Gas Separation Basics

[bonzoboy]

I'm looking at some processes, and trying to be a handle on capturing the basic energy requirements for separating gas A from B (or B/C). It's relatively easy to calculate the free energy change that occurs when two gases are allowed to mix, but is that equivalent to the MINIMUM energy required to unmix these same two gases, at 100% efficiency. When I look at the analysis in Perry's, I don't come close to the numbers published in the sixth edition (on Cryogenics).

Can anyone direct me to some basics on gas separation, either by cryogenics or other methods (inert, non-reactive gases).

Bonzoboy

 

[MortenA]

It dosnt require any energy to SEPARATE the phases (that must be a gravity process). The enrgy is require for phase changes (changes to pressure while maintaining temperature and/or volume).

But if you are talking about a mixture of e.g. methane and ethane in a gas mixture and you want to separate these two components - the i belive its the same as what goes into mixing them.

Best regards

Morten

 

[davefitz]

It has been over 30 yrs since I was in the cryogenic business, but the energy used to seperate the gases using a cryogenic process is several times the minimum needed based on a process entropy dictate.

I recall from a grad course in compressible fluid mechanics a theoretical scheme whereby gases could be seperated at very low energy consumption rates if they are expanded thru a turbine with exhaust velocities at choked conditions- the centrifugal forces would be sufficient to cause 2 different constituent streams- but it has never been implemented for that specific purpose ( but I think something similar is used in the Oak Ridge type of industry).

 

[pmover]

davefitz,

google search with text ABB Randall.

hopefully the website will contain useful info. if not, post again and i'll get you a little more explanatory info (may take awhile to dig up the docs).

hope this helps!
good luck!
-pmover

 

[davefitz]

pmover:
As I read the Randall webpages, their use of the turbo expander is soley to produce a cyryogenic gas mixture due to isentropic expansion of the inlet stream- that process is not much a change from the Linde process of 50 yrs ago. The sepeartion would still need to utuilzies a distilaltion column and liquifaction, as I understand it. Certainly, one can today recover most of the energy produced by the turbine as useful electrical energy using the same technoligies used by Capstone microturbines for synchronizing to the grid a variable speed turbine, as opposed to the old Linde techique of wasting the turbine energy by powering a parasitic fan.

What I was referring to was a direct seperation of the mixture as gases, at the turbine exhuast due to the gas constituents' different densities, choked velocities, and other effects whose description no longer reside in my brain cells.

 

[pmover]

davefitz,
your understanding for additional process equipment is correct. what you describe is an interesting concept, but i'm not familiar with what you describe; that is direct gas seperation without other process equipment. i do know that at one time some fellow engrs looked into permeable membranes for gas seperation, but i do not know the outcome/results nor the gases to be seperated (think it may have been CO2 from combustible sources - ICE, gas turbines, heaters, boilers, etc. - not certain though).

sorry not much of a help, but good luck! will keep this concept in mind though . . .
-pmover

 

[UmeshMathur]

bonzoboy:

As a matter of process thermodynamics, no process exists that can reverse the effects of mixing gases. Usually, a huge expenditure of energy - wholly out of proportion to the free energy of mixing - is required.

All real world processes seek to exploit some molecular-level difference between the components. In large-scale chemical processing, this is generally the boiling point difference; hence the use of cryogenic distillation.

The use of semi-permeable membranes has worked for separating hydrogen economically for many years. Similarly, adsorption and pervaporation have been considered as alternatives o distillation when there is a substantial difference in molecular functional groups.

For such processes, in my opinion, a theoretical analysis of entropy, exergy, free energy, etc. is primarily of academic interest. Mixing of gases is not a reversible process by any stretch of the imagination. You might recall discussions of "Maxwell's Devil" in your introductory thermodynamics classes as a way to understand the underlying phenomena.

 

[bonzoboy]

THanks for the feedback. I realize that a real process would not be isentropic, and would require substantial energy input. I am curious if whether my reasoning was correct, that the bare-absolute minimum requirement (actual energy requirements would be several times that) is the difference in free energy between the mixed gases and the unmixed gases.

 

[davefitz]

As I recall from thermodynamics, the min energy needed for seperating the mixed gases is T*(delta S), where T is the absolute temperature, and (delta S) is the increase in entropy caused by mixing 2 pure gases together by the imagined process of bursting an intervening diaphragm.

The energy used in the either the cryogenic process or PSA processes is several times that minimum energy.