The coolant is between tow heated walls. Thank you

Hi! Heat transfer (by conduction, convection, and radiation) in a nuclear reactor is well described in classic textbooks. I suggest R. A. Knief - Nuclear Engineering [1992] (see Chapter 7, also Figure 7-4).

However, that treatise assumes a constant bulk temperature of the coolant and only 1D geometry (from fuel pellet center outwards). Also, the last step - nucleate boiling for transferring the heat from the cladding to the bulk coolant - is not modelled explicitly.

If I understand correctly, you would like a solution between two fuel rods, and at low/zero reactor power, where the natural convection is the principal force for coolant movement (i.e. no forced flow by pumps). The approach above might still be applicable. Or even more applicable, as there is less turbulent flow and (possibly) no nucleate boiling.

Does that help you?

Hi! Heat transfer affects the density of water, hence the weight, and pressure is created. The temperature decreases, the water becomes heavier and sinks down. The temperature increases, the water becomes lighter and rises up. If there is an organized circulation loop, the process will flow.