Phase Equilibrium and Microstructure Examinations of Eutectic Fe-C-Mn-B Alloys

In this study, we analyzed the quaternary Fe-C-Mn-B system to create new eutectic cast alloys for coating deposition and additive manufacturing. Experimental samples were fabricated via the wire arc manufacturing method with argon shielding using Kemppi Pro 5200 Evolution equipment. Annealing was performed in a vacuum electric furnace at 1273 K for 350 h. For phase analyses, Jeol Superprobe 733 equipment was used. Metallographic and differential thermal analyses were used to reveal the eutectic structure of the samples. Examinations of the quaternary Fe-C-Mn-B system demonstrated that several eutectic alloys existed in the system. Four isothermal pseudo-ternary sections of the Fe-C-Mn-B system were studied: “Fe3B”-Fe3C-“Fe3Mn”; Fe2B-“Fe2C”-“Fe2Mn”; “Fe3B”-Fe3C-“Fe1.2Mn”; “Fe23B6”-“Fe23C6”-“Fe23Mn”. Broad eutectic concentrations enabled us to overcome parameter fluctuations during additive manufacturing. In each isothermal section, two dissimilar phase regions were determined: one with a ternary Fe-C-B composition and the other with a ternary Fe-C-Mn composition. Depending on the manganese content, two types of solid solutions could be formed: (Fe, Mn)α or (Fe, Mn)γ.


Introduction
As natural micro-and nanocomposites, eutectic materials have long been the focus of engineers. These materials have a narrow solidification temperature range and enable the spontaneous formation of various structural cooling rate-controlled microstructures. The ductile matrix is reinforced by hard particles or lamellas formed by carbides or intermetallics. Many excellent tools and wear-resistant materials are eutectics or sintered materials with pseudo-eutectic structures. Their good casting properties also make these materials promising for additive manufacturing. Usually, the coarse microstructure formed during solidification promotes low material toughness. Fast cooling rates, however, may help to overcome this shortcoming.
Additive manufacturing developed from well-known welding processes used for decades to restore worn machine parts. The welding seam and heat-affected zone may reduce the mechanical properties of joining parts. Weld decay corrosion and weld embrittlement often result in joint failure. Recently, metal fusion during additive manufacturing has become so controlled that it is now possible to fabricate the most complex shapes with minor material consumption. It was reported that the additive-manufactured eutectic aluminum alloy AlSi 12 has three to four times better tensile strength than conventional sand-casting [1,2]. The faster cooling rate results in higher tensile strength with minimal reduction in fatigue strength [2]. Nevertheless, both of these characteristics are remarkably higher than those of cast alloys. A faster cooling rate promotes the formation of very fine coral-like silicon eutectics in the near-eutectic alloy AlSi 10 Mg, thereby improving and predetermined useful regions to control the produced material's microstructure [13,18]. The construction of phase diagrams also helps determine the compatibility of coating and substrate materials.
Steels are often used to produce metal matrix composites reinforced by SiC, NbC, and Al 2 O 3 microparticles. Multiple carbides can be formed in Fe-Ti and Fe-Mn systems, and iron-based amorphous materials can be obtained using fast-cooling manufacturing methods [19]. The reconditioning of worn surfaces is another important application of additive manufacturing processes. In a previous study, Co-Ni secondary hardening steels were used to recover aircraft parts via the laser deposition method [20]. However, care must be taken in the condition of the substrate. For example, reheating can cause changes in the microstructure (bainite to martensite in [21]). It is vital to consider the corrosion resistance of additively manufactured (AMed) materials, as AMed stainless steels are often superior to conventionally manufactured steels [22]. The mechanical properties of AMed stainless steels are also better than those of cast steels due to rapid cooling (10 5 -10 7 K/s), resulting in a finer microstructure. The sizes of the inclusions and alloying element-depletion zones are also smaller. Attention should be given to avoid the formation of concentration and structural corrosion cells (e.g., on the boundaries of the molten pool).
