Purification of Aluminium Cast Alloy Melts through Precipitation of Fe-Containing Intermetallic Compounds

Transcription

1 metals Article Purification of Aluminium Cast Alloy Melts through Precipitation of Fe-Containing Intermetallic Compounds Marina Gnatko *, Cong Li, Alexander Arnold and Bernd Friedrich IME Institute of Process Metallurgy and Metal Recycling, RWTH Aachen University, Aachen, Germany; (C.L.); (A.A.); (B.F.) * Correspondence: mgnatko@gmx.net; Tel.: +49-(0) Received: 19 August 2018; Accepted: 1 October 2018; Published: 4 October 2018 Abstract: Aluminium secondary materials are often contaminated by impurities such as iron. As alloy properties are affected by impurities, it is necessary to refine aluminium melts. The formation of Fe intermetallics in aluminium melts can be used to develop a purification technology based on removal of intermetallic compounds. In this study, temperature range for effective separation of intermetallics was determined in an industrial-relevant Al Si Fe Mn system with 6 to 10 Si wt. %, 0.5 to 2.0 Fe wt. %, and 0 to 2.0 Mn wt. %. Based on DTA (Differential Thermal Analysis) and SEM (scanning electron microscope) results and following rules of phase boundary drawing, isopleths were drawn. This method allows to derive temperature ranges of intermetallic phase stability and can be applied for assessment of melt-refining parameters. Keywords: aluminium purification; iron removal; intermetallic formation; polyrmal section 1. Introduction In order to achieve legal recycling rate requirements (e.g., regarding end-of-life vehicles, 95% of materials must be recycled), material cycles must be almost completely closed. The recovery of all metals in ir pure form, however, is not possible. Secondary recovered materials are often contaminated. The complexity of such materials leads to difficulties in sorting, as well as to impurity pickup during mechanical treatment processes. As property formation is affected by impurities, aluminium end-of-life scrap is normally used for production of cast alloys. Since impurities such as iron accumulate in aluminium secondary alloys at values of up to 2 wt. %, it is difficult to produce Al recycling alloys which conform to standards (Table 1). Therefore, it is necessary to refine aluminium melts, as current practice of diluting primary aluminium is becoming uneconomical. Among all impurities that need to be removed, iron is a serious challenge. Table 1. Composition of some Al cast alloys, data from [1,2]. Alloy Identification Alloy Composition Limits, wt. % Numerical Chemical Si Fe Cu Mn Mg Or cast alloys for pressure casting EN AC EN AC-AlSi12(Fe) EN AC EN AC-AlSi9Cu3(Fe) cast alloys for common application EN AC EN AC-AlSi12(a) EN AC EN AC-AlSi8Cu While many efforts have been made for removal of iron from primary aluminium and high-purity aluminium [3 6], only limited attention has been paid to that of secondary aluminium, Metals 2018, 8, 796; doi: /met

2 EN AC EN AC-AlSi8Cu The addition of furr alloying elements results in formation of quaternary or higher alloy systems with complex phase relations. Ternary and quaternary intermetallic compounds with iron are formed in Al Si Fe Mn system, and iron solubility decreases to 0.29 wt. % at eutectic point [12]. The current German standards regarding maximum Fe content in cast Al Si alloys range While many efforts have been made for removal of iron from primary aluminium and highpurity aluminium [3 6], only limited attention has been paid to that of secondary aluminium, which Metals 2018, 8, of 12 contains usually more than 2 wt. % Fe. Conventional ways of iron removal from high iron-containing aluminium melts include filtration, centrifugal separation, and electromagnetic (EM) separation [7 9]. All which se contains methods usually are based more on than 2 principle wt. % Fe. of Conventional precipitation ways of Fe-enriched of iron removal phases. fromit is high a wellknown iron-containing fact that in aluminium Al Si Fe melts system, includea filtration, variety of centrifugal binary and separation, ternary and compounds electromagnetic with (EM) Al exist, including separation Al3Fe, [7 9]. Al5FeSi, All se Al8Fe2Si, methods Al3FeSi, are based and on principle of precipitation of Fe-enriched phases. Al4FeSi2 [10,11]. On one hand, precipitation of It is a well-known fact that in Al Si Fe system, a variety of binary and ternary compounds se phases impacts quality of end products. On or hand, it can provide a basis for with Al exist, including Al 3 Fe, Al 5 FeSi, Al 8 Fe 2 Si, Al 3 FeSi, and Al 4 FeSi 2 [10,11]. On one hand, development of a refining technology with help of physical separation process, e.g., filtration. precipitation of se phases impacts quality of end products. On or hand, it can Thus, provide it was a basis aim for of a development six-year project of a at refining IME (Institute technology IME with Process help Metallurgically of physical separation and Metal Recycling) process, to e.g., find filtration. elements Thus, that itinfluence was aim ofresidue melt a six-year project composition at IME (Institute in order IME to Process reduce concentration Metallurgically of impurities, and Metal Recycling) above all toiron. find elements Even if that intermetallic influence compounds residue meltare composition formed, conditions in orderand to reduce separation concentration technique considered of impurities, are above very important all iron. Even for ifreaching intermetallic highest compounds grade of purity. are The formed, aim of conditions this work and was separation to determine technique suitable considered temperature are very ranges important in foral Si Fe Mn reaching system in highest industrially grade of purity. relevant The aim concentration of this workareas was of to determine 6 to 10 Si wt. suitable %, 0.5 temperature to 2.0 Fe wt. ranges %, and in 0 to 2.0 Mn wt. Al Si Fe Mn %, in which system separation in industrially of intermetallics relevant concentration becomes effective. areas of 6 to 10 Si wt. %, 0.5 to 2.0 The Fe eutectic wt. %, and iron 0 to content 2.0 Mn in wt. a pure %, in which binary Al Fe separation melt is of 1.8 intermetallics wt. % at 655 becomes C [10]. effective. Therefore, in The eutectic iron content in a pure binary Al Fe melt is 1.8 wt. % at 655 C [10]. Therefore, in case of hypereutectic alloys (over 1.8 wt. % Fe), iron content cannot be reduced by segregation case of hypereutectic alloys (over 1.8 wt. % Fe), iron content cannot be reduced by segregation below this value. Iron precipitates in form of intermetallic compound Al3Fe, if below this value. Iron precipitates in form of intermetallic compound Al3Fe, if temperature temperature falls below falls below liquidus line liquidus (Figure line 1). Since (Figure this 1). intermetallic Since this intermetallic phase has a melting phase point has a of melting 1060 C point of 1060 and C is insoluble and is insoluble moltenin aluminium, molten aluminium, it can be mechanically it can be mechanically removed from removed molten aluminium, from molten aluminium, e.g., by filtration. e.g., by filtration. Neverless, Neverless, this system has this no system industrial has significance. no industrial significance. Industrial cast cast alloy compositions are are based based on on binary binary system system Al Si, where Al Si, where ternary eutectic ternary eutectic iron iron content content is reduced is reduced to 0.7 wt. to % 0.7 atwt. 577% C at [10,11]. 577 C In[10,11]. Al corner In ofal this corner system, of iron this issystem, present in iron is present phases in phases Al 3 Fe, Al Al3Fe, 8 2 Si, Al8Fe2Si, 5 FeSi, Al5FeSi, and 4 FeSi and 2 (Figure Al4FeSi2 2). (Figure 2). Metals 2018, 8, x FOR PEER REVIEW Figure Figure Al Fe Al Fe phase phase diagram calculated with with FactSage. 3 of 12 FactSage. Figure 2. Liquidus surface in in Al Al corner corner of of Al Si Fe Al Si Fe system system [10]. [10].

