Development of a new high entropy alloy for wear resistance: FeCoCrNiW0.3 and
FeCoCrNiW0.3 + 5 at. % of C.
Marco Gabriele Polet1 , Gianluca Fiore1 , Flavia Gili2 , Davide Mangherini2 , Livio Battezzat1.
1 Dipartimento di Chimica, Università di Torino, Via P. Giuria 7 10125 Torino. Italy 2 GML – Metals, C.R.F. S.C.p.A, Corso Settembrini 40, Torino – Italy
A new high entropy alloy (HEA) to has been synthesized using a predictive method recently developed. The alloy performance is compared to that of a benchmark commercial alloys (Stellite®6) as wear resistance material for coating automotive engine valves.
Using step by step predictive parameters the occurrence of an fcc solid solution with entropy of mixing higher than the conventional limit of 1.5 R was singled out and verified in the FeCoCrNiW0.3 composition.
The as-cast ingot displays W segregation within the fcc grains as revealed by XRD patterns which were fitted by assuming the arc melted alloy contains two solid solutions. The segregation is eliminated by annealing at 1200°C for 3 hours. In these sample a (Co-Fe)7W6-type phase is found embedded in the solid solution
matrix. Slip step patterns around indentation marks suggest low stacking fault energy of FeCoCrNiW0.3. The
HEA has been hardened by adding 5 % at. C to form carbides inside the ductile matrix. Conventional hardness, scratch and oxidation resistance tests show that the alloy compares well with Co-based Stellite®6 being the content of the expensive and strategical elements halved with respect to it.
Keywords: High Entropy Alloy; XRD; Scratch, Hardness; dispersion strengthening. Highlights
FeCoCrNiW0.3 high entropy alloy predicted and synthetized.
Slip step patterns around indentation marks indicate alloy ductility.
Addition of 5 at. % of C to FeCoCrNiW0.3 produces hard carbides in HEA matrix.
Good resistance to oxidation in 700-900 °C range
Hardness and scratch hardness compare well with conventional more expensive alloys.
Multicomponent solid solutions containing elements in nearly equiatomic amount, called High Entropy Alloys (HEAs), have attracted increasing interest during the last ten years both for fundamental studies of the inner part of multicomponent phase diagrams and because some alloys display outstanding properties, strictly related to the complexity of such solutions [1].
Since several elements are needed for alloy formulation, predictive models have been developed to guide the synthesis of new multicomponent compositions [2;3;4;5;6]. These should also account for requirements posed by the performance necessary for industrial applications with respect to current commercial alloys. The aim of this work is to design a new HEA composition to be tested as wear resistance coating in the automotive field, e. g. for the seat of engine valves, using a predictive method previously put forward and to compare its performance to those of benchmark alloys (Co-based superalloys, Stellite® 6 (a registered trademark of Kennametal Stellite ), Eatonite® 6 or Pyromet® 31b (a registered trademark of Crs Holdings Inc.) [7]
[
8]
)mettiamo solo la stellite dato che le altre non sono state usate?. Co-based alloys are widely used as wear resistance materials for their good resistance to oxidation, galling and adhesive wear [9;10]. Specifically, the Stellite® alloys family retain at temperature lower than 417 °C the high temperature fcc phase of Co having a low staking fault energy [10] which is thought to convey good resistance to galling to these materials since it favors a stress induced martensitic transformation during high load contact wear [11]. Their Chromium content of about 30% in weight confers the oxidation resistance. W or Mo and Cr are employed also to form hard carbides.The design strategy pursued in this work has been to improve the mechanical properties (in particular the hardness) of the quaternary FeCoCrNi equimolar composition by adding W while retaining the fcc solid solution. Experimental evidence of the formation of a disordered solid solution in the FeCoCrNi system with absence of long range order was provided earlier in various papers [12, 13, 14] by means of conventional XRD and SEM analyses. This has been questioned on the basis of further XRD measurements, suggesting that the alloy is constituted by two solid solutions with very close lattice parameter. The lack of microstructural analysis did not allow a firm conclusion on the alloy constitution [15]. Irrespective of this uncertainty, the amount of W to be added to the equimolar FeCoCrNi has been determined using the maps developed in previous work which reported a method to identify the occurrence of compact solid solutions in multicomponent alloys by combining parameters obtained from Hume-Rothery rules and the regular solution model [6]. The maps highlight the regions in the space of electronegativity-radius mismatch, valence electrons-itinerant electrons, critical solubility temperature-radius mismatch parameters where single-phase HEAs are expected. In order to be inside the region where the formation of a solid solution is likely, the higher W content allowed in the alloy is about 7% atomic, i. e. with a ratio of 0.3 with respect to the molar quantity of Fe, Co, Cr, and Ni. In fact, adding W to the quaternary system implies increasing the overall radius and electronegativity mismatch causing the solid solution to be less stable and favoring segregation and the occurrence of intermetallic compounds.
