Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Corrosion in metal pipe carrying DC current?

  1. Feb 23, 2015 #1
    Hey
    Given the simple electric circuit below, the connection between the negative end of the DC source and the load (R1) is a pipe where heavy oil flow with some H2S and CO2. The pipe is made of iron and copper alloy.
    1.PNG
    Without the DC current, H2S and CO2 will corrode the pipe. After introducing the current, will corrosion rate decreases, stay the same or increase? Is the pipe considered under "Cathodic Protection"?

    Thank you,
     
    Last edited: Feb 23, 2015
  2. jcsd
  3. Feb 24, 2015 #2

    CalcNerd

    User Avatar
    Education Advisor
    Gold Member

    Since you provide no real values and your question is relatively straight forward, here is the way to approach this problem.
    You have galametric corrosion, it is stated without DC current, you will corrode the pipe.
    You provide H2s and CO2 as having polluted the oils (you don't provide ion details, so we have to assume the current will be in the correct orientation to the Fe-Cu alloy). If you do have tables and values, you could calculate this, but if it is by trial and error, you would simply adjust the current so as to offset the voltage of the galametric junction to zero. Your understanding of this will provide you with the answer.
     
  4. Feb 24, 2015 #3
    Thank you for the reply.
    I provided no values because I don't have any yet. my question was on the concept, will DC current in the above system make corrosion worse or better?
    Are you saying that knowing the values of H2S and CO2 concentrations we can calculate the required voltage necessary to cancel out the corrosion reaction? What if the voltage was higher than the required voltage? what do you mean by "the current will be in the correct orientation to the Fe-Cu alloy"?
    As you probably noticed from my questions, this is very far away from my field
     
  5. Feb 24, 2015 #4

    CalcNerd

    User Avatar
    Education Advisor
    Gold Member

    Did you look up Galvametric action?

    Any time to two metals touch, they form a dissimilar metal junction ie a voltage across the junction. This voltage can cause corrosion if there are ions present that can attach to the metals. H2S can behave like an acid, eating away at the metals. These materials can be researched and with a good chemistry book you can determine ionic reaction rates. You are not required to do so, as your question doesn't ask, it only asks about corrosion.

    The answer is it will decrease (assuming you did introduce the current in the correct way as my earlier statement implied) and you are providing Cathodic protection.

    Large buildings in metropolitan areas that are close to electric subways certainly DO have provide cathodic protection to their steel structures to avoid corrosion to the steel in the building foundations.
     
  6. Feb 24, 2015 #5
    Please correct me if I am mistaken: Galvanic corrosion occurs when two different metals come in touch with each other, I understood from what you wrote that it also occur in objects made from the same alloy, which (the alloy) made of different metals.

    In this particular case, the H2S will be the "ions that attach to the metals" and thus cause the galvanic corrosion.

    Did I get everything right?

    Thank you very much ..
     
    Last edited: Feb 24, 2015
  7. Feb 25, 2015 #6

    Baluncore

    User Avatar
    Science Advisor

    No, it is not cathodic protection.

    Cathodic protection requires there to be a sacrificial electrode that will be gradually consumed by the reactants flowing in the pipeline. That would be a very expensive way of removing the reactive contaminants in an oil, an oil that will coat surfaces and is probably an insulator.

    The voltage gradient along the pipe resulting from your externally applied current will be insufficient to make a significant difference in the reactions between the pipe and it's contents. Even if it did make a difference by lowering the rate of corrosion at one end of the pipe, it would increase the sacrifice of pipe at the other end.
     
  8. Feb 25, 2015 #7

    Borek

    User Avatar

    Staff: Mentor

    I can be wrong, but my understanding is that what the OP asks for is not different in principle from impressed current cathodic protection (ICCP), which is routinely used for pipelines protection (and doesn't require a sacrificial cathode).
     
  9. Feb 25, 2015 #8

    Baluncore

    User Avatar
    Science Advisor

    My understanding was that the OP was trying to prevent corrosion on the inside of the pipe by the fluid flowing within. The OP shows the pipeline as being a resistive load and a loop current path.

    ICCP is used to protect the outside of structures where there is contact with the ground electrolyte. The current that flows is that needed to oppose corrosion of the protected metal structure. ICCP requires a ground reference anode somewhere.
    The circuit used with ICCP connects to the conductive structure at one point. The other terminal is a ground reference anode. The ground anode used with ICCP may still corrode or become polarised like a charged battery.
     
  10. Feb 25, 2015 #9
    Thank you Borek and Baluncore.
    Yes I am trying to prevent corrosion on the inside of the pipe from the H2S or at least not increase corrosion rate when introducing the current.

    How about the system in the picture below: Oil will flow in pipes from point 1 to point 3. pipe from 1 to 2 is the same (made from normal metal alloy). where the metal of the pipe from 2 to 3 is made from active metal to act as a sacrificing anode (like Zinc). Will this stop/ reduce the corrosion?

    2.jpg

    R1 at the top = R1 in the bottom
     
  11. Feb 25, 2015 #10

    Baluncore

    User Avatar
    Science Advisor

    The conductive metal pipeline will form an effective short circuit. There will be little ground to pipe current. The majority of the voltage will be dropped across the resistors and the wires on the RHS of your diagram.

