HIGH TEMPERATURE MECHANICAL PROPERTIES OF Al-Al4C3 COMPOSITE PRODUCED BY MECHANICAL ALLOYING

A method of mechanical alloying process is describe d. It was proved that the transformation efficiency of carbon to Al 4C3 by heat treatment of aluminium with the porous furn ace black and electrographite at temperatures 723 – 873 K and tim e periods 1 10 hours is higher than that of the hard cracked graphite. The size of Al 4C3 dispersed phase was measured on thin foil and it was constant and as small as 30 nm. Subgrain size m easured in the range of 100 grains in thin foils depended on the carbon type.. It ranged from 0.3 to 0.7 μm. Mechanical properties were analyzed in a microstructural matrix (after extrusi on) as well as in nano-matrix The temperature dependence of ductility, and reduction of area in t he temperature range of 623 – 723 K and strain rate of 10 s, indicated a considerable increase of properties. In a case when the volume fraction of Al4C3 changes from lower to higher, the grain rotation m echanism dominates instead of the grain boundary sliding. Creep parameters for Al-Al 4C3 systems and commercial IN9021 and IN9052 systems were compared. The dependence of the minimum deflection rate on the applied force as well as the dependence of the time o fracture on the applied force for two temperature levels (623 and 723 K) by small punch t esting is depicted. The anisotropy of the creep properties and fracture using small punch tes ts for the Al-Al4C3 system produced by ECAP were analyzed.


Introduction
Since strengthening of aluminium alloys by (noncoherent) Al 4 C 3 particles can be expected to have great potentiality for increasing their high temperature mechanical properties, creep and creep fracture in such alloys have been investigated extensively [1,2].Aluminium alloys strengthened by Al 4 C 3 particles are produced by mechanical alloying and/or reaction milling DOI 10.12776/ams.v20i3.307p-ISSN 1335-1532 e-ISSN 1338-1156 followed by consolidation and hot extrusion.The alloys are typically fine grained (the mean grain diameter is always less than 1 μm) and always contain a volume fraction of Al 2 O 3 particles besides that of Al 4 C 3 , [3].The particles size is typically 20 nm.Mechanical alloying is the process of production of macroscopically homogeneous materials from heterogeneous mixtures.The process is implemented in laboratories in attritors, on the industrial scale in high energy ball mills.It is based on deformation, repeated disintegration and welding of power particles during intensive dry milling [4,5].Originally, this technology was developed for production of Ni and Fe superalloys.Later, it was shown that mechanical alloying is capable of producing several stable and metastable phases, including solid solutions, metastable crystalline and quasicrystalline phases, as well as amorphous phases.It is remarkable that the crystalline phases are usually of the nanometric size.Because these effects are similar to other ones, obtained by nonequilibrium techniques, such as rapid melt solidification, the system is characterized by increased mechanical properties.The advantage of preparing amorphous materials by mechanical alloying in comparison with rapid solidification technology is the fact that it allows larger quantities of materials to be produced, and the possibilities for alloying have therefore been extended.Mechanical alloying technology is now applied to different types of material systems, namely metals, ceramic and polymers, and considerable attention is being paid to it throughout the world.In the literature, the term for mechanical alloying is stated as ´mechanical alloying´ if elementary powders are used, and as ´mechanical milling´ if a prealloyed powder is used as the starting material.A series of material systems has been developed using mechanical alloying, mainly for high-temperature applications [6].The aim of this paper is to study the influence of the various graphite types (when mixed with Al powder) and heat treatment procedure on the microstructure, mechanical properties and fracture of dispersion strengthened aluminium alloys in the Al 4 C 3 system with micro and nanostructures.

Experimental material and methods
The experimental material -dispersion strengthened aluminium with Al 4 C 3 particles was prepared by intense milling of aluminium powder with different types of carbon, as shown in Table 1.The prime aluminium powder grain size was 100 µm with the carbon content of 0.6 -3 wt.%.The aluminium composite dispersion strengthened by Al 4 C 3 particles has been prepared by the method of mechanical alloying.The milling time used in all tested specimens was 90 min.The final carbide content was in the range of 2.5 -12 vol.%.The obtained mixture was compacted at 600 MPa and heat treated at 723, 773, 823, and 873 K whereas treatment times of 1, 3, 10, and 30 hours were employed.The final compacting by hot extrusion at a temperature of 823 K and a reduction of 94 % in the cross circular section of extruded bars was applied [7].The experimental material has been prepared at the Institute for Chemical Technology of Inorganic Materials, TU Vienna.
A microstructural material Al-Al 4 C 3 with 4 vol.% of Al 4 C 3 was transformed by the ECAP (equal channel angular pressing) method by two passes into a nanocomposite material.The experimental material was pressed through two right angled (90°) channels of a special die by route "C", [8,9].The ECAP technology allows obtaining the very fine-grained micorstructurenanostructure by multiple pressing through the die, Fig. 1.