3 Metals 2018, 8, of 12 The addition of furr alloying elements results in formation of quaternary or higher alloy systems with complex phase relations. Ternary and quaternary intermetallic compounds with iron are formed in Al Si Fe Mn system, and iron solubility decreases to 0.29 wt. % at eutectic point [12]. The current German standards regarding maximum Fe content in cast Al Si alloys range between 0.2 and 0.9 wt. % (depending on alloying class) [13]. For current investigation, Al Si Fe Mn system was applied because numerous intermetallics are formed in this system, and residue melt composition can be influenced depending on Mn/Fe ratio [10 12,14,15]. Table 2 summarizes phases to be expected in Al Si Fe Mn system. Table 2. Published data on expected phases in Al corner of Al Si Fe Mn system, data from [10 12,14,15]. s Al 8 Fe 2 Si Al 5 FeSi <0, Al 16 (FeMn) 4 Si Al 15 Mn 3 Si < Al 4 FeSi < Until now, no quaternary phase has been clearly identified in this system [10,12,15]. Initially, it was believed that an area of solid solutions existed between Al 8 Fe 2 Si and Al 15 Mn 3 Si 2. Later, this assumption was rejected on basis of fact that se compounds had different crystal structures (hexagonal and cubic). The currently accepted version of phase diagram illustrates a broad range of solid solutions based on compound Al 15 Mn 3 Si 2 extending towards Al Si Fe surface [10]. In this variant, manganese is replaced with iron to form compound with composition 31 wt. % Fe, 1.5 wt. % Mn, 8 wt. % Si. This broad range of homogeneity is considered as quaternary phase Al 15 (FeMn) 3 Si 2 [10]. On or hand, Zakharov A. et al. studied alloys containing wt. % Si, 0 3 wt. % Fe, 0 4 wt. % Mn, and proposed existence of quaternary compound Al 16 (FeMn) 4 Si 3 [12]. The formation of this phase would allow a quasi-ternary section Al Al 16 (FeMn) 4 Si 3 Si and formation of two secondary systems on both sides of this section: Al Al 16 (FeMn) 4 Si 3 Si Al 5 FeSi and Al Al 16 (FeMn) 4 Si 3 Si Al 15 Mn 3 Si 2. According to reference [10], solid solution of iron in Al 15 Mn 3 Si 2 phase has a cubic structure with a lattice parameter which decreases because of an increase of Fe content from nm (0 wt. % Fe) to 1.25 nm (31.1 wt. % Fe). The quaternary phase found in reference [12] has a face-centered cubic structure with a lattice parameter of a = ± 0.04 nm. The similar lattice parameters mean that it cannot be determined which version of Al Si Fe Mn phase diagram is correct. In references [11,15], it was proposed that non-equilibrium crystallization had a significant effect on phase composition, especially in Al Si Fe alloys. This is because of inhibition of peritectic reactions, which take a long time to be completed. However, due to numerous intermetallics, this system shows a potential for removing iron and manganese from Al Si melts. diagrams are a useful tool for presenting required relations in a metal system. In comparison with binary systems (only two dimensions), ternary and multi-phase phase diagrams (here and after in this article, Multi- refers specially to more than three) are rar complicated. A ternary phase diagram is shown in Figure 3a, where composition plane forms base triangle, and phase variations caused by temperature change are illustrated vertically (Figure 3a). Vertical sections (Figure 3b) of a ternary phase diagram also known as isopleths have been widely used because of ir similarities to binary diagrams. Such sections are two-dimensional planes constructed by cutting three-dimensional diagrams with a slice which is vertical to base composition triangle. Once phase areas in an isopleth are clearly clarified, liquidus and solidus temperatures for certain alloy compositions can be readily read from it.

