Microstructure and Microhardness of 3D-Printed Samples of Steel X45CrSi 9-3

Abstract

The main aim of this study is to investigate the microstructure and microhardness of Wire Arc Additive Manufactured (WAAM) samples produced under different layer deposition strategies and corresponding interlayer temperature conditions. Experimental samples were produced using the WAAM process with X45CrSi9-3 steel. During the experiments, both the number of layers and the thermal conditions (heating and cooling) were systematically varied. This was achieved by fabricating samples consisting of five layers with three beads per layer. The layer deposition procedure was implemented in two different ways: (i) with a waiting period after each layer to allow cooling to room temperature, and (ii) without such a waiting period. Thermal cycles at selected locations within the samples were calculated using simulation modeling. By combining these thermal cycles with the continuous cooling transformation (CCT) diagram, the expected microstructures in the vicinity of these locations were determined. These predictions were supplemented by microstructural analysis and hardness measurements. Particular emphasis was placed on the influence of interlayer temperature and repeated heating and cooling cycles. The analyses enabled the identification of process parameters that facilitate control over microstructure, microhardness, and property gradients. It can be concluded that the interlayer holding time provides an effective means of controlling the microstructure of the workpiece, ranging from predominantly austenitic to predominantly martensitic. Depending on the thermal cycles, the measured microhardness varied within the range of 360–900 HV. Metallographic examination revealed a wide spectrum of non-equilibrium microstructures, including martensite with varying degrees of tempering, retained austenite, pearlite, and bainite. The application of a thermal model to the conducted experiments, combined with the CCT diagram, indicated that the expected microstructures consist predominantly of martensite with varying degrees of tempering, retained austenite, carbides, and, in some cases, up to 5% pearlite.

1. Introduction

The development of additive manufacturing (AM) technologies has created new opportunities for producing components from materials that were previously considered unsuitable for such processes. However, parts and test samples fabricated by layer-by-layer deposition exhibit specific characteristics that are inherently linked to the manufacturing process. These characteristics are strongly influenced by the selected process parameters, which necessitates a detailed investigation of the resulting material properties under additive manufacturing conditions. WAAM technology has been applied to a wide range of materials, including high-strength low-alloy (HSLA) steels [1], stainless steels [2], Ni–Ti alloys [3], low-carbon steels [4], and duplex stainless steels [5]. In the present study, welding wire made of AX600 steel (1.4718–X45CrSi9-3) was used as the feedstock material. At room temperature, the material exhibits a yield strength of approximately 600 MPa and an ultimate tensile strength in the range of 800–900 MPa [6]. In addition, it demonstrates a relative elongation of about 20% and a reduction in area of approximately 40%. At elevated temperatures, the steel retains relatively high strength, which makes it suitable for applications involving thermal loading [6]. The high silicon content enhances oxidation resistance, making the material particularly suitable for use in oxidizing environments under cyclic thermal conditions. Due to this combination of properties, AX600 steel is widely used for highly loaded components operating at elevated temperatures, such as internal combustion engine valves. Its excellent heat resistance, high strength, and good resistance to abrasive wear and oxidation make it especially suitable for applications in the automotive and aerospace industries, as well as in power engineering, where components are exposed to severe thermal and mechanical loads [7]. Additive and hybrid manufacturing approaches have already been applied in the production of various engineering components, including turbine shafts fabricated by surfacing on 20MnCr5 steel [8], gears [9,10,11,12,13,14,15], shafts and bearing housings [16], and bearing washers [17]. In addition, laboratory-scale samples, including multilayer functionally graded materials, have been produced using surfacing techniques [18,19,20,21]. Both experimental samples and in-service components are of particular interest for evaluating microstructural evolution and property degradation under operational conditions [22,23]. Recent studies have also focused on samples produced by additive manufacturing technologies, including WAAM, which enable controlled layer-by-layer structure formation [24]. A key technological concept underlying many of these studies is “Tailored Forming”, which represents an integrated process chain involving the joining of dissimilar materials (through additive or layering techniques), followed by hot forming and, in many cases, subsequent heat treatment [9,12,14]. Within this framework, laser hot-wire cladding plays a central role, as it enables the formation of high-quality metallurgical bonds between layers while providing precise control over layer thickness and chemical composition [9,11,16]. This technology is often combined with additional processes such as hot volumetric deformation (e.g., die forging) [10,14], cross-wedge rolling [8,16], and machining and finishing operations

Additive manufacturing technologies also play a significant role in this context, including:

• Directed Energy Deposition (DED), as well as its twin-wire variants, for the production of functionally graded materials [18,19,25] and
• Wire Arc Additive Manufacturing (WAAM) [26].

