Dispersed Slug Research Articles

Review and extension of pressure drop models applied to Taylor flow regimes

This paper investigates the pressure drop induced by both liquid–liquid and liquid–gas segmented Taylor flow regimes. A comprehensive experimental programme was completed using four different liquid–liquid and three different liquid–gas combinations over dimensionless slug length, Reynolds and Capillary numbers that spanned several orders of magnitude. Comparisons between the liquid–liquid pressure drop data and the most referenced expressions in the literature highlighted their lack of robustness and demonstrated their inapplicability for use with most practical systems that incorporate liquid–liquid Taylor flow regimes. The experimental pressure drop values obtained for the liquid–gas flows agreed well with existing pressure drop correlations. Interpretation of the liquid–liquid data using liquid–gas models unearthed the existence of a threshold viscosity ratio. Above this threshold, experimental liquid–liquid data was found to agree well with existing liquid–gas models. Below this threshold, results showed that the dispersed slug velocity needed to be considered as the flows were subject to higher interfacial contributions and inertial effects. A modification is proposed to an existing liquid–gas pressure drop correlation. This proposed modification extends the applicability of the correlation to liquid–liquid flows, and furthermore extends the non-dimensional limits of the correlation.

Local Nusselt number enhancements in liquid–liquid Taylor flows

Abstract Thermal management has emerged as a critical requirement to ensure the performance and reliability of electronic devices and systems. System level heat fluxes are approaching the limits of conventional forced air cooling, and there is a need to develop alternative cooling techniques for devices such as processors, power amplifiers and laser arrays. This paper examines the potential heat transfer enhancements of a two phase liquid–liquid Taylor flow regime. The primary focus of the work was to examine the influence of slug length and carrier phase variations on the local Nusselt numbers. An experimental facility was designed and commissioned to subject the flow to a constant heat flux boundary condition, a boundary condition commonly encountered in thermal management applications. Local temperature measurements were acquired using a high resolution infrared thermography system. Experiments were carried out over slug length, Capillary and Prandtl numbers that spanned several orders of magnitude in a minichannel geometry. Reductions in carrier slug length and increases in dispersed slug length were found to augment the heat transfer rates, with the greatest enhancements observed in flows with carrier slug lengths approaching the channel diameter. The thickness of the liquid film separating the dispersed slugs from the heated capillary walls was found to play a significant role in the removal of heat, with increases in film thickness resulting in a reduction in the heat transfer rates. Based on the characteristics identified, a novel correlation is proposed to model the flow in the thermally developing and fully-developed regions.

Size-Exclusion Chromatography in the Measurement of Concentration and Molecular Weight of Some EOR Polymers

Summary Procedures that involve the use of size exclusion chromatography (SEC) for the measurement of concentration and weight-averaged molecular weight, M¯w, of some EOR polymers were developed and found to give improved detectability, accuracy, and/or efficiency. The separation of polymer from low-molecular-weight impurities by size allows unambiguous detection of polymer properties such as concentration and M¯w. A combination of an SEC column of a pore size small enough to exclude the polymer totally and a mobile phase of ionic strength of 1.5 was found suitable for the separation of Polyacrylamide, partially hydrolyzed Polyacrylamide, cationic Polyacrylamide derivative, and xanthan polysaccharide from impurities. Concentration detection of the separated polymer sample with a variable-wavelength ultraviolet (UV) detector was found to give superior detectability over detection by refractive index difference. A wavelength of 214 nm [2,140 Å] was used for the detection of these polymers on the basis of the spectra of samples purified by dialysis. With the active polymer assay determined by reprecipitation into a nonsolvent, the detection limit by UV was determined to be <0.1 μg/cm3 for Polyacrylamide and a cationic Polyacrylamide derivative, <0.2 μg/cm3 for partially hydrolyzed Polyacrylamide, and <0.7 μg/cm3 for a xanthan polysaccharide. The linear calibration range was up to 500 μg/cm3. The precision of the concentration measurement was better than 4% for Polyacrylamide and its derivative and 5% for polysaccharide at a 95% confidence level. On-line measurement of M¯w of four different types of EOR polymers was made by connecting a low-angle laser-light-scattering (LALLS) detector to the SEC system and interfacing the detectors to a computer. The polymer sample was dispersed in the SEC column to give a concentration profile while being separated from the impurities. Continuous measurement of concentration and Rayleigh ratio of the dispersed slug by UV and LALLS detectors allows the calculation of the M¯w for the whole sample. The precision of this measurement is expected to be better than ±5% at a 95% confidence level. The M¯w determined for a Polyacrylamide molecular-weight standard was in good agreement with the reported value (0.98 million compared to 1 million). Such a procedure substantially minimizes the interference by dust and requires only a single injection, thereby improving both the accuracy and efficiency of the M¯w measurement.

Dispersion of gases in laminar flow through a circular tube

The longitudinal dispersion of a finite slug of gas has been measured at various velocities by using a gas (1,3-butadiene) that absorbs in the ultra-violet region and passing the dispersed slug through a narrow beam of ultra-violet light of wavelength 250mμ. To avoid effects of differences in gas density, the second gas (1-butyne) was chosen to have the same molecular weight as butadiene. The range of applicability of Sir Geoffrey Taylor’s virtual coefficient of diffusivity has been discussed. The experimental observation that the peak mean concentration passes through a maximum value with velocity has been explained by considering the relative rates of dispersion by convection, longitudinal molecular diffusion and radial molecular diffusion.