failure of pump in phosphoric acid plant

[psamieyan]

in our plants we have 10 centrifugal low pressure pumps, their impeller which is made by cast 20 alloy is corroded extremely, the plant is phosphoric acid and total analysed F is measured 14.5% and SiO2 15% and Cl 65 ppm, P2O5 is also 21.1%, now I want to know that with this kind of components: is it possible to have extreme corrosion which can cause failure of pump after only 1 month, if yes which material is better for this equipment.

I'll be happy to recieve your comments.

 

[randi]

Checkout with Hastelloy, Inconel, Monel, Haynes, and Tantalum alloy vendors.

 

[BobPE]

yes, anything is possible, but I would be very interested in the hydraulics of the system just to make sure. Can you give us the pressures (suction & discharge) design flow and actual flows, how you operate the pumps, pipe sizes and what valves you have. Just anothert approach...

BobPE

 

[hacksaw]

Alloy 20 is no longer proprietary and there are at least 5-6 variations. Easy to get nailed.

Depending on your temperature 65 ppm chlorides will eat Alloy 20 like candy...although the cast Alloy 20 is better than some of the forged grades. Pumps,valves,pipe, generally are vulnerable...lined pumps are an option so are some of the proprietary grades (Hast. C (actually C276), Carp. 20, etc.). Hopefully you have enough silicate to control the florides.

 

[BRT549]

Your fluoride content is extremely high. Silica amount is also unreal. The chlorides don't sound too bad, 65 ppm means that is not your worst problem.

I've never heard of Fluoride numbers that high in filter acid. Where is your ore coming from? Even the worst Morrocan rock is only 4% F. Is it raw rock which has not been through a flotation process to beneficiate the phosphate up and drop the other impurities down? Having 21% P2O5 in the filter acid is also quite low; we usually get 26-27% in our dihydrate process. What is your free sulfate amount in the reactor? Yes, Alloy 20 might only last a month in your service.

Hacksaw - good point on the F/Si ratio.
To keep the hydrofluoric acid from chewing up your pumps you need at least 4 atoms of silica for each fluorine atom. A more typical fluoride amount in the filter acid is 1 - 2%. You are lucky if only the pumps are getting chewed up - what is the rest of your plant made of?

As far as metallurgy for your pumps, if you truely have those F and SiO2 numbers, you are going make some pump vendor rich. You need lots of chrome and moly. High cost choices would be Inconel 625, Alloy 59, or Hastelloy G-30. Medium cost - Ferralium 255 (UR52N+), 904L, or Cd4MCu would do better than Alloy 20 (we use Cd4MCu). There might be a non-metallic or lined pump out there, but in a phos acid plant the acid is never clean enough to prevent erosion. A pump manufacturer that might be able to help is Wilfley

http://www.wilfley.com/

 

[ghopkins]

I'm not sure about metallics, but we're helping out a phos acid plant right now who uses lots of Teflon-FEP lined fiberglass pipes and tanks. This material seems to be working out exceptionally well for them. Perhaps the wetted surfaces of your pump can be coated with Teflon FEP or PFA.

Greg
ghopkins@rl-industries.com

 

[SASC]

BelZona coating has proven to work exceptionally well in these applications. Contact your area BelZona representative to arrive at the exact BelZona formulation to match your operational parameters. This material is just plain hard to beat and wears like iron, while it adheres better than any material we've ever come across....as good or better than many epoxy coatings.

Regards,
SASC