ASTM G48 and G28 Corrosion Testing: Buyer's Guide
Date: 2026年9月29日 Categories: News Views: 329
By Shanghai Hangbo Alloy Group Co., Ltd. - China Supplier | ISO 9001:2015 | Est. 2012
Contact: sales@hangboalloy.com | hangbo@nickel-alloy.com | WhatsApp (Lisa): +86 13611656360
Quick Answer: How Do I Specify ASTM G48 or G28 on a Purchase Order?
Name the test method and edition, the test temperature and duration, the specimen orientation and surface condition, and a numerical acceptance criterion. Order the test on the delivered product in the supplied condition, and if the component will be welded, require a test specimen that includes the weld. A test without a stated acceptance criterion is not a requirement.
Key Takeaways
- A corrosion test without an acceptance criterion is a dispute waiting to happen. The specification must state the method, the conditions, the specimen preparation and the pass or fail limit, because a laboratory can only report results against a criterion someone has defined.
- ASTM G48 measures pitting and crevice resistance in ferric chloride, and the method matters as much as the test name. The pitting and crevice methods, and the electrochemical critical-temperature methods, answer different questions and produce results that cannot be compared across methods.
- ASTM G28 covers intergranular corrosion, principally through the Huey test and the ferric sulfate-sulfuric acid test. It is the appropriate check where a material has been sensitised by welding or by an incorrect heat treatment rather than where pitting resistance is the concern.
- The condition of the material decides the result. Solution-annealed material must be tested in the condition supplied, and testing a heat-treated coupon rather than the delivered product defeats the purpose of the test.
- Where the fabrication is welded, the specimen must contain a weld. Pitting and intergranular attack in service concentrate at welds and heat-affected zones, and a parent-metal specimen tests the part of the component least likely to fail.
- The cost of corrosion testing is small relative to the cost of a corrosion failure. A properly specified test adds a modest amount to the order price and converts the material's corrosion performance from an assumption into a verified result.
Why Corrosion Testing Must Be Specified, Not Assumed
A nickel alloy is usually selected for its corrosion resistance, and the certificate that accompanies it normally reports chemistry, tensile properties and heat treatment. None of those three items verifies corrosion resistance. A chemistry certificate shows that the elements are within their ranges; it does not show that the heat treatment produced a microstructure free of sensitisation or secondary-phase precipitation. A tensile test shows strength and ductility; it does not show whether the grain boundaries will resist attack. A heat treatment record shows what was intended; it does not show whether the resulting condition is the one the application requires. Corrosion testing is the only check that measures the property the material was chosen for.
The gap between a compliant certificate and a corrosion-resistant component is a real operating risk rather than a theoretical one, because the two most common corrosion failures in these materials are both invisible to routine testing. The first is sensitisation in austenitic grades, where chromium carbides precipitate at the grain boundaries during slow cooling through, or exposure to, a critical temperature range, leaving narrow chromium-depleted zones that corrode preferentially. The second is the precipitation of molybdenum-rich secondary phases in the high-molybdenum Ni-Cr-Mo alloys, which produces a similar localised depletion and which is why those alloys avoid post-weld heat treatment. Both conditions pass a chemistry check, both pass a tensile test, and both are readily detected by a corrosion test in the correct medium.
There is also a commercial reason to specify the test rather than assume it. Nickel alloys are expensive, and a corrosion failure in a chemical plant, a scrubber, a bleach stage or a seawater system is not only a material cost: it is a shutdown, a repair, a schedule impact and sometimes a safety event. The incremental cost of a specified corrosion test is a small fraction of the material cost, and it converts an unverified assumption into a recorded result that can be compared against the acceptance criterion and, later, against service experience. Where a plant has suffered corrosion failures in a particular duty, the test requirement is usually the first clause to be added at the next order, and it would have been cheaper to add it earlier.
Two qualifications keep the requirement proportionate. First, corrosion testing is not necessary on every nickel alloy order: where the alloy was selected for strength or for elevated-temperature service rather than for corrosion resistance, the test adds cost without adding information, and the money is better spent on the checks that address the actual failure mode. Second, the test must be matched to the mechanism. Pitting and crevice resistance, intergranular corrosion, stress-corrosion cracking and general wastage are different failure modes with different test methods, and a G48 result says nothing about intergranular resistance in a sensitisation-prone grade. The sections that follow set out which method addresses which mechanism, and how to write the requirement so that the result means what the buyer intends.
Chemical Composition: The Elements That Decide Pitting and Intergranular Resistance
Corrosion test results are predictable from chemistry, which is why the chemistry certificate and the corrosion test complement one another rather than duplicating each other. The elements that determine pitting and crevice resistance are chromium, molybdenum, nitrogen and tungsten, and the relationship between them is captured by the pitting resistance equivalent number used as a ranking tool across the stainless and nickel alloys.
