3 replies to this topic

[polipo]

Dear guys,

I am unclear about the equation to estimate minimum number of trays for distillation column according to Fenske-Underwood-Gilland method. The equation is Nm = log[(d_LK/b_LK)(b_HK/d_LK)]/log (relative volatility of light key to heavy key)

where d is distillate rate and b is bottom rate (LK =Light key, HK=Heavy key)

relative volatility (a) = [(a)_top(a)_bottom]^0.5

For my distillation column, I know only the pressure on top and bottom but I don't know the temperature. So I wonder that what temperature I should use for estimating relative volatility at the top and bottom of the column.

In my column, there is mixture of HC ranging from C2 to C4 (paraffin and olefins). The distillate is designed to obtain 99% ethylene. So my light key component is ethylene whereas heavy key is propylene.

I don't know whether I did it wrong or not. The temperature at the top is its bubble point at that pressure. On top, I assume that only ethylene 99% and propylene 1% mole. On the other hand, the temperature at the bottom is bubble point of the mixture at that pressure. At the bottom, I estimated the bubble point from every component, not only ethylene and propylene.

I'd appreciate if you could comment on my idea to calculate this one.
Thank you.

Edited by polipo, 25 August 2010 - 10:06 AM.

[Propacket]

polipo,

You can’t calculate relative volatility unless you have temperatures available. You said that “temperature at the top is its bubble point at that pressure. On top, I assume that only ethylene 99% and propylene 1% mole. On the other hand, the temperature at the bottom is bubble point of the mixture at that pressure”. Now problem is how to calculate the bubble point temperatures? In distillation, you must have studied a term K (Equilibrium Ratio). This K value is used to estimate the bubble point and dew point temperatures. For this, you will need K- Value charts. You can find this in various datebooks. Procedure for estimating the bubble point is as follows.

• For each component, select a temperature in chart and find the K-Value at operating pressure.

• Multiply the K-Value of each component with its mole fraction (N). You will get a figure NK for each component.

• Sum up NK values for all the components.

• At bubble point temperature, this sum must be equal to one. If it is not so, change the temperature and repeat the same procedure until you get sum equal to one.

For dew point temperature:

• You need to divide mole fraction (N) by K-Value. Now, you will get a figure N/K.

• Sum up N/K values for all the components.

• At dew point temperature, this sum must be equal to one. If it is not so, change the temperature and repeat the same procedure until you get sum equal to one.

Thanks

Edited by P.Engr, 26 August 2010 - 04:02 AM.

[polipo]

Yep
But I don't even know the temperature. So I can't determine K value though. That's why I ask whether assuming bubble point for bottom and dew point for the top is appropriate or not. If it is alright, I can use this to calculate K value. And I've made an assumption on top that 99% ethylene because that's the design specification.

Thanks

[Propacket]

polipo,

Please read my earlier post carefully. While reading the K-Chart, you need to make a random assumption of temperature. Read the K-Value for each component from chart at this assumed temperature. Obtain NK values and add them. If the sum is equal to one, your assumption was right and assumed temperature will be bubble point temperature. Otherwise you should assume a new temperature and repeat the same procedure until you get the sum equal to one.

Hope it will be much clear now.

Thanks

Edited by P.Engr, 27 August 2010 - 03:58 AM.