Having tested the stability of the fcc phase in such composition, an amount of C is added to harden the material by forming carbides similarly to commercial wear resistant materials [16] while the matrix should remain a high entropy phase. Since in HEAs there is no principal element, it is envisaged that the properties arising from complex interaction between several constituents sharing the same lattice would give rise to improved mechanical behavior [17].
Some authors have tested the wear resistance properties of HEAs containing four of the elements employed in the present work. The role of Al in AlxCoCrCuFeNi [18] was ascertained showing that as long as the alloy is
fcc, i. e. low Al content, it is ductile and heavily grooved and delaminated. Abrasive wear was found in bcc AlCoCrFeMo0.5Ni [19]. The delamination wear was confirmed in AlxCo1.5CrFeNi1.5Tix having low Al and Ti
resistance was shown for CuCoNiCrAlFeBx alloys [21]. The role of Ni content was examined in
Al2CrFeCoCuTiNix [22] showing that the maximum wear esistance is achieved at the equiatomic
composition with respect to the other transition elements. It is apparent that an fcc matrix should be added of hard particles to combine ductility and hardness.
The comparison of alloys performance is made by using scratch testing as fast technique to evaluate wear response.
The design concept of mixing several elements in nearly equimolar ratio allows also to decrease the amount of costly elements whose supply chain is considered strategical: in the HEA developed in this work the content of Cobalt has been halved with respect Co-based wear resistance superalloys.
Experimental
Fe, Co, Cr, Ni, W with purity higher than 98 % at. were weighted in the proper amount to produce 20 g of the desired composition, FeCoCrNiW0.3, by arc melting in a Ti-gettered Argon atmosphere. The ingot was
melted 5 times. The concentration of elements in the ingot was checked by EDS and the ingot re-melted three more times to ensure full dissolution of high-melting elements and to reach homogeneity. The product extracted from the furnace was shining. The EDS analysis confirms that the as cast FeCoCrNiW0.3
composition reproduces the nominal one within the experimental error.
Another batch of about 60 g was prepared with the same procedure by adding 5 % at. C (about 1 % in weight) to the Fe, Co, Cr, Ni and W elements to produce an alloy labelled as FeCoCrNiW0.3 + 5 C at %. The
ingots were cut vertically, grinded and polished to perform XRD diffraction studies in Bragg-Brentano mode employing Cu k wavelength. The lattice constants of phases were determined with the cos2/sen
extrapolation method. The microstructure of the polished cross section was analyzed using a Leica Scanning Electron Microscope equipped with a EDS analyzer. Alloy portions of both compositions having mass of about 200 mg were analyzed in a Setaram High Temperature-Differential Scanning Calorimeter (HT-DSC) up to the highest operative temperature of 1550°C under Helium flux with a scan rate of 10 °C/min. The FeCoCrNiW0.3 HT-DSC sample was then polished in order to collect XRD patterns and SEM images.
A slice cut from the FeCoCrNiW0.3 as cast ingot has been annealed under helium flux at 1200 °C for 3 h using
the HT-DSC furnace reaching the annealing temperature with a scan rate of 10 °C/min. The same rate was used for cooling to room temperature from 1200°C after the anneal. The sample was than polished to perform XRD and SEM analyses.
The Vickers hardness of all samples was measured using a conventional indenter with load of 300 gF ( 2.942 N) for 15 s as well as at a fine scale for the FeCoCrNiW0.3 as cast sample using loads of 10 mN, 200 mN and
2000 mN with a Fischer Nanoindenter averaging the values measured with different loads.
The resistance to oxidation at high temperature of the FeCoCrNiW0.3 + 5 C at % composition and commercial
Stellite® 6 was studied by performing thermogravimetric (TGA) experiments using a Thermal Instrument TGA Q600 equipped with alumina crucibles at 700 °C and 900 °C in air for 4 hours. The specimens was cut approximately to 10 mm x 10 mm x 10 mm and ultrasonically cleaned in ethanol. Heating was performed with scan rate of 20 °C/min from RT to 700 °C or 900 °C under N2 atmosphere.
Scratch tests were performed, using a CETR-Bruker UMT Scratch Tester mounted with a Rockwell tip, on flat polished samples of as cast FeCoCrNiW0.3 and FeCoCrNiW0.3 + 5 % at. C alloys and on the benchmark
Samples of either Stellite® 6 or FeCoCrNiW0.3 + 5 % at. C of about 5 mm in height were used to coat by Tungsten Inert Gas Welding (TIG) a stainless steel AISI 304 support having size of 6 cm x 2,5 cm x 1 cm. The coating surface was than polished and five parallel scratches were made at room temperature at a distance of 1 mm for Scratch Hardness according to ASTM G 171 [23]. The ASTM G 171 [23] standard suggests to select the load avoiding to cause severe damage to the test surface which would imply the scratch width is not clearly identifiable, and the test is invalidated. Also in case of a too fine scratch the measurement of the width can be difficult. For this reason the scratches on the FeCoCrNiW0.3 + 5 % at. C and Stellite® 6 welded samples were made with a load of 20 N, sample speed of 10.0 mm/min producing a scratch of 5.0 mm in length. The scratch width and the profile depth have been measured using an optical microscope and a Veeco Dektat 150 Surface profiler. The scratch hardness number were evaluated from the width of the scratch applying the formula Hs= k·P /(π· w2) where Hs [P], P [N], and w [m] refer to scratch hardness number,
normal load, and scratch width, respectively, k is a geometrical constant (k = 24,98).
