The relentless drive towards increased fuel efficiency and stringent emission regulations in internal combustion engines has led to progressively severe operating conditions for critical components. This is particularly evident during transient events like start-up and shut-down, where components such as exhaust manifolds and turbocharger housings experience significant thermal and mechanical strain cycles, making them susceptible to thermomechanical fatigue failure. Consequently, a thorough understanding of the stress-strain response of materials used in these high-temperature applications under both static and cyclic loading across their entire service temperature spectrum is paramount for accurate durability assessment and lightweight design optimization.

High-silicon molybdenum nodular cast iron, often referred to as ductile cast iron (DCI), is a material of choice for these demanding applications due to its favorable combination of castability, thermal stability, and mechanical properties. A prominent grade is QTRSi4Mo1, characterized by approximately 4% Si and 1% Mo, enabling service temperatures up to 760°C. While the mechanical properties of various nodular cast iron grades have been studied at room and elevated temperatures, detailed investigations into their cyclic stress-strain response over a broad temperature range, which directly simulates the transient thermal stresses encountered in service, are less common. This work presents a comprehensive study on the cast nodular cast iron QTRSi4Mo1, investigating the influence of temperature from ambient to 760°C on its tensile properties and, more critically, its strain-controlled low-cycle fatigue behavior. The aim is to elucidate the evolution of strength, ductility, and cyclic deformation characteristics with temperature, providing essential data and constitutive descriptions for the fatigue life assessment of engine components.
Material and Experimental Methodology
The material under investigation is a ferritic nodular cast iron, QTRSi4Mo1. Test specimens for both static tensile and strain-controlled fatigue tests were machined from as-cast blocks. The chemical composition of the material is detailed in Table 1.
| C | Si | Mo | Mg | Mn | S | P | Fe |
|---|---|---|---|---|---|---|---|
| 3.18 | 4.43 | 1.23 | 0.03 | 0.31 | 0.008 | 0.027 | Bal. |
Uniaxial tensile tests were conducted at a constant strain rate of $$2.5 \times 10^{-4} \, s^{-1}$$ at temperatures of 25°C (Room Temperature, RT), 200°C, 300°C, 400°C, 500°C, 600°C, 700°C, and 760°C. Strain-controlled low-cycle fatigue (LCF) tests were performed under fully reversed axial strain (strain ratio, R = -1) using a triangular waveform at a constant strain rate of $$5 \times 10^{-3} \, s^{-1}$$. The tests were conducted at temperatures of 25°C, 200°C, 400°C, 500°C, and 760°C with various total strain amplitudes, as summarized in Table 2. The fatigue life, $$N_f$$, was defined as the number of cycles to specimen fracture or a 5% drop in the maximum tensile stress.
| Temperature (°C) | Total Strain Amplitude, $\Delta\varepsilon/2$ (%) | Strain Ratio, R | Strain Rate ($s^{-1}$) |
|---|---|---|---|
| 25 | 0.25 | -1 | $5 \times 10^{-3}$ |
| 0.30 | |||
| 0.40 | |||
| 0.50 | |||
| 200 | 0.25 | -1 | |
| 0.30 | |||
| 0.40 | |||
| 400 | 0.25 | -1 | |
| 0.30 | |||
| 0.40 | |||
| 500 | 0.15 | -1 | |
| 0.25 | |||
| 0.40 | |||
| 760 | 0.15 | -1 | |
| 0.30 | |||
| 0.40 |
Influence of Temperature on Tensile Properties
The results from the uniaxial tensile tests are compiled in Table 3. To clearly illustrate the temperature dependence, key properties—yield strength (0.2% offset, $\sigma_{0.2}$), ultimate tensile strength (UTS, $\sigma_u$), elongation to fracture ($\delta$), and reduction of area ($\psi$)—are normalized with respect to their room temperature values and plotted in Figure 1.