4 base triangle, and phase variations caused by temperature change are illustrated vertically (Figure 3a). Vertical sections (Figure 3b) of a ternary phase diagram also known as isopleths have been widely used because of ir similarities to binary diagrams. Such sections are two-dimensional planes constructed by cutting three-dimensional diagrams with a slice which is vertical to Metals base composition 2018, 8, 796 triangle. Once phase areas in an isopleth are clearly clarified, liquidus 4 and of 12 solidus temperatures for certain alloy compositions can be readily read from it. Figure 3. (a) Temperature composition space diagram of a ternary system (b) Isopleth through a ternary system [16]. From metallurgical practice point of view, multi-phase alloy diagrams involving four or more elements are needed more than binary or ternary diagrams. This is because most commercial alloys contain more than three alloying elements, even without taking impurity into consideration. However, temperature composition phase phase diagrams diagrams of multi of components multi components are extremely are inconvenient extremely and inconvenient highly complicated. and highly complicated. In order to to determine phase phase variation variation caused caused by temperature by temperature changes, as changes, well as as composition well as difference composition in complex difference multi-components in complex multi-components system, a feasiblesystem, way is a to feasible draw way corresponding is to draw three- corresponding two-dimensional three- sections, or two-dimensional in which temperature sections, in which and concentration temperature of and certain concentration component(s) of certain are represented component(s) as are variables. represented as variables. For constructionof ofa a two-dimensional isopleth, i.e., i.e., temperature composition diagrams, diagrams, following following information is usually is usually needed: needed: (1) (1) general diagram including number, disposition, and identity of phases and respective invariant reaction, and (2) temperature and compositions along all boundary lines (and surfaces). The most widely used method of constitutional investigation is Differential Thermal Analysis (DTA). It It is is capable of locating of locating liquidus liquidus lines and lines at and same at time same indicating time indicating general disposition general of disposition phases and of phases invariant and reactions invariant reactions system. in Its principle system. Its is principle extremelyis simple: extremely every simple: occurrence every of occurrence phase change of phase is accompanied change is accompanied by exormic by exormic and endormic and endormic effects such effects as heat such from as heat from melt crystallization. melt crystallization. The delay The andelay acceleration and acceleration of cooling of speed cooling compared speed compared to a reference to a reference material is material monitored. is monitored. 2. Materials and Methods In this research work, approximately 60 alloy compositions were prepared by induction melting within following concentrationranges: ranges: 6 6to to10 10 wt. wt. % % Si, Si, 0 to 0 2 towt. 2 wt. % Fe, % Fe, and and 0 to 02 to wt. 2% wt. Mn. % ICP Mn. ICP (Spectro (Spectro ICP-OES ICP-OES Spectro Spectro Ciros Ciros Vision, Vision, Kleve, Kleve, Germany) Germany) analysis analysis was was applied applied to determine to determine composition of samples. Differential Thermal Analysis (DTA) (IME, Aachen, Germany) and Scanning Electron Microscopy (SEM) (JEOL JSM-7000F, Tokyo, Japan) with integrated EDX (Energy Dispersive X-ray analysis) (Oxford Instruments, Oxford, UK) were applied to determine phase precipitations and temperatures of phase transformations. In order to allow an evaluation in form of isopleths, three of four element concentrations were kept constant. The groups of investigated alloys and isopleths are shown in Table 3. The manganese content changed from 0 to 2 wt. % by representation on isopleths in steps of 0.5 wt. %.

5 were kept constant. The groups of investigated alloys and isopleths are shown in Table 3. The manganese content changed from 0 to 2 wt. % by representation on isopleths in steps of 0.5 wt. %. Metals 2018, 8, 796 Table 3. Groups of investigated alloys leading to individual isopleth. 5 of 12 Iron/Manganese Content, wt. % Group Table 3. Groups of investigated alloys leading to individual isopleth. Fe/Mn Step 0.5 Fe/Mn Step 0.5 Fe/Mn Step 0.5 Fe/Mn Step 0.5 AlSi6FeMn 0.5/0 2 Iron/Manganese 1.0/0 2 Content, wt. 1.5/0 2 % 2.0/0 2 Group AlSi8FeMn Fe/Mn0.5/0 2 Step Fe/Mn Step 1.0/ Fe/Mn Step 1.5/ Fe/Mn Step 2.0/ AlSi10FeMn AlSi6FeMn 0.5/ / / / / / / /0 2 AlSi8FeMn 0.5/ / / /0 2 AlSi10FeMn 0.5/ / / /0 2 Extended experimental equipment for Differential Thermal Analysis (DTA) (IME, Aachen, Germany) was built, containing a resistance furnace and a differential rmocouple (Figure 4). The differential Extended rmocouple experimental consists equipment of for two connected Differential rmocouples. Thermal Analysis The (DTA) first (IME, one, Aachen, working Germany) was built, containing a resistance furnace and a differential rmocouple (Figure 4). rmocouple, measured temperature in sample. The second one, reference rmocouple, The differential rmocouple consists of two connected rmocouples. The first one, working measured temperature difference which existed during cooling between samples and rmocouple, measured temperature in sample. The second one, reference rmocouple, reference measured substance. temperature Two crucibles, difference one which with existed reference during cooling substance between (Al2O3) samples and and or with sample, reference were substance. placed Two in a crucibles, steel block oneto with ensure reference same substance external (Al heat 2 O 3 ) conditions and or for with both crucibles during sample, cooling. were placed As steel in ahas steela block lower tormal ensure conductivity same external than heat Al, conditions this block for protected both crucibles crucibles from during temperature cooling. Aschanges steel has in a lower furnace rmalspace. conductivity