Results
The XRD pattern of the as cast FeCoCrNiW0.3 (Fig. 1a) taken on the ingot cross section contains reflections
compatible with an fcc solid solution. Close inspection of their position and asymmetric shape reveals that the peaks occur at slightly varied angles with respect to where they would be expected, should the structure of the phase be compact cubic. The origin of this discrepancy is understood when considering the SEM image in Fig. 2a taken in backscattering mode and the EDS analyses in Tab. 1. They reveal that most of the alloy is constituted by a phase (grey contrast) with composition close to the nominal one. A brighter thin layer, richer in W, decorates the border zones around the primary grains. There is no sharp boundary separating the grey and bright zones. A few bright small particles due to a phase rich in W are embedded in the bright zone. All these evidences suggest that the compact phase constituting the alloy displays segregation of W in the regions in between each grain. This occurred during the relatively rapid solidification of the alloy in contact with the cold hearth of the arc melter. However, the segregation caused the formation of only a small volume fraction of a compound rich in W (see below for its identification) whereas the majority of the excess W remained in the compact phase. A gradient in composition and, consequently, in the lattice constant is expected due to the enrichment in W, the largest atoms among the elements in the alloy. To check this, although approximately, a Rietveld refinement was performed with MAUD [24] by assuming the occurrence of two fcc phases which provided a reasonable fit to the overall pattern. The two corresponding lattice constants are 3.6048 0.0002 Å and 3.625 0,003 Å for the grey matrix and the region enriched in W at its border, respectively.
HT-DSC results of two consecutive scans on heating and cooling made with the same sample of cast alloy (Fig. 3a) show no appreciable signal before reaching the melting point at Tm= 1412 4 °C. The melting range
is narrow since the liquidus temperature taken at the peak of the melting signal is 1440 2 °C. The alloy displays tendency to undercool in the calorimeter cell where the solidification occurs around 1330 °C in both the first and the second cycle. The absence of apparent thermal events in the whole temperature range from room temperature to the melting point suggests that the phases obtained by casting are equilibrium ones. However, EDS-SEM analyses of the sample which underwent two thermal cycles in HTDSC, showed that the matrix has composition close to the alloy nominal one, but the Z contrast is not sharp, in particular around the bright W-rich particles. This indicates that the segregation occurring at the undercooling experienced in HT-DSC (150 K) is not recovered during the programmed cooling at the constant rate of 10 °C/min. Actually, the simulation of the experimental XRD pattern made after HT-DSC
cycling gives reasonable fit by admitting two solid solutions with lattice constants 3.6065 0.0002 Å and 3.619 0,003 Å.
In order to achieve homogenization, another sample of the as cast FeCoCrNiW0.3 was annealed in the
HT-DSC cell at 1200°C for 3 hours and cooled with the scan rate of 10°C/min down to room temperature. The SEM image in Fig. 2b showing the final microstructure indicates that in this case homogenization took place. The single phase matrix has lattice constant of 3.6062 0.0004 Å and composition very close to the nominal one without detectable gradient around the bright precipitates. These increased in volume fraction with respect to both the as cast and HT-DSC cycled samples. Reflections belonging to this phase are clearly discernible (see XRD diffraction pattern of Fig. 1b) in addition to the reflections due to the matrix. They have been indexed as due to a trigonal intermetallic compound having the space group R-3m of Co-Fe)7W6. The addition of 5 % at. C to FeCoCrNiW0.3 leads to the microstructure seen in the backscattered electrons
images of Figs. 2c-e. The dendrites have composition close to the alloy nominal one (gray contrast and EDS analysis in Tab. 1)- In the interdendritic region three phases are recognized: a very bright one, richer in W, dark lamellae, richer in Cr, and gray lamellae (same contrast as the matrix) (Fig. 2e-f and Tab. 1). The XRD pattern of the FeCoCrNiW0.3 + 5 at. % C sample (Fig. 1c) has been indexed by assigning reflection to an fcc
solid solution and two carbides, W2C type and Cr3C2 type. The lattice constant of the fcc solid solution is afcc=
3.598±0.002 Å, comparable to that of FeCoCrNiW0.3 samples, in agreement with EDS analyses.