| Temperature (°C) | Yield Strength, $\sigma_{0.2}$ (MPa) | Ultimate Tensile Strength, $\sigma_u$ (MPa) | Elongation, $\delta$ (%) | Reduction of Area, $\psi$ (%) |
|---|---|---|---|---|
| 25 | 500 | 627 | 13.5 | 7.7 |
| 200 | 451 | 588 | 8.7 | 7.7 |
| 300 | 406 | 558 | 8.6 | 4.2 |
| 400 | 386 | 525 | 11.4 | 11.6 |
| 500 | 306 | 325 | 18.9 | 21.1 |
| 600 | 144 | 157 | 31.3 | 34.8 |
| 700 | 63 | 70 | 36.8 | 44.7 |
| 760 | 39 | 42 | 38.4 | 45.8 |
The strength properties, $\sigma_{0.2}$ and $\sigma_u$, exhibit a continuous decline with increasing temperature. This softening behavior is typical for metallic materials as thermally activated processes like dislocation climb and cross-slip become more prevalent. The relationship between yield strength and temperature for this nodular cast iron can be accurately described by a bilinear function, as shown in Figure 2. A clear transition in the slope occurs around 419°C. The piecewise linear relationship can be expressed as:
$$
\sigma_{0.2}(T) = \begin{cases}
481.80 – 0.2624T & \text{for } T \leq 419^\circ \text{C} \\
806.15 – 1.0365T & \text{for } T > 419^\circ \text{C}
\end{cases}
$$
In stark contrast to the monotonic decrease in strength, the ductility parameters ($\delta$ and $\psi$) show a pronounced non-monotonic trend. They initially decrease, reaching a minimum in the temperature range of 300°C to 400°C, before increasing sharply at temperatures above 500°C. This ductility minimum signifies a regime of embrittlement, a phenomenon often reported in certain grades of nodular cast iron and sometimes referred to as “400°C embrittlement.” This embrittlement is critical for component design as it can lead to unexpectedly low fatigue life under thermal or thermomechanical cycling conditions that traverse this temperature range. Above 500°C, the material exhibits significantly improved plastic deformation capability, which is advantageous for accommodating thermal strains but is accompanied by substantially reduced strength.
Cyclic Stress-Strain Response and Deformation Behavior
The cyclic deformation behavior of nodular cast iron QTRSi4Mo1 is complex and highly temperature-dependent. The evolution of the stress amplitude with the number of cycles at different temperatures and strain amplitudes is depicted in Figure 3.
At room temperature (25°C), the material exhibits a triphasic behavior: initial rapid cyclic hardening, followed by a period of gradual cyclic softening, and concluding with a secondary hardening stage until failure. The intermediate softening stage is less pronounced at lower strain amplitudes. At 200°C, a similar sequence is observed, but the initial hardening peak is more dominant, and the subsequent secondary hardening remains below this peak stress level.
This complex behavior transitions to a simpler one at medium temperatures. At both 400°C and 500°C, the nodular cast iron displays persistent cyclic hardening throughout most of the fatigue life. In the 400°C tests, hardening is nearly monotonic, while at 500°C, after a rapid initial increase, the stress amplitude tends to saturate before the final drop due to crack propagation.
At the highest temperature of 760°C, the behavior inverts completely. The material shows clear and continuous cyclic softening from the first cycle onward for all applied strain amplitudes. This is attributed to the dominance of time-dependent deformation mechanisms such as creep and dynamic microstructural changes (e.g., recovery, carbide coarsening), which overcome any dislocation-based hardening mechanisms.
Cyclic Stress-Strain Curves and Masing Behavior
The stable cyclic response is commonly characterized by the cyclic stress-strain curve (CSSC), which relates the saturated stress amplitude to the applied strain amplitude. The CSSC is effectively described by the Ramberg-Osgood relationship:
$$
\frac{\Delta \varepsilon}{2} = \frac{\Delta \varepsilon_e}{2} + \frac{\Delta \varepsilon_{in}}{2} = \frac{\sigma_a}{E} + \left(\frac{\sigma_a}{K’}\right)^{1/n’}
$$
where $\Delta \varepsilon/2$ is the total strain amplitude, $\sigma_a$ is the stable stress amplitude, $E$ is the Young’s modulus, $K’$ is the cyclic strength coefficient, and $n’$ is the cyclic strain hardening exponent. The hysteresis loops at mid-life (approximately half of $$N_f$$) for various strain amplitudes at each temperature are shown in Figure 4, along with the fitted CSSC based on the Ramberg-Osgood equation. The corresponding parameters obtained from the LCF tests are listed in Table 4.
