Temperature Limits of Duplex and Super Duplex Stainless Steel Pipe: 475 °C Embrittlement, Sigma Phase and Low-Temperature Design

Temperature Limits of Duplex and Super Duplex Stainless Steel Pipe: 475 °C Embrittlement, Sigma Phase and Low-Temperature Design

On This Page

Table of Contents

Technical review: XTD Special Steel, metallurgy and product engineering. Reviewed against ASTM A789/A790, ASME BPVC Section VIII Div. 1, ASME B31.3, EN 13445, NORSOK M-630 and IMOA guidance. Last updated: 31 October 2026 (revision 2: all temperature and heat-treatment values re-verified against ASTM A790/A790M-24 and ASME B31.3-2018).

Quick Answer

Duplex and super duplex stainless steel pipe is normally applied between about −50 °C and 315 °C. The limits are set by phase stability in the ferrite, not by loss of strength.

  • Upper limit (code): ASME Section VIII Div. 1 and B31.3 publish allowable stresses up to 315 °C (600 °F). EN 13445 practice stops at 250 °C (482 °F). IMOA recommends 315 °C to avoid alpha-prime formation.
  • Upper limit (metallurgy): above about 300 °C the ferrite embrittles over time (peak rate near 475 °C). Between 600 °C and 1000 °C sigma phase destroys toughness and corrosion resistance, which is mainly a fabrication risk.
  • Lower limit: below about −50 °C the ferrite approaches its ductile-to-brittle transition. Low-temperature service must be qualified by Charpy V-notch testing, commonly 45 J average at −46 °C.

Key Takeaways

  • The limits are metallurgical, not mechanical. Duplex retains most of its strength at 300 °C — allowable stress for S32205 falls by about 15 % from ambient to the 600 °F table limit — but what fails first is toughness, not strength.
  • The governing code sets the number, not the mill datasheet: ASME allows 315 °C, EN 13445 allows 250 °C.
  • Higher alloy means faster embrittlement: super duplex (S32750, S32760) precipitates intermetallic phases faster than 2205, and 2205 faster than lean duplex.
  • Weld metal and heat-affected zone (HAZ) toughness usually govern the minimum design metal temperature (MDMT), not the parent pipe.
  • Embrittled pipe cannot be repaired in service; it is replaced or fully re-solution annealed in a furnace.
  • Never stress-relieve duplex at 600–650 °C. That range sits inside the sigma and carbide window.

1. Why Duplex Has Temperature Limits

Duplex stainless steel is a two-phase alloy of roughly 50 % austenite and 50 % ferrite. Common grades include 2205 (UNS S32205 / S31803), 2507 (UNS S32750), Zeron 100 (UNS S32760) and lean duplex such as UNS S32101 and S32304.

  • Ferrite carries most of the strength, and the interlocking duplex structure is what resists chloride stress corrosion cracking (SCC), by breaking up the continuous transgranular crack path.
  • Austenite carries most of the toughness, ductility and weldability.

Table 1. Typical minimum mechanical properties compared with 316L

Alt text / table summary: Minimum yield, tensile, elongation and PREN for 316L, lean duplex 2304, 2205, 2507 and S32760 per ASTM A790 and A312.

GradeUNSMin. yield (MPa)Min. tensile (MPa)Min. elongation (%)PREN (typical)
316LS3160317048535≈ 24
Lean duplex 2304S3230440060025≈ 25–28
Duplex 2205S3220545065525≈ 32–35
Super duplex 2507S3275055080015≈ 40–43
Super duplex Zeron 100S3276055075025≈ 40–43

Mechanical values are specification minimums: duplex grades per ASTM A790, 316L per ASTM A312. PREN is calculated from chemistry rather than specified, and varies within the grade range. Check the current edition of the standard for the exact product form and wall thickness.

The two-phase balance is metastable. Ferrite is body-centred cubic (BCC) and holds more chromium and molybdenum in solution than it can retain at intermediate temperatures. Given time at temperature, it rejects those elements as precipitates.

Table 2. Precipitation and transformation windows in duplex stainless steel

Alt text / table summary: Precipitation windows in duplex stainless steel — alpha-prime 300–525 °C, carbide and nitride 450–800 °C, sigma and chi 600–1000 °C — with effects and triggers.

