# Fileset

[65_F-M2024805.pdf](https://mdr.nims.go.jp/filesets/f8ae8970-18ad-4854-ac02-d2b08ad1eee6/download)

## Creator

[Machiko Ode](https://orcid.org/0000-0002-9500-5466), Hisao Esaka, [Akira Ishida](https://orcid.org/0000-0002-5188-1093), [Susumu Takamori](https://orcid.org/0000-0003-3422-9391), [Hideyuki Murakami](https://orcid.org/0000-0001-8220-5816)

## Rights

[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Observation and Numerical Prediction of Concentration Distribution at Cast Coating Interface of Solid Pt, Ir, Re Using Liquid Ni-Based Alloys](https://mdr.nims.go.jp/datasets/1028b94f-4d73-4a26-aeca-f4046424bc95)

## Fulltext

Observation and Numerical Prediction of Concentration Distribution at Cast Coating Interface of Solid Pt, Ir, Re Using Liquid Ni-Based AlloysObservation and Numerical Prediction of Concentration Distribution at Cast CoatingInterface of Solid Pt, Ir, Re Using Liquid Ni-Based Alloys+1Machiko Ode+2, Hisao Esaka, Akira Ishida, Susumu Takamori and Hideyuki MurakamiNational Institute for Materials Science, Tsukuba 305-0047, JapanThe applicability of a cast-coating process for improving the oxidation resistance of cast Ni-based superalloys was evaluated. Specifically,metallic plates of Pr, Ir, and Re expected to improve oxidation resistance when they are enriched on the cast alloys were placed in a mold andcast coating using Ni-10at%Al alloy was performed in order to investigate the formation of the Pt, Ir, or Re-enriched layer on the casting surface.Then the microstructure of the Ni-based alloy/specimen interface was observed. To analyze the concentration profile in the interdiffusion region,solidification and diffusion simulations were performed. It was found that Pt easily dissolves into the molten Ni-based alloy, and Re cannotexpected to modify cast metal surfaces due to its low solubility into the Ni-10at%Al alloy. On the other hand, Ir forms smooth interdiffusionlayer, and numerical calculations predicted that Ir can maintain the modification ability even in a process time of 1 hour, which is equivalent tothe casting time of Ni-based turbine blades. [doi:10.2320/matertrans.F-M2024805](Received December 6, 2023; Accepted February 2, 2024; Published March 8, 2024)Keywords: cast-coating, Ni-base superalloy, solid/liquid interface, solidification simulation1. IntroductionPlatinum group metals and rare metals generally have ahigh melting point and many have excellent mechanicalproperties at high temperatures, oxidation resistance,corrosion resistance and catalytic performance.1) However,since they are very expensive, a method is required, whichcan unevenly distribute them on the surface of the componentin a way that is suitable for each application. For example, inthe case of catalytic applications, in order to cover sufficientsurface area of the catalyst material, it is attached to thesurface of the carrier as fine particles achieving both materialsavings and catalytic performance. Similarly, when usingthe oxidation and corrosion resistance properties of platinumgroup metals, high cost performance can be achieved byforming a dense thin film covering the surface of the basematerial.The plating method is the industrially cheapest thin-filmdeposition process due to its inexpensive equipment, whichproduces little waste of raw materials. However, some raremetals such as W and Re are difficult to plate. On the otherhand, physical vapor deposition methods such as sputteringand PVD are adaptable to various materials but expensiveprocesses in terms of material yield. In principle, the innerwall surface of hollow components cannot be modified.Therefore, there is a need to develop a process that isinexpensive, simple and can uniformly modify all surfacesof components. Ni-based superalloy turbine blades have atwo-layer coating on the outside for oxidation resistanceand heat shielding, and a cooling circuit on the inside for air-cooling the turbine blades. Such improvements in internaland external system allow Ni-based alloy blades to be used atoperating temperatures above their melting point. However,as a result of the higher operating temperatures, oxidation ofthe uncoated inner blade wall has become a problem.The cast-in insert method is one of the casting techniquesemployed to obtain surface-modified metallic materials. Inthis method, a component with a higher melting point thanthe molten metal temperature is placed in the mold inadvance. The casting and the component are added duringcasting to combine different materials while taking advantageof the moldability of casting.2,3) In another method, thecoated casting process,4) the surface of the casting material ismodified by first coating the mold surface with a slurry ofcemented carbide powder such as titanium or WC, and thencasting Mg, Al alloy, steel or cast iron into it.5–7) This is alsoreferred to as a cast coating. In both cases, good control ofthe reaction between the molten metal and the solid metal inthe casting process is necessary to achieve the requiredproperties. The aim of this study was to evaluate whether it ispossible to modify the surface of Ni-based alloy castings byapplying the above metallic elements to the surfaces of moldsand cores for using expensive elements more efficiently.Three elements were selected: the two from platinum groupmetals, Pt and Ir, and a rare metal with a high melting point,Re. The platinum group metal Pt is already in practical useas an oxidation-resistant bond coating for Ni-based super-alloys. The other metals Re and Ir have particularly highmelting