betaFeedback

WorksPhysics and Astronomy

Effective water/water contact angle at the base of an impinging jet

Abstract

The base of a jet impinging on an ultrapure water bath is studied experimentally. At the impact point, a train of capillary waves develops along the jet. By performing Particle Tracking Velocity measurements, we show that there is a boundary layer separation between the jet and the meniscus. We thus describe the shape of this meniscus with a hydrostatic model. A striking observation is the existence of an effective non-zero water/water contact angle between the jet and the meniscus. The rationalization of this finite contact angle requires a full description of the shape of the interface. By doing an analytical matching between the meniscus and the jet, we show that the capillary waves can be considered as reflected waves present to ensure pressure continuity. It is finally shown that the value of the apparent contact angle is fixed by energy minimization, with an excellent agreement between prediction and experiment for small jets.

Liquid-air interfaces are geometrical objects with an intimate link between their shape and their properties. They give rise to purely geometrical questions such as surface minimization problems [1, 2]. A physical illustration is foams, whose mechanical stability is linked to the local bubble packing [3] and whose coarsening is fixed by topology [4, 5]. Interfacial geometry can also play a critical role in dynamical situations. For example, the shape of the cusp at the rear of a sliding drop or an entrained film fixes the wetting transition [6, 7]. More complex interfacial topologies exist in multiphase flows, such as breaking waves [8] or aeration through weirs or cascades [9]. A paradigm for these flows is the plunging jet, which can entrain air when it impacts a bath [8]. Its geometrical features, such as the shape of the cusp in the viscous case [10, 11], or surface disturbances when inviscid [12], control the whole behavior of the system.

In this Letter, we focus on the base of water jets impacting a quiescent water bath. We report on an intriguing observation: the apparition of an apparent non-zero contact angle at the water/water junction, see Figure 1(c). We rationalize the presence of this contact angle and provide a description of the whole shape of this interface.

Experimental images and comparison of theoretical and experimental values

Figure 1. (a) Experimental image of a water jet of radius R=100R = 100 μm\mu\mathrm{m} and speed V=2.1V = 2.1 m.s−1\mathrm{m.s}^{-1} impacting a bath of water. (b) Zoom on the capillary waves developing at the jet surface fitted by a function of the form r(z)=R+acos⁡(kz+φ)exp⁡(−zzatt)r(z) = R + a \cos(kz + \varphi) \exp\left(-\frac{z}{z_{\mathrm{att}}}\right). (c) Zoom on the connection between the meniscus and the jet, where the non-zero contact angle appears. (d) Comparison between the values of zattz_{\mathrm{att}} and λ\lambda extracted from the experimental fit and the ones calculated with Eq. 6 in ref.[13] for a jet of radius R=160R = 160 μm\mu\mathrm{m}.

When a water jet plunges in a water bath, a train of stationary capillary waves will be created at its base, and decay upstream (Figure 1(a) and (Figure 1b)) [14]. Their wavelength can be obtained by equating the phase velocity of the waves with the speed of the jet [15]. The damping coefficient has been modeled and linked to the properties of the fluid [13, 16]. However, less attention has been given to the meniscus at the connection between the jet and the bath, where it is clear that there exists an apparent contact angle. This is surprising as the only phases at play are water and air, hence no contact angle is expected.

To study this contact angle, we produce jets of ultrapure water (resistivity larger than 18.2 MΩ.cm\mathrm{M\Omega.cm}) with radii RR ranging from 50 to 400 μm\mu\mathrm{m}, using a syringe pump

(Harvard Apparatus PHD ULTRA) connected to a nozzle when R<100R < 100 μm\mu\mathrm{m} and a pressurized reservoir otherwise. As shown in [17], a small amount of surfactants in the recipient bath modifies significantly the wave pattern, so great care was taken to avoid contamination of the ultrapure water in our experiments. The pressure in the reservoir is imposed by a pressure controller (Elve-flow OB1 Mk4) connected to a network of compressed air. These setups can create laminar jets with speeds between 1.9 and 8 m.s−1\mathrm{m.s}^{-1} for R<100R < 100 μm\mu\mathrm{m} and between 1.4 and 5 m.s−1^{-1} for bigger ones. The base of the jet is imaged using telecentric lenses (SilverTl 2x or 4x Edmund Optics) to obtain sharp interfaces and a digital camera (Basler a2A1920-160umPRO). The images were typically taken with an exposure time of 700 μ\mus.