Such changes than Al, this could block influence protected temperature crucibles data and fromdistort temperature results. changes in furnace space. Such changes could influence temperature data andin distort order to results. determine an isopleth with sufficient accuracy, a minimum of five alloys must be investigated. In order After to determine melting an isopleth alloy, with sufficient differential accuracy, rmal a minimum analysis ofcommenced. five alloys must The besample, weighing investigated. approximately After melting 20 g, was alloy, placed differential in crucible rmal (Figure analysis 4) and commenced. heated to 750 The C 760 sample, C. This weighing approximately 20 g, was placed in crucible (Figure 4) and heated to 750 C 760 C. temperature value was chosen to allow a sufficient superheat. As according to literature data, This temperature value was chosen to allow a sufficient superheat. As according to literature data, melting point of alloys studied was below or near 700 C. melting point of alloys studied was below or near 700 Subsequently, furnace was switched C. Subsequently, furnace was off, and cooling curve with a rate of approx. 4.5 C/min switched off, and cooling curve with a rate of approx. 4.5 was recorded. C/min was recorded. Figure 4. Equipment for large scale Differential Thermal Analysis (DTA) at IME. 3. Results andfigure Discussions 4. Equipment for large scale Differential Thermal Analysis (DTA) at IME DTA Experimental Results Figure 5 illustrates a cooling curve example for alloy AlSi8Fe2.0Mn1.0 from isopleth AlSi8Fe2.0 Mn. Two curves are indicated: one for sample alloy and one for reference (Al 2 O 3 ). The curve of sample demonstrates two significant effects, whereas reference curve shows four. This is because of special bonding of rmocouples (Figure 4), whereby reference material

6 3.1. DTA Experimental Results Figure 5 illustrates a cooling curve example for alloy AlSi8Fe2.0Mn1.0 from isopleth AlSi8Fe2.0 Mn. Two curves are indicated: one for sample alloy and one for reference (Al2O3). The curve of sample demonstrates two significant effects, whereas reference curve shows four. Metals 2018, 8, of 12 This is because of special bonding of rmocouples (Figure 4), whereby reference material becomes very sensitive and can detect changes with lower evolutions of heat, e.g., at liquidus temperature. becomestherefore, very sensitive was and presumed can detect that changes four phase withchanges lower evolutions occurred in ofthis heat, alloy. e.g., Exemplary at liquidus DTA temperature. results are shown Therefore, in Table it was 4 for presumed isopleths thatalsi8fe0.5-mn, four phase changes AlSi8Fe1-Mn, occurredalsi8fe1.5-mn, in this alloy. Exemplary and AlSi8Fe2.0-Mn; DTA results all are data shown are published in Table 4in for reference isopleths [17]. After AlSi8Fe0.5-Mn, recording and AlSi8Fe1-Mn, evaluating all AlSi8Fe1.5-Mn, cooling curves, and AlSi8Fe2.0-Mn; temperature composition all data are published diagrams inwere reference created [17]. for After se recording isopleths. and evaluating all cooling curves, temperature composition diagrams were created for se isopleths. Figure Figure 5. Cooling 5. Cooling curve curve of of alloy AlSi8Fe2.0Mn1.0 alloy AlSi8Fe2.0Mn1.0 from isopleth from isopleth AlSi8Fe2 Mn. AlSi8Fe2 Mn. Table 4. Results of evaluation of cooling curve effects of alloys from isopleths AlSi8Fe0.5 Mn, Table 4. Results of evaluation of cooling curve effects of alloys from isopleths AlSi8Fe0.5 AlSi8Fe1 Mn, AlSi8Fe1.5 Mn, and AlSi8Fe2.0 Mn. Mn, AlSi8Fe1 Mn, AlSi8Fe1.5 Mn, and AlSi8Fe2.0 Mn. Alloy Mn, Effect 1 Effect 2 Effect 3 Effect 4 Alloy Mn, Effect 1 Effect 2 Effect 3 Effect 4 (Target) wt. % T, C T, C T, C T, C (Target) wt. % T, C T, C T, C T, C AlSi8Fe AlSi8Fe AlSi8Fe0.5Mn AlSi8Fe0.5Mn0.5 AlSi8Fe0.5Mn AlSi8Fe0.5Mn1.0 AlSi8Fe0.5Mn AlSi8Fe0.5Mn2.0 AlSi8Fe AlSi8Fe1.0Mn AlSi8Fe AlSi8Fe1.0Mn AlSi8Fe1.0Mn0.5 Al Si AlSi8Fe1.0Mn1.0 Fe1.0Mn Al Si8 Fe1.0Mn1.5 Al Si Al Si8 Fe1.0Mn2.0 Fe1.0Mn AlSi8Fe AlSi8Fe1.5 AlSi8Fe1.5Mn AlSi8Fe1.5Mn0.5 AlSi AlSi8 Fe1.5Mn1.0 Fe1.5Mn AlSi8Fe1.5Mn1.5 AlSi8Fe1.5Mn AlSi8Fe1.5Mn AlSi8Fe1.5Mn AlSi8Fe AlSi8Fe2.0 AlSi8Fe2.0Mn AlSi8Fe2.0Mn AlSi8Fe2.0Mn AlSi8Fe2.0Mn Precipitated s Figure 6 shows exemplary SEM examination patterns of alloys AlSi8Fe2Mn0.5(a) and AlSi8Fe2Mn1.0(b) performed by GfE (Gemeinschaftslabor für Electronenmikroskopie) RWTH

7 Figure AlSi8Fe2.0Mn1.5 6 shows exemplary 1.36 SEM examination patterns of alloys AlSi8Fe2Mn0.5(a) and AlSi8Fe2.0Mn AlSi8Fe2Mn1.0(b) performed by GfE (Gemeinschaftslabor für Electronenmikroskopie) RWTH (Rheinisch-Westfälische Technische Hochschule) Aachen University. The dark grey crystals are 3.2. Precipitated s eutectic Metals silicon 2018, 8, precipitations. 796 White needle-like precipitations indicate ternary phase Al5FeSi. 7 of 12 The groups Figure of white 6 shows net-forming exemplary precipitations SEM examination (also patterns known of as Chinese alloys script) AlSi8Fe2Mn0.5(a) are clusters and of AlSi8Fe2Mn1.0(b) performed by GfE (Gemeinschaftslabor für Electronenmikroskopie) RWTH quaternary phase Al(FeMn)Si. These descriptions of phase shapes were previously accepted, as in (Rheinisch-Westfälische Technische Hochschule) Aachen University. The The dark dark grey crystals grey crystals are eutectic are references [18,19]. The composition of precipitations was determined by EDX analysis. eutectic silicon silicon precipitations. precipitations. White needle-like White needle-like precipitations precipitations indicate indicate ternary ternary phase Al phase 5 FeSi. Al5FeSi. The groups The groups of whiteof net-forming white net-forming precipitations precipitations (also known (also as Chinese known script) as Chinese are clusters script) of are quaternary clusters of phase quaternary Al(FeMn)Si. phase These Al(FeMn)Si. descriptions These of phase descriptions shapes were of phase previously shapes were accepted, previously as in references accepted, [18,19]. as in references The composition [18,19]. of The composition precipitations of wasprecipitations determined by was EDX determined analysis. by EDX analysis. Figure 6. SEM pattern of microstructure. According to EDX microanalysis Figure Figure6. 