Your first annual survey is two days behind you, and it went fine.
Twenty-some test stations along a coated line, rectifier cycling on an interrupter — three seconds on, one second off — so every station gives you a pair of numbers instead of one. On-potential around −1.28 volts with the current flowing; instant-off around −0.98 volts in the gap. Station after station, the pair holds, and your notebook sits right on top of four years of records from the tech who retired last spring. Nothing to flag.
Then the phone rings. The project manager has a crew coming in to tie new piping in at the compressor station where your rectifier sits, and they want the unit de-energized while they work. Standard request, and an easy yes. You drive out and shut it down.
Your meter is already on the seat, so you pull a reading at the test station just outside the fence before you leave. It comes back −0.91 volts.
That is not the on-potential you wrote down two days ago, and it is not the instant-off either. It is past the instant-off — seventy millivolts less negative than the number you recorded in the interrupter gap two days ago, at this same station. Nothing about the pipe has changed except that the current stopped. You check the leads and the cell. The number is real.
You mention it to a tech with twenty years on you, and he does not sound surprised. It’s depolarizing. Go read it again tomorrow.
So you do. The next day it reads −0.79 volts. Two days after that, −0.68. On day five, with the construction crew packing up, it reads −0.63 volts and it has very nearly stopped moving.
Nothing broke. Nothing leaked. You just watched a number you thought was a property of the pipe turn out to be a property of the current.
Two things left, at two different speeds
Line the numbers up and something splits cleanly in half.
The station, from survey day to day fiveOn-potential (current flowing) ………. −1.28 VInstant-off (1-second gap) ………… −0.98 VTwo hours after shutdown …………. −0.91 VOne day …………………………. −0.79 VThree days ………………………. −0.68 VFive days ………………………… −0.63 VThree hundred millivolts disappeared in one second. Three hundred fifty more took five days.
The first three hundred millivolts — the gap between −1.28 and −0.98 — vanished inside a single one-second window of the interrupter cycle. It was gone before your meter could finish settling. That amount never had anything to do with the condition of the steel.
The other three hundred fifty millivolts — from −0.98 all the way back to −0.63 — took five days to bleed off, and it did not bleed off evenly. It came off fast at first and slower and slower after that: about thirty-five millivolts an hour over the first two hours, five and a half an hour through the rest of the first day, then two, then one. It behaved like something draining, not something switching.
Two different quantities were riding on that reading. One was electrical and instant. One was chemical and slow. Telling them apart is the whole job of this module — and it is the difference between a number that describes your pipe and a number that describes your wiring.
What polarization actually is
Put a piece of bare steel in soil and walk away. It settles at a potential of its own — somewhere around −0.55 to −0.65 volts vs. copper sulfate for ordinary carbon steel, depending on the soil. Nobody is doing anything to it. That is the metal at rest, and the trade calls that its native or free-corroding potential. It is the number your station drifted back toward over five days.
Now push current onto it. The potential moves.
That movement is polarization. It is not a substance and not a setting. It is a displacement — how far you have shoved a metal’s potential away from where it wanted to sit, by making current flow through its surface.
The formal definition. Peabody’s Control of Pipeline Corrosion defines polarization as “the deviation (change) in potential of an electrode as a result of the passage of current.” That sentence packs a great deal into very few words, and it is a hard one to absorb before you have watched it happen. After five days of watching a station drift from −0.98 back to −0.63 volts, it is just a careful way of saying what you already saw: current moved the number, and when the current stopped, the number went home.
Two things follow immediately, and both matter in the field. Polarization costs current: the farther you want to move a structure’s potential, the more current you have to push into it. And polarization takes time: the surface has to change chemically before the potential settles, which is why your station needed five days to let go of it and why a freshly energized system does not hit its final number the same afternoon.
That second point is the one that catches new techs. A rectifier is not a dimmer switch on a light. It is closer to a thermostat — you turn it up, and the room takes its time.
How long that takes depends on what you are protecting. A large or poorly coated structure — a lot of bare steel asking for current — can take days or even weeks to reach its final number. A small, well-coated one can settle far faster. “Give it time” is a habit, not a fixed number of hours.
