A physics preprint reports that the relationship between a droplet's speed and size changes as the surface beneath it is tilted more steeply. In tests of water drops moving through an oil-lubricated inclined system, the velocity-size exponent rose from approximately 3/2 toward 9/4 as the inclination increased. The number is a compact way of describing how sharply speed changes with drop size: the higher it is, the more strongly speed separates as size grows. The result points to an angle-associated change in motion scaling, but not to a directly observed change in the contact region beneath each drop.
How the test was set up
To test the relationship, the researchers used water drops with controlled volumes of 10, 20, 40, 100, 200, 400, 800, 1,400 and 3,000 microlitres. Their equivalent undistorted radii ranged from 1.36 to 8.92 millimetres, with volume-derived radius uncertainty below 0.3%. The drops moved along a 27-centimetre acrylic slope immersed in olive oil. The oil's viscosity was monitored and calibrated at between 60 and 70 millipascal-seconds.
Images were analysed with ImageJ. The measurements included creeping velocity, maximum width along the slope and maximum height, while an oil film formed beneath a moving drop. The team then compared the speed-size relation at fixed inclinations and looked for whether higher-angle results could be represented by one master curve.
Those two endpoint exponents come from the study's theoretical reference cases. The quasi-spherical regime predicts 3/2, while the pancake regime predicts 9/4. The model uses LLD theory to describe the oil-film thickness and places the relevant velocity gradient in the dynamic meniscus. The observed drift from one value toward the other is therefore treated as a crossover in the scaling of motion, not as proof that the whole droplet has already become flat.
Speed and shape diverged
That distinction matters because the visible dimensions did not change in a simple, matching way. At a fixed droplet size, increasing inclination was associated with increases in both measured streamwise width and height. The transverse width, which runs perpendicular to the rolling direction and the slope normal, was not measured; volume conservation implies that it decreased as the other dimensions increased.
Drop size itself also altered the outline. For small equivalent radii, width and height were close to twice the radius. As radius increased, width grew while height fell below the spherical value, and for sufficiently large drops the height tended toward the value set by the capillary length.
At higher inclinations, especially from 8 to 25 degrees, the data were reasonably described by a single master curve. Fitting the reported scaling law produced a numerical multiplier, called the prefactor, that levelled off at 0.11. This means the high-angle measurements shared one approximate scaling form, even though the agreement was not exact.
Most of the data that followed pancake-like velocity scaling still remained far from the pancake-limit height. The study therefore found a mismatch between dynamic behaviour and global appearance: the velocity law approached the pancake prediction more clearly than the measured overall height did. A drop could move according to the flatter regime's scaling without yet having the full macroscopic shape associated with that limit.
What the data cannot settle
The proposed mechanism remains an inference. The effective contact radius was not directly resolved, so the suggested crossover in contact length was inferred from velocity data rather than directly observed. The measurements can show an angle-associated change in the scaling law, but they cannot by themselves establish the precise geometry of the lubricated contact.
There was a second unresolved issue in the fitted number. The saturated prefactor, about 0.11, was roughly three times smaller than 0.34 from the cited bubble experiments and more than four times smaller than the theoretical comparison value of 0.474. The source of that gap remains unclear.
At the onset of motion, some small drops did not move: 10 and 20 microlitre drops at 0.4 degrees and 10 microlitre drops at 1 degree remained stationary, so no velocity measurements were obtained for those conditions. The low-angle, small-drop part of the comparison therefore has missing speed data.
The document is an arXiv preprint, version 1, dated 28 August 2026. The authors report support from JSPS KAKENHI Grant Number JP24K00596.
Paper data and sources
Original title: Inclination-Induced Crossover in the Velocity Scaling of Lubrication-Mediated Droplet Motion
Authors: Haruka Hitomi, Ko Okumura
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-28
DOI: Not available
Original paper · Full text