Eutectoid iron-based alloys are extensively used in structural engineering, whereas eutectic alloys are mostly applied in the form of hardfacing coatings. However, for many applications (massive cutting tools, inserts for mills, etc.) working at high specific loads, a thick coating is not practicable. In such cases, it is better to use a bulk material. The machining of eutectic carbide-strengthened materials is troublesome due to the high hardness of such materials. Additive manufacturing produces net shape products for which only post-grinding is required. In contrast to conventional casting, this process results in a much finer microstructure and better service characteristics. Hence, it is vital to find eutectic concentrations of studied systems to produce tool alloys suitable for additive manufacturing. To produce test samples, it is better to preheat the substrate to avoid cracks [22]. A new layer deposition can reheat the sublayer, thereby providing tempering conditions. As the built material grows, this heat treatment effect becomes dissimilar in different layers; the top layer is fast-cooled, whereas the sublayers are tempered several times. The wear resistance of AMed H13 tool steels in [23] was found to be only one third that of conventionally manufactured counterparts.
The problem one encounters with additive manufacturing is that any particular material can require adjustments to the equipment [19], feedstock size and shape, feed rate, and energy input [18].
The fabrication of alloys in the form of solid solutions, chemical compounds, or eutectics requires the precise balancing of alloy components. Most metallic materials are multicomponent. Binary equilibrium diagrams are already available for a significant number of systems and have been well-described and studied [24][25][26][27]. Ternary phase diagrams are much more complicated, and quaternary systems (particularly Fe, C, Mn, and B) are much less common in the literature. There are few studies on this topic. Multiple carbides and borides present together in steel should greatly improve the cutting performance and wear resistance of the metal. To evaluate this hypothesis, we analyzed two equilibrium ternary systems, Fe-C-Mn and Fe-C-B [28][29][30]. From this review, we can figure out the eutectic regions (where full or partial eutectic exists) of elements of Fe-C-Mn and Fe-C-B systems, which will further be used to construct isothermal sections of pseudo-ternary systems and reduce the number of possible compositions (Table 1).
To produce material with the required wear resistance, the Fe 3 C, Cr 2 B, Fe 2 B, and FeB phases are very promising, as these phases are very hard, resistant to abrasion and corrosion, and thermally stable. Furthermore, the addition of manganese improves the toughness of cementite and enhances the eutectic alloy's properties. Manganese is soluble in iron in a solid state and extends the temperature-concentration region of (Fe, Mn) 3 C mixed carbides. At the same time, manganese refines carbides and promotes their uniform distribution in the matrix material [31]. Boron partially substitutes for carbon in cementite and can form borides with iron. Manganese carbides can also be formed. The Fe-C-B and Fe-C-Mn systems are both very effective from the perspectives of mechanical properties, wear resistance, cost savings, and availability [32]. Indeed, these simple alloys have remarkable capabilities, and, as we determined, combining them into a quaternary Fe-C-Mn-B system will yield many beneficial properties.
The ternary section of this system [27] indicates that the eutectic alloy region is affected only by carbon content within a broad content range (0.2-0.8 wt%). At the same time, Mn has no significant effect, as it is infinitely soluble in iron in a solid state.
Taking into consideration the specificity and technological details of producing eutectic alloys and surface layers, as well as the obtained properties-in particular, the reduction in the tendency to crack-we chose the same carbon and boron contents used in the Fe-Mn-C eutectic alloy. The fundamental constituents of Fe-C-Mn-B, Fe, and Mn have good solubility in Ni and Cr. These elements offer strong beneficial effects on hardness, corrosion resistance, and ductility and could be used to enhance the considered eutectic system's properties. In combinations with even more alloying elements, such as alloy steels, one can obtain versatile materials for casting, additive manufacturing, and coating deposition.

