- Email: [email protected]
Structural investigations of mechanical properties of Al based rapidly solidified alloys
Materials and Design 32 (2011) 4970–4979
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
Structural investigations of mechanical properties of Al based rapidly solidified alloys Ercan Karaköse a, Mustafa Keskin b,⇑ a b
Karatekin University, Faculty of Sciences, Department of Physics, 18100 Çankırı, Turkey Erciyes University, Faculty of Sciences, Department of Physics, 38039 Kayseri, Turkey
a r t i c l e
i n f o
Article history: Received 1 March 2011 Accepted 21 May 2011 Available online 1 June 2011 Keywords: Rapid solidification Intermetallics Mechanical properties
a b s t r a c t In this study, Al based Al–3 wt.%Fe, Al–3 wt.%Cu and Al–3 wt.%Ni alloys were prepared by conventional casting. They were further processed using the melt-spinning technique and characterized by the X-ray diffraction (XRD), scanning electron microscopy (SEM) together with energy dispersive spectroscopy (EDS), transmission electron microscope (TEM), differential scanning calorimetry (DSC) and the Vickers microhardness tester. The rapidly solidified (RS) binary alloys were composed of supersaturated a–Al solid solution and finely dispersed intermetallic phases. Experimental results showed that the mechanical properties of RS alloys were enhanced, which can be attributed to significant changes in the microstructure. RS samples were measured using a microhardness test device. The dependence of microhardness HV on the solidification rate (V) was analysed. These results showed that with the increasing values of V, the values of HV increased. The enthalpies of fusion for the same alloys were determined by DSC. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction Metallurgists and materials scientists have been striving for several centuries to develop new materials which are stronger, stiffer, more ductile and lighter than existing materials which can be used at high temperatures. It is very important to develop structural materials that require less processing cost, and also to improve their properties as elevated temperature strength, density and stiffness. Improved properties can lead to weight reductions resulting in extended lifetime, fuel reduction, etc. The need for these improvements is of particular importance in developing structural aluminum alloys [1]. One way of achieving this objective is to use the rapid solidification processing method to synthesize the fine-grained microstructure and obtain a large fraction of phases in solidified condition. Among the rapid solidification techniques, the melt spinning is most commonly used method that can significantly modify the structure of materials [2]. Rapid solidification processing (RSP) involves exceptionally high rates of cooling (104–108 K/s) during solidification from the molten state. RS allows the synthesis of materials (powders, foils, scales, ribbons) with refined microstructures and extended solid solubility of their constituent elements. The result is materials with enhanced mechanical, physical and chemical properties compared with those
⇑ Corresponding author. Tel.: +90 352 437 49 01x33105; fax: +90 352 437 49 33. E-mail address: [email protected] (M. Keskin). 0261-3069/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2011.05.042
that are conventionally processed [3]. The precipitation of a supersaturated solid solution in melt-spun alloys is usually accompanied by recrystallization, affecting aging microstructures and corresponding properties. The mutual actions between precipitation and recrystallization have been studied for steel, cooper, nickel and aluminum based alloys [4–8]. However, the use of aluminum has been restricted by its low degree of strength and poor corrosion resistance. Both properties can be enhanced by adding certain alloying elements. Iron, copper and nickel have been used to increase the surface hardness of aluminum based intermetallic alloys. These elements have a significant influence on the mechanical properties of the alloy. Similarly, an appropriate selection of the different processing stages can lead to a significant strengthening by grain size refinement and/or fine dispersion of hard second-phases. The intermetallic compounds have long been recognized as potentially useful structural materials for high temperature applications [9]. The great interest in aluminum-based intermetallic systems is due to the important properties they process for potential technological applications, such as high melting temperature, comparatively low density, good oxidation resistance, increase in yield strength with increasing temperature and extreme hardness [10]. It is therefore important to refine the size and change the morphology of intermetallic compounds. They have been employed in the aerospace, mechanical, electro-chemical and environmental industries [11]. In this study, we used Al–3 wt.%Fe, Al–3 wt.%Cu and Al–3 wt.%Ni because it is known that the
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
iron-containing intermetallic compound is formed as an apiculate phase, which can significantly deteriorate fracture toughness and reduce its effectiveness on the microstructure’s stability at high temperatures [12]. Al-rich Al–Fe alloys (containing 2–5 wt.%Fe) are of particular interest to researchers because of their high corrosion resistance at high temperatures and good mechanical properties. Different aspects of the effects of iron on aluminum casting alloys have been summarized in earlier studies [13,14]. Iron forms an Al3Fe intermetallic phase that has a detrimental effect on mechanical properties due to its high degree of hardness. Alloys containing 3–6 wt.%Cu respond best to thermal treatments [15]. Mechanical properties of Al–Cu alloys depend on copper content. Copper is added to aluminum alloys to increase their strength, hardness, fatigue and creep resistance and machinability [16,18]. The first and most widely used aluminum alloys were those containing 3–10 wt.%Cu. Even though additions of copper can reduce hot tear resistance and decrease castability and corrosion resistance, copper substantially improves strength and hardness in the as-cast and heat-treated conditions [17]. It is also clear that the mechanical properties of Al–Ni alloys strongly depend on microstructural parameters, such as phase chemistry, particle size and particle volume fraction. On the other hand, Al–Ni alloys exhibit the Al3Ni intermetallic phase [18]. Al–Ni based intermetallic compounds are characterized by low density, high strength, good oxidation resistance and, for some of them, an improvement in strength with increasing temperature [19,20]. In this study, we correlated the microstructure of the intermetallic Al3Fe, Al2Cu and Al3Ni phases with the cooling rate. It would be interesting to find out how alloying elements influence the microstructural features of aluminum when different cooling rates were applied. The cooling rate was controlled by the angular speed of the copper wheel used in the chill block melt-spinning casting technique. The resulting microstructures were observed and analyzed by the optical microscopy (OM), X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscope (TEM) and differential scanning calorimetry (DSC). Investigations were carried out on the Al–3Fe, Al–3Cu and Al–3Ni (in wt.%) alloys ribbons produced by the melt-spinning technique. The microhardness and tensile-stress values of the conventionally cast samples and meltspun ribbons were also measured. In addition, in order to compare the characteristics of structures and properties, the ingot structure and properties were also studied.
2. Experimental procedure A master alloys with nominal compositions of Al–3 wt.%Fe, Al– 3 wt.%Cu and Al–3 wt.%Ni were prepared in a vacuum furnace by using 99.99% pure Al, Fe, Cu and Ni. The RS ribbons were prepared using the single-roller melt-spinning technique. The melt-spinning was performed under argon. During rapid solidification, a stream of molten alloy with a temperature of 950 °C was ejected by pressured argon from a 0.5 mm diameter orifice onto a brass wheel that was rotating at circumferential velocity 15, 20, 25, 30, 35 m/s. The temperature was approximately 200 °C above the liquids temperature of the materials. The high temperatures ensured fluency of liquid alloys. During the heating of the alloy its temperature has been checked by means of a thermocouple, so that the melt overheating can be controlled and the alloy can be readily ejected. The resulting ribbons were typically several meters long, 2–7 mm wide and 42–138 lm thick. Melt-spun Al–3Fe, Al–3Cu and Al–3Ni ribbons were characterized using optical microscopy (OM), scanning electron microscopy (SEM), differential scanning calorimeter (DSC) and X-ray diffractometry techniques (XRD). The OM observations were conducted using an Olympus BH2 model optical microscope. The SEM
4971
investigations were carried out using a LEO 440 scanning electron microscope operated at 30 kV and linked with an energy dispersive spectrometry (EDS) attachment. The melt-spun ribbons were further examined by an analytical transmission electron microscope JEM 2010 (JEOL, Japan) equipped with a Noran Vantage energy dispersive spectrometer (ThermoNoran, The Netherlands). Thermal analyses were performed using a Perkin–Elmer DSC-7 differential scanning calorimeter with a heating of 10 K/min in Ar protection in a temperature range of 450–750 K. The phase structures of the ingot and melt-spun alloys were determined using an XRD diffractometer (Bruker AXS D8). The diffraction, with experimental parameters of 160 mA, 40 kV and 10°/min respectively, was performed using Cu Ka radiation filtered by graphite. For metallographic observation, the samples were cold-mounted and prepared in the normal manner, ground to a 2500 grind and polished on 6, 3, 1 and 0.25 lm diamond lap wheels. The hardness measurements of the as-cast and ribbons were performed in a Highwood HVDM3 model digital microhardness tester at room temperature. A Vickers pyramidal indenter with different loads (0.098, 0.245, 0.49, 0.98, 1.47, and 1.96 N) and a loading time of 10 s were used to measure the diagonals of indentation with an accuracy of ±0.1 lm. An average of 10 readings at different locations of specimen surfaces was taken for each specimen. The uniaxial tensile tests were performed at room temperature at a strain rate of 103 s1 with a Shimadzu Universal Testing Instrument (Type AG-10KNG), which was designed for testing the stress–strain responses of the as-cast and ribbons. In order to avoid damaging the sample surface, two seals were stuck on the sample instead of the traditional clip gauge. Strains were then measured by observing the displacement between the two seals using a video camera. A computer with data acquisition software was used to collect the data.