6. SEM SEMof pattern pattern investigated of of microstructure. alloys, compositions of phases precipitated were determined and are shown in Table 5. The appearance of above-mentioned According to EDX microanalysis of investigated alloys, compositions of phases phases According to EDX microanalysis of investigated alloys, compositions of phases precipitated depended were on determined ir composition, and are shown and in Table extent 5. The appearance varied with of above-mentioned Mn content of phases alloy, precipitated were determined and are shown in Table 5. The appearance of above-mentioned especially depended for on ir precipitation composition, of and Al(FeMn)Si extent varied phase. with Mn content ofin alloy, quaternary especially for phase phases depended on ir composition, and extent varied with Mn content of alloy, increased precipitation from 8.42 of to Al(FeMn)Si wt. phase. %, and MnFe content content in decreased quaternary phase from increased to from wt. to %, especially for precipitation of Al(FeMn)Si phase. Mn content in quaternary phase correspondingly wt. %, and (Figure Fe content 7). decreased from to wt. %, correspondingly (Figure 7). increased from 8.42 to wt. %, and Fe content decreased from to wt. %, correspondingly (Figure 7). Figure 7. Composition change in Al(FeMn)Si phase with increasing Mn content in alloy group The ternary Al 5 FeSi disappeared after a specific Mn content was reached in alloy, and formation of Al 16 (FeMn) 4 Si 3 was not as clearly determined as reported by A. Zakharov [12]. This was caused by fact that Mn content of our Al(FeMn)Si phases changed with Mn content of alloys. On or hand, diagram version proposed by L. Mondolfo [17] cannot be accepted

8 Metals 2018, 8, of 12 as fundamental information for drawing isopleths because of Al 5 FeSi disappearance (see above). Since phase compositions are significantly influenced by crystallization conditions, deviations in Mn content of Al(FeMn)Si phases, in comparison to A. Zakharov s study, must be considered in consequence of different crystallization conditions. Al 5 FeSi Based on DTA and SEM results, 12 isopleths were drawn (according to Table 4). All isopleths The construction of isopleths was based on following ory as well as on rules of phase (1) The quaternary Al(FeMn)Si are differentiated by Mn/Fe ratio into α-al(femn)si if Mn/Fe 1.1 and β-al(femn)si if Mn/Fe > 1.1. These three systems are formed depending on Mn/Fe ratio of alloy: if Mn/Fe < 1.1, after crystallization, alloys consist of Al α-al(femn)si Si Al 5 FeSi; if Mn/Fe > 1.1, alloys consist of Al α-al(femn)si Si β-al(femn)si; if Mn/Fe = 1.1, (2) Crossing tilted phase boundary line leads to exhaust or precipitation of one phase, whereas passing through horizontal phase boundary line, where eutectic or peritetic reactions occur, boundary results in eir exhaust (precipitation) of two phases or exhaust of one phase and precipitation of or [20]. In case of AlSi8Fe1 Mn, α-al or α-al(femn)si precipitated primarily, and liquidus line (marked by Figure 7. Composition change in Al(FeMn)Si phase with increasing Mn content in alloy group The ternary Al5FeSi disappeared after a specific Mn content was reached in alloy, and caused by fact that Mn content of our Al(FeMn)Si phases changed with Mn content of alloys. On or hand, diagram version proposed by L. Mondolfo [17] cannot be accepted as fundamental information for drawing isopleths because of Al5FeSi disappearance (see above). Since phase compositions are significantly influenced by crystallization conditions, deviations in Mn content of Al(FeMn)Si phases, in comparison to A. Zakharov s study, must be Based on DTA and SEM results, 12 isopleths were drawn (according to Table 4). All isopleths The construction of isopleths was based on following ory as well as on rules of phase (1) The quaternary Al(FeMn)Si are differentiated by Mn/Fe ratio into α-al(femn)si if Mn/Fe 1.1 and β-al(femn)si if Mn/Fe > 1.1. These three systems are formed depending on Mn/Fe ratio of alloy: if Mn/Fe < 1.1, after crystallization, alloys consist of Al α-al(femn)si Si Al5FeSi; if Mn/Fe > 1.1, alloys consist of Al α-al(femn)si Si β-al(femn)si; if Mn/Fe = 1.1, (2) Crossing tilted phase boundary line leads to exhaust or precipitation of one phase, whereas passing through horizontal phase boundary line, where eutectic or peritetic reactions occur, boundary results in eir exhaust (precipitation) of two phases or exhaust of one phase and precipitation of or [20]. In case of AlSi8Fe1 Mn, α-al or α-al(femn)si precipitated primarily, and liquidus line (marked by 1 in Figure8(b)) was drawn by fitting data of primary precipitation temperature. At AlSi8Fe1 side, 2, 3, 4 phase boundaries were extended from corresponding points, which indicates, respectively, precipitation of Al5FeSi, Si, and exhaust of melt. For Mn content from 0.5 to 2 wt. %, exhaust of melts were caused by two four-phase eutectic reactions: Depending on se reactions, 5, 6 phase boundaries were drawn. At nearly 610 C, 7, 8 phase boundaries were drawn because of not only DTA results, but also of fact that a threephase was assumed that precipitation of α-al(femn)si would lead to a decrease of Mn concentration in melt, and refore α-al was assumed to precipitate prior to that of Al5FeSi, according to which in Figure 8b) was drawn by fitting data of primary precipitation temperature. At AlSi8Fe1 side, Figure 7. Composition change in Al(FeMn)Si phase with increasing Mn content in alloy group The ternary Al5FeSi disappeared after a specific Mn content was reached in alloy, and caused by fact that Mn content of our Al(FeMn)Si phases changed with Mn content of alloys. On or hand, diagram version proposed by L. Mondolfo [17] cannot be accepted as fundamental information for drawing isopleths because of Al5FeSi disappearance (see above). Since phase compositions are significantly influenced by crystallization conditions, deviations in Mn content of Al(FeMn)Si phases, in comparison to A. Zakharov s study, must be Based on DTA and SEM results, 12 isopleths were drawn (according to Table 4). All isopleths