| Temperature (°C) | Young’s Modulus, E (GPa) | Cyclic Strength Coefficient, K’ (MPa) | Cyclic Strain Hardening Exponent, n’ |
|---|---|---|---|
| 25 | 166 | 1231.18 | 0.1417 |
| 200 | 127 | 637.67 | 0.0716 |
| 400 | 114 | 648.19 | 0.0517 |
| 500 | 111 | 618.17 | 0.0778 |
| 760 | 51 | 79.26 | 0.0516 |
The Ramberg-Osgood model provides an excellent fit to the cyclic stress-strain data across all temperatures, confirming its applicability for this nodular cast iron. Two other important observations can be made from the hysteresis loops. First, they exhibit tension-compression asymmetry, with the compressive stress magnitude being higher than the tensile one for a symmetric strain-controlled cycle. This is a common feature in cast irons due to the different resistance of graphite nodules to tensile and compressive loads. Second, the loops’ shape and the material’s conformity to the Masing hypothesis are temperature-dependent.
The Masing behavior is investigated by translating individual hysteresis loops so their compressive tips coincide at the origin. A material is said to exhibit Masing behavior if the ascending branches of all stabilized loops superpose to form a single master curve, which is often described by a modified Ramberg-Osgood equation with a factor of 2:
$$
\Delta \varepsilon = \frac{\Delta \sigma}{E} + 2 \left(\frac{\Delta \sigma}{2K’}\right)^{1/n’}
$$
Analysis reveals that the nodular cast iron displays non-Masing behavior at lower temperatures (25°C, 200°C, 400°C), where the ascending branches do not coincide. However, at higher temperatures of 500°C and 760°C, the loops conform well to the Masing assumption (Figure 5). This transition is linked to changes in the dominant deformation and damage mechanisms with temperature. At lower temperatures, mechanisms like persistent slip band formation, crack initiation at graphite-matrix interfaces, and the complex interplay between matrix hardening and damage accumulation lead to a path-dependent cyclic response. At higher temperatures, where thermal activation is high and inelastic deformation is more homogeneous (influenced by creep and recovery), the material’s memory of prior deformation paths is reduced, leading to Masing-type behavior.
Discussion on Mechanisms and Implications
The observed mechanical behavior of the nodular cast iron QTRSi4Mo1 is a direct consequence of the interaction between its microstructure—comprising a ferritic matrix with embedded spherical graphite nodules—and thermally activated processes.
The embrittlement trough around 300-400°C is a critical finding. This phenomenon in nodular cast iron is often associated with dynamic strain aging (DSA), where solute atoms (like silicon in this case) diffuse to and temporarily pin dislocations during plastic deformation. This leads to an increase in flow stress but a severe reduction in ductility due to localized deformation and easier void nucleation, particularly at the graphite/matrix interface. The change in slope of the yield strength-temperature curve near this region supports the presence of a change in the dominant strengthening/softening mechanism.
The evolution of cyclic hardening/softening can be interpreted as follows:
- Low-Temperature Complex Behavior (25-200°C): The initial rapid hardening is due to dislocation multiplication and interaction within the ferritic matrix. The subsequent softening may be attributed to the rearrangement of dislocations into lower-energy configurations or the initiation of micro-cracks that locally relieve stress. The final secondary hardening could result from crack closure effects or continued matrix hardening in the uncracked ligament.
- Medium-Temperature Hardening (400-500°C): In this range, thermal recovery processes become more active but are not yet dominant enough to cause softening. Dislocation multiplication and interaction still prevail, leading to net cyclic hardening. The embrittlement mechanism might also suppress softening by limiting plastic flow to localized bands.