Measuring the observed contact angle is not straightforward, as it is not possible to place the contact line between the jet and the meniscus in a non-arbitrary manner. To overcome this problem, we propose to measure the contact angle by fitting the shape of the interface. Despite the jet having a speed of a few meters per second, the meniscus at the base of the jet looks very similar to the one climbing a static fiber plunged in a liquid bath (visual comparison in Figure 2(a) and (b)), as was previously noticed by Patrascu et al. [13]. We performed PTV measurements below the interface, using small silica particles (of radius between 30 and 80 μ\mum), and a fast camera (Photron FASTCAM NOVA S9) recording 6000 images per second. These images were then analysed with Python. The results can be seen in Figure 2(c) and (d) for two different jet radii and velocities. The measurements confirm that there is a boundary layer separation between the dynamic jet and the static meniscus, with a typical velocity ratio of 20 between the velocity in the dynamic jet and in the still zone of the meniscus. This justifies the use of hydrostatic equations to describe the shape of the meniscus.

Fit of the meniscus and PTV measurements

Figure 2. (a) Fit of the meniscus, on the right and on the left, by equation (1) for a jet of radius R=280 μR = 280\ \mum and speed V=2.65 m.s−1V = 2.65\ {\rm m.s}^{-1} and for (b) a glass fiber of R=250 μR = 250\ \mum. (c) Results of the PTV experiments under the water surface for a jet of radius R=280 μR = 280\ \mum and speed V=2 m.s−1V = 2\ {\rm m.s}^{-1} and a jet of radius R=80 μR = 80\ \mum and speed V=2.8 m.s−1V = 2.8\ {\rm m.s}^{-1}.

The shape of the static meniscus is fixed by a balance between the hydrostatic pressure ρgz\rho g z, with ρ\rho the liquid density, gg the acceleration of gravity and zz the surface elevation and the Laplace pressure γC\gamma C, with γ\gamma the liquid/air surface tension and CC the curvature of the interface. The typical length scale of the meniscus is thus the capillary length ℓc=γρg\ell_c = \sqrt{\frac{\gamma}{\rho g}}. An equation for z(r)z(r), with rr the coordinate in the radial direction, in the non-linear part of the meniscus attached to a fiber of radius r0r_0 with a non-zero contact angle has been proposed by James [18]. The linear small slope solution is z(r)=AK0(rℓc)z(r) = A K_0\left(\frac{r}{\ell_c}\right), where AA is a constant and K0K_0 a modified Bessel function of the second kind of order zero. By matching it with the equation of a catenoid satisfying the boundary condition z'(r_0/r_0) = -\cotan(\theta) where θ\theta is the contact angle, he obtains, in the vicinity of the fiber

z(r)=r0cos⁡(θ)[−ln⁡(ϵ)+ln⁡(4)−Γ−ln⁡(rr0+((rr0)2−cos⁡2(θ))12)],(1)z(r) = r_0 \cos(\theta) \left[-\ln(\epsilon) + \ln(4) - \Gamma- \ln\left(\frac{r}{r_0} + \left(\left(\frac{r}{r_0}\right)^2 - \cos^2(\theta)\right)^{\frac{1}{2}}\right)\right], \tag*{(1)}

where ϵ=ρgR2γ\epsilon= \sqrt{\frac{\rho g R^2}{\gamma}} is the square root of the Bond number and Γ\Gamma is Euler’s constant. We use James’ equation to fit the meniscii in Figure 2(a) and (b). The contact angle is used as a free parameter and can be different on the right and on the left. This leads to slightly different values of the contact angle on the jet, which will be taken into account while evaluating the error bars. On the fiber, the difference is larger, probably due to impurities at its surface. The agreement between the fit and the experimental shape confirms that this region can be considered as static, and that there exists a non-zero contact angle between the meniscus and the jet. Replacing directly the fiber radius r0r_0 in James’ equation with the jet radius RR proves unsatisfactory, for the actual radius is altered with the capillary wavefield. We thus need to write proper matching conditions between the capillary wavefield and the static meniscus.