points, among the elements under research anddevelopment. First, in order to gain a basic understandingof the process, small pieces of Pt, Ir and Re metal were cast-coated on a Ni alloy, and the interface was observed. Then,to gain knowledge about the interaction between the solidmetal and the molten metal, the temperature change of themetallic pieces during casting was predicted by solidificationheat transfer calculations. The theoretical calculations ofthe interfacial diffusion layer were carried out using theobtained temperature changes as input conditions forthermodynamic calculation software. Based on the results,it is discussed whether surface modification with Pt, Ir andRe as the main elements can be obtained by the castingmethod. For clarity, casting for the purpose of surfacemodification will henceforth be referred to as cast-coating inthis paper.+1This Paper was Originally Published in Japanese in J. JFS 94 (2022) 673–683. Some spelling errors were modified.+2Corresponding author, E-mail: ODE.Machiko@nims.go.jpMaterials Transactions, Vol. 65, No. 5 (2024) pp. 541 to 551©2024 Japan Foundry Engineering Societyhttps://doi.org/10.2320/matertrans.F-M20248052. Methods2.1 Casting experimentAbout 2 kg of Ni-10at%Al alloy was melted in a vacuuminduction melting furnace and poured into a mold fixed withsmall pieces of pure Pt, Ir and Re (approximately 20mmlong, 10mm wide and 1mm thick). A schematic diagram ofthe experimental apparatus and a detailed drawing of themold are shown in Fig. 1. The heat flow was controlled inone direction by using a rectangular mold, which had 18mm-thick SUS on one side and ceramic insulation on other sides.There are two methods for fixing the Pt, Ir, and Re samples.In one method, they are directly affixed to the SUS platesurface using an inorganic adhesive. In the other method,they are fixed to a cylindrical quartz glass stand with slitsusing an inorganic adhesive. These assemblies are thenpositioned at 20mm and 40mm from the SUS plate on thebottom of the mold. The contact time of the sample with themolten metal changes according to the distance betweenthe sample and the SUS plate. Therefore, the dependenceof the interfacial reaction of the sample/Ni-based alloy on themolten metal contact time can be verified.Temperature changes in the mold were measured usingthree B type thermocouples. The height of the thermocou-ples in the mold was approximately 20mm from the bottomof the mold. Their positions in the direction of heat flowwere 15mm and 35mm from the SUS plate surface and SUSplate center, respectively. Temperatures were obtained usinga data logger (GRAPHTEC GL200A) with a time intervalset to 0.1 second. Casting experiments were carried out onceeach at different casting temperatures of approximately1650°C, 1600°C and 1500°C. In this experiment, as Ptsamples were found to dissolve easily, the Pt sample wasonly placed on the SUS plate surface in the case of 1600°Ccasting temperature. A sample was cut from the ingot andmirror polished. The microstructure and concentrationdistribution near the interface between the sample and theNi-based alloy were observed using SEM-EDX (HitachiHigh-Tech S4700).2.2 SimulationsTwo different type of computer simulations were carriedout. The first was for estimating the temperature of thesample during casting by comparing temperature measure-ments from casting experiments with solidification heattransfer calculations. The second was for predicting theconcentration near the interface between the sample and theNi-based alloy using the temperature obtained from theheat transfer calculations as input value. For validation ofthe latter, the concentration distribution calculated by thesimulation were compared with experimental observations.The ability to apply cast-coating to long solidificationprocesses such as turbine blade casting was investigatedbased on the verified prediction method.Figure 2 shows the schematic illustration of calculationsystem and temperature distribution. Assuming that theadiabatic conditions are satisfied except for the SUS plate,the heat flow is assumed to be in one direction. The enthalpymethod was used in the heat transfer calculations. Theenthalpy method is outlined below.The enthalpy at temperature T in the solid-liquidcoexistence temperature range is expressed as follows.HðT Þ ¼ H0 þZ TT0CpdT þ ð1� fsÞL ð1ÞFig. 1 Experimental apparatus. (a) Schematic of vacuum-inductionfurnace. (b) Top view of the mold. Approximately 2 kg of Ni-10at%Alalloy was melted in the vacuum-induction furnace and cast into the moldin which pure Pt, It, and Re specimens are fixed.Fig. 2 Schematic illustration of calculation system. The heat flow can beregarded as unidirectional, so that the temperature distribution could berepresented by a one-dimensional calculation.M. Ode, H. Esaka, A. Ishida, S. Takamori and H. Murakami542Where H0 is the enthalpy of the standard state, Cp is thespecific heat, L is the latent heat of solidification and fs is thesolid fraction. Partial differentiation of eq. (1) with temper-ature results in the specific heat in the solid-liquidcoexistence temperature range taking into account the latentheat of solidification, as shown in eq. (2).@H@T¼ Cp �@fs@TL ð2ÞBy defining the effective specific heat CPeffasCPeff¼CpCp �@fs@TL ðTS < T < TLÞ8<: ð3ÞThe relationship between specific heat and enthalpy is asfollows, regardless of whether it is a single-phase or solid-liquid coexisting state.