- High-Temperature Softening (760°C): Here, time-dependent effects dominate. Creep deformation, dynamic recovery, and possibly oxidation at grain boundaries or nodule interfaces lead to a continuous loss of load-bearing capacity, manifesting as cyclic softening from the first cycle.
The tension-compression asymmetry is intrinsic to the composite-like nature of nodular cast iron. Under tension, the graphite nodules act as internal notches and voids, promoting early void formation and lower tensile strength. Under compression, the nodules can withstand more load, and the matrix bears a greater portion of the stress, resulting in a higher flow stress in compression.
Modeling Considerations for Fatigue Life Prediction
For engineering analysis of components made from this nodular cast iron, the temperature-dependent data and models presented are crucial. The cyclic stress-strain curve described by Equation (2) and the parameters in Table 4 can be directly implemented in elastic-plastic finite element analysis to calculate local strains and stresses under thermal-mechanical loading.
Fatigue life prediction often relies on strain-life ($\varepsilon$-$N$) curves, which relate the elastic and plastic strain amplitudes to fatigue life. The total strain amplitude can be partitioned as in the Coffin-Manson relationship:
$$
\frac{\Delta \varepsilon}{2} = \frac{\Delta \varepsilon_e}{2} + \frac{\Delta \varepsilon_{p}}{2} = \frac{\sigma_f’}{E} (2N_f)^b + \varepsilon_f’ (2N_f)^c
$$
where $\sigma_f’$ is the fatigue strength coefficient, $b$ is the fatigue strength exponent, $\varepsilon_f’$ is the fatigue ductility coefficient, and $c$ is the fatigue ductility exponent. These parameters are highly temperature-dependent for this nodular cast iron, especially given the ductility minimum and the shift from cyclic hardening to softening. Life prediction models must account for this temperature dependence, as well as for the mean stress effects arising from the tension-compression asymmetry. A common correction is the Morrow mean stress correction:
$$
\frac{\Delta \varepsilon}{2} = \frac{(\sigma_f’ – \sigma_m)}{E} (2N_f)^b + \varepsilon_f’ (2N_f)^c
$$
where $\sigma_m$ is the mean stress. The non-Masing behavior at lower temperatures further complicates life prediction, as it implies that the cyclic deformation response is not uniquely defined by the stress amplitude, requiring more advanced constitutive models (e.g., kinematic hardening rules with multiple back stresses) for accurate simulation of stress-strain paths under complex loading.
Conclusion
This comprehensive investigation into the temperature-dependent mechanical behavior of nodular cast iron QTRSi4Mo1 from room temperature to 760°C leads to the following key conclusions:
- The tensile strength (yield and ultimate) of the nodular cast iron decreases monotonically with temperature, following a bilinear relationship for yield strength with a transition near 419°C.
- A pronounced ductility minimum, indicative of embrittlement, occurs in the range of 300°C to 400°C. Above 500°C, the material exhibits significantly improved ductility but with greatly reduced strength.
- The cyclic deformation response is profoundly temperature-dependent: complex hardening-softening-hardening sequences occur at low temperatures (25-200°C), persistent cyclic hardening dominates at medium temperatures (400-500°C), and continuous cyclic softening is observed at high temperature (760°C).
- The cyclic stress-strain behavior of this nodular cast iron is well-characterized by the Ramberg-Osgood equation across the entire temperature range studied.
- The material exhibits significant tension-compression asymmetry in its hysteresis loops. It demonstrates non-Masing behavior at lower temperatures (≤400°C) but transitions to Masing behavior at higher temperatures (≥500°C).
These findings underscore the critical importance of considering specific temperature regimes when designing and performing durability assessments on components made from nodular cast iron like QTRSi4Mo1. The embrittlement region poses a particular risk for thermomechanical fatigue, while the high-temperature softening must be accounted for in stress relaxation and creep-fatigue analyses. The provided constitutive data and observed behavioral trends form a essential foundation for developing reliable fatigue life prediction methodologies for engine exhaust systems and other high-temperature applications.
