What is Temper Embrittlement in Steels? Causes, Assessment and Prevention
Temper embrittlement is one of the most insidious degradation mechanisms in low-alloy pressure vessel steels because it does not announce itself through visible cracking or dimensional change. Instead, the steel gradually loses toughness while retaining its hardness and tensile strength, only revealing its embrittled condition when subjected to impact loading or startup thermal shock. Understanding the mechanism, quantifying the risk through established compositional indices, and applying the correct test protocols are essential skills for any engineer working with Cr-Mo reactors, hydroprocessing vessels, or high-temperature alloy steel fabrications.
This guide covers the metallurgical mechanism behind temper embrittlement, the two industry-standard assessment parameters — the Watanabe J factor and the Bruscato X factor — the step-cooling test procedure, the acceptance criterion used in pressure vessel codes, and practical strategies for prevention during steelmaking, welding, and heat treatment. References to ASTM A387 and WRC Bulletin 275 are included throughout.
For engineers working on P91 and P92 Cr-Mo steels, temper embrittlement risk assessment is not optional — it is a procurement and qualification requirement. The compositional limits described below are directly traceable to documented service failures in refinery hydrocracking reactors and steam-reformer equipment.
The Mechanism of Temper Embrittlement
Temper embrittlement is a solid-state phenomenon driven by the preferential segregation of specific impurity elements from the bulk steel matrix to austenite grain boundaries. During the steel’s original heat treatment (austenitising and quenching), these impurities are temporarily dissolved in a supersaturated solution within the grains. When the steel is subsequently reheated to — or slowly cooled through — the critical temperature range of approximately 350 to 575 °C (660 to 1065 °F), thermodynamic driving forces cause the impurity atoms to diffuse toward and concentrate at grain boundaries.
The elements primarily responsible are phosphorus (P), antimony (Sb), tin (Sn), and arsenic (As). Manganese (Mn) and silicon (Si) play a secondary but significant role: they promote the segregation of the primary embrittlers rather than causing embrittlement directly. The effect of these combined elements is captured by the J and X factor formulas described below.
Grain Boundary Cohesion Reduction
Once segregated at grain boundaries, the impurity atoms reduce the interatomic bonding energy between adjacent grains. This lowers the fracture energy required to propagate an intergranular crack. Under impact loading or at low temperature, the material fails by intergranular brittle fracture rather than by the ductile transgranular mechanism seen in un-embrittled steel. The macroscopic result is a marked upward shift in the Charpy V-notch ductile-to-brittle transition temperature (DBTT) and a corresponding reduction in upper-shelf impact energy.
Importantly, embrittlement is reversible at high temperatures. Heating above approximately 600 °C dissolves the segregated layer, and rapid cooling from that temperature prevents re-segregation. This thermal reversal is the metallurgical basis for de-embrittlement heat treatment.
Assessing Susceptibility: The J Factor and X Factor
Two quantitative compositional parameters have been developed to allow engineers to predict and screen for temper embrittlement risk at the procurement stage. Both are routinely specified in material purchase orders for alloy steel pressure vessel plates, forgings, and weld consumables. For a thorough discussion of the Bruscato parameter, see the dedicated X-factor (Bruscato factor) calculator and guide on WeldFabWorld.
1. The Watanabe J Factor (Parent Steel and Weld Metal)
The J factor, developed by Watanabe et al. from research on 2.25Cr-1Mo pressure vessel steels, correlates the concentrations of the secondary promoters (Mn and Si) and the primary embrittlers (P and Sn) into a single index:
J = (Mn + Si)(P + Sn) × 104
All elements expressed in weight percent (wt%)
Acceptance criterion:
J ≤ 180 → Low susceptibility to temper embrittlement
J > 180 indicates increased risk; specify J ≤ 180 at procurement
The multiplication structure of the formula captures the synergistic effect: even if Mn and Si are individually within normal ranges, a higher-than-usual P or Sn content will elevate J significantly. This is why vacuum-degassed and ladle-refined steels with tightly controlled tramp element content are preferred for vessels in embrittling service.
