In this study, we investigate the effects of maximum temperature during thermal fatigue cycling on the resistance to thermal fatigue and microstructural evolution in spheroidal graphite cast iron. Spheroidal graphite cast iron is widely utilized in applications requiring high strength, oxidation resistance, and cost-effectiveness, such as automotive exhaust manifolds, where thermal fatigue is a critical failure mode. The complexity of thermal fatigue behavior arises from numerous factors, including upper and lower temperature limits, cycling rates, dwell times, stress concentrations, and microstructural changes. Our focus is on low-cycle thermal fatigue conditions, examining how varying the maximum temperature (Tmax) impacts the performance and integrity of spheroidal graphite cast iron. We aim to provide insights into the degradation mechanisms and offer guidance for material selection and design in high-temperature environments.
We conducted thermal fatigue experiments on ferritic spheroidal graphite cast iron, with Tmax set at 600°C, 700°C, and 800°C. The lower temperature (Tmin) was maintained at 20±2°C using water quenching. Specimens were subjected to repeated heating and cooling cycles, and we analyzed microstructural changes, hardness variations, and crack propagation behavior. Our findings reveal that increasing Tmax accelerates degradation processes, such as pearlite decomposition, graphite-matrix decohesion, and surface pitting, leading to reduced thermal fatigue resistance. This article elaborates on these phenomena through detailed experimental data, supported by tables and mathematical models to summarize key trends.
Material and Experimental Methodology
The material used in this study is ferritic spheroidal graphite cast iron, conforming to grade QT400-15. Its chemical composition is presented in Table 1. The base microstructure consists primarily of ferrite with a small amount of pearlite, providing a tensile strength of 400 MPa, yield strength of 250 MPa, and elongation of 15%. We machined thermal fatigue specimens with dimensions as illustrated in Figure 1, incorporating an artificial notch to initiate cracks. The notch is 0.1 mm wide, 3 mm deep, and 10 mm high, serving as a stress concentrator to simulate real-world conditions.
| Element | C | Si | Mn | P | S | Mg | RE |
|---|---|---|---|---|---|---|---|
| Content | 3.5 | 3.0 | 0.25 | 0.047 | 0.012 | 0.029 | 0.02 |
Thermal fatigue tests were performed in an SX2-5 box-type resistance furnace. For each cycle, specimens were heated to Tmax for 180 seconds, then rapidly quenched in water. To minimize oxidation from steam, specimens were dried after each cooling phase before reinsertion into the furnace. We monitored the process over multiple cycles, measuring crack lengths, examining microstructures using optical microscopy and scanning electron microscopy (SEM), and assessing hardness with a Rockwell hardness tester. Crack lengths were measured from the artificial notch, excluding graphite spheres at the crack tip. Our analysis emphasizes the role of Tmax in driving microstructural transformations and crack dynamics.
Microstructural Evolution Under Thermal Fatigue
As thermal cycles accumulate, spheroidal graphite cast iron undergoes significant microstructural changes. Initially, the matrix comprises ferrite and pearlite, but heating promotes decomposition of pearlite into ferrite and carbon. The rate of this process is highly temperature-dependent, following the Arrhenius equation for diffusion:
$$ D = D_0 \exp\left(-\frac{Q}{RT}\right) $$
where \( D \) is the diffusion coefficient, \( D_0 \) is a pre-exponential factor, \( Q \) is the activation energy, \( R \) is the gas constant, and \( T \) is the absolute temperature. Higher Tmax values, such as 800°C, enhance carbon diffusion, accelerating pearlite breakdown. We observed that after 40 cycles at Tmax = 800°C, pearlite regions showed pronounced decomposition, with cementite lamellae fragmenting, as depicted in Figure 2c. In contrast, at 600°C and 700°C, changes were subtle even after 20 cycles.
Another critical phenomenon is the decohesion between graphite spheroids and the ferritic matrix. Due to the stark difference in thermal expansion coefficients—graphite expands more rapidly than iron—thermal stresses develop at interfaces. At Tmax = 800°C, these stresses exceed the bonding strength, causing graphite to detach from the matrix. Additionally, surface pitting occurs where graphite spheres underlie the surface; the matrix above graphite cracks and collapses under cyclic stress, forming凹陷s (Figure 2b). This pitting is exacerbated by stress concentrations near the artificial notch.
