Influence of Solidification Conditions on the Microstructure of Laser

materials

Article Inﬂuence of Solidiﬁcation Conditions on the Microstructure of Laser-Surface-Melted Ductile Cast Iron

Damian Janicki 1,* , Jacek Górka 1 , Waldemar Kwa´sny 2 , Wojciech Pakieła 2 and Krzysztof Matus 2 1 Department of Welding, Silesian University of Technology, Konarskiego 18A, 44-100 Gliwice; Poland; [email protected] 2 Department of Engineering Materials and Biomaterials, Silesian University of Technology, Konarskiego 18A, 44-100 Gliwice, Poland; [email protected] (W.K.); [email protected] (W.P.); [email protected] (K.M.) * Correspondence: [email protected]

Received: 21 February 2020; Accepted: 4 March 2020; Published: 6 March 2020

Abstract: The thermal conditions in the molten pool during the laser surface melting of ductile cast iron EN-GJS-700-2 were estimated by using infrared thermography and thermocouple measurements. The thermal data were then correlated with the microstructure of the melted zone. Additionally, the thermodynamic calculations of a Fe-C-Si alloy system were performed to predict the solidification path of the melted zone. It was found that increasing the cooling rate during solidification of the refined ledeburite eutectic but also suppressed the martensitic transformation. A continuous network of plate-like secondary cementite precipitates and nanometric spherical precipitates of tertiary cementite were observed in regions of primary and eutectic austenite. The solidification of the melted zone terminated with the Liquid γ-Fe + Fe C + Fe Si C reaction. The hardness of the → 3 8 2 melted zone was affected by both the fraction of the retained austenite and the morphology of the ledeburite eutectic.

Keywords: laser surface melting; ductile cast iron: secondary cementite; tertiary cementite; thermography; cooling rate

1. Introduction Over the past three decades, considerable research has been devoted to enhancing the wear resistance of different grades of cast iron via laser surface modification methods. Most reported works have focused on shaping the microstructure, and thus wear properties, of surface layers by using laser surface melting (LSM) [1–5]. This process involves rapid solidification and cooling, and it provides a unique opportunity to synthesize non-equilibrium structures [6,7]. Additionally, compared with other laser surface modification methods, such as laser surface alloying and cladding processes, LSM is a relatively simple process since it does not change the chemical composition of the processed surface layer [8–10]. However, the inability to change the chemical composition of the melt also significantly limits the ability to change the microstructure and mechanical properties of the processed surface layer [11]. During LSM, the mechanical properties of the processed layers can typically only be modified by controlling the solidification conditions in the molten pool [12,13]. Consequently, to optimize the wear properties of the working surface of the components that are made of a particular cast iron grade via LSM, it is necessary to comprehensively analyze the microstructure evolution in the processed surface layers under different solidification conditions. Unfortunately, especially in the case of ductile cast iron (DCI), little data are available. Since DCI is finding use in a growing range of

Materials 2020, 13, 1174; doi:10.3390/ma13051174 www.mdpi.com/journal/materials Materials 2020, 13, 1174 2 of 13 industrial applications, comprehensive knowledge of the phenomena that occur during this process would be particularly useful. Such knowledge would provide a means to determine how to shape the microstructure, and thus the mechanical properties, of the resulting surface layers. The thermal conditions in the molten pool of a DCI sample were analyzed by using infrared (IR) thermography. Since emissivity is the most important calibration parameter for temperature measurements when using IR thermography [14,15], this research method included an original calibration procedure of the IR camera to estimate the emissivity values of the melt in the molten pool, as well as regions that were directly adjacent to the molten pool under a given shielding atmosphere. The main purpose of this work was to correlate changes in the selected processing conditions and the resulting changes in the solidiﬁcation conditions in the molten pool with the microstructure of single-pass melted beads (SMB). To comprehensively understand the microstructure development in the melted zone, the developed research methodology combined an experimental approach with computational thermodynamics.

2. Materials and Methods For the research, DCI-grade EN-GJS-700-2 with a pearlitic/ferritic matrix and graphite precipitates with an average diameter of about 30 µm was selected. The chemical composition of the studied DCI substrate is given in Table1. Substrate material (SM) specimens with dimensions of 25 60 10 mm × × were ground to an average roughness (Ra) of approx. 0.5 µm. Prior to laser processing, the SM specimens were cleaned in acetone.

Table 1. Chemical composition of the used DCI grade EN-GJS-700-2 (wt %).

C Si Cu Mn Cr Ni S P Fe 3.60 2.51 0.78 0.25 0.02 0.04 0.008 0.016 balance

The laser processing trials were carried out by using an experimental stand that was equipped with a Roﬁn-Sinar DL 020 2.0 kW continuous-wave high-power direct diode laser (HPDDL, Roﬁn-Sinar Laser GmbH, Hamburg, Germany). The HPDDL had a rectangular beam with a top-hat intensity distribution in the slow-axis direction and a near Gaussian distribution in the fast-axis direction (spot size of 1.5 6.6 mm). Additional laser source details are available elsewhere [16]. During all trials, × the slow-axis of the laser beam was set to be perpendicular to the traverse direction, and the beam focal plane was located at the SM specimen surface. Argon was used to prevent the laser-processed area from oxidation. The processing conditions are presented in Table2.

Table 2. Selected processing conditions of laser surface melting (LSM).

Processing Conditions No. Laser Power (W) Traverse Speed (mm/s) Heat Input 1 (J/mm) A1 2000 3.33 600 A2 1000 1.67 600 A3 1500 1.25 1200 1 deﬁned by the ratio of the laser power and the traverse speed.

In order to examine the thermal conditions in the molten pool, the experimental stand was equipped with an original measurement system that combined IR thermography and a thermocouple. The system used an IR FLIR A600-Series camera (FLIR System, Inc. Wilsonville, OR, USA) with a measurement temperature range from 600 to 2200 ◦C and a spectral range of 7.5–14.0 µm. The maximum frame rate of the thermal image capturing system was 200 Hz. The thermocouple data were acquired and recorded by using a computer-based system with an Agilent 34970A data logger (Agilent Technologies, Inc., Santa Clara, CA, USA), which provided a maximum sampling frequency of 300 Hz. Materials 2020, 13, 1174 3 of 13