2. The Bruscato X Factor (Weld Metal)
The X factor applies specifically to weld metal deposits and includes arsenic (As) in addition to P, Sb, and Sn, with individually weighted coefficients reflecting their relative potency as grain boundary embrittlers:
X = (10P + 5Sb + 4Sn + As) / 100
All elements expressed in parts per million (ppm)
Acceptance criterion:
X < 20 ppm → Acceptable resistance to weld metal temper embrittlement
X ≥ 20 ppm warrants rejection or additional testing
Worked Example: J and X Factor Calculations
| Parameter | Steel A (Vacuum-Degassed) | Steel B (Standard Melt) |
|---|---|---|
| Mn (wt%) | 0.60 | 0.88 |
| Si (wt%) | 0.10 | 0.25 |
| P (wt%) | 0.007 | 0.015 |
| Sn (wt%) | 0.005 | 0.012 |
| J Factor | (0.70)(0.012) × 104 = 84 | (1.13)(0.027) × 104 = 305 |
| Assessment | PASS J ≤ 180 | FAIL J > 180 |
| Element | Weld Metal A (ppm) | Coefficient | Contribution A | Weld Metal B (ppm) | Contribution B |
|---|---|---|---|---|---|
| Phosphorus (P) | 60 | 10 | 600 | 130 | 1300 |
| Antimony (Sb) | 4 | 5 | 20 | 12 | 60 |
| Tin (Sn) | 30 | 4 | 120 | 80 | 320 |
| Arsenic (As) | 50 | 1 | 50 | 120 | 120 |
| Sum / 100 = X factor | 7.9 ppm | 18.0 ppm | |||
| Assessment | PASS X < 20 | MARGINAL — Verify | |||
The Step-Cooling Test
Compositional parameters provide a screening tool, but they cannot fully account for all microstructural variables. The step-cooling test provides a direct, empirical measure of a specific steel or weld metal’s susceptibility by simulating accelerated in-service embrittlement in the laboratory. It is particularly valuable for qualifying materials that fall near the J or X factor limits, or for confirming compliance in thick-section pressure vessel applications.
Principle
The test uses a controlled multi-step heat treatment — a “staircase” of temperatures and hold times — designed to recreate the equilibrium grain boundary segregation state that would accumulate over many years of service in the embrittling temperature range. Because impurity diffusion is thermally activated, a series of holds at progressively lower temperatures within the 350–575 °C range, each of sufficient duration, produces a state equivalent to prolonged low-temperature exposure.
Step-Cooling Cycle (ASTM A387 Supplementary Requirements)
| Step | Temperature (°C) | Temperature (°F) | Hold Time (hours) |
|---|---|---|---|
| 1 | 595 | 1100 | 1 |
| 2 | 540 | 1005 | 15 |
| 3 | 510 | 950 | 24 |
| 4 | 480 | 895 | 60 |
| 5 | 455 | 850 | 100 |
| Furnace cool to RT | — | — | — |
Charpy Testing and Transition Temperature Shift
Charpy V-notch (CVN) impact specimens are tested in both the as-formed (AF) condition and the step-cooled (SC) condition across a range of temperatures. From each dataset, the temperature at which the material achieves 54 J (40 ft-lb) impact energy is determined — this is the Charpy 54 J transition temperature. The upward shift in transition temperature between AF and SC states is the primary measure of embrittlement susceptibility.
The Erwin-Kerr Acceptance Criterion
The transition temperature shift alone does not constitute a pass/fail criterion. The shift must be evaluated in combination with the baseline toughness of the material using the expression developed by Erwin and Kerr (WRC Bulletin 275, 1982). This criterion is used in the design and qualification of pressure vessels that may operate in the embrittling temperature range or pass slowly through it during startup and shutdown:
AF + 2.5(SC − AF) < 38 °C
Where:
AF = As-formed Charpy 54 J transition temperature (°C)
SC = Step-cooled Charpy 54 J transition temperature (°C)
Rearranged:
−1.5 × AF + 2.5 × SC < 38
Example — Material Passes:
AF = −20 °C, SC = +10 °C
−20 + 2.5(10 − (−20)) = −20 + 75 = 55 °C
55 > 38 → this material FAILS
Example — Material Passes:
AF = −30 °C, SC = −5 °C
−30 + 2.5(−5 − (−30)) = −30 + 62.5 = 32.5 °C
32.5 < 38 → PASSES acceptance criterion
The factor of 2.5 in the expression projects the measured laboratory step-cooled condition forward to the worst-case long-term in-service embrittlement state. A material with a higher initial toughness (more negative AF temperature) is better able to absorb a given shift in SC and still remain within the 38 °C limit. This is why specifying low baseline transition temperatures at procurement is just as important as controlling the J and X factors.
