Determination of High-Temperature Fatigue Properties of Alloys

High-temperature fatigue testing of alloys quantifies how metallic materials resist cyclic mechanical loading at elevated temperatures, producing the stress-life, strain-life, and crack-growth data required for the design and life assessment of critical components in aerospace, power generation, and petrochemical engineering. This article explains the specimens, measured parameters, test methods, applicable international standards, and typical applications involved in determining the high-temperature fatigue properties of alloys.

Technical Overview

High-temperature fatigue is the progressive, localized structural damage that develops when an alloy is subjected to repeated cyclic loading at elevated temperatures, typically above 0.3 to 0.4 times its absolute melting point, where time-dependent mechanisms such as creep and oxidation interact with time-independent cyclic plasticity. Determining the high-temperature fatigue properties of alloys is essential because components such as turbine disks, blades, heat exchanger tubes, and pressure vessels operate under combined thermal and mechanical cycling, and failures frequently initiate at stresses well below the static yield strength of the material.

Unlike fatigue at room temperature, high-temperature fatigue behavior depends strongly on test frequency, waveform, dwell periods at peak load or strain, environment, and mean stress. Oxidation accelerates crack initiation at free surfaces, while creep deformation during hold times promotes intergranular cavitation damage. A complete characterization therefore requires systematic testing under controlled conditions that reproduce the service temperature, strain range, and cycle shape experienced by the component, typically performed by accredited third-party laboratories using servo-hydraulic or electrodynamic testing systems equipped with furnaces or induction heating in accordance with ISO, ASTM, EN, GB, and JIS standards.

Testing Samples

Material Forms and Alloys

Specimens are machined from nickel-based superalloys, cobalt-based alloys, titanium alloys, stainless steels, creep-resistant heat-resistant steels, aluminum alloys, and emerging high-entropy alloys in wrought bar, cast, powder metallurgy, or additively manufactured conditions. Material orientation and heat treatment condition must be documented, because fatigue properties are markedly anisotropic in forged, rolled, and directionally solidified products.

Standard Specimen Geometries

• Hourglass specimens with a continuous-radius gauge section, typically 6 to 10 mm in minimum diameter, used for force-controlled high-cycle fatigue tests.
• Uniform-gauge cylindrical specimens with parallel gauge sections used for strain-controlled low-cycle fatigue tests.
• Flat or plate specimens for sheet materials and welded joints.
• Compact tension (CT) or single-edge-notched (SEN) specimens for fatigue crack growth testing.

Gauge-section surfaces are polished longitudinally to a roughness of typically Ra 0.2 to 0.4 micrometers or better, because machining marks act as crack initiators and bias measured lives. Dimensions, surface condition, and residual stress state are verified and reported before testing.

Welded Joints and Component-Specific Samples

For welded structures such as pipework, reactor shells, and combustor casings, specimens are extracted across weld seams so that failure can occur in the base metal, heat-affected zone, or weld metal, allowing the weakest region of the joint to be identified at temperature.

Testing Items

• Stress-life (S-N) curves and high-cycle fatigue strength at 10^6 to 10^8 cycles at the specified temperature.
• Strain-life (epsilon-N) curves, including elastic and plastic strain components and Coffin-Manson parameters.
• Cyclic stress-strain curves and cyclic hardening or softening behavior.
• Creep-fatigue interaction life under strain-hold and stress-hold dwell conditions.
• Thermomechanical fatigue (TMF) life under in-phase and out-of-phase temperature-strain cycling.
• Fatigue crack growth rates (da/dN versus delta-K) and threshold stress-intensity range.
• Creep-fatigue crack growth behavior under cyclic loading with hold times.
• Mean stress and stress ratio (R) sensitivity.
• Fractographic examination of initiation sites, crack propagation modes, and damage mechanisms by scanning electron microscopy.

Testing Methods

Force-Controlled High-Cycle Fatigue Testing

Constant-amplitude axial fatigue tests are performed per ASTM E466, ISO 1099, and GB/T 3075. Specimens are cycled at stress ratios such as -1, 0.1, or 0.5, with sinusoidal waveforms at frequencies from about 0.1 to 20 Hz at elevated temperature, chosen to avoid specimen self-heating. Temperature is maintained within plus or minus 2 to 3 degrees Celsius of the setpoint using resistance furnaces or high-frequency induction coils, verified by calibrated thermocouples attached to the gauge section. S-N curves are generated from replicate specimens at several stress levels, and the fatigue strength corresponding to a run-out life such as 10^7 cycles is established.

Strain-Controlled Low-Cycle Fatigue Testing

Low-cycle fatigue tests follow ASTM E606, ISO 12106, and GB/T 15248. Total strain amplitudes between roughly 0.2 and 2.0 percent are applied as fully reversed triangular waveforms at strain rates of approximately 10^-3 to 10^-2 per second, with strain measured and controlled by high-temperature extensometers using quartz or ceramic rods. Hysteresis loops are recorded continuously to track the evolution of stress amplitude, which reveals cyclic hardening or softening. Failure is defined by complete fracture or by a specified drop (commonly 10 to 50 percent) in peak tensile stress. The data are fitted using the Basquin and Coffin-Manson relationships to characterize elastic and plastic strain contributions to life.