| Grade | UNS number | Cr | Mo | N | W | Cu | Product standard (example form) |
|---|---|---|---|---|---|---|---|
| 904L | N08904 | 19.0-23.0 | 4.0-5.0 | - | - | 1.0-2.0 | ASTM B625 (plate), B677 (tube) |
| 254SMO | S31254 | 19.5-20.5 | 6.0-6.5 | 0.18-0.25 | - | 0.5-1.0 | ASTM A240 (plate), A312 (pipe) |
| AL-6XN | N08367 | 20.0-22.0 | 6.0-7.0 | 0.18-0.25 | - | 0.75 max | ASTM B688 (plate), B676 (tube) |
| Duplex 2205 | S32205 | 22.0-23.0 | 3.0-3.5 | 0.14-0.20 | - | - | ASTM A240 (plate), A276 (bar) |
| Super duplex 2507 | S32750 | 24.0-26.0 | 3.0-5.0 | 0.24-0.32 | - | 0.50 max | ASTM A240 (plate), A276 (bar) |
| Incoloy 825 | N08825 | 19.5-23.5 | 2.5-3.5 | - | - | 1.5-3.0 | ASTM B424 (plate), B425 (bar) |
| Hastelloy C-276 | N10276 | 14.5-16.5 | 15.0-17.0 | - | 3.0-4.5 | - | ASTM B575 (plate), B574 (bar) |
| Alloy 59 | N06059 | 22.0-24.0 | 15.0-16.5 | - | - | 0.50 max | ASTM B575 (plate), B574 (bar) |
| Hastelloy C-22 | N06022 | 20.0-22.5 | 12.5-14.5 | - | 2.5-3.5 | - | ASTM B575 (plate), B574 (bar) |
Table note: The ranges summarised are taken from the ASTM product standards named in the table (latest editions) and are shown to illustrate the compositional basis of corrosion performance; the controlling limits are those of the standard named on the purchase order. The pitting resistance equivalent number is commonly calculated as PREN = Cr + 3.3 (Mo + 0.5 W) + 16 N and is used as a ranking tool rather than as an acceptance criterion, because it does not account for microstructure, heat treatment condition or the actual environment. Composition data reported on a mill certificate should be checked against the specification limits before a corrosion test result is interpreted.
The table shows why corrosion testing is required in addition to chemistry. Two heats of the same grade can both be inside their chromium and molybdenum ranges and still differ materially in corrosion performance, because performance also depends on the microstructure produced by melting, processing and heat treatment. A heat in which chromium carbides have precipitated at the grain boundaries has a chemistry certificate that is entirely correct, and a corrosion test is the check that detects the resulting sensitisation. Similarly, a heat in which molybdenum-rich secondary phases have formed after exposure to an intermediate temperature can show a perfectly acceptable analysis and fail an intergranular test. In both cases the corrosion test measures the consequence of the processing rather than the composition, which is precisely why the two checks are complementary.
The compositional ranking also explains why the choice of test method must be matched to the grade. The high-molybdenum grades, including the super-austenitic stainless steels and the Ni-Cr-Mo alloys, are tested for pitting and crevice resistance because that is the mechanism their alloying is designed to resist, and ASTM G48 in ferric chloride is the conventional method. The austenitic and duplex grades with lower molybdenum contents are more often assessed for intergranular resistance after welding or heat treatment, which is the domain of ASTM G28. Applying the wrong test produces a result that cannot be interpreted: a G28 result on a super-austenitic grade says little about its pitting resistance, and a G48 result on a sensitised austenitic grade says little about its grain-boundary condition. The selection matrix later in this article sets out the pairing we recommend.
Mechanical Properties and Companion Acceptance Requirements
A corrosion test result is meaningful only if the material has already satisfied the mechanical requirements of the order, because a lot that fails its tensile requirements will not be accepted whatever its corrosion performance, and a lot that has been incorrectly heat treated may pass a corrosion test while failing its specified strength. The table below gives typical properties for the grades most often ordered with a corrosion acceptance requirement, with the standard basis stated.
| Grade | Typical tensile strength, annealed | Typical 0.2 % yield | Typical elongation | Companion requirement |
|---|---|---|---|---|
| 904L | ~490-590 MPa | ~220-260 MPa | ~35-45 % | intergranular test after welding |
| 254SMO | ~650-750 MPa | ~300-350 MPa | ~35-45 % | pitting and crevice test, ferrite check for duplex |
| AL-6XN | ~690-760 MPa | ~310-380 MPa | ~35-45 % | pitting and crevice test |
| Duplex 2205 | ~620-700 MPa | ~450-480 MPa | ~25-35 % | ferrite content plus pitting test |
| Super duplex 2507 | ~800-880 MPa | ~550-600 MPa | ~25-30 % | ferrite content plus pitting test |
| Incoloy 825 | ~550-620 MPa | ~240-280 MPa | ~30-40 % | intergranular test for welded service |
| Hastelloy C-276 | ~750-800 MPa | ~355-420 MPa | ~40-50 % | pitting and crevice test |
| Alloy 59 | ~690-780 MPa | ~340-380 MPa | ~50-55 % | pitting and crevice test |
| Hastelloy C-22 | ~750-800 MPa | ~355-420 MPa | ~40-50 % | pitting and crevice test |
Table note: The values shown are typical published annealed properties and are explicitly not standard minima; acceptance minima for a specific product form are fixed by the governing specification named on the purchase order. Mechanical testing is performed to ASTM E8/E8M at room temperature, ASTM E21 at elevated temperature, ASTM E10 or E18 for hardness and ASTM E112 for grain size. For duplex and super duplex grades, ferrite content is normally specified and verified in addition to the mechanical requirements, and the corrosion test result should be interpreted together with the ferrite content because the two are related.
The practical reason for stating the companion requirements on the same order is procedural rather than metallurgical. If a corrosion test is required and the mechanical tests are not explicitly required on the same order, the supplier may perform the corrosion test on the material without having tested its strength, and a failure discovered later leaves the buyer in an awkward position: the material cannot be accepted on strength grounds, and the corrosion result that was supposed to qualify it is of no use because the lot will be rejected anyway. Specifying the full test schedule on one order, with the order of testing understood, avoids both the duplication of effort and the ambiguity about which result governs.
For duplex and super duplex grades there is a specific point that is frequently missed. Their corrosion performance depends on the balance between austenite and ferrite, and an incorrect heat treatment or an unsuitable welding procedure can produce an unbalanced microstructure, an intermetallic phase or an excessive ferrite content, all of which reduce corrosion resistance. The corrosion test detects the consequence, and the ferrite content measurement identifies the cause, so the two requirements belong together. Where a duplex fabrication is welded, the procedure qualification should include a corrosion test on the welded specimen in addition to the ferrite measurement, because the heat-affected zone is where the balance is most likely to have been disturbed.