Other methods employed in this field include friction stir processing, plasma arc additive manufacturing, and powder metallurgy [20,25,27,28]. An important aspect of current research is the application of numerical modeling, particularly finite element methods (FEM), to analyze thermal processes, material flow, and the distribution of properties in the resulting components [13,15,24].

Despite these advances, there is still a lack of comprehensive data on the properties of materials produced by Wire Arc Additive Manufacturing (WAAM) using X45CrSi9-3 steel as a filler material.

Given the similarities between WAAM and conventional surfacing and welding processes, the Scheffler diagram can be used as an initial tool to predict the expected microstructure. According to this approach, the microstructure is expected to consist of austenite and martensite. However, due to the relatively high contents of carbon, chromium, and silicon, a more detailed analysis is required. This is due to the potential formation of various carbides during the initial solidification of the melt, their subsequent modification, and the possible precipitation of new phases during repeated thermal cycles. In addition, the presence of significant amounts of chromium and silicon may promote the formation of silicides. It should also be noted that, in the WAAM process, cooling rates can be extremely high, while the residence time within specific temperature ranges strongly depends on the layer deposition strategy. Since repeated thermal cycling is an inherent characteristic of WAAM, the resulting microstructural evolution under such conditions is of considerable scientific and practical interest.

Microstructural studies demonstrate a clear relationship between the applied processing technology and the resulting microstructure. In surfaced layers of X45CrSi9-3 steel, a predominantly martensitic structure is typically formed, characterized by high hardness and wear resistance [8,11]. When combined processes such as hardfacing followed by hot deformation are applied, recrystallization and grain refinement occur, leading to improved structural homogeneity and the elimination of defects under optimal processing conditions [12,14]. In additively manufactured and multilayer structures, a relatively uniform distribution of alloying elements is observed, along with smooth compositional gradients in functionally graded materials [18,19]. In regions affected by hardfacing and thermal exposure, processes such as elemental diffusion and the formation of carbide phases take place [21,29]. Under conditions of elevated temperatures and cyclic loading, microstructural degradation is observed, including grain growth and carbide coarsening, which ultimately leads to a deterioration of mechanical properties [22].

Jérémie Bouquerel and co-authors conducted a comparative study of the microstructure of two identical internal combustion engine valves, one of which had been in continuous operation while the other remained unused [22]. The results showed that the unused valve exhibited a finer-grained microstructure with significantly smaller carbides. In contrast, in the used valve, the carbides (Fe,Cr)₇C₃ increased in size by a factor of two to three, and additional carbides of the type (Fe,Cr)₂₃C₆ were observed at the grain boundaries. This evolution was primarily attributed to repeated thermal exposure at temperatures on the order of 600 °C.

The CCT diagram of the steel X45CrSi9-3 presented in [6] was used to determine the expected microstructure. Several key features can be identified. The start temperature of martensitic transformation (M_S) is indicated, whereas the finish temperature (M_F) is not specified. When the cooling time from A_C3 to 600 °C is less than 11 s and no holding occurs within the temperature range of 300–600 °C, the resulting microstructure falls within the martensite + carbides region, with a microhardness exceeding 650 HV. Under conditions of continuous cooling from A_C3 to 300 °C over a period exceeding 400 s, the resulting microstructure consists of pearlite, retained austenite, and carbides. At intermediate cooling rates, mixed microstructures containing pearlite, martensite, retained austenite, and carbides are formed. In contrast, rapid cooling to 600 °C (suppressing pearlite formation), followed by holding for more than 500 s within the temperature range of 300–600 °C, leads to a microstructure composed predominantly of retained austenite and carbides. It should also be noted that subsequent reheating to approximately 600 °C may result in tempering of the martensite, further influencing the final microstructure and properties.