Thermal analysis of a FeCoCrNiW0.3 + 5 at. % C as cast sample (Fig. 3b) did not reveal any signal during the
heating scan up to the start of melting at 1305 °C which proceeds in three stages ending around 1480 °C (inset in Fig. 3b). The melting range of about 170° C is wider than that of the alloy not containing C. The solidification signal starts at around 1350 °C in the HT-DSC cooling run and is composed again of three peaks in a range of 90 K with undercooling of 130 K. No other thermal event has been detected during cooling either.By observing the microstructure of the sample, it is deduced that the solid solution is the primary solidifying phase followed by the formation of a mixture of phases in the interdendritic region either via peritectic or eutectic reactions.
The Vickers hardness (Fig. 5a) of the as cast FeCoCrNiW0.3 alloy is 198 ± 4 HV. After annealing at 1200°C for 3
h it increases to 269 ± 9 HV because of the increase in amount of intermetallic phase. The addition of C increases the hardness further to 377 ± 11 HV due to the formation of hard carbides. Slip step patterns have been found around indentation marks both in as cast FeCoCrNiW0.3 (Fig. 4a) and FeCoCrNiW0.3 + 5 at. %
C (Fig. 4b) samples as shown in Fig. 4. The hardness of the benchmark Stellite 6 was determined as 410 ± 5 HV consistent with the technical data provided by the supplier [25] riferimento.
The effective elastic modulus, Eef., was obtained by instrumented indentation from the unloading curves of all the measurements made with 10 mN and 200 mN loads. The alloy elastic modulus, Eexp., was deduced from Eexp.= Eef. (1-2avarage) riferimento where is the Poisson ratio of the phase which was estimated by averaging those of the constituents. The Eexp. is 211 ± 6 GPa. The elastic modulus computed using the rule of mixtures of the pure elements using the concentrations in Tab. 1 is comparable, 229 GPa. The hardness of the cast FeCoCrNiW0.3 sample obtained in the same experiments depends on load: 222 ± 9 HV with the load
of 2000 mN, 235 ± 9 HV with load of 200m N, and 358 ± 14 HV with load of 10 mN. The rise in hardness is attributed to work hardening performed by polishing the sample ().
Samples of both Stellite 6 and FeCoCrNiW0.3 + 5 at. % C were TIG welded to a stainless steel support to
simulate the envisaged application process. The microstructure of the welded samples reproduced that of the as cast material being slightly finer. The hardness of the welded coating of Stellite 6 and FeCoCrNiW0.3 +
5 at. % C were equal to those obtained on the as cast sample: 415 ± 7 HV for Stellite and 371 ± 5 HV for FeCoCrNiW0.3 + 5 at. % C.
Scratch hardness values have been obtained on a polished as cast sample of FeCoCrNiW0.3, FeCoCrNiW0.3 + 5
at. % C and Stellite® 6 as cast samples, and on the TIG welded coating for FeCoCrNiW0.3 + 5 at. % C and
Stellite® 6 (Fig. 5b). The values provided by as cast and coated sample match fully. In Fig. 5c the profiles of the five scratches for the FeCoCrNiW0.3 + 5 at. % C and Stellite coatings are shown. In Fig 5d an
enlargement of the profile of the central scratches are reported for detailed comparison.
The results show that the resistance to abrasion of FeCoCrNiW0.3 is much lower with respect to FeCoCrNiW0.3
+ 5 at. % C and the commercial Stellite® 6 alloy. It is interesting to notice that FeCoCrNiW0.3 + 5 at. % C has
lower Vickers hardness with respect to Stellite® 6 as reported above, but its scratch hardness is 4.5 ± 0.3 GPa fully comparable to that of the Stellite® 6 (4.3 ± 0.3 GPa). In fact, the profile and depth of scratches on Stellite 6 and FeCoCrNiW0.3 + 5 at. % C welded coatings are almost coincident as shown in Fig. 5c-d., also
accounting for higher load applied during scratch testing (20N) with respect micro-indentation (2.9420 N), there is an inversion in hardness values between Stellite® 6 and FeCoCrNiW0.3 + 5 at. % C alloy during static
(micro indentation) or dynamic (scratch test) application of load:
The resistance to oxidation at high temperature of the FeCoCrNiW0.3 + 5 at. % C alloy was evaluated by
measuring the weight gain, normalized with respect to surface area of the samples, in TGA experiments at 700 °C and 900 °C in air for 4 hours. The same experiments have been performed with the Stellite® 6 benchmark alloy. As shown in Fig. 6 the curve for weight gain of both Stellite® 6 and FeCoCrNiW0.3 + 5% at.