    If there is any external effect, it will show as increased corrosion of the outside of the pipeline near the electrode connections. For that reason the pipeline will need to be decommissioned or replaced earlier than scheduled.

    I very much doubt there will be any difference to the corrosion on the inside of the pipeline. If there is a change, it will again be differential and so require early intervention.
     
  12. Feb 25, 2015 #11
    if this is the problem, then what if I connected the battery between the metal pipe and zinc pipe. So that pipe 2-3 (zinc) is an anode and 1-2 is a cathode? :)

    4.PNG
     
  13. Feb 25, 2015 #12

    Baluncore

    User Avatar
    Science Advisor

    You will not solve an internal corrosion problem by trial and error on the outside of the pipe.
    If the conductive pipe is ever an anode it will be corroded by the ground.
     
  14. Feb 25, 2015 #13
    the internal surfaces of pipes will be conducting the electricity from the DC power source and thus the internal surface of the pipes will be a space for electrons movement from the power source from anode to cathode. and H2S will be accessing positively changed anode and a cathode, so I think H2S will corrode the internal surface of the zinc pipe (the anode) and the iron alloy pipe will be protected
    am I right?
     
    Last edited: Feb 25, 2015
  15. Feb 25, 2015 #14
    I don't care if the anode pipe (from 2-3) corrodes
     
  16. Feb 25, 2015 #15

    Baluncore

    User Avatar
    Science Advisor

    There will need to be an insulator where the two pipes meet. If the differential voltage at that point is great enough it will cause corrosion of the anode metal.

    If there is no insulator, the voltage gradient will be very small and so there will be very little current flow through the fluid. Corrosion will continue as before.

    Because the pipeline carries a continuous flow of reactants, the reaction of the fluid will require quantities of sacrificial anode proportional to the flow rate and reactant concentration. At the same time there may be deposition on the inside of the cathode pipe that will gradually block the pipe.

    Resolution of problems like this have been achieved in various ways. As I said previously, trial and error is most unlikely to solve the problem, while it will waste much time. If you were to specify the exact problem, then we might be able to give you a solution.
     
  17. Feb 26, 2015 #16
    putting what you said in consideration, protection of the cathode pipe and amount of sacrificing anode will also depend on the DC current provided?

    I don't have any more details, I am asking about the concept only. whether or not I can protect the pipe with the DC current introduced to the pipe or at least maintain the corrosion rate as before introducing the current
     
  18. Feb 26, 2015 #17

    Baluncore

    User Avatar
    Science Advisor

    The current necessary to protect the pipe is actually the current needed to maintain sufficient protection voltage, (a few volts), between the terminal electrodes. That current is dependent on the conductivity of the fluid.
     
  19. Feb 26, 2015 #18
    So this system provides "ICCP" not galvanic protection? what I mean that, going back to my original question, with proper installation and calculations electric current do help reduce corrosion rate?
     
  20. Feb 26, 2015 #19
    finally, if we have water with high electric conductivity flowing with the oil and H2S, will this change anything? I mean, in the insulator between the pipes, the electricity will probably flow in the water fraction of the flow and H2S will be most likely soluble in the oil
     
  21. Feb 26, 2015 #20

    Baluncore

    User Avatar
    Science Advisor

    It attempts to provide galvanic protection. The protection potential may be provided by an external supply, called ICCP, or by an electrochemical cell such as a sacrificial zinc electrode. When you propose to prevent corrosion by using both a sacrificial zinc pipe and an externally provided current, it is still remains galvanic protection. Part of the potential needed to galvanically protect the pipe comes from the sacrificial anode and part from your proposed external power supply. I see no advantage in such a duplicate system. It costs twice as much as one system and has the disadvantages of both systems.


    It is the impressed voltage or potential that prevents corrosion. The current that flows is a function of electrolyte conductivity. A high forward current may be an indicator of excessive protection voltage, that will involve unnecessary sacrifice of the anode material. You have not yet proposed a “proper” or a sensible installation.

    ICCP is used to protect the outside of conductive structures, not the inside. You cannot protect the inside of a pipe by having an electric current flowing along the conductive wall of that metal pipe. To provide the voltage potential necessary to protect the structure, current must flow between the conductive pipe wall and the corrosive electrolyte in contact with the surface.

    To protect the inside of a pipe using ICCP, you would need a coaxial electrode inside that pipe. The internal longitudinal electrode would need insulated supports to keep it on axis. Those multiple supports would obstruct the flow of fluid. It would be very difficult to examine or replace the internal electrode.

    Until you understand the fundamentals of electro-chemistry and of electrical circuits, you stand little chance of implementing a practical and economic protection system.


    Protection would only be on one side of, and in the immediate vicinity of the insulator. There would be no protection elsewhere in the pipe.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Corrosion in metal pipe carrying DC current?
  1. Copper corrosion (Replies: 14)

  2. Corrosion of pipe (Replies: 1)

  3. Corrosion and rust (Replies: 7)

Loading...