In the following, we propose a matching between James’ description of the meniscus and the jet decorated with capillary waves. The parameters needed to describe the meniscus using James’ equation are r0r_0 and θ\theta. The capillary wavefield is approximated by an attenuated sinusoidal wave r(z)=R+acos⁡(kz+φ)exp⁡(−zzatt)r(z) = R + a \cos(kz + \varphi)\exp\left(-\frac{z}{z_{\mathrm{att}}}\right), with k=2πλk = \frac{2\pi}{\lambda} the wavenumber, λ\lambda the wavelength, aa the amplitude, φ\varphi a phase shift, and zattz_{\mathrm{att}}, the attenuation length. kk and zattz_{\mathrm{att}} are predicted by the model of Patrascu (equation (6) in reference [13]). It agrees very well with the fitted values of zattz_{\mathrm{att}} and λ\lambda, as can be seen in Figure 1d. We then have five unknowns θ\theta, r0r_0, z0z_0, φ\varphi, and aa. The first matching condition is the radial matching position, which gives r0=r(z0)r_0 = r(z_0). The height matching position z0z_0 is given by James’ equation z0=z(r0)z_0 = z(r_0). The slope matching yields ∂r∂z=−tan⁡(θ)\frac{\partial r}{\partial z} = -\tan(\theta). The last condition is the matching between the jet curvature

κjet=1+r′(z)2−r(z)r′′(z)r(z)[1+r′(z)2]3/2\kappa_{\mathrm{jet}} = \frac{1+r'(z)^2-r(z)r''(z)}{r(z)[1+r'(z)^2]^{3/2}}

and the meniscus curvature κm=0\kappa_{\mathrm{m}}=0, as the meniscus is a catenoid. Practically we used a linearized version of κjet\kappa_{\mathrm{jet}} as full nonlinear computations did not show significant differences.

This curvature matching is in fact a pressure matching between the jet, which has an overpressure ΔP=γR\Delta P=\frac{\gamma}{R}, and the meniscus, which has an underpressure ΔP=−ρgz\Delta P=-\rho gz. Capillary wave emission stems from this pressure mismatch. The reflected waves lower the curvature of the impinged jet base to accommodate with the meniscus’ pressure, as sketched in Figure 3a. This is similar to the stationary wave rippling the air cavity entrained behind an object dropped in a liquid[19]. In this case, a reflected stationary wave is needed so that the cavity is attached to the side of the moving object. These situations are reminiscent of classical problems in wave physics, where boundary conditions are impossible to meet without a reflected wave [20].

Sketch of the jet impacting a bath and fitted interfaces of jets

Figure 3. (a) Sketch of the jet impacting a bath. In the first virtual case, a straight jet connects to the meniscus, and exposes the impossibility of a pressure continuity between the two regions. In the second case, the capillary waves allow for pressure continuity. (b) Fitted interfaces of jets of various radii and speeds. The red line is the fitted meniscus, the blue line the wavefield with the calculated amplitude, and the violet cross the calculated matching point (r0,z0)(r_0,z_0).

With five unknowns but only four matching conditions, the problem is under-constrained as stated. However, if we set a θ\theta, we can get all the other values thanks to the following system:

r0=R(R2+zatt2(k2R2−2)+2zattRtan⁡(θ))R2+zatt2(k2R2−1),z0=r0cos⁡(θ)(log⁡(4Rr0ϵ(1+sin⁡(θ)))−Γ),φ=−kz0+arctan⁡(r0−R−zatttan⁡(θ)kzatt(R−r0)),a=(r0−R)exp⁡(z0/zatt)cos⁡(kz0+φ).(2)\begin{aligned} r_0 &= \frac{R\left(R^2+z_{\mathrm{att}}^2(k^2R^2-2)+2z_{\mathrm{att}}R\tan(\theta)\right)}{R^2+z_{\mathrm{att}}^2(k^2R^2-1)},\\ z_0 &= r_0\cos(\theta)\left(\log\left(\frac{4R}{r_0\epsilon(1+\sin(\theta))}\right)-\Gamma\right),\\ \varphi&= -kz_0+\arctan\left(\frac{r_0-R-z_{\mathrm{att}}\tan(\theta)}{kz_{\mathrm{att}}(R-r_0)}\right),\\ a &= \frac{(r_0-R)\exp(z_0/z_{\mathrm{att}})}{\cos(kz_0+\varphi)}. \tag*{(2)} \end{aligned}

where φ\varphi can be shifted by π\pi to get a positive amplitude. To test that our equations of matching indeed set the amplitude of the capillary waves, we follow this fitting procedure: for a jet of given RR and VV, we first get kk and zattz_{\mathrm{att}} by fitting the wave pattern at the base of the jet, or from Patrascu equation (that only depends on RR and VV) when the waves become barely visible on our images. This fixes r0(θ)r_0(\theta) through Eq. (2). We then obtain θ\theta by fitting the meniscus with James’ equation.

