1 reply to this topic

[psychodan]

I'm doing my university design project on the Monsanto process for the production of acetic acid at the rate of ~25 te/hr. The reactor is a gas-liquid stirred tank reactor operating at 40 bara and 180°C. I've worked out the reactor size to be ~110m3 which to me seems quite big for a vessel operating at 40 bara pressure. I've not got much practical experience and I was wondering whether anybody could give me some guidence on whether this is a ridiculous figure?

Thanks,
Dan

[kkala]

I'm doing my university design project on the Monsanto process for the production of acetic acid at the rate of ~25 te/hr. The reactor is a gas-liquid stirred tank reactor operating at 40 bara and 180°C. I've worked out the reactor size to be ~110m3 which to me seems quite big for a vessel operating at 40 bara pressure. I've not got much practical experience and I was wondering whether anybody could give me some guidance on whether this is a ridiculous figure?

It seems that the size of 110 m3 is not excessive, even though it may be on the limit. However this does not mean that economic considerations would support this reactor size instead of (say) two reactors for acetic acid production rate of 25 ton /h. Following data seems to support this view:
1. Alumina is leached from bauxite (digestion) in autoclaves of 100 m3. Temperatures up to 260 0C are encountered, resulting pressures in the autoclave do not differ much to 40 bara (though lower).
2. Assuming an acetic acid reactor of D=3.3 m & H=13.2 m (total capacity=112.9~110 m3), its wall thickness for a design pressure of P=43 barg would be (a) e=0.5*P*D/f (f=steel design stress=140 Nt/m2); so e=0.5*4.3*3300/140=51 mm (total metal weight w = 154*0.051*8=63 ton). Standard steel plates of such thickness can be found in the market (e= up to 5 - 12 in), even if not widely used.
(a)note: Estimate of thickness is simplified, cladding is assumed inside reactor.
3. Assuming two reactors in series of D=2.6 m & Η=10.5 m each (total capacity=112.6~110 m3), s=40 mm (estimated as before) & w=193*0.04*8=62 ton (roughly same as before). In this case two stirrers would operate, but the system will have more flexibility; e.g. production would not cease (but will be reduced) whenever one reactor is out of operation for maintenance.
4. Specific technical literature can clarify the subject; if there is no time for this investigation, I would proceed with two reactors in the exercise and give an explanation. However there are processes preferably using a single reactor for simplicity and lower maintenance cost, such as: ICI's low density polyethylene production in an autoclave, Rhone-Poulenc phosphoric acid leaching from phosphates in a CST Reactor (but other methods use multiple CSTRs for phosphoric acid).

Edited by kkala, 11 April 2010 - 02:25 PM.