ReactionTemperature windowEffect on the pipeTypical trigger
Alpha-prime (α′) spinodal decomposition300–525 °C (peak ≈ 475 °C)Toughness and ductility collapse; hardness rises; slight loss of corrosion resistanceLong service above ~250–300 °C; slow cooling through the band; thick sections
Carbide (M23C6) and chromium nitride (Cr2N)450–800 °CChromium depletion at phase boundaries, preferential attackMultipass welding; insufficient quench
Sigma (σ) phase600–1000 °C (nose ≈ 850 °C)Hard, brittle, Cr- and Mo-rich; severe drop in toughness and pitting resistanceSlow cooling from anneal; holds during hot forming; high heat input welding
Chi (χ) and Laves phase600–900 °CMo/W-rich intermetallics; crack initiators and pitting sitesHigh Mo + W super and hyper duplex
Secondary austenite700–900 °CFine intragranular form is Cr/N-depleted and less corrosion resistantReheated weld HAZ
Scaling in air> 1000 °COxide scale, section loss, pickling neededHeat treatment atmosphere
Solidification1380–1470 °CPrimary ferrite solidification, austenite forms on coolingWeld pool and castings

Two consequences explain most field failures:

  1. Wall thickness is a hidden variable. A 6 mm tube cools from annealing temperature in seconds. A 30 mm header or forged flange spends far longer in the 600–1000 °C window, so thick sections need genuinely fast water quenching, not air cooling.
  2. Time matters as much as temperature. Below 300 °C alpha-prime formation is slow but not zero. It is a kinetic process with no true threshold, only a rate that is acceptable or not relative to design life.

2. Upper Limit I: 475 °C (Alpha-Prime) Embrittlement

Between roughly 280 °C and 525 °C the ferrite decomposes by spinodal decomposition into a chromium-rich α′ phase and an iron-rich α matrix. Nothing is visible under an optical microscope. The damage is at nanometre scale and shows up only as lower impact energy and higher hardness.

What 475 °C embrittlement does

  • Toughness falls before strength does. Charpy energy can drop by 50 % or more while hardness and strength rise only slightly, so a tensile test on an exposed sample proves nothing about embrittlement.
  • Fracture mode shifts from ductile tearing to brittle cleavage. A pipe that passed its hydrotest when new can fail under a pressure surge after tens of thousands of hours at 320 °C.
  • Corrosion resistance drops modestly, because chromium is partitioned into α′ and depleted in the surrounding matrix.

Kinetics: why 250 °C is not a magic number

Time-temperature-transformation (TTT) curves, usually drawn at a 50 % toughness-loss criterion, show three things worth designing around:

  1. Higher alloy grades embrittle faster. At the same temperature, 2507 reaches 50 % toughness loss in a fraction of the time taken by 2205, and lean duplex is slower still.
  2. The curve has a long low-temperature tail. At 300 °C the reaction can take years; at 250 °C it may exceed the design life; at 200 °C it is effectively irrelevant for process plant. The right question is not “what is the maximum temperature?” but “which temperature and time combination is acceptable for this design life?”
  3. Welds embrittle differently from base metal. Coarser HAZ and high-ferrite weld metal decompose at different rates, so a base-metal-only assessment understates risk in a welded header.

Practical rule. For continuous service above 250 °C, answer three questions before releasing the inquiry: (1) what is the actual metal temperature, including insulation, heat tracing and solar gain; (2) what is the design life in hours; (3) does the governing code permit it? A “yes” from the mill datasheet is not one of the three.

How is embrittlement detected?

  • Charpy V-notch impact test at the relevant temperature. This is the most sensitive routine indicator.
  • Hardness testing (Vickers or Rockwell). A rising trend against a baseline suggests α′ formation.
  • ASTM A923 Methods A, B and C screen for sigma and chi, not for alpha-prime. Run them to prove that the anneal worked and that no intermetallics are present; they say nothing about α′ resistance.
  • Metallography does not detect α′ and should not be used alone for this purpose; transmission electron microscopy or atom probe is needed in research work.

Can it be reversed?

Yes, but only by a full re-solution anneal at the ASTM A790 Table 1 temperature (1020–1100 °C for 2205) followed by rapid quenching. It cannot be done on an installed line. In practice, an embrittled pipe in a hot section is a replacement, not a repair.

