Journal Information
Vol. 8. Issue 3.
Pages 2969-2977 (May - June 2019)
Download PDF
More article options
Vol. 8. Issue 3.
Pages 2969-2977 (May - June 2019)
Original Article
DOI: 10.1016/j.jmrt.2019.05.004
Open Access
Microstructure homogeneity of milled aluminum A356–Si3N4 metal matrix composite powders
Heydi Fernándeza,
Corresponding author

Corresponding author.
, Stella Ordoñeza, Hector Pesentib, Rodrigo Espinoza Gonzálezc, Matteo Leonid
a Departamento de Ingeniería Metalúrgica, Universidad de Santiago de Chile, Av. Lib. Bdo. ÓHiggins 3363, Estación Central, Santiago, Chile
b Facultad de Ingeniería, Universidad Católica de Temuco, Rudecindo Ortega 02950, Temuco, IX Región de la Araucania, Chile
c LabMAM, Departamento de Ingeniería Química Biotecnología y Materiales, FCFM, Universidad de Chile, Av. Beauchef 851, Santiago, Chile
d DICAM, University of Trento, Mesiano, 38123 Trento, Italy
This item has received

Under a Creative Commons license
Article information
Full Text
Download PDF
Figures (6)
Show moreShow less

A metal matrix composite was produced by co-milling an A356 aluminum alloy powder obtained by rotating electrode off-equilibrium solidification, with different mass fractions (10, 20 and 30%) of Si3N4. The structural and microstructural modifications occurring during the milling were investigated with X-ray powder diffraction (XRPD). Whole powder pattern modeling (WPPM) of the XRPD reveals the inhomogeneous nature of the material in terms of silicon content and allows the crystallite size distribution and dislocation content to be followed in detail for all phases present in the powder. Neither microscopy nor the traditional Scherrer equation can reveal such a detailed picture in this case. Short milling times are sufficient to homogenize the microstructure and to obtain nanoscale crystallites. Long milling times are advantageous to increase the dislocation density that might be favorable for subsequent sintering.

Mechanical milling
Aluminum matrix composite
Microstructure inhomogeneity
Scherrer equation
Full Text

Due to their low density, aluminum and its alloys find many applications in several industrial fields ranging e.g. from aeronautics to automotive [1,2]. The applicability range is however limited by the poor strength and hardness of these materials. This is not just an intrinsic structural limitation, but also the result of the production and forming processes. During the conventional solidification, in fact, several microconstituents form, resulting in macrosegregation and porosity [3]. In non-equilibrium solidification, like in the rotating electrode process (REP), a more refined microstructure with smaller dendritic arm spacing (DAS) is obtained [4].

Much research has been carried out to increase the properties of light alloys. The creation of metal matrix composites (MMCs) is a viable solution, as the toughness of the metal is combined with the hardness and wear resistance of a reinforcing component [5–7]. The best properties are obtained when the reinforcing particles are nano-sized and uniformly distributed in the metal matrix. Several aluminum matrix composites (AMCs) have been proposed in the literature, by dispersing a tough ceramic (e.g. Al2O3, TiC, SiC, B4C, TiB2) in an aluminum or aluminum alloy (e.g. 2014, 6061, 7075) matrix [8–11].

Powder metallurgy (PM) is one of the most common fabrication processes for MMC components, as it can guarantee a better microstructural control and minimizing the interactions between matrix and reinforcement [12]. The MMC powder can be produced using different techniques: mechanical milling is probably the most convenient, as it can assurance a larger homogeneity in the distribution of the reinforcement, low processing temperature and can activate the material for sintering at lower temperatures.

In order to optimize the powder production process, it would be necessary to estimate the effectiveness of the milling and to establish the optimal conditions (especially in terms of milling time), to obtain a homogeneous nanocomposite. Microscopy observation and mechanical characterization would give a direct information on the characteristics of the milled powder. Microscopy should be done however on cross-sectioned powder grains and the mechanical characterization would be limited to micro or nano-hardness testing, being the material in a powdered state. A possible fast and non-destructive alternative is that given by X-ray diffraction: the diffraction pattern bear information on the phases present in the powder, on their structure and on the overall microstructure (size of the crystallites, lattice distortion), which are expected to vary during the milling.