3. Results and discussion 3.1. X-ray diffraction Figs. 1–3 show the XRD patterns of the ingot and melt-spun Al– Fe, Al–Cu and Al–Ni ribbons at the same wheel speed of 25 m/s, respectively. Fig. 1a displays the XRD pattern of an Al–3Fe ingot sample that has two phases, a-Al and intermetallic Al3Fe. For the melt-spun Al–3Fe ribbons, only the diffraction peaks of the Al solid solution are distinguished in the XRD pattern as shown in Fig. 1b. The XRD pattern indicates that Al–3Cu ingot alloy has two different phases Al2Cu intermetallic and a-Al. Looking at except a-Al, other diffraction peaks, the intermetallic Al2Cu phase is not detected in Al–3Cu ribbons, as shown in Fig. 2a and b, respectively. The phase constitutions of the ingot and melt-spun Al–3Ni alloy are shown in Fig. 3a and b. Whereas diffraction peaks belonging to the intermetallic Al3Ni and a-Al phases are present in the conventionally cast Al–3Ni sample (Fig. 3a), only the diffraction peaks of the Al solid solution are present in the melt-spun Al–3Ni ribbon (Fig. 3b). Table 1 summarizes the crystallographic data obtained from XRD analysis. Rapid solidification has an effect on the alloy’s phase constitution. It can be seen from Figs. 1–3 that the substitution of Fe, Ni and Cu for Al, changes the phase structure of ingot alloys. All melt-spun alloys display a single phase structure. Clearly this result is not in complete harmony with that of the SEM observations discussed later. As mentioned above, in the melt-spun alloys, no Al3Fe, Al2Cu or Al3Ni intermetallic phases were detected in the XRD patterns, probably because the amounts of the intermetallic phases are so small that they were undetectable by the XRD with a comparatively fast scanning rate (10°/min). Apart from the X-ray diffraction’s limitation in detecting low volume concentrations (typically 5 vol.% [21]), the absence of intermetallic phases in the
4972
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
Fig. 1. The XRD pattern of the Al–3%Fe alloy taken from (a) the ingot and (b) the wheel side surface of the melt-spun ribbons at 25 m/s.
Fig. 3. The XRD pattern of the Al–%3Ni alloy taken from (a) the ingot and (b) the wheel side surface of the melt-spun ribbons at 25 m/s.
Table 1 Crystallographic data of phases present in the Al–%3Fe, Al–%3Cu and Al–%3Ni alloys. Phase
Class
Space group
Lattice parameters (nm)
a-Al Al3Fe
Cubic Cubic
Fm3m Fm3m
Al3Ni
Orthorhombic
Pnma
Al2Cu
Body-centered tetragonal
I4/mcm
a = 0.4038–0.4051 a = 1.5489, b = 0.8083, c = 1.2476 a = 0.6598, b = 0.7352, c = 0.4802 a = 0.6066, c = 0.4874
X-ray diffraction pattern also revealed the extended solid solubility of the matrix. Nevertheless, considering the equilibrium solid solubility values of Fe, Cu and Ni in Al, the present results indicate that these values were extended by the melt-spinning method.
3.2. Thermal stability and crystallization
Fig. 2. The XRD pattern of the Al–%3Cu alloy taken from (a) the ingot and (b) the wheel side surface of the melt-spun ribbons at 25 m/s.
In order to examine the thermal stability and the crystallization of the melt-spun alloys, the DSC analysis was conducted. The resulting profiles shown in Fig. 4 revealed that during heating the melt-spun alloys. The enthalpy of fusion (DH) and specific heat (Cp) were calculated from the area under the peak by a numerical integration. It is clear that sharp peaks were observed for the meltspun alloys as shown in Fig. 4. As can be seen from Fig. 4a, the sharp exothermic reaction of Al–3Fe melt-spun alloy was detected to be 632 K. The sharp peak 632 K is clearly linked with the precipitation of a crystalline mixture of fcc-Al and Al3Fe intermetallic phases. The values of the enthalpy of fusion (DH) and the specific heat (Cp) for Al–3Fe melt-spun alloy were also calculated to be
4973
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
(a)
30 µm
(b)
Fig. 4. DSC traces from the melt-spun (a) Al–%3Fe, (b) Al–%3Cu and (c) Al–%3Ni alloys.
14.90 J/g, and 0.042 J/g K, respectively, from the graph of the heat flow versus temperature. As seen in Fig. 4b, the DSC scan exhibits a sharp exothermic peak at 598 K and this curve implies the precipitation of Al2Cu intermetallic phase. The values of the enthalpy of fusion (DH) and the specific heat (Cp) for Al–3Cu melt-spun alloy were also calculated to be 24.80 J/g, and 0.076 J/g K, respectively. On the other hand, Fig. 4c shows only sharp exothermic peak at 622 K. This sharp peak is clearly linked with the precipitation of Al3Ni intermetallic phase as the Al–3Ni melt-spun alloy. Similarly, the values of the enthalpy of fusion (DH) and the specific heat (Cp) for Al–3Ni melt-spun alloy were also calculated to be 42.80 J/g, and 0.122 J/g K, respectively. Resulting values of DH and Cp are in a reasonable agreement with those reported earlier [22–26].
30 µm
(c)
3.3. The cooling rate Cooling rate plays a key role in the formation of the RS microstructures mentioned above. It is known that in the RS process, as the size of particles decreases the convective cooling rate increases, which contributes to a large undercooling in the smaller particles. The cooling rate R was roughly estimated from the equation [27]
R ¼ hðT 1 T 0 Þ=C P qt;
30 µm
ð1Þ Fig. 5. The optic micrographs of the ingot (a) Al–%3Fe, (b) Al–%3Cu and (c) Al–%3Ni alloys.
where h is the heat transfer coefficient between the ejected molten alloy and the rotating substrate, T1 and T0 are the molten and the substrate temperature, respectively, q the density and CP the specific heat of the alloy, and t the thickness of the solidified ribbon specimen. In this work, a value of 1.0 cal (cm2 K s)1 was assigned
to h for the copper roller [28]. The density of the alloy, q, may be estimated, as a first approximation, by
Table 2 Ribbons thickness and cooling rates of melt-spun Al–%3Fe, Al–%3Cu and Al–%3Ni alloys. Wheel speed
15 (m/s)
20 (m/s)
25 (m/s)
30 (m/s)
35 (m/s)
Al–3Fe
Thickness (lm) Cooling rate (K s1)
98 4.2 105
93 4.4 105
72 5.7 105
63 6.5 105
42 9.8 105
Al–3Cu
Thickness (lm) Cooling rate (K s1)
105 1.7 106
82 2.2 106
76 2.3 106
53 3.4 106
48 3.7 106
Al–3Ni
Thickness (lm) Cooling rate (K s1)
138 8.2 105
117 9.7 105
93 1.2 106
86 1.3 106
71 1.6 106
4974
q¼
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
X
X i qi ;
ð2Þ
where Xi represents the mole fraction of component i in the alloy, and qi is the density of the pure component i. The density of Al– 3Fe, Al–3Cu and Al–3Ni alloys was determined to be 2.855 gcm3, 2.887 gcm3 and 2.886 gcm3, respectively. Thus the calculated cooling rates are presented in Table 2. The estimated cooling rates remarkably varied within approximately 6.12 105 K s1, 2.66 106 K s1 and 1.17 106 K s1, for the Al–3Fe, Al–3Cu and Al–3Ni alloys, respectively. The RSP methods focused upon the attainment of high cooling rates in the range of 104–108 K s1 [29]. The RS alloys resulted in refined and equalized grain morphology in contrast to the coarse structure, with severe microsegregation, observed in conventionally cast alloys. Therefore, the limited
diffusion of atoms during the rapid solidification process is expected to suppress the formation of intermetallic phases and change the morphology. In addition, the high cooling rate drives the alloy away from equilibrium conditions, and, as a result, the solid solubility limit of the alloying elements increases. Similar results were reported by Matyja et al. [30]. 3.4. The effect of cooling rate on microstructure Fig. 5 shows the optical micrograph of the microstructure for ingot samples. Fig. 5a illustrates the microstructure of the Al–3Fe ingot alloy, comprising the primary needle-like Al3Fe phase embedded in the a-Al matrix. As seen from Fig. 5b and c, each micrograph shows a discontinuous microstructure in which the
(a)
(d)
(b)
(e)
(c)
(f)
Fig. 6. The SEM micrographs of melt-spun ribbons: (a) Al–%3Fe ribbon with a wheel speed of 15 m/s, (b) Al–%3Fe ribbon with a wheel speed of 35 m/s, (c) Al–%3Cu ribbon with a wheel speed of 15 m/s, (d) Al–%3Fe ribbon with a wheel speed of 35 m/s, (e) Al–%3Ni ribbon with a wheel speed of 15 m/s and (f) Al–%3Ni ribbon with a wheel speed of 35 m/s.
4975
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
intermetallic phases are dispersed in the Al-matrix as dentritic Cu and Ni particles, respectively. Those microstructures are expected for conventionally Al based Al–Fe, Al–Cu and Al–Ni cast alloys and similar results have also been reported in literature [19,31– 35]. In order to obtain a more detailed microstructural characterization, the melt-spun ribbons were also studied by SEM photographs. Rapid solidification changes phase constitution and also has a significant effect on the microstructure of the Al based intermetallics. Fig. 6 illustrates the microstructure of the melt-spun Al– 3Fe, Al–3Cu and Al–3Ni alloys that rapidly solidified under different wheel speeds (V = 15 and 35 m/s). Decreases in particles sizes are accompanied by an increase in their numbers with increasing wheel speeds (Fig. 6a–f). Fig. 6a and b illustrate RS alloys at minimum V = 15 m/s wheel speed and at maximum V = 35 m/s wheel speed of the melt-spun Al–3Fe alloy. As shown in Fig. 6a, a relatively large amount of the small dendrite Al3Fe phase exists in the sample at V = 15 m/s low wheel speed (cooling rate 4.2 105 K s1), while in the specimen at wheel speed V = 35 m/s (cooling rate 9.8 105 K s1), dendrite structures have smaller and more homogenous distribution than in the Al–3Fe melt-spun alloy seen in Fig. 6b. The morphology of the intermetallic Al3Fe is also influence by the solidification rate. The average dimension of dendrites is measured approximately 0.5 lm at 35 m/s (Fig. 6a), which is about five times smaller than that measured for the minimum wheel speed at 15 m/s (2.5 lm) of the same alloy. As seen in Fig. 6c and d, a relatively large amount of the polygonal Al2Cu phase exists in the sample at V = 15 m/s and 35 m/s wheel speeds, respectively. In the specimen at wheel speed V = 15 m/s (cooling rate 1.7 106 K s1), the polygonal structures have coarsened as seen in Fig. 6c. While in the specimen at wheel speed V = 35 m/s (cooling rate 3.7 106 K s1), the polygonal structures have smaller and more homogenous distribution than those seen in Fig. 6d, these polygonal which are in size approximately between 2 and 4 lm. On the other hand, Fig. 6e and f show the SEM micrographs of melt-spun Al–3Ni alloy prepared using a circumferential wheel speed of V = 15 m/s (cooling rate 8.2 105 K s1) and 35 m/s (cooling rate 1.6 106 K s1), respectively. As seen in Fig. 6e, there are clear indications of dentritic formations and grain colonies. While in the specimen solidified at a rate of V = 35 m/s as shown in Fig. 6f, the dentritic formations and grain colonies disappear and very fine homogeneous structures are observed; hence, the fast cooling rate changes the melt-spun structure. The microstructure of the NiAl dendrites has also been explored in more detail by Sheng et al. [33,34]. Evidently, a higher solidification rate leads to smaller grain size and increases particle number. This indicates that the high solidification rate of melt-spinning process effectively improves the alloy’s composition homogeneity. For a more detailed microstructural characterization, the meltspun ribbons were also studied by the TEM. The results obtained from the TEM characterization confirmed the results of the SEM studies. The TEM observation (Fig. 7) shows that the microstructure is much homogeneous and finer with the grain size around 10–50 nm in the samples of melt-spun ribbons. Some more irregular and coarse intermetallic Al3Fe, Al2Cu and Al3Ni grains exist in the melt-spun samples. In TEM exploration, the grains forming the microstructure of melt-spun alloys Al–3Fe and Al–3Ni were found to be similar grains, as disclosed by the SEM examination of these samples. The similar results have also been reported [21,35]. However, in the melt-spun alloy Al–3Cu, the grains were found by TEM to be almost polygonal. In order to observe the secondary and intermetallic phases, the aluminum matrix was removed by electrochemical etching and the resulting structure was observed using SEM with the EDS spectrometer with a possible error of up to 1%. The results are shown in Fig. 8. The results obtained by energy dispersive spectrometry (EDS) indicate that the major phases of the melt-spun alloys were
(a)
50 nm
(b)
50 nm
(c)
50 nm Fig. 7. Bright field TEM images showing the microstructure of the Al–%3Fe, Al– %3Cu and Al–%3Ni at 35 m/s melt-spun alloys.