3. Upper Limit II: Sigma, Chi and Laves Phase (600–1000 °C)

Above about 600 °C diffusion is fast enough for intermetallic precipitation, led by sigma phase nucleating at ferrite–austenite boundaries. Sigma is hard and brittle, contains roughly 30 % Cr and 5–7 % Mo, and grows by consuming exactly the elements that provide corrosion resistance. Impact toughness can fall to a few joules, and pitting resistance of the surrounding matrix drops.

Why this window matters more in the shop than in service

Duplex pipe rarely operates at 800 °C, so sigma is a fabrication defect. It forms during slow cooling from solution annealing, holds during hot forming, heat treatment of bends and fittings, and multipass welds with excessive heat input. The nose of the precipitation curve for 2205 sits around 850–900 °C with incubation times of minutes; for 2507 and hyper duplex it moves to shorter times. A pipe furnace-cooled at 30 °C/h instead of quenched can meet chemistry and tensile requirements and still fail a Charpy test.

Table 3. Relative susceptibility to intermetallic precipitation

Alt text / table summary: Lean, standard, super and hyper duplex compared by PREN and speed of intermetallic precipitation.

FamilyTypical gradesPRENPrecipitation kinetics
Lean duplexS32101, S32304, S32202≈ 25–28Slowest; most tolerant of thick sections and modest cooling rates
Standard duplexS31803, S32205≈ 32–35Moderate; quench required above ~10 mm section
Super duplexS32750, S32760, S32550≈ 40–43Fast; tightest welding and cooling windows
Hyper duplexS32707, S33207≥ 45Fastest; seamless tube only, specialist fabrication

PREN = %Cr + 3.3 × %Mo + 16 × %N; tungsten-bearing grades such as S32760 also credit 1.65 × %W. PREN ranks pitting resistance only. It says nothing about temperature capability.

4. What the Codes Actually Permit

This is where most procurement arguments start. The numbers differ by more than 60 °C, and none is wrong: each embeds a different assumption about exposure time and acceptable risk.

Table 4. Maximum service temperature by governing document

Alt text / table summary: Maximum duplex service temperature by code — ASME B31.3 and VIII Div.1 at 315 °C, EN 13445 practice at 250 °C, NORSOK M-630 and IMOA guidance.

Governing documentMax. service temperatureBasis and caveats
ASME BPVC Section VIII Div. 1 and ASME B31.3315 °C (600 °F)Allowable stress tables for S31803/S32205, S32750 and S32760 stop at 600 °F. The code publishes a stress value; it does not guarantee metallurgical suitability at that temperature.
European practice under PED (EN 13445-3)250 °C (482 °F)Conservative ceiling adopted to exclude long-term α′ embrittlement over an unlimited design life. Typical of PED-assessed equipment. Note that EN 13445-3 itself sets no single duplex limit: 250 °C comes from the permitted material temperature range in the material standards plus TÜV / notified-body assessment practice.
IMOA guidance315 °C (600 °F)Industry guidance, not a code. Recommends 315 °C maximum to avoid alpha-prime precipitation.
NORSOK M-630 data sheetsPer data sheet — e.g. D41/D51 (22Cr), D55/D57 (25Cr)Each MDS states its own design temperature range, impact test temperature and hardness limit. Quote the MDS number; a generic range is not a specification.
ASTM A789 / A790No numeric ceilingThe scope warns that these steels are susceptible to embrittlement in prolonged elevated-temperature use. The specification controls annealing, not service temperature.
NACE MR0175 / ISO 15156-3Sour-service envelopeHardness and H2S partial pressure limits as a function of temperature and pH. Independent of the thermal ceiling.

Design temperature is not metal temperature. Insulation, heat tracing, steam-out, adjacent hot equipment, solar gain on an unshaded rack and blocked-in no-flow conditions all push metal temperature above line temperature. A line tagged 240 °C can sit near 290 °C during a summer shutdown with tracing energised.

Which number should go on the datasheet?

  • ASME-governed project: 315 °C is the absolute ceiling; a margin to 280–300 °C is good practice for a 20-year life.
  • PED / EN-governed project: 250 °C. This is a conformity issue, not a preference.
  • Offshore / NORSOK: take the value from the applicable material data sheet.
  • Above 300 °C with a long design life: request TTT data and consider a super austenitic or nickel alloy.

