3 replies to this topic

[DJCoulton]

Our group is currently in a technological selection stage for a Ammonia Production design study. We're producing ammonia via the Haber-Borsch process and we're getting our Hydrogen feed from coal gasifier. The gasifier is air blown and thus we have a large amount of N2 throughout the system.

 

After wet scrubbing, sour-shift and H2S removal, we're considering using PSA or H2-selective membrane to generate a (relatively) pure H2 feed for our ammonia synthesis loop.

 

My question is this; is PSA a viable technological choice? There is a lot of literature on PSA for removing traces amount of N2, however at this stage our N2 is roughly 46 vol% which is quite significantly higher. Will the PSA be able to separate H2 from what is essentially N2 dominant stream?

 

 

According to literature membrane will work wonderfully but it is a much less proven technology.

 

[gegio1960]

My 2 cents....

Both PSA and Membranes are widely used for H2 purification.

PSA is preferred when you need high purity of H2 at high pressure

Membranes are preferred when you need very high purity of H2 product and you've differential pressure to utilize.

Good luck!

[Art Montemayor]

DJ:

 

The attached, old magazine article that I've kept as a fond memory should serve to educate you on how practical and proven the PSA process is.

 

In fact, I kept this article because I worked for Howe-Baker Engineers in Tyler, Texas during the years 1973 -76 and during this time we were designing and fabricating skid-mounted, pure hydrogen (99.999% vol) generators using steam reformers.  I was instrumental in getting one of these units into a Belgium Furfuryl Alcohol plant (Geel, Belgium) in 1974 (I was the Project Manager that installed that FA unit) and the last time I checked on the Internet, I could still see the PSA unit was working.  Does that tell you something (besides impressing you with Texas engineering capabilities) about the feasibility of using PSA for Hydrogen production?

 

The process works - obviously - but the tradeoff is that you must have an economical use for the inevitable purge gas generated.

 

 1979 Union Carbide Polybed PSA.pdf   3.12MB   29 downloads

 

[DJCoulton]

Thank you guys - you've given me a lot to think about.

 

Another point I'd like to rise is I've read somewhere that there's a minimum H2 % of 50% in the feed to make the size of the PSA bed economically feasible. Our gas contain 37% H2. Whereas for membrane the minimum H2 level is at 15%.

 

However having said that there is Jacob's Ammonia process which utilizes PSA to reduce N2 from excess to stoichiometric level so I believe PSA will work.

Edited by DJCoulton, 15 August 2013 - 08:19 PM.