*5.1. Pearlitic Ductile Irons*

One of the first works in the context of pearlitic DIs was carried out by Mathur and Molian [47]. The authors applied an LST treatment with a Gaussian CO2 laser on both gray and ductile iron samples. The latter, named ASTM class 80-55-06, had a matrix with approximately 50% of pearlite. Three power levels (0.4, 0.8, and 1.2 kW) were defined, coupled with scanning velocities ranging from 4.2 to 169.3 mm/s. In addition, a manganese phosphate coating was used to increase the absorptivity of the material. The ductile iron changed its microstructure, initially pearlitic with graphite nodules surrounded by

ferritic shells to one made up of two well-defined areas. The closest zone to the surface was identified as the fusion zone, where graphite nodules completely dissolved and solidified to give rise to a martensitic-like structure, while the HAZ was evidenced by graphite nodules surrounded by martensite rings. Furthermore, in the cross-sections to the laser passing direction, the shape of the transformed zone was parabolic, similar to the original laser distribution. In addition, the researchers qualitatively determined, from several attempts to fit the experimental data, the relationship between the variables of the experiment (laser power, diameter, and scanning velocity) and the size or depth of the transformed zone. It was concluded that the reached depth was directly proportional to the power of the laser and inversely proportional to the diameter of the laser and the scanning speed.

Moreover, Molian and Mathur [48] explored the differences of applying the same treatment but changing the laser circular shape to a square and elliptical type and making single or multiple passes through the working section. The authors were able to determine a linear correlation between the depth of the transformed area and the square-shaped laser parameters considered in the term P/ √v. On the other hand, one-dimensional (1D) and three-dimensional (3D) heat-conduction models were used to simulate the temperature on the surface of the cast iron at constant power and different scanning velocities, with sufficiently high dispersions to conclude that the thermal response of the sample to the treatment could not be described by any simplified model.

The influence of the heat input and the solidification rates was further addressed by Chen et al. [49], where a 1 mm–diameter circular CO2 laser was used at different powers and velocities to transform the structure of a pearlitic DI. For solidification rates greater than 5 × 10<sup>4</sup> K/s, the resulting microstructure consisted of primary austenite dendrites inserted in a continuous interdendritic cementite network, while martensitic microstructures with lamellar ferrite–cementite arrays were obtained for low rates.

These transformations are consistent with the work of Gadag et al. [50], where both a CO2 laser at different levels of constant power (1 to 2.5 kW) and a 400 W Nd:YAG laser were used, coupled with different sets of scanning velocities. In this case, three zones were identified after LSM, aided by XRD patterns: a stationary region with a homogeneous microstructure consisting mainly of ledeburite, eutectic austenite, and cementite; a slowly decreasing section below the melting zone, with martensitic and fine pearlitic microstructure; and a rapid transition to the as-cast state. Moreover, a numerical simulation of the temperature during the treatment and the cooling rate at different locations was performed based on a 3D model of the heat equation that is written as follows:

$$
\rho \mathbb{C}\_p \left( \frac{\partial T}{\partial t} \right) = \nabla (k \nabla T) - lL\rho \mathbb{C}\_p \left( \frac{\partial T}{\partial x} \right), \tag{2}
$$

where *T* represents the temperature, *t* is the time, *ρ* is the density of the material, *Cp* is the specific heat for ductile iron, *k* is its thermal conductivity, and *U* is the constant laser scanning velocity. The finite-difference solution for the temperature at different depths was determined as a function of time. With this dataset, a successful comparison with the experimental HAZ depth was achieved, both for its magnitude and parabolic shape, thus validating the considered model.

Fernández-Vicente et al. [51] focused on the formation of cracks during the application of LSH and LSM treatments on pearlitic and bainitic DI castings used in the production of hot industrial rolls. A surface energy density of 60 J/mm<sup>2</sup> was experimentally found to be the transition between LSH and LSM treatments, using scanning velocities from 2 to 14 mm/s and powers between 2.5 and 4.4 kW. Visible cracks were observed on the samples, as illustrated in Figure 8. Only for the lowest energy density LSH, the transformation occurred without the generation of microscopically observable cracks. The homogeneous structure of the fusion layer determined by Gadag et al. [50] was also present in this work, as evidenced by a matrix composed of primary martensite (M) dendrites, together with interdendritic cementite (C), residual blocks of retained austenite (A), plate-shaped carbides (D), and ledeburite (LD).

**Figure 8.** Microstructure of LSM ductile iron, formed by a parabolic remelted layer (RL), a heataffected zone (HAZ), and the base material (BM), with visible cracks indicated by arrows (**top**); magnification of RL microstructure (**bottom**) (Reprinted with permission from ref. [51]. 2012 Elsevier).

A first approach on the hardening effect of LSH/LSM in pearlitic DIs was made in References [47,49], where Vickers tests were performed to measure an average hardness of 800–1000 HV in the fusion zone, compared to an original hardness between 220 and 290 HV, similar to the results found later in Molian and Baldwin [52]. However, the authors emphasized the difficulty of obtaining generalizable values in the HAZ, due to the influence of graphite nodules and their heterogeneous location in the microstructure. Moreover, in Reference [48], the effect of overlapping reported in References [40,41] for gray irons is verified for LSM.