Materials 2020, 13, x FOR PEER REVIEW 3 of 14 An FLIR A600-Series thermal camera was used to investigate the temperature on the surface of the moltenAn FLIR pool A600-Series and the regions thermal that camera were was directly used adjacent to investigate to the moltenthe temperature pool on the on laser-processedthe surface of surface.the molten To pool calibrate and the the regions IR camera, that were an original directly calibration adjacent toprocedure the molten waspool developed on the laser-processed that allowed forsurface. the estimation To calibrate of the the IR emissivity camera, an of original the melt calibration in the molten procedure pool and was the developed regions thatthat wereallowed directly for adjacentthe estimation to the of molten the emissivity pool under of athe given melt shielding in the molten atmosphere. pool and The the emissivity regions that data were were directly used to reconstructadjacent to two-dimensionalthe molten pool under temperature a given profiles shieldin alongg atmosphere. the molten The pool emissivity and the surroundingdata were used regions to onreconstruct the laser-processed two-dimensional surface. temperature The developed profiles procedure along providedthe molten a direct pool correlationand the surrounding between the dataregions of theon IRthe camera laser-processed and the thermocouple surface. The measurements. developed procedure The scheme provided of this a method direct correlation is presented inbetween Figure 1the. During data of thethe aboveIR camera calibration and the procedure, thermocouple the HPDDLmeasurements. beam heatedThe scheme and partially of this method melted theis presented sample under in Figure an argon 1. During atmosphere the above that wascalibration provided procedure, by the shielding the HPDDL nozzle beam that washeated arranged and coaxiallypartially melted with the the laser sample beam under (Figure an 1argona). The atmosp dimensionshere that of was the sampleprovided and by itsthe alignment shielding relativenozzle tothat the was laser arranged beam spotcoaxially are presentedwith the laser in Figure beam1 (Figureb. By using1a). The the dimensions samples of of the the as-received sample and DCI its andalignment the samples relative with to the the laser melted beam layer, spot itare was presen possibleted in toFigure estimate 1b. By the using emissivity the samples of the of region the as- in frontreceived of the DCI molten and the pool samples and thewith solidified the melted material layer, it of was the possible processed to estimate bead behind the emissivity the molten of pool.the Theregion two-step in front calibration of the molten procedure pool and included the solidified IR thermography material of the and processed thermocouple bead behind measurements the molten that requiredpool. The the two-step use of samples calibration with identical procedure dimensions. included In IR the firstthermography stage, an IR and camera thermocouple recorded the temperaturemeasurements fields that during required the the heating use of and samples melting with of aidentical sample. dimensions. In the second In stagethe first of thestage, procedure, an IR bycamera using recorded the same the heating temperature conditions fields of during the sample, the heating the thermocouple and melting providedof a sample. a pointwise In the second check ofstage the temperatureof the procedure, fields by that using were the previously same heating recorded conditions by the IR of camera. the sample, The datathe thermocouple from these two measurementprovided a pointwise stages were check acquired of the temperature and recorded fields at a that rate were of 200 previously Hz. The acquisitionrecorded by and the recordingIR camera. of theThe thermocouple data from these data two were measurement performed by stages using were a computer-based acquired and systemrecorded with at ana rate Agilent of 200 34970A Hz. The data logger.acquisition Depending and recording on the location of the thermocouple of the measuring data point were on performed the sample by (Figure using1 b)a computer-based and the resulting rangesystem of with recorded an Agilent temperatures, 34970A data a standard logger. K-typeDependin thermocoupleg on the location and non-standard of the measuring high-temperature point on the Mo–Wsample thermocouple(Figure 1b) wereand used.the resulting The Mo–W range thermocouple of recorded was calibratedtemperatures, by using a standard the data reportedK-type bythermocouple Michalski et and al.[ 17non-standard]. high-temperature Mo–W thermocouple were used. The Mo–W thermocouple was calibrated by using the data reported by Michalski et al. [17].

(a) (b)

FigureFigure 1. Diagram showing showing ( (aa)) the the experimental experimental setup setup for for emissivity emissivity measurements measurements of of the the ductile ductile cast cast ironiron (DCI)(DCI) inin thethe as-received condition condition and and after after laser laser processing processing (during (during IR IR camera camera measurements) measurements) andand ((b)) thethe samplesample dimensions, thermocouple thermocouple locations, locations, and and alignment alignment of of the the sample sample relative relative to to the the laserlaser beambeam spot.

MicrostructuralMicrostructural characterization was was performed performed by by using using scanning scanning electron electron microscopy microscopy (SEM) (SEM) andand transmission electron microscopy (TEM). (TEM). SEM SEM investigations investigations were were conducted conducted by by using using a aPhenom Phenom ProXProX (Phenom-World,(Phenom-World, Eindhoven, Netherlands) Netherlands) and and ZEISS ZEISS SUPRA SUPRA 35 35 (Carl (Carl Zeiss Zeiss Microscopy Microscopy GmbH, GmbH, Jena,Jena, Germany)Germany) microscopes. SEM SEM observations observations were were carried carried out out in in the the backscattered backscattered electron electron (BSE) (BSE) andand in-lensin-lens image image modes. modes. For For the the SEM SEM investigation, investigatio samplesn, samples were were polished polished by using by 0.04usingµm 0.04 colloidal µm silicacolloidal and silica etched and with etched 12% with Nital. 12% AnNital. area An (volume) area (volume) fraction fraction of the of eutectic the eutectic cementite cementite phases phases was calculatedwas calculated from from the SEM the micrographsSEM micrographs by using by theusin NIS-Elementsg the NIS-Elements image processingimage processing software software (version (version 3.13, Nikon Corporation, Tokyo, Japan).Measurements were performed on the SEM images of the undersurface area of the SMBs (~200 µm beneath the bead surface) over a total area of 10 mm² for each bead. The TEM analysis was conducted by using an FEI Titan 80/300 microscope (Scientific

Materials 2020, 13, 1174 4 of 13

3.13, Nikon Corporation, Tokyo, Japan).Measurements were performed on the SEM images of the undersurface area of the SMBs (~200 µm beneath the bead surface) over a total area of 10 mm2 for each bead. The TEM analysis was conducted by using an FEI Titan 80/300 microscope (Scientific and Technical Instruments, Hillsboro, OR, USA). During the TEM investigation, the selected area electron diffraction (SAED) patterns were recorded for phase identification. The phase composition of SMBs and the unit cell parameters of the austenite phase were determined by using X-ray diffraction (XRD). The XRD analysis was performed using a PANalytical X’Pert PRO MPD X-ray diffractometer (Malvern PANalytical, Malvern, UK). XRD patterns were recorded by using an X’Celerator detector (Malvern PANalytical, Malvern, UK) and a Cu-Kα (λ = 0.15406 nm) source operating at 40 kV and 30 mA. The XRD data were obtained over an angular range of 30–125◦. The martensite and retained austenite fractions in the SMBs were evaluated by using the Rietveld refinement method in the FullProf software. Hardness measurements were conducted on polished SMB cross-sections by using a Wilson Wolpert 401 MVD Vickers microhardness indenter (Wilson Wolpert Instruments, Aachen, Germany) with a 200 g load and a 10 s dwell time. Thermodynamics calculations of the Fe-C-Si ternary alloy system were made under both metastable equilibrium (considering the formation of the cementite phase) and non-equilibrium conditions by using commercial Thermo-Calc software. The Scheil–Gulliver model was used to predict the non-equilibrium solidification path of the Fe-C-Si system and to estimate the melt composition in the final stage of solidification.