The absence of a clearly defined martensitic transformation finish (M_F) curve provides a basis for investigating microstructural evolution at lower temperatures. Jaswin et al. [30] studied the influence of cryogenic treatment on the microstructure and wear resistance of X45CrSi9-3 and X53Cr22Mn9Ni4N steels. The materials were subjected to sub-zero cryogenic treatments at 193 K (shallow cryogenic treatment, SCT) and 85 K (deep cryogenic treatment, DCT). The resulting microstructures were analyzed using optical microscopy and scanning electron microscopy (SEM—Hitachi 5× to 300,000×, Gurgaon, India, [30]) and compared with those obtained after conventional heat treatment. The authors reported that complete elimination of retained austenite was not achieved in either SCT- or DCT-treated samples. However, a reduction in the amount of retained austenite was observed compared to conventionally heat-treated samples. Furthermore, it was concluded that cryogenic treatment promotes the precipitation of finer carbides with a higher volume fraction and a more uniform distribution throughout the microstructure. The formation of these ultrafine carbides within the martensitic matrix contributes to an increase in the hardness of cryogenically treated steels.

Behrens and co-workers investigated the microstructure and microhardness of shafts and gears [16] manufactured from C22.8 steel (AISI 1022M) with surfaced layers of X45CrSi9-3 steel (AISI HNV3). Following surfacing, the components were subjected to an integrated thermomechanical processing route involving forging and heat treatment. As a result of this combined processing, recrystallization and significant grain refinement were observed, accompanied by increases in hardness and tensile strength. The analysis indicates that the initially coarse bead microstructure underwent recrystallization due to the applied thermomechanical treatment. Among the process parameters, the cooling strategy was found to have the most pronounced influence on the resulting hardness and tensile strength. High cooling rates, achieved through air–water quenching, were identified as the primary strengthening mechanism. Furthermore, the integration of heat treatment into the forging process—via water–air jet quenching followed by self-tempering—enables effective control over the mechanical properties of both the deposited metal and the base metal. This technological combination allows for the production of components with tailored hardness gradients through the use of dissimilar materials and customized time–temperature profiles. A related study [16] also emphasizes the distribution of residual stresses. In both cases, achieving the desired microstructure is governed by precise control of the thermomechanical cycle.

The results of hardness and microhardness measurements demonstrate a significant improvement in surface properties when applying deposited layers. In several studies, microhardness values of up to 700–715 HV have been reported, which are characteristic of martensitic microstructures [9,11]. It has been established that hardness is strongly influenced by several factors, including the coating thickness and its distribution [8,10], the cooling regime after hot deformation [8], and the parameters of subsequent heat treatment [23,31,32]. In functionally graded materials, a gradual variation in hardness along the cross-section is typically observed, which reduces the likelihood of stress concentration [18,19]. However, during long-term exposure to elevated temperatures, a reduction in microhardness has been reported [22].

Wear resistance is one of the key properties improved through the use of hybrid materials and deposited coatings. The results indicate that the local application of high-strength materials in highly loaded zones leads to a significant increase in service performance and operational resistance [17]. Studies on additively manufactured samples further demonstrate that technological parameters have a strong influence on wear behavior, with optimized processing regimes resulting in improved wear resistance and stable tribological performance [26]. In addition, it has been established that higher hardness generally correlates with improved wear resistance. The microstructure—particularly the presence of martensite and carbide phases—plays a crucial role in determining tribological behavior. Furthermore, fatigue and frictional resistance are enhanced in multilayer and functionally graded structures [28,33]. Tribological testing of real components, such as bearing washers, has confirmed that hybrid structures can successfully withstand cyclic loading and contact fatigue [17].

The analysis of the reviewed literature indicates that current research is primarily focused on surfacing techniques for the production of hybrid-material components. As is well established, this approach offers significant economic advantages. The combination of different materials enables local optimization of properties, resulting in improved performance and reduced manufacturing costs [9,12,17]. From a technological perspective, laser-based surfacing processes dominate the field. In contrast, studies on additive manufacturing methods using X45CrSi9-3 steel remain limited. Although laser technologies are widely applied, and some works report their combination with WAAM, systematic investigations of WAAM with this steel grade are still scarce. These technologies typically require complex equipment setups and precise control of process parameters to ensure appropriate thermal conditions, cooling rates, and material deposition [18,19].