When θ\theta is fitted, we inject it in system 2 and deduce the values of z0z_0, φ\varphi and aa. In Figure 3b, the resulting wavefield is plotted on the image, together with James’ equation for the fitted θ\theta. We can see that, for various radii and jet velocities, the obtained amplitude agrees very well with the experimental one. This confirms that the pressure matching between the meniscus and the jet sets the amplitude of the capillary wavefield.

To get a complete model of the interface, we miss a prediction for θ\theta, as all θ\theta are compatible with our equations.

of matching. However, the total energy of the system depends on θ\theta. As can be seen in Figure 4a, when θ\theta is small, the meniscus joins the jet at a higher altitude, resulting in an excess surface compared to a flat meniscus. When θ\theta is large, the amplitude of the capillary waves is larger, so the excess of surface compared to an unperturbed jet is higher. The kinetic energy excess caused by the capillary waves is equal to the surface energy excess according to Chandrasekhar (equation (154) and (57) of chapter XII in [21]). It is then possible to calculate the difference in energy ΔE\Delta E with a virtual cylindrical waveless jet and flat meniscus as a function of θ\theta:

Energy excess and apparent contact angle for jets of various radii

Figure 4. (a) Energy excess of the meniscus (ΔEjet\Delta E_{\mathrm{jet}}), of the jet (ΔEmeniscus\Delta E_{\mathrm{meniscus}}) and their sum (ΔEtot\Delta E_{\mathrm{tot}}) plotted against the speed for a jet of radius R=190 μmR = 190\ \mu\mathrm{m}. The inserted sketches show the difference between the shape of the interface for a low and high contact angle. (b) Apparent contact angle fitted and predicted by our model for jets of various radii, plotted against the speed of the jet. The two experimental images show two reasons for the deviation from our model, the first one being instationarity (at the top) and the second one (at the bottom) the loss of axisymmetry, even when the stationarity is restored.

ΔEtot(θ)=2πγ(2∫z0(θ)∞[R(z,θ)1+∂R(z,θ)2∂z−R]dz+∫r0(θ)∞r(1+∂z(r,θ)2∂r−1)dr)+ρg∫r0(θ)∞πrz(r,θ)2dr(3)\Delta E_{\mathrm{tot}}(\theta)=2\pi\gamma\left(2\int_{z_0(\theta)}^{\infty}\left[R(z,\theta)\sqrt{1+\frac{\partial R(z,\theta)^2}{\partial z}}-R\right]\mathrm{d}z+\int_{r_0(\theta)}^{\infty}r\left(\sqrt{1+\frac{\partial z(r,\theta)^2}{\partial r}}-1\right)\mathrm{d}r\right)+\rho g\int_{r_0(\theta)}^{\infty}\pi r z(r,\theta)^2\mathrm{d}r \tag*{(3)}

where the last term accounts for the gravitational potential energy of the meniscus [22]. The sum of the jet and meniscus energies is plotted in Figure 4a, revealing a value θmin⁡\theta_{\min} which minimizes ΔEtot\Delta E_{\mathrm{tot}}. This θmin⁡\theta_{\min} depends on VV and RR as these variables set kk and zattz_{\mathrm{att}}, which affect ΔEjet\Delta E_{\mathrm{jet}}. In Figure 4b, the prediction of θmin⁡\theta_{\min} is plotted together with the experimental data as a function of VV for different radius. The uncertainty for the experimental value of θw\theta_w comes from the numerical uncertainty on the fit of the meniscus. For small jets (R=56 μmR = 56\ \mu\mathrm{m} and R=76 μmR = 76\ \mu\mathrm{m}), the model agrees well with the experiments without any fitting parameters, especially when V>3 m.s−1V > 3\ \mathrm{m.s}^{-1}. For lower speeds, a/Ra/R is not negligible and capillary waves are not well described by an attenuated sinusoid [23] (see the first image in Figure 3b), so we expect a bigger difference with our model. For wider jets, non stationarities, and a certain degree of swirl [13] can appear (see Figure 4b), which result in deviations from our model, as our assumptions of stationarity and axisymmetry are no longer fulfilled.

In conclusion, the observation of an apparent contact angle when a jet impacts a bath has been rationalized. Capillary waves are due to a pressure matching between the jet and the meniscus, and energy minimization fixes the effective contact angle at least for small jets. Our data are in excellent agreement with this prediction for tiny jets. Discrepancies appearing for wider jets underline the necessity to better understand the non-stationary and asymmetric regimes.