3. Results and Discussion

3.1. Examination of Thermal Conditions in the Molten Pool IR thermography measurements of the temperature distribution on the surface of the molten pool during LSM under processing conditions A1, A2, and A3 (Table2) are presented in Figure2. The corresponding temperature proﬁles at the surface of the molten pool along the SMB centerline are shown in Figure3. Figure4 presents the estimated emissivity values of the unprocessed DCI and the DCI after LSM as a function of temperature under an argon atmosphere. Table3 summarizes the cooling rate and temperature gradient at the surface of the molten pool along the SMB centerline for the investigated processing conditions.

Table 3. Values of the cooling rate (GR) and temperature gradient (G) at the surface of the molten pool along the SMB centerline for diﬀerent processing conditions.

Temperature Range (◦C) Processing Conditions No. 1200–1090 2 1090–800 (Table2) 1 1 1 1 GR (◦C/s) G (◦C/mm) GR (◦C/s) G (◦C/mm) A1 495 38 149 12 326 32 98 10 ± ± ± ± A2 611 28 366 17 337 18 202 10 ± ± ± ± A3 97 22 78 17 127 11 102 9 ± ± ± ± 1 Mean value and the standard deviation of three measurements; 2 the solidus temperature of Fe-3.6 wt% C-2.5 wt% Si alloy system that was determined by using the Scheil calculation.

The data implied that the cooling rate at the trailing edge of the molten pool decreased when the heat input level increased. Furthermore, an increase in the heat input signiﬁcantly decreased the cooling rate behind the molten pool. From 1090 to 800 ◦C, the cooling rate of the region of the solidiﬁed melt-bead material behind the liquid/solid interface was about 360 and 130 ◦C/s for heat input levels of 600 and 1200 J/mm, respectively (note that the solidus temperature of the investigated DCI was estimated from the Scheil calculation to be approx. 1090 ◦C; Section 3.2). The temperature gradient in the trailing edge of a molten pool directly depends on the molten pool’s geometry, which in turn is Materials 2020, 13, x FOR PEER REVIEW 4 of 14

and Technical Instruments, Hillsboro, OR, USA). During the TEM investigation, the selected area electron diffraction (SAED) patterns were recorded for phase identification. The phase composition of SMBs and the unit cell parameters of the austenite phase were determined by using X-ray diffraction (XRD). The XRD analysis was performed using a PANalytical X’Pert PRO MPD X-ray diffractometer (Malvern PANalytical, Malvern, UK). XRD patterns were recorded by using an X’Celerator detector (Malvern PANalytical, Malvern, UK) and a Cu-Kα (λ = 0.15406 nm) source Materialsoperating2020 ,at13 ,40 1174 kV and 30 mA. The XRD data were obtained over an angular range of 30–125°.5 of The 13 martensite and retained austenite fractions in the SMBs were evaluated by using the Rietveld refinement method in the FullProf software . significantly affected by the traverse speed [18,19]. In the case of the investigated DCI, for a given heat Hardness measurements were conducted on polished SMB cross-sections by using a Wilson input level, the change in the traverse speed also significantly affected the cooling rate at the trailing Wolpert 401 MVD Vickers microhardness indenter (Wilson Wolpert Instruments, Aachen, Germany) edge of the molten pool. At a heat input of 600 J/mm (A1 and A2 processing conditions; Table2), with a 200 g load and a 10 s dwell time. increasing the traverse speed notably decreased both the temperature gradient and the cooling rate at Thermodynamics calculations of the Fe-C-Si ternary alloy system were made under both the trailing edge of the molten pool. For the A1 and A2 processing conditions, the cooling rates at metastable equilibrium (considering the formation of the cementite phase) and non-equilibrium the trailing edge of the molten pool were 495 38 and 611 28 ◦C/s, respectively (calculated from conditions by using commercial Thermo-Calc± software. The± Scheil–Gulliver model was used to 1200–1090 ◦C). This was mainly attributed to the difference in the molten pool shape. Due to the higher predict the non-equilibrium solidification path of the Fe-C-Si system and to estimate the melt traverse speed (3.33 mm/s), the A1 processing condition led to a significantly elongated molten pool. composition in the final stage of solidification. As a result, the distance over which the temperature decreased from its maximum value to the solidus temperature of the DCI substrate was significantly larger than in the A2 processing condition (traverse 3. Results and Discussion speed of 1.66 mm/s). On the other hand, it is reasonable to suggest that the large difference in the laser power3.1. Examination between the of Thermal A1 and A2Conditions processing in the conditions Molten Pool (2000 and 1000 W, respectively) may have also affected the amount of energy that was absorbed by the molten pool. This suggestion is supported by the factIR that thermography the cross-sectional measurements area of theof the bead temperature that was produced distribution at a on laser the power surface of of 2000 the W molten was aboutpool 40%during larger LSM than under that processing of the bead conditions that was produced A1, A2, and at 1000 A3 W.(Table Anincreased 2) are presented absorption in Figure of laser 2. lightThe withcorresponding an increase temperature in laser intensity, profiles at aat constant the surfac traversee of the speed, molten was pool reported along bythe Trapp SMB etcenterline al. [20]. Theare coolingshown ratein Figure values 3. of the region of the solidified melt-bead in the temperature range of 1090–800 ◦C did not significantly vary with changes in the traverse speed at a heat input of 600 J/mm.

(a) (b)

(c)

Figure 2. IR camera images obtained during LSM of the DCI under processing conditions: (Table2) (a) A1, (b) A2, and (c) A3. (Selected emissivity value, E = 0.13). The arrows and dashed lines indicate the traverse direction of laser beam and the single-pass melted beads (SMB) centerline, respectively. Materials 2020, 13, x FOR PEER REVIEW 5 of 14

Figure 2. IR camera images obtained during LSM of the DCI under processing conditions: (Table 2) (a) A1, (b) A2, and (c) A3. (Selected emissivity value, E = 0.13). The arrows and dashed lines indicate the traverse direction of laser beam and the single-pass melted beads (SMB) centerline, respectively.