Microstructural analyses generally show relatively consistent results, namely the formation of martensitic structures and an associated improvement in mechanical properties. However, some studies report undesirable microstructural changes, such as grain coarsening and carbide growth at elevated temperatures and during prolonged service [22]. This raises important questions regarding the evolution of properties during the additive manufacturing process and their long-term stability, which are not sufficiently addressed in the majority of the available literature.

With regard to hardness and microhardness, the results generally correlate well with the observed microstructure. Nevertheless, it has been reported in some cases that high hardness does not necessarily ensure high wear resistance, particularly in the presence of structural defects or residual stresses [3,16]. This highlights the need for a more comprehensive approach to property evaluation.

Overall, despite significant progress in the field, several key challenges remain, which are addressed in this study:

• insufficient available data on the material properties developed during additive manufacturing processes and
• the need for improved integration of experimental investigations and numerical modeling approaches.

As previously noted, a key feature of Wire Arc Additive Manufacturing (WAAM) is the pronounced difference in the microstructure and properties of the deposited metal compared to those obtained through conventional plastic deformation processing. In addition, WAAM directly produces near-net-shape components rather than intermediate products from which parts are subsequently manufactured. This makes the investigation and control of the resulting material properties critically important. The literature review indicates that the WAAM processing of samples or components fabricated from AX600 steel remains insufficiently investigated. The aim of this study is to investigate the microstructure and microhardness of WAAM-fabricated samples produced under different layer deposition strategies and corresponding interlayer temperature conditions. In parallel, numerical simulations of thermal processes are employed to predict the expected microstructural evolution. To validate these predictions, experimental microhardness measurements were performed on 3D-printed samples, accompanied by metallographic analysis of the resulting microstructures.

2. Materials and Methods

The experiments were performed using Wire Arc Additive Manufacturing (WAAM) with a 1 mm diameter electrode wire made of AX600 steel (1.4718–X45CrSi9-3). The chemical composition of the wire (wt%) was: 0.458 C, ≤0.6 Mn, 3.0 Si, ≤0.05 Ni, and 9.0 Cr. The deposition process was carried out using the following parameters: arc current I = 140 A, arc voltage U = 19.1 V, and travel speed V_S = 12 cm/min. A substrate made of low-carbon steel S355JR with dimensions of 15 × 40 × 400 mm used. The chemical composition of this steel was C-0.024%, Mn-1.6%, Si-0.55%, P-0.035% and S-0.035%. The deposition process was initiated along the 15 × 400 mm side of the substrate. Five layers were fabricated in total (Figure 1a), with each layer consisting of three beads. The bead deposition sequence was b1–b2–b3. After each bead deposition, a waiting time was applied before starting the next bead. The inter-bead waiting time varied between layers but remained constant within each individual layer, as follows: layer 1—230 s, layer 2—190 s, layer 3—150 s, layer 4—110 s, and layer 5—70 s.

Two deposition strategies were applied:

• Sample A (Option 1): interlayer waiting time equal to the inter-bead waiting time of the previous layer.
• Sample B (Option 2): cooling to room temperature was allowed between successive layers.

After fabrication, samples were sectioned from the deposited walls. Samples for metallographic analysis and microhardness testing were prepared from the region indicated in Figure 1b. Temperature–time profiles at selected points within the samples were evaluated using numerical simulation to analyze thermal history and cooling behavior during the deposition process.