Diffraction can be the only viable alternative for the study of MMCs with a complex microstructure (e.g. hypo or hyper-eutectic alloys). This is the case of the reinforcement of the A356 aluminum alloy powders with silicon nitride Si3N4. The presence of the silicon both in the matrix and in the reinforcement limits the capabilities of microscopy to understand the evolution of the microstructure during the production of the AMC powder. The aim of this work is to evaluate the production of A356–Si3N4 AMC powders by mechanical milling, and in particular to analyze the evolution of the microstructure with the milling time as a function of the reinforcement content.

2Experimental procedure2.1Sample preparation and processing

Aluminum, silicon, and magnesium of commercial purity were employed for the production of the A356 aluminum alloy (Al, 7wt% Si, 0.25% Mg). The solids were molten in an Inductotherm Power-Trak 35-96 induction furnace and the resulting alloy cast into cylindrical sand molds. The cylinders were machined and subsequently atomized by the rotating electrode process (REP) in a glove box under controlled argon atmosphere.

Alpha silicon nitride (Si3N4) Grade M11, manufactured by H.C. Starck, was added to the metal powder in order to obtain A356–Si3N4 composites with different mass fractions (10, 20 and 30%). The powders were prepared in a RETSCH PM400 planetary ball mill using stainless steel jars (500ml) and hardened steel balls (12mm in diameter). The powders were milled for 3, 6, 9, 12 and 15h with a ball/powder weight ratio of 20:1 at a sun wheel speed Ω=360rpm and a planet/sun speed ratio ω/Ω=2:1. Stearic acid (200mg) was used as milling agent. The milling process was discontinuous, consisting of 1h of milling followed by 0.5h of resting, to limit the heating of the jars. The loading and unloading of the jars were carried out in a glove box under controlled argon atmosphere, as to limit the oxidation of the powder.

For conciseness, the specimens will be named x%-nh where x is the quantity of reinforcement and n is the milling time. As an example, 10%-3h represents the powder with 10% Si3N4 milled for 3h and 0%-0h the as-prepared A356 aluminum alloy.

2.2Optical microscopy

For each of the samples, a small quantity of powder was cold-assembled in epoxy resin. After a curing of 24h, the test pieces were polished using wetted 500, 600, 1500 and 2500 sandpaper and finished with 2μm alumina paste. The samples were then characterized in a Leica DM LM/P optical microscope connected to a 5 MP digital camera.

2.3X-ray diffraction measurements and analysis

The materials were investigated by X-ray powder diffraction (XRPD) with a Bruker D8 Advance diffractometer using Bragg Brentano θ–2θ flat-plate geometry and a LynxEye position-sensitive detector. Cu Kα radiation (λCu=0.15406nm) produced at 45kV, 40mA was employed. The patterns were collected in the 10–90° 2θ range with a scan step of 0.02° and a time of 0.2s per step. A Ni-filter was used to reduce background and beta radiation contamination. To align the instrument and to characterize its contribution to the diffraction line profiles, the pattern of the NIST SRM 660a LaB6 line position and line shape standard powder was collected.

Phase identification was performed using the ICDD SIEVE+ software and the ICDD PDF4+ database (release 2018). The whole powder pattern modeling (WPPM) was employed for the analysis of the microstructure and the determination of the unit cell parameters of the phases identified in the powders [13]. The WPPM uses physical models for the microstructure to synthesize the observed profiles and a nonlinear least squares procedure to determine the best set of parameters that reproduce the experimental observation. Several models were tested using the PM2K software implementing the WPPM [14]. The average contrast factor, necessary to determine the dislocation density in the various phases, was determined using the general procedure outlined in Ref. [15]. The data needed for the calculation will be specified on a case-by-case basis.

