NACE MR0175 / ISO 15156 — Petroleum and Natural Gas Industries — Materials for Use in H2S-Containing Environments

What the standard is for, and the one it is confused with

ANSI/NACE MR0175/ISO 15156 is titled Petroleum and natural gas industries, materials for use in H2S-containing environments in oil and gas production. Every word of that title is load bearing. It is a production standard: wells, wellheads, trees, flowlines, gathering systems, field processing and the transport of unprocessed sour fluid. It addresses cracking, specifically sulfide stress cracking, stress corrosion cracking, hydrogen induced cracking with its stress-oriented and stepwise variants, soft zone cracking and galvanically induced hydrogen stress cracking.

The standard it is confused with is NACE MR0103, published as ISO 17945, which covers materials resistant to sulfide stress cracking in corrosive petroleum refining environments. Refineries handle wet H2S too, but the mechanisms, the temperatures and the fabrication practices differ enough that the two documents diverge. Writing MR0175 into a refinery fabrication specification produces requirements the fabricator cannot always meet while omitting the ones that matter, and it is one of the most common specification errors in downstream procurement.

Since NACE International and SSPC merged, the document is maintained under AMPP, while the ISO identity is unchanged. Older purchase orders reference NACE MR0175 alone, some reference ISO 15156 alone, and a few still reference the pre-2003 NACE MR0175, which was a materials list rather than the three-part environment-based standard in force today. If a supplier quotes against that old document, the technical content of the offer is not what the buyer believes it to be.

When service is sour at all

Sour service is not a description, it is a calculated condition. For a gas system, MR0175 treats the service as sour when the total absolute pressure reaches 0.4 MPa, about 65 psi, and the partial pressure of H2S in the gas reaches 0.3 kPa, which is 0.05 psi. Multiphase and oil systems have their own criteria, based on gas-oil ratio and the composition of the gas phase at the bubble point.

Those numbers are small. A stream carrying a few parts per million of H2S at a few hundred psi crosses the threshold comfortably. That is precisely why the calculation must be performed rather than eyeballed, and why the H2S content used should be the design maximum over the life of the field rather than the composition in the first commissioning sample. Souring over field life has forced more retrospective material changes than any analytical error ever has.

Once the service is sour, Part 2 places carbon and low-alloy steels on a severity map built from H2S partial pressure and in-situ pH, dividing the domain into regions 0 through 3. Region 0 is not sour service. Regions 1 to 3 are increasingly severe, and the region decides which materials and which qualification evidence are open to you. In-situ pH is calculated rather than measured in the field, so a wrong pH assumption can move a design a full region in either direction.

Part 2: the numbers that get quoted, and the ones that get missed

For carbon and low-alloy steels the number everyone knows is 22 HRC. It is a real limit and it is not the whole requirement. Part 2 also caps the nickel content of carbon and low-alloy steels at 1 percent by mass, restricts free-machining steels, limits cold-deformed components unless they are stress relieved, and requires that the material fall within one of the tabulated material groups with its associated conditions. A 22 HRC result on a steel outside those groups proves nothing at all.

For weldments the controlling number is hardness measured in Vickers with a 10 kilogram load. Weld metal, heat-affected zone and base metal are all required to sit at or below 250 HV10 for carbon steel weldments in sour service. HRC and HV are not interchangeable in this context; a portable Rockwell reading on a weld cap is not a substitute for a Vickers traverse, because the indentation samples a volume larger by an order of magnitude while the hardest microstructure is local and narrow.

Bolting is the component people forget. Exposed bolting in sour service cannot be ASTM A193 Grade B7. It is B7M, heat treated to a maximum of 235 HBW, with A194 Grade 2HM nuts, marked accordingly so the grade can be verified without a laboratory. A flange assembly with sour-qualified body material and B7 studs is a non-compliance sitting in plain sight, and it turns up on walkdowns constantly.

Part 3: alloys inside an envelope, not alloys on a list

Part 3 does not approve corrosion-resistant alloys in the abstract. It tabulates each material type against an operating envelope: maximum temperature, maximum H2S partial pressure, maximum chloride concentration, presence or absence of elemental sulfur, and in some cases pH and applied stress. An alloy is acceptable inside its envelope and unacceptable outside it, which means the same grade can be compliant on one line of a plant and non-compliant on the next.