The construction of isopleths was based on following ory as well as on rules of phase (1) The quaternary Al(FeMn)Si are differentiated by Mn/Fe ratio into α-al(femn)si if Mn/Fe 1.1 and β-al(femn)si if Mn/Fe > 1.1. These three systems are formed depending on Mn/Fe ratio of alloy: if Mn/Fe < 1.1, after crystallization, alloys consist of Al α-al(femn)si Si Al5FeSi; if Mn/Fe > 1.1, alloys consist of Al α-al(femn)si Si β-al(femn)si; if Mn/Fe = 1.1, (2) Crossing tilted phase boundary line leads to exhaust or precipitation of one phase, whereas passing through horizontal phase boundary line, where eutectic or peritetic reactions occur, boundary results in eir exhaust (precipitation) of two phases or exhaust of one phase and precipitation of or [20]. In case of AlSi8Fe1 Mn, α-al or α-al(femn)si precipitated primarily, and liquidus line (marked by 1 in Figure8(b)) was drawn by fitting data of primary precipitation temperature. At AlSi8Fe1 side, 2, 3, 4 phase boundaries were extended from corresponding points, which indicates, respectively, precipitation of Al5FeSi, Si, and exhaust of melt. For Mn content from 0.5 to 2 wt. %, exhaust of melts were caused by two four-phase eutectic reactions: Depending on se reactions, 5, 6 phase boundaries were drawn. At nearly 610 C, 7, 8 phase boundaries were drawn because of not only DTA results, but also of fact that a threephase was assumed that precipitation of α-al(femn)si would lead to a decrease of Mn concentration in melt, and refore α-al was assumed to precipitate prior to that of Al5FeSi, according to which, Figure 7. Composition change in Al(FeMn)Si phase with increasing Mn content in alloy group The ternary Al5FeSi disappeared after a specific Mn content was reached in alloy, and caused by fact that Mn content of our Al(FeMn)Si phases changed with Mn content of alloys. On or hand, diagram version proposed by L. Mondolfo [17] cannot be accepted as fundamental information for drawing isopleths because of Al5FeSi disappearance (see above). Since phase compositions are significantly influenced by crystallization conditions, deviations in Mn content of Al(FeMn)Si phases, in comparison to A. Zakharov s study, must be Based on DTA and SEM results, 12 isopleths were drawn (according to Table 4). All isopleths The construction of isopleths was based on following ory as well as on rules of phase (1) The quaternary Al(FeMn)Si are differentiated by Mn/Fe ratio into α-al(femn)si if Mn/Fe 1.1 and β-al(femn)si if Mn/Fe > 1.1. These three systems are formed depending on Mn/Fe ratio of alloy: if Mn/Fe < 1.1, after crystallization, alloys consist of Al α-al(femn)si Si Al5FeSi; if Mn/Fe > 1.1, alloys consist of Al α-al(femn)si Si β-al(femn)si; if Mn/Fe = 1.1, (2) Crossing tilted phase boundary line leads to exhaust or precipitation of one phase, whereas passing through horizontal phase boundary line, where eutectic or peritetic reactions occur, boundary results in eir exhaust (precipitation) of two phases or exhaust of one phase and precipitation of or [20]. In case of AlSi8Fe1 Mn, α-al or α-al(femn)si precipitated primarily, and liquidus line (marked by 1 in Figure8(b)) was drawn by fitting data of primary precipitation temperature. At AlSi8Fe1 side, 2, 3, 4 phase boundaries were extended from corresponding points, which indicates, respectively, precipitation of Al5FeSi, Si, and exhaust of melt. For Mn content from 0.5 to 2 wt. %, exhaust of melts were caused by two four-phase eutectic reactions: Depending on se reactions, 5, 6 phase boundaries were drawn. At nearly 610 C, 7, 8 phase boundaries were drawn because of not only DTA results, but also of fact that a threephase was assumed that precipitation of α-al(femn)si would lead to a decrease of Mn concentration in melt, and refore α-al was assumed to precipitate prior to that of Al5FeSi, according to which, Figure 7. Composition change in Al(FeMn)Si phase with increasing Mn content in alloy group The ternary Al5FeSi disappeared after a specific Mn content was reached in alloy, and caused by fact that Mn content of our Al(FeMn)Si phases changed with Mn content of alloys. On or hand, diagram version proposed by L. Mondolfo [17] cannot be accepted as fundamental information for drawing isopleths because of Al5FeSi disappearance (see above). Since phase compositions are significantly influenced by crystallization conditions, deviations in Mn content of Al(FeMn)Si phases, in comparison to A. Zakharov s study, must be Based on DTA and SEM results, 12 isopleths were drawn (according to Table 4). All isopleths The construction of isopleths was based on following ory as well as on rules of phase (1) The quaternary Al(FeMn)Si are differentiated by Mn/Fe ratio into α-al(femn)si if Mn/Fe 1.1 and β-al(femn)si if Mn/Fe > 1.1. These three systems are formed depending on Mn/Fe ratio of alloy: if Mn/Fe < 1.1, after crystallization, alloys consist of Al α-al(femn)si Si Al5FeSi; if Mn/Fe > 1.1, alloys consist of Al α-al(femn)si Si β-al(femn)si; if Mn/Fe = 1.1, (2) Crossing tilted phase boundary line leads to exhaust or precipitation of one phase, whereas passing through horizontal phase boundary line, where eutectic or peritetic reactions occur, boundary results in eir exhaust (precipitation) of two phases or exhaust of one phase and precipitation of or [20]. In case of AlSi8Fe1 Mn, α-al or α-al(femn)si precipitated primarily, and liquidus line (marked by 1 in Figure8(b)) was drawn by fitting data of primary precipitation temperature. At AlSi8Fe1 side, 2, 3, 4 phase boundaries were extended from corresponding points, which indicates, respectively, precipitation of Al5FeSi, Si, and exhaust of melt. For Mn content from 0.5 to 2 wt. %, exhaust of melts were caused by two four-phase eutectic reactions: Depending on se reactions, 5, 6 phase boundaries were drawn. At nearly 610 C, 7, 8 phase boundaries were drawn because of not only DTA results, but also of fact that a threephase was assumed that precipitation of α-al(femn)si would lead to a decrease of Mn concentration in melt, and refore α-al was assumed to precipitate prior to that of Al5FeSi, according to which phase boundaries were extended from corresponding