A6 Nanoparticle Reinforced Al Casting Alloys Cecilia Borgonovo, Hao Yu Report No.

09-01

Table of contents 1. Why nanocomposites? 2. The promise of nanocomposite aluminum structural components 2.1 Mechanical properties 2.2 High-temperature properties 3. Conventional manufacturing processes 3.1 Introduction 3.2 Liquid-state techniques 3.3 Solid-state processing 4. Selected manufacturing processes 4.1 Ultrasonic cavitation based solidification 4.2 Electromagnetic stirring solidification 4.3 In-situ chemical reactions 5. Proposed approach at ACRC 5.1 Electromagnetic stirring solidification 5.2 In-situ chemical reactions 6. Bio-sketches (Hao and Cecilia)

1
 


1. Why Nano-Composites? The term "Metal Matrix Nano-Composites (MMNCs)" broadly refers to a composite system which is based on metal or alloy substrate, combined with metallic or non-metallic nano-scale reinforcements. The main advantages of MMNCs include excellent mechanic performance, high working temperature, wear resistance, low creep rate and etc. MMNCs are broadly used in aerospace industry and other high technology fields. In the past years, MMNCs have been extensively studied, especially the fabrication methods. A variety of new manufacturing processes, such as semi-solid casting and spray deposition, promoted the developments of MMNCs, lowered the cost of fabrication and enabled the applications of MMNCs extended from high technology to automobile industry.

Improved Properties Metal matrix composites (MMCs) such as continuous carbon or boron fiber reinforced aluminum and magnesium,and silicon carbide reinforced aluminum have been used for aerospace applications due to their lightweight and tailorable properties[cite]. In the past few decades, there is an increasing interest to produce Metal Matrix Nano-composite that incorporate nanoparticles for structural applications, as these materials show better performances as compared to composites with micron-sized reinforcements. For example, the tensile strength of a 1 vol.%SI3N4(10nm)-Al composite has been found to be comparable to that of a 15 vol.%SiCp(3.5µm)-Al composite, with the yield stress of the nano-metric MMC being significantly higher than that of the micro-metric MMC [1].

Low Cost Solutions The key to realize the commercial application of MMNCs is the need to find out one effective and low cost method to produce these materials. Most of the existent fabrication routes involves the use of powder metallurgy techniques, which are not only high cost, but also result in the presence of porosity and impurity. Solidification processing methods, such as squeeze casting, stir mixing and pressure infiltration have superiority over the other processes in rapidly and low cost producing large and complex near-net shape components [2]. However, less effort has been taken in this area in the fabrication of nano-composties. In this report, the advantages of MMNCs are explored and a variety of fabrication methods of MMNCs are introduced and compared. Solidification processing of MMNCs assisted with electromagnetic stirring and in situ method are elaborated in this report. 2
 


References 1. Y.C. Kang, S.L. Chan, Materials Chemistry and Physics, Vol.85, (2004), p438. 2. L.Fischer, Literature Survey Report: Nano-Dispersion Strengthening of Aluminum, Introduction to research, 2004, University of Colorado

2. The promise of nanocomposite Al structural component 2.1 Mechanical Properties The study of metal matrix composite dates back to the 1960s'. In the year of 1963, research on the W/Cu composite was reported, followed by the research on SiC/Al, Al2O3/Al composite [1]. In 1978, US reported the application of B/Al composite on Columbia space shuttle [2]. In 1982, Toyota [3] reported the application of Al2O3.SiO2/Al composite in the fabrication of piston, which is the first attempt to realize the civil application of metal matrix composite. After that, the research on metal matrix composite has obtained more and more interest from the scientists, a large quantity of new metal matrix composites, such as C/Cu, C/Mg, SiCw/Al composite were investigated and developed, many of which have come into commercial application in Aerospace and automobile industries. Particle reinforced metal matrix composite offer attractive advantages, such as simple fabrication technology, low equipment requirements, in favor of mass production and relatively low cost, which enables being extensively studied in the past few years. The main matrix systems, which are being explored include Aluminim matrix, Magnesium matrix, Copper matrix, Zinc matrix and Titanium matrix, the reinforcements include SiC, Al2O3, TiC, TiB2 and B4C. Over the past two decades aluminum matrix composites have been used in numerous structural, non-structural and functional applications of different engineering sectors. It has been proved that they can lead to high performances, and to economic and environmental benefits. The compelling need to achieve superior properties combined with top level results in weight saving, implies the massive use of these materials in transport sector and especially in diesel engines components. Aluminum matrix composites provide outstanding mechanical properties, especially high specific strength and high specific modulus. Reinforced particles and fibers, such as carbon fibers and SiC particles, have extremely high strength and modulus. For instance, the maximum strength of carbon fiber, whose density is 1.85 g/cm3, could be 7000 MPa, which is ten times of the strength of aluminum alloys. Moreover, the modulus of graphite fibers could be as high as 900 GPa. A very small amount of added reinforcements could significantly increase the strength and modulus of Aluminum composites.