Milling and carbonisation kinetics
The dependence of the transformation rate on temperature and holding time for the four carbon types is shown in Fig. 2. The good susceptibility to transformation for porous furnace black (a 1 and a 2 ) and that of electro-graphite (b) is evident.The porous carbon types are incorporated into the matrix by friction during milling, the distribution is uniform, and clustering is small.On the other side, hard graphite (c) resists disintegration and the granules are large.
Light microscopy microstructure analysis of the produced compacts proved a high homogeneity of dispersed particle distribution in the direction perpendicular to the direction of hot extrusion.
In the longitudinal direction of the bar as a result of hot extrusion the Al 4 C 3 carbide particles were arranged into bands.
During the milling of Al-1C system with the carbon type I microstructure changes occurred in the fracture processes and welding of the particles with incorporation of C phase and forming of final granules.The following micromechanisms, which take place during mechanical alloying, have been defined as the decisive ones: 1. Plastic deformation and deformation strengthening of the starting Al powders and their transformation to Al lamellae.2. Creation of non-equiaxial aggregates from powder particles of the lamellae by the mechanism of cold welding with the origin of coherence failures being the same.3. Forming and compacting of non-equiaxial aggregates and their transformation to equiaxial ones.4. Exhaustion of deformability of aggregates or granules and their crumbling activated by weak surfaces on their structure.Electron microscopy analysis was conducted using carbon replicas and thin foils.TEM of thin foils offered better results.For all the tested carbon combinations from the A to I labels, thin foils were produced for the heat treatment 723 K/30 h.The Al 4 C 3 particle size and the subgrain size were measured using the thin foils.The dispersed phase Al 4 C 3 particle size was measured on 100 to 200 thin foil structures, and it was constant and as small as 30 nm.The particle size was influenced neither by the carbon type nor by the heat treatment applied.Subgrain size measured in the range of 100 grains in thin foils depended on the carbon type, as well.It ranged from 0.3 to 0.7 µm.The stability of properties, resulting from graphite type I (KS 2.5), led to the highest production and utilization of this type of dispersion strengthening, [10].