The picture the whole trade argues from
There is one drawing that sits underneath every polarization conversation in this industry, and it is worth carrying in your head even at this level.
Before you look at any version of this drawing, know one thing: the potential axis runs with positive at the top and negative at the bottom. That is upside down from the survey plots you are used to, where more negative usually runs up the page. Every Evans diagram you will ever meet is drawn this way, so it is worth getting straight once. Same numbers, opposite direction — and mixing the two up is the fastest way to read the picture backwards.
Picture a graph. Potential runs up and down the side, more negative toward the bottom. Along the bottom runs current density — how much current is crossing each square foot of steel — but plotted on a compressed scale, where each step to the right is not one more unit but ten times more current.
Two lines get drawn on it. One line describes the metal dissolving: steel giving up electrons and going into the soil. That is the corrosion side. The other describes the reaction that consumes those electrons at the surface — in buried pipe, most often oxygen being reduced. The two lines slope toward each other and cross.
Where they cross is where an unprotected pipe actually lives. That crossing point names the potential you would measure with a meter and no current applied, and it names how fast the steel is being eaten — the corrosion current on that axis is the corrosion rate, just expressed as current. Both facts, one intersection.
Cathodic protection works by refusing to let the metal sit at that crossing point. You supply electrons from somewhere else — an anode bed, a rectifier — and drag the potential down below it. And because the bottom axis is a ten-times-per-step scale, dragging the potential down a modest distance does not trim the corrosion current a little. It collapses it.
The compressed axis is the entire reason cathodic protection is practical. If corrosion rate fell off in proportion to potential, protecting a pipeline would take impossible current. It falls off by factors of ten instead, which is why a few hundred millivolts and a modest rectifier can shut down a line that would otherwise rust through.
You will never be asked to construct that graph on a job. You need to know it exists, which way the lines run, and why the compressed axis makes small potential shifts pay off large. The deck and the article in your downloads carry the drawing itself.
Why the number is a hundred millivolts
Somewhere in your first year, somebody will tell you that a hundred millivolts of polarization shift demonstrates protection. It gets quoted like a speed limit — a number handed down, obeyed, and rarely explained. It is not arbitrary. It falls straight out of that graph.
For carbon steel in ordinary soil, the corrosion side of that drawing has a measured steepness: the potential has to move roughly a hundred millivolts for the corrosion rate to change by a factor of ten. That figure is not something you calculate on a job. It is a property of steel, measured in laboratories, and it is stable enough that the industry built a criterion on it.
Which gives you the one piece of arithmetic worth carrying out of this module.
Polarization shift, converted to a corrosion-rate cutdecades = shift (mV) ÷ 100 mV per decadecorrosion rate is cut by 10 ^ decadesSteel moves about one decade — one factor of ten — per hundred millivolts of shift.
A hundred millivolts is one decade. One decade is a factor of ten. So the criterion is not saying “a hundred millivolts is a safe number.” It is saying: move the steel a hundred millivolts and you have cut what it loses to about a tenth of what it was losing. That is the practical line between a pipe actively corroding and a pipe essentially not.
Worked example — the station you just watched depolarize
Setup. With the rectifier on the interrupter, the station read −0.98 volts instant-off. After five days with the current off, it settled at −0.63 volts. How much polarization was the system actually buying, and what was it worth?
Step 1 — the shiftshift = 0.98 V − 0.63 V = 0.35 V = 350 mV
Step 2 — convert to decadesdecades = 350 mV ÷ 100 mV per decade = 3.5
Step 3 — convert to a rate cut10 ^ 3.5 = about 3,000Roughly three thousand times less metal loss than the bare, unprotected condition.
Answer. Three hundred fifty millivolts of real polarization — three and a half decades, about a three-thousand-fold cut in corrosion rate. That system was not scraping past the criterion. It was three and a half times past it, and the five-day drift you stumbled into was the proof.
Now run the same three steps on a station that is telling you something much less comfortable.
Worked example — the station that only just clears
Setup. A different line, a station near the far end of the protected length. Native potential measured at −0.61 volts. Polarized potential, current interrupted, −0.73 volts.