Materials and Methods
The Fe-C-Mn-B alloys were first fabricated as powder feedstock. To fabricate the test samples of eutectic alloys to study the Fe-C-Mn-B equilibrium system, we used high purity components: Fe (99.98%), Mn (99.6%), amorphous B (99.4%), and synthetic graphite (99.95%) powders. The component concentration in each section changed with a 10 or 20 molar part pitch. Iron concentrations were in the range of 0.67-0.79 at% [28][29][30]. To prevent oxidation, minor additions of silicon were used. Silicon also enhances carbon diffusion in iron. Silicon, manganese, and boron can also effectively deoxidize steel, which is good for surface deposition technologies. Liquid metal was argon-atomized. Powders were then placed in a low-carbon (analog of AISI 1020 steel) filler wire to consider the metal's effects on the final alloy composition. Wire arc manufacturing was done using Kemppi Pro 5200 Evolution equipment (Kemppi, Lahti, Finland) in semi-automated mode. Parameters of the manufacturing process: current = 270 A, voltage 30 V, the distance from the electrode tip to the base plate was 6 mm at the travel speed of 10 cm/min, the diameter of powder-cored wire-2.6 mm, the thickness of the weld layer 4-5 mm. To protect the alloy from oxidation, the work area was shielded by Argon gas flow (0.8 m 3 /s). The base plate (AISI 1045 steel) was preheated to 200 • C to reduce possible thermal-induced stresses. Then, samples were cut from the plate and annealed at 1273 K for 350 h. For heat treatment, the samples were enclosed in pressurized quartz containers. Figure 1 shows alloys 1-4 ( Table 2), which were welded in as-manufactured condition.
Before the tests, samples were annealed (1273 K, 350 h). In order to study solid state transformations (thermal effects), we used differential scanning calorimetry (Netzsch DSC 404 F1 Pegasus) and thermal analyses (VDTA-8M; argon gas medium; maximum temperature, 1873 K; cooling rate, 80 K/min). The sample sections were polished and etched using a 3 wt.% HNO 3 solution in ethyl alcohol. For microstructure examinations, we used Zeiss Neophot-32 (Zeiss, Jena, Germany) and MIM-8 (Lomo, St. Petersburg, Russia) light microscopes. The concentrations of chemical elements were detected using Jeol Superprobe 733 equipment (Jeol, Tokyo, Japan). A DRON-3.0 X-ray diffractometer (Burevestnik, St. Petersburg, Russia) was utilized to study the phase composition of the samples (Bragg-Brentano 2θ configuration, monochromatic Cu Kα radiation). All tests were carried out under standard atmospheric conditions. Table 2. Phase composition of Fe-C-Mn-B samples (isothermal section of "Fe 3 B"-Fe 3 C-"Fe 3 Mn" system ( Figure 2a)). Then, samples were cut from the plate and annealed at 1273 K for 350 h. For heat treatment, the samples were enclosed in pressurized quartz containers. Figure 1 shows alloys 1-4 (Table 2), which were welded in as-manufactured condition.  Table 1) on the base plate in as-manufactured condition. Table 2. Phase composition of Fe-C-Mn-B samples (isothermal section of "Fe3B"-Fe3C-"Fe3Mn" system (Figure 2a)).  Before the tests, samples were annealed (1273 K, 350 h). In order to study solid state transformations (thermal effects), we used differential scanning calorimetry (Netzsch DSC 404 F1 Pegasus) and thermal analyses (VDTA-8M; argon gas medium; maximum temperature, 1873 K; cooling rate, 80 K/min). The sample sections were polished and etched using a 3 wt.% HNO3 solution in ethyl alcohol. For microstructure examinations, we used Zeiss Neophot-32 (Zeiss, Jena, Germany) and MIM-8 (Lomo, St. Petersburg, Russia) light microscopes. The concentrations of chemical elements were detected using Jeol Superprobe 733 equipment (Jeol, Tokyo, Japan). A DRON-3.0 X-ray diffractometer (Burevestnik, St. Petersburg, Russia) was utilized to study the phase composition of the samples (Bragg-Brentano 2θ configuration, monochromatic Cu Kα radiation). All tests were carried out under standard atmospheric conditions.