5. Lower Limit: Charpy Toughness and the Ductile-to-Brittle Transition

Austenitic grades such as 304L and 316L are face-centred cubic and keep toughness to −196 °C. Duplex is half ferrite, and ferrite is BCC with a genuine ductile-to-brittle transition temperature (DBTT). That single fact excludes duplex from most cryogenic duty.

What governs low-temperature toughness

  • Ferrite fraction and morphology. More, coarser or continuous ferrite raises the DBTT. A weld that solidified at 70 % ferrite is far more brittle at −46 °C than a balanced 45 % base metal.
  • Nickel content. Nickel lowers the DBTT and is the main lever in low-temperature duplex grades.
  • Impurity control. Carbon, phosphorus and sulphur are the elements that damage low-temperature toughness and must be held low. Nitrogen is the opposite: it is a deliberate addition — 0.14–0.20 % in S32205, 0.24–0.32 % in S32750 — that stabilises austenite, raises PREN and improves toughness. Clean steelmaking matters more than the datasheet headline.
  • Section size and restraint. Thick, restrained joints turn a marginal result into brittle fracture.
  • Prior embrittlement. Sigma or α′ formed in fabrication consumes the toughness margin. The upper and lower limits interact.

How low-temperature suitability is proven

No code table says “duplex is good to −60 °C.” Suitability is demonstrated by Charpy V-notch (KV2) testing at the MDMT:

  • 45 J average at −46 °C is a common project requirement for offshore and low-temperature duplex piping (NORSOK-style data sheets), usually with a minimum single value and, for critical service, lateral expansion criteria. Take the exact acceptance criteria — average, single minimum, test location and lateral expansion — from the applicable material data sheet or project specification, not from a generic figure.
  • ASTM A923 Method B is the standard impact method for detecting detrimental intermetallic phase; Method A (metallographic) and Method C (ferric chloride corrosion) are companion checks.
  • ASME VIII-1 UHA-51: duplex lacks the deep cryogenic exemptions given to austenitic grades; in practice impact testing applies once MDMT falls below about −29 °C unless the UHA-51 exemption rules (material group, thickness, stress ratio, impact energy basis) are satisfied item by item — never assume an exemption.
  • Weld procedure qualification must include the HAZ. A base-metal result at −46 °C is not evidence for the joint.

Table 5. Indicative Charpy V-notch performance (guidance only; demonstrate by test on the actual heat and procedure)

Alt text / table summary: Indicative Charpy V-notch energy at −46 °C and −100 °C for 2205 and 2507 base metal, weld metal and HAZ, and sigma-affected material.

Material / conditionAt −46 °CAt −100 °CComment
2205 base metal, correctly annealedTypically 100–200 JNot normally qualifiedLarge margin; disappears if sigma is present
2205 weld metal / HAZ (2209 filler)Typically 45–100 JNot qualifiedGoverning location for MDMT
2507 base metalTypically 80–150 JGrade-dependentHigher strength, faster precipitation, tighter welding window
Low-temperature duplex (Ni-balanced)> 100 J27–100 J with qualified procedureNeeds purpose-designed chemistry and full qualification

Where duplex should not go: LNG at −162 °C, liquid nitrogen, liquid CO2 let-down, and any duty below about −60 °C without specific qualification. Use 304L/316L, high-manganese austenitic, 9 % Ni steel or a nickel alloy.

6. Thermal Cycling and Fatigue: The Limit Nobody Specifies

Thermal limits are usually written as one steady-state number, but cyclic duty fails by a different route.

  • Phase-level mismatch. Ferrite (about 11–12 × 10⁻⁶/K) and austenite (about 17–18 × 10⁻⁶/K) expand differently. Each cycle strains the phase boundaries; with α′ embrittlement this is a recognised low-cycle fatigue mechanism in heat exchangers and scrubber ducting.
  • Component-level advantage. Bulk expansion of duplex (~13 × 10⁻⁶/K) is well below 316L (~17 × 10⁻⁶/K), giving lower restraint loads and thinner expansion joints.
  • Thermal shock. Cold-water CIP on a hot line, steam-out of a cold vessel or emergency depressurisation impose gradients no steady-state rating captures. State cycle count and ramp rate in the inquiry.
  • Accumulated exposure. Cycling between 250 °C and ambient accumulates α′ time on the hot side, so a cyclic duty at 280 °C can behave worse than a steady one.