Type 316 austenitic stainless steel is the standard illustration. It appears in Part 3 with a maximum temperature of 60 degrees Celsius, 140 Fahrenheit, together with limits on H2S partial pressure and chloride content, and it is excluded where elemental sulfur is present. Engineers who remember 316 as sour-service acceptable and forget the temperature ceiling specify it into hot separator service, where it is outside the envelope entirely and the mistake is invisible on the material certificate.

Duplex and super duplex grades, nickel-based alloys and precipitation-hardened materials each carry their own tables, and several carry hardness limits and heat treatment conditions on top of the environmental limits. The correct citation in a material requisition is therefore not the standard number alone. It is the standard, the part, the material type and the specific table entry the selection relies on. That one extra line prevents a fabrication dispute six months later.

Qualification testing, and the criteria MR0175 refuses to give you

Where a material is not covered by a tabulated entry, or where the buyer wants evidence rather than a table, MR0175 points to laboratory qualification. Sulfide stress cracking resistance is demonstrated by NACE TM0177, whose Method A uses a uniaxial tensile specimen, Method B a bent beam, Method C a C-ring and Method D a double cantilever beam. The classic Method A exposure runs 720 hours in an acidified brine saturated with H2S at ambient temperature.

Resistance to hydrogen induced cracking is demonstrated by NACE TM0284, which sections exposed coupons and reports crack length ratio, crack thickness ratio and crack sensitivity ratio. Here is the trap: MR0175 does not set acceptance values for those ratios. It requires HIC-resistant steel where the service demands it and points to the test method, but the numeric limits, commonly written as crack length ratio 15 percent, crack thickness ratio 5 percent and crack sensitivity ratio 2 percent averaged over three sections, come from the purchaser's own specification.

If a purchase order says HIC tested to NACE TM0284 and stops there, the mill can supply plate with cracks in it and a technically compliant certificate. The acceptance criteria have to be stated in the order, along with the number of test locations, the sampling frequency per heat, the test solution and whether Solution A or Solution B applies. Reviewing that requisition wording before it is issued is far cheaper than rejecting plate on arrival, and it is routine ASNT Level III consulting work.

Welding, hardness surveys and the finding that repeats

Most MR0175 non-conformances found in the field are hardness related, and most of those are survey related rather than metallurgy related. The requirement is that weld metal, heat-affected zone and base metal all satisfy the limit. The heat-affected zone is a narrow band, and its hardest region is not usually where the convenient flat surface happens to be.

The recurring finding is a hardness survey performed only on the weld cap, or performed with a portable rebound instrument whose indentation samples a volume far larger than the heat-affected zone itself. Both approaches produce comfortable numbers and neither demonstrates compliance with anything. A macro section with a Vickers traverse across the fusion line, or a properly conditioned in-situ survey with defined indentation locations, is what the requirement is actually asking for.

Post-weld heat treatment is the second recurring misunderstanding. PWHT usually lowers weldment hardness and often brings a joint into compliance, but performing PWHT is not itself evidence of compliance. Hardness has to be measured after the treatment on the qualification weld and monitored in production. A file containing the treatment chart and no hardness results is proof that the oven ran, not proof that the weldment meets 250 HV10.

What the documentation has to prove

MR0175 compliance is not certified by any body. There is no stamp, no registration and no third-party authorisation scheme. Responsibility sits with the equipment user, and the only defence is the record: the material specification and heat number, the mill certificate with chemistry and mechanical properties, hardness results tied to a specific location and method, the welding procedure qualification with its hardness traverse, and the Part 2 or Part 3 table entry the selection relies on.

That set of records has to be assembled per item and retrievable years later, usually during an incident review or a change-of-service study when the original engineers have moved on. Sour-service documentation held as scanned PDFs in project folders is effectively lost. Holding it against the equipment record, where the material basis, the hardness evidence and the inspection history sit on the same asset, is the difference between answering a question in an hour and reopening a fabrication file. That is the argument for an inspection data management system.