2.4Transmission electron microscopy analysis

Transmission electron microscopy (TEM) and energy-dispersive X-ray spectroscopy (EDS) analysis were conducted in a FEI Tecnai F20 operated at 200kV in scanning TEM mode. The sample powder was dispersed in isopropanol by ultrasonic stirring for 15min and the dispersion was dropped in carbon-copper grids. In order to remove organic contamination, the grids were exposed to an argon plasma cleaning before microscopy measurements.

3Results and discussion

Phase identification performed on the X-ray powder diffraction patterns of the milled specimens reveals the presence of:

  • a major fcc phase similar to bulk aluminum (ICDD PDF-4+ card 00-004-0787, space group Fm3m);

  • a cubic diamond phase similar to bulk silicon (ICDD PDF-4+ card 00-027-1402, space group Fd3m);

  • hexagonal α-Si3N4 (ICDD PDF-4+ card 04-005-5074, space group P31c);

  • hexagonal β-Si3N4 (ICDD PDF-4+ card 00-033-1160, space group P63/m).

The first two phases can be justified on the basis of the Si–Al phase diagram, showing a limited miscibility of the two elements. Two polymorphs of silicon nitride can be identified in all patterns; the initial powder (guaranteed α-Si3N4 fraction af 90%) is therefore a mixture of the α-Si3N4 and the β-Si3N4 polymorphs.

3.1Microstructural evolution analysis

The traditional tool employed to determine the size of the coherently scattering domains (crystallites) from XRPD data is Scherrer equation [16]:

where Dhkl is the so-called “average crystallite size”, K is a constant depending on the shape and close to 1 [17], λ is the wavelength of the X-rays, βhkl is the full width at half maximum (FWHM) of the peak in rad and θhkl is the position of the peak. The hypotheses under which Eq. (1) was derived are usually quite different from those met in a real specimen and in a modern XRPD pattern. As a consequence, deviations from the true picture of the material are expected. A pseudo-Voigt function is usually employed to fit the most intense peak of the phase under study, in order to obtain, in an easy way, the peak position and FWHM.

As an example, Fig. 1(a) shows the pseudo-Voigt fit of the most intense diffraction peak of the fcc phase for the 10%-3h specimen. The presence of both the Kα1 and Kα2 components, was taken into account. The same operation can be done for all collected data to estimate, via Eq. (1), a trend of size versus milling time. The result for the present specimens is shown in Fig. 1(b).

Fig. 1.

Experimental pattern (black line), pseudo-Voigt fit (red line) and residual (blue line) for the (111) peak of the powder with 10% Si3N4 milled for 3h (a), values of crystallite size of the alloy with 10%, 20% and 30% Si3N4 at different milling times obtained via Scherrer equation using K=0.9 (b) (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).


A progressive reduction of the crystallite size down to an asymptotic limit is observed for the major fcc phase in the powder. The presence of the ceramic reinforcement is to effectively increase the efficacy of the milling.

A similar trend of the crystallite size with the milling time is typically observed in the production of nanostructured materials via mechanical milling (see e.g. [18] or see [19] for a powder similar to those analyzed here). The result is however always obtained by using Eq. (1) i.e. without considering the actual details of the peak shape [20–22]. To understand the possible issues related to this fact, Fig. 2(a) shows the pseudo-Voigt fit of a region containing a nitride peak and the (200) fcc peak of the same pattern previously analyzed in Fig. 1(a). The shape of the experimental peak is clearly different from the model (see the features in the difference plot), but it should in fact provide the same fit quality and result obtained for the (111) reflection. As a matter of fact neither the pseudo-Voigt function nor Scherrer equation are expected to model the diffraction data ab initio, as they are not based on realistic/physical models for the microstructure.

Fig. 2.