Heat Treatment and Condition: Why the Wrong Condition Invalidates the Test
The condition of the material determines its corrosion performance, and a test performed on material in the wrong condition produces an answer to the wrong question. This is the most common technical error in ordering corrosion tests, and it takes two forms: testing a coupon that has been given a different heat treatment from the delivered product, and testing parent metal in isolation from a weld that will exist in service.
| Condition issue | Effect on corrosion behaviour | How the order should address it |
|---|---|---|
| Solution annealed versus as-formed | cold work and residual stress alter localised corrosion behaviour | require testing in the condition supplied |
| Sensitised condition, austenitic grades | chromium-depleted grain boundaries corrode preferentially | require intergranular test to ASTM G28 |
| Secondary phases in Ni-Cr-Mo grades | molybdenum-depleted boundaries reduce corrosion resistance | avoid post-weld heat treatment; test the delivered condition |
| Incorrect duplex balance | reduced pitting resistance, intermetallic formation | require ferrite content plus pitting test |
| Weld and heat-affected zone | localised attack in service, the most common failure location | require specimens containing a weld |
| Surface condition: mill finish, pickled, ground | passive film formation and surface defects affect results | state the surface condition required for the specimen |
| Heat treated test coupon instead of product | result describes the coupon, not the delivered material | require testing on the delivered product |
Table note: The condition effects listed are established metallurgical behaviour for these material families; the applicable test methods are ASTM G48 for pitting and crevice corrosion, ASTM G28 for intergranular corrosion and ASTM G150 for the electrochemical critical pitting temperature of stainless steels. Where a specification permits a test coupon to represent a lot, the coupon must be taken from the material being supplied and given the same heat treatment, and this requirement should be stated explicitly because the alternative interpretation allows a separately processed coupon to be substituted.
The single most valuable clause in a corrosion test specification is the requirement that the specimen be taken from the delivered product in the condition in which it will be used. That clause closes off the most common route by which a corrosion test produces a misleading pass: a coupon prepared and heat treated specifically for the test, representing a condition that the delivered material may not be in. Where the order requires testing on the delivered product, the result describes the material that will be installed, and the buyer has something that can be compared with the acceptance criterion and with subsequent service experience.
The second most valuable clause concerns welds. In service, pitting, crevice attack and intergranular corrosion in welded fabrications concentrate at the weld and the heat-affected zone, and in many failure investigations the parent metal is found to be intact while the weld zone has perforated. A corrosion test on parent metal alone therefore tests the part of the component least likely to fail, and it gives no information about the region where the risk actually sits. Where the component will be welded, the specification should require the corrosion test to be performed on a specimen that includes a production-representative weld, with the weld cap and root in the as-welded condition unless the order requires dressing or treatment. This requirement costs slightly more and addresses the dominant failure mode.
A third practical point concerns surface condition, which is routinely overlooked. Corrosion test results depend on the surface state of the specimen: a pickled surface, a machined surface and a ground surface have different passive film formation behaviour and different levels of residual cold work, and a specimen prepared differently from the delivered product can give a result that does not describe the installation. The specification should state the required surface condition of the specimen and, where the production material is supplied in a specific finish, that the specimen should match it. Where a result is close to the acceptance limit, the difference in surface preparation can be enough to change the pass or fail outcome, and stating the requirement removes that variable.
ASTM G48 Methods Explained
ASTM G48 is a set of methods that use ferric chloride solution to assess the pitting and crevice corrosion resistance of stainless steels and related alloys, and the method chosen determines what the result means. The methods differ in the type of attack they measure, in the specimen and apparatus they use, and in whether they produce a mass-loss result at a fixed temperature or a critical temperature. A specification that names ASTM G48 without naming the method has not defined the test.
| Method family | What it measures | Typical output | When to use it |
|---|---|---|---|
| Pitting test in ferric chloride at a specified temperature | resistance to pitting attack in an oxidising chloride medium | mass loss per unit area, or pass/fail against a limit | ranking and acceptance for super-austenitic and Ni-Cr-Mo grades |
| Crevice corrosion test in ferric chloride | resistance to attack under a crevice, which is more severe than open pitting | mass loss with crevice formers in place | gasketed joints, flange faces, under-deposit conditions |
| Second crevice arrangement (alternate crevice geometry) | crevice attack under a different former geometry | mass loss by area | where a specific crevice condition must be reproduced |
| Critical pitting temperature (electrochemical) | the temperature at which stable pitting initiates | temperature in °C | comparative material ranking and quality control |
| Critical crevice temperature (electrochemical) | the temperature at which stable crevice attack initiates | temperature in °C | comparative ranking where crevices are unavoidable |
| Related electrochemical method for stainless steels (ASTM G150) | critical pitting temperature by a defined electrochemical procedure | temperature in °C | austenitic and duplex stainless grade qualification |
Table note: ASTM G48 is a multi-method standard and the specific method designations, their numbering and their detailed requirements are those of the current edition of the standard, which should be consulted directly and named on the purchase order. The table is arranged by what each method family measures rather than by method letter, because the numbering in the standard has evolved across editions and a specification that relies on a remembered letter is unreliable. ASTM G150 is a separate standard for the electrochemical determination of the critical pitting temperature of stainless steels and is not a substitute for ASTM G48 where a ferric chloride mass-loss result is required.
