The possibility of molten salt freezing in pipe flow systems is a key concern for the
solar-energy industry and a safety issue in the Generation-IV molten-salt reactors,
worthy of careful consideration. The overriding aim of this thesis is to address this
issue by providing an approach to quantify the solidification of molten salts in piping
systems (in terms of mass build-up, effect on the flow and heat transfer, etc.).
In light of this aim, several aspects needed to be investigated which affect how
molten salt solidification can be predicted. Specifically, the work described in this
thesis is hereby described: 1) An experimental method was developed to measure
the thermal conductivity of molten salts, whose uncertainties significantly affect
further modelling efforts. The method can be applied to measure the thermal
conductivity of molten salts up to temperatures around 760 K, with an overall
error better than 4%. 2) The thermal conductivities of NaNO3 -NaNO2 -KNO3
eutectic (HTS) and LiCl -KCl eutectic were measured up to temperatures of 700 K
and 760 K respectively. In addition, data in the literature were re-evaluated by
taking into account the thermal losses present in a particular experimental apparatus;
the revised results were found to be in good agreement with other studies.
These re-evaluated data and the measurements conducted in the present study
were used to critically review and suggest the values of the thermal conductivities
of common salts, including FLiNaK. 3) A 1-dimensional thermo-hydraulic model
was developed under the steady state assumption and validated to predict transient
freezing in internal pipe flows. The model can be incorporated in standard
thermo-hydraulic codes and can be used to predict the solidi cation process in
complex piping system where CFD is computationally expensive. 4) An experimental
apparatus equipped with laser-based diagnostic measurement techniques
was built to measure the growing thickness of an ice layer in contact with a cold
surface and liquid water flow. The developed freezing model was validated against
these experimental data and the discrepancies were considered. 5) The freezing
model was then applied to study the behaviour of the Direct Reactor Auxiliary
Cooling System (DRACS) under Loss of Forced Circulation (LOFC) with blackout.
DRACS was found to be prone to failure due to freezing in the molten salt/air
heat exchanger; its transient response was characterised and discussed.