References

  1. [1]Richard Courant. The existence of a minimal surface of least area bounded by prescribed jordan arcs and prescribed surfaces. Proceedings of the National Academy of Sciences, 24(2):97–101, 1938.
  2. [2]Joseph Antoine Ferdinand Plateau. Statique expérimentale et théorique des liquides soumis aux seules forces moléculaires, volume 2. Gauthier-Villars, 1873.
  3. [3]S Heitkam, W Drenckhan, and J Fröhlich. Packing spheres tightly: influence of mechanical stability on close-packed sphere structures. Physical review letters, 108(14):148302, 2012.pubmed.ncbi.nlm.nih.gov/22540826
  4. [4]John Von Neumann. Metal interfaces. American Society for Metals, Cleveland, 108:108–110, 1952.
  5. [5]Sascha Hilgenfeldt, Andrew M Kraynik, Stephan A Koehler, and Howard A Stone. An accurate von neumann’s law for three-dimensional foams. Physical review letters, 86(12):2685, 2001.pubmed.ncbi.nlm.nih.gov/11290011
  6. [6]Nolwenn Le Grand, Adrian Daerr, and Laurent Limat. Shape and motion of drops sliding down an inclined plane. Journal of Fluid Mechanics, 541:293–315, 2005.DOI
  7. [7]Jacco H Snoeijer, Giles Delon, Marc Fermigier, and Bruno Andreotti. Avoided critical behavior in dynamically forced wetting. Physical review letters, 96(17):174504, 2006.pubmed.ncbi.nlm.nih.gov/16712304
  8. [8]Kenneth T Kiger and James H Duncan. Air-entrainment mechanisms in plunging jets and breaking waves. Annual Review of Fluid Mechanics, 44:563–596, 2012.DOI
  9. [9]DA Ervine. Air entrainment in hydraulic structures: A review. Proceedings of the Institution of Civil Engineers-Water Maritime and Energy, 130(3):142–153, 1998.
  10. [10]Jens Eggers. Air entrainment through free-surface cusps. Physical review letters, 86(19):4290, 2001.pubmed.ncbi.nlm.nih.gov/11328157
  11. [11]Élise Lorenceau, David Quéré, and Jens Eggers. Air entrainment by a viscous jet plunging into a bath. Physical review letters, 93(25):254501, 2004.pubmed.ncbi.nlm.nih.gov/15697900
  12. [12]Yonggang Zhu, Hasan N Oguz, and Andrea Prosperetti. On the mechanism of air entrainment by liquid jets at a free surface. Journal of Fluid Mechanics, 404:151–177, 2000.DOI
  13. [13]Claudiu Patrascu and Corneliu Balan. Decay of stationary capillary waves on impinging liquid jets. Physics of Fluids, 33(2), 2021.DOI
  14. [14]D Slavkentes. Wave patterns on a water column. American Journal of Physics, 65(3):225–230, 1997.
  15. [15]J. Lighthill. Waves in Fluids. Cambridge University Press, 1978.
  16. [16]KM Awati and T Howes. Stationary waves on cylindrical fluid jets. American Journal of Physics, 64(6):808–811, 1996.
  17. [17]Matthew J Hancock and John WM Bush. Fluid pipes. Journal of Fluid Mechanics, 466:285–304, 2002.DOI
  18. [18]David F James. The meniscus on the outside of a small circular cylinder. Journal of Fluid Mechanics, 63(4):657–664, 1974.DOI
  19. [19]Torben Grumstrup, Joseph B Keller, and Andrew Belmonte. Cavity ripples observed during the impact of solid objects into liquids. Physical review letters, 99(11):114502, 2007.pubmed.ncbi.nlm.nih.gov/17930442
  20. [20]John Billingham and Andy C King. Wave motion. Number 24. Cambridge university press, 2000.
  21. [21]Subrahmanyan Chandrasekhar. Hydrodynamic and hydromagnetic stability. Courier Corporation, 2013.
  22. [22]The meniscus energy term has here been computed using the uniformly valid composite solution of James, see §6 of (author?) [18].
  23. [23]J.-M. Vanden-Broeck, T. Miloh, and B. Spivack. Axisymmetric capillary waves. Wave Motion, 27(3):245–256, apr 1998.

Paper details

vlnc:W_arpk4dfc
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents