Rouge is a reddish, orange, or black iron-oxide film that forms on the interior surfaces of stainless steel process and high-purity water systems. It is not general corrosion of the base metal. It is a discolouring deposit of iron oxide — principally hematite (Fe2O3) or magnetite (Fe3O4) — that either migrates from elsewhere in the loop or grows in place where the chromium oxide passive layer has been depleted. Rouge is classified in three types, Class I, Class II and Class III, and each type requires a different derouging chemistry to remove.
What is rouge and why does it form on stainless steel?
Austenitic stainless steels such as 304 and 316L resist corrosion because chromium at the surface reacts with oxygen to form a thin, self-repairing chromium oxide film only one to five nanometres thick. That film is what makes the steel “passive.” Rouge appears when that film is locally depleted, damaged, or overwhelmed, and iron rather than chromium becomes the dominant species at the surface.
Two distinct mechanisms produce it. In the first, iron released somewhere else in the system — a pump seal, a poorly passivated weld, a carbon-steel tool mark, or the feedwater itself — travels through the loop and deposits downstream. That is migratory rouge. In the second, the passive layer breaks down in place and iron oxide grows directly out of the parent metal. That is in-situ rouge, and it is considerably harder to remove because the deposit is chemically bonded to the substrate rather than sitting on top of it.
Rouge matters because regulators and auditors treat visible discolouration in a product-contact system as a control failure until proven otherwise. Under 21 CFR 211.67, equipment must be cleaned and maintained at intervals that prevent contamination that would alter the safety, identity, strength, quality or purity of the drug product. A rouged distribution loop invites the question of whether particulate iron is reaching product.
What causes rouging in a high-purity system?
Rouge is almost always the visible symptom of a specific, identifiable condition. In our experience the causes fall into five categories:
- Temperature. Hot systems rouge faster. Water-for-injection loops held above 65 °C and clean steam systems operating above 250 °F are the most commonly affected, which is why Class III magnetite is a steam-system phenomenon.
- Chlorides. Chloride attacks the passive layer directly. Feedwater chlorides above roughly 50 ppb in a purified water system, or above 5 ppm in WFI service, materially accelerate breakdown. Cleaning agents containing more than about 100 ppm chloride should never be used on product-contact stainless.
- Dissolved oxygen. Above approximately 20 ppb, dissolved oxygen accelerates oxide growth at hot surfaces.
- Incomplete or absent passivation. A system commissioned without a documented passivation to ASTM A967 starts life with free iron and weld heat tint still on the surface. Rouge then appears within months rather than years.
- Fabrication defects. Dead legs longer than six pipe diameters, velocities below 3 ft/s, crevices at poorly fitted welds, and surface finishes rougher than about 0.5 µm Ra all give iron oxide somewhere to nucleate and stay.
Rouge Class I vs Class II vs Class III — how do they differ?
The three-class scheme is the practical basis for choosing a derouging chemistry. Misidentifying the class is the most common reason a derouging attempt fails: Class I chemistry applied to a Class III deposit will not touch it.
| Property | Class I | Class II | Class III |
|---|---|---|---|
| Oxide present | Hematite, Fe2O3 | Mixed iron oxides | Magnetite, Fe3O4 |
| Origin | Migratory — deposited from elsewhere | In situ — grows from the parent metal | In situ, high-temperature |
| Typical color | Orange to light red | Red to red-brown | Purple-black to blue-black |
| Wipe test | Wipes off onto a clean cloth | Does not wipe off | Does not wipe off; tightly adherent |
| Typical thickness | Surface film | 0.1–0.5 mm | Up to 1 mm |
| Surface Cr:Fe ratio | >0.5 | 0.1–0.5 | Very low |
| Where it is found | Ambient and hot water loops, downstream of a source | Purified water and WFI systems | Clean steam and pure steam systems |
| Removal difficulty | Low | Moderate | High |
Particle size across all three classes typically falls between 0.1 and 10 µm, which is why rouge is often first detected as a rising particulate count or as staining on a filter housing rather than by visual inspection of the pipe interior.
How do you identify and characterize rouge?