7. Solution Annealing and Hot Forming Windows

All duplex pipe is delivered solution annealed, and the window is narrow: too cold and sigma survives; too hot and ferrite fraction and grain size rise, degrading toughness. Annealing values below are from ASTM A790/A790M Table 1; A789 applies the same logic to tubing.

Table 6. Solution annealing and quenching requirements

Alt text / table summary: ASTM A790 solution annealing temperature, quench medium and hot-forming range for S32101, S32304, S31803, S32205, S32750, S32760 and S32707.

UNSCommon nameSolution anneal (°C)QuenchTypical hot-forming range (°C)*
S32101LDX 2101 / lean duplex1020Water or rapid cooling1150 → 900
S323042304925–1050Water or rapid cooling1150 → 950
S318032205 (original chemistry)1020–1100Water or rapid cooling1230 → 950
S322052205 (tightened chemistry)1020–1100Water or rapid cooling1230 → 950
S32550255 / Ferralium1040 minWater or rapid cooling1200 → 1000
S327502507 super duplex1025–1125Water or rapid cooling1230 → 1025
S32760Zeron 100 super duplex1070–1140Water or rapid cooling1230 → 1050
S32707Hyper duplex1080–1120Water or rapid cooling1200 → 1050

*Hot-forming ranges are shop practice, not code requirements; the anneal range is the mandatory value. Below the lower forming limit the material must be re-annealed. For seamless pipe, ASTM A790 permits quenching directly from hot-forming temperature if the pipe is still at or above the minimum solution temperature, and requires each length to be individually quenched or rapidly cooled by other means, not stack-cooled. Water is the usual medium, but the standard does not make it the only one.

Five heat-treatment mistakes that become field failures

  1. Air cooling a thick section. A 25 mm wall air-cooled from 1050 °C crosses 850 °C slowly enough to precipitate sigma.
  2. Re-using 2205 parameters on super duplex. A 1050 °C setpoint sits inside the S32750 range (1025–1125 °C) and is a valid A790 anneal, so temperature alone is not the problem: retained sigma comes from a short soak and slow cooling. Run 2507 in the upper part of its range and guarantee an individual quench.
  3. No soak time for large charges. Time at temperature must be based on thermocouple evidence, not furnace air temperature.
  4. Wrong atmosphere or no descaling. Annealing atmosphere and post-anneal pickling affect corrosion resistance independently of the thermal cycle.
  5. Re-annealing without re-testing. A re-heat-treated heat needs fresh mechanical and corrosion tests; the original MTC no longer describes the product.

8. Welding Controls That Decide the Real Limit

In welded systems the weld metal and HAZ usually govern the minimum service temperature, while the HAZ is normally the first location to lose toughness at the upper limit, because it is coarser and carries a higher ferrite fraction than the parent pipe. Both are controlled in the welding procedure, not the purchase order.

Table 7. Fabrication parameters commonly specified

Alt text / table summary: Welding and fabrication parameters for 2205 versus 2507/S32760 — filler metal, heat input, interpass temperature, shielding gas and ferrite limits.

Parameter2205 (S32205)2507 / S32760
Filler metalER2209 / E2209 (Ni over-alloyed)ER2594 / E2594
Heat input0.5–1.5 kJ/mm0.5–1.5 kJ/mm, tighter upper bound preferred
Interpass temperature≤ 150 °C≤ 100–150 °C per data sheet
PreheatNoneNone
Root purgeArgon, or Ar + N2 to restore austeniteAr + N2, oxygen < 50 ppm
Weld ferrite35–65 % (internal target)35–60 % (internal target)
Post-weld heat treatmentNot required, normally harmfulNot required; never stress-relieve
Post-weld cleaningPickle and passivateRequired after welding or heat treatment unless bright-annealed

Never stress-relieve duplex. A conventional 600–650 °C PWHT sits directly in the sigma and carbide window. If a client specification calls for PWHT on “all stainless steels,” challenge that clause in writing before fabrication, and cite ASME B31.3-2018 Table 331.1.1 Note (4): for P-No. 10H duplex, PWHT is not required, and if it is performed the temperature must be a full solution anneal — 1020–1100 °C for S31803/S32205 and 1025–1125 °C for S32750 — followed by rapid cooling, not a 600–650 °C stress relief. That citation is far stronger than “PWHT is harmful.”

9. Selection Matrix: When Duplex Is the Wrong Answer

Temperature is only one of the variables, but it eliminates whole families of alloys before cost is even discussed. The matrix below maps the duty window to the material that is normally correct, and flags the cases where duplex is the wrong answer. Use it to frame the inquiry, then confirm against the governing code and the project data sheet.