The distinction that matters most in practice is between mass-loss testing at a fixed temperature and the determination of a critical temperature. A mass-loss test at, for example, a specified temperature for a specified duration produces a number in grams per square metre or a corrosion rate, and the acceptance decision is a comparison against a limit. A critical temperature determination produces a temperature value, and the acceptance decision is a comparison against the service temperature with an agreed margin. Both are legitimate, and they answer different questions: the mass-loss test answers "does this material resist attack under these conditions", while the critical temperature test answers "at what temperature does this material start to fail". For quality control on a delivered lot, the mass-loss test at a defined temperature against a defined limit is usually the more directly useful requirement, because it produces an unambiguous pass or fail on the material in hand.
The crevice test deserves specific attention because it is more severe than the pitting test and because it corresponds more closely to real plant conditions than an exposed surface does. Crevice attack occurs under gaskets, under deposits, at flange faces, under fastener heads and in any geometry where the solution becomes stagnant and the local chemistry changes; a material that performs well in a pitting test can fail in a crevice test in the same solution. Where the component will contain flanged joints, gasketed connections, or areas where solids can settle, the crevice test is the more representative requirement, and specifying the pitting test alone is a common and consequential oversight. Where the application has no crevices, the pitting test alone may be sufficient, and this should be a considered decision rather than a default.
Two practical points apply to both families of G48 testing. First, the test temperature is the most influential single variable, and the specification must state it explicitly and, where relevant, state whether the acceptance criterion is evaluated at that temperature as a mass-loss limit or as a temperature threshold. Second, the test solution and its preparation are defined by the standard, and any deviation - a different ferric chloride concentration, a different solution volume to specimen area ratio, or a different specimen preparation - changes the result, so the specification should require the current edition of the standard to be followed in full rather than reproducing a remembered procedure in the purchase order.
ASTM G28 and Intergranular Corrosion Testing
ASTM G28 addresses intergranular corrosion in wrought nickel-rich, chromium-bearing alloys, and it is the appropriate check where the risk is attack along grain boundaries rather than pitting of the surface. The risk arises from sensitisation in austenitic grades and from grain-boundary precipitation in the high-molybdenum nickel alloys, and both conditions are consequences of processing and thermal history rather than of composition.
| Method family | What it measures | Typical conditions | When to use it |
|---|---|---|---|
| Boiling nitric acid test (Huey) | intergranular attack in a strongly oxidising acid | boiling nitric acid, multiple test periods with mass loss per period | austenitic grades and alloys where sensitisation is the concern |
| Ferric sulfate - sulfuric acid test | intergranular attack in a less strongly oxidising medium | boiling acid solution with ferric sulfate addition | nickel-chromium-molybdenum alloys and welds |
| Weight-loss evaluation across periods | whether attack accelerates, indicating grain-boundary penetration | mass loss per period, ratio between periods | all methods, to detect progressive attack |
| Weld specimen evaluation | intergranular attack in the weld and heat-affected zone | specimen containing a production-representative weld | welded fabrications in sensitisation-prone grades |
| Microstructural examination in support | confirmation of the mechanism where a mass-loss result is borderline | metallographic section through the specimen | investigation and acceptance disputes |
Table note: ASTM G28 is a multi-method standard covering boiling nitric acid and ferric sulfate-sulfuric acid procedures, and the specific method designations and requirements are those of the current edition, which should be named on the purchase order. The mass-loss evaluation across successive test periods is significant because intergranular attack typically produces accelerating loss as the grain boundaries are penetrated, whereas uniform corrosion produces a steady rate; a single-period result cannot always distinguish the two. Where an acceptance decision is close to the limit, microstructural examination of the tested specimen gives the mechanism directly.
The distinction between the Huey method and the ferric sulfate-sulfuric acid method follows the chemistry of the environment rather than a preference. The boiling nitric acid test is highly oxidising and is used for austenitic stainless steels and for alloys where the expected exposure is oxidising; the ferric sulfate-sulfuric acid test is used in the same class of materials and is particularly relevant for the nickel-chromium-molybdenum alloys and for weld evaluation, because it discriminates between correctly and incorrectly processed material in those grades. Selecting the method requires knowing both the material and the environment, and where the user is unsure, the correct course is to specify the method that corresponds to the service medium rather than to default to whichever method the supplier offers.
The weld specimen requirement is as important for G28 as for G48 and for the same reason. Sensitisation occurs in the heat-affected zone when chromium carbides precipitate during welding and cooling, and the resulting chromium-depleted boundaries are where intergranular attack begins. A test on parent metal alone can pass while the weld zone in the delivered fabrication is sensitised, which is the outcome the test was intended to prevent. Where a fabrication will be welded, the specification should require an intergranular corrosion test on a specimen containing a production-representative weld, in the as-welded condition, and the acceptance criterion should be the same as for the parent metal unless the application justifies a different limit.
Two further points help when reading a G28 result. First, the result is a mass loss or a corrosion rate, and its significance depends on the exposure time and on how the test was structured: a rate that appears acceptable over a short period can deteriorate sharply in later periods if attack is intergranular, which is why the successive-period evaluation matters. Second, the result describes the specimen, not the lot. Where an order covers multiple pieces or multiple heats, the sampling plan determines how much the result tells you about the delivered quantity, and the specification should state how many tests are required and how the material is to be sampled. A single test result on a multi-heat delivery is a partial verification, and the specification should reflect that honestly.