Steels and Applications Most Affected
While temper embrittlement is theoretically possible in any alloy steel containing the relevant impurities, it is a primary design concern in the following situations:
| Steel Grade | Composition | Typical Application | TE Risk Level |
|---|---|---|---|
| 2.25Cr-1Mo (P22 / Grade 22) | 2.25Cr, 1Mo | Hydroprocessing reactors, boiler headers | High |
| 2.25Cr-1Mo-0.25V | 2.25Cr, 1Mo, 0.25V, Cb | High-pressure hydrogen reactors | High |
| 1.25Cr-0.5Mo (P11 / Grade 11) | 1.25Cr, 0.5Mo | Steam pipework, pressure vessels | Moderate |
| 3.5Ni-1.75Cr-0.5Mo-0.1V | 3.5Ni, 1.75Cr | Low-temperature pressure vessels | High |
| P91 / Grade 91 (9Cr-1Mo-V) | 9Cr, 1Mo, V, Nb | Power plant steam headers, high-temp piping | Moderate (X factor required) |
| Mn-Mo-Ni steels (ASTM A533) | Mn, Mo, Ni | Nuclear reactor pressure vessels | Moderate |
For engineers working on P91 welding procedures, controlling the X factor in weld metal is a direct requirement in many owner-engineer specifications, even though P91 sees service above the embrittling range. This is because startup and shutdown cycles can place thick-wall components in the critical zone for extended periods.
Post-Weld Heat Treatment and Temper Embrittlement
Post-weld heat treatment (PWHT) of Cr-Mo steels typically takes place at temperatures of 680–750 °C, which is above the embrittling range. The risk arises during cooling from the PWHT temperature. If the cooling rate through 350–575 °C is slow — as is common in thick-section pressure vessels or large forgings cooled in a furnace — significant embrittlement can accumulate during that single thermal cycle.
The heat treatment guide on WeldFabWorld provides detailed guidance on controlled cooling rates for Cr-Mo steels. As a general rule, cooling rates above approximately 55 °C/hour through the critical range are preferred for embrittlement-sensitive applications, though exact requirements depend on section thickness, composition, and the applicable engineering specification.
Relationship to Reheat Cracking
Temper embrittlement and reheat cracking in welds share overlapping root causes. In both phenomena, grain boundary embrittlement by impurity segregation and the precipitation of alloy carbides play central roles. The difference lies in the service temperature: reheat cracking occurs at elevated PWHT temperatures where grain boundary films become fully liquid or semi-solid, while temper embrittlement operates at lower temperatures through solid-state diffusion. Reducing impurity levels in both base and weld metal therefore provides protection against both mechanisms simultaneously.
Prevention and Mitigation Strategies
1. Steelmaking and Composition Control
The most effective long-term defence against temper embrittlement is minimising the impurity content of the steel at the point of production. Modern electric arc furnace and ladle metallurgy routes, combined with vacuum degassing, routinely achieve phosphorus levels below 0.010 wt%, antimony below 5 ppm, tin below 15 ppm, and arsenic below 25 ppm in clean-steel practice. Specifying J ≤ 150 (instead of the standard limit of 180) provides a further safety margin for the most critical applications.
2. Weld Consumable Selection
Use low-impurity weld filler materials for Cr-Mo applications. Review certificates of conformance for heat chemistry, and request additional analysis for P, Sb, Sn, and As if the application warrants it. Calculate the X factor and verify compliance before committing to a consumable lot. The welding consumable nomenclature guide explains how to read and interpret test certificates for different electrode classifications.