@H@T¼ CPeffð4Þwhere TL is the liquidus temperature and TS is the solidustemperature.The governing equations are derived with a focus onnumerical computations. First, we consider a small unitvolume (computational grid) with a surface area ¦A andvolume ¦V. Let J be the amount of heat (heat flux) flowingthrough the unit surface per unit time ¦t, then relationshipbetween J and the time variation of temperature can bederived from the energy conservation law and the continuityequation asCPeff�T�t¼ �A�VJ ð5ÞUsing ¦A/¦V = 1/¦x for one-dimension case, where ¦xis the unit length, and substituting eq. (4) into eq. (5), thefollowing equation is obtained.�H�t¼ J�xð6ÞThe heat flux J was classified into four categories,including inside the ingot and three types of boundaries, asshown below.i) Inside the ingotThe heat flux inside the SUS plate is assumed to followsFourier’s law.J ¼ ­@T@xð7Þwhere ­ is the thermal conductivity of SUS.ii) Because surfaces other than the SUS plate were thermallyinsulated,J ¼ 0 ð8Þiii) Assuming that the heat dissipation from the SUS platesurface follows Newton’s law of cooling, the heat flux is,J ¼ hairðTsus1 � TfurnaceÞ ð9Þwhere hair, the heat transfer coefficient is assumed constantin this study, Tsus1 is the outer surface temperature of the SUSplate and Tfurnace is the internal temperature of the furnace,which was set to 50°C based on actual measurements beforecasting.iv) The heat transfer between the SUS plate and the Ni-basedalloy is assumed to be the same as shown in eq. (9) above thesolidification initiation temperature, Tc, because the Ni-basedalloy is a fluid while it is in the liquid phase. On the otherhand, after the start of solidification, a contact resistance, R¹1,exists between the SUS plate and the solidified layer, whichcan be shown as follows.J ¼ hliqðTsus2 � TNi=susÞ TNi=sus � TcJ ¼ R�1ðTsus2 � TNi=susÞ TNi=sus < Tc ð10Þ1/R = ¬­/¦x, where ¬ is the contact conductance, the secondline of eq. (10) is equivalent to Fourier’s equation (7).The values of hair, hliq, R¹1 and Tc in each equation weredetermined by simulation and measured temperatures and areregarded as constants. The heat transfer coefficients are notconstants and may vary considerably with temperature andtime. However, the aim of this calculation is to estimate thesample temperature, not to obtain detailed values for thetransfer coefficients. The transfer coefficients are assumedto be constants, because if the coefficients are assumed asfunctions of temperature and time, an infinite number ofplausible combinations will be found. The heat transfer at thecasting/mold interface decreases rapidly once solidificationhas started due to solidification shrinkage. Therefore, thenumber of parameters required for the calculation wasreduced by setting R¹1 = ¡ · hliq, (¡ = 0.1) to simplify thecalculation for thermal resistance. If ¡ is set to differentvalues less than 1, the corresponding changes in hliq havelittle effect on the desired computational prediction results.In the numerical calculations, the heat flux is first obtainedfrom the temperature distribution from eqs. (7)–(10) and thetime variation of enthalpy from eq. (5). The temperaturechange is then determined from the temperature-enthalpyrelationship in eq. (1). The temperature-enthalpy relationshipwas obtained from the Ni-10at%Al alloy equilibriumsolidification calculation using CompuTherm’s thermody-namic database PanNi2020a and the phase diagramcalculation software PANDAT. The database is available forany binary combination of the five elements Ni, Al, Pt, Ir andRe in the entire composition range. For ternary systems, thereare no compositional restrictions for the major phases in theNi-Al-Pt and Ni-Al-Re systems and for the fcc phase andits equilibrium phase in the Ni-Al-Ir system.8,9) From thecalculation results, the temperature-enthalpy relationship canbe approximated by eq. (11). The thermodynamic propertiesused in the calculation are given in Table 1. The enthalpy-Table 1 Thermodynamic properties of Ni-10at%Al alloy used in simu-lations.Observation and Numerical Prediction of Concentration Distribution at Cast Coating Interface of Solid Pt, Ir, Re 543temperature relationship thus obtained is shown in Fig. 3.The values in Table 1 were selected giving priority to thereproducibility around the solid-liquid coexistence temper-ature. Therefore, the maximum error is observed in thetemperature range away from the solid-liquid coexistencetemperature, but the value is within about 2.5% even around1000°C, where the error is the largest. The thermalconductivity of the alloys was obtained by linearlycombining the data for pure Ni and pure Al from the MetalsData Book10) by composition ratio.HðT Þ ¼ HTl þ clðT � TlÞ : T > TlHðT Þ ¼ HTs þ@H@TðT � TsÞ : Ts � T � TlHðT Þ ¼ HTs þ csðT � TsÞ : T < Tsð11ÞThere are three parameters in the calculation: hair, hliq and Tc.These parameters were determined using the length of theliquidus temperature holding time of the thermocouple in themold, which was obtained from actual measurements, as anindicator.The governing equations were solved numerically inPython.The second concentration field prediction was carried outusing the Pandat diffusion module and the diffusion mobilitydatabase for Ni-based alloys (PanNi2020_MB). In thedatabase, only the fcc phases of the Al-Ni, Al-Pt, Ni-Ir, Ni-Pt and Ni-Re systems relevant to this study were assessed. Asthe alloys used in this study have a high concentration of Ni-90at%, the calculated concentration profiles of the Ir and Ptsamples and Ni can be used for discussion. The hcp phase hasnot been assessed, but reaction diffusion is expected to occurbetween the Re sample and the Ni-based