Visual inspection tells you that something is present. It does not tell you which class, and therefore does not tell you what chemistry to use. Analytical characterization is what makes a derouging plan defensible to an auditor.
| Method | What it establishes | Approximate detection limit |
|---|---|---|
| Visual inspection and borescope | Presence, distribution, color, extent | Qualitative |
| Wipe or swab test | Migratory (Class I) vs adherent (Class II/III) | 1–10 µg per swab |
| ICP-MS of loop water | Dissolved and particulate iron concentration | 0.1 ppb |
| SEM with EDX | Elemental composition of the deposit | 0.1–1.0 wt% |
| X-ray diffraction (XRD) | Which iron oxide phase is present | 1–5 wt% |
| XPS or AES | Surface Cr:Fe ratio, passive layer condition | Surface-sensitive, nm scale |
| Electrochemical testing (ASTM G61) | Pitting and crevice susceptibility | Comparative |
For a system already in GMP service, the sequence we recommend is a borescope survey to map extent, a wipe test at several points to separate migratory from in-situ deposit, and ICP-MS on loop water to establish a baseline iron figure that the derouging can later be measured against.
How is rouge removed? Derouging chemistry by class
Derouging is a controlled chemical cleaning operation, not a rinse. The chemistry, concentration, temperature, contact time and flow velocity are all specified in advance, verified during execution, and recorded in the turnover package.
| Class | Chemistry | Concentration | Temperature | Contact time | Velocity |
|---|---|---|---|---|---|
| Class I | Citric acid | 0.5–2% | 50–60 °C | 2–4 hours | >5 ft/s |
| Class II | Citric acid with EDTA chelation | 4–6% | 60–70 °C | 4–8 hours | >5 ft/s |
| Class III | Phosphoric acid based | 10–15% | 70–80 °C | 12–24 hours | >5 ft/s |
Solution pH is held between 1.5 and 3.0 throughout. Going below pH 1.5 does not remove rouge faster; it risks attacking the base metal and destroying the very surface you are trying to restore. Turbulent flow is essential — a chemistry circulated below about 3 ft/s will clean the bottom of a horizontal run and leave the crown untouched.
Derouging must always be followed by re-passivation. The acid that strips iron oxide also strips the chromium oxide film. A system that is derouged and not re-passivated to ASTM A967 will re-rouge faster than it did originally, because it has been returned to service with a bare, iron-rich surface.
What is the difference between passivation and derouging?
They are different operations with different objectives, and they are routinely confused in specifications.
Passivation removes free iron from a clean surface and encourages the chromium oxide layer to reform. It is governed by ASTM A967 and removes essentially no base metal. It is what you do to a new system before it enters service, and to any system after a chemical clean.
Derouging removes an established iron-oxide deposit that has already formed. It is more aggressive, runs longer, and is followed by passivation. Cleaning and descaling of the surface prior to either operation falls under ASTM A380.
The practical rule: passivation is preventive and restorative; derouging is corrective. A specification that calls for “passivation” on a visibly rouged system is under-scoped and will not achieve the intended result.
Can citric acid remove existing rouge?
Only partially, and only Class I. Citric acid at the concentrations used for passivation — typically 4 to 10 percent — is an effective chelator of free iron but is not sufficient on its own to lift an established, adherent Class II or Class III deposit. Removing those requires either citric acid enhanced with EDTA or a stronger phosphoric-acid-based formulation, run hotter and longer.
This is the single most common misconception we encounter. A plant runs a standard citric passivation on a rouged loop, sees a modest improvement in appearance, and concludes the system is clean. The Cr:Fe ratio tells a different story.
Can rouge be prevented? Design and operating limits
Largely, yes — and prevention is far cheaper than remediation. The limits below are the ones that actually govern whether a system rouges.
| Parameter | Target | Why it matters |
|---|---|---|
| Conductivity, WFI | <1.3 µS/cm | USP <645> limit; ionic load drives attack |
| Conductivity, purified water | <5 µS/cm | Same mechanism, less stringent service |
| Total organic carbon | <500 ppb | USP <643>; organics support biofilm, which shelters corrosion |
| Chlorides, WFI | <5 ppm | Direct attack on the passive layer |
| Chlorides, purified water | <50 ppb | As above |
| Dissolved oxygen | <20 ppb | Accelerates oxide growth at temperature |
| Feedwater iron | <0.3 ppm | Primary source of migratory Class I rouge |
| Operating temperature | <65 °C where process allows | Rouge rate rises sharply with temperature |
| Loop velocity | >3 ft/s | Keeps particulate in suspension; prevents settling |
| Dead leg length | L/D <6, ASME BPE prefers <2 | Stagnant volume is where rouge starts |
| Surface finish | Ra <0.5 µm, electropolished | Smoother surface gives oxide less to key into |
| Material | 316L with 2–3% Mo, SF4 finish | Molybdenum improves chloride resistance |
Is rouge a regulatory or compliance problem?