intermetallic phases (denoted as Fig. 8a–c for Al–3Fe, Al–3Cu and Al–3Ni, respectively). The melt-spun Al–3Fe, Al–3Cu and Al–3Ni alloys were perceived to be composed mainly of the a-Al phase (dark
4976
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
(a) Al3Fe intermetallic phase
-Al rich phase
(b) Al2Cu intermetallic phase
-Al rich phase
(c) Al3Ni intermetallic phase
-Al rich phase
Fig. 8. The chemical composition analysis of melt-spun alloys using SEM with EDS, (a) Al–%3Fe ribbon with wheel a speed of 20 m/s, (b) Al–%3Cu ribbon with a wheel speed of 20 m/s and (c) Al–%3Ni ribbon with a wheel speed of 40 m/s.
gray phase) and intermetallic Al3Fe, Al2Cu and Al3Ni phases (gray phase), as shown in Fig. 8. The general compositions of these phases were 98.02%Al and 1.98%Fe, 97.13%Al and 2.87%Cu, 98.19%Al and 1.81%Ni, for the Al–3Fe, Al–3Cu and Al–3Ni meltspun alloys, respectively. The EDS traces from particles in the melt-spun alloys showed that a considerable amount of Fe, Cu
and Ni had dissolved into the intermetallic compound, and was thus consistent with the XRD results seen in Figs. 1b, 2b and 3b for Al–3Fe, Al–3Cu and Al–3Ni, respectively. In other words, rapid solidification resulted in an increase in the solubility of the Al-matrix. According to the results of the EDS analysis, the peaks of Al, Ni and Cu were clearly seen, and the element
4977
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
composition percentages were similar to those compositions before measurements.
3.5. Mechanical properties One of the purposes of this investigation was to learn the relationships among solidification processing parameters, cooling rate or wheel speed and microhardness (HV) for conventionally cast and
Table 3 Dependence of the microhardness (HV) on the different wheel speeds (V) for meltspun Al–3 wt.%Fe, Al–3 wt.%Cu and Al–3 wt.%Ni alloys (in wt.%) ribbons and ingot alloys. Microhardness (kg/mm2)
Samples
Chill casting Ribbon (15 m/s) Ribbon (20 m/s) Ribbon (25 m/s) Ribbon (30 m/s) Ribbon (35 m/s)
Al–3 wt.% Fe alloys
Al–3 wt.% Cu alloys
Al–3 wt.% Ni alloys
84.47 ± 8.25 165.69 ± 4.63 173.48 ± 2.86 179.93 ± 7.98 180.92 ± 6.65 182.74 ± 3.78
76.82 ± 6.77 125.78 ± 2.28 128.07 ± 5.66 137.73 ± 8.94 141.94 ± 3.54 144.42 ± 8.23
80.84 ± 4.33 137.46 ± 6.43 142.54 ± 7.32 156.48 ± 8.21 163.33 ± 5.54 160.52 ± 2.82
rapidly solidified Al–3Fe, Al–3Cu and Al–3Ni alloys. The mechanical properties of any solidified material are usually determined with the hardness test, tensile strength test, ductility test, etc. Since true tensile strength testing of solidified alloys gave inconsistent results with a wide scatter due to their strong dependence on solidified sample surface quality, the mechanical properties were monitored by hardness testing, which is one of the easiest and most straightforward techniques. In this study, the mechanical properties of rapidly solidified ribbons and their conventionally cast counterparts were determined by Vickers microhardness measurements. As performed in previous literature studies [17,19,36], first we used one applied load (0.098 N) to measure the Vickers hardness value (HV) of both ingot and melt-spun specimens. The values were calculated using the standard Vickers formula:
microhardness (kgf/mm2)
200 Al-3wt %Fe Al-3wt %Cu Al-3wt %Ni
175
150
125
10
15
20
Fig. 11. Typical nominal stress–strain curves for rapidly solidified Al–%3Fe, Al– %3Cu and Al–%3Ni alloys.
25
30
35
40
wheel speed (m/s) Fig. 9. Variation of Vickers microhardness with wheel speed of melt-spun Al–%3Fe, Al–%3Ni and Al–%3Cu.
Fig. 10. Typical nominal stress–strain curves for conventionally cast Al–%3Fe, Al– %3Cu and Al–%3Ni alloys.
HV ¼
2P sinðh=2Þ 2
d
¼
1:8544ðPÞ d
2
;
ð3Þ
where P is the indentation force, d is the average diagonal length and 1.8544 is a geometrical factor for the diamond pyramid. However, in order to complete the investigation of microstructures, we also measured microhardness. Variations of the values of microhardness (HV) on the different wheel speed (v) are given in Table 3 and Fig. 9 for the each alloy system. Table 3 lists the Vickers microhardness values of the Al–3Fe, Al–3Cu and Al–3Ni ribbons and those of the conventionally cast alloys having the same composition. For the melt-spun Al–3Fe, Al–3Cu and Al–3Ni alloys, the Hv values were approximately 176.55, 135.58 and 152.06 kg/mm2, respectively. These values agree Takeda et al. [37], Hsu et al. [32] and Audebert et al. [38]. However, the microhardness of melt-spun Al–3Fe, Al–3Cu and Al–3Ni ribbons was approximately 2.09, 1.76 and 1.88 times higher than those of the original ingot alloys, respectively, thus they closely matched Al–Fe, Al–Cu and Al–Ni based melt-spun alloys [17,36–41]. We should also mention that the high hardness for all the melt-spun alloys compared with their ingot counterparts can be attributed to the super saturated solid solution of the a-Al phase. The atomic radius difference between the three elements in the solid solution provides a strain field, which interacts with dislocations. This solid solution strengthening mechanism is also supported by the XRD analyses SEM and TEM examinations. Another aim of this study was to compare the tensile-stress (r) values of conventionally cast and melt-spun the Al–3Fe, Al–3Cu, Al–3Ni alloys. Typical stress–strain curve from tensile test in con-
4978
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
ventionally cast Al–3Fe, Al–3Cu and Al–3Ni samples are shown in Fig. 10. It can be observed that all the conventionally cast alloys had tensile-stress values and elongation in the range of 250– 320 MPa, 35–45%, respectively. The highest tensile-stress value was achieved by the Fe containing Al–3Fe alloy. On the other hand, the tensile stress–strain curves of Al–3Fe, Al–3Cu and Al–3Ni ribbons are shown in Fig. 11. As seen from this figure, the tensilestress values of the Al–3Fe, Al–3Cu and Al–3Ni ribbons were 378, 289 and 328 MPa, respectively. The elongation was also, respectively, elevated from 1.4%, 1.7% and 1.5%. Rapidly solidified ribbons show superior mechanical properties compared to conventionally cast counterparts which are attributed to precipitates of intermetallic compound in the Al matrix and refinement the microstructures. Additionally, tensile stress value of Al–3Fe alloy is greater than Al–3Ni and Al–3Cu alloys. It can be concluded that the addition of 3 wt.%. Fe increases the tensile strength of the rapidly solidified alloy at room temperature, but slightly affects the elongation. This is attributed to the presence of more intermetallic compound Al3Fe and refinement the crystal size. Also, this behavior can be explained in terms of the alloy structure and resulting properties.