C has a decreasing slope during the isothermal heating at 700 °C, indicative of oxidation of samples via a diffusion-controlled process. Then it reaches a plateau suggesting the formation of an oxide layer able to prevent further oxidation. In the TGA experiment at 900 °C with FeCoCrNiW0.3 + 5 at. % C the weight gain per unit area, w, as a function of time, t, is fast during the first hour with seemingly parabolic trend and can be expressed as: w2= k * t [s] - C with k = 5,10*e-10 C=2,29 e- 7). mettiamo la parabola nella figura? At the same temperature the weight gain of Stellite® 6 reaches a plateau but then starts again to increase after 2.5 h with a rate comparable to that of FeCoCrNiW0.3 + 5 at. % C. TGA experiments highlights a comparable
oxidation resistance at high temperature of the new HEA FeCoCrNiW0.3 + 5 at. % C and the benchmark
Stellite® 6 alloys.
Discussion
The XRD and SEM results (Fig. 1 and 2a-b) suggest that the FeCoCrNiW0.3 alloy is very close to be a single fcc solid solution. Although segregation of W occurs in the as cast condition and even after controlled cooling in HT-DSC, homogeneity is achieved in the sample after annealing for 3 h at 1200°C. A minority secondary phase made of fine crystals, rich in W, embedded into the matrix, was identified as isostructural with the trigonal R-3m (Co-Fe)7W6 type intermetallic compound.. The method for alloy design is apparently successful with limited scatter. Should only one phase be desired at equilibrium in the high entropy alloy, the amount of W should be broght down to about 5.5 at.%. This, however, is not needed for the purpose of this work since the excess W reacts with C to give a strengthening carbide.
The lattice parameters found for the fcc phases (Fig. 1) are in agreement with the presence of W, the element with the largest atomic radius among the constituents of the alloy. The lattice constant of the base quaternary fcc FeCoCrNi solid solution has been reported as 3.561 Å [14], 3.575 Å in [26] and 3.5643 Å in [27] to be compared with afcc= 3.6048 ± 0.0002 Å for the matrix of the present FeCoCrNiW0.3 as cast sample. In order to check if the solid solution is ideal, i. e. complies with Vegard’s law, hypothetical fcc lattice parameters of the bcc metals Cr and W were obtained by extrapolating those of the fcc solid solutions in the binary Ni-Cr and W-Ni systems [28], respectively. Using the EDS concentrations of the matrix inTab. 1 it has
been found that the computed lattice constant is an agreement with the experimental one within the error of both experimental and computed values.
In the literature the effect of the addition of 7% at. of a fifth element to the quaternary equimolar alloy FeCoCrNi, as for W in this work, i. e. 0.3 times the content of the other elements, has been studied in the case of Mo [29], Ti [14] and Al [30]. In the as cast condition the alloy with 7 % at. Mo has dendritic microstructure composed of two fcc solid solutions and a low amount of a Cr-Mo phase. Considering the maps discussed in [6], the addition of 7 % at. Mo to FeCoCrNi would have implied the formation of a solid solution because of favorable radius and electronegativity mismatches. The alloy with Ti, having larger atomic radius than W and lower electronegativity and hence higher electronegativity and radius mismatch, is composed of a high entropy matrix with lower content of the fifth element (Ti) (3.5 % at) than in the case of W (5.6 % at.). The matrix is surrounded by a brighter region but secondary phases (a Ni-Cr rich and a likely phase) The Al addition, in accordance with the parameters for the occurrence of solid solutions, leads to the formation of nano-precipitates of a fcc solid solution, with L12 order [30].
The addition of 7 at. % W to FeCoCrNi increases the hardness of the alloy from 135 HV [14] to 198 HV due to the lattice distortion introduced in the solid solution by the larger W atoms. The straight slip bands around indentation marks (Fig. 4a-d) are a consequence of the cooperative movements of dislocations [36] reaching the surface of the sample. They were found also in another HEA constituted by an fcc solid solution of composition Al0.5CrCuFeNi2 [35], and indicate low probability of cross slip of dislocations and preferential
formation of stacking faults [32,33].
The tendency of a material to give rise to planar slip instead of cross-slip is related to low stacking fault energy (SFE) [33] as well as to solute content and atomic size misfit because of the frictional stress imposed by the solute atom on the joining of partial dislocation [37]. Actually, stacking faults have been observed by TEM in FeCoCrNiMn after small plastic deformation [38, 16]. Moreover, the SFE in FeCoCrNi and in equiatomic and non equiatomic FeCoCrNiMn alloys was estimated by coupling XRD measurements and ab-initio calculations [26], revealing that it depends on the elemental concentration, i. e. SFE decreases on decreasing the Ni and increasing the Cr content. It can be hypothesized that in the FeCoCrNiW0.3
composition the W should decrease the SFE favouring the formation of straight slip patterns ( Fig. 4). The apparent low stacking fault energy is expected to confer resistance to galling to the material [34].
In order to increase the hardness to confer wear resistance to the alloy, 5 at. % of C has been added to the FeCoCrNiW0.3 to form of Cr and W carbides but retaining the chance that the fcc FeCoCrNiW0.3 matrix still presented good resistance to galling. The amount of C has been devised by considering the content of C present in commercial alloys (about 1 % in weight). The combination of both Cr and W should also leave enough Cr in the matrix to retain the oxidation resistance of the alloy.