It becomes one quickly. Rouge itself is not named in the regulations, but the conditions it creates are. Under 21 CFR 211.67 equipment must be maintained so that it does not alter product quality. Visible discolouration in a product-contact surface is the kind of observation that appears on a Form 483 and then requires a documented impact assessment: is particulate iron reaching product, and what is the elemental impurity position under USP <232> and USP <233>?
The defensible position is a documented rouge monitoring program with defined action levels, characterization data, and a derouging and re-passivation history. That is a far stronger answer to an investigator than an assertion that the discolouration is cosmetic.
How often should systems be inspected for rouge?
A reasonable baseline program, adjusted to the criticality and temperature of the system:
- Quarterly — visual and borescope inspection at defined sample points, particularly tank interiors, heat exchanger surfaces and the return leg of the distribution loop.
- Monthly — dissolved and particulate iron by ICP-MS on loop water, trended rather than read in isolation. A rising trend matters more than any single result.
- Annually — a full internal inspection with surface analysis at representative locations, tied to the system periodic review.
Hot WFI loops and clean steam systems warrant the tighter end of these intervals. Ambient purified water loops can often be extended, provided the trend data supports it.
Does electropolishing prevent rouge?
It delays it substantially; it does not make a system immune. Electropolishing removes a thin layer of the surface, preferentially dissolving iron and leaving a chromium-enriched, smoother finish — typically taking Ra below 0.5 µm and raising the Cr:Fe ratio before the system ever sees water. Rouge needs somewhere to nucleate, and an electropolished surface offers far fewer sites.
But electropolishing does not change the water chemistry, the operating temperature, or the dead legs left by a fabricator. A well electropolished system running hot on high-chloride feedwater will still rouge.
Can heat tint cause rouge formation?
Yes, and it is one of the most preventable causes. Weld heat tint is a chromium-depleted oxide scale formed when the weld zone is exposed to oxygen at temperature. The straw and blue colors seen around an unpurged weld indicate an oxide layer 20 to 100 nm thick over a chromium-poor region — precisely the condition rouge needs.
ASTM A380 requires heat tint to be removed before passivation, because passivation chemistry will not remove it. A system commissioned with heat tint left in place has rouge initiation sites built into every weld.
Rouge in your system? Get it characterized before you clean it
Paul Industries derouges and re-passivates pharmaceutical, biotech and food-grade process and high-purity water systems nationwide. We characterize the deposit first, select the chemistry to match the class, execute under controlled parameters, and hand over documented Cr:Fe and iron results — not just a clean-looking pipe.
Frequently asked questions about rouge and derouging
What is rouge in stainless steel?
Rouge is a reddish, orange or black iron-oxide film that forms on the interior surfaces of stainless steel process and high-purity water systems. It is a deposit of iron oxide, principally hematite or magnetite, rather than general corrosion of the base metal, and it appears where the chromium oxide passive layer has been depleted or where iron has migrated from elsewhere in the loop.
What are the three classes of rouge?
Class I is migratory hematite that wipes off and originates elsewhere in the system. Class II is in-situ rouge, typically 0.1 to 0.5 mm thick, that grows from the parent metal and does not wipe off. Class III is magnetite, purple-black, up to 1 mm thick, and is found in clean steam and pure steam systems.
How do you remove rouge from a stainless steel system?
Rouge is removed by circulating an acid chemistry matched to its class: 0.5 to 2 percent citric acid at 50 to 60 degrees C for Class I, 4 to 6 percent citric acid with EDTA at 60 to 70 degrees C for Class II, and a 10 to 15 percent phosphoric acid formulation at 70 to 80 degrees C for Class III. Flow must stay above about 5 ft/s and pH between 1.5 and 3.0, and the system must be re-passivated afterwards.
What is the difference between passivation and derouging?
Passivation removes free iron from a clean surface and restores the chromium oxide layer under ASTM A967, removing essentially no base metal. Derouging removes an established iron-oxide deposit that has already formed, using more aggressive chemistry over a longer contact time. Derouging is always followed by passivation.
Can citric acid remove existing rouge?
Light Class I surface rouge, yes, and citric is often the preferred chemistry for it because it chelates iron selectively without attacking chromium. Adherent Class II and embedded Class III magnetite are a different matter: those generally need stronger chemistry, elevated temperature and repeat cycles with inspection between them, and citric alone will frequently under-perform. That is why classifying the oxide before selecting chemistry matters, and why a quotation issued without inspecting the system is a guess.
Why does rouge form in WFI systems?
Water-for-injection loops run hot, often above 65 degrees C, which accelerates iron oxide growth. Combined with any chloride ingress, dissolved oxygen above about 20 ppb, incomplete initial passivation, or dead legs and low-velocity zones, those conditions deplete the passive layer and allow rouge to establish.