4. Conclusions The present work investigated the influence of cooling rate (wheel speed) on the microstructure and microhardness of Al based (Al–3 wt.%Fe, Al–3 wt.%Cu and Al–3 wt.%Ni) rapidly solidified alloys. The different intermetallic compounds that formed under rapid solidification were identified and characterized. The conclusions may be summarized as follows: (1) XRD patterns revealed melt-spun Al–3Fe, Al–3Cu and Al–3Ni ribbons whose peaks related to two phases, namely the a-Al and intermetallic phases (Al3Fe, Al2Cu and Al3Ni). It was noticed that no peaks corresponding to intermetallic phases were seen in Figs. 1b–3b. The solidification rate was high enough to hold most of the alloying elements in the Al matrix as a solid solution, and in addition, the rapid solidification had an effect on phase constitution. There were intermetallic phase formations between Al, Fe, Cu and Ni. (2) The Al–3Fe, Al–3Cu and Al–3Ni ribbons were heated with a heating rate of 10 K/min from room temperature to 750 K. From the trace of heat flow versus temperature, Al–3Fe, Al–3Cu and Al–3Ni alloys, the melting temperatures were detected to be 632, 598 and 622 K, respectively. The enthalpy of fusion values (DH) for Al–3Fe, Al–3Cu and Al– 3Ni alloys were found to be 14.90 J/g, 24.80 J/g and 42.80 J/ g, respectively. On the other hand, the specific heat values for (Cp) Al–3Fe, Al–3Cu and Al–3Ni alloys were found to be 0.042 J/g K, 0.076 J/g K and 0.122 J/g K, respectively. The resulting DH and Cp values are in reasonable agreement with those reported in previous studies [22–26]. (3) The micrographs recorded using the OM, SEM and TEM gave information about the size and microstructure of the intermetallic phases. They also indicated that cooling rates clearly influence the morphology of the Al–3Fe, Al–3Cu and Al–3Ni alloys. Lower cooling rates produced coarse needle-like and coarse dentritic structures in the ingot alloys, while higher cooling rates produced more regular minor dentritic or polygonal structures in the rapidly solidified alloys. (4) The microhardness value for Al based (Al–3 wt.%Fe, Al– 3 wt.%Cu and Al–3 wt.%Ni) rapidly solidified alloys increased with the value of solidification rates. The relationships between microhardness and the solidification rate for Al– 3Fe, Al–3Cu and Al–3Fe aloys were found to be approxi-
mately the Hv value 176.55, 135.58 and 152.06 kg/mm2, respectively. The microhardness values agreed with the previous experimental studies [32,37,39]. The tensile stresses of the rapidly solidified sample are about 378 and 328 MPa, respectively. Acknowledgments This work was supported by the Turkish State Planning Organization (DPT), Grant No. 2003K120510, and Erciyes University Research Funds, Grant No. FBD-08-472. References [1] Katgerman L, Dom F. Rapidly solidified aluminium alloys by meltspinning. Mater Sci Eng 2004;A375–377:1212–6. [2] Anantharman TR, Suryanarayana C. Rapidly solidified metals. Switzerland: Trans Tech Publisher; 1987. p. 5. [3] Jones H. Microstructure rapidly solidified materials. Mater Sci Eng 1984;A65:145–56. [4] Hsu CJ, Kao PW, Ho NJ. Ultrafine-grained Al–Al2Cu composite produced in situ by friction stir processing. Scr Mater 2005;53:341–5. [5] Guan YC, Zhou W, Li ZL, Zheng HY. Study on the solidification microstructure in AZ91D Mg alloy after laser surface melting. Appl Surf Sci 2009;255:8235–8. [6] Tsau CH, Chen YC. The coarsening of the precipitates in melt-spun Al–Ti–Zr ribbons. Mater Chem Phys 2002;73:111–7. [7] Su~nola JJ, Saurina J, Bruna P. Structural and thermal changes induced by mechanical alloying in a Fe–Ni based amorphous melt-spun alloy. Mater Chem Phys 2009;114:996–9. [8] Sieber H, Park JS, Weıssmüller J, Perepezko JH. Structural evolutıon and phase formatıon ın cold-rolled alumınum–nıckel multılayers. Acta Mater 2001;49:1139–51. [9] Pope DP, Darolia R. In: Westbrook JH, editor. Applications of intermetallic compounds. MRS bulletin, vol. 21; 1996. p. 26. [10] Colin J, Serna S, Campillo B, Flores O, Juarez-Islas J. Microstructural and lattice parameter study of as-cast and rapidly solidified NiAl intermetallic alloys with Cu additions. Intermetallics 2008;16:847–53. [11] Yong LH, Hun JS, Yong LS, Hyun KK. Alloying of cold-sprayed Al–Ni composite coatings by post-annealing. Appl Surf Sci 2002;253:3496–502. [12] Rajabi M, Simchi A, Vahidi M, Davami P. Effect of particle size on the microstructure of rapidly solidified Al–20Si–5Fe–2X (X = Cu, Ni, Cr) powder. J Alloys Compd 2008;466:111–8. [13] Meyer M, Mendoza-Ze´lis L, Sa´nchez FH, Clavaguera-Mora MT, Clavaguera N. On the mechanism of milling-induced disordering in AlFe. Phys Rev B 1999;60:3206–12. [14] Karpe N, Larsen KK, Bottiger J. Phase formation induced by ion irradiation and electrical resistivity of aluminum-3d-transition-metal alloys. Phys Rev B 1992;46:2686–92. [15] Spinelli JE, Rosa DM, Ferreira IL, Garcia A. The effect of melt temperature profile on the transient metal/mold heat transfer coefficient during solidification. Mater Sci Eng 2004;A408:317–25. [16] Takeda M, Maeda Y, Yoshida A, Yabuta K, Konuma S, Endo T. Discontinuity of GP(I) zone and h00 -phase in an Al–Cu alloy. Scr Mater 1999;41:643–9. [17] Karaköse E, Keskin M. Morphological characteristic of the conventional and melt-spun Al–10Ni–5.6Cu (in wt.%) alloy. Mater Charact 2009;60:1569–77. [18] Osorio WR, Spinelli JE, Ferreira IL, Garcia A. The roles of macrosegregation and of dendritic array spacings on the electrochemical behavior of an Al– 4.5 wt.%Cu alloy. Electrochim Acta 2007;52:3265–73. [19] Sahoo KL, Wollgarten M, Haug J, Banhart J. Effect of La on the crystallization behavior of amorphous Al94–XNi6LaX (x = 4–7) alloys. Acta Mater 2005;53:3861–70. [20] Kim TS, Hong SJ, Lee BT. Hardness behavior of the partially crystallized amorphous Al86Ni9Mm5 alloys. Mater Sci Eng 2003;A363:81–5. [21] Ünlü N, Genç A, Öveçog˘lu ML, Lavernia EJ, Froes FH. Microstructural evolution of the melt-spun ternary hypoeutectic Al–7.6Si–3.3Fe (in wt%) alloy. J Alloys Compd 2002;343:223–33. [22] Fiellstedt J, Fredriksson H. On the crystallization process of hypoeutectic Al– %6Cu, unmodified and Sr modified Al–%2Si solidified alloys. Adv Eng Mater 2003;5:1–2. [23] Kim CH, Ree T. Activation enthalpies for plastic deformation. Bull Kor Chem Soc 1980;1:39–43. [24] Muller S, Wang LW, Zunger A. Coherent phase stability in Al–Zn and Al–Cu fcc alloys: the role of the instability of fcc Zn. Phys Rev B 1999;60:16448–62. [25] Lee JM, Kang SB, Sato T, Tezuka H, Kamio A. Evolution of iron aluminide in Al/ Fe in situ composites fabricated by plasma synthesis method. Mater Sci Eng 2003;A362:257–63. [26] Zeng Q, Baker I. The effects of local versus bulk disorder on the magnetic behavior of stoichiometric Ni3Al. Intermetallics 2007;15:417–27. [27] Jones H. Ultrarapid quenching of liquid alloys, treatise on materials science and technology. New York: Academic; 1981. p. 76. [28] Burden MN, Jones H. Determination of cooling rate in splat-cooling from scale of micro-structure. J Inst Metals 1970;98:249.
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979 [29] Perepezko JH. Solidification reactions in undercooled alloys. Mater Sci Eng 1994;A179–180:52. [30] Matyja H, Giessen BC, Grant NJ. The effect of cooling rate on the dendrite spacing in splat-cooled aluminium alloys. J Inst Metals 1968;96:30. [31] Rios CT, Bolfarini C, Botta WJ, Kiminami CS. Rapidly solidified Al92Fe3Cr2Mn3 alloy. Mater Sci Eng 2007;A449–451:1057–61. [32] Hsu CJ, Kao PW, Ho NJ. Intermetallic-reinforced aluminum matrix composites produced in situ by friction stir processing. Mater Lett 2006;61:1315–8. [33] Sheng LY, Zhang W, Guo JT, Wang ZS, Ye HQ. Microstructure evolution and elevated temperature compressive properties of a rapidly solidified NiAl– Cr(Nb)/Dy alloy. Mater Design 2009;30:2752–5. [34] Sheng LY, Guo JT, Tian YX, Zhou LZ, Ye HQ. Microstructure and mechanical properties of rapidly solidified NiAl–Cr(Mo) eutectic alloy doped with trace Dy. J Alloys Compd 2009;475:730–4. [35] Sheng LY, Guo JT, Ye HQ. Microstructure and mechanical properties of NiAl– Cr(Mo)/Nb eutectic alloy prepared by injection-casting. Mater Des 2009;30:964–9.
4979
[36] Kaya H, Çadırlı E, Böyük U, Marasßlı N. Variation of microindentation hardness with solidification and microstructure parameters in the Al based alloys. Appl Surf Sci 2008;255:3071–8. [37] Takeda M, Komatsu A, Ohta M, Shirai T, Endo T. The influence of Mn on precipitation behavior in Al–Cu. Scr Mater 1998;39:1295–300. [38] Audebert F, Mendive C, Vidal A. Structure and mechanical behaviour of Al–Fe– X and Al–Ni–X rapidly solidified alloys. Mater Sci Eng 2004;A375– 377:1196–200. [39] Shaw L, Luo H, Villegas J, Miracle D. Effects of internal strains on hardness of nanocrystalline Al–Fe–Cr–Ti alloys. Scr Mater 2004;51:449–53. [40] Gogebakan M, Uzun O, Karaaslan T, Keskin M. Hardness and microstructural characteristics of rapidly solidified Al–8–16 wt.%Si alloys. J Mater Proc Technol 2003;142:87–92. [41] Fernandes JC, Anjinho C, Amaral PM, Rosa LG, Rodriguez J, Martinez D, et al. Characterisation of solar-synthesised TiCx (x = 0. 50, 0.625, 0.75, 0.85, 0.90 and 1.0) by X-Ray diffraction, density and Vickers microhardness. Mater Chem Phys 2002;77:711–8.
