Spline-based Interface Modeling and Optimization (SIMO) for Surface Tension and Contact Angle Measurements

Karan Jakhar, Ashesh Chattopadhyay, Atul Thakur, Rishi RajarXiv 2019 · DOI · arXiv

Abstract

Surface tension and contact angle measurements are fundamental characterization techniques relevant to thermal and fluidic applications. Drop shape analysis techniques for the measurement of interfacial tension are powerful, versatile and flexible. Here we develop a Spline-based Interface Modeling and Optimization (SIMO) tool for estimating the surface tension and the contact angle from the profiles of sessile and pendant drops of various sizes. The employed strategy models the profile using a vector parametrized cubic spline which is then evolved to the eventual equilibrium shape using a novel thermodynamic free-energy minimization-based iterative algorithm. We perform experiments to show that in comparison, the typical fitting-based techniques are very sensitive to errors due to image acquisition, digitization and edge detection, and do not predict the correct surface tension and the contact angle values. We mimic these errors in theoretical drop profiles by applying the Gaussian noise and the smoothing filters. We then demonstrate that our optimization algorithm can even drive such inaccurate digitized profiles to the minimum energy equilibrium shape for the precise estimation of the surface tension and the contact angle values. We compare our scheme with software tools available in public domain to characterize the accuracy and the precision of SIMO.

Introduction: The Challenge of Noisy Drop Profiles

In fields ranging from thermal management to microfluidics, accurately determining the surface tension of a liquid and its contact angle on a surface is critical. The standard approach, Drop Shape Analysis (DSA), involves capturing an image of a drop—either sessile (sitting on a surface) or pendant (hanging from a needle)—and mathematically analyzing its profile.

However, experimental images are rarely perfect. Factors such as camera resolution, lighting conditions, and edge-detection artifacts introduce "noise" into the digitized profile. Traditional techniques, including polynomial fitting and Axisymmetric Drop Shape Analysis (ADSA), are notoriously sensitive to these imperfections. Even minor errors in edge detection can lead to significant inaccuracies in surface tension calculations. This research introduces SIMO (Spline-based Interface Modeling and Optimization), a method designed to use physical constraints to drive noisy data toward its true thermodynamic equilibrium state.

Methodology: The SIMO Framework

The SIMO approach consists of two primary stages:

  1. Spline-based Representation: The drop profile is modeled using a vector-parametrized cubic spline. This mathematical framework provides a continuous representation of the interface, allowing for precise calculations of local curvature, surface area, and volume.
  2. Thermodynamic Optimization: This core innovation utilizes a Pinning (PN) perturbation operator. Instead of a simple "best fit" line, the algorithm identifies areas of high and low "equivalent curvature" based on the Young-Laplace equation. It iteratively simulates fluid flow by moving tiny volumes from areas of high pressure to low pressure, evolving the noisy profile into a theoretical shape that satisfies thermodynamic equilibrium.

To identify the true physical properties, the algorithm evaluates the Root Mean Square Deviation ($\sigma_{\chi_{eq}}$) between the noisy profile and various evolved equilibrium shapes. The shape with the minimum deviation identifies the correct Bond number ($Bo$), which is directly related to surface tension.

Key Results and Performance

The researchers evaluated SIMO using both experimental data and synthetic "corrupted" datasets. In tests on theoretical images intentionally degraded with Gaussian noise, standard fitting methods showed massive errors (predicting a Bond number of 2.84 when the true value was 1.0). In contrast, SIMO predicted a $Bo$ of 1.05, demonstrating remarkable robustness to image quality.

Drop profiles and optimization curves for different Bond numbers.

As shown in Figure 2, the algorithm effectively identifies the minimum energy state across different initial conditions. Figure 2a illustrates drop profiles for various Bond numbers at a contact angle of $133^\circ$, while Figure 2b shows the corresponding minimization of the equivalent curvature deviation. The clear minima in the $\sigma_{\chi_{eq}}$ plots confirm the algorithm's ability to converge on the correct physical parameters.

Experimental validation using water drops at $20^\circ$C yielded a surface tension of $72.75 \pm 0.31$ mN/m, matching the accepted literature value of $72.75 \pm 0.36$ mN/m. Benchmarking against open-source tools like OpenDrop and DropSnake further confirmed SIMO's superior accuracy, particularly in contact angle measurements below $90^\circ$.

Conclusion

SIMO represents a significant advancement in drop shape analysis by providing a physics-based "safety net" for experimental data. While future work may refine optimization parameters like drop volume and contact point coordinates, the current iteration offers a robust solution for researchers working with imperfect imaging setups. By prioritizing thermodynamic equilibrium over simple geometric fitting, SIMO ensures that experimental noise does not compromise scientific accuracy.

Summary updated

Abstract

Surface tension and contact angle measurements are fundamental characterization techniques relevant to thermal and fluidic applications. Drop shape analysis techniques for the measurement of interfacial tension are powerful, versatile and flexible. Here we develop a Spline-based Interface Modeling and Optimization (SIMO) tool for estimating the surface tension and the contact angle from the profiles of sessile and pendant drops of various sizes. The employed strategy models the profile using a vector parametrized cubic spline which is then evolved to the eventual equilibrium shape using a novel thermodynamic free-energy minimization-based iterative algorithm. We perform experiments to show that in comparison, the typical fitting-based techniques are very sensitive to errors due to image acquisition, digitization and edge detection, and do not predict the correct surface tension and the contact angle values. We mimic these errors in theoretical drop profiles by applying the Gaussian noise and the smoothing filters. We then demonstrate that our optimization algorithm can even drive such inaccurate digitized profiles to the minimum energy equilibrium shape for the precise estimation of the surface tension and the contact angle values. We compare our scheme with software tools available in public domain to characterize the accuracy and the precision of SIMO.
BibTeX
@misc{jakhar2019spline,
  author = {Karan Jakhar and Ashesh Chattopadhyay and Atul Thakur and Rishi Raj},
  title = {{Spline-based Interface Modeling and Optimization (SIMO) for Surface Tension and Contact Angle Measurements}},
  year = {2019},
  eprint = {1909.05943},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1909.05943},
}

Loading related content...

Figure