Step 1 — the shiftshift = 0.73 V − 0.61 V = 0.12 V = 120 mV
Step 2 — convert to decadesdecades = 120 mV ÷ 100 mV per decade = 1.2
Step 3 — convert to a rate cut10 ^ 1.2 = about 16Sixteen times less metal loss — real protection, but a thinner margin.
Answer. A hundred twenty millivolts clears the mark, and on paper that station passes. But sixteen times is not three thousand times — two stations, the same criterion, and almost two hundred times the difference in margin. A number that passes and a number that is comfortable are not the same number.
The criterion is a floor, not a target. Knowing the arithmetic behind it is what lets you tell the difference between a station that is protected and a station that is one dry summer away from not being.
Three kinds that are real, and one that is not
Polarization does not come from one mechanism. Three different things slow the reactions at a steel surface, and a fourth quantity gets called polarization out of habit even though it is nothing of the kind.
Activation polarization
The reaction itself is sluggish. Electrons crossing between the metal and the things in the soil have to get over a hump, and pushing that reaction faster costs potential. This is the kind that dominates when there is plenty of reactant available and the current is modest — a wet, active soil, a system running at ordinary output.
Concentration polarization
The reaction runs out of something to react with. In buried pipe the usual bottleneck is oxygen: the surface consumes it faster than fresh oxygen can work its way in through the soil. The reaction is no longer limited by the steel at all — it is limited by delivery.
This one has a signature you can spot from the truck. Turn a rectifier up and the potentials move. Turn it up again and they move less. Turn it up a third time and they barely move at all, while your anode bed works harder and your power bill climbs. Nothing is broken. You have simply reached the point where the surface is already pulling oxygen in as fast as the soil can supply it, and more current buys almost nothing.
When more output stops producing more shift, the honest first question is not “what broke?” It is “have I hit the delivery limit?” Techs have chased shorted casings and bad connections for days over a system that was simply out of oxygen to reduce.
Environmental polarization
The current changes the chemistry right at the surface, and the changed chemistry helps. Cathodic reactions leave the film against the steel more alkaline than the bulk soil around it, and that alkaline film is a less hospitable place for corrosion to proceed. The protection you paid for with current partly maintains itself through the environment it creates — which is also why a system that has been running for months holds its numbers better than one energized last week.
Resistance drop — the one that is not polarization
And then there is the three hundred millivolts that disappeared in one second.
When current flows from an anode bed, through soil, onto your pipe, that soil has resistance. Current through resistance produces a voltage, and your reference cell sits out in that soil where it reads the voltage along with everything else. Every on-potential you take with the current flowing carries that amount baked in.
It is not polarization. Nothing about the steel has changed. It is a measurement artifact, and it disappears the instant the current does — which is exactly why interrupters exist, and why the pair of numbers you take at a station is worth more than either number alone.
Resistance drop is not real polarization, but it looks exactly like it on a meter. That is the trap. A confident on-potential can make an under-protected pipe look comfortably protected, and the pipe will not argue with you.
What moves these numbers between one survey and the next
The same station, the same rectifier setting, does not always give you the same answer — and most of the time nothing is wrong. Four conditions move polarization behavior, and all four move with the weather.
- Oxygen. The master depolarizer. More oxygen at the steel means the cathodic reaction runs harder and your current demand goes up. Dry, sandy, well-aerated ground is thirstier than saturated clay for exactly this reason.
- Temperature. Warm speeds reactions up. A line that holds its numbers in February can slip in August without a single component changing.
- Moisture and movement. Water carries reactants in and sweeps products away, which erases concentration polarization and raises demand. A line under a creek crossing does not behave like the same line under a hilltop.
- Soil chemistry. Acidity changes which reaction dominates at the surface, and different reactions polarize differently.
When a station that read comfortably last fall reads marginal in July, the useful reaction is not to go hunting for a fault. It is to ask what changed in the ground — drier, warmer, better-aerated soil raises the current it takes to hold the same shift.
There is a ceiling on the other end too. This set already named it: push a structure far past what protection requires and you start trading one problem for another — coating disbondment and hydrogen effects at the steel surface. More is not automatically better. The goal is enough polarization, held steadily, not the most a rectifier can produce.