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
Structural investigations of mechanical properties of Al based rapidly solidified alloys Ercan Karaköse a, Mustafa Keskin b,⇑ a b
Karatekin University, Faculty of Sciences, Department of Physics, 18100 Çankırı, Turkey Erciyes University, Faculty of Sciences, Department of Physics, 38039 Kayseri, Turkey
a r t i c l e
i n f o
Article history: Received 1 March 2011 Accepted 21 May 2011 Available online 1 June 2011 Keywords: Rapid solidification Intermetallics Mechanical properties
a b s t r a c t In this study, Al based Al–3 wt.%Fe, Al–3 wt.%Cu and Al–3 wt.%Ni alloys were prepared by conventional casting. They were further processed using the melt-spinning technique and characterized by the X-ray diffraction (XRD), scanning electron microscopy (SEM) together with energy dispersive spectroscopy (EDS), transmission electron microscope (TEM), differential scanning calorimetry (DSC) and the Vickers microhardness tester. The rapidly solidified (RS) binary alloys were composed of supersaturated a–Al solid solution and finely dispersed intermetallic phases. Experimental results showed that the mechanical properties of RS alloys were enhanced, which can be attributed to significant changes in the microstructure. RS samples were measured using a microhardness test device. The dependence of microhardness HV on the solidification rate (V) was analysed. These results showed that with the increasing values of V, the values of HV increased. The enthalpies of fusion for the same alloys were determined by DSC. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction Metallurgists and materials scientists have been striving for several centuries to develop new materials which are stronger, stiffer, more ductile and lighter than existing materials which can be used at high temperatures. It is very important to develop structural materials that require less processing cost, and also to improve their properties as elevated temperature strength, density and stiffness. Improved properties can lead to weight reductions resulting in extended lifetime, fuel reduction, etc. The need for these improvements is of particular importance in developing structural aluminum alloys [1]. One way of achieving this objective is to use the rapid solidification processing method to synthesize the fine-grained microstructure and obtain a large fraction of phases in solidified condition. Among the rapid solidification techniques, the melt spinning is most commonly used method that can significantly modify the structure of materials [2]. Rapid solidification processing (RSP) involves exceptionally high rates of cooling (104–108 K/s) during solidification from the molten state. RS allows the synthesis of materials (powders, foils, scales, ribbons) with refined microstructures and extended solid solubility of their constituent elements. The result is materials with enhanced mechanical, physical and chemical properties compared with those
⇑ Corresponding author. Tel.: +90 352 437 49 01x33105; fax: +90 352 437 49 33. E-mail address: [email protected] (M. Keskin). 0261-3069/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2011.05.042
that are conventionally processed [3]. The precipitation of a supersaturated solid solution in melt-spun alloys is usually accompanied by recrystallization, affecting aging microstructures and corresponding properties. The mutual actions between precipitation and recrystallization have been studied for steel, cooper, nickel and aluminum based alloys [4–8]. However, the use of aluminum has been restricted by its low degree of strength and poor corrosion resistance. Both properties can be enhanced by adding certain alloying elements. Iron, copper and nickel have been used to increase the surface hardness of aluminum based intermetallic alloys. These elements have a significant influence on the mechanical properties of the alloy. Similarly, an appropriate selection of the different processing stages can lead to a significant strengthening by grain size refinement and/or fine dispersion of hard second-phases. The intermetallic compounds have long been recognized as potentially useful structural materials for high temperature applications [9]. The great interest in aluminum-based intermetallic systems is due to the important properties they process for potential technological applications, such as high melting temperature, comparatively low density, good oxidation resistance, increase in yield strength with increasing temperature and extreme hardness [10]. It is therefore important to refine the size and change the morphology of intermetallic compounds. They have been employed in the aerospace, mechanical, electro-chemical and environmental industries [11]. In this study, we used Al–3 wt.%Fe, Al–3 wt.%Cu and Al–3 wt.%Ni because it is known that the
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
iron-containing intermetallic compound is formed as an apiculate phase, which can significantly deteriorate fracture toughness and reduce its effectiveness on the microstructure’s stability at high temperatures [12]. Al-rich Al–Fe alloys (containing 2–5 wt.%Fe) are of particular interest to researchers because of their high corrosion resistance at high temperatures and good mechanical properties. Different aspects of the effects of iron on aluminum casting alloys have been summarized in earlier studies [13,14]. Iron forms an Al3Fe intermetallic phase that has a detrimental effect on mechanical properties due to its high degree of hardness. Alloys containing 3–6 wt.%Cu respond best to thermal treatments [15]. Mechanical properties of Al–Cu alloys depend on copper content. Copper is added to aluminum alloys to increase their strength, hardness, fatigue and creep resistance and machinability [16,18]. The first and most widely used aluminum alloys were those containing 3–10 wt.%Cu. Even though additions of copper can reduce hot tear resistance and decrease castability and corrosion resistance, copper substantially improves strength and hardness in the as-cast and heat-treated conditions [17]. It is also clear that the mechanical properties of Al–Ni alloys strongly depend on microstructural parameters, such as phase chemistry, particle size and particle volume fraction. On the other hand, Al–Ni alloys exhibit the Al3Ni intermetallic phase [18]. Al–Ni based intermetallic compounds are characterized by low density, high strength, good oxidation resistance and, for some of them, an improvement in strength with increasing temperature [19,20]. In this study, we correlated the microstructure of the intermetallic Al3Fe, Al2Cu and Al3Ni phases with the cooling rate. It would be interesting to find out how alloying elements influence the microstructural features of aluminum when different cooling rates were applied. The cooling rate was controlled by the angular speed of the copper wheel used in the chill block melt-spinning casting technique. The resulting microstructures were observed and analyzed by the optical microscopy (OM), X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscope (TEM) and differential scanning calorimetry (DSC). Investigations were carried out on the Al–3Fe, Al–3Cu and Al–3Ni (in wt.%) alloys ribbons produced by the melt-spinning technique. The microhardness and tensile-stress values of the conventionally cast samples and meltspun ribbons were also measured. In addition, in order to compare the characteristics of structures and properties, the ingot structure and properties were also studied.
2. Experimental procedure A master alloys with nominal compositions of Al–3 wt.%Fe, Al– 3 wt.%Cu and Al–3 wt.%Ni were prepared in a vacuum furnace by using 99.99% pure Al, Fe, Cu and Ni. The RS ribbons were prepared using the single-roller melt-spinning technique. The melt-spinning was performed under argon. During rapid solidification, a stream of molten alloy with a temperature of 950 °C was ejected by pressured argon from a 0.5 mm diameter orifice onto a brass wheel that was rotating at circumferential velocity 15, 20, 25, 30, 35 m/s. The temperature was approximately 200 °C above the liquids temperature of the materials. The high temperatures ensured fluency of liquid alloys. During the heating of the alloy its temperature has been checked by means of a thermocouple, so that the melt overheating can be controlled and the alloy can be readily ejected. The resulting ribbons were typically several meters long, 2–7 mm wide and 42–138 lm thick. Melt-spun Al–3Fe, Al–3Cu and Al–3Ni ribbons were characterized using optical microscopy (OM), scanning electron microscopy (SEM), differential scanning calorimeter (DSC) and X-ray diffractometry techniques (XRD). The OM observations were conducted using an Olympus BH2 model optical microscope. The SEM
4971
investigations were carried out using a LEO 440 scanning electron microscope operated at 30 kV and linked with an energy dispersive spectrometry (EDS) attachment. The melt-spun ribbons were further examined by an analytical transmission electron microscope JEM 2010 (JEOL, Japan) equipped with a Noran Vantage energy dispersive spectrometer (ThermoNoran, The Netherlands). Thermal analyses were performed using a Perkin–Elmer DSC-7 differential scanning calorimeter with a heating of 10 K/min in Ar protection in a temperature range of 450–750 K. The phase structures of the ingot and melt-spun alloys were determined using an XRD diffractometer (Bruker AXS D8). The diffraction, with experimental parameters of 160 mA, 40 kV and 10°/min respectively, was performed using Cu Ka radiation filtered by graphite. For metallographic observation, the samples were cold-mounted and prepared in the normal manner, ground to a 2500 grind and polished on 6, 3, 1 and 0.25 lm diamond lap wheels. The hardness measurements of the as-cast and ribbons were performed in a Highwood HVDM3 model digital microhardness tester at room temperature. A Vickers pyramidal indenter with different loads (0.098, 0.245, 0.49, 0.98, 1.47, and 1.96 N) and a loading time of 10 s were used to measure the diagonals of indentation with an accuracy of ±0.1 lm. An average of 10 readings at different locations of specimen surfaces was taken for each specimen. The uniaxial tensile tests were performed at room temperature at a strain rate of 103 s1 with a Shimadzu Universal Testing Instrument (Type AG-10KNG), which was designed for testing the stress–strain responses of the as-cast and ribbons. In order to avoid damaging the sample surface, two seals were stuck on the sample instead of the traditional clip gauge. Strains were then measured by observing the displacement between the two seals using a video camera. A computer with data acquisition software was used to collect the data.