alloys, where hcpphase is expected to precipitate phase diagram. This can bediscussed qualitatively from the viewpoint of the equilibriumphase diagram as well as diffusion. The details of the specificinput conditions are described in the section, temperatureprediction by enthalpy method in results and discussion.3. Results and Discussion3.1 Temperature measurements and numerical predic-tions3.1.1 Results of temperature measurementsFigure 4 shows the temperature variation during castingover a period of 600 s from the time of casting. In thetemperature measurement at 1600°C, the thermocouple at35mm was disconnected at around 350 s. The temperaturecurve looks very smooth compared to the other twoexperiments because it is not influenced by the electricalnoise from the power supply.In the case of unidirectional solidification, the columnardendrites grow in the direction opposite to the heat flow.During solidification, the growing tip is maintained at theliquidus temperature, so that the maximum temperature onthe liquidus side decreases with time from the castingtemperature and then remains at the liquidus temperature.Therefore, the temperature curve inside the casting forms aplateau that maintains the liquidus temperature for a certaintime.11) In all cases in this experiment, the thermocoupleinside the casting formed a constant temperature plateau atabout 1440°C. Furthermore, in all cases, the plateautemperature holding time at 15mm from the SUS plate wasshorter than that at 35mm, indicating that solidification wasproceeding unidirectionally from the SUS plate side. Theholding time will be discussed in detail in the heat transfercalculation section. Figure 5 shows the calculated phasediagram of the Ni-Al binary system on the Ni-rich side. Theliquidus temperature of the Ni-10at%Al alloy indicated byFig. 3 Enthalpy change depending on temperature for Ni-10at%Al alloyobtained by equilibrium solidification simulation using PANDAT andPanNickel thermodynamic database.Fig. 4 Temperature change in the mold measured by thermocouples. The casting temperature is (a) 1650, (b) 1600 and (c) 1550°C,respectively. In all cases, the temperature curve obtained inside the casting shows a constant temperature plateau at about 1440°C and theholding time at 35mm is longer than that at 15mm. This result, “the farther the longer” suggests that the solidification progressesunidirectionally from the SUS plate side.M. Ode, H. Esaka, A. Ishida, S. Takamori and H. Murakami544the arrow is about 1445°C, which is in good agreement withthe experimentally measured plateau temperature.The temperature of the thermocouple inside the SUS platerises by absorbing heat from the molten metal. Under allcasting conditions, the temperature quickly rises to around600°C within about one minute after casting, and thencontinues to rise slowly before maintaining an almostconstant temperature. This is because as the temperatureof the SUS plate rises, the heat absorption of the SUS platedecreases while the amount of heat released into the furnaceincreases. The heat balance is practically achieved at atemperature, which remains almost constant until thegeneration of latent heat of solidification ceases. Thetemperature in the SUS plate at 600 s were 770°C, 707°Cand 690°C, consistent with the casting temperatures of1650°C, 1600°C and 1500°C, respectively.3.1.2 Temperature prediction by enthalpy methodFigure 6 shows an example of the calculation results usingthe enthalpy method described in Section 2.2. It presents thevariation of the temperature over time within the mold, asdepicted in the schematic diagram in Fig. 2 (Fig. 6(a)). Byplotting the relationship of temperature changes over time atboth the thermocouple and sample positions, we can obtaintemperature changes similar to those measured by thethermocouples (Fig. 6(b)). The temperature gradient in thesolidifying casting across the solid liquid interface is higheron the solid side than on the liquid side due to the latent heatof solidification. After solidification is finished, the temper-ature distribution becomes smooth throughout the castingand the temperature difference within the casting becomessmaller. In other words, under the temperature conditionshown in Fig. 2, the temperatures of the samples andthermocouples decrease maintaining a constant temperaturedifference even after the solidification process, as shown inFig. 6(b). However, the temperatures of the two thermocou-ples shown in Fig. 4 relatively quickly converged after theybecomes smaller than the plateau temperature at about 600seconds. This may be due to the fact that the heat flow in themold is not completely one-dimensional, but also flowstowards the adiabatic wall and the top of the mold.However, the macroscopic shrinkage cavity, which wereconsidered to be the final solidification region, were locatednear the gate. Thus, it can be concluded that unidirectionalsolidification was achieved at least up to the gate area.Specifically, unidirectional solidification is reliably achievedup to a sample temperature of 1440°C and it continuesuntil approximately 1200°C, based on the results shown inFig. 