Results of X-ray diffraction examinations were used to determine phase analyses of the samples. Based on these results, Tables 1-4 were composed. Basically, the X-ray diffraction patterns most common for this study are presented in Figure 3a-f. The analyses of XRD patterns allow to state on the existence of two basic solid solutions in the system: Fe α and Fe γ alloyed by Mn, designated as (Fe, Mn) α and (Fe, Mn) γ , respectively. Commonly, only one of them is present, but in some alloys ( Table 2, ##11, 12, Table 5, ##10, 13, and other compositions) they may exist simultaneously. In addition, traces of unalloyed (or with an undetectable amount of alloying elements) Fe α were detected (Table 3, ##1-5, Table 5, #11). This is the evidence of the partitioning of alloying elements (or segregation) which was studied by authors on other alloy systems [34,36].
In the Fe-C system, if alloying elements are added, the cementite structure may be modified. Particularly in the considered system, iron atoms may be replaced by manganese, and carbon atoms by boron [36]. Thus, this constituent is called alloyed cementite and we designate it as Fe 3 (C, B) ( Figure 3a,d,e). In other phase regions (alloys ##8, 10-12 in Table 2, 9-17 in Table 3 and others) more complex borocarbide is formed: (Fe, Mn) 23 (C, B) 6 .
In alloys with a higher concentration of boron and lower carbon content, iron borides are formed: Fe 2 B (Figure 3e, alloys 6 and 9, Table 3). In the majority of alloy compositions, boron does not form a separate chemical compound, but a part of alloyed cementite and borocarbide. In alloy compositions 8 and 9 (Table 2) alloyed iron boride was detected ( Figure 3b) as a minor constituent The DSC curves were used to check the assumption about the eutectic concentration of the alloy. This was done based on the property of eutectic alloy to melt at a definite temperature. Typically, this temperature is the lowest in the system, but in our case of the quaternary system, we assumed the formation of Fe-C-Mn and Fe-C-B type eutectics. The DSC curves indicating the eutectic temperature of these separate ternary subsystems within a quaternary Fe-C-Mn-B system are presented in Figure 4. According to our results, the melting point of Fe-C-Mn type eutectic is in the region of 1192 • C, which is in good agreement with [37], and the melting point of Fe-C-B type eutectic was estimated at Materials 2022, 15, 4393 7 of 16 1109 • C. Thus, using DSC, the existence of two separate eutectic regions was proved. Nevertheless, we should take in mind that these temperatures may be affected by the presence of admixture atoms (B in the case of Fe-C-Mn type eutectic and Mn in Fe-C-B type eutectic). Table 3. Phase composition of Fe-C-Mn-B samples with iron content of 66.6 at.% (isothermal section of the Fe 2 B-"Fe 2 C"-"Fe 2 Mn" system (Figure 2b)). of XRD patterns allow to state on the existence of two basic solid solutions in the system: Feα and Feγ alloyed by Mn, designated as (Fe, Mn)α and (Fe, Mn)γ, respectively. Commonly, only one of them is present, but in some alloys ( Table 2, ##11, 12, Table 5, ##10, 13, and other compositions) they may exist simultaneously. In addition, traces of unalloyed (or with an undetectable amount of alloying elements) Feα were detected (Table 3, ##1-5, Table 5, #11). This is the evidence of the partitioning of alloying elements (or segregation) which was studied by authors on other alloy systems [34,36].  