Atmospheric Ammonia Storage Tank 2

[sheridan]

I'm looking for some advice concerning the decommissioning of an atmospheric ammonia storage tank. Tank in question does not have a drain, only connection at bottom of tank is pump suction line (DN250). This line ends about 100 mm above the tank-bottom. Set pressure of RV is 165 mbarg.

With normal emptying procedures the tank level doesn't get below approx 450 mm.

I've been toying with the idea of pushing a flexible hose (approx. DN32) through the DN250 line and pumping more liquid out and afterwards evaporating the remaining liquid by adding hot ammonia vapour (approx. 50 ºC) through the DN250 line.

There's no scrubber available, no flare and no possibility to use aqueous ammonia (to make it interesting).

Your remarks/suggestions are welcome.

S. Gabrielse

 

[BenThayer]

I have been involved with emptying tanks that were 170 ft diameter and 64 feet sidewall for 30,000 tons of liquid ammonia at -28°F.

We would use a small transfer pump from a side sump until we got to the last 6" in the tank and then turn up the tank's foundation heaters to vaporize the rest.

The requirements to empty the tank really need to be addressed during the design of the tank but rarely really are.

Our tanks' floor had a slight cone shape so that the center was a few inchs higher than the outside. This helped reduce the amount of liquid when there was only 6" remaining at the outside edge of the tank.

Depending on the sump arrangement and the pressure design of the tank, you can raise the tank's operating pressure a few inches water column.

We typically operated the tank at 6" to 8" of water column but would raise the pressure to 16" to keep suction to the transfer pump longer. Our transfer pump was a temporary pump and much smaller than the normal pumps used for loading. You have to be careful doing this because the relief devices could be set at 20 to 27" of water column. But we could do this with the refrigeration compressors keeping up quite well because with so little liquid in the tank, there was very little heat gain.

Good luck. If you are emptying the tank for inspection/repair purposes, you will want to do a search for Ammonia Stress Cracking and pay careful attention to removing the oxygen when you place the tank back in service.

 

[sheridan]

Thanks for the response.

I forgot to mention that the decommissioning had to be done within 15 days (that's including purging with N2 and air).

To my knowledge there is no cone shaped floor (unfortunately) and I have no information about the foundation heaters.

 

[BenThayer]

I am unsure where you are located but if you look at the quantity of gas at atmospheric pressure in the tank, you may be rather surprised. For our 30,000 ton tank we had essentially the equivalent of a truckload or 40 tons of vapor when the tank was completely empty of liquid.

A flare is essential to safely empty the tank. 15 days is not enough time in my opinion.

The basics of clearing the tank (after all liquid is out, difficult task in its own right) is to purge with nitrogen until you have it below the lower explosive limit and then you can consider adding air to the tank. Use extreme care during this transition period or you will have a huge danger. Even after you get air into the tank, it will take days/weeks to get the ammonia fumes to a limit where you can enter the tank. Calculate the nitrogen volume you need to get it below the LEL. Since this is NOT a pressure tank, there are severe limits to how fast you add/remove the nitrogen. Then calculate the air volume you need to dilute the ammonia to an acceptable level within the limits of the tank's pressure ratings.

This is a very long and tedious procedure. To vent 40 tons of ammonia vapor without a flare over several days will be a significant risk to anyone in the area.

You may look at some old AIChE Ammonia Safety Symposium meeting documents to see if there are any references to how long others took to empty their tanks.

I worked for a producer and we were fortunate to have liquid nitrogen vaporizers on site and a flare to vent through and it still took us an entire summer to inspect a tank. At least 4-6 weeks to empty, 4 weeks to inspect and 2 weeks to return to service.

Placing in service was to purge to less than 1% oxygen using nitrogen and ammonia vapor to remove the nitrogen and then cooldown the tank to where we could freely add liquid without concern.

Good Luck!!

 

[gramya]

I know of an expert who has handled 10 such storage planning commissioning deccommissioning in India. He may be of some help. you may try to contact for some useful tips. I am also trying to contact him, but he is in Doha Qatar for training programs on a relevant topic, Hazardous Chemical Handling & management, at Qatar Petroleum.
He may be contacted at rjsingh1@gramya.com or grai_1999@yahoo.com

NK Roy

Narendra K. Roy
Gramya Research Analysis Institute,
PO box 4016, Vadodara 390015, India
Website:

http://www.gramya.com

;

http://www.charismaglobal.com

 

[sheridan]

Mr Thayer,

can you please contact me at s.gabrielse@continentalengineers.nl ?

 

[edcat]

I had a similar situation in 1984 trying to deinventory a 30,000 toa NH3 storage tank. Authored a paper in 1985 and presented it at the AIChE NH3 Symposium. We did it by taking vapors off the top of the tank through the NH3 refrigeration system and instead of returning them to the tank we sent the vapors from the refrig system to a small auxiliary tank that sat above the submerged transfer pumps where the compressed ammonia flashed, cooled and formed a liquid level that was sufficient head for the transfer pumps. We were able to pump the tank out until only puddles remained on the floor. Also had some interesting problems in purging the tank with hot nitrogen due to the nature of ammonia, but that's another story. Contact me at efiesinger@zephyrenv.com for details.