3. Results and discussion 3.1. X-ray diffraction Figs. 1–3 show the XRD patterns of the ingot and melt-spun Al– Fe, Al–Cu and Al–Ni ribbons at the same wheel speed of 25 m/s, respectively. Fig. 1a displays the XRD pattern of an Al–3Fe ingot sample that has two phases, a-Al and intermetallic Al3Fe. For the melt-spun Al–3Fe ribbons, only the diffraction peaks of the Al solid solution are distinguished in the XRD pattern as shown in Fig. 1b. The XRD pattern indicates that Al–3Cu ingot alloy has two different phases Al2Cu intermetallic and a-Al. Looking at except a-Al, other diffraction peaks, the intermetallic Al2Cu phase is not detected in Al–3Cu ribbons, as shown in Fig. 2a and b, respectively. The phase constitutions of the ingot and melt-spun Al–3Ni alloy are shown in Fig. 3a and b. Whereas diffraction peaks belonging to the intermetallic Al3Ni and a-Al phases are present in the conventionally cast Al–3Ni sample (Fig. 3a), only the diffraction peaks of the Al solid solution are present in the melt-spun Al–3Ni ribbon (Fig. 3b). Table 1 summarizes the crystallographic data obtained from XRD analysis. Rapid solidification has an effect on the alloy’s phase constitution. It can be seen from Figs. 1–3 that the substitution of Fe, Ni and Cu for Al, changes the phase structure of ingot alloys. All melt-spun alloys display a single phase structure. Clearly this result is not in complete harmony with that of the SEM observations discussed later. As mentioned above, in the melt-spun alloys, no Al3Fe, Al2Cu or Al3Ni intermetallic phases were detected in the XRD patterns, probably because the amounts of the intermetallic phases are so small that they were undetectable by the XRD with a comparatively fast scanning rate (10°/min). Apart from the X-ray diffraction’s limitation in detecting low volume concentrations (typically 5 vol.% [21]), the absence of intermetallic phases in the
4972
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
Fig. 1. The XRD pattern of the Al–3%Fe alloy taken from (a) the ingot and (b) the wheel side surface of the melt-spun ribbons at 25 m/s.
Fig. 3. The XRD pattern of the Al–%3Ni alloy taken from (a) the ingot and (b) the wheel side surface of the melt-spun ribbons at 25 m/s.
Table 1 Crystallographic data of phases present in the Al–%3Fe, Al–%3Cu and Al–%3Ni alloys. Phase
Class
Space group
Lattice parameters (nm)
a-Al Al3Fe
Cubic Cubic
Fm3m Fm3m
Al3Ni
Orthorhombic
Pnma
Al2Cu
Body-centered tetragonal
I4/mcm
a = 0.4038–0.4051 a = 1.5489, b = 0.8083, c = 1.2476 a = 0.6598, b = 0.7352, c = 0.4802 a = 0.6066, c = 0.4874
X-ray diffraction pattern also revealed the extended solid solubility of the matrix. Nevertheless, considering the equilibrium solid solubility values of Fe, Cu and Ni in Al, the present results indicate that these values were extended by the melt-spinning method.
3.2. Thermal stability and crystallization
Fig. 2. The XRD pattern of the Al–%3Cu alloy taken from (a) the ingot and (b) the wheel side surface of the melt-spun ribbons at 25 m/s.
In order to examine the thermal stability and the crystallization of the melt-spun alloys, the DSC analysis was conducted. The resulting profiles shown in Fig. 4 revealed that during heating the melt-spun alloys. The enthalpy of fusion (DH) and specific heat (Cp) were calculated from the area under the peak by a numerical integration. It is clear that sharp peaks were observed for the meltspun alloys as shown in Fig. 4. As can be seen from Fig. 4a, the sharp exothermic reaction of Al–3Fe melt-spun alloy was detected to be 632 K. The sharp peak 632 K is clearly linked with the precipitation of a crystalline mixture of fcc-Al and Al3Fe intermetallic phases. The values of the enthalpy of fusion (DH) and the specific heat (Cp) for Al–3Fe melt-spun alloy were also calculated to be
4973
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
(a)
30 µm
(b)
Fig. 4. DSC traces from the melt-spun (a) Al–%3Fe, (b) Al–%3Cu and (c) Al–%3Ni alloys.
14.90 J/g, and 0.042 J/g K, respectively, from the graph of the heat flow versus temperature. As seen in Fig. 4b, the DSC scan exhibits a sharp exothermic peak at 598 K and this curve implies the precipitation of Al2Cu intermetallic phase. The values of the enthalpy of fusion (DH) and the specific heat (Cp) for Al–3Cu melt-spun alloy were also calculated to be 24.80 J/g, and 0.076 J/g K, respectively. On the other hand, Fig. 4c shows only sharp exothermic peak at 622 K. This sharp peak is clearly linked with the precipitation of Al3Ni intermetallic phase as the Al–3Ni melt-spun alloy. Similarly, the values of the enthalpy of fusion (DH) and the specific heat (Cp) for Al–3Ni melt-spun alloy were also calculated to be 42.80 J/g, and 0.122 J/g K, respectively. Resulting values of DH and Cp are in a reasonable agreement with those reported earlier [22–26].
30 µm
(c)
3.3. The cooling rate Cooling rate plays a key role in the formation of the RS microstructures mentioned above. It is known that in the RS process, as the size of particles decreases the convective cooling rate increases, which contributes to a large undercooling in the smaller particles. The cooling rate R was roughly estimated from the equation [27]
R ¼ hðT 1 T 0 Þ=C P qt;
30 µm
ð1Þ Fig. 5. The optic micrographs of the ingot (a) Al–%3Fe, (b) Al–%3Cu and (c) Al–%3Ni alloys.
where h is the heat transfer coefficient between the ejected molten alloy and the rotating substrate, T1 and T0 are the molten and the substrate temperature, respectively, q the density and CP the specific heat of the alloy, and t the thickness of the solidified ribbon specimen. In this work, a value of 1.0 cal (cm2 K s)1 was assigned
to h for the copper roller [28]. The density of the alloy, q, may be estimated, as a first approximation, by
Table 2 Ribbons thickness and cooling rates of melt-spun Al–%3Fe, Al–%3Cu and Al–%3Ni alloys. Wheel speed
15 (m/s)
20 (m/s)
25 (m/s)
30 (m/s)
35 (m/s)
Al–3Fe
Thickness (lm) Cooling rate (K s1)
98 4.2 105
93 4.4 105
72 5.7 105
63 6.5 105
42 9.8 105
Al–3Cu
Thickness (lm) Cooling rate (K s1)
105 1.7 106
82 2.2 106
76 2.3 106
53 3.4 106
48 3.7 106
Al–3Ni
Thickness (lm) Cooling rate (K s1)
138 8.2 105
117 9.7 105
93 1.2 106
86 1.3 106
71 1.6 106
4974
q¼
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
X
X i qi ;
ð2Þ
where Xi represents the mole fraction of component i in the alloy, and qi is the density of the pure component i. The density of Al– 3Fe, Al–3Cu and Al–3Ni alloys was determined to be 2.855 gcm3, 2.887 gcm3 and 2.886 gcm3, respectively. Thus the calculated cooling rates are presented in Table 2. The estimated cooling rates remarkably varied within approximately 6.12 105 K s1, 2.66 106 K s1 and 1.17 106 K s1, for the Al–3Fe, Al–3Cu and Al–3Ni alloys, respectively. The RSP methods focused upon the attainment of high cooling rates in the range of 104–108 K s1 [29]. The RS alloys resulted in refined and equalized grain morphology in contrast to the coarse structure, with severe microsegregation, observed in conventionally cast alloys. Therefore, the limited
diffusion of atoms during the rapid solidification process is expected to suppress the formation of intermetallic phases and change the morphology. In addition, the high cooling rate drives the alloy away from equilibrium conditions, and, as a result, the solid solubility limit of the alloying elements increases. Similar results were reported by Matyja et al. [30]. 3.4. The effect of cooling rate on microstructure Fig. 5 shows the optical micrograph of the microstructure for ingot samples. Fig. 5a illustrates the microstructure of the Al–3Fe ingot alloy, comprising the primary needle-like Al3Fe phase embedded in the a-Al matrix. As seen from Fig. 5b and c, each micrograph shows a discontinuous microstructure in which the
(a)
(d)
(b)
(e)
(c)
(f)
Fig. 6. The SEM micrographs of melt-spun ribbons: (a) Al–%3Fe ribbon with a wheel speed of 15 m/s, (b) Al–%3Fe ribbon with a wheel speed of 35 m/s, (c) Al–%3Cu ribbon with a wheel speed of 15 m/s, (d) Al–%3Fe ribbon with a wheel speed of 35 m/s, (e) Al–%3Ni ribbon with a wheel speed of 15 m/s and (f) Al–%3Ni ribbon with a wheel speed of 35 m/s.
4975
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
intermetallic phases are dispersed in the Al-matrix as dentritic Cu and Ni particles, respectively. Those microstructures are expected for conventionally Al based Al–Fe, Al–Cu and Al–Ni cast alloys and similar results have also been reported in literature [19,31– 35]. In order to obtain a more detailed microstructural characterization, the melt-spun ribbons were also studied by SEM photographs. Rapid solidification changes phase constitution and also has a significant effect on the microstructure of the Al based intermetallics. Fig. 6 illustrates the microstructure of the melt-spun Al– 3Fe, Al–3Cu and Al–3Ni alloys that rapidly solidified under different wheel speeds (V = 15 and 35 m/s). Decreases in particles sizes are accompanied by an increase in their numbers with increasing wheel speeds (Fig. 6a–f). Fig. 6a and b illustrate RS alloys at minimum V = 15 m/s wheel speed and at maximum V = 35 m/s wheel speed of the melt-spun Al–3Fe alloy. As shown in Fig. 6a, a relatively large amount of the small dendrite Al3Fe phase exists in the sample at V = 15 m/s low wheel speed (cooling rate 4.2 105 K s1), while in the specimen at wheel speed V = 35 m/s (cooling rate 9.8 105 K s1), dendrite structures have smaller and more homogenous distribution than in the Al–3Fe melt-spun alloy seen in Fig. 6b. The morphology of the intermetallic Al3Fe is also influence by the solidification rate. The average dimension of dendrites is measured approximately 0.5 lm at 35 m/s (Fig. 6a), which is about five times smaller than that measured for the minimum wheel speed at 15 m/s (2.5 lm) of the same alloy. As seen in Fig. 6c and d, a relatively large amount of the polygonal Al2Cu phase exists in the sample at V = 15 m/s and 35 m/s wheel speeds, respectively. In the specimen at wheel speed V = 15 m/s (cooling rate 1.7 106 K s1), the polygonal structures have coarsened as seen in Fig. 6c. While in the specimen at wheel speed V = 35 m/s (cooling rate 3.7 106 K s1), the polygonal structures have smaller and more homogenous distribution than those seen in Fig. 6d, these polygonal which are in size approximately between 2 and 4 lm. On the other hand, Fig. 6e and f show the SEM micrographs of melt-spun Al–3Ni alloy prepared using a circumferential wheel speed of V = 15 m/s (cooling rate 8.2 105 K s1) and 35 m/s (cooling rate 1.6 106 K s1), respectively. As seen in Fig. 6e, there are clear indications of dentritic formations and grain colonies. While in the specimen solidified at a rate of V = 35 m/s as shown in Fig. 6f, the dentritic formations and grain colonies disappear and very fine homogeneous structures are observed; hence, the fast cooling rate changes the melt-spun structure. The microstructure of the NiAl dendrites has also been explored in more detail by Sheng et al. [33,34]. Evidently, a higher solidification rate leads to smaller grain size and increases particle number. This indicates that the high solidification rate of melt-spinning process effectively improves the alloy’s composition homogeneity. For a more detailed microstructural characterization, the meltspun ribbons were also studied by the TEM. The results obtained from the TEM characterization confirmed the results of the SEM studies. The TEM observation (Fig. 7) shows that the microstructure is much homogeneous and finer with the grain size around 10–50 nm in the samples of melt-spun ribbons. Some more irregular and coarse intermetallic Al3Fe, Al2Cu and Al3Ni grains exist in the melt-spun samples. In TEM exploration, the grains forming the microstructure of melt-spun alloys Al–3Fe and Al–3Ni were found to be similar grains, as disclosed by the SEM examination of these samples. The similar results have also been reported [21,35]. However, in the melt-spun alloy Al–3Cu, the grains were found by TEM to be almost polygonal. In order to observe the secondary and intermetallic phases, the aluminum matrix was removed by electrochemical etching and the resulting structure was observed using SEM with the EDS spectrometer with a possible error of up to 1%. The results are shown in Fig. 8. The results obtained by energy dispersive spectrometry (EDS) indicate that the major phases of the melt-spun alloys were
(a)
50 nm
(b)
50 nm
(c)
50 nm Fig. 7. Bright field TEM images showing the microstructure of the Al–%3Fe, Al– %3Cu and Al–%3Ni at 35 m/s melt-spun alloys.