Oxidation also plays a detrimental role. Surface oxides, appearing as fluffy aggregates (Figure 2d), are brittle and prone to cracking, providing pathways for crack initiation. We used energy-dispersive spectroscopy (EDS) to analyze oxide composition, confirming high carbon content at grain boundaries. During heating, carbon from graphite and decomposing pearlite diffuses into the matrix; during rapid cooling, it precipitates as cementite at grain boundaries due to non-equilibrium conditions. This precipitation is more pronounced at higher Tmax, as shown in Figure 2f. Over many cycles, accumulated heating effects simulate graphitization annealing, leading to fine graphite formation at grain boundaries.
To quantify microstructural stability, we measured hardness variations. Table 2 summarizes hardness data at different cycle numbers for each Tmax. Hardness trends are plotted in Figure 3, revealing an initial increase followed by a decrease. This behavior stems from competing mechanisms: strain hardening and phase transformations increase hardness, while pearlite decomposition and oxidation decrease it. At Tmax = 800°C, martensite forms upon quenching from above the eutectoid temperature, significantly boosting hardness. However, with continued cycling, martensite and pearlite decompose, and oxidation intensifies, causing hardness decline. The rates of increase and decrease are more rapid at higher Tmax, as modeled by:
$$ \frac{dH}{dN} = k_1 \exp\left(-\frac{E_a}{RT_{\text{max}}}\right) – k_2 N $$
where \( H \) is hardness, \( N \) is cycle number, \( k_1 \) and \( k_2 \) are constants, and \( E_a \) is an activation energy.
| Cycle Number (N) | Tmax = 600°C | Tmax = 700°C | Tmax = 800°C |
|---|---|---|---|
| 0 | 85 | 85 | 85 |
| 20 | 86 | 87 | 92 |
| 40 | 87 | 88 | 95 |
| 60 | 88 | 89 | 96 |
| 80 | 87 | 88 | 94 |
| 100 | 86 | 87 | 90 |
| 120 | 85 | 86 | 87 |
Crack Initiation and Propagation Behavior
Crack formation in spheroidal graphite cast iron during thermal fatigue arises from multiple sources: thermal stresses from rapid temperature changes, stresses due to thermal expansion mismatch between graphite and matrix, and phase transformation stresses. We identified four primary crack initiation sites: (1) the artificial notch, where stress concentration exceeds the ferrite’s tensile strength; (2) irregular contours of graphite spheroids, causing localized stress in adjacent ferrite (Figure 2e); (3) matrix overlaying graphite, which fractures under cyclic expansion and contraction; and (4) interfaces where graphite detaches from the matrix. At lower Tmax (600°C and 700°C), initiation is dominated by the first two mechanisms, while at 800°C, all four contribute due to higher stresses.
Crack propagation paths are consistent across Tmax variations. Major cracks originate from the artificial notch, traverse low-strength ferrite regions, link with pre-existing cracks at graphite or pits, and extend toward the specimen’s opposite end (Figure 4). Oxidation along crack flanks accelerates propagation by embrittling the material. We measured crack lengths over cycles, as shown in Table 3 and plotted in Figure 5. Crack growth rates (da/dN) were derived using the Paris law for subcritical extension:
$$ \frac{da}{dN} = C (Y \Delta \sigma \sqrt{\pi a})^n $$
where \( a \) is crack length, \( \Delta \sigma \) is the thermal stress range, and \( C \), \( Y \), and \( n \) are material constants. Thermal stress can be estimated from:
$$ \Delta \sigma = E \alpha \Delta T $$
with \( E \) as Young’s modulus, \( \alpha \) as the coefficient of thermal expansion, and \( \Delta T = T_{\text{max}} – T_{\text{min}} \). For spheroidal graphite cast iron, \( \alpha \) is anisotropic due to graphite presence, but we approximate it as a composite value.