Table 2), (b) alloy contains (Fe, Mn)α solid solution and particles of (Fe, Mn)2B boride, and (Fe, Mn)23(C, B)6 borocarbide (#8 Table 3), (c) alloy #10 Table 5 contains two solid solutions: alloyed ferrite and alloyed austenite with particles of borocarbide (Fe, Mn)23(C, B)6, (d) alloy 13, Table 5 contains alloyed ferrite and austenite with the particles of boron-alloyed cementite Fe3(C, B), (e) diffraction pattern of alloy 6 from Table 3 Table 2), (b) alloy contains (Fe, Mn) α solid solution and particles of (Fe, Mn) 2 B boride, and (Fe, Mn) 23 (C, B) 6 borocarbide (#8 Table 3), (c) alloy #10 Table 5 contains two solid solutions: alloyed ferrite and alloyed austenite with particles of borocarbide (Fe, Mn) 23 (C, B) 6 , (d) alloy 13, Table 5 contains alloyed ferrite and austenite with the particles of boron-alloyed cementite Fe 3 (C, B), (e) diffraction pattern of alloy 6 from Table 3 Table 4 consists of borocarbide (Fe, Mn) 23 (C, B) 6 particles in (Fe, Mn) γ solid solution.  and we designate it as Fe3(C, B) (Figure 3a,d,e). In other phase regions (alloys ##8, 10-12 in Table 2, 9-17 in Table 3 and others) more complex borocarbide is formed: (Fe, Mn)23(C, B)6. In alloys with a higher concentration of boron and lower carbon content, iron borides are formed: Fe2B (Figure 3e, alloys 6 and 9, Table 3). In the majority of alloy compositions, boron does not form a separate chemical compound, but a part of alloyed cementite and borocarbide. In alloy compositions 8 and 9 (Table 2) alloyed iron boride was detected (Figure 3b) as a minor constituent

Component Contents
The DSC curves were used to check the assumption about the eutectic concentration of the alloy. This was done based on the property of eutectic alloy to melt at a definite temperature. Typically, this temperature is the lowest in the system, but in our case of the quaternary system, we assumed the formation of Fe-C-Mn and Fe-C-B type eutectics. The DSC curves indicating the eutectic temperature of these separate ternary subsystems within a quaternary Fe-C-Mn-B system are presented in Figure 4. According to our results, the melting point of Fe-C-Mn type eutectic is in the region of 1192 °C, which is in good agreement with [37], and the melting point of Fe-C-B type eutectic was estimated at 1109 °C. Thus, using DSC, the existence of two separate eutectic regions was proved. Nevertheless, we should take in mind that these temperatures may be affected by the presence of admixture atoms (B in the case of Fe-C-Mn type eutectic and Mn in Fe-C-B type eutectic).

Study on the Isothermal Section of the "Fe3B"-Fe3C-"Fe3Mn" Pseudo-Ternary System
Studying the "Fe3B"-Fe3C-"Fe3Mn" isothermal pseudo-ternary section evidenced two dissimilar phase regions (Figure 2a; Table 2 (Table 2) contained a solid solution of (Fe, Mn)α. It also should be mentioned that in phase region I
The 19 samples of this pseudo-ternary equilibrium system were then examined. Three constituents were determined: cementite, (Fe, Mn) 23 (C, B) 6 , and (Fe, Mn) 3 (C, B) phases with the structure of Fe 23 C 6 and Fe 3 C. Samples ##6 and 10 presented phases with a structure of cementite and the following lattice parameters: a = 0.5091, b = 0.6650, and c = 0.4559 nm.