A correlation between surface and microhardness as a function of LST parameters was also carried out by Gadag et al. [50], where it was evidenced that, the longer the interaction time (i.e., lower scanning velocities), the greater the HAZ depth. Moreover, the surface hardness increased with laser power (heat input). On the other hand, Nd:YAG produced greater surface hardness than CO2 lasers under the same conditions, which was attributed to greater absorptivity and efficiency.

Recent investigations performed by Ghaini et al. [53] allowed us to develop a model that is able to address the absolute thermal efficiency of LSH on GGG-60 cast irons in terms of a hardening efficiency index (HEI) and a hardening ratio. A 600 W fiber laser with shielding argon gas was used at two different sets of power and four beam travel speeds to ensure a reasonable hardened area without melting. After microscopical observation and Vickers tests, the hardening efficiency index was defined as follows:

$$HEI = \frac{\Delta Hardness \text{ (average)} \times Hardended area}{Heat \text{ Input}},\tag{3}$$

where the maximum achievable hardness was around 1020 HV0.3, the hardened area (or volume per unit length) was obtained from the microhardness profiles and the overall transformed region in cross-section micrographs, and the heat input was measured as the ratio between beam power and laser travel speed. A linear relationship between HEI and

the term *<sup>P</sup>*5/*v* was found, and it was compared to those of tool steels at similar operating parameters. Finally, the hardening ratio was established as follows:

$$\text{Hardening ratio} = \frac{A \times \rho \times (\int\_{T\_1}^{Ac3} \mathbb{C}\_p dT + \Delta H\_d)}{P/v},\tag{4}$$

where *A* is the area of the hardened case in cross-section, *ρ* is the density, *T*1 is the ambient temperature, *Ac3* is the full austenite formation temperature, *Cp* is the specific heat, *dT* is the temperature differential, and Δ*Ha* is the austenization enthalpy. The results were correlated with the obtained values for HEIs, and it was determined that LST with 500 W laser power and 2 mm/s travel speed achieved the maximum thermal efficiency of 15.3%.

The wear mechanisms and resistance of laser-treated pearlitic DIs were initially evaluated by Chen et al. [54], based on their previous treatments [49]. From surface erosion, abrasion, and scratch tests, it was revealed that the rate of material loss from the lasertreated bodies was reduced by a factor of up to four times, depending on the surface hardness. Subsequently, transmission electron microscopy (TEM) and scanning electron microscopy (SEM) allowed them to conclude that, for continuous high applied loads, severe plastic deformation on the surface of the treated samples generated a state of internal stresses that was responsible for the growth of microcracks that propagated, favoring the wear rate of the material.

Gadag and Srinivasan [55] also provided evidence of the improvement in the behavior of DIs treated with LSM against cavitation erosion. The authors used three classes: a pearlitic type (600/3 class), a ferritic one (400/12 class), and a 600/3 DI treated with LSM. The researchers compared the corrosion resistance in different environments, and the rate of material loss was measured from observation by optical microscopy (OM) and SEM. For the erosion tests, the three samples were exposed to artificial seawater. A linear relationship between corrosion wear and exposure time was identified, while LSM reduced the erosion rate between 6 and 8 times compared to the as-cast specimens. This improvement was attributed to the transformation of the material into a finer grain structure (consisting of ledeburite in the melting region and martensite in the hardened area) with a uniform distribution of plastic deformations and a reduction in the free path for movement of dislocations.

Molian and Baldwin [52] suggested that the improvement in erosion resistance by LST could be attributed to the emergence of beneficial compressive residual stresses during the transformation to fine martensite and retained austenite microstructures. Fernández-Vicente et al. [51] further studied the nature and magnitude of both thermal and transformational stresses due to LSM. First, thermal stresses were estimated from the temperature variations during the cycle. In particular, the only contribution to the strain was the difference between the peak temperature of the heating cycle and the temperature after the irradiation ceased. These temperatures were also approximated by simplified solutions for the heat equation, and then the value of the thermal strain was measured as follows:

$$
\varepsilon = (1+\nu)\mathfrak{a}\Delta T,\tag{5}
$$

where *ε* is the thermal strain, *ν* is the Poisson's ratio, *α* is the thermal expansion coefficient, and Δ*T* is the temperature variation. Then, following the linear elastic theory, the thermal stresses were estimated as follows:

$$
\sigma = \frac{E}{2(1+\nu)} \varepsilon,\tag{6}
$$

where *E* is the elastic modulus, and *σ* is the thermal stress. On the other hand, the authors indicated that, for higher retained austenite contents (over 40%), the observed cracking in LSM treated samples was due to an increment in the tensile stresses from the lower specific volume of retained austenite compared to martensite or bainite (Figure 9). Finally, the retention of nucleated cracks in LSH specimens was associated with the relatively high

fracture toughness of the Fe3C carbides present in the microstructure, and this contributes to the absorption of the thermal and transformational stresses generated during the treatment.

**Figure 9.** Laser processing window for LSH/LSM treatments as a function of retained austenite content, scanning velocity, and surface temperature (Reprinted with permission from ref. [51]. 2012 Elsevier).