| Cycle Number (N) | Tmax = 600°C | Tmax = 700°C | Tmax = 800°C |
|---|---|---|---|
| 20 | 0.15 | 0.20 | 0.30 |
| 40 | 0.30 | 0.45 | 0.75 |
| 60 | 0.45 | 0.70 | 1.35 |
| 80 | 0.60 | 0.95 | 2.10 |
| 100 | 0.75 | 1.20 | 3.00 |
| 120 | 0.90 | 1.50 | 4.50 |
From the data, da/dN is relatively constant in the Paris regime for Tmax = 600°C and 700°C, with average rates of approximately 3 μm/cycle and 6 μm/cycle, respectively. At Tmax = 800°C, da/dN is about 9 μm/cycle initially but increases sharply after 100 cycles to 16 μm/cycle, as decohesion and pitting provide easy crack paths. This transition marks a shortened subcritical expansion phase, highlighting the severe impact of high temperature on spheroidal graphite cast iron’s durability.
Discussion on Thermal Fatigue Mechanisms
The degradation of spheroidal graphite cast iron under thermal fatigue is a multifactorial process. We can model the cumulative damage \( D \) as a function of Tmax and cycle number:
$$ D = \int_0^N f(T_{\text{max}}, \sigma, \text{microstructure}) \, dN $$
where \( f \) incorporates effects like oxidation kinetics, phase transformations, and crack growth. Oxidation follows a parabolic rate law:
$$ x^2 = k_p t $$
with \( x \) as oxide thickness and \( k_p \) as a temperature-dependent constant. Higher Tmax increases \( k_p \), accelerating surface degradation.
Graphite-matrix interaction is crucial. The thermal expansion mismatch generates interfacial stresses \( \sigma_i \) given by:
$$ \sigma_i = \frac{E_m E_g (\alpha_g – \alpha_m) \Delta T}{E_m (1-\nu_g) + E_g (1-\nu_m)} $$
where subscripts \( m \) and \( g \) denote matrix and graphite, and \( \nu \) is Poisson’s ratio. For spheroidal graphite cast iron, \( \alpha_g \gg \alpha_m \), leading to high \( \sigma_i \) that promotes decohesion.
Phase transformations also contribute. At Tmax ≥ 800°C, partial austenitization occurs, and quenching produces martensite, increasing hardness but also introducing transformation stresses. Over cycles, tempering effects soften the martensite, reducing hardness. We can express the phase fraction \( f_p \) of martensite as:
$$ f_p = f_0 \left[1 – \exp\left(-\frac{N}{\tau}\right)\right] $$
where \( f_0 \) is the maximum possible fraction and \( \tau \) is a time constant dependent on Tmax.
To optimize spheroidal graphite cast iron for thermal fatigue resistance, control of Tmax is essential. Our data suggest that keeping Tmax below 700°C prolongs service life by minimizing microstructural damage. However, in applications like exhaust systems, temperatures can exceed 800°C, necessitating alloy modifications or coatings.
Conclusions
Our investigation into the influence of maximum temperature on thermal fatigue resistance of spheroidal graphite cast iron yields several key findings. First, increasing Tmax from 600°C to 800°C accelerates microstructural degradation, including pearlite decomposition, graphite-matrix decohesion, and surface pitting. These changes are driven by enhanced diffusion and stress generation, as described by Arrhenius and stress models. Second, hardness exhibits a non-monotonic trend, rising initially due to strain hardening and phase transformations, then falling as decomposition and oxidation dominate. The rates are markedly higher at Tmax = 800°C. Third, crack propagation behavior is consistent across temperatures, with major cracks extending from notches through ferrite, connecting defects. However, crack growth rates increase substantially with Tmax, reducing the subcritical expansion period. Finally, we conclude that spheroidal graphite cast iron’s thermal fatigue resistance deteriorates at elevated temperatures, underscoring the need for careful thermal management in design. Future work could explore alloying elements or heat treatments to mitigate these effects.
This study comprehensively analyzes the thermomechanical behavior of spheroidal graphite cast iron, providing a foundation for predictive models and material improvement. The integration of experimental data with mathematical formulations offers a robust framework for assessing performance in cyclic thermal environments.