Materials 2020, 13, x FOR PEER REVIEW 5 of 14

Figure 2. IR camera images obtained during LSM of the DCI under processing conditions: (Table 2) (a) A1, (b) A2, and (c) A3. (Selected emissivity value, E = 0.13). The arrows and dashed lines indicate the traverse direction of laser beam and the single-pass melted beads (SMB) centerline, respectively. Materials 2020, 13, 1174 6 of 13

(a) (b)

(c)

Figure 3. Temperature profiles at the surface of the molten pool along the SMB centerline under processing conditions: (Table 2) (a) A1, (b) A2, and (c) A3. Emissivity values were selected based on Figure 4. The dashed lines indicate the solidus temperature of the DCI that was determined by using the Scheil calculation (Section 3.2). (c) FigureFigure 3. 4 presentsTemperature the estimated proﬁles at theemissivity surface ofvalu thees molten of the unprocessed pool along the DCI SMB and centerline the DCI under after LSM as a processingfunction of conditions: temperature (Table under2)( a) A1,an argon (b) A2, atmosphere. and (c) A3. Emissivity Table 3 summarizes values were selected the cooling based rate on and temperatureFigure4. The gradient dashed at lines the indicatesurface theof the solidus molten temperature pool along of the the DCI SMB that centerline was determined for the by investigated using Figure 3. Temperature profiles at the surface of the molten pool along the SMB centerline under processingthe Scheil conditions. calculation (Section 3.2). processing conditions: (Table 2) (a) A1, (b) A2, and (c) A3. Emissivity values were selected based on Figure 4. The dashed lines indicate the solidus temperature of the DCI that was determined by using the Scheil calculation (Section 3.2).

Figure 4 presents the estimated emissivity values of the unprocessed DCI and the DCI after LSM as a function of temperature under an argon atmosphere. Table 3 summarizes the cooling rate and temperature gradient at the surface of the molten pool along the SMB centerline for the investigated processing conditions.

(a) (b)

Figure 4. Variation of the emissivity of the surface of the investigated DCI as a function of temperature in (a) the as-received condition (after grinding to a surface ﬁnish of 0.5 µm Ra) and (b) after laser surface melting.

3.2. Thermodynamic Calculations The metastable equilibrium phase diagram (i.e., considering the formation of the cementite phase instead of graphite) and(a) the Scheil solidification path resulting from the thermodynamic(b) calculations of a ternary Fe-3.6 wt% C-2.5 wt% Si alloy system are presented in Figures5 and6a, respectively. The calculation showed that the first phase to precipitate in the molten pool was an austenite phase (γ-Fe). Further cooling indicated that a binary eutectic reaction occurred, and this led to the formation of ledeburite eutectic (γ-Fe+Fe3C). In contrast to the metastable equilibrium diagram, the Scheil model predicted that the solidification path ended with the formation of ternary eutectic γ-Fe/Fe3C/Fe8Si2C. MaterialsMaterials 20202020,, 1313,, xx FORFOR PEERPEER REVIEWREVIEW 77 ofof 1414

Fe/FeFe/Fe33C/FeC/Fe88SiSi22C.C. Generally,Generally, thethe ScheilScheil simulationsimulation resultsresults ofof thethe Fe-C-SiFe-C-Si alloyalloy systemsystem suggestedsuggested thethe followingfollowing solidificationsolidification sequencesequence inin thethe DCIDCI substrate:substrate: Materials(1)(1) 2020NucleationNucleation, 13, 1174 ofof primaryprimary austeniteaustenite (~1165(~1165 °°C):C): 7 of 13 LiquidLiquid →→ LiquidLiquid ++ γγ-Fe.-Fe. Generally, the Scheil simulation results of the Fe-C-Si alloy system suggested the following solidiﬁcation (2) Binary eutectic reaction (~1120 °°C ): sequence(2) Binary in the eutectic DCI substrate: reaction (~1120 C):

(1) Nucleation of primary austeniteLiquidLiquid (~1165 →→ LiquidLiquid◦C): ++ γγ-Fe-Fe ++ FeFe33C.C.

(3) Ternary eutectic reaction (~1090 °C ): (3) Ternary eutectic reaction (~1090Liquid °C): Liquid + γ-Fe. → LiquidLiquid →→ LiquidLiquid ++ γγ-Fe-Fe ++ FeFe33CC ++ FeFe88SiSi22C.C. (2) Binary eutectic reaction (~1120 ◦C ): AccordingAccording toto thethe literatureliterature [21,22],[21,22], thethe formationformation ofof thethe ternaryternary eutecticeutectic γγ-Fe/Fe-Fe/Fe33C/FeC/Fe88SiSi22CC inin anan Fe-C-SiFe-C-Si alloyalloy systemsystem occursoccurs atat SiSi contentscontentsLiquid higherhigherLiquid thanthan+ γ 3.03.0-Fe wtwt+ Fe %.%. C. ThisThis isis inin goodgood agreementagreement withwith thethe → 3 ScheilScheil calculationcalculation resultsresults forfor segregatedsegregated alloyingalloying elementselements duringduring thethe solidificationsolidification ofof thethe investigatedinvestigated DCIDCI (Figure(3)(Figure Ternary 6b),6b), eutectic whichwhich indicatedindicated reaction (~1090 thatthat thethe◦C CC ): andand SiSi cocontentsntents inin thethe liquidliquid atat thethe finalfinal solidificationsolidification stagestage reachedreached 4.14.1 andand 4.34.3 wtwt %,%, respectively.respectively. TheThe presencepresence ofof eutecticeutectic γγ-Fe/Fe-Fe/Fe33C/FeC/Fe88SiSi22CC inin thethe Liquid Liquid + γ-Fe + Fe C + Fe Si C. microstructuremicrostructure ofof thethe SMBsSMBs waswas confirmedconfirmed→ byby thethe microstructuralmicrostructural3 8 investigationsinvestigations2 thatthat areare presentedpresented inin SectionSection 3.3.3.3.

FigureFigure 5. 5. IsoplethIsopleth atat 2.52.5 wt%wt% SiSi in in the the metastable metastable equilibriumequilibrium equilibrium Fe-C-SiFe-C-Si Fe-C-Si phasephase diagram.diagram.

(a)(a) (b)(b) FigureFigure 6.6. ((aa)) ScheilScheil solidificationsolidification pathpath forfor thethe Fe-3.6Fe-3.6 wt%wt% C-2.5C-2.5 wt%wt% SiSi alloyalloy system;system; ((bb)) thethe Figure 6. (a) Scheil solidiﬁcation path for the Fe-3.6 wt% C-2.5 wt% Si alloy system; (b) the corresponding correspondingcorresponding compositioncomposition ofof thethe liquidliquid asas thethe functionfunction ofof temperature.temperature. composition of the liquid as the function of temperature.