The equipment used for WAAM fabrication and subsequent characterization of the samples is listed below:

• The WAAM process was carried out using a Parweld XTM 211 D1 power source (Parweld, Bewdley, UK) in combination with a modified Wanhao D12/500 3D printer (Wanhao, Jinhua, China) adapted for metal deposition.
• Sample preparation (sectioning, grinding, and polishing) was performed according to standard metallographic procedures, followed by chemical etching using two solutions: 5 g FeCl₃ + 30 mL HCl + 70 mL H₂O, and 4% HNO₃.
• Microhardness measurements were carried out using a UCI UT200 hardness tester (BAQ, Braunschweig, Germany).
• Microstructural analysis was performed using a ZEISS optical microscope (Jena, Germany) equipped with a digital camera and ToupView 3.7 image analysis software (ToupTek Photonics, Hangzhou, China), as well as a HIROX 5500 scanning electron microscope (HIROX Europe, Limonest, France) equipped with a BRUKER EDS system (Bruker, Bremen, Germany).

3. Results and Discussion

3.1. Thermal Analysis

In order to analyze the thermal conditions during the conducted experiments and estimate the resulting microstructure, a thermal model was developed based on an approach involving transient thermal analysis with a moving heat source, layer-by-layer geometry addition, and weak coupling with hardness and phase transformation calculations [34]. The heat source model was calibrated using the bead width and penetration depth by solving an optimization problem aimed at achieving the solidus temperature at points along the fusion line [35]. Within each bead and layer, characteristic monitoring points were defined to track the evolution of temperature during sample fabrication. The obtained thermal cycles were subsequently superimposed onto the CCT diagram in order to qualitatively assess the expected microstructural transformations. The selected monitoring points were chosen to represent the general thermal history of the deposited beads. To improve interpretability, the calculated thermal cycles are additionally presented in relation to the CCT diagram in Figure 2, Figure 3, Figure 4, Figure 5 and Figure 6.

Figure 2a,b show the temperature–time histories for a bead in the first layer (L1–b1). As can be observed, the thermal cycles exhibit similar behavior for both samples. However, in Sample B, the maximum temperatures reached during subsequent layer deposition are lower due to cooling to room temperature between the deposition of individual layers. This difference in thermal history leads to distinct thermal accumulation effects. Repeated heating and cooling create conditions favorable for martensite tempering, which is expected to influence the final microstructure and mechanical properties.

Significant differences in thermal behavior are observed during the deposition of the fifth layer. In both process variants (Figure 3a,b), no conditions favorable for subsequent tempering are present after completion of the layer deposition. A clear distinction is observed between the two samples. In Sample A, the material remains at elevated temperatures for a relatively long time before crossing the martensite start (M_S) line. This approximately 350 s dwell time in the high-temperature range is expected to promote a stabilizing effect on austenite. In addition, a very short transition through the pearlite + carbide region is observed. In contrast, in Sample B, the cooling curve intersects the M_S line already during the deposition of the first bead. As a result, tempered martensite formation can be expected due to subsequent thermal exposure during the deposition of the remaining beads within the layer. Figure 4 compares the cooling curves for characteristic points within the fifth layer of Sample A. It can be observed that the L5_b2 and L5_b3 locations experience a shorter residence time in the temperature range between 300 °C and 600 °C prior to crossing the M_S line. This behavior suggests that slightly higher microhardness values are expected in these regions compared to the material of the central bead.

Figure 5 presents the temperature–time curves for the central bead locations of layers 1 and 5 in Sample B (L1_b1 and L5_b1), highlighting clear differences in thermal history during deposition. A distinct variation in heating and cooling behavior is observed between the two layers. For the first layer (L1_b1), the thermal cycle includes a temperature range favorable for martensite tempering, indicating that subsequent thermal exposure contributes to microstructural modification of the as-formed martensitic phase. In contrast, for the fifth layer (L5_b1), after the third intersection with the martensite start (M_S) line, the cooling process continues uninterrupted down to room temperature. This indicates the absence of additional thermal cycles capable of inducing further tempering effects in this region.