Creep-Fatigue Interaction Testing

Dwell periods at peak tensile strain, lasting from tens of seconds to many hours, are introduced into the strain cycle to reproduce the conditions of thick-section components. Stress relaxation and creep strain accumulation during holds are recorded, and the resulting life reduction relative to continuous cycling quantifies the creep-fatigue interaction. Data reduction may employ strain-range partitioning or frequency-modified damage models.

Thermomechanical Fatigue Testing

Specimens are cycled simultaneously in temperature, for example between 200 and 950 degrees Celsius, and in mechanical strain, either in-phase (maximum strain at maximum temperature) or out-of-phase, in accordance with ASTM E2368 and ISO 12111. Induction or quartz-lamp heating with programmable temperature controllers provides controlled heating and cooling rates, while the control system compensates for thermal expansion. TMF testing is the most realistic laboratory simulation of start-up and shut-down cycling in engines and is widely used for hot-section component qualification.

Fatigue Crack Growth Testing

Pre-cracked CT or SEN specimens are cycled under constant load amplitude or decreasing delta-K conditions at the service temperature per ASTM E647 and ISO 12108. Crack length is monitored using compliance, direct current potential drop, or optical methods, and Paris-law constants together with the threshold stress-intensity range are determined. Creep-fatigue crack growth can additionally be characterized using hold times in accordance with ASTM E1457.

Equipment, Calibration, and Quality Assurance

Closed-loop servo-hydraulic or electrodynamic frames are used with load cells verified per ISO 7500-1 or ASTM E4. Load-train alignment is verified per ASTM E1012 so that bending strain remains below approximately 5 percent of the applied axial strain range, since misalignment severely shortens measured fatigue life. Furnace temperature profiles are surveyed with at least three calibrated thermocouples, and extensometer calibration follows ASTM E83, with Class B-1 or better typically required for strain-controlled testing.

Data Analysis and Deliverables

• S-N and strain-life data plotted on logarithmic coordinates with statistical treatment of scatter.
• Cyclic stress-strain curves and hysteresis loop evolution throughout the test.
• Stress relaxation and creep strain profiles recorded during dwell periods.
• Crack growth rate curves with fitted Paris-regime parameters and thresholds.
• Scanning electron microscopy fractography distinguishing transgranular fatigue striations, intergranular creep cavitation, and oxidation-assisted crack initiation.

Applications

• Aerospace: life assessment and qualification of turbine disks, blades, combustor liners, and casings in aero engines, including additively manufactured parts.
• Power generation: integrity evaluation of gas turbine hot-section components, steam turbine rotors and blades, and boiler components in fossil and combined-cycle plants.
• Nuclear energy: generation of fatigue design curves and aging management data for reactor internals, piping, and fuel cladding operating at 300 to 650 degrees Celsius.
• Petrochemical: fitness-for-service evaluation of reformer tubes, reactor vessels, and catalytic units subject to thermal cycling.
• Automotive: durability validation of turbocharger wheels, exhaust manifolds, and cylinder-head materials.
• Materials development: ranking of new superalloys, high-entropy alloys, coatings, and welding processes for high-temperature cyclic service.

FAQ

What temperature range can be covered in high-temperature fatigue testing?

Resistance furnaces typically support testing from ambient up to approximately 1,100 degrees Celsius, while induction and infrared heating can extend the range to about 1,400 degrees Celsius or higher, optionally under vacuum or inert gas atmospheres to suppress oxidation when required.

How are creep-fatigue interactions evaluated?

Hold times are inserted at peak strain, tests are run at reduced strain rates, or TMF cycles are applied. The resulting lives are compared with continuous-cycling baselines and interpreted using strain-range partitioning, frequency-modified, or energy-based cumulative damage models.

How long does a typical test program take?

Individual strain-controlled tests may run from several hours to several weeks depending on strain range and hold time, while high-cycle tests to 10^7 cycles at reduced frequencies can take days per specimen. A complete S-N or strain-life campaign with replication and statistical analysis generally requires several weeks to a few months.

Which standards govern high-temperature fatigue testing of alloys?

The principal standards include ASTM E466, ASTM E606, ASTM E2368, ASTM E647, ASTM E1457, ASTM E1012, ISO 1099, ISO 12106, ISO 12111, ISO 12108, and the corresponding GB/T and JIS national standards series.

Why is strain-controlled testing preferred at high temperature?

At elevated temperatures, stress relaxation and creep redistribute the stress state, so the strain range, rather than the stress range, remains the controllable and physically meaningful quantity that governs the deformation and damage accumulated in each cycle.