Corrosion Chemistry in Pipeline Fluids

February on a ridgetop production pad in the Appalachian Basin. You’re out for a routine sample run, and the line on the work order is a wet-gas flowline — wellhead to separator, maybe a half mile of buried steel. The paperwork tells you what it carries: gas comes off the wellhead at about 95 °F, but the ground it’s buried in is sitting around 45 °F this time of year. The gas is about 2% CO₂ with 30 ppm H₂S, the stream is cutting about 5% water, and the flow is slow and stratified — gas riding on top, liquid laying along the bottom of the pipe.

Here’s the question that matters, and it’s the question this whole module teaches you to answer: is this line corroding right now?

You can’t look. That’s the defining fact of internal-corrosion work. There’s no walking this line and spotting the problem from above the ground — the damage, if it’s happening, is on the inside wall of a buried pipe that’s in service and staying in service. Nobody is cutting it open to check. The only witnesses you get are the fluids: what the line carries, what drops out of it, what comes off a pig run, what collects at the drips. If you can read the chemistry, the fluids will tell you what the steel is going through.

The module before this one gave you the landscape — what internal corrosion is, the four big drivers by name, where it shows up, who does what. This module is where you learn to read. Not mechanism by mechanism — that’s the rest of this set. This is the chemistry underneath all of them: what has to be true inside a line for any internal corrosion to run, and what makes it run faster. By the end, you’ll walk back to that flowline and answer the question.

The diagnostic frame for the rest of your career

The four ingredients

Internal corrosion needs four things in the same place at the same time. Every mechanism in this set — sweet, sour, oxygen, microbial, all of them — is some version of these four coming together:

• Water on the pipe wall. Not water in the line — water touching the steel. Liquid water is where all of this chemistry lives. No water on the wall, and the rest of the list doesn’t matter.
• Something aggressive dissolved in it. CO₂, H₂S, oxygen, or the acids and sulfides that bacteria make. On its own, pure water barely touches carbon steel. It’s what the water is carrying that does the work.
• The steel surface. The pipe itself — the iron that’s available to give up.
• Conditions that let them meet and stay met. Temperature in the working range. Flow slow enough (or geometry low enough) that water holds against the wall. Time. A low spot, a dead leg, a cool stretch where water drops out and sits.

The power of this list is what happens when something is missing. Take away any one ingredient and significant internal corrosion essentially stops. A genuinely dry sweet-gas transmission line can run for decades with minimal internal damage — same steel, same CO₂ in the gas, no liquid water on the wall, no problem. Let water into that same line and it can light up. That’s why the first question an IC tech asks about any segment — before the lab work, before the coupons, before anything — is the simplest one: does water reach this pipe wall, and where?

Keep the flowline from the hook in your pocket — it scores four-for-four on this list, and at the end of the module we’ll run the full check on it with everything the sections between now and then add to your read.

The one reaction under all of it

When the four ingredients do come together, what actually happens to the steel is the same event everywhere, in every mechanism this set covers. An iron atom at the surface lets go and leaves the wall as a dissolved ion:

You don’t need to derive that reaction, balance it, or do bookkeeping with it. You need the picture it paints: your pipe wall is dissolving into the water, one iron atom at a time, and the speed of it depends on what the water hands back. Aggressive water keeps the reaction fed and running. Mild water starves it down to nearly nothing.

The iron doesn’t vanish, either — it goes somewhere, and where it goes is something you can see and sample. Dissolved iron reacts with what’s around it: meet sulfide and it becomes black iron-sulfide scale, meet oxygen and it becomes the familiar orange-brown rust and tubercles, meet carbonate and it can lay down a gray mineral scale. The gunk in a pig receiver is, in part, where the iron went. When a water sample comes back with dissolved iron in it, that iron came off a pipe wall somewhere upstream. Iron counts, scale types, deposit colors — they’re all the same reaction, read from its leftovers.

So when the modules ahead each teach a mechanism, understand what they’re really teaching: not new reactions, but new ways the same reaction gets fed. That’s the thread that makes this set one story instead of six.