Materials 2020, 13, 1174 8 of 13

According to the literature [21,22], the formation of the ternary eutectic γ-Fe/Fe3C/Fe8Si2C in an Fe-C-Si alloy system occurs at Si contents higher than 3.0 wt %. This is in good agreement with the Scheil calculation results for segregated alloying elements during the solidification of the investigated DCI (Figure6b), which indicated that the C and Si contents in the liquid at the final solidification stage reached 4.1 and 4.3 wt %, respectively. The presence of eutectic γ-Fe/Fe3C/Fe8Si2C in the microstructure of the SMBs was confirmed by the microstructural investigations that are presented in Section 3.3.

3.3. Microstructural Analysis Figure7 shows BSE SEM images that were taken of the undersurface area of SMBs that were produced under the processing conditions listed in Table2. The corresponding microstructural parameters of SMBs, based on a quantitative analysis of micrographs and XRD patterns, are presented in Table4. In general, the microstructure of the melted zone contained primary austenite ( γp) dendrites transformed to martensite (partially or fully) and a ledeburite eutectic in the interdendritic regions that, in turn, was composed of a eutectic austenite (γe) phase and a eutectic cementite phase. Additionally, the formation of fine eutectic-like structures was observed in the interdendritic regions, as shown in Figure8a,b. Based on the results of thermodynamic calculations (Section 3.2) that predicted a ternary eutectic reaction at the final solidification stage under non-equilibrium conditions, it is reasonable to assume that the above fine eutectic-like structures were ternary eutectic α-Fe/Fe3C/Fe8Si2C. This assumption is also supported by the fact that the morphology of the observed ternary eutectic is consistent with that previously reported for the Fe-C-Si alloy system [23]. The above results indicate a good agreement between the experimental results and those that were calculated under the Scheil condition. Due to the high carbon content in the processed DCI (3.6 wt%), the eutectic regions (ledeburite eutectic structure) formed a continuous network surrounding the γp dendrites. The investigated solidification conditions did not provide a substantial change in the fraction of the ledeburite eutectic. The fraction of the eutectic cementite was approx. 53 vol%. In contrast, the cooling rate altered the morphology of the γp dendrites and ledeburite eutectic. Generally, as the cooling rate decreased, the γp dendrites and ledeburite eutectic became coarser (Figure7). However, the di fference in the solidification conditions between the two analyzed cases at a heat input of 600 J/mm (SMB1 and SMB2, Tables3 and4) had no marked influence on the dimensions of the solidification structure. The metallographic data presented in Table4 indicate a strong correlation between the fraction of the retained austenite (γr) and the cooling rate. Generally, an increase in the cooling rate increases the γr fraction. At a given heat input, the γr fraction is affected by the traverse speed. When comparing the microstructural data of SMB1 and SMB2, a larger fraction of γr existed in the SMB2 microstructure. When considering that the cooling rate behind the molten pool was essentially the same in both cases, this difference can be attributed to the different cooling rates during solidification (at the trailing edge of the molten pool). In both the casting process and laser surface treatment of cast irons, it has been reported that under non-equilibrium cooling conditions, the austenite becomes supersaturated with carbon and suppresses the martensitic transformation [24–26]. Thus, it seems very reasonable to assume that the faster cooling rates during solidification increased the carbon content in the austenite phase and inhibited the martensitic transformation. This assumption is supported by the values of the γr austenite lattice parameter, which were estimated to be 0.3597 and 0.3611 nm in SMB1 and SMB2, respectively. This confirms an increase in the carbon content in the γr austenite as the cooling rate increased during solidification. Considering the microstructural parameters of SMB3, it is reasonable to conclude that the cooling rates during the fabrication of this melted zone ensured a nearly complete martensitic transformation of the γp and γe austenite phases. Materials 2020, 13, x FOR PEER REVIEW 8 of 14

Materials 2020, 13, x FOR PEER REVIEW 9 of 14 (a) (b)

(c) Figure 7. Backscattered electron (BSE) SEM micrographs that were taken of the undersurface area of Figure 7. Backscattered electron (BSE) SEM micrographs that were taken of the undersurface area of (a) SMB1, (b) SMB2, and (c) SMB3 (Table 4). (a) SMB1, (b) SMB2, and (c) SMB3 (Table4).

TableTable 4. Microstructural parameters of SMBs.

Bead Processing Conditions α-Fe (Martensite) Retained Austenite EutecticEutectic Cementite Processing Retained DesignationBead No. (Table2) αFraction-Fe (Martensite) (wt %) Fraction (wt %) FractionCementite (vol %) Conditions no. Austenite Fraction DesignationSMB1 A1Fraction 25 3(wt %) 22 3 53Fraction4 ± ± ± SMB2(Table A2 2) 16 3 30 (wt3 %) 52 4 ± ± (vol± %) SMB3 A3 44 4 3 1 53 5 SMB1 A1 25± ± 3 ± 22 ± 3 ± 53 ± 4 SMB2 A2 16 ± 3 30 ± 3 52 ± 4 SMB3Detailed SEM investigations A3 (Figures8–10) revealed44 ± 4 the presence of 3 bright± 1 nanometric precipitates 53 ± 5

that exhibited plate-like and spherical morphologies in the γp and γe regions. Figure 11 shows the TEM micrographs of the γp region (grain) in SMB3. The plate-like and spherical precipitates were identiﬁed as the cementite phase by SAED. As shown in Figures8–10, in the entire range of investigated solidiﬁcation conditions, a continuous network of plate-like cementite precipitates was formed in both the γp and γe regions. Spherical cementite precipitates were also observed in all SMBs; however, SMB2—the bead that was produced at the highest cooling rate—showed the lowest fraction of spherical cementite in both the γp and γe regions. This indicates that increasing the cooling rate inhibited the precipitation of spherical cementite. According to a TEM micrograph (Figure 11), plate-like cementite

(a) (b)

Materials 2020, 13, x FOR PEER REVIEW 9 of 14

(c) Figure 7. Backscattered electron (BSE) SEM micrographs that were taken of the undersurface area of (a) SMB1, (b) SMB2, and (c) SMB3 (Table 4).

Table 4. Microstructural parameters of SMBs.