Specifying, Witnessing and Reporting the Test
A corrosion test requirement is only as good as the wording that defines it, and the wording has to cover six items: the method, the specimen, the conditions, the acceptance criterion, the witnessing arrangement and the report content. The checklist below sets out what each item should contain, and it doubles as a review list for a specification that is about to be issued.
| Requirement | What the specification must state | Common omission |
|---|---|---|
| Method | the standard, the edition and the specific method | method named without edition or method designation |
| Specimen | orientation, location, size, surface condition, whether a weld is included | no specimen orientation or surface condition stated |
| Conditions | temperature, duration, solution per the standard | temperature not stated |
| Acceptance criterion | numerical limit and the units, or the temperature threshold | no criterion, leaving pass or fail undefined |
| Sampling | how many tests, from which items, for which heats | one test assumed to represent the whole delivery |
| Witnessing | whether third-party witnessing is required and at which stage | inspection required without scope or timing |
| Report content | method, conditions, individual results, criterion, conformity statement | result reported without conditions or criterion |
| Material condition | the condition in which the material is tested and supplied | coupon condition not tied to the delivered product |
Table note: The checklist reflects common industrial practice for specifying corrosion testing on metal purchase orders and is intended as a drafting aid rather than as a standard. The test methods referenced are ASTM G48, G28 and G150 as published in their current editions, and the inspection document framework is that of EN 10204 where a formal inspection certificate is required. Where a customer, code or end user imposes additional requirements such as accreditation of the testing laboratory or witnessing by a nominated inspector, those requirements should be stated explicitly on the purchase order.
The acceptance criterion is the item most often missing, and its absence is the reason corrosion test disputes are common. A laboratory asked to perform a G48 test without a stated limit will report a result, and the question of whether the result is acceptable then falls to be argued after the event, usually with the material already produced and sometimes already shipped. A stated numeric limit, with its units, removes that argument entirely: the report either satisfies the criterion or it does not, and both parties know which before the test is requested. Where the appropriate limit is uncertain, it should be agreed at enquiry stage, ideally by reference to the service conditions and to published data for the grade, and recorded on the order.
Witnessing arrangements deserve the same precision. Where third-party witnessing is required, the order should state which party witnesses, what they witness, at what stage the material is presented and what document they issue. A witnessing clause without scope is satisfied by an inspector attending and signing, which may or may not correspond to what the project needed. Where the corrosion test itself is to be witnessed, the inspection plan should show it as a hold point so that the material is not cut or dispatched before the result is available, since a specimen that has been removed from a finished component cannot be re-tested.
Finally, the report content should be specified, because a result without its conditions cannot be compared with anything. A useful test report identifies the material and heat, the specification and method, the specimen orientation and surface condition, the test conditions including temperature and duration, the individual results, the stated acceptance criterion and a conformity statement, together with the laboratory performing the test and, where required, the witnessing inspector. Reports of that kind can be compared across suppliers and across years, and they become useful when a plant is investigating a corrosion problem and needs to know what was actually verified when the material was purchased.
Matching the Test to the Material and the Service
The test requirement should follow from the material and from the failure mechanism the service can produce, and the pairing below reflects that logic. The right question is not "which test is standard for this grade" but "what would fail first in this service, and which test detects it".
| Material and service | Failure mechanism of concern | Recommended test requirement |
|---|---|---|
| Super-austenitic stainless, seawater or chloride cooling | pitting and crevice attack | pitting and crevice test to ASTM G48 with a stated temperature and limit |
| Super-austenitic stainless, flanged or gasketed joints | crevice attack under seals and deposits | crevice method with crevice formers included |
| Duplex and super duplex, welded fabrication | unbalanced microstructure and pitting in the heat-affected zone | ferrite content plus G48 pitting test on a welded specimen |
| Austenitic grade, welded, oxidising service | sensitisation and intergranular attack | intergranular test to ASTM G28 on a welded specimen |
| Ni-Cr-Mo alloy, chemical or bleach service | pitting, crevice and grain-boundary precipitation | G48 for pitting and crevice, G28 where weld processing is critical |
| Ni-Cr-Mo alloy after any intermediate-temperature exposure | secondary-phase precipitation at grain boundaries | G28 on the delivered condition; avoid post-weld heat treatment |
| Nickel-copper and nickel grades in seawater | general and localised attack in flowing or stagnant conditions | immersion or pitting test matched to the medium |
| Material for a benign environment, selected for strength | none requiring a corrosion test | no corrosion test; spend the effort on mechanical verification |
| Thin-wall tube for heat exchangers | through-wall pitting and crevice attack under deposits | pitting and crevice test plus eddy current examination where specified |
| Material for a specific process liquor not covered by a standard test | environment-specific attack | immersion test in the actual liquor with an agreed criterion |
Table note: The pairings reflect common industrial practice and the mechanisms that the respective test methods are designed to detect; the governing specification, the client requirement and the applicable code determine what is actually mandatory. Where the service medium is not represented by a standard test method, an immersion test in the actual process liquor, with a defined temperature, duration and acceptance criterion, is often more informative than a standard test in a surrogate medium, and the criterion should be agreed in writing before the test is performed.
Two general rules emerge from the table. First, match the test to the mechanism: pitting and crevice tests for the grades whose alloying is designed to resist those mechanisms, intergranular tests where the risk arises from thermal history, and environment-specific tests where the medium is unusual. Second, always test the region that will fail. In welded fabrications that is the weld and heat-affected zone, in gasketed equipment it is the crevice, and in thin-wall tube it is the through-wall section under a deposit; the parent-metal specimen in an uncreviced holder tests the least vulnerable part of the component and gives correspondingly little assurance.
The rule about not testing where no risk exists is worth stating explicitly, because over-testing has a cost in both money and lead time. Where a nickel alloy has been selected for elevated-temperature strength or for low-temperature toughness, a corrosion test adds nothing to the assurance the application needs, and the same budget is better spent on the checks that address the actual failure mode, such as grain size verification, impact testing or dimensional verification. The discipline is to identify the mechanism that would cause the component to fail and to verify that mechanism, rather than to add tests because they are available or because a previous project included them.