Mechanical properties of microstructural system
In our previous works [11][12][13], we have evaluated the distance between the particles by point object simulation methods.This includes the mean interparticle distance λ µ , the mean minimum distance λ ρ , the mean visibility λ ν , and the mean free spherical contact distance λ ϑ .The characteristics and properties of these parameters have been analyzed in [11].The mean free spherical contact distance λ ϑ . is the optimal parameter which has a physical interpretation, too.
During the last years, a new approach to the description of point systems has been developed intensively, which is referred to as polygonal methods [13].The dual representation formed in the above way describes completely the given point system.Properties of Voronoi tessellation and their various generalizations are being very intensively studied now the state of this study is given in the monograph [14].Intermediate stages of evaluation for thin foil (a), outlines of particles (b), and of reference points (c) are documented in Fig. 3.
The Al-Al 4 C 3 system with 4 vol.% of Al 4 C 3 was tested under different tensile conditions, where three different strain rates and different testing temperatures up to 723 K were used [15][16].The results are shown in Fig. 4. The deformation mechanism and fracture mechanism were analyzed corresponding to different testing conditions.For the higher strain rates of 10 -1 s -1 at 723 K, a significant growth of plastic properties was observed.The high uniform elongation A 5 of the specimen gauge length, and corresponding reduction values of the reduction in area Z were manifested in Fig. 5.
Fig. 4 Stress-strain dependences at 723 K (◆ ε& =10 -1 s -1 , □ ε& =10 -3 s -1 , ▲ ε& =2.The ductility anomalies are showing an onset of a type of superplasticity.According to [17][18][19] it was proved that for materials Al -12Al 4 C 3 , the main mechanism responsible for superplastic behaviour is the grains rotation process and not sliding. The comparison of real F-∆l curves for material Al-Al 4 C 3 with 12% Al 4 C 3 is given in Fig. 6.In the curves for 293 and 573 K the first part with deformation strengthening is followed immediately by the loss of plastic stability, plastic deformation localization and fracture.On the other side curves for high strain rate and temperatures of 673 and 723 K showed after part I, a definitely linear part II with near constant true stress, sometimes called "plateau", with dynamic recovery, and only after that the part III, with the loss of plastic stability.The plastic properties for different strain rates are expressed by ductility A 5 in Fig. 7.For low strain rates a decrease of plastic properties is characteristic with the increase of temperature up to 723 K.A peak like significant increase of plastic properties is visible at 673 and 723 K in all plots (Fig. 8) for the highest strain rate ε& = 1•10 -1 s -1 .According to our experience, ductility A 5 is influenced by differences of deformation, and fracture in part III of the test.Cut-outs and thin foil TEM micrographs were produced also from the fracture area.The total values of ductility and reduction of area are much lower than those for known superplastic materials.During deformation of samples with high dispersed phase content at high strain rate, the recovery is not as perfect as known for superplastic materials.Local deformation with neck formation and fracture with limited reduction of area are the signs showing the limits of these materials.We suggest characterizing more precisely the deformation process in the Al-Al 4 C 3 system in terms of quasi-superplastic behaviour.The comparison of the test results and changes in fracture for the Al-Al 4 C 3 system at temperatures from 293 to 673 K and strain rates from ε& = 2.5•10 -5 s -1 to ε& = 1•10 -1 s -1 is summarized in [20].The composition of the alloys investigated is given in Table 2.The alloys IN9021 and IN9052 were commercially produced by NOVAMET Co., USA, and supplied in the form of extruded bars 45 mm in diameter.The Al-Al 4 C 3 alloys were laboratory produced at the Technical University Vienna, and supplied as extruded bars of 6 mm in diameter, [21].[22] holds: ) where C and p are constants; p = 0.99 ± 0.10 for both alloys, too [23].The fact that Q f < Q c and m f < m c is related to the temperature and stress dependence of the strain to fracture ε c .Creep behaviour of the composite based on aluminium matrix, reinforced by 4 vol.% Al 4 C 3 was studied at temperatures from 623 to 723 K by small punch (SP) testing with a constant force.SP technique represents one of the recent methods for estimation of mechanical properties of solids [25,26].The time dependence of the central deflection was registered and the minimum deflection rate was determined.The dependence of this quantity on the applied force can be described by a power function with relatively high value of the power.The dependence can be rationalized by an analysis in terms of the threshold concept.Analytical procedure for comparison of the threshold force in small punch experiments and threshold stress in conventional creep testing are given.
The dependence of the minimum deflection rate on the applied force for two temperature levels is given in Fig. 12 in bilogarithmic coordinates.The dependence of the time to fracture, t F on the applied force F is given in Fig. 13.

Mechanical properties of nanostructural system
The material after ECAP is on the border of nanostructural materials.Microstructure of the starting material is in Fig. 14.TEM micrograph, Fig. 15 shows that the mean grain size was 100-200 nm, dislocations are present in nanograins, too, but mostly on the boundaries.In Fig. 15 these dislocations are weakly visible due to the tilting of the specimen.The nanostructure formation takes place most probably by a three-stage mechanism, described in [27].This model has been experimentally verified only for several specific materials but in our case it seems to be probably usable.The model includes creation of cell structure, then formation of transitory cell nanostructure with large angle disorientation, and finally formation of nanostructural grains with size of approx.100 nm.However, here one has to consider the retarding effect due to present dispersoid particles.
The composite was tested in two different states: a) as received by mechanical alloying with hot extrusion (HE) as the final operation and b) with ECAP superimposed on the hot extruded material, [28].The ECAP does not improve the observed creep resistance.The reduction of force leading to the same deflection rate is not very significant.This points out that the ECAP process of the present composite, which produces substantial strengthening at lower temperatures, is not accompanied by pronounced weaking of creep resistance at elevated temperatures.The threshold force in the ECAP material is about 5 N weaker than in the HE material.The dependence of the minimum deflection rate on the applied force for two temperature levels is given in Fig. 16.The dependence can be described by the power-law relationship of the form: and deflection at fracture were different in specimen with different orientation with respect to the axis of ECAP deformation [30,31].A fracture surface analysis of the tested small punch specimens was conducted.Fractures have the transcrystalline ductile character.The fracture dimples are equiaxial as well as elongated in the dependence on strain direction.Fracture dimples are of two categories, the small ones sized from 0.1 to 0.5μm and large ones ranging from 3 to 6μm.