The effect of C addition in low quantity (0.5 % at.) to high entropy alloys has been studied earlier to improve cryogenic mechanical properties of FeCoCrNiMn revealing an increased tendency to the formation of deformation twins at the nano scale [39]. The addition of 2 at. % C to the AlCoCrFeNi composition has been experimented in [40] obtaining an fcc solid solution and (Cr,Fe,)7C3 carbide in the as cast condition and
(Cr,Fe,)23C6 after heat treatment at 1000 °C. The addition of 5 at. % of C to the fcc high entropy FeCoCrNiW0.3
solid solution performed in this work leads to an as cast microstructure where the matrix is still composed of a high entropy phase accompanied by two complex carbides either rich in Cr or W, as in the case of Co-based superalloys. The hardness is the increased of about 180 HV up to 377 HV, comparable to that of Stellite® 6. The matrix remains ductile as shown by the occurrence of straight slip lines around indents ( Fig. 4f).
HT-DSC measurements have shown that the solidification range of FeCoCrNiW0.3 + 5% at. C is rather narrow,
actually comparable with that of commercial Stellite® 6, which represents a favorable condition for casting and welding processes. Moreover FeCoCrNiW0.3 + 5% at. C has oxidation resistance comparable to Stellite® 6
(Fig. 6), although with lower Cr content, i. e. 18% in weight with respect to about 30 % , in the operative temperatures around 700°C at which Stellite® 6 is used as a coating for engine valves.
The microstructure composed of a high entropy matrix and of hard carbides gives positive response to scratch testing indicative of favorable wear resistance of the material. Finally, it must be emphasized that the amount of costly and critical materials in the new alloy, Co and W together, is 1/3 lower with respect to Stellite® 6.
Conclusions
The aim of this work was to produce a new multicomponent High Entropy Alloy to be tested as wear resistance material. Using the criteria previously developed the prediction of stable HEAs, the maximum content of W to be added to the quaternary FeCoNiCr system has been defined at around 7 at. %. The new FeCoCrNiW0.3 alloy is almost entirely constituted by an fcc phase once equilibrium is reached. A minor
fraction of a Fe7W6 type intermetallic compound accounts for the excess 1% W not soluble in the main
phase. On the contrary, the as cast alloys is supersaturated in W and displays a gradient in W content in the main high entropy phase due to segregation during dendrite growth.
The solid solution constituting the alloy possesses low stacking fault energy as suggested by the occurrence of slip step patterns around Vickers indentation marks. On the other hand, the hardness is low for a material to be used for wear resistance. To circumvent this, 5 at. % C has been added to the high entropy phase to obtain Cr- and W- based hard carbides embedded into the fcc solid solution.
The properties of the alloy designed FeCoCrNiW0.3 + 5 % at. C have been shown to be at least as good as those of wear resistant Co-based superalloys in terms of hardness, scratch hardness test and oxidation resistance.
These results encourage further work on composition refinement and further testing in view of promoting the new alloy for industrial use. A clear advantage would be that the amount of costly and critical materials, Co and W, is all together 1/3 lower with respect to the benchmark material, Stellite® 6.
Acknowledgments
This work has been performed in the frame of the EU-7FP Accelerated Metallurgy Project (ACCMET, contract NMP4-LA-2011-263206).
This work has been performed in the frame of the EU-7FP Accelerated Metallurgy Project (ACCMET, contract NMP4-LA-2011-263206).
Figure 1. XRD diffraction patterns (Bragg-Brentano method, Cu-source): as cast FeCoCrNiW0.3 (a);
Figure 2. SEM backscattered electron mode images: as cast FeCoCrNiW0.3 (a); FeCoCrNiW0.3 sample after
Figure 3. HT-DSC cycling performed at 10°C/min. up to 1550° C: as cast FeCoCrNiW0.3 (a); as cast
Figure 4. SEM backscattered electron mode images of the indentation marks after Vickers hardness
measurements (performed using a load of 2000 mN) on the as cast FeCoCrNiW0.3 sample (a) and on the
Figure 5. a) Hardness of the FeCoCrNiW0.3 and FeCoCrNiW0.3 + 5 at. % C samples obtained using a load of
300gF (2.942 N) for 15s with a Vickers indenter. Scratch test performed on polished welded coating
sample for Stellite 6 and FeCoCrNiW0.3 + 5 at. % C alloys, on as cast sample in the case of FeCoCrNiW0.3 ,
both of them were obtained with a load of 20N and using a Rockwell indenter (b), scratch profiles
obtained with Veeco Dektat 150 Surface profiler for the Stellite 6 and FeCoCrNiW0.3 + 5 at. % C coating
Figure 6. Weight gain per unit area versus time for the FeCoCrNiW0.3 + 5 at. % C alloy (continuous line)
and for commercial Stellite® 6 (dotted lines) obtained with a TGA TA Q600 instrument during isothermal annealing at 700 and 900°C for 4 h under air flux.
