Temper Embrittlement in Alloy Steels
Temper embrittlement is a reversible loss of notch toughness in alloy steels caused by trace impurity atoms segregating to prior austenite grain boundaries during slow cooling through, or prolonged holding within, a specific intermediate temperature band. This guide covers the segregation mechanism, which compositions are most susceptible, the J-factor and X-factor prediction tools, step-cooling testing, and why temper embrittlement remains a live consideration for Cr-Mo creep-resistant pressure equipment steels, including the modern P91/P92 family.
Key Takeaways
- Temper embrittlement is caused by phosphorus, antimony, tin, and arsenic segregating to prior austenite grain boundaries during slow cooling or holding in roughly the 350–575°C range.
- It typically produces little change in hardness or tensile strength, but can raise the DBTT by up to about 100°C and shifts fracture mode from transgranular to intergranular.
- Nickel, chromium, and manganese increase susceptibility; molybdenum (and to a lesser extent tungsten) suppresses it by slowing impurity segregation.
- Susceptibility is predicted compositionally using the Watanabe J factor (weld metal) and the Bruscato X factor (base metal), and verified experimentally with the step-cooling test.
- Classical temper embrittlement is reversible: reheating above ~600°C followed by rapid cooling redissolves the segregated layer and restores toughness.
- Historically most significant in heavy-section 2.25Cr-1Mo pressure vessels and Ni-Cr-Mo-V rotor forgings; modern P91/P92 steels are melted to tight impurity limits specifically to control this risk.
What Is Temper Embrittlement?
Temper embrittlement is a solid-state segregation phenomenon, not a bulk microstructural transformation: it does not appreciably change hardness, tensile strength, or the parent tempered martensite or bainite microstructure visible under an optical microscope. Instead, trace impurity elements — principally phosphorus, antimony, tin, and arsenic, which are present at low concentrations even in clean commercial steels — diffuse to and concentrate at prior austenite grain boundaries when the steel spends sufficient time in a critical intermediate temperature range, commonly cited as roughly 350–575°C (660–1065°F). This atomic-scale segregation weakens grain boundary cohesion, shifting fracture mode from ductile, transgranular tearing to brittle, intergranular fracture along the weakened boundaries, and raising the ductile-to-brittle transition temperature, in some documented cases by as much as 100°C.
Equilibrium Grain Boundary Segregation
The underlying driving force is equilibrium segregation thermodynamics: impurity atoms lower the total free energy of the system by occupying grain boundary sites rather than remaining dissolved in the bulk lattice, and the equilibrium segregation concentration is temperature-dependent. At very high temperature, thermal agitation limits segregation; at very low temperature, diffusion is too slow for meaningful segregation to develop within practical timescales. The critical embrittling range sits between these limits, where diffusion is fast enough to move impurity atoms to boundaries yet the equilibrium segregation level at that temperature remains significant.
Reversible vs Related Embrittlement Phenomena
Classical (Type II, reversible) temper embrittlement, described above, is distinguished from several related but mechanistically distinct low-alloy steel embrittlement phenomena that occur in overlapping or adjacent temperature ranges:
| Phenomenon | Temperature Range | Mechanism | Reversible? |
|---|---|---|---|
| Classical temper embrittlement (Type II) | ~350–575°C | Equilibrium impurity segregation to prior austenite grain boundaries | Yes, by reheat + rapid cool |
| One-step / “500°F” embrittlement (Type I) | 200–450°C tempering | Combined effect of austenitizing-stage segregation and carbide film formation during low-temperature tempering | No |
| 475°C embrittlement | ~400–550°C | Spinodal decomposition of ferrite into Fe-rich/Cr-rich regions (ferritic/duplex stainless steels) | Yes, by solution anneal |
| Blue brittleness | ~200–300°C | Dynamic strain aging (carbon/nitrogen interstitial locking of dislocations) | N/A, strain-rate/temperature dependent |
Effect on Mechanical Properties
The signature of temper embrittlement is a Charpy transition curve shift with little accompanying change in room-temperature hardness or tensile properties, which is exactly why routine hardness or tensile checks alone cannot detect it — dedicated impact testing before and after simulated or actual thermal exposure is required. Fractography confirms the mechanism directly: embrittled specimens show intergranular fracture facets along prior austenite grain boundaries, in clear contrast to the transgranular cleavage or ductile dimple rupture typical of non-embrittled material at the same test temperature.
Susceptible Compositions
Because segregation kinetics and grain boundary chemistry both depend on bulk alloy content, not just trace impurity level, certain alloying strategies markedly increase or decrease susceptibility for a given impurity content.
| Element | Effect on Susceptibility | Notes |
|---|---|---|
| Phosphorus, antimony, tin, arsenic | Primary embrittlers | Segregate directly to grain boundaries; keep as low as practical |
| Nickel | Increases susceptibility | Common in rotor and heavy forging steels (e.g. NiCrMoV) |
| Chromium | Increases susceptibility | Present in most creep-resistant Cr-Mo steels by design |
| Manganese, silicon | Increase susceptibility (co-segregation) | Promote segregation of primary embrittlers rather than embrittling directly |
| Molybdenum | Reduces susceptibility | Slows impurity segregation kinetics; a key reason Mo is retained in Cr-Mo pressure vessel steels |
| Tungsten | Reduces susceptibility (secondary) | Similar but generally weaker effect than molybdenum |
Predicting Susceptibility: J-Factor and X-Factor
Because full-scale step-cooling tests are time-consuming, compositional screening parameters are widely used at the material and welding procedure specification stage to flag high-risk heats or weld metals before fabrication.