points, which indicates, respectively, precipitation of Al5FeSi, Si, and exhaust of melt. For Mn content from 0.5 to 2 wt. %, exhaust of melts were caused by two four-phase eutectic reactions: (1) L + α-al + α-al(femn)si + Al 5 FeSi = α-al + α-al(femn)si + Si + Al 5 FeSi and Depending on se reactions, Figure 7. Composition change in Al(FeMn)Si phase with increasing Mn content in alloy group The ternary Al5FeSi disappeared after a specific Mn content was reached in alloy, and caused by fact that Mn content of our Al(FeMn)Si phases changed with Mn content of alloys. On or hand, diagram version proposed by L. Mondolfo [17] cannot be accepted as fundamental information for drawing isopleths because of Al5FeSi disappearance (see above). Since phase compositions are significantly influenced by crystallization conditions, deviations in Mn content of Al(FeMn)Si phases, in comparison to A. Zakharov s study, must be Based on DTA and SEM results, 12 isopleths were drawn (according to Table 4). All isopleths The construction of isopleths was based on following ory as well as on rules of phase (1) The quaternary Al(FeMn)Si are differentiated by Mn/Fe ratio into α-al(femn)si if Mn/Fe 1.1 and β-al(femn)si if Mn/Fe > 1.1. These three systems are formed depending on Mn/Fe ratio of alloy: if Mn/Fe < 1.1, after crystallization, alloys consist of Al α-al(femn)si Si Al5FeSi; if Mn/Fe > 1.1, alloys consist of Al α-al(femn)si Si β-al(femn)si; if Mn/Fe = 1.1, (2) Crossing tilted phase boundary line leads to exhaust or precipitation of one phase, whereas passing through horizontal phase boundary line, where eutectic or peritetic reactions occur, boundary results in eir exhaust (precipitation) of two phases or exhaust of one phase and precipitation of or [20]. In case of AlSi8Fe1 Mn, α-al or α-al(femn)si precipitated primarily, and liquidus line (marked by 1 in Figure8(b)) was drawn by fitting data of primary precipitation temperature. At AlSi8Fe1 side, 2, 3, 4 phase boundaries were extended from corresponding points, which indicates, respectively, precipitation of Al5FeSi, Si, and exhaust of melt. For Mn content from 0.5 to 2 wt. %, exhaust of melts were caused by two four-phase eutectic reactions: Depending on se reactions, 5, 6 phase boundaries were drawn. At nearly 610 C, 7, 8 phase boundaries were drawn because of not only DTA results, but also of fact that a threephase was assumed that precipitation of α-al(femn)si would lead to a decrease of Mn concentration in melt, and refore α-al was assumed to precipitate prior to that of Al5FeSi, according to which, Figure 7. Composition change in Al(FeMn)Si phase with increasing Mn content in alloy group The ternary Al5FeSi disappeared after a specific Mn content was reached in alloy, and caused by fact that Mn content of our Al(FeMn)Si phases changed with Mn content of alloys. On or hand, diagram version proposed by L. Mondolfo [17] cannot be accepted as fundamental information for drawing isopleths because of Al5FeSi disappearance (see above). Since phase compositions are significantly influenced by crystallization conditions, deviations in Mn content of Al(FeMn)Si phases, in comparison to A. Zakharov s study, must be Based on DTA and SEM results, 12 isopleths were drawn (according to Table 4). All isopleths The construction of isopleths was based on following ory as well as on rules of phase (1) The quaternary Al(FeMn)Si are differentiated by Mn/Fe ratio into α-al(femn)si if Mn/Fe 1.1 and β-al(femn)si if Mn/Fe > 1.1. These three systems are formed depending on Mn/Fe ratio of alloy: if Mn/Fe < 1.1, after crystallization, alloys consist of Al α-al(femn)si Si Al5FeSi; if Mn/Fe > 1.1, alloys consist of Al α-al(femn)si Si β-al(femn)si; if Mn/Fe = 1.1, (2) Crossing tilted phase boundary line leads to exhaust or precipitation of one phase, whereas passing through horizontal phase boundary line, where eutectic or peritetic reactions occur, boundary results in eir exhaust (precipitation) of two phases or exhaust of one phase and precipitation of or [20]. In case of AlSi8Fe1 Mn, α-al or α-al(femn)si precipitated primarily, and liquidus line (marked by 1 in Figure8(b)) was drawn by fitting data of primary precipitation temperature. At AlSi8Fe1 side, 2, 3, 4 phase boundaries were extended from corresponding points, which indicates, respectively, precipitation of Al5FeSi, Si, and exhaust of melt. For Mn content from 0.5 to 2 wt. %, exhaust of melts were caused by two four-phase eutectic reactions: Depending on se reactions, 5, 6 phase boundaries were drawn. At nearly 610 C, 7, 8 phase boundaries were drawn because of not only DTA results, but also of fact that a threephase was assumed that precipitation of α-al(femn)si would lead to a decrease of Mn concentration in melt, and refore α-al was assumed to precipitate prior to that of Al5FeSi, according to which phase boundaries were drawn. At nearly 610 C, Figure 7. Composition change in Al(FeMn)Si phase with increasing Mn content in alloy group The ternary Al5FeSi disappeared after a specific Mn content was reached in alloy, and caused by fact that Mn content of our Al(FeMn)Si phases changed with Mn content of alloys. On or hand, diagram version proposed by L. Mondolfo [17] cannot be accepted as fundamental information for drawing isopleths because of Al5FeSi disappearance (see above). Since phase compositions are significantly influenced by crystallization conditions, deviations in Mn content of Al(FeMn)Si phases, in comparison to A. Zakharov s study, must be Based on DTA and SEM results, 12 isopleths were drawn (according to Table 4). All isopleths The construction of isopleths was based on following ory as well as on rules of phase (1) The quaternary Al(FeMn)Si are differentiated by Mn/Fe ratio into α-al(femn)si if Mn/Fe 1.1 and β-al(femn)si if Mn/Fe > 1.1. These three systems are formed depending on Mn/Fe ratio of alloy: if Mn/Fe < 1.1, after crystallization, alloys consist of Al α-al(femn)si Si Al5FeSi; if Mn/Fe > 1.1, alloys consist of Al α-al(femn)si Si β-al(femn)si; if Mn/Fe = 1.1, (2) Crossing tilted phase boundary line leads to exhaust or precipitation of one phase, whereas passing through horizontal phase