Analysis result of the (200) peak of the 10%-3h powder: raw data (black line), model (red line) and residual (blue line). Fit using a pseudo-Voigt curve (a), WPPM using a lognormal distribution (b), a free-form size distribution (c), two fcc phases with independent lognormal distribution (d). WPPM of the pattern using two fcc phases and two nitride phases (e) and corresponding size distributions for the fcc phases (f). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


The WPPM has been developed to allow for the finer details of the microstructure to be modeled based on a one-to-one correspondence between microstructure feature and profile shape (and therefore also FWHM) [13]. The analysis is performed by fitting the whole diffraction pattern with a single model for the specimen microstructure, whose parameters can be tuned until all experimental features are correctly described. Fig. 2(b) shows an example of the modeling performed under the hypothesis that the crystallites are on average spherical and distributed according to a lognormal curve. The presence of defects, under the form of dislocations, was also assumed. The single crystal elastic constants needed to calculate the average contrast factor for the various phases present in the system are not available, being the phases not pure. For the fcc phase, the single crystal elastic constants of Al were considered [23] to calculate the average contrast factor for the primary slip system ½{111}<110> [24]. Albeit more physical than the result of Fig. 2(a), the model is still unable to fully describe the observed evidence: some extra intensity is present on the low-angle tail of the (200) peak. This might be the result of a deviation of the actual size distribution from the lognormal curve, usually taken as standard in the analysis of nano-sized powders. To verify this hypothesis, Fig. 2(c) shows a modeling of the same data by using arbitrarily-distributed spheres (see [25] for details on the model). The tip of the peak is not well modeled; we should remember that the algorithm tries to accommodate simultaneously all peaks in the system and variations in the peak shape at a given angle has to be compatible with the variation observed at all other angles.

It is likely that a distribution of sizes and of cell parameters is simultaneously present. An inhomogeneous microstructure can be predicted on the basis of the Al–Si phase diagram [26] considering that the A356 aluminum alloy is hypoeutectic. The alloy obtained by conventional melting and casting is expected to be composed of a primary dendritic phase (fcc Al-rich), a lamellar eutectic mixture of α-Al and Si, and some Si segregation regions [27]. However, the powder analyzed here is obtained by the rotating electrode process, where solidification occurs out of equilibrium. As a result, the interdendritic spacing is smaller and the quantity of eutectic is greatly reduced. The microstructure is then composed of a large number of primary fcc dendrites supersaturated in Si and a small amount of fine eutectic, as evident in the optical micrograph of Fig. 3.

Fig. 3.

Optical micrograph of the atomized A356 aluminum alloy.


In the unmilled powder, two fcc peaks are therefore expected in the diffraction pattern: one for the primary supersaturated α-phase and one, very small, for the eutectic α-phase. During the milling, the quantity of supersaturated phase reduces, as some of the material starts to move toward the equilibrium composition. As soon as the milling starts, we expect to see a major fcc phase peak corresponding to the primary phase in transition, and a tail (at higher angle) corresponding to the core of the dendrites that needs longer time for a complete Si diffusion. This situation has been effectively modeled by considering two independent fcc phases in the WPPM (cf. Fig. 2(d) for the (200) peak and in Fig. 2(e) for the whole pattern). The two size distributions are shown in Fig. 2(f); as expected, a longer tail (i.e. the presence of larger domains) is observed for the supersaturated phase. The power of the presented method to highlight fine details in the microstructure is evident here and further suggests that Scherrer equation should be employed only qualitatively for the analysis of complex systems.

As the milling time and the reinforcement content increases, the peaks become more symmetrical and a single fcc phase suffices to model data. This is due to the rejection of the solute from the supersaturated phase caused by the energy introduced in the milling process.

The evolution of the crystallite size with milling time for the major fcc α-Al is shown in Fig. 4(a). As an example, Fig. 4(c) proposes the trend of the size distributions for the major fcc phase in the sample containing 30% Si3N4; such plots could be compared directly with TEM results. Unfortunately, TEM cannot provide, in this case, such a size distribution, as the contribution of the various fcc phases (similar composition) could hardly be separated. Due to the presence of the reinforcement, a separation based on the Si content at the scale needed for the analysis would not be viable. A direct comparison between the WPPM and the Scherrer results is in general not possible (see e.g. [28]). In fact, the WPPM result in the true mean size (first moment of the size distribution), whereas the result of Scherrer analysis is, in the best case, the ratio between the third and the second moment of the size distribution. The large differences at low milling time are mainly due to the fact that the Scherrer calculation considers a compound breadth that includes the two different α phases present in the system. At higher milling times, the peak shape differences introduced by an evolving size distribution can be captured by the WPPM resulting in a more physical trend of the data.