Price Reference (2026, EXW Shanghai)
The cost of corrosion testing is a small addition to the cost of nickel alloy material, and the price ranges below are for the materials themselves on a reference basis. Because testing, documentation and inspection requirements vary between orders, the position of any specific quotation within a range is determined more by the specification than by the market.
| Material (reference form) | Grade | Reference range, EXW Shanghai | Note |
|---|---|---|---|
| Super-austenitic plate | 904L / 254SMO / AL-6XN | USD 18-45/kg | molybdenum content drives the band |
| Duplex and super duplex bar | 2205 / 2507 | USD 12-35/kg | 2507 carries a substantial premium over 2205 |
| Nickel-iron-chromium bar | Incoloy 825 | USD 25-42/kg | corrosion testing and documentation move the band |
| Nickel-chromium-molybdenum plate | Hastelloy C-276 / C-22 | USD 38-68/kg | solution annealed and rapidly quenched |
| Nickel-chromium-molybdenum plate | Alloy 59 | USD 42-68/kg | narrower production base than C-276 |
| Duplex tube for heat exchangers | 2205 / 2507 | USD 22-50/kg | size and wall thickness dominate |
| Super-austenitic tube | 254SMO / AL-6XN | USD 30-65/kg | non-standard sizes carry a premium |
| Nickel-copper bar | Monel 400 | USD 20-34/kg | lowest-cost nickel alloy of this group |
Table note: Reference range only, 2026, EXW Shanghai, USD/kg. These figures float with the LME nickel price and with the molybdenum market and are not a quotation. Corrosion testing adds cost through specimen preparation, laboratory time and the material consumed, and the addition is small relative to the material price for the grades in this table; the figure depends on the method, the number of specimens and whether the test is witnessed. Where a specification requires welded specimens, the cost includes the preparation of a qualified weld, and the requirement should therefore be considered at enquiry stage rather than added after the order is placed.
The commercial judgement in this area is straightforward once the framing is right. These are expensive materials, and they are bought for corrosion resistance in services where a failure means lost production, repair cost and sometimes environmental or safety consequences. The incremental cost of specifying a corrosion test with a clear acceptance criterion is a small percentage of the material value, and it converts the central assumption of the purchase into a documented result. Where a plant has experienced corrosion failures, the cost of a single unplanned shutdown usually exceeds the corrosion testing cost of every order placed that year, which is why the test requirement tends to be added to every specification once an organisation has been through the experience. Adding it deliberately, at the point where the material is specified, is the cheaper route.
Conversely, the cost of over-specification is real but smaller, and it consists mainly of lead time. Requiring a full corrosion test programme on material for a benign duty adds a laboratory step to the schedule for no benefit. The correct approach is therefore the one described throughout this article: identify the mechanism of concern, specify the test that detects it with a stated criterion, test the region that will fail, and leave the material alone where the application presents no corrosion risk that testing could quantify.
Standard Index
| Standard | Title / scope | Covers | Form |
|---|---|---|---|
| ASTM G48 | Pitting and crevice corrosion resistance of stainless steels and related alloys by use of ferric chloride solution | test methods | - |
| ASTM G28 | Detecting susceptibility to intergranular corrosion in wrought, nickel-rich, chromium-bearing alloys | test methods | - |
| ASTM G150 | Electrochemical critical pitting temperature testing of stainless steels | test method | - |
| ASTM G31 | Laboratory immersion corrosion testing of metals | test method | - |
| ASTM G36 | Evaluating stress-corrosion cracking resistance in boiling magnesium chloride solution | test method | - |
| ASTM G46 | Examination and evaluation of pitting corrosion | test method | - |
| ASTM A262 | Detecting susceptibility to intergranular attack in austenitic stainless steels | test methods | - |
| ASTM B575 / B574 | Nickel-chromium-molybdenum alloy plate, sheet and strip, and rod and bar | composition + mechanical | plate, bar |
| ASTM B625 / B677 | Nickel-chromium-molybdenum-copper alloy (UNS N08904) plate and tube | composition + mechanical | plate, tube |
| ASTM A240 / A276 / A789 / A790 | Stainless and duplex stainless plate and sheet, bars, and tube | composition + mechanical | plate, bar, tube |
| ASTM B688 / B676 | Nickel-chromium-molybdenum alloy (UNS N08367) plate and tube | composition + mechanical | plate, tube |
| ASTM E8 / E8M / E21 | Tension testing at room and elevated temperature | test method | - |
| ASTM E10 / E18 | Brinell and Rockwell hardness testing | test method | - |
| ASTM E112 | Determining average grain size | test method | - |
| ASTM E1476 / E572 | Metals identification and analysis by PMI and X-ray spectrometry | test method | - |
| EN 10204 | Metallic products - types of inspection documents (2.2, 3.1, 3.2) | inspection documents | all forms |
Table note: Standards are listed by number and scope; where an edition year is not quoted, the latest edition applies and the controlling edition is the one named on the purchase order. ASTM G48 and ASTM G28 are both multi-method standards, and the specific method designation must be stated on the order because the methods within each standard measure different things and produce results that cannot be compared across methods. Accreditation requirements for the testing laboratory, and any requirement for witnessed testing, must be stated explicitly because they are not implied by the material specification.
FAQ
Q1: Do I need a corrosion test on every nickel alloy order?
No. A corrosion test is required where the alloy was selected for its corrosion resistance, because in that case the material's performance in the intended medium is the property the purchase depends on and it is not verified by a chemistry or tensile certificate. Where the alloy was selected for elevated-temperature strength, for low-temperature toughness or for structural purposes in a benign environment, a corrosion test adds cost and lead time without adding assurance, and the verification effort is better directed at the property that actually governs the application, such as grain size, impact toughness or dimensional accuracy. The correct approach is to identify the failure mechanism that would end the component's life and to verify that mechanism, which sometimes means a corrosion test and sometimes means something else entirely. In our enquiry review we ask for the service environment before proposing a test requirement, and where the environment presents no corrosion risk we say so rather than adding a test that will not inform the decision.