What makes pipeline water aggressive

If water on the wall is the necessary condition, the obvious next question is: what makes one water dangerous and another nearly harmless? A handful of levers, and operating conditions pull all of them:

• pH. Carbon steel in water protects itself with thin films of oxide and carbonate — as long as the water lets them survive. Above a pH of about 6, those films stay reasonably stable and slow everything down. Below about 6, they start dissolving and the bare steel underneath goes back to work. Below about 5, attack runs fast. Most of the mechanisms in this set are, at bottom, ways of pushing pH down — or stripping the films some other way.
• Chlorides. Salt. Once chlorides climb past a few hundred mg/L, protective films get less protective and damage stops being even — it concentrates into pits. Produced waters in our part of the world can carry chlorides into the thousands and tens of thousands of mg/L, so this lever is pulled hard and often.
• Dissolved gases. CO₂, H₂S, and O₂ — each one a different chemistry, all three feeding the same iron-dissolution reaction. The next section gives each its survey-level introduction.
• Temperature. An amplifier on almost everything else. Big enough to get its own section below.
• Flow. Fast flow can sweep water and solids through before they settle — or, fast enough, strip protective films right off the wall. Slow or stopped flow lets water lay against the steel and lets deposits and biofilms take up residence. Flow doesn’t corrode anything by itself; it decides where and how long the other ingredients get to work.

A tech who knows these levers can look at a water analysis and the line’s operating data and make a fair first guess at how aggressive the environment is — before any direct measurement goes in. That’s not a parlor trick; it’s how monitoring locations get chosen.

The four aggressive solutes, by name

The module before this one named four big chemical drivers. Now you’ve got the frame to see what each one actually does to the water — surveyed here, taught deep in the modules ahead.

CO₂ — the sweet one

Carbon dioxide dissolves into water and a slice of it becomes carbonic acid:

“Weak” is doing some heavy lifting in that sentence. Carbonic acid is mild compared to anything in a lab cabinet, but a pipeline doesn’t get one dose of it — it gets a fresh supply made on the spot, around the clock, for years. It pushes pH down, dissolves the protective films, and eats steel in patterns that can be startlingly aggressive. The more CO₂ pressure in the gas, the more acid in the water. This is sweet corrosion, and it’s the bread-and-butter internal threat in upstream gas work. The next module takes it end to end.

H₂S — the sour one

Hydrogen sulfide also dissolves into water and also makes a weak acid — so it attacks steel along the same general lines, and it leaves a calling card: black iron-sulfide scale and that rotten-egg smell. But H₂S brings a second problem the others don’t. It lets atomic hydrogen slip into the steel itself, which can crack and blister pipe from the inside of the metal — damage that isn’t metal loss at all, and that doesn’t wait politely for walls to thin. One sentence is all that gets here, because it’s a serious enough subject to own a large piece of its own module later in this set. For now: sour service means acid attack plus a hydrogen problem, and it changes how careful everybody has to be.

O₂ — the one that doesn’t belong

Formation fluids come up from the reservoir with essentially no oxygen in them — the downhole world is oxygen-dead, and the production system is designed to keep it that way. So unlike CO₂ and H₂S, which arrive with the gas, oxygen is an intruder. It sneaks in through tank vents and breathing roofs, worn pump packing, leaks on the suction side, open lines after maintenance, and with injected fluids that sat in open tanks.

It doesn’t take much. Oxygen at parts-per-billion levels measurably speeds up attack, and at parts-per-million it’s a serious problem — it is hungry for the electrons iron gives up, so wherever it’s present, the iron-dissolution reaction stays fed and runs hard. Worse, oxygen rarely spreads itself evenly: where it reaches some patches of steel and not others, the starved spots take concentrated attack. And it amplifies nearly every other mechanism on this list. Orange-brown rust and tubercles in a system that should be oxygen-free mean one thing: air is getting in somewhere, and finding the door is the job. Its own module comes later in this set.

Bacteria — the chemistry factories

The fourth driver isn’t a chemical — it’s alive, and it makes chemicals. Pipeline systems carry whole communities of microbes, and a few kinds matter to us. Acid-producing bacteria generate organic acids and drive pH down right where they live. Sulfate-reducing bacteria manufacture H₂S in place — a line with no sour gas in it can still develop sour chemistry at the wall, courtesy of its residents. Slime-formers build biofilms: sticky roofs over the steel that trap water and solids underneath, keep treatment chemicals out, and turn a patch of pipe wall into a private little environment with its own chemistry. Microbiologically influenced corrosion — MIC — gets its own module in this set, and the monitoring methods for it run deep in the track beyond.

