304 or 316: The Grade Decision Chloride and Temperature Make on a Seawater Header

Picture a coastal mill on a Friday afternoon: a seawater cooling-header replacement order on the desk, and one field on it still blank — grade. The medium is marine cooling water, the hottest stretch of the loop runs near 40 °C, and the drawing is dense with flanged joints, threaded instrument stubs and full-penetration welds. On the quote sheet, two lines decide the conversation. Type 304 lands at the number the budget expected. Type 316 lands roughly 30 percent above it, and nobody at the table can yet say what that extra money actually buys. The order will not wait for a long study, because the shop wants to cut the first length of pipe next week. What is missing is not more supplier quotes. It is one number from the process side, and the rule for reading it.

From One Seawater Cooling Header to the Question That Settles the Grade

The order that explains the stakes was not a steel mill's. In 2024, a chemical processing facility in Shandong Province installed 304 stainless steel cooling headers on its seawater heat exchange system. The procurement team picked 304 over 316 and booked the difference as a win: roughly 12,000 saved on material cost, delivered on schedule, no engineering change order. The win survived exactly until the pipe walls failed. Chloride in the seawater had been attacking the tube walls and the geometry around them the whole time, and that grade carried no molybdenum to stop it. What the team treated as a purchasing decision turned out to be a service-condition decision wearing a purchasing decision's clothes. That is the pattern a coastal mill inherits every time it replaces a seawater-cooled header, at Oxelösund or anywhere else with salt in the water.

So the question to settle first is not which grade is better. It is narrower and more useful: which operating boundary does this line sit on, and where does that boundary fall relative to the grades on the quote sheet. A boundary is made of three things — the chemistry travelling through the pipe, the temperature at the point where that chemistry is hottest, and the shape of the metal at its most exposed spots. Grades are not ranks on a scale of quality, because 304 is not a lower version of 316 and 316 is not a premium edition of 304. They are two different answers to two different chloride loads, sold at two different prices. Getting that order of thinking right is what keeps a plant from buying corrosion protection it never needed, or skipping the protection it did.

Sorting by unit price fails here for a mechanical reason. For neutral, indoor, chloride-free service, Type 304 is the lowest-cost specification; the moment chloride exceeds about 50 ppm at ambient temperature, or about 25 ppm once service runs above 50 °C, the specification flips to 316 or 316L. Nothing in the invoice signals that flip, because the purchase order is written before the failure and the corrosion claim is written after it. The two documents sit years apart, so the cost that decides the grade arrives years after the decision that needed it. That delay is what makes a price-first habit so durable and so expensive: the feedback lands too late to correct the behavior it should have corrected.

Two Percent Molybdenum Rewrites What a Pipe Wall Can Survive

The mechanism starts with what molybdenum does to a passive film. Stainless steel resists corrosion through a thin chromium-oxide layer that repairs itself; chloride ions attack that layer locally, and if the repair loses the race, a pit opens and keeps going. Molybdenum slows the attack at precisely those weak points. That is why 316 and 316L are the grades specified for marine, chemical and pharmaceutical service, and why 316 tubing appears in specifications such as ASTM A312 TP316/316L rather than as generic “better stainless.” The distinction is not marketing. It is a specific alloying addition solving a specific failure mode that a chloride-free application never triggers — which means the same extra money spent indoors buys nothing at all.

Put the two chemistries side by side and the difference reduces to one line on a mill certificate. Type 316 stainless steel carries 2 to 3 percent molybdenum as an additional element; Type 304 carries essentially none of it. Everything downstream descends from that single addition — the marine rating, the saltwater rating, the high-chloride industrial rating — because molybdenum is what blocks chloride-based corrosion. Type 304 is built for atmospheric conditions and low-level chemical exposure, and it does that job well and cheaply. Type 316 is built for the water that would be lethal to 304. Reading the two as quality levels instead of as two different coverage areas is the misreading that puts 304 on a seawater header and 316 on an indoor tank, wasting money in one direction and pipe in the other.

For engineers who want that chemistry compressed into one number, PREN does the job: pitting resistance equivalent number, roughly 19 for 304, 25 for 316, 24 for 316L. The useful part is not the arithmetic but the line it draws. Above PREN 25 is where a grade starts handling chlorides reliably. That is the threshold a specification should be checked against before price enters the conversation, and it explains why 316L, with a marginally lower PREN than 316, still qualifies for the same family of service. A grade sitting under the line stays under it no matter how carefully the piping is installed, and a grade above it stays viable even when the loop runs hot for most of the year.

The Chloride Threshold Line Decides Where the Header Can Be Installed

The threshold numbers are the part worth memorizing. At 40 °C, Type 304 fails at roughly 300 ppm chloride; Type 316 holds to about 1,000 ppm at the same temperature. Those are not laboratory curiosities but the operating envelope the grade was bought for. The second pair matters just as much: the switch point sits at roughly 50 ppm chloride in ambient service, and once the loop runs above 50 °C the figure drops to about 25 ppm. Notice what the temperature change does to the acceptable concentration — it nearly halves it. A loop that looked safely below the line at commissioning can cross it simply by running warmer, and nothing about the pipe changes at the moment it does.

Then correct the location, because the average concentration of the medium is the wrong number to design against. Crevices fail first: gaskets, threaded connections and weld roots. In those gaps chloride concentrates and oxygen can no longer replenish the passive film, so pitting starts at joints while the bulk pipe still looks sound. That is how the Shandong cooling headers went — seawater duty, 304 in the wall and no molybdenum in the alloy, until the pipe walls suffered complete chloride attack. A specification written around the average chloride reading would have passed that line. A specification written around its gaskets, threads and weld roots would not have.

Service category narrows which band applies. Marine and coastal cooling water sits high on the scale almost by definition, and pharmaceutical and chemical duty sits high for a different reason: the process fluid itself is aggressive, and cleaning regimes reintroduce chlorides on a schedule. Neutral, indoor, chloride-free service sits low, and 304 there is the lowest-cost correct answer rather than a compromise. Between the bands, the deciding evidence is the worst case rather than the typical case — the hottest day, the most concentrated cleaning cycle, the joint that cannot be flushed. Where a line carries both a mild bulk fluid and an aggressive cleaning step, the cleaning step sets the grade, because it happens on purpose, repeatedly, for the life of the plant.

Thirty Percent More on the Invoice, Priced Against a Shutdown

The premium has a number and a payback. As of April 2026 FOB Asia, 316L runs about 28 to 35 percent above 304L, and the older rule of thumb of a 30 percent material premium lands in the same place. On chloride-exposed lines, that gap typically closes in under 5 years, with 3 to 5 years as the usual band, through avoided shutdowns, avoided tube replacement and avoided second construction passes. The arithmetic only works in one direction on a given line. Choosing 316 where 304 would have done the job wastes about 30 percent of the material budget. Choosing 304 where 316 was required buys a failure with a date on it. Both errors are paid in full; only one of them is paid in material.

Supplier pricing puts the same wedge in blunter terms: 316 costs 30 to 40 percent more than 304, and that is the number a procurement team actually sees. It is also the figure the Shandong team was optimizing in 2024, when it chose 304 headers for a seawater exchange loop to save roughly 12,000. The saving was real; so was the chloride. What the ledger never showed was the second entry: wall metal eaten through by chloride, an unscheduled replacement, and the same header ordered again in the grade that should have come first. A 12,000 line item is not a small thing in a project budget. It is simply not the whole cost of the decision.

The same logic travels to other metals, which is why the grade habit is worth building. Sourcing aluminum sheet, a buyer meets 3003, 5052 and 6061: 3003 is the general-purpose alloy with good workability and moderate strength for panels, signs and HVAC work; 5052 pairs excellent corrosion resistance with higher strength and shows up in marine environments and fuel tanks; 6061 is the strong structural alloy. The names change, the structure of the decision does not. Aluminum's baseline properties set the family-level case before alloy selection begins — roughly one-third the density of steel, toughness retained at low temperature rather than turning brittle, conductivity near 62 percent of annealed copper at equal cross-section and 204 percent at equal weight.

Back at the Order: One Rule to Carry Into the Next Specification

Return to the order on the desk. Even after the grade question resolves, the specification is not finished: form and alloy code still have to be written down, the same layer aluminum buyers already know from sheet, coil, plate and treadplate guides and from reading alloy codes rather than brand names. Steel is no different. The grade answers what the metal is made of; the standard and the form answer what is being delivered — seamless or welded, tube or plate, which dimensional range. Fix the grade first, because grade is the only one of those choices the chloride load can decide for you, and it is the one that cannot be corrected after installation.

The rule that comes out of all this fits in a few lines. Take the chloride concentration at the point of the loop that runs hottest, not the figure averaged across the loop. Add the temperature at that same point, because above 50 °C the acceptable concentration falls to roughly half the ambient figure. Then look for crevices — gaskets, threaded stubs, weld roots — and treat them as the critical location, since they fail before the bulk wall does. Compare that worst case against the threshold rather than against the price column, and let the answer name the grade. If chloride approaches 50 ppm at ambient, or 25 ppm above 50 °C, or if crevices are present at all, 316 or 316L is the specification.

What the rule does not do is finish the paperwork. It sorts a service condition into a grade family; it does not confirm the exact grade-level parameters for a specific application, and those belong on official technical documentation rather than in a planning heuristic. Where a line is unusual — mixed fluids, unusual temperatures, an aggressive cleaning cycle — the threshold comparison is a starting filter, not a final answer. The useful outcome is that the first question becomes the right one. Chloride and temperature go onto the specification review ahead of price, and the grade that follows from them is defensible in front of both the process engineer and the budget.

When the quote sheet puts two grades in front of you at a 30 percent spread, the rule is to price the boundary before you price the pipe. Molybdenum sets the threshold; chloride, temperature and crevice geometry decide whether that threshold is the one your line needs; the invoice only tells you what crossing it costs. A grade chosen from the worst point of the loop is an engineering judgment with a number behind it. A grade chosen from the cheapest line of the quote is a bet that the loop stays average for the next five years — and average is not a condition any loop actually holds.