[Discussion] Effect on Shell Tube Exchangers of varying gas Molecular Weight

[Iomcube]

This is a discussion not a question because it just happened to me that how can it effect my future designs...

Lets keep the discussion to S&T exchangers both in vertical & horizontal orientation.

We have two gases wet-Cl2 & wet-H2 coming out from electrolyzers at 80C & are going in exchangers to get dry... Because latter on they have to burn in HCl furnaces along with other factors.

Now for Cl2 Mr=71; for H2=2 & for H2O=18

Recall humid air is lighter than dry air; so in essence wet Cl2 is lighter than dry Cl2 & wet H2 is heavier than dry H2!

Now both are being cooled by cooling water at 30C BUT Cl2 exchanger is vertical while H2 horizontal (I will not discuss which orientation is better...)

These wet gases will be going through partial condensation within their respective exchangers & along the length Cl2 will be getting heavier, relatively & H2 lighter (water is being removed). Will this aspect has any significance in the exchanger design or selection?

 

[TiCl4]

How much water is there? Slightly wet chlorine should have HCl and HOCl, not free water, correct? Or is there not enough energy to drive that reaction to completion at 80 C? I'm curious here.

Most condensers will typically have two zones in them - a relatively isothermal condensing zone and a non-condensing cooling zone that is after the majority of the condensing is complete. For condensers, much of the heat transfer in the condensing zone is accomplished by the actual condensation process - the collapse of the vapor into liquid creates a localized low-pressure zone, and encourages radial mixing. If there is a large amount of condensable gases, then the change in molecular weight shouldn't make a huge difference along the condensing zone of the condenser.

If you have very low %s of condensable gases, the system will likely more behave as a simple cooled gas. For a simple cooled gas, you can evaluate how the Nusselt number changes with change in composition using correlations (Gnielinski) for forced convection in tubes. Without evaluating myself, I feel that the increase in gas MW (thus density) will tend to increase the heat transfer coefficient of a stream. Following that, I assume the H2 stream would become highly inefficient at cooling once all the water has been condensed out necessitating a larger Hx area in the non-condensing zone. Vice-versa, the Cl2 stream would become slightly more efficient.

I'm not sure what % condensables you are starting with, but I would guess it is a relatively low %. Thus my guess in the second paragraph would be more apropos to your situation.

How has the jived with your experience?