J = (Mn + Si)(P + Sn) × 104 All elements in weight percent. Lower J indicates lower susceptibility. Common maximum acceptance value for critical pressure equipment weld metal: J ≤ 180 (project-specification dependent)
X = (10P + 5Sb + 4Sn + As) / 100 All elements in parts per million (ppm). Lower X indicates lower susceptibility. Values below roughly 15-20 are generally considered low risk for critical, heavy-section applications
Screening Tools, Not Guarantees
J-factor and X-factor calculations are useful compositional screening tools for comparing heats and ranking relative risk, but they do not replace actual mechanical testing. Section thickness, actual thermal history (cooling rate through the critical range, post-weld heat treatment cycles, and service temperature history), and grain size all influence the real degree of embrittlement a component experiences, none of which a purely compositional formula captures.
The Step-Cooling Test
The step-cooling test simulates the cumulative effect of years of slow cooling or elevated-temperature service exposure within an accelerated laboratory schedule. A specimen is held at a sequence of progressively lower temperatures within the embrittling range for extended periods, then furnace cooled between steps, before Charpy impact testing is used to compare the transition temperature before and after the treatment.
Step 1: 593°C, hold 1 hour, furnace cool Step 2: 538°C, hold 15 hours, furnace cool Step 3: 524°C, hold 24 hours, furnace cool Step 4: 496°C, hold 48 hours, furnace cool Step 5: 468°C, hold 72 hours, furnace cool Step 6: 343°C, hold 70 hours, furnace cool Report: shift in Charpy transition temperature (delta FATT or delta DBTT) between as-received and step-cooled condition
Temper Embrittlement in Creep-Resistant Cr-Mo Steels
Temper embrittlement has historically been most significant in two families of heavy-section, long-service-life components: 2.25Cr-1Mo pressure vessel steel used in petrochemical hydrocracker and hydrotreater reactors, and NiCrMoV low-pressure steam turbine rotor forgings. Both applications combine large section thickness (which inherently produces slow cooling rates through the critical range during heat treatment), decades of intended service, and, for reactors, sustained operating temperatures that fall partly within or near the embrittling range — a combination that makes both J-factor/X-factor screening and step-cooling qualification standard practice in their material specifications.
Modern 9Cr-1Mo modified steels, commonly known by their ASME designations P91 and P92, are melted to considerably tighter impurity limits than older Cr-Mo grades, specifically to control temper embrittlement risk alongside other property targets. In practice, however, the dominant long-term degradation concerns reported for P91/P92 weldments in creep service are Type IV cracking (premature creep failure in the fine-grained HAZ) and general creep cavitation, rather than classical temper embrittlement, reflecting how effectively modern steelmaking has controlled the impurity levels that drive it. Even so, impurity limits and post-weld heat treatment practice for these grades continue to be specified with temper embrittlement susceptibility explicitly in mind, and step-cooling or J-factor screening remains part of qualification testing for critical service.
| Steel Family | Typical Application | TE Relevance |
|---|---|---|
| 2.25Cr-1Mo (and 3Cr-1Mo) | Hydrocracker/hydrotreater reactor vessels | Classical, well-documented TE risk; heavy section, long service life |
| NiCrMoV rotor steel | Steam turbine low-pressure rotor forgings | High Ni and Cr content raise susceptibility; large forgings cool slowly |
| P91 / P92 (9Cr-1Mo modified) | High-temperature/high-pressure power piping and headers | Tight impurity limits reduce TE risk; Type IV cracking is the more prominent long-term concern |
Prevention and Mitigation
- Specify low residual impurity content (P, Sn, Sb, As) at the material procurement stage through clean steelmaking practice, particularly for thick sections and long service-life components.
- Favour molybdenum-bearing compositions, consistent with strength and creep requirements, since Mo directly suppresses segregation kinetics.
- Cool rapidly through the critical temperature range after tempering or post-weld heat treatment wherever the process and section size allow, rather than furnace cooling slowly through it.
- Where service temperature falls within or near the critical range for extended periods, factor temper embrittlement into inspection planning and periodic toughness verification over the component’s operating life.
- Use J-factor/X-factor screening at the specification stage and step-cooling qualification testing for critical, heavy-section, or long-life applications.
Frequently Asked Questions
What is temper embrittlement?
Does temper embrittlement change the hardness or strength of steel?
Which alloying elements increase susceptibility to temper embrittlement?
What is the Watanabe J factor?
What is the Bruscato X factor?
Is temper embrittlement reversible?
How is temper embrittlement susceptibility tested?
Is temper embrittlement a concern for P91 and P92 creep-resistant steels?
What is the difference between temper embrittlement and 475 degree C embrittlement?
How can temper embrittlement be prevented?
Recommended Reference Books
ASM Handbook, Volume 4: Heat Treating
Covers temper embrittlement mechanisms alongside conventional tempering practice for alloy steels.
View on AmazonASM Handbook, Volume 19: Fatigue and Fracture
Reference data on intergranular embrittlement, Charpy transition testing, and toughness degradation mechanisms.
View on AmazonSteels: Processing, Structure, and Performance by G. Krauss
Authoritative treatment of temper embrittlement, tempered martensite embrittlement, and related phenomena.
View on AmazonCreep-Resistant Steels (Woodhead Publishing)
Focused reference on Cr-Mo and 9-12% Cr power plant steels, including P91/P92 metallurgy and degradation.
View on AmazonDisclosure: MetallurgyZone participates in the Amazon Associates programme. If you purchase through these links, we may earn a small commission at no extra cost to you. This helps support free technical content on this site.