EDS analysis FeCoCrNiW0.3 as cast
EDS analysis FeCoCrNiW0.3 after
1200°C 3 h
Element
Matrix
Bright
precipitate
Matrix
Bright precipitate
at % st. dev. at % st. dev. at % st. dev. at % st. dev.
Cr 22,4 0,3 20 2 23 1 20 2
Fe 24,0 0,7 18 3 24 1 16 1
Co 24,2 0,7 20 2 23,1 0,9 19 2
Ni 23,8 0,3 13 5 24 1 10 1
W 5,6 0,1 30 10 5,7 0,3 35 1
EDS analysis FeCoCrNiW0.3 + 5 % at. C
Matrix
Carbides
Element
Dark
Light
at % st. dev. at % St. dev. at % st. dev.
Cr 19 1 49 10 27,5 0,5 Fe 24,8 0,4 18 2,8 13 1 Co 24 1 15 4,5 15 1 Ni 26 1 12 4 10 1 W 5,7 0,7 5 0,9 34 3 C - - - -
-Tab 1 . EDS analysis results for the microstructures presented in Fig. 2.
[1] Cantor, B.; Chang, I.; Knight, P.; Vincent, A., Microstructural development in equiatomic multicomponent alloys. Mat Sci Eng A 375 (2004) 213-218.
[2] I. Toda-Caraballo, P.E.J. Rivera-Díaz-del-Castillo. A criterion for the formation of high entropy alloys based on lattice distortion. Intermetallics 71 (2016) 76-87
[3] Zhang, Y.; Zhou, Y. J.; Lin, J. P.; Chen, G. L.; Liaw, P. K., Solid solution phase formation rules for multi‐ ‐ component alloys. Adv Eng Mat 10 (2008) (6), 534-538.
[4] King, D. J. M.; Middleburgh, S. C.; McGregor, A. G.; Cortie, M. B., Predicting the formation and stability of single phase high-entropy alloys. Acta Mater 104 (2016) 172-179.
[5] Ye, Y. F.; Wang, Q.; Lu, J.; Liu, C. T.; Yang, Y., Design of high entropy alloys: A single-parameter thermodynamic rule. Scripta Mater 104 (2015) 53-55.
[6] M.G. Poletti, L. Battezzati, Electronic and thermodynamic criteria for the occurrence of High Entropy Alloys in metallic systems Acta Mater. 75 (2014) 297-306.
[7] R. Zhao , G. C. Barber , Y. S. Wang & J. E. Larson. Wear Mechanism Analysis of Engine Exhaust Valve Seats with a Laboratory Simulator, Tribol T 40 2 (1997) 209-218.
[8] http://www.eaton.in/EatonIN/AutoExpo2014/HollowValves/index.htm
[9] D.H.E. Persson, S. Jacobson, S. Hogmark. Effect of temperature on friction and galling of laser processed Norem 02 and Stellite 21. Wear 255 (2003) 498–503.
[10] K.C. Antony. Wear-Resistant Cobalt-Base Alloys. JOM(1983) 52-60.
[11] R.T. Smith, T. Lolla, D. Gandy, L. Wu, G. Faria, A.J. Ramirez, S.S. Babu and P.M. Anderson. In situ X-ray diffraction analysis of strain-induced transformations in Fe- and Co-base hardfacing alloys. Scripta Mater 98 (2015) 60–63.
[12] M. S. Lucas, G. B. Wilks, L. Mauger, J. A. Muñoz, O. N. Senkov, E. Michel, J. Horwath, S. L. Semiatin, M. B. Stone, D. L. Abernathy, and E. Karapetrova. Absence of long-range chemical ordering in equimolar FeCoCrNi. Appl Phys Lett 100 (2012) 251907.
[13] H.P. Chou, Y.S. Chang, S.K. Chen, J.W. Yeh. Microstructure, thermophysical and electrical properties in AlxCoCrFeNi (0≤x≤2) high-entropy alloys. Mat Sci Eng B 163 (2009) 184–189.
[14] T.T. Shun, L.-Y. Chang, M. H. Shiu. Microstructures and mechanical properties of multiprincipal component CoCrFeNiTix alloys. Mater Sc Eng A 556 (2012) 170–174.
[15] U. Dahlborg , J. Cornide, M. Calvo-Dahlborg, T.C. Hansen, A. Fitch, Z. Leong, S. Chambreland , R.
Goodall. Structure of some CoCrFeNi and CoCrFeNiPd multicomponent HEA alloys by diffraction techniques. J Alloy Compd 681 (2016) 330-341.
[16] H. Ocken. The galling wear resistance of new iron-base hardfacing alloys: a comparison with established cobalt- and nickel-base alloys. Surface and Coatings Technology 76-77 (1995) 456-461.