Q2: What is the difference between ASTM G48 and ASTM G28?
ASTM G48 assesses pitting and crevice corrosion resistance using ferric chloride solution, and it is the appropriate test for the grades whose alloying is designed to resist localised chloride attack, including the super-austenitic stainless steels and the nickel-chromium-molybdenum alloys. ASTM G28 assesses susceptibility to intergranular corrosion in wrought nickel-rich, chromium-bearing alloys, and it addresses the attack that follows sensitisation or grain-boundary precipitation rather than surface pitting. The two tests answer different questions and neither substitutes for the other: a good G48 result says nothing about whether the grain boundaries have been depleted by chromium carbide precipitation, and a good G28 result says nothing about whether the surface will pit in chloride service. Our stainless and special alloy range and our nickel-chromium-molybdenum range are the two families where this distinction most often decides which test is ordered.
Q3: Which ASTM G48 method should I specify?
Specify the method that measures the mechanism you are protecting against, and state the method explicitly together with the current edition of the standard, because ASTM G48 is a multi-method document and a reference to the standard alone does not define the test. Where the risk is open-surface pitting, the pitting method with a stated temperature and a mass-loss or pass-fail criterion is appropriate. Where the component contains gasketed joints, flange faces, fastener assemblies or areas where solids can settle, the crevice method is the more representative requirement, because crevice attack is more severe than open pitting and a material that performs well in one test can fail in the other. Where comparative ranking between candidate materials is the purpose, the electrochemical methods that determine critical pitting or crevice temperature produce a more discriminating result. The method designation and its detailed requirements should be taken from the current edition rather than from memory, because the numbering and content of the methods have evolved.
Q4: What acceptance criterion should I state for a G48 test?
State a numerical limit with its units, together with the temperature and duration at which the limit applies, or state the temperature threshold where a critical-temperature method is used. Without a criterion the laboratory can only report a result and the question of acceptability falls to be argued after the material has been produced, which is the most common cause of corrosion test disputes. Where the appropriate limit is not obvious, the correct approach is to agree it at enquiry stage by reference to the service conditions and to published data for the grade and product form, and to record the agreed criterion on the purchase order so that both parties are working to the same number. A criterion that reflects the actual service medium is more useful than a generic limit taken from a previous project in a different environment, because the severity of the service, and therefore the margin required, varies with the chloride level, the temperature, the oxidation state and whether crevices are present.
Q5: Why should the corrosion test be performed on a welded specimen?
Because in welded fabrications the weld and heat-affected zone are where corrosion failures actually occur, and a test on parent metal verifies the part of the component least likely to fail. Welding alters the microstructure locally: it can sensitise austenitic grades by precipitating chromium carbides in the heat-affected zone, it can unbalance duplex grades, and it can leave thermally altered zones in the nickel-chromium-molybdenum alloys in which secondary phases form. Corrosion failures in welded plant equipment are found at welds and their immediate surroundings far more often than in the parent plate, which is why a corrosion test specification that does not include a production-representative weld leaves the dominant risk unverified. The requirement costs slightly more because a qualified weld must be prepared for the specimen, which is a reason to decide it at enquiry stage rather than after the order has been placed. Our other inspection and materials articles cover the related welding and heat treatment questions.
Q6: Why does the heat treatment condition affect the corrosion test result?
Because corrosion resistance depends on microstructure as well as on composition, and microstructure is produced by processing and heat treatment. Two heats with identical chemistry can differ substantially in corrosion performance if one has been correctly solution annealed and rapidly quenched and the other has been slowly cooled through an intermediate temperature range, allowing chromium carbides or molybdenum-rich secondary phases to precipitate at the grain boundaries. Those precipitates drain the surrounding zones of the elements that provide corrosion resistance, and the result is preferential attack along the grain boundaries that a chemistry certificate cannot detect. This is why the specimen must be taken from the delivered product in the condition in which it will be used, rather than from a separately prepared coupon, and why a specification should require the tested condition to be stated in the report.
Q7: How many specimens should be tested on a multi-heat order?
State the sampling plan on the order, because a single test result on a multi-heat delivery verifies one piece rather than the lot. The appropriate number depends on the consequence of failure and on how the material will be used: a single heat supplied as one item may reasonably need one test, while a delivery assembled from several heats to be used in a critical service justifies a test per heat and, where the material is welded, a welded specimen from a representative production weld. It is also worth stating how the specimen is to be selected and identified, so that the correspondence between the test result and the delivered material can be demonstrated later. Where a supplier proposes a reduced sampling plan, the proposal should be assessed against the criticality of the application rather than accepted on the basis of cost, and any agreed reduction should be recorded in writing.
Q8: Should third-party inspection witness the corrosion test?
Witnessing is appropriate where the consequence of a corrosion failure is high, where the client or an applicable code requires independent verification, or where the buyer has reason to want an independent record of the test being performed on the correct material. Where witnessing is required, the order should state who witnesses, what they witness, at what stage the material is presented, and what document they issue; a clause that says only that inspection is required leaves the scope to be decided by the inspector and may not cover the test at all. The corrosion test should be shown as a hold point in the inspection plan so that material is not cut or dispatched before the result is available, because a specimen removed from a finished component cannot be re-tested. Where the test is witnessed, the report should identify the witnessing party and the document reference so that the record is traceable.
Q9: What should a corrosion test report contain?
