1
$\begingroup$

I have modelled a multi-tubular reactor performing a Fischer-Tropsch synthesis of alkanes and I was wondering about the stability of such a reactor. Even a very tiny deviation of the wall temperature can lead to a potential destruction of the catalyst if not more...

The Fischer-Tropsch synthesis I am modelling as an adiabatic temperature around 1200 K. This is the result I get when the wall temperature varies from 201.000 °C to 201.470 °C.

Wall temperature Fischer Tropsch synthesis

So theoretically a variation of less than 0.5 K can lead to a reactor temperature variation of more than 150 K.


How to control the temperature along the reactor (it can be more than 10 meter long) as precisely as let 0.1 K say of difference to avoid such unpleasant drift leading to a potentially bad situation?

$\endgroup$
2
  • $\begingroup$ We need to see the model of the reactor and the heat transfer. Something is driving the exponential increase in temperature along the tube and until we find what it is, we can do nothing. $\endgroup$ Commented May 18, 2019 at 12:32
  • $\begingroup$ what was your resaerch so far? Googling gives me some ideas, but I'm no sure if they are new to you. Does your model assume a uniform T_w along the reactor length? Jeffrey: FT is strongly exothermal. $\endgroup$
    – mart
    Commented Jun 19, 2019 at 9:10

1 Answer 1

2
$\begingroup$

I've had this problem before, albeit not with such a tight temperature range.

I would recommend immersing the reactor in a fluid with high thermal mass, then I would circulate and temperature control the fluid. The circulation will minimize the effects of temperature differences.

Water is ideal, but dangerous since it would have to be under pressure, and it would explode into steam if there was a breach. Oil might work fine.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.