6(a). From the above, when estimating the sampletemperature by numerical calculation, hair, hliq and Tc weredetermined to minimize the sum of the squares of thedifference between the actual measurement of the plateautemperature holding time and its predicted time in the hightemperature range, which have the greatest influence ondiffusion.For the prediction of the temperature plateau retentiontime, three parameters were explored in the realistic ranges asfollows:1000 < hair < 50000 ðincrement: 500Þ1000 < hliq < 50000 ðincrement: 1000ÞTc ¼ Ts � x0 < x < 100 ðincrement: 10ÞFigure 7(a) shows the sum of squares of the differencebetween the experimentally measured and calculated temper-ature plateau holding time for a casting temperature of1650°C. To visualize the error between the measured valuesand predicted values more clearly, the parameter x is fixed atFig. 5 Calculation phase diagram of Ni-Al binary system. The liquidustemperature of Ni-10at%Al is about 1455°C.Fig. 6 Time change of the calculated temperature. (a) Temperature profilealong heat flow. Each curve is drawn every 100 sec. Temperature gradientvaries at the solid-liquid interface. (b) Temperature change by fixed pointmeasurement. In the case of experiment (Fig. 4), the temperaturedifference between two points, d = 15mm and d = 35mm, showsmaximum when the point, d = 35mm, solidifies and then decreases. Onthe other hand, in the present simulation, the temperature difference hardlychanges over a wide range.Observation and Numerical Prediction of Concentration Distribution at Cast Coating Interface of Solid Pt, Ir, Re 5450°C, and the two horizontal axes are assigned to the heattransfer coefficients, hair and hliq. The figure shows that theheat transfer coefficient between the SUS plate interface andthe casting, hliq, is the dominant factor in the temperatureprediction accuracy at Ts ¹ Tc = 0°C. Figure 7(b) shows thesum of the squared time differences used for the vertical axisin Fig. 7(a) as the radius of the sphere, with Ts ¹ Tc on thevertical axis. The right side of the horizontal axis is hair andthe left side is hliq. The smaller the radius of the sphere, thesmaller the error is in predicting the plateau time, as shownin Fig. 7(a). The prediction accuracy is highly dependent onthe change in hliq and is less influenced by hair and Ts ¹ Tc.Considering the mold designed for one-dimensional heatflow and the calculation conditions, it is natural to expect thatthe heat is lost to the outside due to the heat transfer betweenthe SUS/casting, and this is the dominant factor in theplateau holding time. The minimum prediction error is athair = 6500W/m2K, hliq = 8000W/m2K, Ts ¹ Tc = 0°C,with an error of 17 s2. Although hair value is very largecompared to the typical value of heat transfer coefficient innatural convection,12) it is apparent from Fig. 7 that thedominant factor influencing prediction errors is hliq. Thecontribution of hair to these errors is negligible. Therefore,in order to appropriately optimize the value of hair, norefinements to the model such as multidimensionalization thecomputational domain or parameterizing h as a function oftime or temperature will be pursued. Table 2 summarizesthe predicted values of the calculation and experimentalresults. Similar parametric optimization calculations werealso performed at 1600°C. They showed that when hliq asdifferent (9000W/m2K) from the optimum parameter set at1650°C, the plateau time was predicted with an error of lessthan 1 s2. On the other hand, when the molten metaltemperature was 1500°C, the model could not predict theplateau retention time accurately. There are several reasonsfor this. Figure 4 shows that at casting temperatures of 1650and 1600°C, the thermocouples in the molds recordedtemperatures above the liquid phase line temperatureimmediately after casting. On the other hand, at 1500°C,the thermocouples show temperatures below the liquidus linetemperature, suggesting that solidification has already startedsomewhere in the casting when the molten metal reachesthe thermocouples. If solidification progresses unidirection-ally from the SUS plate, the time for the thermocouple inthe mold to hold the plateau should increase as the castingtemperature increases. At the 35mm position, the plateauholding time at 1500°C and 1600°C was 165 s and 170 srespectively. The higher the casting temperature, the longerthe plateau retention time, and this is considered reasonablecasting. However, at the 15mm position, the plateau retentiontime was 90 s at 1500°C, which is 30 s longer than the castingtemperature of 1600°C and is inconsistent with the timeexpected from the casting temperature. This result suggeststhat as the molten metal was not sufficiently superheated inthe experiment at 1500°C, solidification and melting mayhave occurred inhomogeneously in time and space on themold surface, resulting in the heat flow being morecomplicated and unidirectional than at 1650°C and 1600°C.As a result, it is expected that the model assumingunidirectional solidification could not reproduce the plateautime accurately by parameter fitting. However, the liquidphase temperature holding time of the thermocouple waslonger at the 35mm position than at 15mm. The results ofcutting the castings showed that the shrinkage cavities, whichwere thought to be the final solidification area, were locatedjust below the gate (as shown in Fig. 1), suggesting thatTable 2 Comparison of plateau time interval (s) between experimental andsimulation results.* predicted by experimentFig. 