3. Controlled Cooling After PWHT
Avoid slow continuous cooling through 350–575 °C. Where possible, use step-cooling simulation data to set minimum cooling rates specific to the heat of steel being processed. For pressure vessel components, document and record cooling thermocouple data as part of the heat treatment record.
4. Minimum Pressurisation Temperature (MPT)
For vessels already in service that may have undergone some degree of embrittlement, the primary engineering control is the minimum pressurisation temperature (MPT). The vessel is not allowed to be pressurised beyond a specified fraction of design pressure until the shell temperature exceeds a defined MPT. This ensures the material is sufficiently tough to resist brittle fracture initiation during start-up, even if some grain boundary segregation has occurred.
5. De-Embrittlement Treatment
Because grain boundary segregation is thermodynamically reversible, a short heat treatment above 600 °C, followed by rapid cooling, will dissolve the segregated layer and restore toughness. This de-embrittlement heat treatment is occasionally specified for vessels that have experienced extended service in the embrittling range, prior to fitness-for-service evaluation. The procedure must be carefully designed to avoid new thermal damage and must be followed by re-testing to confirm toughness restoration.
- Specify J ≤ 180 (or J ≤ 150 for critical service) on the material purchase order
- Specify X < 20 ppm for weld consumable lot qualification
- Require step-cooling test and Erwin-Kerr criterion check for vessels operating in the 350–575 °C range
- Define minimum cooling rate through the embrittling range in the PWHT procedure
- Establish MPT limits in the vessel design documentation
- Retain full tramp element chemistry data for each plate and forging heat
Relevant Standards and References
| Document | Relevance to Temper Embrittlement |
|---|---|
| ASTM A387 | Standard specification for Cr-Mo alloy steel pressure vessel plates; includes supplementary requirements for step-cooling test and transition temperature shift |
| WRC Bulletin 275 (Erwin & Kerr, 1982) | Foundational research document establishing the AF + 2.5(SC − AF) < 38 °C criterion for 2.25Cr-1Mo vessels |
| ASTM STP 755 (Watanabe et al., 1980) | Source of the J factor formula for 2.25Cr-1Mo steels |
| API 934-A | Materials and fabrication of 2.25Cr-1Mo and 2.25Cr-1Mo-V steel heavy-wall vessels for high-temperature, high-pressure hydrogen service; references J and X factor limits |
| ASME Section VIII Div. 1 / Div. 2 | Pressure vessel design code; references impact testing requirements relevant to embrittled material assessment |
| EN 10028-2 | European specification for weldable fine-grain pressure vessel steels; includes Cr-Mo grades with J factor controls |
Recommended Books on Alloy Steel Metallurgy and Pressure Vessel Design
Frequently Asked Questions
What is temper embrittlement?
Temper embrittlement is a loss of toughness and impact resistance in alloy steels caused by exposure to a critical temperature range, typically 350–575 °C (660–1065 °F). It results from the segregation of trace impurity elements such as phosphorus, antimony, tin, and arsenic to austenite grain boundaries during slow cooling or prolonged hold in this range. The steel may appear hard and strong, yet fail by brittle fracture at operating or start-up temperatures. The phenomenon is reversible by heating above 600 °C and quenching, which redissolves the segregated layer.
What is the J factor and how is it calculated?
The Watanabe J factor is a compositional parameter used to assess temper embrittlement susceptibility in Cr-Mo parent steels and weld metals. It is calculated as J = (Mn + Si)(P + Sn) × 104, where all elements are expressed in weight percent. A J factor at or below 180 indicates low susceptibility to embrittlement. Higher values suggest the steel is prone to grain boundary embrittlement. Specifying J ≤ 180 in the material purchase order is standard practice for hydroprocessing vessel procurement.
What is the Bruscato X factor?
The Bruscato X factor is applied specifically to weld metals and is defined as X = (10P + 5Sb + 4Sn + As) / 100, where all elements are in parts per million (ppm). A value below 20 ppm indicates acceptable resistance to temper embrittlement. The X factor is particularly important when qualifying weld consumables for pressure vessels operating in the embrittling temperature range. For a full calculator and derivation, see the dedicated X-factor guide on WeldFabWorld.