Eutectic Materials 2020, 13, 1174 Processing Retained 10 of 13 Bead α-Fe (Martensite) Cementite Conditions no. Austenite Fraction Designation Fraction (wt %) Fraction (Table 2) (wt %) precipitates were located at the martensite lath boundaries, whereas spherical precipitates(vol seemed %) to be locatedSMB1 at both the martensiteA1 lath boundaries25 ± and3 inside the martensite22 ± 3 laths. Note that53 ± the 4 γr austeniteSMB2 fraction in SMB3 wasA2 negligible. Based16 on ± these 3 observations,30 it is± reasonable3 to assume52 ± 4 that the nanometricSMB3 plate-like andA3 spherical precipitates44 were ± 4 secondary and tertiary3 ± 1 cementite, respectively.53 ± 5

Materials 2020, 13, x FOR PEER REVIEW 10 of 14

(a) (b)

(c) (d) FigureFigure 8. 8. In-lIn-lensens SEM SEM images images ta takenken of of SMB3 SMB3 (~150 (~150 µmµm beneath beneath the the bead bead su surface)rface) showing showing (a) ( ath)e the

distribution of ternary eutectic α-Fe/Fe3C/Fe8Si2C; (b) a detail from (a) showing the morphology of distribution of ternary eutectic α-Fe/Fe3C/Fe8Si2C; (b) a detail from (a) showing the morphology of the α-Fe/Fe3C/Fe8Si2C eutectic (marked area); (c,d) a distribution of plate-like and spherical cementite the α-Fe/Fe3C/Fe8Si2C eutectic (marked area); (c,d) a distribution of plate-like and spherical Materialsprecipitates 2020, 13, x FOR in (c) PEER the primaryREVIEW austenite grain and (d) eutectic austenite regions (Table4). 11 of 14 cementite precipitates in (c) the primary austenite grain and (d) eutectic austenite regions (Table 4). The metallographic data presented in Table 4 indicate a strong correlation between the fraction of the retained austenite (γr) and the cooling rate. Generally, an increase in the cooling rate increases the γr fraction. At a given heat input, the γr fraction is affected by the traverse speed. When comparing the microstructural data of SMB1 and SMB2, a larger fraction of γr existed in the SMB2 microstructure. When considering that the cooling rate behind the molten pool was essentially the same in both cases, this difference can be attributed to the different cooling rates during solidification (at the trailing edge of the molten pool). In both the casting process and laser surface treatment of cast irons, it has been reported that under non-equilibrium cooling conditions, the austenite becomes supersaturated with carbon and suppresses the martensitic transformation [24–26]. Thus, it seems very reasonable to assume that the faster cooling rates during solidification increased the carbon content in the austenite phase and inhibited the martensitic transformation. This assumption is supported by the values of the γr austenite lattice parameter,(a) which were estimated to be 0.3597 and(b) 0.3611 nm in SMB1 and SMB2, Figurerespectively. 9. In-lens This SEM confirms images an showing increase thein thedistribution carbon content of plate-like in the andγr austenite spherical as cementite the cooling Figure 9. In-lens SEM images showing the distribution of plate-like and spherical cementite precipitates rate increasedprecipitates during throughout solidification. the undersurface Considering area of the SMB1 microstructural (~150 µm beneath parameters the bead ofsurface): SMB3, ( ait) is throughout the undersurface area of SMB1 (~150 µm beneath the bead surface): (a) primary austenite reasonable to conclude that the cooling rates during the fabrication of this melted zone ensured a grain;primary (b) austenite ledeburite grain; eutectic (b) region.ledeburite eutectic region. nearly complete martensitic transformation of the γp and γe austenite phases. Detailed SEM investigations (Figures 8–10) revealed the presence of bright nanometric precipitates that exhibited plate-like and spherical morphologies in the γp and γe regions. Figure 11 shows the TEM micrographs of the γp region (grain) in SMB3. The plate-like and spherical precipitates were identified as the cementite phase by SAED. As shown in Figures 8–10, in the entire range of investigated solidification conditions, a continuous network of plate-like cementite precipitates was formed in both the γp and γe regions. Spherical cementite precipitates were also observed in all SMBs; however, SMB2—the bead that was produced at the highest cooling rate— showed the lowest fraction of spherical cementite in both the γp and γe regions. This indicates that increasing the cooling rate inhibited the precipitation of spherical cementite. According to a TEM micrograph (Figure 11), plate-like cementite precipitates were located at the martensite lath boundaries, whereas spherical precipitates seemed to be located at both the martensite lath boundaries and inside the martensite laths. Note that the γr austenite fraction in SMB3 was negligible. (a) (b) Based on these observations, it is reasonable to assume that the nanometric plate-like and spherical Figure 10. In-lens SEM images showing the distribution of plate-like and spherical cementite precipitates were secondary and tertiary cementite, respectively. precipitates throughout the undersurface area of SMB2 (~150 µm beneath the bead surface): (a) primary austenite grain; (b) ledeburite eutectic region.

3.4. Hardness Analysis The hardness profiles of the SMBs are compared in Figure 12. It is interesting that, despite the considerable differences in the solidification conditions, the overall hardness profiles of the SMBs were quite similar. Due to the high fraction of the eutectic cementite, the hardness of the melted zone was influenced by both the γr fraction and the morphology of the ledeburite eutectic. The formation of very fine ledeburite due to the high cooling rate during solidification elevated the overall hardness of the melted zone. The effect of the ledeburite eutectic morphology on the hardness of the melted zone can be observed when comparing hardness profiles of SMB1 and SMB3. Despite its significantly higher γr fraction, SMB1 showed a higher average hardness (895 HV0.2) than SMB3 (837 HV0.2). This difference can be directly attributed to a difference in the morphology of the ledeburite eutectic. In turn, comparing the hardness data of SMB1 and SMB2, i.e., beads with a similar ledeburite eutectic morphology, shows that increasing the γr fraction decreased the overall hardness of the melted zone.

Materials 2020, 13, x FOR PEER REVIEW 11 of 14

(a) (b) Figure 9. In-lens SEM images showing the distribution of plate-like and spherical cementite Materialsprecipitates2020, 13, 1174 throughout the undersurface area of SMB1 (~150 µm beneath the bead surface): (a11) of 13 primary austenite grain; (b) ledeburite eutectic region.