7 Effect of calculation parameters on difference between measuredand calculated holding time of temperature plateau. (a) The parameter Tcis fixed to Ts and X and Yaxis represents heat transfer coefficients, hliq andhair, respectively. The effect of hliq on the prediction error is much largerthan that of hair. (b) The parameter, Tc is treated as a valuable and assignedto the Z axis. The difference between calculated and measured time isexpressed as the radius of the sphere at each parameter space. Theparameter, hliq, is dominant factor of the plateau time.M. Ode, H. Esaka, A. Ishida, S. Takamori and H. Murakami546solidification was unidirectional overall even in the 1500°Csample. Therefore, plateau retention time at the sampleposition at a casting temperature of 1500°C was assumed tobe proportional to the distance from the SUS plate, and thevalue estimated from the 1500°C thermocouple measurementwas used.The results show that the time, when the sample is incontact with the liquid phase, does not exceed 300 s, evenfor samples located far from the SUS plate. The plateautemperature time thus obtained from the calculations wasemployed in the concentration distribution simulation.3.2 Microstructural observation and concentrationdistribution predictionFigure 8 shows SEM image of the interface in the Irsample placed at 40mm from the surface of the SUS plateand the concentration distribution. The distance from the topof the sample on the opposite side of the SUS plate were6mm (1650°C) and 3mm (1500°C), respectively. In the caseof the Ir sample, the cast-coating layer with the Ni-basedalloy was in a good diffusion bonding state and no differencewas observed in the concentration distribution between themeasurement positions (SUS side and opposite side, in theheight direction from the bottom of mold). The thicknessof the diffusion layer is indicated by the arrows: 25 µm at amolten temperature of 1650°C (Fig. 8(a)) and 20 µm at1500°C (Fig. 8(b)). Here, the Al concentration in the Irsample appears to vary with casting temperature is due toan analytical error and is not considered to be a significantdifference. This is because the Al-K¡ line in the EDSspectrum is too close to the Ir-Mζ line peak, so even if thereis no Al in the Ir alloy sample, it can be misidentified ascontaining a small amount of Al.Figure 9 shows the results of the Re sample placed at40mm and the molten temperature was 1500°C. The Resample is characterized by the formation of a solute diffusionlayer on the Re sample side only, even though no cracks wereobserved between the sample and the Ni-based alloy. Ascan be seen from the cross section, the Re sample has asintered microstructure with a density of about 90%. Thereason for the formation of the diffusion layer only on theRe side is not due to the effect of molten metal penetrationinto the sintered microstructure, as will be explained in thecomparison with the diffusion simulation to be shown later.The Pt sample affixed directly to the SUS304 plate remainedpartially at both 1650°C and 1500°C, but the sample placedinside the casting was completely melted. The melting pointof Pt is 1769°C, which is approximately 700°C lower thanthat of Ir (2443°C) and 1400°C lower than that of Re(3180°C). The equilibrium phase diagram also explains theease of melting of Pt; Fig. 10 shows the Pt/Ni-10Alat%longitudinal sectional phase diagram. The liquid phase regionextends well towards the low temperature side, and theliquidus temperatures of Ni-10at%Al and Ni-5Al-50at%Ptcan be regarded as almost identical. In other words, themelting point of the Pt sample drops rapidly on the surfacein contact with the Ni-10at%Al alloy due to interdiffusion,indicating that the Pt sample melts quickly even at around1500°C.The temperature change at the sample was estimated, andusing this as an input condition, the PANDAT diffusionmodule and PanNi2020 mobility were used to calculate theconcentration change near the sample and molten metalinterface due to solute diffusion. The calculation system isone-dimensional and assumes a diffusion couple between oneof the precious metal samples and the Ni-10at%Al alloy. AsFig. 8 Cross section in vicinity of Ir specimen/Ni-Al cast alloy interface and concentration profile obtained using SEM-EDS. Thetemperature of molten Ni-Al alloy is (a) 1650°C (b) 1500°C, respectively. In both cases, concentration profiles are smooth and there is nosecondary phase in the diffusion zone.Observation and Numerical Prediction of Concentration Distribution at Cast Coating Interface of Solid Pt, Ir, Re 547the calculation condition, the temperature was set to beslightly higher than the liquidus of Ni-10at%Al and thetime was the plateau time obtained from the heat transfercalculations, because the solute diffusion layer is mainlyformed when the Ni-based alloy is in the liquid state. Thespecific set temperature was 1450°C, approximately 4°Chigher than the liquidus of the Ni-10at%Al alloy (1445.4°C).This is to avoid numerical instability on the Ni-base alloyside in the diffusion couple, where the liquid/solid phaseratio fluctuates greatly due to inevitable numerical errors indiffusion calculation when the calculations were performednear the liquidus temperature.Figure 11 shows the calculation results for the Ir samplewhich placed at 40mm and molten temperature was 1650°C/1500°C, and the plateau time was 267 s/184 s, respectively.The concentration profiles of both Ni and Ir can be regardedas symmetrical with respect to the original interface set at50 µm. The length indicated by the arrow in the figure is25 µm. The experimental results (Fig. 8) and the calculateddiffusion distance of Ir are in good agreement. The alloyingelement, Al does not decrease monotonically towards the Irside, which is due to the up-hill diffusion of Al. In fact,according to the diffusion database the diagonal term ofthe interdiffusion coefficient, DAlIr (Al diffusion caused byIr concentration gradient), showed positive when the Irconcentration was below about 40 at% in