intermetallic phases (denoted as Fig. 8a–c for Al–3Fe, Al–3Cu and Al–3Ni, respectively). The melt-spun Al–3Fe, Al–3Cu and Al–3Ni alloys were perceived to be composed mainly of the a-Al phase (dark
4976
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
(a) Al3Fe intermetallic phase
-Al rich phase
(b) Al2Cu intermetallic phase
-Al rich phase
(c) Al3Ni intermetallic phase
-Al rich phase
Fig. 8. The chemical composition analysis of melt-spun alloys using SEM with EDS, (a) Al–%3Fe ribbon with wheel a speed of 20 m/s, (b) Al–%3Cu ribbon with a wheel speed of 20 m/s and (c) Al–%3Ni ribbon with a wheel speed of 40 m/s.
gray phase) and intermetallic Al3Fe, Al2Cu and Al3Ni phases (gray phase), as shown in Fig. 8. The general compositions of these phases were 98.02%Al and 1.98%Fe, 97.13%Al and 2.87%Cu, 98.19%Al and 1.81%Ni, for the Al–3Fe, Al–3Cu and Al–3Ni meltspun alloys, respectively. The EDS traces from particles in the melt-spun alloys showed that a considerable amount of Fe, Cu
and Ni had dissolved into the intermetallic compound, and was thus consistent with the XRD results seen in Figs. 1b, 2b and 3b for Al–3Fe, Al–3Cu and Al–3Ni, respectively. In other words, rapid solidification resulted in an increase in the solubility of the Al-matrix. According to the results of the EDS analysis, the peaks of Al, Ni and Cu were clearly seen, and the element
4977
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
composition percentages were similar to those compositions before measurements.
3.5. Mechanical properties One of the purposes of this investigation was to learn the relationships among solidification processing parameters, cooling rate or wheel speed and microhardness (HV) for conventionally cast and
Table 3 Dependence of the microhardness (HV) on the different wheel speeds (V) for meltspun Al–3 wt.%Fe, Al–3 wt.%Cu and Al–3 wt.%Ni alloys (in wt.%) ribbons and ingot alloys. Microhardness (kg/mm2)
Samples
Chill casting Ribbon (15 m/s) Ribbon (20 m/s) Ribbon (25 m/s) Ribbon (30 m/s) Ribbon (35 m/s)
Al–3 wt.% Fe alloys
Al–3 wt.% Cu alloys
Al–3 wt.% Ni alloys
84.47 ± 8.25 165.69 ± 4.63 173.48 ± 2.86 179.93 ± 7.98 180.92 ± 6.65 182.74 ± 3.78
76.82 ± 6.77 125.78 ± 2.28 128.07 ± 5.66 137.73 ± 8.94 141.94 ± 3.54 144.42 ± 8.23
80.84 ± 4.33 137.46 ± 6.43 142.54 ± 7.32 156.48 ± 8.21 163.33 ± 5.54 160.52 ± 2.82
rapidly solidified Al–3Fe, Al–3Cu and Al–3Ni alloys. The mechanical properties of any solidified material are usually determined with the hardness test, tensile strength test, ductility test, etc. Since true tensile strength testing of solidified alloys gave inconsistent results with a wide scatter due to their strong dependence on solidified sample surface quality, the mechanical properties were monitored by hardness testing, which is one of the easiest and most straightforward techniques. In this study, the mechanical properties of rapidly solidified ribbons and their conventionally cast counterparts were determined by Vickers microhardness measurements. As performed in previous literature studies [17,19,36], first we used one applied load (0.098 N) to measure the Vickers hardness value (HV) of both ingot and melt-spun specimens. The values were calculated using the standard Vickers formula:
microhardness (kgf/mm2)
200 Al-3wt %Fe Al-3wt %Cu Al-3wt %Ni
175
150
125
10
15
20
Fig. 11. Typical nominal stress–strain curves for rapidly solidified Al–%3Fe, Al– %3Cu and Al–%3Ni alloys.
25
30
35
40
wheel speed (m/s) Fig. 9. Variation of Vickers microhardness with wheel speed of melt-spun Al–%3Fe, Al–%3Ni and Al–%3Cu.
Fig. 10. Typical nominal stress–strain curves for conventionally cast Al–%3Fe, Al– %3Cu and Al–%3Ni alloys.
HV ¼
2P sinðh=2Þ 2
d
¼
1:8544ðPÞ d
2
;
ð3Þ
where P is the indentation force, d is the average diagonal length and 1.8544 is a geometrical factor for the diamond pyramid. However, in order to complete the investigation of microstructures, we also measured microhardness. Variations of the values of microhardness (HV) on the different wheel speed (v) are given in Table 3 and Fig. 9 for the each alloy system. Table 3 lists the Vickers microhardness values of the Al–3Fe, Al–3Cu and Al–3Ni ribbons and those of the conventionally cast alloys having the same composition. For the melt-spun Al–3Fe, Al–3Cu and Al–3Ni alloys, the Hv values were approximately 176.55, 135.58 and 152.06 kg/mm2, respectively. These values agree Takeda et al. [37], Hsu et al. [32] and Audebert et al. [38]. However, the microhardness of melt-spun Al–3Fe, Al–3Cu and Al–3Ni ribbons was approximately 2.09, 1.76 and 1.88 times higher than those of the original ingot alloys, respectively, thus they closely matched Al–Fe, Al–Cu and Al–Ni based melt-spun alloys [17,36–41]. We should also mention that the high hardness for all the melt-spun alloys compared with their ingot counterparts can be attributed to the super saturated solid solution of the a-Al phase. The atomic radius difference between the three elements in the solid solution provides a strain field, which interacts with dislocations. This solid solution strengthening mechanism is also supported by the XRD analyses SEM and TEM examinations. Another aim of this study was to compare the tensile-stress (r) values of conventionally cast and melt-spun the Al–3Fe, Al–3Cu, Al–3Ni alloys. Typical stress–strain curve from tensile test in con-
4978
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979
ventionally cast Al–3Fe, Al–3Cu and Al–3Ni samples are shown in Fig. 10. It can be observed that all the conventionally cast alloys had tensile-stress values and elongation in the range of 250– 320 MPa, 35–45%, respectively. The highest tensile-stress value was achieved by the Fe containing Al–3Fe alloy. On the other hand, the tensile stress–strain curves of Al–3Fe, Al–3Cu and Al–3Ni ribbons are shown in Fig. 11. As seen from this figure, the tensilestress values of the Al–3Fe, Al–3Cu and Al–3Ni ribbons were 378, 289 and 328 MPa, respectively. The elongation was also, respectively, elevated from 1.4%, 1.7% and 1.5%. Rapidly solidified ribbons show superior mechanical properties compared to conventionally cast counterparts which are attributed to precipitates of intermetallic compound in the Al matrix and refinement the microstructures. Additionally, tensile stress value of Al–3Fe alloy is greater than Al–3Ni and Al–3Cu alloys. It can be concluded that the addition of 3 wt.%. Fe increases the tensile strength of the rapidly solidified alloy at room temperature, but slightly affects the elongation. This is attributed to the presence of more intermetallic compound Al3Fe and refinement the crystal size. Also, this behavior can be explained in terms of the alloy structure and resulting properties.