Table 8. Temperature-driven material selection

Alt text / table summary: Material selection by duty temperature from cryogenic to above 300 °C, showing where duplex, super austenitic, nickel alloy and 9 % Ni steel apply.

DutyTypical choiceReason
−50 to 250 °C, chlorides, high pressure2205 / 2507 duplexBest strength-to-weight with SCC resistance; inside all code ceilings
250–300 °C, long design lifeDuplex with written justification, or 6Mo super austeniticα′ kinetics acceptable for a finite life; 6Mo avoids the ferrite problem
> 300 °C, corrosive dutySuper austenitic (S31254, N08367) or nickel alloy (N06625, N10276)Fully austenitic: no α′ and no DBTT
> 400 °C, oxidation-limited321 / 347 / 310S or heat-resistant alloyDuplex is out of scope well before oxidation limits
−50 to −100 °C, qualifiedLow-temperature duplex with full CVN qualificationOnly with Ni-balanced chemistry and documented Charpy results
Below −100 °C304L / 316L, high-Mn austenitic, 9 % Ni or nickel alloyFerrite DBTT makes duplex unsuitable
Sour service (H2S)Duplex within ISO 15156-3 envelopeHardness and pH2S limits apply on top of the thermal ceiling

10. What to Specify and What to Check on the MTC

Temperature capability is only as good as the evidence behind the certificate. For duplex or super duplex pipe near either limit, specify and verify:

  1. Standard and grade. Order to ASTM A790 (pipe) or A789 (tube) and quote the UNS number, not just “2205.” Accept dual S32205/S31803 certification only if nitrogen and molybdenum meet the tighter S32205 ranges.
  2. Heat treatment evidence. Annealing range, quench medium and furnace type traceable to the actual charge.
  3. ASTM A923 Method A, B and/or C as agreed, reported per heat and per lot.
  4. Charpy V-notch at the MDMT, including weld metal and HAZ for welded pipe, with test temperature and each absorbed energy value, not an average only.
  5. Ferrite content by ASTM E562 point count or electromagnetic method, for base metal and weld.
  6. ASTM G48 Method A where pitting resistance is critical, with test temperature stated. A result at 20 °C says nothing about 50 °C.
  7. Hardness against both the product standard and the sour-service limit where applicable.
  8. Certification: EN 10204 3.1 as a minimum; 3.2 with independent witness for critical offshore and pressure duty.
  9. PMI on 100 % of lengths for super duplex, to catch grade substitution in mixed stock.
  10. Hydrostatic test or NDT per ASTM A999/A999M.
  11. HT-O option on welded pipe. ASTM A790 clause 6.3.2.1 allows welded pipe to be supplied without the final anneal only where A923 Method B or C verifies parent metal, weld and HAZ, and the pipe is stencil-marked “HT-O”. If the MTC carries HT-O, the A923 report is not an optional extra.

Red flags in a supplier’s offer

  • Annealing temperature quoted as one number, with no range or quench description.
  • Impact values quoted only as an average, with no test temperature.
  • “Equivalent to 2205” with no UNS number or standard.
  • Sigma risk dismissed because “the tensile test passed.”
  • PWHT offered as standard practice for duplex.

Operating at a Temperature Boundary?

Send us the design temperature, design life, governing code (ASME, EN, NORSOK or project data sheet) and the required Charpy test temperature. We will confirm grade suitability, propose the annealing and welding envelope, and quote seamless duplex or super duplex pipe with the test documentation your specification calls for.

Request a quote for duplex / super duplex pipe →

11. Frequently Asked Questions

1) What is the maximum service temperature of 2205 (UNS S32205) pipe?

It depends on the governing code. ASME Section VIII Div. 1 and B31.3 permit design up to 315 °C (600 °F); EN 13445 practice limits duplex to 250 °C (482 °F); IMOA recommends 315 °C as a practical ceiling. The limit prevents alpha-prime embrittlement of the ferrite, which begins slowly above about 300 °C and is fastest near 475 °C.

2) Is super duplex more sensitive to high temperature than 2205?

Yes. Higher chromium, molybdenum and tungsten accelerate alpha-prime and intermetallic precipitation. 2507 and S32760 reach a given loss of toughness faster than 2205 at the same temperature, and they need tighter cooling and welding control. The service ceiling on code is the same, but the safety margin is smaller.