boundary line, where eutectic or peritetic reactions occur, boundary results in eir exhaust (precipitation) of two phases or exhaust of one phase and precipitation of or [20]. In case of AlSi8Fe1 Mn, α-al or α-al(femn)si precipitated primarily, and liquidus line (marked by 1 in Figure8(b)) was drawn by fitting data of primary precipitation temperature. At AlSi8Fe1 side, 2, 3, 4 phase boundaries were extended from corresponding points, which indicates, respectively, precipitation of Al5FeSi, Si, and exhaust of melt. For Mn content from 0.5 to 2 wt. %, exhaust of melts were caused by two four-phase eutectic reactions: Depending on se reactions, 5, 6 phase boundaries were drawn. At nearly 610 C, 7, 8 phase boundaries were drawn because of not only DTA results, but also of fact that a threephase was assumed that precipitation of α-al(femn)si would lead to a decrease of Mn concentration in melt, and refore α-al was assumed to precipitate prior to that of Al5FeSi, according to which, Figure 7. Composition change in Al(FeMn)Si phase with increasing Mn content in alloy group The ternary Al5FeSi disappeared after a specific Mn content was reached in alloy, and caused by fact that Mn content of our Al(FeMn)Si phases changed with Mn content of alloys. On or hand, diagram version proposed by L. Mondolfo [17] cannot be accepted as fundamental information for drawing isopleths because of Al5FeSi disappearance (see above). Since phase compositions are significantly influenced by crystallization conditions, deviations in Mn content of Al(FeMn)Si phases, in comparison to A. Zakharov s study, must be Based on DTA and SEM results, 12 isopleths were drawn (according to Table 4). All isopleths The construction of isopleths was based on following ory as well as on rules of phase (1) The quaternary Al(FeMn)Si are differentiated by Mn/Fe ratio into α-al(femn)si if Mn/Fe 1.1 and β-al(femn)si if Mn/Fe > 1.1. These three systems are formed depending on Mn/Fe ratio of alloy: if Mn/Fe < 1.1, after crystallization, alloys consist of Al α-al(femn)si Si Al5FeSi; if Mn/Fe > 1.1, alloys consist of Al α-al(femn)si Si β-al(femn)si; if Mn/Fe = 1.1, (2) Crossing tilted phase boundary line leads to exhaust or precipitation of one phase, whereas passing through horizontal phase boundary line, where eutectic or peritetic reactions occur, boundary results in eir exhaust (precipitation) of two phases or exhaust of one phase and precipitation of or [20]. In case of AlSi8Fe1 Mn, α-al or α-al(femn)si precipitated primarily, and liquidus line (marked by 1 in Figure8(b)) was drawn by fitting data of primary precipitation temperature. At AlSi8Fe1 side, 2, 3, 4 phase boundaries were extended from corresponding points, which indicates, respectively, precipitation of Al5FeSi, Si, and exhaust of melt. For Mn content from 0.5 to 2 wt. %, exhaust of melts were caused by two four-phase eutectic reactions: Depending on se reactions, 5, 6 phase boundaries were drawn. At nearly 610 C, 7, 8 phase boundaries were drawn because of not only DTA results, but also of fact that a threephase was assumed that precipitation of α-al(femn)si would lead to a decrease of Mn concentration in melt, and refore α-al was assumed to precipitate prior to that of Al5FeSi, according to which phase boundaries were drawn because of not only DTA results, but also of fact that a three-phase was assumed that precipitation of α-al(femn)si would lead to a decrease of Mn concentration in melt, and refore α-al was assumed to precipitate prior to that of Al 5 FeSi, according to which phase composition area Metals 2018, 8, x FOR PEER REVIEW 9 of 12 phase composition area 9 was determined. Lastly, according to rules of phase boundary drawing, phase boundaries were added in diagram for a complete isopleth. It is worth noting that in case of Al Si8Fe1.5 Mn isopleth, ternary phase Al5FeSi or quaternary α-al(femn)si precipitated primarily, whereas in case of AlSi8Fe2-Mn isopleth, ternary phase Al8Fe2Si or quaternary α-al(femn)si precipitated primarily. In AlSi8Fe2-Mn isopleth, one more phase change occurred before eutectic equilibrium: L + α-al + Al8Fe2Si + α-al(femn)si = L + α-al + Al5FeSi + α-al(femn)si (shown as a dotted horizontal line at 591 C in Figure 8d). Therefore, ternary Al8Fe2Si was absent in microstructure of solid alloys. was determined. Lastly, according to rules of phase boundary drawing, Metals 2018, 8, x FOR PEER REVIEW 9 of 12 phase composition area 9 was determined. Lastly, according to rules of phase boundary drawing, phase boundaries were added in diagram for a complete isopleth. It is worth noting that in case of Al Si8Fe1.5 Mn isopleth, ternary phase Al5FeSi or quaternary α-al(femn)si precipitated primarily, whereas in case of AlSi8Fe2-Mn isopleth, ternary phase Al8Fe2Si or quaternary α-al(femn)si precipitated primarily. In AlSi8Fe2-Mn isopleth, one more phase change occurred before eutectic equilibrium: L + α-al + Al8Fe2Si + α-al(femn)si = L + α-al + Al5FeSi + α-al(femn)si (shown as a dotted horizontal line at 591 C in Figure 8d). Therefore, ternary Al8Fe2Si was absent in microstructure of solid alloys. Metals 2018, 8, x FOR PEER REVIEW 9 of 12 phase composition area 9 was determined. Lastly, according to rules of phase boundary drawing, phase boundaries were added in diagram for a complete isopleth. It is worth noting that in case of Al Si8Fe1.5 Mn isopleth, ternary phase Al5FeSi or quaternary α-al(femn)si precipitated primarily, whereas in case of AlSi8Fe2-Mn isopleth, ternary phase Al8Fe2Si or quaternary α-al(femn)si precipitated primarily. In AlSi8Fe2-Mn isopleth, one more phase change occurred before eutectic equilibrium: L + α-al + Al8Fe2Si + α-al(femn)si = L + α-al + Al5FeSi + α-al(femn)si (shown as a dotted horizontal line at 591 C in Figure 8d). Therefore, ternary Al8Fe2Si was absent in microstructure of solid alloys. phase boundaries were added in diagram for a complete isopleth. It is worth noting that in case of Al Si8Fe1.5 Mn isopleth, ternary phase Al 5 FeSi or quaternary α-al(femn)si precipitated primarily, whereas in case of AlSi8Fe2-Mn isopleth, ternary phase Al 8 Fe 2 Si or quaternary α-al(femn)si precipitated primarily.