Temperature is an amplifier, not a hazard gauge

It’s tempting to file temperature as “hotter = worse.” The field is messier and more interesting than that. The honest rule: ask what the temperature is doing to the chemistry the steel sees.

• Reaction speed. As a working rule of thumb, chemical reaction rates roughly double for every 10 °C (about 18 °F) of temperature rise. Warmer water does the same chemistry, faster.
• Microbes have a comfort zone. Most of the bacteria we care about do their best work roughly between 20 and 40 °C (about 70–105 °F) — which is, inconveniently, the operating range of an awful lot of gathering systems.
• Scales flip sides. Some corrosion-product films, like iron carbonate, actually get denser and more protective at higher temperatures — hot sections of a sweet system sometimes fare better than warm ones. Other films destabilize as temperature climbs. Same lever, opposite directions, depending on the chemistry.
• Gradients make water. Where a warm wet gas meets a cool wall — a winter-buried flowline, an above-ground bend in the wind, the first hundred feet downstream of a pressure cut — water condenses out of the gas right onto the steel. Temperature difference puts water on walls that the bulk numbers say should stay dry.

That last one is the quiet giant in our part of the world. The hook line’s 95 °F gas against a 45 °F winter wall is a condensation machine — and what condenses is not ordinary water, as the next two sections explain.

Where the water comes from — and why it matters which

Not all pipeline water is the same water, and knowing the source tells you a lot about the chemistry before the lab report ever comes back:

• Produced water came up out of the formation with the hydrocarbons. It’s salty — often heavily — mineral-loaded, oxygen-dead, and it frequently carries dissolved minerals that buffer it against acid. High chlorides, but some built-in armor too.
• Condensed water dropped out of the gas onto a cool wall. It’s essentially distilled: fresh, clean, no salts — and no armor whatsoever. We’ll see in a moment why that makes it more dangerous than it sounds, not less.
• Injection and disposal water is whatever the operation is moving — often high-chloride brine, sometimes carrying bacteria and whatever it picked up in tanks on the way.
• Hydrotest residual is the water left behind after a pressure test — the line gets purged with air or displaced with product when it’s loaded, but the purge never gets it all, and what’s left settles into low spots, sometimes for a long time.
• Atmospheric ingress rides in with breathing tanks and opened lines — and brings oxygen with it.

A single system can hold several of these at once. The hook flowline carries produced water in the stream and manufactures condensed water against its cool wall — two different chemistries, two different threats, same half mile of pipe.

The carbonate buffer — the most under-read line on the lab report

Here’s the piece that explains one of the oldest mysteries in this work: two lines, same gas, same pressure — one corrodes, one doesn’t. The answer is very often not in the gas at all. It’s in the water’s alkalinity — its stock of dissolved bicarbonate and carbonate minerals.

Think of alkalinity as the water’s shock absorber against acid. When CO₂ makes carbonic acid in a water that’s rich in bicarbonate, the bicarbonate soaks the acid up before the pH can fall far. The attack still costs something, but the water pushes back. A typical produced water, loaded with formation minerals, often has that cushion built in — it can carry a gas’s worth of CO₂ at a pH that stays civil.

Now run the same gas over condensed water — the fresh, distilled film on that cool pipe wall. No minerals, no bicarbonate, no cushion. The same CO₂ drives the pH of that thin film down hard and fast, and the wall under it pays full price. Fresh water sounds harmless and is anything but: condensate, frac flowback, leftover hydrotest water — the “clean” waters are the unbuffered ones. Same gas, very different result, and the difference is the water’s chemistry, not the gas’s.

On a lab report, alkalinity is one line, usually reported as mg/L as CaCO₃ — and it might be the most-skipped line on the page. Stop skipping it. High alkalinity (many hundreds of mg/L) means the water has armor. Low alkalinity (tens of mg/L) means every bit of acid the gas makes lands directly on the pH — and on the steel.