3) Can duplex pipe be used at −46 °C?

Yes, when proven by test. Correctly annealed 2205 and 2507 base metal typically exceeds 45 J at −46 °C. The weld metal and HAZ usually govern, so the weld procedure must be qualified by Charpy testing at the MDMT.

4) Can duplex be used at cryogenic temperatures such as −196 °C?

No. Ferrite has a ductile-to-brittle transition, so standard duplex is not suitable for LNG, liquid nitrogen or other cryogenic duty. Use 304L/316L, high-manganese austenitic, 9 % nickel steel or a nickel alloy.

5) What happens to duplex at 475 °C?

The ferrite decomposes by spinodal decomposition into chromium-rich α′ and iron-rich α. Hardness and strength rise slightly, while impact toughness and ductility can fall by half or more, with a modest loss of corrosion resistance. It is the fastest embrittlement temperature, though reaction is detectable from about 280 °C over long exposure.

6) Why do EN 13445 and ASME give different maximum temperatures?

ASME publishes allowable stresses up to 315 °C and leaves metallurgical suitability to the designer. EN 13445 practice adopts a conservative 250 °C to exclude long-term α′ embrittlement for an unlimited design life. Each is correct within its own assumptions.

7) Is a short temperature excursion above the limit acceptable?

Often yes, because embrittlement is time-dependent, but it must be assessed. Brief excursions below the sigma range (for example a few hours at 300–350 °C) accumulate little α′, while any excursion into 600–1000 °C, such as a fire or welding repair, can precipitate sigma in minutes. Document the cumulative time and consider a Charpy check after significant excursions.

8) Why must duplex never be post-weld heat treated?

Conventional stress relief at 600–650 °C sits in the sigma, carbide and nitride window and would reduce toughness and corrosion resistance. Duplex welds are controlled by heat input, interpass temperature and filler selection instead. If solution annealing is required, it must be done at the full A790 temperature followed by rapid quench.

9) Which tests prove a duplex pipe is suitable for a low-temperature or high-temperature duty?

Charpy V-notch at the MDMT (base metal, weld metal and HAZ), ASTM A923 Methods A, B and C, ferrite content measurement, ASTM G48 corrosion testing at a stated temperature, hardness testing, and an EN 10204 3.1 or 3.2 certificate showing heat-treatment evidence traceable to the charge.

12. References

  • ASTM A789/A789M: Seamless and Welded Ferritic/Austenitic Stainless Steel Tubing for General Service.
  • ASTM A790/A790M-24: Seamless and Welded Ferritic/Austenitic Stainless Steel Pipe, Table 1 (Heat Treatment) and clause 6.3 for the HT-O option.
  • ASTM A923: Detecting Detrimental Intermetallic Phase in Wrought Duplex Austenitic/Ferritic Stainless Steels.
  • ASTM G48: Pitting and Crevice Corrosion Resistance of Stainless Steels by Use of Ferric Chloride Solution.
  • ASME Boiler and Pressure Vessel Code, Section VIII Div. 1 (UHA-51) and Section II Part D.
  • ASME B31.3-2018: Process Piping — Table A-1 (allowable stresses) and Table 331.1.1, Note (4) (PWHT of P-No. 10H duplex).
  • EN 13445-3: Unfired Pressure Vessels, Design.
  • NORSOK M-630: Material Data Sheets and Element Data Sheets for Piping.
  • NACE MR0175 / ISO 15156-3: Materials for Use in H2S-Containing Environments.
  • IMOA: Practical Guidelines for the Fabrication of Duplex Stainless Steels, 2nd edition.
  • ISSF: Duplex Stainless Steels, Guidance and Information.

Zhejiang Xintongda Special Steel Manufacturing Co., Ltd. is a dedicated manufacturer of high-performance stainless steel seamless pipes and tubes with over a decade of engineering expertise. Our manufacturing capabilities focus on austenitic stainless steels, duplex and super duplex grades, as well as advanced nickel alloys designed for critical and high-integrity applications. With a strong emphasis on metallurgical precision, process stability, and full traceability, Xintongda delivers seamless pipe solutions engineered to perform reliably in high-pressure, high-temperature, and corrosive environments worldwide.

Catalogue Phone E-mail Whatsapp