The lattice parameters of this cementite-like structure corresponded to Fe 3 C. Samples ##1-3, 5, and 7 contained the Fe 3 C phase with increased lattice parameters. This phase was identified as borocarbide Fe 3 (C, B). Many of the samples consist of solid solution, (Fe, Mn) 23 (C, B) 6 phase and/or Fe 2 B, (Fe, Mn) 2 B borides. Samples ##1-14 (Table 2) contained a solid solution of (Fe, Mn) α . It also should be mentioned that in phase region I (Figure 2a), a hypereutectic structure was achieved for alloy #3 with a composition of Fe 75 C 12.5 Mn 2.5 B 10 (at.%). It consists of dendritic (Fe, Mn) 23 (C, B) 6 carbides (light field) and (Fe, Mn) α + Fe 3 (C, B) eutectic (dark field) between them (Figure 5a). These dendrites are oriented along the basic crystallographic directions [36]. High-temperature annealing at 1273 K during 350 h led to carbides coarsening (Figure 5b). In carbon-rich alloys, the (Fe, Mn) 23 (C, B) 6 compound occurred in the form of lamellas that were 250-450 µm in length and 10-30 µm in width. Carbide lamellas as long as the studied sample were also identified ( Figure 5c).  Table 2. Here and on other following microstructures, carbide particles-light field, solid solution-dark field.
Eutectic carbide particles grew from these lamellas and ran parallel to each other. The microstructure here is dendritic. All other alloys in phase regions I and II were of the solid solution type and consisted of (Fe, Mn)α and (Fe, Mn)γ solid solutions containing (Fe, Mn)2B boride, Fe3(C, B), or (Fe, Mn)23(C, B)6 borocarbide particles. The Fe-C-B type eutectic was located in region A, whereas the Fe-C-Mn type eutectic was in region B (Figure 2a).
Eutectic carbide particles grew from these lamellas and ran parallel to each other. The microstructure here is dendritic. All other alloys in phase regions I and II were of the solid solution type and consisted of (Fe, Mn) α and (Fe, Mn) γ solid solutions containing (Fe, Mn) 2 B boride, Fe 3 (C, B), or (Fe, Mn) 23 (C, B) 6 borocarbide particles. The Fe-C-B type eutectic was located in region A, whereas the Fe-C-Mn type eutectic was in region B (Figure 2a).

Study on the Isothermal Section of the Fe 2 B-"Fe 2 C"-"Fe 2 Mn" Pseudo-Ternary System
The sample composition and phases detected in the pseudo-ternary Fe 2 B-"Fe 2 C"-"Fe 2 Mn" system ( Figure 2b) are listed in Table 3. The alloys contained the (Fe, Mn) γ solid solution and Fe 3 (C, B), (Fe, Mn) 23 (C, B) 6 , Fe 2 B chemical compounds. The iron content for these samples was 66.6 at.%. Here, we separated the two phase regions (Figure 2b; Table 3). Region I contained Fe 3 (C, B) particles in the Fe α solid solution. Hypereutectic region II contained (Fe, Mn) γ and (Fe, Mn) 23 (C, B) 6 . The Fe-C-B type eutectic alloy was located in field A, whereas the Fe-C-Mn type eutectic was in field B (Figure 2b). Hypereutectic alloys contained 66.6 at.% of iron and 3.3-6.7 at.% of manganese, as well as 10.0-23.3 at.% of carbon and 6.7-16.6 at.% of boron. Hypereutectic samples ##3-4 contained primary Fe 3 (C, B) crystals and dendrites of (Fe, Mn) 23 (C, B) 6 iron-manganese borocarbide (Figure 6a,b).
According to the micro X-ray and microstructural analyses performed for the "Fe3B"-Fe3C-"Fe1.2Mn" pseudo-ternary section, two phase regions were identified (Figure 2c
When Mn content in the hypoeutectic alloys increased, the primary dendrites of (Fe, Mn) γ formed first. The eutectic consisting of (Fe, Mn) γ and (Fe, Mn) 23 (C, B) 6 was located in the areas between these dendrites (Figure 7a). After high-temperature annealing, these microstructures of the alloys presented homogenized and coagulated (Fe, Mn) γ crystals (Figure 7b). The contents of the components in these samples were as follows:  Table 4): (a,b) sample #3, (c) sample #10; (a,c) as cast; (b) annealed 350 h at 1273 K.
According to the micro X-ray and microstructural analyses performed for the "Fe3B"-Fe3C-"Fe1.2Mn" pseudo-ternary section, two phase regions were identified (Figure 2c
(  (Figure 2a,b). The hypereutectic concentrations of the system are outlined in Figure 2 and Table 6.