The water you sample is not the water on the wall

One more idea completes the chemistry frame, and the rest of the track leans on it constantly. Everything you measure from a sample point — pH, chlorides, iron, bacteria counts — describes the bulk fluid: the average of what’s flowing by. But corrosion doesn’t happen in the average. It happens at the wall, and the wall can be running a completely different environment:

• Under a deposit — sand, scale, sludge laying at the 6 o’clock — chlorides can concentrate to several times the bulk reading (in bad cases ten-fold), oxygen runs out, and the trapped pocket of water turns into its own little acid bath that no sample tap will ever see.
• Inside a biofilm, acid-producing bacteria can hold the local pH two to three units below the bulk water flowing past it. Your sample says 6.5 and looks fine; the steel under the slime is living at 4.
• Under old mill scale or pigging debris, the wall has its own microclimate that the bulk numbers know nothing about.

This is why a clean-looking water analysis doesn’t close the case, and why deposits and solids deserve as much respect as the water itself. It’s also, looking one module-set ahead, exactly why direct measurement tools live on the wall — they sit where the corrosion actually happens, not where the sampling is convenient.

Reading a sample, not just recording it

Put the whole frame to work. Everything this module has built — the levers, the solutes, the buffer, the bulk-versus-wall caution — comes together the moment a lab report hits your inbox. The difference between a tech who records data and a tech who reads it looks like this:

From a reading to a plan — how the data flows

So a sample told a story. What happens next? This is the data-flow chain the whole IC track is organized around, and it’s worth meeting now so every later module slots into a place you already know:

1. Indirect data. Everything in this module so far — fluid, gas, and solids samples from pig runs, drips, and sample points; field tests for pH, bacteria, dissolved gases; lab confirmation. Indirect data tells you internal corrosion is possible, and roughly how bad the environment is. It never proves metal is being lost at a rate that matters.
2. The repeatability gate. One aggressive sample is a note in the file — sampling has too many ways to lie for one result to move money. Two consecutive positives, telling the same story, is a signal worth acting on. That’s the gate between “keep watching” and “go measure for real.”
3. Direct data. Hardware on the wall. Corrosion coupons are the workhorse — weighed pieces of steel installed through fittings into the live line, left to soak, retrieved under pressure, and sent to the lab to answer the real questions: how fast is metal leaving (corrosion rate), and is it pitting (pitting rate)? Probes and in-line inspection add to the picture. The methodology — placement, exposure, retrieval discipline — fills a whole group of modules later in the track.
4. Mitigation, designed upstream. When direct data says the line has a real problem, the response — inhibitor program, biocide treatment, pigging cadence, drying the gas, operational changes — is designed by an integrity engineer or corrosion specialist. That’s not a courtesy; it’s how the field is structured. You bring the data and the read; they own the design.
5. Execution and follow-up. The tech runs the program in the field — and keeps sampling, because the follow-up data is what says whether the mitigation is working. The chain loops back to step one and never really ends.

Notice where this module lives: it’s the literacy that makes step one worth anything. A tech who can read chemistry collects indirect data that means something, places coupons where they’ll actually catch the problem, and walks into the conversation with the integrity engineer carrying a story instead of a stack of numbers.

Back to the flowline

February, a ridgetop pad in the Appalachian Basin, half a mile of buried steel between a wellhead and a separator. Is it corroding right now?

Now you can actually work the question. Four ingredients: water on the wall — yes, a 5% cut laying in the bottom of a stratified line, plus condensation wherever that 95 °F gas touches the 45 °F wall, and that condensed film is the unbuffered kind that takes CO₂ acid at full strength. Aggressive solutes — yes: 2% CO₂ making carbonic acid around the clock, 30 ppm H₂S adding sour chemistry and its hydrogen baggage, and slow stratified flow giving bacteria every opportunity to set up housekeeping at the 6 o’clock. Steel — yes. Conditions — yes, in every direction you look.

So: possible — emphatically. Likely enough to deserve real attention. And you know what attention looks like now, because it isn’t panic and it isn’t a dig: it’s indirect data. Water samples at the separator, read the whole report — pH, iron, chlorides, alkalinity. Gas analysis for the acid-gas numbers. Bacteria cultures from the water and from anything solid that comes out of the line. Field-test what changes fast, before the chemistry in the bottle drifts. Then do it again next month — because one read is a note, and two are a signal.

That’s the answer this module promised. Not “yes it’s corroding” — nobody can say that from the surface. The honest answer: everything this line needs to corrode is present, here’s where the attack would live, and here’s exactly what I’d collect to find out. A tech who can say that, and back it, is doing internal-corrosion work.