1 reply to this topic

[chemehou]

Hi All

 

I am a student working on a project. The project is about a fixed bed reactor for methane combustion at relatively low temperature (450 C). Now I have made a model using comsol to see the heat transfer performance of a resistance wire inside the bed. Due to the low thermal conductivity of the catalyst pellets, the coil surface temperature is very high while the bed temperature is low at steady state. Could you please give me some suggestion on how to deal with this?

 

Truly thankful for every reply!

[Art Montemayor]

You are a student and I can understand your frustrations in trying to deal with this application of heat transfer.  I think I understand what you are proposing.  If I am not correct, then please clarify what I understand.

 

I have never heard of, read, or imagined this type of heat transfer being applied - or even attempted.  In my opinion you are wasting your time trying to apply effective heat transfer in a packed bed as derived from an electric resistance element or a "heat" pipe flowing with steam or some other hot fluid.  The same applies if you try to cool the bed using a cold fluid in an embedded pipe coil.  It wont' work.  What you are describing is that there will be intimate contact between the surface of the electric element and all the catalyst in the bed.  You state that the low thermal conductivity of the catalyst pellets is what is limiting your required heat transfer.  I say that you are wrong.  The low thermal conductivity of the pellets is NOT the culprit.  The bad feature of what you propose is the limited pellet SURFACE in contact with the heating element.  Try to visualize a billiard table and the billiard balls on top  of it.  If the table were very hot, would you expect the balls to be receptive to receiving the heat from the table - bearing in mind that the only heat transfer taking place is that existing between the contact surfaces of the balls and the table.  Eventually you could possibly heat the balls but it would probably take hours.  You are trying to heat by conduction, so you need contact surface.  Instead, try convection heating.

 

The way I've heated and cooled fixed bed reactors and other vessels is to heat or cool a heat transfer fluid (preferably inert) externally to the vessel and subsequently flow the same fluid (whether gas or liquid) through the bed.  Common sense would tell you to use an inert gas such as nitrogen preferably instead of a liquid in the interest of eliminating the need to drain and purge the bed afterwards.

 

You describe your process as one where you are combusting methane within the bed.  I assume once you heat the bed and initiate the combustion, the bed will remain hot.  Therefore, all you need is the initial heat up.  Is that correct?