(a) (b) Figure 10. In-lens SEM images showing the distribution of plate-like and spherical cementite Figure 10. In-lens SEM images showing the distribution of plate-like and spherical cementite precipitates precipitates throughout the undersurface area of SMB2 (~150 µm beneath the bead surface): (a) throughout the undersurface area of SMB2 (~150 µm beneath the bead surface): (a) primary austenite primary austenite grain; (b) ledeburite eutectic region. Materialsgrain; 2020 (,b 13) ledeburite, x FOR PEER eutectic REVIEW region. 12 of 14 3.4. Hardness Analysis The hardness profiles of the SMBs are compared in Figure 12. It is interesting that, despite the considerable differences in the solidification conditions, the overall hardness profiles of the SMBs were quite similar. Due to the high fraction of the eutectic cementite, the hardness of the melted zone was influenced by both the γr fraction and the morphology of the ledeburite eutectic. The formation of very fine ledeburite due to the high cooling rate during solidification elevated the overall hardness of the melted zone. The effect of the ledeburite eutectic morphology on the hardness of the melted zone can be observed when comparing hardness profiles of SMB1 and SMB3. Despite its significantly higher γr fraction, SMB1 showed a higher average hardness (895 HV0.2) than SMB3 (837 HV0.2). This difference can be directly attributed to a difference in the morphology of the ledeburite eutectic. In turn, comparing the hardness data of SMB1 and SMB2, i.e., beads with a similar ledeburite eutectic morphology, shows that increasing the γr fraction decreased the overall hardness of the melted zone.

(a) (b) Figure 11. (a) Bright field and (b) dark-field TEM images showing the nanometric plate-like (P) and Figure 11. (a) Bright ﬁeld and (b) dark-ﬁeld TEM images showing the nanometric plate-like (P) and spherical (S) cementite precipitates in the primary austenite grains in SMB3 (Table 4). spherical (S) cementite precipitates in the primary austenite grains in SMB3 (Table4).

3.4. Hardness Analysis

The hardness profiles of the SMBs are compared in Figure 12. It is interesting that, despite the considerable differences in the solidification conditions, the overall hardness profiles of the SMBs were quite similar. Due to the high fraction of the eutectic cementite, the hardness of the melted zone was influenced by both the γr fraction and the morphology of the ledeburite eutectic. The formation of very fine ledeburite due to the high cooling rate during solidification elevated the overall hardness of the melted zone. The effect of the ledeburite eutectic morphology on the hardness of the melted zone can be observed when comparing hardness profiles of SMB1 and SMB3. Despite its significantly higher γr fraction, SMB1 showed a higher average hardness (895 HV0.2) than SMB3 (837 HV0.2). This difference can be directly attributed to a difference in the morphology of the ledeburite eutectic. In turn, comparing the hardness data of SMB1 and SMB2, i.e., beads with a similar ledeburite eutectic morphology, shows that increasing the γr fraction decreased the overall hardness of the melted zone. Figure 12. Hardness profiles of SMBs (Table 4).

4. Conclusions The influence of solidification conditions on the microstructure and hardness of laser-surface- melted DCI was investigated. The thermal conditions in the molten pool were examined with IR thermography. The results showed that increasing the cooling rate during solidification increased the carbon content in the austenite phase, which inhibited the martensitic transformation. Additionally, increasing the cooling rate during solidification refined the ledeburite eutectic. The hardness of the melted zone was affected by the fraction of the retained austenite, as well as the morphology of ledeburite eutectic. Due to the high fraction of ledeburite, its morphology had a significant influence on the hardness of the melted zone. Under the investigated processing conditions, the solidification of the melted zone terminated with the Liquid → γ-Fe + Fe3C + Fe8Si2C reaction. Furthermore, it was revealed that the primary and eutectic austenite regions contained a network of plate-like secondary cementite precipitates and nanometric spherical tertiary cementite precipitates. Increasing the cooling rate inhibited the precipitation of the spherical tertiary cementite.

Author Contributions: Conceptualization, D.J.; methodology, D.J.; formal analysis, D.J., J.G. and W.K; investigation, D.J., J.G., W.K., W.P. and K.M.; resources, D.J.; writing—original draft preparation, D.J.; writing— review and editing, D.J.; visualization, D.J.; supervision, D.J.; project administration, D.J.; funding acquisition, D.J., J.G. and K.M. All authors have read and agreed to the published version of the manuscript.

Materials 2020, 13, x FOR PEER REVIEW 12 of 14

(a) (b) MaterialsFigure2020 11., 13 (a,) 1174 Bright field and (b) dark-field TEM images showing the nanometric plate-like (P) and 12 of 13 spherical (S) cementite precipitates in the primary austenite grains in SMB3 (Table 4).

FigureFigure 12. 12.HardnessHardness profiles proﬁles of SMBs of SMBs (Table (Table 4).4 ).

4. Conclusions 4. ConclusionsThe influence of solidification conditions on the microstructure and hardness of laser-surface-meltedThe influence of solidification DCI was investigated. conditions on Thethe microstructure thermal conditions and hardness in the molten of laser-surface- pool were meltedexamined DCI withwas investigated. IR thermography. The thermal The resultsconditio showedns in the that molten increasing pool were the coolingexamined rate with during IR thermography.solidification increasedThe results the showed carbon that content increasing in the the austenite cooling phase, rate during which solidification inhibited the increased martensitic the carbontransformation. content in Additionally,the austenite phase, increasing which the inhibited cooling ratethe martensitic during solidification transformation. refined Additionally, the ledeburite increasingeutectic. Thethe hardnesscooling rate of the during melted solidification zone was a ffreectedfinedby the the ledeburite fraction ofeutectic. the retained The hardness austenite, of as the well meltedas the morphologyzone was affected of ledeburite by the eutectic.fraction Dueof the to theretained high fractionaustenite, of ledeburite,as well as itsthe morphology morphology had of a ledeburitesignificant eutectic. influence Due on to the the hardness high fraction of the meltedof ledeburite, zone. Underits morphology the investigated had a significant processing influence conditions, the solidification of the melted zone terminated with the Liquid γ-Fe + Fe C + Fe Si C reaction. on the hardness of the melted zone. Under the investigated processing→ conditions,3 the solidification8 2 ofFurthermore, the melted zone it was terminated revealed with that thethe primaryLiquid → and γ-Fe eutectic + Fe3C austenite + Fe8Si2C regionsreaction. contained Furthermore, a network it was of revealedplate-like that secondary the primary cementite and eutectic precipitates austenite and re nanometricgions contained spherical a network tertiary of cementiteplate-like precipitates.secondary cementiteIncreasing precipitates the cooling an rated inhibitednanometric the precipitationspherical tertiary of the ceme sphericalntite tertiaryprecipitates. cementite. Increasing the cooling rate inhibited the precipitation of the spherical tertiary cementite. Author Contributions: Conceptualization, D.J.; methodology, D.J.; formal analysis, D.J., J.G. and W.K; Authorinvestigation, Contributions: D.J., J.G., Conceptualization, W.K., W.P. and K.M.;D.J.; methodology, resources, D.J.; D.J. writing—original; formal analysis, draftD.J., preparation,J.G. and W.K; D.J.; writing—review and editing, D.J.; visualization, D.J.; supervision, D.J.; project administration, D.J.; funding investigation, D.J., J.G., W.K., W.P. and K.M.; resources, D.J.; writing—original draft preparation, D.J.; writing— acquisition, D.J., J.G. and K.M. All authors have read and agreed to the published version of the manuscript. review and editing, D.J.; visualization, D.J.; supervision, D.J.; project administration, D.J.; funding acquisition, D.J.,Funding: J.G. andPublication K.M. All supportedauthors have by ownread scholarshipand agreed fundto the of published the Silesian version University of the of manuscript. Technology, in a year 2018. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Gadag, S.P.; Srinivasan, M.N.; Mordike, B.L. Eﬀect of laser processing parameters on the structure of ductile iron. Mater. Sci. Eng. A 1995, A196, 145–151. [CrossRef] 2. Grum, J.; Sturm, R. Microstructure analysis of nodular iron 400-12 after laser surface melt hardening. Mater. Charact. 1996, 37, 81–88. [CrossRef] 3. Alabeedi, K.F.; Abboud, J.H.; Benyounis, K.Y. Microstructure and erosion resistance enhancement of nodular cast iron by laser melting. Wear 2009, 266, 925–933. [CrossRef] 4. Janicki, D. Direct diode laser surface melting of nodular cast iron. Appl. Mech. Mater. 2015, 809–810, 423–428. [CrossRef] 5. Pagano, N.; Angelini, V.; Ceschini, L.; Campana, G. Laser remelting for enhancing tribological preferences of a ductile iron. Procedia CIRP 2016, 41, 987–991. [CrossRef] 6. Steen, W.M. Laser surface treatment. In Laser Processing: Surface Treatment and Film Deposition, NATO ASI Series; Mazumer, J., Conde, O., Villar, R., Steen, W., Eds.; Springer: Berlin/Heidelberg, Germany, 1996; Volume 307, pp. 1–20. Materials 2020, 13, 1174 13 of 13

