid phase viscosity has been increased systematically from I to 36 mPa s. Test matrix implemented varied between 0- 32 m/s and 0.003 - 1.3 m/s for gas and liquid phase superficial velocities respectively.

Plotting corresponding liquid hold-up against gas superficial velocity for each take off point elucidates chum - annular transition boundary at gas superficial velocity of 15 m/s. Liquid hold-up is seen to increase as liquid viscosity and fraction of gas taken off increases suggesting corresponding increase in partial separation of phases. However, effect of liquid viscosity does not become significant until a threshold is exceeded when fraction of gas taken off equals 0.40. In all cases examined, periodicity of flow structures is observed to increase as liquid viscosity increases.

Considering the results of the three investigations carried out, it can be concluded that periodicity of two-phase flow structure increases as liquid viscosity increases and transition to co-current annular flow occurs at gas superficial velocity of 21 m/s.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Azzopardi, B.J.
Robinson, J.
Keywords: Gas-liquid interfaces, multiphase flow, viscosity
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA 357 Fluid mechanics
Faculties/Schools: UK Campuses > Faculty of Engineering > Department of Chemical and Environmental Engineering
Item ID: 12228
Depositing User: EP, Services
Date Deposited: 18 Oct 2011 10:49
Last Modified: 16 Dec 2017 18:01
URI: https://eprints.nottingham.ac.uk/id/eprint/12228

Actions (Archive Staff Only)

Edit View Edit View