Is rouge a GMP compliance problem?
It becomes one. Under 21 CFR 211.67 equipment must be maintained so that it does not alter the safety, identity, strength, quality or purity of the product. Visible discolouration on a product-contact surface typically triggers a documented impact assessment, including elemental impurity considerations under USP chapters 232 and 233.
How often should a system be inspected for rouge?
A common program is quarterly visual and borescope inspection, monthly dissolved and particulate iron testing by ICP-MS with results trended over time, and a full internal inspection with surface analysis annually. Hot WFI and clean steam systems warrant the tighter intervals.
Does electropolishing prevent rouge?
Electropolishing substantially delays rouge by removing surface iron, raising the chromium-to-iron ratio and taking surface roughness below about 0.5 micrometers Ra, which leaves fewer nucleation sites. It does not make a system immune, because it does not change water chemistry, operating temperature or fabrication defects such as dead legs.
Can heat tint cause rouge?
Yes. Weld heat tint is a chromium-depleted oxide scale 20 to 100 nanometres thick formed when a weld zone is exposed to oxygen at temperature. ASTM A380 requires it to be removed before passivation, because passivation chemistry will not remove it and it provides a ready initiation site for rouge.
What Cr:Fe ratio should a passivated surface achieve?
ASME BPE sets a minimum surface chromium-to-iron ratio of 1.3 to 1. Citric acid passivation typically achieves 1.7 to 2.0 or better, while nitric acid passivation typically returns 1.4 to 1.6. Ratios below 1.0 indicate an iron-rich surface that will rouge quickly.
Do you have to re-passivate after derouging?
Yes. The acid chemistry that dissolves iron oxide also strips the chromium oxide passive layer. A system returned to service without re-passivation to ASTM A967 has a bare, iron-rich surface and will re-rouge faster than it did before the cleaning.
Article sources
More questions we are asked
What does rouge mean in a WFI system?
Rouge is a discoloured iron oxide film that forms on stainless surfaces in hot high-purity water and clean steam service. In a WFI loop it appears as an orange, red, brown or black tint on the interior, and it matters for three reasons: it releases iron into water that is meant to be free of it, its rough surface gives biofilm somewhere to establish, and it can conceal pitting underneath. It forms because hot ultrapure water is aggressive toward the passive layer and mobilizes iron from the metal, which then redeposits. It is not a sign of poor-quality stainless; well-built WFI loops rouge as a normal consequence of temperature and time, which is why inspection intervals exist.
What is rouge in pharmaceutical water systems?
Rouge is classified into three types and the class determines the treatment. Class I is migratory: it originates elsewhere in the system, often from a pump or valve, travels and deposits on surfaces, is typically orange to red, and wipes off. Class II forms in place from the surface itself, is more adherent, and is frequently associated with chloride attack. Class III is black magnetite, forming at high temperature and most common in clean steam systems and the hottest parts of a WFI loop; it is the most tenacious and can be abrasive, migrating into valve seats and instruments. Identifying the class before selecting chemistry is what prevents either an ineffective treatment or an unnecessarily aggressive one.
How to prevent rouge contamination in WFI systems?
Rouge cannot be eliminated in hot service, only slowed, and the levers are specification and monitoring. Specify electropolished ASME BPE SF4 surfaces on product contact, since electropolishing leaves a chromium-enriched, low-iron surface that resists iron release far better than mechanical polish. Passivate thoroughly after fabrication and after every modification, and never passivate before the final tie-ins are made. Control chlorides, since they attack the passive film independently of temperature. Avoid unnecessary thermal cycling, because repeated heating and cooling accelerates iron release. Then monitor rather than assume: annual borescope inspection against commissioning baseline photographs, with conductivity, total organic carbon and periodic iron testing trended, is what catches rouge while it is still a wipe-down.
Best methods for WFI system derouging
Method follows the rouge class. Class I frequently responds to a circulated chelating chemistry, commonly citric-based, at moderate temperature. Class II generally needs a stronger acid treatment, often phosphoric or a formulated blend, with longer contact time. Class III magnetite usually requires a reducing agent such as oxalic acid or a proprietary formulation, because chelation alone does not break down magnetite effectively. In all cases the chemistry is circulated rather than soaked, so velocity assists removal, it is heated and held, and the system is rinsed to a conductivity endpoint rather than for a fixed time. Every derouging must end in re-passivation, because the treatment strips the protective film along with the oxide, and a system derouged without re-passivation rouges again faster than before.