An experimental and theoretical investigation of evaporating
meniscus dynamics and instabilities
Continuing advances in electronics are leading to a wider array of applications and higher thermal
dissipation demands. To address these increasing demands, phase change heat transfer devices such as
heat pipes can be used in which their operation is being pushed to regions of transients and instability.
The study of meniscus based phase change and any associated dynamics such as those found in heat
pipes has led to two principle questions:
What aspects govern evaporating meniscus dynamics?
What are the principle contributors in thin-film stability?
A set of theoretical and experimental analyses were performed using cylindrical capillary tubes and
evaporating menisci to study meniscus dynamics. The dynamics model as developed, describes an
evaporating meniscus with a unidirectional flow and a simplified interface. The model was validated
using different capillary tube sizes and fluids. The experiment and model were found to have varying
degrees of consistency in addition to the discovery of a correlation between meniscus height and the
applied superheat driving evaporation.
When a meniscus is host to phase change, the interface adjusts to form an asymptotically bounded
curved interface. The resulting curved evaporating thin-film may have potential to destabilize giving way
to meniscus dynamics. The stability of a curved evaporating thin-film was investigated with
consideration to a variety of forces resulting in a non-linear differential equation. Using a linear stability
analysis, the potential for instability was studied for three cases with the results indicating a spatial
dependence for non-trivial interface geometries.
Experimental studies of meniscus instability used heated glass channels of varying widths and three
fluids to incite evaporation driven meniscus instability. Two kinds of instability were found for the
alkanes, while acetone only had one kind of instability. The first kind of instability unique to the alkanes
was localised while the second kind was common to all fluids tested. Using the superheat and meniscus
height relation from meniscus dynamics, the second kind of instability was found to require larger
superheats with decreasing channel widths. For the same channel width, the superheat required to
initiate instability was found to be different for each fluid.