the calculatedconcentration distribution at 1450°C. The concentrationdistribution in Fig. 8, the width of the Al concentrationchange is narrower than that of Ir and Ni and the change isobserved on the high Ir concentration side. In Fig. 8(a), theAl concentration peak at a distance of about 22 µm is alsothought to be caused by the up-hill diffusion effect of Al.The calculated Al concentration in the interdiffusion layerdecreases at about 12 µm on the Ni alloy side from the initialinterface. This concentration drop corresponds to the solid-liquid interface at 1450°C and results in a slight deviationfrom the experimental observation in which the sample isfurnace-cooled to room temperature.The result from the Re sample shows that the change in theconcentration distribution is observed only on the Re side(Fig. 12), which is in good agreement with the experimentalresult (Fig. 9). Since the Re sample is not assumed to be asintered body in the numerical calculation, the concentrationdistribution appearing only in the Re sample is not due to theRe sample being a sintered body, but it is a characteristic ofthe Re element. Here, the distance of concentration changesFig. 10 Vertical section of the calculated Ni-Al-Pt ternary phase diagram.When Pt is added to Ni-10% Al alloy to increase the melting point, Ptmust be added at least 50 at%.Fig. 11 Calculated diffusion profile in vicinity of Ir/Ni-10at%Al interface.The length of the diffusion region indicated by the arrow is about 25 µmwhich is in good agreement with experimental results.Fig. 9 Cross section in vicinity of Re specimen (d = 40mm)/Ni-Al castalloy interface and the concentration profile obtained using SEM-EDS.The casting temperature of Ni-Al alloy is 1500°C. Solute partitioningoccurs at the interface.M. Ode, H. Esaka, A. Ishida, S. Takamori and H. Murakami548on the Re side is much longer than that in the experiment.Calculation of the diffusion database shows that the diffusioncoefficient of Ni in the Re-rich hcp phase is about3 © 10¹10m2/s at 1450°C, which is about 10,000 timeslarger than that in Ir and Pt at the same temperature.Extrapolating the experimental diffusion coefficients of Ni-Re binary alloys13) to 1450°C, the diffusion coefficients arealmost equal to those in Ir and Pt, so the calculated diffusiondistance into the Re sample is not highly reliable.Figures 13 and 14 show the longitudinal sectional phasediagrams for pure substance X (X = Ir, Re) and Ni-10at%Al,and isothermal cross sections at 1450°C. The dotted line inthe figure corresponds to the temperature of 1450°C in thelongitudinal section and the concentration of Ni-10at%Al andX (X = pure Ir or Re) in the isothermal section, respectively.The phase diagram shows that no large concentration gap isformed at the solid-liquid interface between pure Ir and Ni-10at%Al alloy, and that a smooth concentration distributionis formed near the interface because the fcc phase is stableregardless of Ir concentration in the solidified state.In the case of the Re/Ni-10at%Al system, on the otherhand, the formation of a Re-enriched layer on the Ni-basedalloy surface is difficult due to the characteristics of the phasediagram, as shown in Fig. 14. The longitudinal sectionalphase diagram of the Re system is not a simple completesolubility type as in the Ir system. The isothermal crosssection shows that the molten metal is expected to be inequilibrium with two solid phases, the hexagonally closedpacked (hcp) and face-centered-cubic (fcc) structures. Asshown in Fig. 12(b), near the interface between the Resample and Ni-10at%Al, the phases are arranged in the orderhcp/hcp+fcc/fcc/fcc+liquid/liquid. The concentrations atthe interface positions indicated by ① and ② in the figurecorrespond to the phase boundary concentrations in the phasediagram. In other words, the dotted line in the longitudinalcross section (1450°C) and the isothermal cross section showthat Re is not soluble in the liquid phase, i.e., on the Ni-richside, and thus cannot form a diffusion layer, and that onlyNi can diffuse into the hcp phase up to 20 at%. In theexperiment, Al is distributed below 10 at% in both the NiFig. 12 Calculated diffusion profile in vicinity of Re/Ni-10at%Al interface. (a) Concentration distribution over the entire calculation area,(b) Enlarged concentration profile near the initial interface; in the case of Re, two phase boundaries, hcp/fcc and fcc/liq interfaces, areformed.Fig. 13 Vertical and isothermal (1450°C) section of calculated Ni-Al-Ir ternary phase diagram.Observation and Numerical Prediction of Concentration Distribution at Cast Coating Interface of Solid Pt, Ir, Re 549alloy and the Re sample. As in the case of Ir, the Al-K¡and Re-Mζ peaks are close together, making it difficult tomeasure dilute Al in high concentrations of Re by EDSanalysis. When EDS analysis was carried on the areas, wherethe Ni-based alloy and Re were not diffusion bonded dueto voids and cracks, Al was detected at about 10 at% in theRe sample, though Ni was not detectable. This means that itis difficult to measure the Al concentration distribution inthe Re sample by EDS analysis, except in areas where Reconcentration is reduced due to the interdiffusion of Ni, andAl does not appear to have diffused into the Re sample asshown in Fig. 9. Qualitatively, the concentration distributionin the Re sample shown in Fig. 9 can be explained bythermodynamic analysis using the diffusion simulation andequilibrium phase diagrams, indicating that the solutediffusion layer is only observed in the Re sample is not theresult of inherent