Fig. 4.

Mean values of the crystallite size (a), and dislocations density (b) of the A356 aluminum alloy composites with 10%, 20% and 30% Si3N4 at different milling times. Size distributions of the major fcc phase in 30% Si3N4 as a function of milling time (c).


The dislocation density gives a good idea about the effectiveness of the milling process in terms of the possible negative effect of heating (dynamic recovery) that would slow down the storage of mechanical energy under the form of severe deformation. Fig. 4(b) shows the trend of the dislocation density versus milling time for the powders studied here. During the first hours of milling we observe an overall stability in the dislocation density, that starts to increase when the domain size stabilizes. This is in agreement with the finding of [29] in pure Al: at low milling times, the material is subjected to recovery and most of the new grain boundaries that form are low-angle ones (LAGBs). At higher milling times the high-angle grain boundaries (HAGBs), increase in number. The local strain and the effective number of dislocations associated with a HAGB is larger than for a LAGB and as a consequence the observed dislocation density increases more rapidly. The hard reinforcement, acting as extra milling agent, and the smaller domain size (i.e. larger grain boundary area) contribute to the further increase in the dislocation density observed in the specimens containing 30% Si3N4.

3.2Lattice parameter variation

The lattice parameter estimated for the major (primary) fcc phase of 10%-3h and 20%-3h (see Fig. 5(a)) is higher than that of pure Al (a0,Al=0.40494nm, ICDD PDF-4+ card 00-004-0787). It is not possible to estimate the cell parameter for the atomized unmilled powder, as a large number of non-equilibrium fcc phases seem to be present. As already pointed out, the Al–Si system forms a eutectic mixture: the AlSi bonds are therefore weaker than both the AlAl and SiSi bonds (see e.g. [30]). Thus, there is some degree of repulsion between Al and Si. As a consequence, the lattice parameter of Al is expected to increase when Si is added to it, even if the rule of mixtures based on the atomic radius would indicate otherwise. The supersaturation of the fcc α-Al phases in the system is the cause for the excessively larger lattice parameter of the starting powder, which reduces during the milling process. This diminution is associated with a gradual stabilization of the fcc α-Al phase due to the rejection of solute, till the solubility limit is reached. It is therefore expected that the lattice parameter reaches an asymptotic value as the milling proceeds. As a further source of cell parameter decrease, the increase in the free volume caused by the introduction of vacancies during mechanical milling, has also to be considered [18].

Fig. 5.

Values of Al lattice parameter of the A356 aluminum alloy composites with 10%, 20% and 30% Si3N4 at different milling times (a), relative change in calculated lattice parameter (closed symbols) in comparison to experimental values (open symbols) during milling of A356–Si3N4 composites (b).


The initial decrease of the lattice parameter is lost when a higher content of reinforcing particles is added to the system. The (hard) reinforcing particles enhance the milling process by increasing the amount of fracture and the local deformation exerted on the metal matrix. As a result, similar milling effects are obtained in a shorter time.

As can be seen in Fig. 5(a), the fcc α-Al lattice parameter tends to increase for all alloy composites after a certain milling time. As stated by Zhang [31], the accumulation of vacancies in the powder particles increases dramatically at the initial stage of milling, until it reaches a maximum value. After that, a higher dislocation density sink causes faster vacancy annihilation rate. As result, the Al lattice parameter increases. The literature proposes contrasting results for the variation of the lattice parameter in milled Al alloys [32–34] at high milling times.

