An experimental investigation of two-phase air–water flow through a horizontal circular micro-channel

Abstract

This paper is a continuation of the authors’ previous work. Two-phase air–water flow experiments are performed in a horizontal circular micro-channel. The test section is made of a fused silica tube with an inner diameter of 0.15 mm and a length of 104 mm. The flow phenomena, which are liquid/unstable annular alternating flow (LUAAF), liquid/annular alternating flow (LAAF), and annular flow, are observed and recorded by a high-speed camera mounted together with a stereozoom microscope. A flow pattern map is presented in terms of the phase superficial velocities and is compared with those of other researchers obtained from different working fluids. Image analysis is performed to determine the void fraction, which increases non-linearly with increasing volumetric quality. It is revealed that the two-phase frictional multiplier data show a dependence on flow pattern rather than mass flux. Based on the present data, a new pressure drop correlation is proposed for practical applications. According to the present study, in general the data for the two-phase air–water flow characteristics are found to comply with those of working fluids other than air–water mixtures.

Introduction

Two-phase flow and heat transfer mechanisms in mini- and micro-channels are found to become highly prevalent in recent years. The attractive phenomena appearing in the very small flow passages are necessary for the development of the modern technology applications such as medical devices, high heat-flux compact heat exchangers, cooling systems for high performance micro-electronics and so on.

Two-phase flow and heat transfer characteristics in small channels such as micro- and mini-channels are likely to be strongly dependent on surface tension effects in addition to viscosity and inertia forces, resulting in significant differences in two-phase flow phenomena between ordinarily sized channels and small channels. Several investigators have proposed criteria to address the definition of a micro-channel. The proposed channel classifications are often based on different dimensionless parameters. For instance, arbitrary channel classifications based on the hydraulic diameter D_h have been proposed. Recently, Mehendale et al. [1] employed the hydraulic diameter as an important parameter for defining heat exchangers and Kandlikar [2] proposed criteria for small flow channels used in engineering applications.

As we know, however, a clear understanding of the micro-scale effects on two-phase flow and heat transfer characteristics is still lacking. Thus, further investigations are essential for optimum design and process control of micro-systems. In the present study, the characteristics of air–water flow instead of N₂–deionised water flow, which have never been seen before, are investigated in a horizontal circular micro-channel with a diameter of 0.15 mm.

As compared with the reported two-phase flow mechanisms in ordinarily sized channels, which are available in a relatively large number of publications, hydrodynamic and transport phenomena in mini- and micro-channels tend to show different behaviours due to the effects of the limited, confined space. Recent studies associated with flow patterns, void fractions and pressure drops of two-phase flows in small channels are outlined in the following paragraphs.

It is not possible to understand two-phase flow phenomena without a clear understanding of the flow patterns encountered. It is expected that flow patterns will influence the two-phase pressure drop, holdup, system stability, exchange rate of momentum, heat, and mass during the phase-change heat transfer process. The ability to accurately predict the type of flow is necessary before relevant calculation techniques can be developed.

A flow visualisation study of two-phase air–deionised water (DI water) flow through micro-channels with hydraulic diameters ranging from 1.1 to 1.5 mm was performed by Triplett et al. [3]. The flow patterns observed were bubbly, slug, churn, slug-annular, and annular.

Chen et al. [4] conducted experiments to study the adiabatic two-phase flow of nitrogen–water in circular micro-channels with two different diameters, 1.0 and 1.5 mm. Four common flow patterns including bubbly, slug, churn, and annular flow were observed and a so-called bubble-train slug flow previously reported in a large channel under micro-gravity effects [5] was also observed and discussed in this work.

Serizawa et al. [6] investigated visualisation of the two-phase flow pattern in circular micro-channels. Flowing mixtures of air and water in channels of 20, 25, and 100 μm in diameter and steam and water in a channel of 50 μm in diameter were observed experimentally. Two-phase flow patterns obtained from both air–water and steam–water flows were quite similar and their detailed structures were described.

Chung and Kawaji [7] performed an experiment to distinguish two-phase flow characteristics in micro-channels from those in mini-channels. Four different circular diameters ranging from 50 to 526 μm were employed to examine the scaling effect on nitrogen–DI water two-phase flow.