properties of the sintered microstructure.Together with the results for the Ir system, the diffusionsimulation allowed for the reasonably accurate prediction ofthe properties of the cast-coating interface.Finally, the applicability of cast-coating with the preciousmetals used in this study to Ni-based superalloy componentsis discussed. As shown by the results, the application of cast-coating is difficult for Pt because of liquidus depressioncaused by contact with Ni-based alloy. It is difficult for Realso, because it is not diffusively absorbed into the surface ofNi-based alloy. On the other hand, Ir has the strong potentialfor application.In general, the unidirectional solidification process used inmanufacture of turbine blades, more than one hour is requiredfor the liquid phase to completely solidify. Therefore,calculations were carried out under the same conditions asin Fig. 11, with only the holding time at the liquidustemperature extended to one hour, to predict the extent of Irdiffusion into the molten Ni-10at%Al alloy. The result isshown in Fig. 15. According to the result, the thickness ofthe diffusion layer in the sample was about 200 µm. Thisvalue is considered to be sufficiently thin for Ir to localize inthe Ni-based alloy and act as a surface modification layer.Therefore, it is expected that Ir elements can be cast-coatedin the unidirectional solidification process even as a purestate, while the pure Pt is currently impractical but variousconsiderations are being made based on the results of thepresent study.4. SummaryIn this study, the feasibility of a process for surfacemodification of Ni-based alloys simultaneously with castingby cast-coating with high melting point rare metals wasinvestigated using a combination of casting model experi-ments and numerical calculations. Pt, which is already inpractical use as an oxidation-resistant bond coat for Ni-basedsuperalloys, and Re and Ir, which have particularly highmelting points, were selected as candidate elements for cast-coating.Fig. 14 Vertical and isothermal (1450°C) section of calculated Ni-Al-Re ternary phase diagram.Fig. 15 Calculated diffusion profile in vicinity of Ir/Ni-10at%Al interface.The system temperature is held at the liquidus for 1 hour. The calculateddiffusion length is acceptable as a surface modification formed in a cast-coating process.M. Ode, H. Esaka, A. Ishida, S. Takamori and H. Murakami550Casting model experiments:Small specimens of Pt, Re and Ir pure materials wereplaced inside a mold designed for unidirectional solid-ification, and molten Ni-10at%Al alloy was poured into themold to measure temperature and observe the concentrationat the interface between each elemental sample and the Nialloy. The small specimens were engulfed in solidapproximately 100–300 s after casting; Pt was completelydissolved except for some parts, Ir formed an interdiffusionlayer of 20–30 µm, and Re was hardly soluble in Ni alloyside.Numerical calculations:The sample temperature was predicted from the solid-ification heat transfer analysis (enthalpy method) using thetemperature plateau time recorded by thermocouplesinstalled inside the mold as an indicator. The heat transferparameters required for the prediction were determined tominimize the error between the actual measurements andthe calculations. The obtained thermal conditions wereapplied to the diffusion simulation, and the experimentalconcentration distribution at the interface between the Reand Ir specimens and the Ni-10at% alloy showed goodagreement with the calculation results. This shows thatdiffusion simulation can accurately predict the applicabilityof cast-coating. Furthermore, Ir showed high potential ascast-coating even under turbine blade manufacturingconditions where the molten metal is held for long periodsof time.AcknowledgementsThis work was supported by Innovative Science andTechnology Initiative for Security Grant Number JPJ004596,ATLA, Japan.REFERENCES1) T. Okamoto and K. Nose: The Latest Technological Trend of RareMetals, (CMC Publishing Co., Ltd., Tokyo, 2012) p. 170.2) H. Nakae: Engineering Casting, (Sangyo Tosho Publishing Co., Ltd.,Tokyo, 1995) p. 178.3) T. Noguchi and S. Kamota: J. JFS 70 (1998) 920–927.4) T. Watanabe: Imono 55 (1983) 381.5) S. Aso, M. Nakanishi, S. Goto, H. Ike and Y. Shobuzawa: J. JFS 73(2001) 155–160.6) K. Okada, S. Goto, S. Aso and Y. Komatsu: J. JFS 74 (2002) 497–505.7) Y. Tsunekawa: J. JFS 76 (2004) 487–493.8) https://computherm.com/docs/thermodynamic_manual.pdf.9) C. Zhang, J. Zhu, A. Bengtson, D. Morgan, F. Zhang, W.-S. Cao andY.A. Chang: Acta Mater. 56 (2008) 2576–2584.10) Metals Handbook, 3rd ed., (Maruzen publishing, Tokyo, 2000) p. 13.11) K. Gunji: Tetsu-to-Hagané 80 (1994) N266–N280.12) I. Ohnaka: Introduction to Computer Heat Transfer and SolidificationAnalysis, (Maruzen publishing, Tokyo, 2021) p. 336.13) C. Neubauer, D. Mari and D. Dunand: Scr. Metall. Mater. 31 (1994)99–104.Observation and Numerical Prediction of Concentration Distribution at Cast Coating Interface of Solid Pt, Ir, Re 551https://doi.org/10.11279/jfes.70.920https://doi.org/10.11279/jfes.73.155https://doi.org/10.11279/jfes.73.155https://doi.org/10.11279/jfes.74.497https://doi.org/10.11279/jfes.76.487https://computherm.com/docs/thermodynamic_manual.pdfhttps://computherm.com/docs/thermodynamic_manual.pdfhttps://computherm.com/docs/thermodynamic_manual.pdfhttps://doi.org/10.1016/j.actamat.2008.01.056https://doi.org/10.2355/tetsutohagane1955.80.6_N266https://doi.org/10.1016/0956-716X(94)90102-3https://doi.org/10.1016/0956-716X(94)90102-3