904L (UNS N08904) resists sulfuric acid at 0 to 98% concentration at up to roughly 35 to 40 °C, wet-process phosphoric acid, and reducing organic acids. Its chloride tolerance runs about 2,000 ppm at 25 °C, falling to about 100 ppm at 100 °C. The rate is never one number, because it depends on four conditions as much as on the alloy.
Ask five suppliers for 904L’s corrosion resistance and you’ll get five answers. One puts the maximum service temperature at 400 °C, another says 450 °C. One quotes a pitting temperature of 30 °C, another 60 °C. One tells you 904L covers the full sulfuric acid range; the next tells you it fails in the middle of it.
Every one of those numbers is defensible, because each was measured under a different set of conditions. None arrived with the conditions attached. That is the problem this page fixes.
This is the media data behind our complete 904L stainless steel guide: every rate carries the medium, concentration, temperature and aeration state that produced it, including the duties where 904L is the wrong answer. We supply 904L, and we will tell you when your service is outside it.
Key Takeaways
- 904L corrosion resistance is a property of the system, not the alloy. The rate depends on medium, concentration, temperature, aeration and flow, and a number quoted without those five is not data.
- In pure sulfuric acid, 904L covers 0 to 98% concentration up to roughly 35 to 40 °C. The 40 to 60% band is the first region to fail as temperature or chloride rises, and it defeats 904L as readily as 316L.
- 904L is limited to about 1 to 2% hydrochloric acid, is worse than 304L and 310L in nitric acid, and is not a warm-seawater or FGD grade. 254 SMO and 2507 outperform it above about 40 °C.
- The chloride ceiling falls roughly 20 ppm per °C: about 2,000 ppm at 25 °C, about 100 ppm at 100 °C. Wet-process phosphoric acid is tighter still, about 500 ppm chloride for 904L against about 300 ppm for 316L.
- In NACE MR0175 / ISO 15156 sour service, 904L is limited to 60 °C, either at 1.5 psia H₂S with no chloride restriction or at 50 psia H₂S with chloride capped at 50 mg/L, in the solution-annealed condition.
- Every 904L corrosion resistance rate below carries its condition. Where the literature contradicts itself, we say so and explain why. The 20% sulfuric acid figure for 316L varies by a factor of two between sources because the test conditions differ.
904L Corrosion Resistance at a Glance
The table below is a screening tool for 904L corrosion resistance. Use it to rule your duty in or out before reading further. The rated cells are selection decisions; the numeric cells are measured data with the condition attached. Where a cell carries a rating rather than a rate, no defensible published number exists for that condition.
| Medium | Condition | 904L result | Corrosion rate | If 904L is not enough |
|---|---|---|---|---|
| Sulfuric acid, pure, dilute | 5 to 25% H₂SO₄, up to 40 °C | Usable with margin | <0.1 mm/yr | Not needed |
| Sulfuric acid, pure, mid | 40 to 60% H₂SO₄, 25 to 40 °C | Low temperature only; fails as temperature rises | Rating, not a rate | Alloy 20, C-276 |
| Sulfuric acid, pure, concentrated | 93 to 98% H₂SO₄, 95 to 115 °C | Worse than 304L and 316L | ~0.17 mm/yr | 304L, 310L, Alloy 20 |
| Sulfuric acid, boiling dilute | 10% H₂SO₄, boiling | Fails | ~2.57 mm/yr | Alloy 20, C-276 |
| Sulfuric acid plus chloride | 20% H₂SO₄ with 200 to 2,000 ppm Cl⁻ | Envelope collapses | Rating, not a rate | 254 SMO, C-276 |
| Phosphoric acid, wet process | ~50 °C with impurities | Good | 0.056 mm/yr (316L: 0.88) | 254 SMO |
| Acetic acid, boiling | 20% CH₃COOH | Good, but no better than 316L | 0.02 mm/yr (316L: <0.01) | 316L |
| Formic acid, boiling | 45% HCOOH | Moderate | 0.20 mm/yr (316L: 0.28) | 254 SMO, Alloy 28 |
| Oxalic acid, boiling | 10% (COOH)₂ | Moderate | 0.69 mm/yr (316L: 1.02) | 254 SMO |
| Hydrochloric acid | 1% HCl, boiling | Marginal | 0.55 mm/yr (316L: >5) | C-276 |
| Hydrochloric acid | Above 2% HCl | Not suitable | Rating, not a rate | C-276, C-22 |
| Nitric acid | Any concentration | Worse than molybdenum-free grades | Rating, not a rate | 304L, 310L |
| Seawater, chlorinated | 25 °C | Good | Surficial attack only | Not needed |
| Seawater, chlorinated | 50 °C | Fails | Rating, not a rate | 254 SMO, 2507 |
| FGD condensate | Simulated, pH dependent | Below 254 SMO | Rating, not a rate | 254 SMO, 2507 |
| Tall oil distillation | 235 °C | Good | 0.06 mm/yr (254 SMO: 0.01) | 254 SMO |
| Bleach plant, C-washer liquid | Service condition | Unsuitable | 1.01 mm/yr (254 SMO: 0.03) | 254 SMO, 654 SMO |
| Sour brine plus elemental sulfur | 177 °C | Fails | Rating, not a rate | Annealed C-276 |
One structural number sits behind the table. 904L’s pitting resistance equivalent number works out to about 35.6: high-alloy austenitic territory, but below the ≥40 threshold that NORSOK-style guidance applies to ambient seawater. See our 904L properties page for how that PREN is derived.
Why 904L Corrosion Resistance Is Not a Single Number
Five variables set the rate: medium, concentration, temperature, aeration or redox state, and flow velocity. That is why a single figure cannot be carried from one duty to another. A corrosion rate is a measurement of a system, not a property of a metal.
The same applies to the 400 °C against 450 °C split in the opening. Both trace to the same alloy behaviour seen through different limits: about 400 °C is the practical continuous-service ceiling in oxidising service, 550 °C is where sigma-phase formation starts to matter, and the lower figures engineers actually work to, such as 350 °C for welded ER385 joints, come from fabrication rather than from the base metal. None of those is wrong. Each answers a different question.
Sulfuric and phosphoric acid are reducing at low concentration and become oxidising as their oxidiser content rises. That distinction decides which alloy element helps. 904L’s copper content, 1.0 to 2.0%, is what buys its performance in the reducing regime, and 316L has none. Its molybdenum, 4.0 to 5.0%, buys chloride resistance and actively hurts in strongly oxidising acid.
The aeration effect explains why field experience so often contradicts the datasheet. Dissolved oxygen, ferric ions (Fe³⁺) and cupric ions (Cu²⁺) shift the redox potential positive, and a positive shift can make a lower-alloy grade behave like 904L. In roughly 20% sulfuric acid copper leach liquor, where copper is dissolved in the process stream, 904L corroded at 0.02 mm/yr and Type 316 at 0.03 mm/yr. That difference is inside test scatter.
That single result dismantles the “904L is three times more corrosion resistant than 316” claim on supplier pages. The multiple is real in one medium and roughly onefold in another, depending on whether the system is oxidising or reducing.
Velocity matters too. Near an active-passive boundary, flow can strip the passive film and accelerate attack; in stagnant conditions chlorides concentrate in crevices instead. The number in a pipe is not the number in a tank.
904L in Sulfuric Acid: The Full Concentration Map
904L was developed for sulfuric acid service, and this is where its copper earns its place. It is also where the industry’s summary of 904L is most wrong.
The 0 to 98% Envelope and the Temperature That Sets It
In pure sulfuric acid, 904L can be used across essentially the whole 0 to 98% concentration range at up to roughly 35 to 40 °C. The copper addition is what makes the alloy workable across that span at ambient temperature. Above it, the window narrows quickly and concentration starts to govern.
It isn’t accurate to describe 904L as covering the “20 to 85% gap” that 316L cannot. The mid-concentration region is where the trouble begins, not where 904L is strongest.
Sulfuric acid concentration map (pure acid, no chlorides)
| Concentration band | Temperature limit | 904L rate where published | 316L rate where published | Verdict |
|---|---|---|---|---|
| 0 to 25% | Up to ~40 °C | <0.1 mm/yr | >1.2 mm/yr at 60 °C | 904L, with margin |
| 25 to 40% | Materially below 40 °C | Rating only | Rating only | Marginal; verify by test |
| 40 to 60% | Well below 40 °C | Rating only | Rating only | 904L weak point |
| 60 to 85% | Up to ~35 °C | <0.1 mm/yr | Rating only | 904L, ambient only |
| 85 to 98% | Up to ~35 °C at ambient; fails hot | ~0.17 mm/yr at 95 to 115 °C | Better than 904L when hot | 904L at ambient; 304L or Alloy 20 when hot |
The Reducing Band Where 904L Also Fails
The 40 to 60% concentration region is heavily reducing. In the words of the Australian Stainless Steel Development Association, whose technical FAQ 8 documents the boundary, that band “defeats even the high nickel 904L and 254/654 grades.” This is the honest limit, and it is the one the search results do not state. Within that band the usable temperature falls well below 40 °C, and the equipment the industry actually builds, evaporators and reactors that run hot, is where 904L gets chosen wrongly.
The concentrated end is a second trap. In 93 to 98% H₂SO₄ at 95 to 115 °C, 904L corrodes at roughly 0.17 mm/yr, and about 0.19 mm/yr in 99% acid at 100 to 120 °C. That is worse than 304L or 316L, because at high concentration and temperature the acid is strongly oxidising and molybdenum stops helping. The better answer there is a molybdenum-free grade or Alloy 20.
Boiling dilute acid is a straight failure case: roughly 2.57 mm/yr in boiling 10% H₂SO₄. Any boiling dilute sulfuric duty beyond trace scale sits outside 904L.
Where 904L fails in sulfuric acid
| Condition | 904L rate | Comparison | Better choice |
|---|---|---|---|
| 10% H₂SO₄, boiling | ~2.57 mm/yr | 304L and 316L also fail | Alloy 20, C-276 |
| 93 to 98% H₂SO₄, 95 to 115 °C | ~0.17 mm/yr | 304L and 316L lower | 304L, 310L, Alloy 20 |
| 99% H₂SO₄, 100 to 120 °C | ~0.19 mm/yr | 304L and 316L lower | 304L, 310L, Alloy 20 |
| 40 to 60% H₂SO₄, above ~40 °C | Rate not defensibly published | Defeats 254 and 654 SMO too | Alloy 20, C-276, or non-metallic |
Chloride Contamination Collapses the Envelope
Between 200 and 2,000 ppm chloride in sulfuric acid dramatically reduces 904L’s resistance and extends the failed zone, with the 40 to 60% band affected first. Several grades remain potentially suitable below about 20% sulfuric acid even with significant chloride present. Above that concentration, chlorides move the selection decision.
The practical consequence is that the chloride content of the process stream, not just the acid strength, decides whether 904L is adequate. Ask for both figures in the enquiry.
When the Acid Is Oxidising, 316L Catches Up
Restating the leach-liquor result where it belongs: 904L at 0.02 mm/yr against Type 316 at 0.03 mm/yr in roughly 20% sulfuric acid copper leach liquor, with aeration, Fe³⁺ and Cu²⁺ keeping the surface passive. The Nickel Institute’s corrosion data for sulfuric acid and related compounds carry both the isocorrosion curves and that leach field data.
That result also resolves a contradiction in the published literature. The familiar 316L figure for 20% H₂SO₄ appears as more than 1.2 mm/yr at 60 °C, and as 0.50 to 0.65 mm/yr at 80 °C. A higher rate at a lower temperature can’t hold for a single test condition; the probable cause is a difference in aeration, flow or chloride between the tests.
So quote the number with its condition, or don’t quote it at all. For where 316L’s envelope actually ends, see our 904L vs 316L comparison.
904L in Phosphoric Acid: Wet-Process Service
The Wet-Process Rate Column
904L performs well across the phosphoric acid range, including wet-process acid carrying chloride, fluoride, sulfate and organic impurities. The copper addition forms a protective copper-phosphate surface layer, a mechanism 316L lacks. This is the clearest case on the page where the copper content does the work.
Phosphoric acid rate comparison at approximately 50 °C
| Grade | Corrosion rate |
|---|---|
| 316L | 0.88 mm/yr |
| 317LMN | 0.29 mm/yr |
| 904L | 0.056 mm/yr |
| 254 SMO | 0.01 mm/yr |
For contrast, in boiling 20% H₃PO₄ the pair runs 316 at about 0.18 mm/yr and 904L at about 0.01 mm/yr.
That is a tenfold-plus improvement over 316L at about two to two and a half times the material cost. This is where 904L is genuinely the right answer, stated plainly so the warnings elsewhere on this page are not read as a blanket caution.
The Chloride and Fluoride Ceiling
In wet-process acid, 316L is reported suitable only below roughly 300 ppm chloride. 904L extends that ceiling to roughly 500 ppm. Fluoride is the second impurity to watch, since it attacks the passive film directly and its effect is not captured by a chloride figure alone.
The design implication is straightforward: the chloride analysis of the acid, not its strength, usually selects between 316L, 904L and a 6Mo grade. On the sulfuric axis the paired grade decision, 904L against Alloy 20, is covered in our 904L vs Alloy 20 for sulfuric acid service comparison.
904L in Organic Acids: Acetic, Formic and Oxalic
The boiling-solution data is published widely, so this section’s value is the honest reading of it rather than the numbers themselves.
Boiling organic-acid corrosion rates
| Acid | 904L | 316 / 316L | Read |
|---|---|---|---|
| 20% acetic acid | 0.02 mm/yr | <0.01 mm/yr | 316L is equal or better |
| 45% formic acid | 0.20 mm/yr | 0.28 mm/yr | 904L modestly ahead |
| 10% oxalic acid | 0.69 mm/yr | 1.02 mm/yr | 904L ahead, both high |
In boiling 20% acetic acid, 316L is as good as or better than 904L. Acetic acid isn’t a reason to buy 904L, and a supplier who cites it as a 904L differentiator is selling rather than advising. The genuine organic-acid advantages for 904L sit in formic and oxalic acid, which are reducing acids, and in acid mixtures.
The mechanism explains the split. Acetic acid is a weak acid whose performance is dominated by water content: at high water content 904L is excellent, while in glacial-grade acid, or where chlorides are present, corrosion rises markedly. Formic and oxalic acids are reducing, and their aggressiveness falls when contaminants such as air, Fe³⁺, Cu²⁺ or peroxides shift the redox potential oxidising.
The Nickel Institute publishes a 0.1 mm/y isocorrosion diagram for formic acid covering 304L, 316L, 904L, Alloy 28, 2507 and C-276. Read the boundary from that diagram rather than quoting a single rate. Acetic and formic acid plants are a real 904L application, but the reason to specify it is chloride-laden or reducing duty, not the acetic acid itself.
The same nuance applies to tall oil distillation at 235 °C, where Outokumpu’s published rate data puts 904L at 0.06 mm/yr against 0.88 mm/yr for 316L and 0.01 mm/yr for 254 SMO. That’s a strong second, not the leader.
Where 904L Fails: Hydrochloric and Nitric Acid
Hydrochloric Acid: 1 to 2% and Nothing More
904L is limited to roughly 1 to 2% hydrochloric acid concentration. Within that window, it is far better than conventional stainless steel. In boiling 1% HCl, 904L corrodes at about 0.55 mm/yr against more than 5 mm/yr for Type 316.
The temperature sensitivity is the more useful comparison. 904L exceeds 5 mpy in 1% HCl only at about 50 °C, against about 30 °C for 316L. That 20 °C of extra headroom is the entire HCl case for 904L.
Above 2%, no stainless steel is the answer. One documented case makes the escalation concrete: a 904L hydrochloric acid vapour condenser perforated after 14 months and was replaced with C-276, which ran past twelve years below 0.1 mm/yr. Wall thickness would not have fixed that; the grade was wrong.
Hydrochloric acid limit and escalation path
| Concentration | 904L verdict | Escalation grade |
|---|---|---|
| Trace to 1% | Usable, with temperature limit | C-276 above ~50 °C |
| 1 to 2% | Marginal, low temperature only | C-276, C-22 |
| Above 2% | Not suitable | C-276, C-22 |
| HCl vapour or condensate | Not suitable | C-276 |
Nitric Acid: 904L Is Worse Than the Cheap Grade
904L performs worse than molybdenum-free grades such as 304L and 310L in nitric acid. Nitric acid is strongly oxidising, and molybdenum actively harms performance in that regime. This is the cleanest demonstration of the article’s central thesis: adding alloy content can make the material worse.
Treat it as a specification warning, because it is a real mis-specification route. Buyers upgrade from 304L to 904L assuming more alloy is always better. In nitric acid it is not, and the upgrade buys a shorter life at higher cost. For the equipment these media belong to, see our 904L applications by industry.
Chlorides: Pitting, Crevice Corrosion, Seawater and FGD
This is the second load-bearing section, and it’s the one that most often changes a specification.
CPT and CCT Are Different Numbers
The critical pitting temperature (CPT) and the critical crevice temperature (CCT) answer different questions. CPT is measured on an open, polished surface with no artificial crevice. CCT is measured under a crevice former, and it is always lower for the same alloy and environment.
Those two facts explain why search results quote 904L critical temperatures anywhere from 12 to 60 °C. CCT also depends on the crevice geometry while CPT does not: a crevice has a longer diffusion path than a pit, so crevice corrosion initiates at a lower temperature, and the measured CCT moves with the crevice former used.
| Measurement | Test solution | 904L | 316L | Standard |
|---|---|---|---|---|
| CPT | 10% FeCl₃ | ~42 °C | Not comparable | ASTM G48 Method A |
| CPT | 3.5% NaCl | ~50 °C | ~15 °C | Potentiostatic |
| CPT | 1 M NaCl | 56 to 60 °C | Not published | Potentiostatic |
| CCT | 10% FeCl₃, PTFE washer | ~12 °C | Not published | ASTM G48 Method B |
| CCT | Distributor data | ~18 °C | Not published | Not stated |
The practical rule for a gasketed, flanged or fouled system is that the governing number is the CCT, not the CPT. Select on the lower of the two. A datasheet that gives you one bare figure has told you half a story, and if it does not name the ASTM G48 method, it has told you nothing you can design against.
The Chloride Ceiling Falls About 20 ppm per °C
904L tolerates roughly 2,000 ppm chloride at 25 °C, falling to about 100 ppm at 100 °C. The slope is the important part: a limit that falls about 20 ppm per °C is a design constraint, not a rule of thumb. A 60 °C service means roughly 1,100 ppm, not 2,000.
Chloride tolerance in acid and aqueous service
| Medium | 316L ceiling | 904L ceiling | Next grade up |
|---|---|---|---|
| Sulfuric acid | Envelope collapses rather than a ppm ceiling | Same, plus 200 to 2,000 ppm extends the failed zone | 254 SMO, C-276 |
| Wet-process phosphoric acid | ~300 ppm | ~500 ppm | 254 SMO |
| Seawater, ambient | PREN ~24, well below threshold | PREN ~35.6, below the ≥40 threshold | 254 SMO, 2507, 654 SMO |
NORSOK-style guidance requires PREN ≥40 for ambient seawater service, and 904L’s 35.6 does not meet it. 904L is a chloride-tolerant acid grade, not a seawater grade. That is the boundary the whole 904L library draws, and the grade decision on the chloride axis belongs in 904L vs 254 SMO.
Seawater and Desalination: Good at 25 °C, Failed at 50 °C
In chlorinated and unchlorinated Arabian Gulf seawater, 904L and 2205 duplex both resisted well at 25 °C and both failed at 50 °C, while higher-alloy grades held at every condition tested. The crevice data agrees: 904L showed no crevice corrosion at 25 °C within 180 days but corroded at 50 °C, and in a separate assessment recorded 19 attacks out of 40 sites.
The application consequence is that 904L belongs in the less aggressive, lower-temperature zones of a desalination or seawater-cooling system: pre-treatment piping, chemical dosing lines, and acid-contaminated rather than clean seawater. The equipment map is a separate subject and sits on our 904L applications page rather than here.
FGD Condensate and Bleach Plants: Where 904L Loses
904L is routinely sold for flue gas desulfurization. The published evidence says rank it third. This is the strongest differentiator on the page and the least comfortable to write.
Peer-reviewed testing in simulated FGD condensate found that as molybdenum content rises, the passive region widens and the pitting potential shifts more positive. The ranking is 254 SMO ahead of 904L ahead of 316L, with 254 SMO assessed as the most suitable for FGD service. That work appeared in the Journal of the Chinese Society for Corrosion and Protection.
In the green-death solution at three temperatures, the picture is sharper. Type 316 showed severe pitting at all three. 904L showed no obvious pitting at 20 °C, typical pitting at 40 °C, and severe pitting damage at 70 °C, while 254 SMO and 2507 showed none at any of the three temperatures. In a 70 °C crevice test, 904L recorded roughly 40 crevice sites against about 13 for 254 SMO.
Field evidence agrees. In bleach-plant C-washer liquid, 904L corroded at 1.01 mm/yr against 254 SMO’s 0.03, and 904L nozzles have been observed pitted where an S31254 (254 SMO) replacement was unaffected. The overall bleach-plant ranking runs 654 SMO, then 254 SMO, then 904L and 2205 roughly level, then 316L.
Chloride-heavy service ranking
| Environment | Best | Second | 904L position |
|---|---|---|---|
| FGD condensate, simulated | 254 SMO | 904L | Second, ahead of 316L |
| Green-death solution, 70 °C | 254 SMO, 2507 | Not comparable | Severe pitting |
| FGD crevice test, 60 °C, pH 3.5 | 2507 | 254 SMO | Third, ahead of 2205 |
| Bleach plant, C-washer liquid | 654 SMO | 254 SMO | Unsuitable |
| Chlorinated Gulf seawater, 25 °C | 904L and 2205 hold | n/a | Acceptable |
| Chlorinated Gulf seawater, 50 °C | Higher-alloy grades | n/a | Fails |
| Tall oil distillation, 235 °C | 254 SMO (0.01 mm/yr) | 904L (0.06 mm/yr) | Good, second |
Picture a materials engineer at a coal-fired plant replacing a set of absorber spray nozzles. The supplier recommends 904L, the price is accepted, and the nozzles go in. Two years later they show pitting on the upstream face while the adjacent 254 SMO components are unaffected.
That sequence is drawn from documented field experience rather than a named customer, and the published evidence predicted it. In 50 to 70 °C chloride-laden condensate, the escalation path is a 6Mo grade or super duplex, not a thicker wall.
The direct advice: for FGD absorber internals, quench sections and outlet ducts, specify a 6Mo grade or super duplex rather than 904L. Use 904L for the cooler, less chloride-loaded zones, where its cost and its weldability earn their place.
The chloride escalation ladder, drawn from the data above
- Under 40 °C, low chloride: 316L where it fits, 904L where the acid requires it.
- 40 to 60 °C with chloride present: 904L for acid-dominated duty, 254 SMO where chloride dominates.
- 60 to 80 °C with chlorides: super duplex 2507 or 254 SMO.
- Above 80 °C, or chloride plus an oxidiser: 654 SMO or a nickel alloy such as C-276.
Chloride Stress Corrosion Cracking
904L’s roughly 25% nickel gives it substantially better chloride SCC resistance than 316 or 317. That’s resistance, not immunity. Severe environments, notably boiling magnesium chloride testing, can still crack it, and cold-worked material carries higher risk because residual stress does the rest of the work.
For 904L corrosion resistance in chloride service, the regulator-published band is the framing that helps a designer: about 80 to 120 °C for external marine atmospheres, but only 15 to 30 °C for internal seawater, widening further in sour service. Operator-specified limits vary, so quote the band, name the variability, and test the specific duty. Where the risk is real, solution-treat after cold work and minimise residual tensile stress.
Sour Service: 904L Under NACE MR0175 / ISO 15156
NACE MR0175 / ISO 15156 governs carbon and low-alloy steels plus corrosion-resistant alloys in sour oilfield service. NACE MR0103 / ISO 17945 governs refining. 904L is accepted under the first in the solution-heat-treated condition, and it is not hardenable by heat treatment, so a cold-worked or partially annealed delivery is a compliance failure rather than a performance preference.
904L sour-service limits under NACE MR0175 / ISO 15156
| H₂S partial pressure | Maximum temperature | Chloride limit | Elemental sulfur |
|---|---|---|---|
| 1.5 psia (10 kPa) | 60 °C | No restriction | Not permitted |
| 50 psia (345 kPa) | 60 °C | 50 mg/L | Not permitted |
904L falls in the highly alloyed austenitic category, defined by Ni% plus twice Mo% greater than 30 with Mo at 2% or more. The general austenitic hardness cap of 22 HRC applies, and the solution heat-treated condition is required rather than optional.
The failure case gives the table teeth. In deaerated 25% NaCl brine with 1 g/L elemental sulfur at pH 4.5 under 0.7 MPa H₂S, 904L failed at 177 °C within months, alongside Inconel 625, Incoloy 825, Alloy 28 and duplex grades. Only annealed C-276 lasted twelve months, and it held to 232 °C.
In elemental-sulfur sour service, 904L is a 60 °C material, and the answer above that ceiling is a nickel alloy.
Verifying 904L Corrosion Resistance: Tests and Certification
The corrosion claim only matters if the delivered material can produce it. For 904L, the two elements that decide acid performance are exactly the two a substitution most obviously degrades.
The Two Elements to Shoot With PMI
Request positive material identification against copper (1.0 to 2.0%) and molybdenum (4.0 to 5.0%) specifically, not just a UNS N08904 confirmation. A heat that lands low on copper will underperform in reducing acid, precisely where 904L was bought to perform; a heat low on molybdenum will underperform in chlorides. Both are visible on the incoming PMI shot in seconds.
The Delivery Condition Is Part of the Corrosion Specification
Specify solution-annealed and water-quenched. Cold-worked or partially annealed material carries residual stress, which raises both SCC and intergranular risk regardless of how good the chemistry is. A perfect MTR and a wrong delivery condition is a common and avoidable failure. Avoid the 800 to 900 °C range, where intermetallic phases can form within minutes.
Ask for the Right Test, and Name the Method
For intergranular corrosion, ASTM A262 Practice E (the Strauss test) is the practical acceptance test for 904L; Practice C (Huey) also appears. For any pitting or crevice figure, state the ASTM G48 method, because Method A and Method B use different specimens and return different numbers. An unqualified “CPT” on a certificate is not comparable to a design value.
Documentation
- EN 10204 3.1 for routine supply, listing chemistry, heat treatment condition, mechanical results and the test standards actually applied.
- EN 10204 3.2 where a project, EPC contractor or third-party inspector requires independent witness.
- Third-party inspection through SGS, BV or TÜV. It adds cost, but it’s often the only route to satisfying an owner’s specification.
- A NACE MR0175 / ISO 15156 compliance statement where sour service applies. It must accompany the heat treatment condition, not replace it.
The Specification in One Line
Send the medium, the concentration, the temperature, the chloride level and the aeration state, and our metallurgical team will confirm whether 904L holds or whether the duty needs a different grade. Send your operating conditions
904L Corrosion Resistance FAQ
Is 904L resistant to sulfuric acid?
Yes, across 0 to 98% concentration in pure acid at up to roughly 40 °C. The 40 to 60% band is the first region to fail as temperature or chloride rises. Boiling dilute acid, at about 2.57 mm/yr in 10%, and hot concentrated acid, at about 0.17 mm/yr in 93 to 98% at 95 to 115 °C, both sit outside its range.
What concentration of sulfuric acid can 904L handle?
Essentially any concentration in pure acid up to about 40 °C. The practical limit is temperature first, then chloride. Above 2,000 ppm chloride, the envelope closes in, and the 40 to 60% band is the worst region of the map.
Is 904L resistant to hydrochloric acid?
Only to about 1 to 2%. In boiling 1% HCl it runs about 0.55 mm/yr, compared with 316L’s more than 5 mm/yr. Above 2%, the correct material is a nickel alloy such as C-276, and no increase in wall thickness changes that.
What is 904L’s maximum chloride level?
About 2,000 ppm at 25 °C, falling to roughly 100 ppm at 100 °C, which is a slope of about 20 ppm per °C. In wet-process phosphoric acid, the practical ceiling is about 500 ppm, against about 300 ppm for 316L.
What is 904L’s critical pitting temperature?
About 42 °C in 10% ferric chloride, about 50 °C in 3.5% NaCl, and 56 to 60 °C in 1 M NaCl. The corresponding critical crevice temperature is far lower, about 12 °C in 10% FeCl₃. For a gasketed or fouled system, the crevice figure governs the selection.
Is 904L suitable for FGD scrubbers?
For cooler, low-chloride zones, yes. For absorber internals, quench sections and outlet ducts, no. Peer-reviewed FGD condensate data ranks 254 SMO ahead of 904L, and in 70 °C green-death solution 904L pits severely while 254 SMO and 2507 do not.
Is 904L NACE MR0175 compliant?
Yes, in the solution heat-treated condition, limited to 60 °C: at 1.5 psia H₂S with no chloride restriction, or at 50 psia H₂S with chloride capped at 50 mg/L. Elemental sulfur is not permitted in either case.
Conclusion
Five conclusions carry the decision. 904L’s rate is a property of the system, meaning medium, concentration, temperature, aeration, and flow, and no number is useful without them. The usable envelope is pure sulfuric acid from 0 to 98% up to about 40 °C, wet-process phosphoric acid, and the reducing organic acids. The 40 to 60% sulfuric band and any boiling dilute acid sit outside it, and nitric acid is worse in 904L than in 304L.
It is a chloride-tolerant acid grade rather than a seawater or FGD grade, and 254 SMO and 2507 take over above about 40 °C. In sour service, it is a 60 °C material.
The reason this page exists is that the industry quotes 904L’s corrosion resistance without its conditions, and a rate without a condition isn’t data. We publish the conditions, including where the answer is another grade.
Request 904L plate, pipe or bar with full MTR, PMI verification on the copper and molybdenum lines, and the solution-annealed condition documented, and send your operating conditions with the enquiry. Quotation within 24 hours. For the full grade profile, see our 904L stainless steel guide.