Figure 6 presents the temperature–time curves for the central bead of layer 2 for Samples A and B, highlighting significant differences in thermal exposure during the WAAM process. For Sample A, after the final intersection with the martensite start (M_S) line, the temperature remains within the range of 200–400 °C for more than 2000 s. This extended residence time provides favorable conditions for martensite tempering. In addition, a short-term exposure to the pearlite + carbide transformation region is also observed. In contrast, Sample B exhibits markedly different thermal conditions. The maximum temperatures reached during subsequent layer deposition vary significantly: approximately 700 °C during bead L3_b1 of layer 3, around 500 °C during beads L3_b2 and L3_b3, 320–360 °C during layer 4, and 220–320 °C during layer 5. Overall, the residence time above 200 °C is considerably shorter compared to Sample A, resulting in reduced conditions for martensite tempering. The thermal cycles of layers 3 and 4 show similar trends, with the main difference being the reduced time spent above 200 °C in Sample B. Additionally, in Sample A, progressively higher peak temperatures are reached during the deposition of subsequent beads and layers, indicating stronger thermal accumulation.

Furthermore, the minimum temperatures after crossing the M_S line are lower in Sample B due to the reduced interlayer temperature. This promotes the formation of a higher fraction of martensite compared to Sample A.

3.2. Microhardness

The measured microhardness distribution is shown in Figure 7. The indicated boundaries between the layers were determined based on the average layer height. In sample A, microhardness values exceeding 630 HV are observed in the fourth and fifth layers, whereas reduced microhardness was detected in the central region of the fourth layer. Microhardness values below 270 HV are recorded only in the substrate and outside the penetration zone. The remaining regions of the sample exhibit alternating areas with microhardness values ranging from 270 HV to 630 HV.

In sample B, the same microhardness ranges are observed; however, their spatial distribution differs. In this case, the region with microhardness values between 270 HV and 630 HV is primarily located in the second and third layers (excluding the penetrated portion of the substrate, where the chemical composition differs).

The microhardness maps of the five layers in samples A and B (Figure 7) were analyzed using the image-processing software ImageJ (version 1.54p), and the relative area fractions (%) of regions exhibiting similar microhardness values were calculated. The results for both samples are presented graphically in Figure 8 and Figure 9. In the first three layers of sample A, the hardness is lower but uniformly distributed. In the middle layers of sample B, there are cores with lower hardness (400–500 HV), surrounded by regions of higher hardness exceeding 600 HV. The fourth and fifth layers of both samples exhibit the highest hardness, above 700 HV. The fourth layer of sample B is almost homogeneous, whereas in the center of sample A there is a zone with hardness of 450–500 HV.

The typical microhardness ranges of the main structural constituents are as follows: ferrite, 80–200 HV; pearlite, 200–300 HV; upper bainite, 300–400 HV; lower bainite, 400–600 HV; martensite, 600–1000 HV and above; austenite, 150–250 HV; retained austenite, 120–250 HV; and transformed martensite, 250–550 HV [36,37]. This allows us to infer the structural constituents from which the investigated layers are composed.

The clearly defined variation in microhardness as a function of thermal cycling indicates that the resulting microstructure, and consequently the mechanical properties of the deposited metal, can be effectively controlled by the WAAM processing parameters and the bead deposition strategy.

3.3. Microstructure

The microstructure of samples A and B was examined in the regions around the marked points along the central axis of the macrosections presented in Figure 10. The five deposited layers are clearly distinguishable on the macrosections, as they exhibit different color intensities after etching. Layers with similar morphology and spatial distribution along the sample length are also observed in the hardness distribution maps (Figure 7 and Figure 10).

The analysis of the microstructure and hardness distribution in sample A (Figure 9 and Figure 10) indicates that each deposited layer can be conditionally divided into two sublayers, which differ in both microhardness and microstructure. The lower sublayer, adjacent to the substrate, exhibits lower hardness. In the central region of each layer, an area of increased hardness is observed, followed by a decrease in hardness in the upper part of the layer. The main structural constituents throughout the sample are predominantly retained austenite and martensite, with their relative fractions and grain sizes varying across different regions.

In layer 1, the microstructure has undergone multiple thermal cycles, which influence the relative amounts of retained austenite and martensite in the two sublayers (Figure 11a,b). The highest hardness is observed in the central region of the layer, reaching approximately 600–700 HV, which corresponds to a martensitic microstructure [36,37]. A similar distribution of hardness zones is also observed in layer 2 (Figure 11c,d); however, the high-hardness region in this layer is wider than that in layer 1, with hardness values reaching 700–800 HV.

