The existence and behaviour of large diameter Taylor bubbles

Abstract

Large tube filling bubbles rising up through quiescent fluid in a vertical tube are commonly known as Taylor bubbles. Their apparent simplicity of form and behaviour has led to them being viewed and modelled as a paradigm for both large bubble dynamics, where there is no continuous gas flow, and slug flow for the case of continuous gas flow. Central to this approach is the question: what diameter tubes support stable Taylor bubbles? In this paper we examine the case of low viscosity Taylor bubbles through experiments and theory and show that they exist in much wider diameter tubes than had previously been reported. In order for the bubbles to be stable a settling period is required to allow the column to sufficiently quiesce. This settling period is compared favourably with the classical stability analysis of Batchelor (1987). We also observe such bubbles rising in an oscillatory manner if the gas input is abruptly curtailed. The oscillations match theoretical predictions well.

Numerical modeling of oscillating Taylor bubbles

“…This oscillatory behavior, similar to that observed in the experiments reported by Pringle et al (2015), has been inferred to occur within volcanoes during Strombolian-type eruptions. Vergniolle, Brandeis, and Mareschal (1996) recorded acoustic pressure in the air during a period of explosive activity at the Stromboli volcano and observed oscillations in these readings during the rise of the bubble before eruption.…”

“…In this paper, commercial computational fluid dynamics (CFD) software is used to simulate the flow observed in the experiments detailed in Pringle, Ambrose, CONTACT D. M. Hargreaves david.hargreaves@nottingham.ac.uk Azzopardi, and Rust (2015), with the ultimate goal of producing a validated model suitable for running simulations under conditions that are difficult to create in a laboratory (e.g., larger-scale, higher-dynamic viscosity). The experiments of Pringle et al (2015) were conducted in a vertical Perspex pipe 10 m in length and with an internal diameter of 0.29 m. The pipe was partially filled with water and open to the atmosphere at the top.…”

“…The experiments of Pringle et al (2015) were conducted in a vertical Perspex pipe 10 m in length and with an internal diameter of 0.29 m. The pipe was partially filled with water and open to the atmosphere at the top. Compressed air was injected into the base of the pipe via a system of nozzles, with the ensuing small bubbles coalescing into a Taylor bubble.…”

“…These dimensionless numbers are, for completeness, where w is the vertical component of the rise velocity, g is the acceleration due to gravity, D is the diameter of the pipe, ρ is the density of the liquid, μ is the dynamic viscosity of the liquid and σ is the surface tension. The rise velocity of the bubble is assumed to be in the 'Independent of viscosity and surface tension' regime proposed by White and Beardmore (1962, Figure 6, p. 357) and the bubble rise rates observed by Pringle et al (2015) are consistent with theoretical predictions and empirical models for this regime (Dumitrescu, 1943;Viana, Pardo, Yanez, Trallero, and Joseph, 2003). Pringle et al (2015) demonstrated that if air is injected into quiescent water then Taylor bubbles can exist in pipes that are 0.29 m in diameter, which is significantly larger than previously suggested for the air-water system.…”

“…The rise velocity of the bubble is assumed to be in the 'Independent of viscosity and surface tension' regime proposed by White and Beardmore (1962, Figure 6, p. 357) and the bubble rise rates observed by Pringle et al (2015) are consistent with theoretical predictions and empirical models for this regime (Dumitrescu, 1943;Viana, Pardo, Yanez, Trallero, and Joseph, 2003). Pringle et al (2015) demonstrated that if air is injected into quiescent water then Taylor bubbles can exist in pipes that are 0.29 m in diameter, which is significantly larger than previously suggested for the air-water system. They further went on to show that if the gas injection ends abruptly then the bubble length and apparent rise rate oscillate.…”