Where the evidence already exists but its quality is unknown, an independent read is the fastest way to find out what you have. Report validation applied to mill certificates, procedure qualification hardness traverses and positive material identification records regularly turns up certificates that cite the wrong table, hardness values recorded with no method stated, and heat numbers that do not appear anywhere on the part.

How Atlantis works a sour-service scope

Sour-service verification happens at the supplier's works, not at the client's desk. Atlantis mobilises inspection teams from Houston and Hyderabad to the mill, the forge shop or the fabrication yard, and once mobilised the team stays deployed for the duration of the contract rather than rotating between visits. The inspector who witnessed the procedure qualification weld is the inspector who reviews the production hardness results months later.

For a sour-service package that normally means positive material identification, witnessing of hardness surveys and mechanical testing, review of procedure qualification records against the applicable part and table, verification of bolting grade and marking, and a documented reconciliation of every heat against the requisition. Continuity matters here because a supplier's substitutions become visible only when someone remembers exactly what was agreed at the pre-inspection meeting.

To scope supplier surveillance or a materials compliance review against MR0175 and ISO 15156, start with a consultation describing the service conditions, the referencing project specification and the supplier locations involved.

The document trail an operator needs to prove MR0175 compliance years later

MR0175 compliance is established at the point of material selection and qualification, but the evidence that it was actually applied correctly is frequently needed years afterward — during an incident investigation, a fitness-for-service assessment, or a change-of-service review — by which point the people who made the original selection may no longer be with the organisation. What the standard itself does not specify is how that evidence should be retained, which leaves each operator to build its own documentation practice, and the quality of that practice varies enormously.

The documentation that actually survives a later audit links three things for each sour-service component: the H2S partial pressure and pH conditions the selection was qualified against, the specific material specification and any hardness or qualification testing performed, and the MR0175 edition and part relied upon at the time — because the standard has been revised repeatedly and a component qualified under an older edition is not automatically re-qualified simply because a newer edition is now current.

Where that chain is broken — commonly because material selection was documented in an engineering calculation that was archived separately from the procurement and construction records — reconstructing MR0175 compliance for an existing asset becomes a materials engineering exercise in its own right, undertaken years after the decision it is trying to verify was actually made. Treat the standard's title literally: it names materials for H2S environments, not a general corrosion-resistance guarantee, and qualifying a material against MR0175 says nothing about its suitability against chloride, amine, or naphthenic acid attack in the same circuit.

Frequently asked

Is NACE MR0175 the same document as ISO 15156?

They are the same technical content published under two identities. The joint document is normally cited as ANSI/NACE MR0175/ISO 15156 and is issued in three parts. Purchase orders referencing only one identity are pointing at the same requirements. What is not the same is the pre-2003 NACE MR0175, which was a materials list rather than the environment-based three-part standard, and old references to it should be corrected before they reach a supplier.

Who maintains MR0175 now that NACE has become AMPP?

NACE International and SSPC merged to form the Association for Materials Protection and Performance, and the standard is maintained within that organisation while remaining an ISO standard through ISO/TC 67. The designation in common use has not changed, so documents citing NACE MR0175/ISO 15156 remain correct and do not need rewording.

What is an SSC region?

Part 2 maps the sour environment onto a domain of H2S partial pressure against in-situ pH and divides it into regions 0 to 3. Region 0 falls outside sour service. Regions 1 to 3 are increasingly severe and determine which carbon and low-alloy steel options and which qualification evidence are available. Because in-situ pH is calculated rather than measured, the assumptions behind it should be recorded alongside the material selection.

Does post-weld heat treatment guarantee compliance?

No. PWHT usually reduces weldment hardness and often brings a joint into compliance, but the standard requires hardness to be demonstrated rather than assumed. It has to be measured on the procedure qualification weld in the correct locations and monitored in production. A heat treatment chart with no hardness results is evidence that the cycle ran, not that the weldment meets the limit.

Does MR0175 cover corrosion rates or material loss?

No. It addresses environmental cracking only. Weight-loss corrosion, erosion-corrosion, CO2 corrosion modelling, inhibitor selection and corrosion allowance sit outside its scope entirely and have to be handled by separate materials and corrosion engineering. A material can be fully compliant with MR0175 and still be the wrong choice on general corrosion grounds.