The studied eutectic alloys were dispersion-strengthened natural composites consisting of relatively soft matrix phase: alloyed austenite or pearlite, strengthened with a hard and wear-resistant (Fe, Mn)23(C, B)6 phase. By changing the contents of the alloying elements, it is possible to obtain two or more phases in equilibrium [17,[38][39][40][41][42][43][44][45][46][47][48][49][50]. By balancing the four components, one can obtain an alloy with the required combination of carbide content and composition. Eutectic and hypereutectic alloys have good casting properties and do not form cracks when wire arc-manufactured. This material, moreover, can be modified by other alloying elements to enhance its properties.

Conclusions
Wire arc-manufactured alloy samples were found to be free of cracks, and the thickness of the built products was nearly identical. XRD analyses also demonstrated that the chemical composition of the alloys was nominal. Microstructural studies revealed different eutectic structures depending on chemical composition. This fact could allow one to change the microstructure according to the requirements of any casting or additive manufacturing process, not only wire arc manufacturing.
The main results are listed below: Comparing the results for the four isothermal sections, we observed that the ironmanganese borocarbide phase (Fe, Mn) 23 6 increases, and carbon content decreases (Figure 2a,b). The hypereutectic concentrations of the system are outlined in Figure 2 and Table 6.
The studied eutectic alloys were dispersion-strengthened natural composites consisting of relatively soft matrix phase: alloyed austenite or pearlite, strengthened with a hard and wear-resistant (Fe, Mn) 23 (C, B) 6 phase. By changing the contents of the alloying elements, it is possible to obtain two or more phases in equilibrium [17,[38][39][40][41][42][43][44][45][46][47][48][49][50]. By balancing the four components, one can obtain an alloy with the required combination of carbide content and composition. Eutectic and hypereutectic alloys have good casting properties and do not form cracks when wire arc-manufactured. This material, moreover, can be modified by other alloying elements to enhance its properties.

Conclusions
Wire arc-manufactured alloy samples were found to be free of cracks, and the thickness of the built products was nearly identical. XRD analyses also demonstrated that the chemical composition of the alloys was nominal. Microstructural studies revealed different eutectic structures depending on chemical composition. This fact could allow one to change the microstructure according to the requirements of any casting or additive manufacturing process, not only wire arc manufacturing.
The main results are listed below: • The alloys of the "Fe 3 B"-Fe 3 C-"Fe 1.2 Mn" isothermal section at 66.6 at.% for iron contained no α-phase. This result was only observed when the iron content was higher than 75 at.%; • In the eutectic region with the highest Fe concentration (79; 75 at.%), the (Fe, Mn) γ , (Fe, Mn) α , Fe 3 (C, B), and (Fe, Mn) 23 (C, B) 6 phases were in equilibrium. When the iron concentration reduced to 66.6 at.% (the "Fe 3 B"-Fe 3 C-"Fe 1.2 Mn" isothermal section), the alloys presented no γ-α transition in the Fe-based solid solution. In this case, Fe 3 (C, B) + (Fe, Mn) γ or (Fe, Mn) 23 (C, B) 6 + (Fe, Mn) γ were in equilibrium; • The eutectic regions in the isothermal sections of the "Fe 3 B"-Fe 3 C-"Fe 3 Mn", Fe 2 B-"Fe 2 C"-"Fe 2 Mn", and "Fe 23 B 6 "-"Fe 23 C 6 "-"Fe 23 Mn" pseudo-ternary systems were found in both phase regions. The "Fe 3 B"-Fe 3 C-"Fe 1.2 Mn" system featured a eutectic Fe-C-Mn structure in region II only. Basically, the Fe-C-B type eutectic was formed when the boron and carbon contents were high, whereas the Fe-C-Mn type eutectic was found in the region with a high Mn concentration.