Hostname: page-component-74656f9884-2dmjx Total loading time: 0 Render date: 2026-11-08T07:48:09.988Z Has data issue: false hasContentIssue false

Eccentricity effect on horizontal capillary emptying

Published online by Cambridge University Press:  29 July 2022

Dongwen Tan
Affiliation:
School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, PR China
Xinping Zhou*
Affiliation:
School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, PR China State Key Laboratory of Digital Manufacturing Equipment and Technology, Huazhong University of Science and Technology, Wuhan 430074, PR China
Gang Zhang
Affiliation:
School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, PR China Department of Mechanics, Huazhong University of Science and Technology, Wuhan 430074, PR China
Chengwei Zhu
Affiliation:
School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, PR China Department of Mechanics, Huazhong University of Science and Technology, Wuhan 430074, PR China
Chenyu Fu
Affiliation:
School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, PR China
*
†Email address for correspondence: xpzhou08@hust.edu.cn

Abstract

This paper theoretically studies the effect of eccentricity on the conditions of capillary emptying (determined by critical Bond number) in a horizontal annular tube in a downward gravity field. Experiments are conducted to compare with theoretical results. We find that non-horizontal eccentricity can lead to the occurrence of a re-entrant liquid-state transition (from liquid non-occlusion to liquid plug to liquid non-occlusion) with increasing Bond number, when the eccentricity (e) or inner-to-outer radius ratio (χ) is large enough, and the two liquid non-occlusion states correspond to different emptying mechanisms dominated by the gravity effect and the ‘wedge’ effect, respectively. Existence of the re-entrant transition is accompanied by occurrence of unconditional liquid non-occlusion at large enough or small enough contact angles regardless of Bond numbers. The critical Bond numbers at a contact angle γ for vertical upward eccentricity are equal to those at a contact angle 180° − γ for vertical downward eccentricity. In a parameter space (γ, e/(1 − χ)), the region with the re-entrant transition becomes larger with the eccentric angle varying from 0° (horizontal) to 90° (vertical). Optimization of geometrical parameters and inner and outer contact angles can lead to better effect of capillary emptying. This paper provides a very effective scheme for removing a liquid blockage from a capillary in optofluidics/microfluidics.

Information

Type
JFM Papers
Copyright
© The Author(s), 2022. Published by Cambridge University Press
Figure 0

Figure 1. Schematic of general cross-section of liquid partially filling a horizontal circular tube of radius Ro with a rod of radius Ri inserted into it at an eccentric position in a downward gravity field. The displacement d is defined as the distance of the rod centre O′ away from the outer tube centre O. The eccentric angle θ is measured counterclockwise, starting from the positive x axis. The gas–liquid interface $\varGamma$ meets the inner and outer walls at contact points with inner and outer contact angles γi and γo. The annular gap has total area $\varOmega$ and total inner and outer perimeters ${\varSigma _i}$ and ${\varSigma_o}$, while the liquid has area ${\varOmega^\ast }$ and wetting inner and outer perimeters $\varSigma_i^\ast $ and $\varSigma_o^\ast $.

Figure 1

Table 1. Computational procedure to obtain the critical Bond number(s) for a given capillary tube geometry and specified contact angles when the critical Bond number(s) exists.

Figure 2

Figure 2. Critical Bond numbers of annular tubes with different vertical upward eccentricities for different radius ratios (a) χ = 0.1, (b) χ = 0.4 and (c) χ = 0.7 versus contact angle γo = γi = γ. The squares denote the representative points each corresponding to a cross-section of the part of the interface that extends to infinite length at the critical Bond number. In (b), the circles on the black thick line denote three key points at e = 0.4. Points I, J and K correspond to the intersection point of the lower and upper critical emptying lines (i.e. the lines IJ and IK), Bocl = 0 and Bocu = 0, respectively. In the right-hand part of (b), 3-D interfaces (an oblique view) computed by Surface Evolver (Brakke 1992) at γ = 30° for e = 0.4 for different Bond numbers correspond to the triangles denoting the points P1, P2, P3, P4 and P5, respectively. For points P1 and P2, the upper tongue is longer than the lower tongue (the phenomenon for point P2 does not seem apparent due to the oblique view, and accordingly a grey circle through the tip of the lower tongue denoting a cross-section of the outer tube is plotted to help understand the phenomenon for point P2), while for points P3, P4 and P5, the lower tongue is longer than the upper tongue.

Figure 3

Figure 3. Minimum energy ${\varPhi_{min}}$ (calculated at the fourth step of table 1) for an annular tube of vertical upward eccentricity (χ = 0.4 and e = 0.4) for (a) the cases γo = γi = γ < 90° and (b) the cases γo = γi = γ > 90° versus Bond number. Points C–E and I–K correspond to C–E and I–K, respectively, as presented in figure 2(b). The dot-dashed line denotes the straight line for ${\varPhi_{min}} = 0$. In (a), each dashed line (a discontinuous segment of the line of ${\varPhi_{min}}$) represents a jump due to the transition between different non-occluded liquid topologies corresponding to minimal total free energy.