4. Conclusions The present work investigated the influence of cooling rate (wheel speed) on the microstructure and microhardness of Al based (Al–3 wt.%Fe, Al–3 wt.%Cu and Al–3 wt.%Ni) rapidly solidified alloys. The different intermetallic compounds that formed under rapid solidification were identified and characterized. The conclusions may be summarized as follows: (1) XRD patterns revealed melt-spun Al–3Fe, Al–3Cu and Al–3Ni ribbons whose peaks related to two phases, namely the a-Al and intermetallic phases (Al3Fe, Al2Cu and Al3Ni). It was noticed that no peaks corresponding to intermetallic phases were seen in Figs. 1b–3b. The solidification rate was high enough to hold most of the alloying elements in the Al matrix as a solid solution, and in addition, the rapid solidification had an effect on phase constitution. There were intermetallic phase formations between Al, Fe, Cu and Ni. (2) The Al–3Fe, Al–3Cu and Al–3Ni ribbons were heated with a heating rate of 10 K/min from room temperature to 750 K. From the trace of heat flow versus temperature, Al–3Fe, Al–3Cu and Al–3Ni alloys, the melting temperatures were detected to be 632, 598 and 622 K, respectively. The enthalpy of fusion values (DH) for Al–3Fe, Al–3Cu and Al– 3Ni alloys were found to be 14.90 J/g, 24.80 J/g and 42.80 J/ g, respectively. On the other hand, the specific heat values for (Cp) Al–3Fe, Al–3Cu and Al–3Ni alloys were found to be 0.042 J/g K, 0.076 J/g K and 0.122 J/g K, respectively. The resulting DH and Cp values are in reasonable agreement with those reported in previous studies [22–26]. (3) The micrographs recorded using the OM, SEM and TEM gave information about the size and microstructure of the intermetallic phases. They also indicated that cooling rates clearly influence the morphology of the Al–3Fe, Al–3Cu and Al–3Ni alloys. Lower cooling rates produced coarse needle-like and coarse dentritic structures in the ingot alloys, while higher cooling rates produced more regular minor dentritic or polygonal structures in the rapidly solidified alloys. (4) The microhardness value for Al based (Al–3 wt.%Fe, Al– 3 wt.%Cu and Al–3 wt.%Ni) rapidly solidified alloys increased with the value of solidification rates. The relationships between microhardness and the solidification rate for Al– 3Fe, Al–3Cu and Al–3Fe aloys were found to be approxi-
mately the Hv value 176.55, 135.58 and 152.06 kg/mm2, respectively. The microhardness values agreed with the previous experimental studies [32,37,39]. The tensile stresses of the rapidly solidified sample are about 378 and 328 MPa, respectively. Acknowledgments This work was supported by the Turkish State Planning Organization (DPT), Grant No. 2003K120510, and Erciyes University Research Funds, Grant No. FBD-08-472. References [1] Katgerman L, Dom F. Rapidly solidified aluminium alloys by meltspinning. Mater Sci Eng 2004;A375–377:1212–6. [2] Anantharman TR, Suryanarayana C. Rapidly solidified metals. Switzerland: Trans Tech Publisher; 1987. p. 5. [3] Jones H. Microstructure rapidly solidified materials. Mater Sci Eng 1984;A65:145–56. [4] Hsu CJ, Kao PW, Ho NJ. Ultrafine-grained Al–Al2Cu composite produced in situ by friction stir processing. Scr Mater 2005;53:341–5. [5] Guan YC, Zhou W, Li ZL, Zheng HY. Study on the solidification microstructure in AZ91D Mg alloy after laser surface melting. Appl Surf Sci 2009;255:8235–8. [6] Tsau CH, Chen YC. The coarsening of the precipitates in melt-spun Al–Ti–Zr ribbons. Mater Chem Phys 2002;73:111–7. [7] Su~nola JJ, Saurina J, Bruna P. Structural and thermal changes induced by mechanical alloying in a Fe–Ni based amorphous melt-spun alloy. Mater Chem Phys 2009;114:996–9. [8] Sieber H, Park JS, Weıssmüller J, Perepezko JH. Structural evolutıon and phase formatıon ın cold-rolled alumınum–nıckel multılayers. Acta Mater 2001;49:1139–51. [9] Pope DP, Darolia R. In: Westbrook JH, editor. Applications of intermetallic compounds. MRS bulletin, vol. 21; 1996. p. 26. [10] Colin J, Serna S, Campillo B, Flores O, Juarez-Islas J. Microstructural and lattice parameter study of as-cast and rapidly solidified NiAl intermetallic alloys with Cu additions. Intermetallics 2008;16:847–53. [11] Yong LH, Hun JS, Yong LS, Hyun KK. Alloying of cold-sprayed Al–Ni composite coatings by post-annealing. Appl Surf Sci 2002;253:3496–502. [12] Rajabi M, Simchi A, Vahidi M, Davami P. Effect of particle size on the microstructure of rapidly solidified Al–20Si–5Fe–2X (X = Cu, Ni, Cr) powder. J Alloys Compd 2008;466:111–8. [13] Meyer M, Mendoza-Ze´lis L, Sa´nchez FH, Clavaguera-Mora MT, Clavaguera N. On the mechanism of milling-induced disordering in AlFe. Phys Rev B 1999;60:3206–12. [14] Karpe N, Larsen KK, Bottiger J. Phase formation induced by ion irradiation and electrical resistivity of aluminum-3d-transition-metal alloys. Phys Rev B 1992;46:2686–92. [15] Spinelli JE, Rosa DM, Ferreira IL, Garcia A. The effect of melt temperature profile on the transient metal/mold heat transfer coefficient during solidification. Mater Sci Eng 2004;A408:317–25. [16] Takeda M, Maeda Y, Yoshida A, Yabuta K, Konuma S, Endo T. Discontinuity of GP(I) zone and h00 -phase in an Al–Cu alloy. Scr Mater 1999;41:643–9. [17] Karaköse E, Keskin M. Morphological characteristic of the conventional and melt-spun Al–10Ni–5.6Cu (in wt.%) alloy. Mater Charact 2009;60:1569–77. [18] Osorio WR, Spinelli JE, Ferreira IL, Garcia A. The roles of macrosegregation and of dendritic array spacings on the electrochemical behavior of an Al– 4.5 wt.%Cu alloy. Electrochim Acta 2007;52:3265–73. [19] Sahoo KL, Wollgarten M, Haug J, Banhart J. Effect of La on the crystallization behavior of amorphous Al94–XNi6LaX (x = 4–7) alloys. Acta Mater 2005;53:3861–70. [20] Kim TS, Hong SJ, Lee BT. Hardness behavior of the partially crystallized amorphous Al86Ni9Mm5 alloys. Mater Sci Eng 2003;A363:81–5. [21] Ünlü N, Genç A, Öveçog˘lu ML, Lavernia EJ, Froes FH. Microstructural evolution of the melt-spun ternary hypoeutectic Al–7.6Si–3.3Fe (in wt%) alloy. J Alloys Compd 2002;343:223–33. [22] Fiellstedt J, Fredriksson H. On the crystallization process of hypoeutectic Al– %6Cu, unmodified and Sr modified Al–%2Si solidified alloys. Adv Eng Mater 2003;5:1–2. [23] Kim CH, Ree T. Activation enthalpies for plastic deformation. Bull Kor Chem Soc 1980;1:39–43. [24] Muller S, Wang LW, Zunger A. Coherent phase stability in Al–Zn and Al–Cu fcc alloys: the role of the instability of fcc Zn. Phys Rev B 1999;60:16448–62. [25] Lee JM, Kang SB, Sato T, Tezuka H, Kamio A. Evolution of iron aluminide in Al/ Fe in situ composites fabricated by plasma synthesis method. Mater Sci Eng 2003;A362:257–63. [26] Zeng Q, Baker I. The effects of local versus bulk disorder on the magnetic behavior of stoichiometric Ni3Al. Intermetallics 2007;15:417–27. [27] Jones H. Ultrarapid quenching of liquid alloys, treatise on materials science and technology. New York: Academic; 1981. p. 76. [28] Burden MN, Jones H. Determination of cooling rate in splat-cooling from scale of micro-structure. J Inst Metals 1970;98:249.
E. Karaköse, M. Keskin / Materials and Design 32 (2011) 4970–4979 [29] Perepezko JH. Solidification reactions in undercooled alloys. Mater Sci Eng 1994;A179–180:52. [30] Matyja H, Giessen BC, Grant NJ. The effect of cooling rate on the dendrite spacing in splat-cooled aluminium alloys. J Inst Metals 1968;96:30. [31] Rios CT, Bolfarini C, Botta WJ, Kiminami CS. Rapidly solidified Al92Fe3Cr2Mn3 alloy. Mater Sci Eng 2007;A449–451:1057–61. [32] Hsu CJ, Kao PW, Ho NJ. Intermetallic-reinforced aluminum matrix composites produced in situ by friction stir processing. Mater Lett 2006;61:1315–8. [33] Sheng LY, Zhang W, Guo JT, Wang ZS, Ye HQ. Microstructure evolution and elevated temperature compressive properties of a rapidly solidified NiAl– Cr(Nb)/Dy alloy. Mater Design 2009;30:2752–5. [34] Sheng LY, Guo JT, Tian YX, Zhou LZ, Ye HQ. Microstructure and mechanical properties of rapidly solidified NiAl–Cr(Mo) eutectic alloy doped with trace Dy. J Alloys Compd 2009;475:730–4. [35] Sheng LY, Guo JT, Ye HQ. Microstructure and mechanical properties of NiAl– Cr(Mo)/Nb eutectic alloy prepared by injection-casting. Mater Des 2009;30:964–9.
4979
[36] Kaya H, Çadırlı E, Böyük U, Marasßlı N. Variation of microindentation hardness with solidification and microstructure parameters in the Al based alloys. Appl Surf Sci 2008;255:3071–8. [37] Takeda M, Komatsu A, Ohta M, Shirai T, Endo T. The influence of Mn on precipitation behavior in Al–Cu. Scr Mater 1998;39:1295–300. [38] Audebert F, Mendive C, Vidal A. Structure and mechanical behaviour of Al–Fe– X and Al–Ni–X rapidly solidified alloys. Mater Sci Eng 2004;A375– 377:1196–200. [39] Shaw L, Luo H, Villegas J, Miracle D. Effects of internal strains on hardness of nanocrystalline Al–Fe–Cr–Ti alloys. Scr Mater 2004;51:449–53. [40] Gogebakan M, Uzun O, Karaaslan T, Keskin M. Hardness and microstructural characteristics of rapidly solidified Al–8–16 wt.%Si alloys. J Mater Proc Technol 2003;142:87–92. [41] Fernandes JC, Anjinho C, Amaral PM, Rosa LG, Rodriguez J, Martinez D, et al. Characterisation of solar-synthesised TiCx (x = 0. 50, 0.625, 0.75, 0.85, 0.90 and 1.0) by X-Ray diffraction, density and Vickers microhardness. Mater Chem Phys 2002;77:711–8.