Back at the test station
Everything above collapses into a few habits at the box.
A single potential with no note about whether current was flowing is half a measurement. The pair is what carries meaning — the on-potential with the resistance drop riding along, the instant-off with it stripped away. Record which is which every time; six months from now neither you nor the person auditing the file can reconstruct it.
A potential that moves after the current stops is not a fault, it is polarization doing what it does, and the speed of that movement is information. Fast decay in the first hours, then slower and slower, is the ordinary shape; something that decays to native almost immediately never held much polarization to begin with. And a large or poorly coated system just energized has not finished — reading it the same afternoon and calling it underperforming is a mistake made every season.
Day five, and the crew is packing up
The construction is finished and the rectifier goes back on. If you have been paying attention to the last five days, you already know what will not happen: the station will not snap back to −0.98 volts the moment you close the breaker.
The resistance drop returns instantly — that part is electrical, and it never left the steel to begin with. But the polarization has to be rebuilt, and it rebuilds the same way it drained, only backwards. Fast at first, then slower, over hours and days, until the surface chemistry settles where it was before. Read that station on the way out and you will get an honest number for that afternoon and a misleading one for the line. Read it next week and you will have your survey back.
Which is the whole idea. A potential is not a fixed property of a pipeline. It is where the steel currently sits, given the current currently flowing, in the ground as it currently is — and every one of those can change.
EC-014 closes the Electrochemistry & the Galvanic Series set. Seven modules ago this started with atoms and bonding, and ran through the corrosion cell, oxidation and reduction, the galvanic series, the arithmetic of metal loss, and the reference cell that measures all of it. This module was the last piece: what those potentials do once current is moving, and why the numbers the trade agreed on are the numbers they are. Pass the knowledge check and the certificate for this set is yours.
From here the work turns outward — from what corrosion is to the forms it takes in the ground and the surveys that find it. You will be reading potentials for the rest of your career. You now know what they are telling you.
Key takeaways
- Polarization is potential moving in response to current — how far a metal has been shoved off the potential it would sit at on its own. It costs current to create and it takes time to build or drain.
- Two things ride on an on-potential — real polarization, which decays over hours and days, and resistance drop, which vanishes the instant the current stops. Only one of them describes your steel.
- Steel moves about one decade per hundred millivolts — shift the potential 100 mV and the corrosion rate falls by roughly a factor of ten. That is where the hundred-millivolt criterion comes from; it is a floor, not a target.
- Do the arithmetic: decades = shift ÷ 100 mV, rate cut = 10 ^ decades. A 350 mV shift is 3.5 decades, about 3,000 times less loss. A 120 mV shift clears the same criterion at about 16 times — both pass, one has far more margin.
- Three real kinds, one impostor — activation (sluggish reaction), concentration (reactant delivery runs out), and environmental (the current improves the surface chemistry). Resistance drop is a measurement artifact, not polarization.
- When more current stops buying more shift, suspect the delivery limit before you suspect a fault — the surface may already be consuming oxygen as fast as the soil can supply it.
- Oxygen, temperature, moisture, and soil chemistry move current demand season to season. A station that reads marginal in July and comfortable in February usually has weather behind it, not a fault — and over-polarizing past what protection requires trades corrosion for coating and hydrogen problems.
References & further reading
- Evan’s Diagram — Field Notes from RCS article on the potential-versus-current picture, the kinds of polarization, and how cathodic protection shifts the operating point.
- Examples of Depolarizers — Field Notes from RCS article on oxygen and the other agents that keep cathodic reactions running.
- Influencing Factors: Temperature, Oxygen & Relative Movement — Field Notes from RCS article on the environmental conditions that change current demand.
- Peabody’s Control of Pipeline Corrosion — long-standing field reference on polarization behavior and cathodic protection criteria.
- Corrosion Basics: An Introduction — foundational text on the electrochemistry and kinetics behind polarization.
- NACE SP0169 — Control of External Corrosion on Underground or Submerged Metallic Piping Systems; the source of the polarization-shift criterion.
- NACE TM0497 — Measurement Techniques Related to Criteria for Cathodic Protection; on-potential versus instant-off measurement practice.