Conclusions
The results can be summarized as follows: 1.It was shown that the transformation efficiency of carbon to Al 4 C 3 by heat treatment of aluminium with the porous furnace black a) and electrographite b) is higher, than that of the hard cracked graphite c).Microstructure changes occurred in the fracture processes and welding the particles with incorporation of C phase and forming final granules.2. The dispersed phase Al 4 C 3 particle size was measured on thin foil structures, and it was constant and as small as 30 nm.The particle size was influenced neither by the carbon type nor by the heat treatment technology applied.Subgrain size measured in the range of 100 grains in thin foils depended on the carbon type, as well.It ranged from 0.3 to 0.7 µm.3. The comparison of the tensile test results and changes in fracture for the Al-Al 4 C 3 system at temperatures from 293 to 673 K and strain rates from ε& = 2.5•10 -5 s -1 to ε& = 1•10 -1 s -1 is summarized.The results are considered relevant to changes in micromechanism of deformation and fracture (transition from transcrystalline ductile fracture to intercrystalline fracture).was studied at temperatures from 623 to 723 K by small punch (SP) testing with a constant force.The dependence of the minimum deflection rate on the applied force as well as the dependence of the time to fracture on the applied force was depicted.6.The very fine-grained microstructure -nanostructure with the mean grain size of 100-200 nm was obtained a material Al-Al 4 C 3 with 4 vol.% Al 4 C 3 transformed by the ECAP technology by two passes.7. The small punch tests of a mechanically alloyed Al composite with a superimposed ECAP procedure were performed in the constant force regime at the temperatures of 623 K and 723 K.The force dependencies of the minimum deflection rate and of the time to fracture are quantitatively comparable with the analogical dependencies of the hot extruded material.The ECAP does not improve the observed creep resistance.The anisotropy of the creep properties and fracture using small punch tests for the Al-Al 4 C 3 system produced by ECAP were analyzed.A fracture surface analysis of the tested small punch specimens was conducted.Fractures have transcrystalline ductile character.The fracture dimples are equiaxial, elongated in the dependence on strain direction.Fracture dimples are of two categories, the small ones sized from 0.1 to 0.5 μm and large ones ranging from 3 to 6 μm.

Fig. 2
Fig. 2 Dependence of carbon to the carbide transformation rate on heat treatment temperature and holding time for four carbon types

Fig. 3
Fig. 3 Intermediate stages of evaluation for thin foil (a), outlines of particles (b), and reference points (c)

Fig. 5
Fig. 5 Reduction of area Z as a function of strain rate and temperature

Fig. 6 Fig. 7
Fig. 6 Comparison of real F-∆l curves of the composite materials Al-Al 4 C 3 for temperatures from 293 to 723 K and strain rate of 10 -1 s -1

Fig. 8
Submicrostructure (foil) of Al-12Al 4 C 3 material, (a) the transverse direction, (b) longitudinal direction a) b) Fig. 9 TEM micrograph of the substructure of Al-4Al 4 C 3 in the transverse direction (a), and in the longitudinal direction (b) a) b) Fig. 10 Supposed reordering mechanisms: a) slip, b) rotation

Fig. 11
Fig. 11 Relation between "mean" creep rate and steady state creep rate for Al-Al 4 C 3 alloys at temperatures 623 K and 723 K

Fig. 12 Fig. 13
Fig. 12 Dependence of the minimum deflection rate on the applied force

Fig. 17
Fig. 17Dependence of the time to fracture on the applied force

4 .
It follows from the model of superplastic deformation, that if the volume fraction of Al 4 C 3 changes from lower to higher values, the grain rotation mechanism dominates instead of the grain boundary sliding.5. Creep behaviour of the composite based on aluminium matrix, reinforced by 4 vol.% Al 4 C 3 DOI 10.12776/ams.v20i3.307p-ISSN 1335-1532 e-ISSN 1338-1156

Table 1
Types of different carbon types

Table 2
Dispersion strengthened aluminium alloys investigated: Nominal composition, grain size d, apparent activation energies of creep Q c and creep fracture Q f , apparent stress exponent of steady state creep rate m c and of time to creep fracture m f Mg and Si in wt.%, Al 4 C 3 and Al 2 0 3 in vol.% + ) Cu, ++ ) after annealing at 773 K for 14,4 ks +++ ) in as received condition In Fig. 11, relations between the "mean creep rate" ε c / t f and steady state creep rate s ε& is shown for the Al-Al 4 C 3 alloys at temperatures 623 K and 723 K; ε c is the creep strain to fracture.It can be seen that the well known Monkman-Grant relation as modified introducing ε c ) where δ & is the minimum deflection rate, F is the acting force and S The times to fracture in ECAPed material are shorter than in the hot-extruded material, though the relative difference is smaller than in the previous case of deflection rate: Assuming the same value of power F n in both materials, the time to facture is about 43 times shorter at 623 K and 14 times shorter at 723 K.The ratio at 723 K should be taken with caution since its force dependence has to be admitted.The anisotropy of the creep properties and fracture using small punch tests for the Al-Al 4 C 3 system produced by ECAP were analysed.Small punch creep tests under constant force were performed at the temperature of 623 K.It was shown that the fracture results, i.e. time to fracture