In layer 2 (Figure 11c,d), layer 3 (Figure 11e,f), and the central region along the axis of layer 4 (Figure 12a,b), the microstructure is metastable and largely similar. It consists of retained austenite, martensite, and bainite [38]. The hardness distribution is most uniform in layer 3, where values range from 400 to 600 HV. Microstructural analysis indicates that both the lower region of layer 3 around point 5 (Figure 11e) and the upper region around point 6 (Figure 11f) exhibit comparable proportions of structural constituents. In this layer, the tendency for the highest hardness zone (approximately 700 HV) to be located in the middle region is also preserved.

In the microstructure of layers 2, 3, and 4, a small amount of fine precipitates is observed along the boundaries of prior austenite grains and in their vicinity. The conducted EDS analysis indicates an increased concentration of chromium and silicon in these regions. In layer 4, at a distance of approximately 2 mm from the central axis, the hardness increases significantly, reaching values of up to 900 HV (Figure 12c,d). Subsequently, the high-hardness region gradually extends to cover the entire fifth layer. Microstructural analysis shows that layer 5 consists of fine acicular structures corresponding to a martensitic phase, as confirmed by hardness values (Figure 12e,f) [36,37].

In the macrosection of sample B (Figure 10b), dark regions resembling nucleation sites and lighter underlying layers are clearly distinguishable, resulting from etching of areas with different microstructures formed during the WAAM process. The delineation of deposited layer boundaries indicates that regions with similar microhardness are not strictly confined within individual layers, and that each subsequent layer influences the microstructure, and consequently the microhardness, of the previously deposited layers. This observation is consistent with the results of the thermal analysis. In deposited layer 1, adjacent to the metal substrate, a very fine acicular phase is observed within prior austenite grains (Figure 13a).

Hardness values in the range of 700–800 HV observed in the hardness distribution maps (Figure 7 and Figure 10) allow this region to be identified as predominantly martensitic. In the vicinity of point 2 (Figure 13b–d) and above it in layers 2 and 3, a dark phase is observed along prior austenite grain boundaries and within the grains, which is likely associated with pearlitic transformation products. The light matrix within the grains exhibits a non-equilibrium microstructure consisting of martensite, bainite, and retained austenite [39]. Around points 3 and 4, located in the second deposited layer, the microstructure is analogous to that observed at point 2 (Figure 13e,f). Similarly, points 5 and 6 (Figure 14a,b), located in the third layer, exhibit microstructural characteristics comparable to those at points 3 and 4. The microstructure around points 7, 8, 9, and 10 can be more confidently described as predominantly martensitic, with minor amounts of bainite and retained austenite (Figure 14c,f) [40]. The clearly defined variations in microstructure and microhardness across the individual deposited beads indicate that the mechanical properties of the built-up metal can be controlled through the WAAM process parameters and the selected layer deposition strategy.

4. Conclusions

The study demonstrates that the microstructure and microhardness of the investigated X45CrSi9-3 steel samples can be significantly influenced by the interlayer temperature and the thermal effects of subsequent layers.

Depending on the applied thermal cycles, the measured microhardness varies in the range of 360–900 HV. Metallographic analysis reveals a wide spectrum of non-equilibrium microstructures, including martensite with varying degrees of tempering, retained austenite, pearlite, and bainite. The application of numerical thermal analysis in combination with continuous cooling transformation (CCT) diagrams enables a comprehensive evaluation of the process. The results of both numerical and experimental analyses provide strong evidence for the formation of the expected microstructural constituents, including tempered martensite, retained austenite, and carbides. With decreasing interlayer dwell time, an increase in carbide precipitation is expected, and under certain conditions, the formation of pearlite may also occur.

Depending on the applied printing strategy, variations in microstructural distribution and material properties within the manufactured components can be achieved. It is evident that, although process parameters such as heat input and travel speed are primarily governed by the requirement for defect-free fabrication, their direct use for microstructural control becomes possible. The same generally applies to interpass delay. The interlayer dwell time appears to be a key parameter for practical control of the resulting microstructure and properties.

The good agreement between numerical and metallographic results indicates that computational modeling can be effectively applied both for predicting and for actively controlling the WAAM process in order to obtain steel components with tailored properties or property gradients.