[17] ZiJiao Zhang, M.M. Mao, Jiangwei Wang, Bernd Gludovatz, Ze Zhang, Scott X. Mao, Easo P. George. Nanoscale origins of the damage tolerance of the high-entropy alloy CrMnFeCoNi. Nature Comm 6 (2015) 10143.
[18] J.M. Wu, S.J. Lin, J.W. Yeh , S.K.Chen, Y.S. Huang, H.C. Chen. Adhesive wear behavior of AlxCoCrCuFeNi
high-entropy alloys as a function of aluminum content. Wear 261 (2006) 513–519
[19] C.Y. Hsu, T.S. Sheu, J.W. Yeh, S.K.Chen. Effect of iron content on wear behavior of AlCoCrFexMo0.5Ni
high-entropy alloys. Wear 268 (2010) 653–659.
[20] M.H.Chuang, M.H. Tsai, W.R.Wang, S.J. Lin, J.W.Yeh. Microstructure and wear behavior of AlxCo1.5CrFeNi1.5Tiy high-entropy alloys. Acta Mater 59 (2011) 6308–6317
[21] C.Y. Hsu, J.W. Yeh, S.K. Chen, T.T. Shun. Wear Resistance and High-Temperature Compression Strength of Fcc CuCoNiCrAl0.5Fe Alloy with Boron Addition. Metall Mater TransA 35A (2004) 1465-1469.
[22] X.W. Qiu, C.G. Liu. Microstructure and properties of Al2CrFeCoCuTiNix high-entropy alloys prepared by
laser cladding. J Alloy Compd 553 (2013) 216–220.
[23] ASTM G171-03(2009)e2, Standard Test Method for Scratch Hardness of Materials Using a Diamond Stylus, ASTM International, West Conshohocken, PA, 2009, www.astm.org
[24] L. Lutterotti, Total pattern fitting for the combined size-strain-stress-texture determination in thin film diffraction, Nuclear Inst. and Methods in Physics Research, B, 268 (2010) 334-340.
[25] stellite
[26] A.J. Zaddach, C. Niu, C.C. Koch, and D.L. Irving. Mechanical Properties and Stacking Fault Energies of NiFeCrCoMn High-Entropy Alloy. JOM 65 (2013) 12.
[27] M. S. Lucas, L. Mauger, J. A. Muñoz, Yuming Xiao, A. O. Sheets,S. L. Semiatin, J. Horwathand Z. Turgut. Magnetic and vibrational properties of high-entropy alloys. J App Phys 109 (2011) 07E307.
[28] W.B. 27. A Handbook of lattice spacing and structures of metals and alloys. Pergmanon press Ltd., London 1958.
[29] Tao-Tsung Shun, Liang-Yi Chang, Ming-Hua Shiu. Microstructure and mechanical properties of multiprincipal component CoCrFeNiMox alloys. Mater Charact 70 (2012) 63-67.
[30] W.R. Wang, W.L. Wang, S.C. Wang, Y.C. Tsai, C.H. Lai, J.W. Yeh. Effects of Al addition on the microstructure and mechanical property of AlxCoCrFeNi high-entropy alloys.Intermetallics 26 (2012) 44-51.
[31] T.T. Shun, Y.C. Du. Microstructure and tensile behaviors of FCC Al0.3CoCrFeNi high entropy alloy. J Alloy
Compd 479 (2009) 157–160.
[32] K.A. Nibur, D.F. Bahr. Identifying slip systems around indentations in FCC metals. Scripta Mater 49 (2003) 1055–1060.
[33] K. A. Nibur, F. Akasheh , D. F. Bahr. Analysis of dislocation mechanisms around indentations through slip step observations. J Mater Sci 42 (2007) 889–900.
[35] C. Ng, S. Guo, J. Luan, Q. Wang, J. Lu, S. Shi , C.T. Liu. Phase stability and tensile properties of Co-free Al0.5CrCuFeNi2 high-entropy alloys. J Alloy Compd 584 (2014) 530–537.
[36] J. Weiss, M. Montagnat. Long-range spatial correlations and scaling in dislocation and slip patterns 2007. Philos Mag (2007) 1161–1174.
[37] S.I. Hong, and C. Laird. Mechanism of slip mode modification in f.c.c. solid solutions. Acta metal. mater. 38 (1990) 1581-1594.
[38] F. Otto, A. Dlouhy, Ch. Somsen, H. Bei, G. Eggeler, E.P. George. The influences of temperature and microstructure on the tensile properties of a CoCrFeMnNi high-entropy alloy. Acta Mater 61 (2013) 5743– 5755.
[39] Z. Wu, C.M. Parish, H. Bei. Nano-twin mediated plasticity in carbon-containing FeNiCoCrMn high entropy alloys J Alloy Compd 647 (2015) 815-822.
[40] T.T. Shun, YC Du. Age hardening of the Al0.3CoCrFeNiC0.1 high entropy alloy. J Alloy Compd 478 (2009)
