Search

• Three-dimensional symmetric phase-change lattice Boltzmann model is built up to investigate sessile droplet boiling on micro-pillar array surface in the nucleate boiling regime. • Effects of micro-pillar size on bubble evolution behaviors accompanied with droplet boiling morphologies, manifested in bubble nucleation, growth, coalescence and rupture are studied. • Quantitatively assessed include nucleation activation time, droplet total evaporation time, isolated bubble nuclei density, bubble size and substrate heat flux. • Temperature, confined space and fluid flow are discussed as key factors affecting the preferential activation location and the distribution of nucleation sites. • Understanding of droplet/bubble two-phase dynamics gives some insights to optimizing droplet nucleate boiling heat transfer performance by manipulating surface properties. Despite spray cooling in the form of droplet boiling on a textured surface being a very promising phase-change heat dissipating method, the understanding of droplet/bubble two-phase dynamics in the nucleate boiling is extraordinarily limited. In this study, we report sessile droplet boiling on micro-pillar array surface in the nucleate boiling regime using a three-dimensional lattice Boltzmann model comprehensively. Effects of micro-pillar size on bubble behaviors inside droplet are discussed in detail, covering bubble nucleation, growth, coalescence, and rupture. For the micro-pillars with large side length or small spacing, nucleation sites are activated around micro-pillar top surface. The preferential activation location of nucleation sites is determined by temperature, confined space and fluid flow. In bubble growth stage, the variation of bubble radius with time follows the square root law, being consistent with previous experiments. Bubbles merge into a large central bubble beneath droplet for the short micro-pillars while into a vapor layer for the long micro-pillars. Emergence of large central bubble prolongs droplet lifetime but deteriorates heat transfer. In addition, increasing micro-pillar side length or decreasing micro-pillar height can delay activation of nucleation sites. [ABSTRACT FROM AUTHOR]

• Effects of dense internals' layouts on bubble column hydrodynamics are revealed. • Increase in column scale increases the Sauter mean diameter. • The damped liquid turbulence level in columns with internals deteriorates the mass transfer. • Liquid mixing is enhanced by large sub-channels. • New correlations are proposed accounting for the impact of the internals. Bubble column reactors with exothermic reactions are often equipped with dense tube bundle heat exchangers. While there is some knowledge about the impact of such internals on hydrodynamics and mass transfer for narrow columns, its role in pilot-scale columns is less clear. In this paper we report on a study of hydrodynamics and mass transfer in a BCR of 4.2 m height and 0.392 m diameter. We investigated different tube arrangements with triangular and square pitch and tube diameters of 32 × 10−3 m and 45 × 10−3 m at the same cross-sectional coverage (∼25%). The column was operated at homogeneous and heterogeneous flow conditions. A customized three-layer wire-mesh sensor was utilized to visualize gas phase dynamics and liquid mixing characteristics in the column's cross-section. We found that sub-channel size is the most crucial geometric design parameter. Tracer mixing experiments revealed that internals reduce the mixing time due to the induction of large-scale liquid circulation. Mass transfer was studied with the oxygen stripping method. Here we found, that the effect of the internals on the gas-liquid mass transfer is almost negligible. Eventually, correlations for gas holdup, axial liquid dispersion and the volumetric gas-liquid mass transfer coefficient are given, which take the internals' geometry into account. [ABSTRACT FROM AUTHOR]

Several techniques have been developed in the past to measure gas and liquid phase dynamics; however, reported data were mostly gathered individually for either liquid velocity, or volume fraction (phase holdup), but never when both are measured in the same system. In this work, arguably for the first time, bubble column hydrodynamics have been investigated using two complementary advanced non-invasive measurement techniques, namely Ultrafast X-ray Computed Tomography (UXCT) and Radioactive Particle Tracking (RPT). The UXCT experimental data in terms of gas phase structure is used in a supportive way to explain the liquid velocity profiles of the RPT data. Results of both experimental techniques are verified in a complementary manner using the mass conservation calculation. The results show good agreement. It is envisioned that the presented data would be helpful in the development and validation of numerical models for better predicting the flow profiles in bubble columns. [ABSTRACT FROM AUTHOR]

Magnetic resonance (MR) imaging techniques have been used to study gas phase dynamics during co-current up-flow in a column of inner diameter 43 mm, packed with spherical non-porous elements of diameters of 1.8, 3 and 5 mm. MR measurements of gas hold-up, bubble-size distribution, and bubble-rise velocities were made as a function of flow rate and packing size. Gas and liquid flow rates were studied in the range of 20–250 cm 3 s −1 and 0–200 cm 3 min −1 , respectively. The gas hold-up within the beds was found to increase with gas flow rate, while decreasing with increasing packing size and to a lesser extent with increasing liquid flow rate. The gas hold-up can be separated into a dynamic gas hold-up, only weakly dependent on packing size and associated with bubbles rising up the bed, and a ‘static’ hold-up which refers to locations within the bed associated with temporally-invariant gas hold-up, over the measurement times of 512 s, associated either with gas trapped within the void structure of the bed or with gas channels within the bed. This ‘static’ gas hold-up is strongly dependent on packing size, showing an increase with decreasing packing size. The dynamic gas hold-up is comprised of small bubbles – of order of the packing size – which have rise velocities of 10–40 mm s −1 and which move between the packing elements within the bed, along with much larger bubbles, or agglomerates of bubbles, which move with higher rise velocities (100–300 mm s −1 ). These ‘larger’ bubbles, which may exist as streams of smaller bubbles or ‘amoeboid’ bubbles, behave as a single large bubble in terms of the observed high rise velocity. Elongation of the bubbles in the direction of flow was observed for all packings. [ABSTRACT FROM AUTHOR]