Chung et al. [8] explored nitrogen–DI water two-phase flow through a square channel with a hydraulic diameter of 96 μm. Discussion of the effect of channel shape on transition boundaries for the observed flow patterns was presented.

A flow visualisation study to clarify the flow patterns of vertical upward gas–liquid two-phase flow in rectangular mini-channels with hydraulic diameters ranging from 1.95 to 5.58 mm was carried out by Satitchaicharoen and Wongwises [9]. The flow phenomena, which were bubbly, cap-bubbly, slug, churn, and annular, were observed and recorded by a high-speed camera. The effects of gap size, channel width, and liquid viscosity on flow pattern transitions were discussed.

Saisorn and Wongwises [10] reported the influence of the working fluid on two-phase flow pattern in a 0.53 mm diameter channel. Air, nitrogen gas, water, and deionised water were used as working fluids. The results of the two-phase air–water system were found to correspond well with those of working fluids other than air–water mixture.

Void fraction is also one of the most important parameters needed to evaluate the gravitational and accelerational terms in the total pressure drop of two-phase gas–liquid flow in various channels. Different methods available for void fraction estimation have been used by previous researchers.

Conventional methods such as a constant electric current method generally used to measure void fraction in ordinarily sized channels were carried out by Kariyasaki et al. [11] to obtain measured void fractions in channels with three different diameters, 1, 2.4, and 4.9 mm. In a method based on quick-closing valves, Wongwises and Pipathattakul [12] measured void fraction in an inclined narrow annular channel with a hydraulic diameter of 4.5 mm.

For channels with hydraulic diameters of less than around 1 mm, it is convenient to estimate void fraction by image analysis with some assumptions associated with the shapes formed by the gas–liquid interface. With this kind of method, previous researchers such as Triplett et al. [13], Serizawa et al. [6], Kawahara et al. [14], Chung et al. [8], Chung and Kawaji [7], and Saisorn and Wongwises [10] obtained reasonable results. A method other than image analysis was carried out by Chen et al. [4]. Based on the drift flux model, they measured the velocity of bubbles flowing in tubes with diameters of 1 and 1.5 mm to evaluate the void fraction. However, this method is most applicable to certain flow patterns such as bubbly and plug flow.

The important hydrodynamic aspects of two-phase pressure drop in horizontal mini- and micro-channels have been studied by different researchers. Similarly to in the ordinarily sized channel, frictional pressure drop in the micro-channel still plays the dominant role when compared with other pressure drop components such as accelerational pressure drop and pressure drop due to an abrupt flow area [14], [10].

Air–water two-phase pressure drops in the circular micro-channel with an abrupt flow area section were experimentally studied by Abdelall et al. [15]. Two stainless steel tubes were connected to form such a section. The two-phase pressure drops were found to be considerably smaller than those predicted by the homogeneous flow model, indicating significant velocity slip in the neighbourhood of the flow area change. They also developed an empirical correlation for the flow area contraction two-phase multiplier.

Chung and Kawaji [7] found that the two-phase frictional multiplier data showed dependence on mass flux for channels with diameters of 250 and 526 μm. However, there was no dependence of mass flux on the frictional multiplier for channels with diameters of 50 and 100 μm. Saisorn and Wongwises [10] conducted experiments with a 0.53 mm diameter channel and reported the dependence of mass flux as well as flow pattern on the frictional multiplier.

The applicability of several widely used viscosity models to the pressure drop prediction of air–water flow through a 0.53 mm diameter channel was examined by Saisorn and Wongwises [16]. The other relevant models for the two-phase flow pressure drop prediction in micro-channels were also examined by Kawahara et al. [14], Chung and Kawaji [7], Saisorn and Wongwises [10], and Triplett et al. [13].

Although some information is currently available on flow pattern, void fraction, and pressure drop in micro-channels, there remains room to be filled in with our data which is a continuation of the authors’ previous work. In this study, the influence of the working fluid is investigated in a smaller sized micro-channel (0.15 mm ID) to address whether nitrogen gas can still be replaced by air and water can still be used instead of DI water without significant differences. The present results obtained from an air–water system will be compared with those from other working fluids reported by previous researchers.