How does the step-cooling test work?
The step-cooling test subjects steel specimens to a controlled, multi-step heat treatment that simulates many years of in-service exposure to the embrittling temperature range in a matter of days. Charpy impact tests are performed on specimens in both the as-formed condition and the step-cooled condition across a range of temperatures. The shift in 54 J transition temperature between the two states is used to evaluate toughness retention. The test is described in ASTM A387 supplementary requirements, and the shift is evaluated against the Erwin-Kerr criterion to determine acceptance.
What is the acceptance criterion for temper embrittlement in pressure vessels?
For pressure vessels that may operate in or pass through the embrittling temperature range, the criterion AF + 2.5(SC − AF) < 38 °C must be satisfied, where AF is the as-formed Charpy 54 J transition temperature and SC is the step-cooled 54 J transition temperature. This expression, developed by Erwin and Kerr (WRC Bulletin 275), ensures the material retains acceptable toughness throughout its service life. The factor 2.5 accounts for projecting the laboratory step-cooled state to the maximum expected in-service embrittlement condition.
Which steels are most susceptible to temper embrittlement?
Low-alloy Cr-Mo steels are the primary concern, particularly 2.25Cr-1Mo (Grade 22), 1.25Cr-0.5Mo, and 3.5Ni-1.75Cr-0.5Mo-0.1V grades used in petrochemical reactors and hydroprocessing vessels. Manganese-silicon steels and nickel-chromium steels can also be susceptible. P91 steels are a moderate risk, primarily during startup and shutdown cycling. Steels with very low phosphorus, antimony, tin, and arsenic contents achieved through vacuum degassing are significantly more resistant.
How is temper embrittlement prevented or mitigated?
Prevention relies primarily on steelmaking practice: vacuum degassing, ladle refining, and tight compositional control to minimise P, Sb, Sn, and As. Specifying J ≤ 180 and X < 20 at procurement are the first line of defence. During fabrication, controlling the cooling rate through 350–575 °C after PWHT avoids additional embrittlement from the heat treatment itself. In service, minimum pressurisation temperature (MPT) limits prevent brittle fracture during cold startups. De-embrittlement heat treatment (heating above 600 °C followed by rapid cooling) can restore toughness in embrittled components.
How is temper embrittlement related to reheat cracking?
Temper embrittlement and reheat cracking share a common mechanism: grain boundary weakening by impurity segregation and carbide precipitation. In Cr-Mo steels, the same elements responsible for temper embrittlement (P, Sb, Sn, As) can also reduce grain boundary ductility during post-weld heat treatment, contributing to stress-relief cracking. Managing impurity levels and controlling PWHT heating and cooling rates therefore addresses both failure modes simultaneously. Mechanical testing including CVN impact tests after PWHT provides direct confirmation that neither mechanism has compromised the joint.
References
- Ishiguo, T.; Murakami, Y.; Ohnishi, K. and Watanabe, J.: “2.25%Cr-1%Mo pressure vessel steels with improved creep rupture strength”; Proceedings of the symposium on Applications of 2.25%Cr-1%Mo steel for thick-wall pressure vessels, ASTM STP 755, 1980, pp. 129–147.
- Bruscato, R. M.: “Embrittlement factors for estimating temper embrittlement in 2.25Cr:1Mo, 3.5Ni-1.75Cr-0.5Mo-0.1V and 3.5Ni steels”; ASTM conference, Miami, Florida, 1987.
- Erwin, W.E. and Kerr, J.G.: “The use of Quenched and Tempered 2¼Cr-1Mo Steel for Thick Wall Reactor Vessels in Petroleum Refinery Processes: An Interpretive review of 25 Years of Research and Application”; WRC Bulletin 275, February 1982.
- ASTM A387: “Standard specification for pressure vessel plates, Alloy steel, Chromium-molybdenum”.
- API 934-A: “Materials and Fabrication of 2-1/4Cr-1Mo and 3Cr-1Mo Steel Heavy Wall Pressure Vessels for High-Temperature, High-Pressure Hydrogen Service”.