A useful report identifies the material and heat number, the specification and the test method with its edition, the specimen orientation, location and surface condition, whether a weld was included, the test conditions including temperature, duration and solution, the individual results with units, the acceptance criterion that was applied, a conformity statement, and the identification of the testing laboratory together with the witnessing party where applicable. A report that gives a result without the conditions cannot be compared with anything and is of limited value in a future investigation. Reports of this completeness also allow a plant to compare results across suppliers, across heats and across years, which is how a corrosion testing programme becomes a source of engineering knowledge rather than a single acceptance decision. Our technical knowledge centre covers how we record and interpret these results for the grades we supply.
Q10: Can I use an immersion test in the actual process liquor instead of a standard test?
Yes, and where the service medium is unusual or is not represented by a standard test, an immersion test in the actual process liquor is often more informative than a standard test in a surrogate medium. The requirement then has to be written carefully: state the liquor composition and its source, the temperature, the duration, whether the conditions are aerated or deaerated, whether a weld is to be included, and above all a numerical acceptance criterion expressed as a mass loss, a corrosion rate or a penetration depth, agreed in writing before the test is performed. Where the medium varies between normal operation and upset conditions, the test should represent the condition that governs the material choice, and that should be stated. The main limitation of a liquor test is comparability: results are specific to that liquor, so they provide less basis for comparing suppliers than a standard test does, which is why the two are often used together.
Q11: Why does duplex 2205 need a ferrite check as well as a corrosion test?
Because the corrosion performance of duplex stainless steels depends on the balance between austenite and ferrite in the microstructure, and that balance is established by heat treatment and disturbed by welding. An incorrect solution annealing temperature or cooling rate can produce an unbalanced structure or allow intermetallic phases to form, and welding introduces a heat-affected zone in which the balance is altered; all of these reduce corrosion resistance. A corrosion test detects the consequence, and the ferrite content measurement identifies whether the cause is a microstructural imbalance, so the two requirements work together rather than duplicating each other. This is why we specify ferrite content together with a corrosion test on duplex and super duplex orders, and why a welded specimen is particularly valuable for these grades. Our duplex and super-austenitic range covers the grades we supply for these duties.
Q12: What information should I provide to get the right corrosion test specified?
Provide the material and product form, the service environment with its chemical species, concentration, temperature, pH and oxidation state, whether crevices or deposits will be present, whether the component will be welded, the surface condition in service, and the consequence of a corrosion failure. Those answers determine which mechanism is of concern and therefore which test method applies, and they also determine the acceptance criterion, because the margin required depends on how severe the service is. It is also useful to say whether the material is destined for a coded or client-approved application, since that may dictate the test method, the laboratory accreditation and the inspection document type. Send your service data through our contact page and we will propose the test requirement with a stated criterion and, where the medium is not covered by a standard method, an immersion test programme using your actual liquor. Our other corrosion and quality articles cover the wider materials selection and verification questions.
Conclusion and Selection Rules
Specifying corrosion testing correctly comes down to four decisions, and each one is written into the order: the method, chosen to match the failure mechanism; the specimen, taken from the delivered product, in the supplied condition and, where the component will be welded, containing a production-representative weld; the conditions, stated explicitly with temperature and duration; and the acceptance criterion, stated numerically. An order containing those four items produces a result that means what the buyer intends, and it converts the material's corrosion performance from an assumption into a recorded verification.
Two rules are worth stating as rules. First, a corrosion test without a stated acceptance criterion is not a requirement, because it leaves the pass or fail decision to be argued after the material has been produced. Second, a test on parent metal in a welded fabrication verifies the least vulnerable region of the component, and where a weld will exist in service the specimen should include one. Everything else in corrosion test specification is a question of matching the test to the mechanism and the material, and of keeping the requirement proportionate to the risk.
Shanghai Hangbo Alloy Group Co., Ltd. supplies super-austenitic stainless, duplex and super duplex stainless, Incoloy, Monel and nickel-chromium-molybdenum alloys in plate, sheet, tube, pipe, bar and forgings, with mill test certification, chemistry analysis to ASTM E572 and PMI to ASTM E1476, corrosion testing to ASTM G48 or G28 on request with agreed acceptance criteria, ferrite content verification for duplex grades, and third-party inspection by SGS, BV or TUV including witnessed testing where required. Send your service conditions through our contact page and we will propose the test requirement and the acceptance criterion for the duty, and quote the material with the testing scope stated explicitly. Our other corrosion and materials selection articles cover the wider questions behind the test requirement.
Contact & Complete Product Range
Shanghai Hangbo Alloy Group Co., Ltd. - China Supplier
Email: sales@hangboalloy.com | WhatsApp (Lisa): +86 13611656360 | www.nickel-alloy.com
All Grades
Ni: NI200-N6 | Monel: 400-502, K-500, R-405 | Inconel: 600-740H, X-750, 725 | Incoloy: 800-27-7MO | Hastelloy: C-276-HYBRID-BC1, N | Super Austenitic: 904L-AL-6XN | Chromium-Rich: Alloy 33 | High-Strength Stainless: Nitronic 50/60, PH13-8Mo | Chlorination: Alloy 45 | Nimonic: 75-263 | Haynes: 25-HR-160 | RA: 330/333 | Precision: 1J22-4J45 | High-Strength: 18Ni250-AerMet100 | Heating: Cr20Ni80/60 | Specialty: Alloy 20-602CA
Product Forms: Bars(6-500mm) | Tubes(OD 6-610mm) | Plates(0.1-100mm) | Forgings | Welded Pipes | Fittings | Wire(0.05-10mm) | Strip(0.02-4.0mm)
SGS/BV/TUV. www.nickel-alloy.com