Figure 4

Figure 4. Schematic of evolution of the 3-D interface (an oblique view) (a) as the Bond number gradually decreases (from left to right) and approaches the lower critical Bond number and (b) as the Bond number gradually increases (from left to right) and approaches the upper critical Bond number. Some representative 3-D interfaces computed using Surface Evolver (Brakke 1992) are integrated in a horizontal eccentric annular tube.

Figure 5

Figure 5. Critical Bond numbers of annular tubes with vertical upward (θ = 90°) and downward (θ = −90°) eccentricities e = 0.4 for representative radius ratio χ = 0.4 versus contact angle γo = γi = γ.

Figure 6

Figure 6. Critical Bond numbers of annular tubes with different horizontal eccentricities for representative radius ratio χ = 0.4 versus contact angle γo = γi = γ. An enlarged view of the line segments around the left kink of the curve for e = 0.5 is shown. Contact angle refinement for e = 0.4 is conducted in order to avoid the appearance of false kinks.

Figure 7

Figure 7. Critical Bond numbers of annular tubes with (a) different radius ratios and different eccentricities for eccentric angle θ = 45° and with (b) radius ratio χ = 0.4 and eccentricity e = 0.4 for different eccentric angles versus contact angle γo = γi = γ.

Figure 8

Figure 8. (a) Phase diagram in a space (γ, e/(1 − χ)) for different eccentric angles for χ = 0.4 and (b) boundary lines in the parameter space for vertical eccentricity θ = 90° and different radius ratios χ = 0.1, 0.4 and 0.7. Equal inner and outer contact angles are used: γo = γi = γ. In (a), boundaries among four states are presented for different eccentric angles. With the eccentric angle varies, the boundary line (dashed line) between the left unconditional liquid non-occlusion region and the two-Boc region changes, while the two boundary lines (solid lines) corresponding to zero critical Bond number do not change. Points I–K correspond to I–K, respectively, as presented in figure 2(b). The blue text annotation corresponds to the region between the blue dashed line and the solid line through point J, representing the two-Boc region in the parameter space for the case of vertical eccentricity θ = 90°. In (b), the dashed line corresponds to the case denoted by the solid line in the same colour.

Figure 9

Figure 9. (a) Boundary lines in parameter space (γo, e/(1 − χ)) at χ = 0.4 and θ = 90° for different inner contact angles versus outer contact angle γo and (b) outer contact angle ranges of approximate and unconditional liquid non-occlusions Δγo for different inner contact angles versus e/(1 − χ). In (a), the dashed line corresponds to the case denoted by the solid line in the same colour.

Figure 10

Figure 10. (a) Theoretical predictions for lower and upper ‘critical emptying lines’ (Boc(γ), where γo = γi = γ) for a vertically eccentric annular capillary (χ = 0.5 and e = 0.27) and experimental results. The circles (e.g. Q1 and Q3) indicate liquid non-occlusion and the crosses (e.g. Q2) indicate that the liquid plug remains. For the three experimental cases Q1, Q2 and Q3, different diameters of outer tubes at 5.95, 7.96 and 12.11 mm and the corresponding different sizes of rods are used, respectively, thus leading to different Bond numbers. Note that, as predicted by theory, a small tube (see Q1) and a large tube (see Q3) become non-occluding after rod insertion whereas a medium-sized tube remains blocked (see Q2). (b) Photographs of the liquid occlusion or non-occlusion for the three experimental cases Q1, Q2 and Q3 representing the liquid states related to the re-entrant transition with increasing Bo. Note that the contact angles at the lower section of the tube in the picture of Q2 seem much larger than 30°. This is attributed to the 3-D liquid surface (meniscus) concave towards the gas and the position of the camera relative to the horizontal capillary.

Figure 11

Figure 11. Schematic of cross-section of a horizontal tube $\varOmega$ in an arbitrary geometrical shape and the cross-section of another horizontal tube ${\varOmega^ + }$ whose perimeter is symmetric to the perimeter of the cross-section $\varOmega$ with respect to the horizontal axis.

Figure 12

Figure 12. Non-occluded liquid configurations with the interface meeting the outer and inner walls at zero gravity for a set of inner and outer contact angles γi and γo. The liquid may occupy (a) the narrow part or (b) the wide part. In (a), α is the angular position of the outer contact point on the outer circle, β is the angular position of the inner contact point on the inner circle and ${\varphi _{o}}$ and ${\varphi _i}$ are the angular positions of the outer and inner contact points on the interface arc, respectively. These angular positions are measured counterclockwise, starting from the positive x axis.