3.3Lattice parameter dependence of the crystallite size

A non-monotonic variation of the lattice parameter, similar to the one found here, has been reported e.g. by Rane et al. [35] and by Srivastav et al. [36]. A mathematical model has been proposed to interpret the observed variations in the cell parameter as a function of the microstructure. The authors found that the lattice parameter increases when the excess free volume increases and decreases when the enhanced interfacial energy with the grain size is reduced. A critical grain size exists at which the reversal of the trend occurs.

In order to determine the relative lattice parameter change Δa/a0 with decreasing crystallite size D, the following Eq. (2) was used [35]:

where Δa1 is the effect of interface stress, Δa2 is the effect of excess free volume at the grain boundaries, ξ is the grain-boundary thickness, a0 is the lattice parameter of the undistorted coarse-grained material, σs is the surface/interface tension, K is the bulk modulus and ΔV is the excess free volume at the grain boundary, ΔV=((D+ξ/2)2−D2)/D2. For the calculation we used a grain boundary thickness ξ=1nm [35], a bulk modulus K=75.2GPa [37] and σs=0.96 J/m2[38] and the WPPM average sizes.

Except for the values at high milling times, the calculated values exhibit a poor correspondence with the experimental data (cf. Fig. 5(b)). The model proposed in Ref. [35] is perhaps too simple in this case, as it was developed for pure elements and considers only the effects of excess free volume at the grain boundaries and interface stress. The system under study, on the contrary, contains multiple interacting phases that change with the milling time and a ceramic reinforcement. A simple model considering all those factors cannot be easily built on diffraction data alone.

3.4Morphological characterization of the composites A356–Si3N4

Representative TEM images of the A356–Si3N4 composites particles at 6h of milling are depicted in Fig. 6. The punctual STEM analysis shows that it was possible to identify the Si3N4 particles embedded in the A356 alloy for the three compositions. This is indicated in the images by the little circles that correspond to the adjacent EDX spectra (EDX 1 and EDX 2). The EDX 1 point corresponds to the reinforcement particle (Si3N4) in which the high Si content and the presence of N in the EDX spectra is characteristic. On the contrary, the EDX 2 point corresponds to the matrix (A356 alloy). The slight amount of Fe present in most of the EDX analysis, is due to contamination from the milling process. Careful observations allow to conclude that in the three composites, the morphologically similar reinforcing particles are close to 100nm in size and homogeneously dispersed embedded in the matrix.

Fig. 6.

TEM images of the alloy with 10% (a), 20% (b) and 30% (c) of Si3N4 at 6h of milling.


An aluminum matrix composite was obtained by co-milling a A356 aluminum alloy powder – obtained by means of the rotating electrode process – with Si3N4 particles. The whole powder pattern modeling of the diffraction data shows a rather complex microstructure that evolves rapidly in the first hours of milling. In the early stages of milling the inhomogeneous distribution of silicon resulting from the production is gradually leveled out. The crystallite size reduces rapidly with the milling to an asymptotic value. When a homogeneous composition is obtained, the crystallite size remains approximately constant and the dislocation density starts to increase. The trend in the average lattice parameter can hardly be explained in terms of the available models that consider just grain boundary energy and free volume increase with decreasing domain size. The reinforcing particles act as milling agents: when the amount increases, the refinement of the microstructure is faster. The reinforcing particles are embedded in the matrix and homogeneously dispersed in the composites.

Conflicts of interest

The authors declare no conflicts of interest..


The authors would like to thank for the financial support to “Fondo Nacional Desarrollo Científico y Tecnológico de Chile”, FONDECYT project No. 1150268 and CONICYT, Doctorado Nacional 2015 Ph.D. Grant.

