I am specifying pipes for a water system with 500 ppm chloride in solution. Ordinarily we would specify 316L stainless steel to limit the corrosion, but I was asked if we could instead specify carbon steel pipe with a corrosion allowance. So instead of the typical schedule 40 pipe, could I realistically use schedule 80 or 120 etc?

My gut reaction is that you will never really even have a "protective" oxide layer with water continually flowing, so you will just continually corrode the pipes out. All that will be gained with thicker pipe is that it will last longer before failure, but you will have increased corrosion products to remove from the flow stream over time. I could not really find any good resources talking about chloride effects on carbon steel pipe since the general consensus is that stainless should be used for this application.


Corrosion can usually, but not always, be modeled as a 0-th order reaction which means linear with time. ASME B31 piping code treads lightly into this, mainly discussing factors to be used when specifying a corrosion allowance, and no more. However, we really just need a rate of reaction and some basic principals, and we can derive whatever we want! Chemistry Stack Exchange and a California Science Fair Project are used here. We basically see that the reaction for rusting is: $4Fe +3O_2 +6H_2O \rightarrow 4Fe(OH)_3$, with an overabundance of iron and water in a pipe, the limiting factor is oxygen, and how conductive the solution is.

While oxygen is the limiting factor, concentration of dissolved oxygen can heavily impact corrosion because it would be cubed in the reaction rate equation, if not for the fact that the entire inside surface of the pipe is covered in iron, which typically dominates the reactions. The only reason the reaction doesn't occur faster is because there is such a small concentration of dissolved oxygen, and the multiple reaction steps.

The science test showed that about 0.001" of steel was lost every year for salt water solutions (1 mpy). Running your own tests with the water you will be treating would be the only way to know for certain. But, as a basis for design, it can be seen that for a 50 year lifetime the pipe would need to be $50 * 0.001{in} = 0.05{in}$ thicker for a corrosion allowance. Such a large corrosion allowance is typically practical, but in the same article water with process contaminated was found to be 10 times more aggressive. Hence why 316 SST is typically used until the veracity of the attack can be determined.

  • $\begingroup$ Thank you for the very thorough explanation. Part of the basis of my question was that all of the resources I have found go into detail on what SS alloy to use in the presence of chlorides, but I can't really find anything on the effects on carbon steel. Is the point that carbon steel is prone to rusting and mostly unaffected by the chlorides, but IF you choose to use stainless, you will be introducing the potential vulnerability to chlorides? $\endgroup$ – Secundus Jun 29 '17 at 0:12
  • $\begingroup$ The main point is that chlorides and other ions raise the conductivity of the solution. The oxygen needs electrons to change to oxygen ions. The iron gives electrons to make ferrous and ferric ions. With pure water, the water has so little conductivity that the iron can't exchange the electrons. The chlorine helps the path come together and enhances the conductivity. Up to a point. Too much chlorine and the reaction favors the formation of impermeable ferrous chloride, halting the reaction. Chemistry is weird. $\endgroup$ – Mark Jun 29 '17 at 0:18
  • $\begingroup$ As to the second point in the comments - yes, the choice of stainless is about the contaminants you're fighting. Once you chose stainless, some alloys are vulnerable as they begin to dissolve away in the presence of chlorides, or lose their protective coatings they form. Fluorides and acids are especially bad. There almost isn't a steel that can handle 50% sulfuric, but nearly any plastic can! Yet above 85% sulfuric almost every steel is perfectly safe, yet the plastics dissolve. To reiterate, chemistry is weird. $\endgroup$ – Mark Jun 29 '17 at 0:22
  • $\begingroup$ Thanks! Yeah, when I got to college chemistry I promptly found myself in over my head, so while I SHOULD know plenty about corrosion, I avoided it like the plague after I graduated and now here I am. I like it. I think I will hang up a sign at work: "Chemistry is Weird. ~ Mark" $\endgroup$ – Secundus Jun 29 '17 at 0:23
  • $\begingroup$ Your best advice - go to your piping supplier. I make pipe, and know what attacks it and what doesn't. If not, second best advice for specifying is to state they need to demonstrate to you they have experience in this and to specify their corrosion rates and the expected lifetime. $\endgroup$ – Mark Jun 29 '17 at 0:27

Oxygen and pH will generally be the most important corrosion factors in aqueous corrosion. If this is a potable water as described ,there is no reason to consider anything but carbon or galvanized steel. If the Cl is present as HCl you have a problem that 316 won't solve ; If you you have temperatures above 150F ,you could have stress corrosion cracking of 316 ( although unlikely below 200 F ). If you have experience with similar conditions , that would be the most useful. Otherwise figure on 5 mpy ( 0.005" ) as a worst case.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.