7. Lisiecki, A. Titanium matrix composite Ti/TiN produced by diode laser gas nitriding. Metals 2015, 5, 54–69. [CrossRef] 8. Sun, G.; Zhou, R.; Li, P.; Feng, A.; Zhang, Y. Laser surface alloying of C-B-W-Cr powders on nodular cast iron rolls. Surf. Coat. Tech. 2011, 205, 2747–2754. [CrossRef] 9. Tański,T.; Pakieła, W.; Janicki, D.; Tomiczek, B.; Król, M. Properties of the aluminium alloy EN AC-51100 after laser surface treatment. Arch. Metall. Mater. 2016, 61, 199–204. [CrossRef] 10. Lin, P.Y.; Zhang, Z.H.; Kong, S.H.; Zhou, H.; Liang, Y.; Tong, X.; Ren, X.Q. Mechanical Properties of Vermicular Graphite Cast Iron Processed by Selective Laser Surface Alloying with Ultra-fine ZrO Ceramic Particulates. Acta Metall. Sin. Engl. 2016, 29, 985–992. [CrossRef] 11. Janicki, D. Microstructure and sliding wear behaviour of in-situ TiC-reinforced composite surface layers fabricated on ductile cast iron by laser alloying. Materials 2018, 11, 75. [CrossRef] 12. Chen, C.H.; Ju, C.P.; Rigsbee, J.M. Laser surface modification of ductile iron: Part 1 microstructure. Mater. Sci. Technol. 1988, 4, 161–166. [CrossRef] 13. Fernández-Vicente, A.; Pellizzari, M.; Arias, J.L. Feasibility of laser surface treatment of pearlitic and bainitic ductile irons for hot rolls. J. Mater. Process. Tech. 2012, 212, 989–1002. [CrossRef] 14. Usamentiaga, R.; Venegas, P.; Guerediaga, J.; Vega, L.; Molleda, J.; Bulnes, F.G. Infrared Thermography for Temperature Measurement and Non-Destructive Testing. Sensors 2014, 14, 12305–12348. [CrossRef] [PubMed] 15. Górka, J.; Janicki, D.; Fidali, M.; Jamrozik, W. Thermographic Assessment of the HAZ Properties and Structure of Thermomechanically Treated Steel. Int. J. 2017, 38, 183. [CrossRef] 16. Janicki, D. Fabrication of high chromium white iron surface layers on ductile cast iron substrate by laser surface alloying. Stroj. Vestn. 2017, 63, 705–714. [CrossRef] 17. Michalski, L.; Eckersdorf, K.; Kucharski, J.; McGhee, J. Temperature Measurements, 2nd ed.; Wiley and Sons: Hoboken, NJ, USA, 2001. 18. Rai, R.; Roy, G.G.; DebRoy, T. A computationally efficient model of convective heat transfer and solidification characteristics during keyhole mode laser welding. J. Appl. Phys. 2007, 101, 054909. [CrossRef] 19. Kik, T.; Moravec, J.; Nováková, I. Application of Numerical Simulations on 10GN2MFA Steel Multilayer Welding. In Dynamical Systems in Applications; Awrejcewicz, J., Ed.; Springer: Berlin/Heidelberg, Germany, 2018; Volume 249, pp. 193–204. [CrossRef] 20. Trapp, J.; Rubenchik, A.M.; Guss, G.; Matthews, M.J. In situ absorptivity measurements of metallic powders during laser powder-bed fusion additive manufacturing. Appl. Mater. Today 2017, 9, 341–349. [CrossRef] 21. Lacaze, J.; Sundman, B. An assessment of the Fe-C-Si system. Metall. Trans. A 1991, 22, 2211–2223. [CrossRef] 22. González-Martínez, R.; de la Torre, U.; Lacaze, J.; Sertucha, J. Effects of high silicon contents on graphite morphology and room temperature mechanical properties of as-cast ferritic ductile cast irons. Part I–Microstructure. Metall Trans. A 2018, 712, 794–802. [CrossRef] 23. Lin, Y.C.; Chen, H.M.; Chen, Y.C. Analysis of microstructure and wear performance of SiC clad layer on SKD61 die steel after gas tungsten arc welding. Mater. Des. 2013, 47, 828–835. [CrossRef] 24. Zarubova, N.; Kraus, V.; Cermak, J. Mechanisms of phase transformations during laser treatment of grey cast iron. J. Mater. Sci. 1992, 27, 3487–3496. [CrossRef] 25. Janus, A.; Stachowicz, M. Thermodynamic stability of austenitic Ni-Mn-Cu cast iron. Metalurgija 2014, 3, 353–356. 26. Davis, J.R. (Ed.) Cast Irons. ASM Speciality Handbook; ASM International: Novelty, OH, USA, 1996.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).