H. Arik.
Production and characterization of in situ Al4C3 reinforced aluminum-based composite produced by mechanical alloying technique.
Mater Des, 25 (2004), pp. 31-40
D. Ge, M. Gu.
Mechanical properties of hybrid reinforced aluminum based composites.
Materials (Basel), 49 (2001), pp. 334-339
S. Ordóñez, O. Bustos, R. Colás.
Thermal and microstructural analysis of an A356 aluminium alloy solidified under the effect of magnetic stirring.
Int J Met, 3 (2009), pp. 37-41
L.Y. Zhang, Y.H. Jiang, Z. Ma, S.F. Shan, Y.Z. Jia, C.Z. Fan, et al.
Effect of cooling rate on solidified microstructure and mechanical properties of aluminium-A356 alloy.
J Mater Process Technol, 207 (2008), pp. 107-111
A. Kostka, J. Lelatko, M. Gigla, H. Morawiec, A. Janas.
TEM study of the interface in ceramic-reinforced aluminum-based composites.
Mater Chem Phys, 81 (2003), pp. 323-325
K. Hiraga, M. Mabuchi, K. Higashi.
Effects of addition of magnesium on interface structure and high-strain-rate composites.
Acta Mater, 46 (1998), pp. 6009-6020
H. Arik.
Effect of mechanical alloying process on mechanical properties of a-Si3N4 reinforced aluminum-based composite materials.
Mater Des, 29 (2008), pp. 1856-1861
W.S. Lee, W.C. Sue, C.F. Lin.
The effects of temperature and strain rate on the properties of carbon-fiber-reinforced 7075 aluminum alloy metal–matrix composite.
Compos Sci Technol, 60 (2000), pp. 1975-1983
B.L. Zhang, M.S. Maclean, T.N. Baker.
Hot deformation behaviour of aluminium alloy 6061/SiCp MMCs made by powder metallurgy route.
Mater Sci Technol, 16 (2000), pp. 897-902
G. Govender, L. Ivanchev, H. Burger, A. Mulaba, H. Chikwande.
Evaluation of HEBM mechanical alloying of Al2O3—356/7075 powder mixture.
Open J Compos Mater, 02 (2012), pp. 48-53
B.V. Ramnath, C. Elanchezhian, T.S.A. Atreya, V. Vignesh.
Aluminum metal matrix composites – a review.
Rev Adv Mater Sci, 38 (2014), pp. 55-60
C. Edil da Costa, F. Velasco López, M. Torralba Castelló.
Materiales compuestos de matriz metálica. II parte. Métodos de procesado y consolidación de MMCs reforzados con partículas.
Rev Metal, 36 (2000), pp. 193-197
P. Scardi, M. Leoni.
Whole powder pattern modelling.
Acta Crystallogr Sect A, 58 (2002), pp. 190-200
M. Leoni, T. Confente, P. Scardi.
PM2K: a flexible program implementing whole powder pattern modelling.
Zeitschrift Für Krist Suppl, 23 (2006), pp. 249-254
J. Martinez-Garcia, M. Leoni, P. Scardi.
A general approach for determining the diffraction contrast factor of straight-line dislocations.
Acta Crystallogr Sect A, 65 (2009), pp. 109-119
P. Scherrer.
Bestimmung der Größe und der inneren Struktur von Kolloidteilchen mittels Röntgenstrahlen. Nachrichten von Der Gesellschaft Der Wissenschaften Zu Göttingen, Math Klasse n.d..
(1918), pp. 98-100
J.I. Langford, A.J.C. Wilson.
Scherrer after sixty years: a survey and some new results in the determination of crystallite size.
J Appl Crystallogr, 11 (1978), pp. 102-113
C. Suryanarayana.
Mechanical alloying and milling.
Prog Mater Sci, 46 (2001), pp. 1-184
S. Ordoñez, P. Palominos, F. Martínez, H. Fernández, O. Bustos, J. Lisboa.
Evolución microestructural de una aleación de aluminio A356 con y sin reforzamiento de SiC sometida a molienda mecánica..
Matéria(Rio de Janeiro), 23 (2018),
J. Bystrzycki, T. Czujko, R.A. Varin, D. Oleszak, T. Durejko, W. Darlewski, et al.
Nanocrystalline hydrogen absorbing Mg–Ni alloy processed by mechanical (ball) milling.
Rev Adv Mater Sci, 5 (2003), pp. 450-454
S. Rajeshkanna, O. Nirmalkumar.
Synthesis and characterization of Cu nanoparticle using high energy ball milling route and compare with Scherrer equation.
Int J Sci Eng Res, 2 (2014), pp. 30-35
T. Mahalingam, C. Selvakumar, E. Ranjith Kumar, T. Venkatachalam.
Structural, optical, morphological and thermal properties of TiO2–Al and TiO2–Al2O3 composite powders by ball milling.
Phys Lett Sect A Gen Atom Solid State Phys, 381 (2017), pp. 1815-1819
R.F.S. Hearmon.
Temperature dependence of the elastic constants of aluminum.
Solid State Commun, 37 (1980), pp. 915-918
P. Scardi, M. Leoni, M. D’Incau.
Whole powder pattern modelling of cubic metal powders deformed by high energy milling.
Zeitschrift Für Krist, 222 (2007), pp. 129-135
M. Leoni, P. Scardi.
Nanocrystalline domain size distributions from powder diffraction data.
J Appl Crystallogr, 37 (2004), pp. 629-634
T.B. Massalski, H. Okamoto, A.S.M. International, P.R. Subramanian, L. Kacprzak.
Binary alloy phase diagrams.
ASM International, (1990),
M. Merlin, G.L. Garagnani.
Mechanical and microstructural characterisation of A356 castings realised with full and empty cores.
Metall Sci Technol, 27 (2009), pp. 21-30
P. Scardi, M. Leoni, R. Delhez.
Line broadening analysis using integral breadth methods: a critical review.
J Appl Crystallogr, 37 (2004), pp. 381-390
S.S. Razavi-Tousi, J.A. Szpunar.
Microstructural evolution and grain subdivision mechanisms during severe plastic deformation of aluminum particles by ball milling.
Philos Magn, 95 (2015), pp. 1425-1447
D.R. Gaskell.
Introduction to the thermodynamics of materials.
fifth edition, Taylor & Francis, (2003),
B.Q. Zhang, L. Lu, M.O. Lai.
Evolution of vacancy densities in powder particles during mechanical milling.
Phys B Condens Matter, 325 (2003), pp. 120-129
D. Jeyasimman, S. Sivasankaran, K. Sivaprasad, R. Narayanasamy, R.S. Kambali.
An investigation of the synthesis, consolidation and mechanical behaviour of Al 6061 nanocomposites reinforced by TiC via mechanical alloying.
Mater Des, 57 (2014), pp. 394-404
D. Jeyasimman, K. Sivaprasad, S. Sivasankaran, R. Ponalagusamy, R. Narayanasamy, V. Iyer.
Microstructural observation, consolidation and mechanical behaviour of AA 6061 nanocomposites reinforced by γ-Al2O3 nanoparticles.
Adv Powder Technol, 26 (2015), pp. 139-148
D. Jeyasimman, K. Sivaprasad, S. Sivasankaran, R. Narayanasamy.
Fabrication and consolidation behavior of Al 6061 nanocomposite powders reinforced by multi-walled carbon nanotubes.
Powder Technol, 258 (2014), pp. 189-197
G.K. Rane, U. Welzel, S.R. Meka, E.J. Mittemeijer.
Non-monotonic lattice parameter variation with crystallite size in nanocrystalline solids.
Acta Mater, 61 (2013), pp. 4524-4533
A.K. Srivastav, N. Chawake, B.S. Murty.
Grain-size-dependent non-monotonic lattice parameter variation in nanocrystalline W: the role of non-equilibrium grain boundary structure.
Scripta Mater, 98 (2015), pp. 20-23
S. Raju, K. Sivasubramanian, E. Mohandas.
The high temperature bulk modulus of aluminium: an assessment using experimental enthalpy and thermal expansion data.
Solid State Commun, 122 (2002), pp. 671-676
R.C. Cammarata.
Surface and interface stress effects in thin films.
Prog Surf Sci, 46 (1994), pp. 1-38
Copyright © 2019. The Authors
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.