3
 


The reinforcements of Aluminum composite involve particle size ranging from around 10 nm up to 500 μm. Composite with a dispersion of particles in the range of 10 nm to 1 μm are termed "nano-composites". Much research has been conducted on aluminum composites, however, the majority of the work has focused on micro-metric particle dispersions which are easier to achieve than nano-sized particle distribution, but less effective in strengthening [4]. Aluminum matrix nano-composites offer significant opportunities for developing structural materials with combined physical and mechanical properties which can’t be achieved by monolithic alloys or micro-sized reinforcements. Aluminum nano-composite exhibits attractive mechanical properties, such as high yield and ultimate strength, creep and fatigue resistance. Especially, nano-composite can overcome drawbacks that cannot be avoided by micro-composite, such as low ductility. The manner by which the SiC particles affect the tensile strength of the aluminum alloy composite can be best described in terms of increased work hardening [5]. To strengthen a material by the method of macroscopically speaking, plastic deformation (in which the nano-scopic effect increases the material's dislocation density) is defined as work hardening. More dislocations will be prevented from nucleating (which resists the dislocation development) as the material becomes saturated with new born dislocations. The strengthening then will be observed with the process of resistance effecting on plastic deformation. The defects in the form of dislocations (which created by fluctuations in stress fields in the materials culminating in a lattice rearrangement as the dislocations propagate through the lattice) often carry out irreversible deformation on a microscopic scale in metallic crystals. Annealing does not annihilate the dislocations at normal temperature. While the dislocations accumulate, interact with each other and serve as pinning points or blocks that impede the motion as well. The result is an increase in the yield strength of the material and decrease in ductility. Dislocations, which surrounded by strained bonds are defined as line defects in a material’s crystal structure in materials science. This is the reason why these bonds break first during plastic deformation. In terms of the thermodynamics, the crystal tends to lower their energy through bond formation between constituents of the crystal. This also explains the interaction in between the dislocations and atoms. It leads a lower but energetically favorable energy conformation of the crystal. Dislocations are just vacancies in the host medium which does exist, this feature makes them “negative-entity” in that they do not exist. As a result, the atoms or ions do not move a lot. Instead, the motion in a bonding pattern of largely stationary atoms is much greater. Characterization of the strained bonds around a dislocation is done by lattice strain fields. As an example, in the compressive strain fields and tensile strain fields, the compressively strained bonds are directed next to an edge dislocation and strained bonds beyond (in tension) the end of an edge dislocation. It is interesting to find out that strain fields can be compared to electric fields

4
 


in certain ways. For example, the strain fields of dislocations follow the rule of attraction and repulsion. As speaking of the deformation, microscopic dislocation motion results in visible plastic deformation. Work hardening can be measured by the increase in the number of dislocation. When work and energy are done on a material, plastic deformation will occur. Furthermore, large amounts of fast applied energy not only moves existing dislocation, but also produce a great number of new dislocations by jarring and working the substance sufficiently. In a cold-worked material, the yield strength is increased. By using the lattice strain fields, the movement of dislocations is shown to be hindered by the surrounding filled with other dislocations. Once dislocation motion is hindered, plastic deformation will not be able to occur at normal stresses. When stress is applied just over the yield strength of non-cold-worked material, elastic deformation will be the only method for cold-worked material to continue to deform, with the normal scheme of stretching or compressing of electrical bond continue to occur and unchanging the elastic modulus. At last, to overcome the strain-field interactions and plastic deformation resumes, the stress should be large enough. On the other hand, decreasing of the work-hardened material ductility will occur. Ductility is the limitation of a certain material that can perform plastic deformation, in another word, the extent of a material can be plastically deformed before fracture. While a cold-worked material has already been extended through part of its allowed plastic deformation. If dislocation accumulation hinders the dislocation motion and plastic deformation, and at the same time, stretching of electronic bonds and elastic deformation cannot be extended any more, fracture will be the third mode of deformation. The following equation reflects the dependency of the dislocation stress (T) on the shear modulus (G), the lattice constant (b) and the dislocation density (

).

Where τ0 is the intrinsic strength of the material with low dislocation density and α is a correction factor specific to the material.

5
 


According to the equation and graph above, work hardening depends half root on the number of dislocations. Either high levels of dislocation or no dislocation will make the material exhibits high strength. Typically, low strength will be the result if the dislocation level is moderate.

2.2 High temperature properties Dispersed ceramic particles are stable at temperatures up to the melting temperature of the matrix metal and do not tend to coarsen at elevated temperatures, meaning there is very little drop off in the dispersion strengthening effect. Fischer[4] shows that the elevated temperature strength of aluminum matrix composite with a micrometric dispersion of Al4C3 particles is considerably better than that of aluminum alloys which have a higher strength at room temperature. In addition to this the it loses very little hardness in annealing at temperatures up to 500ºC, whereas age hardened aluminum alloys can decrease in strength by up to 50%. Yang et al.[6] found that at as the temperature increase, tensile strength and Young's modulus of A356 alloy composite fall rapidly. At 673 K the tensile strength is the 80% less than that at room temperature, and the Young's modulus the 50%. Despite that, the abovementioned properties of the composite are still higher than the unreinforced alloy ones. Samuel [7] found evidence that the reinforced composite has a better stiffness than Al-2014 alloy at all temperature tests. This result is considered to be due to the good bonding boundaries of fibers with the matrix material. Vedani et al. [8] investigated the behavior of different composites. At the lowest stain rate (10-3 /S), the fracture elongation of the composites in the temperature range between 450 °C and 500°C reached the same value of the unreinforced alloy (fig. 1).

6
 


Figure 1: tensile strength versus temperature [8] 

Figure 2: fracture elongation versus temperature [6] 

It is noticed that at high temperature all composites are significantly stronger than the conventional Al-2014 alloy. At 350 C° the ultimate tensile strength is the double of the unreinforced 2014, while the yield strength is 60% higher. It can also be seen that above 270 C°, the variation in the dispersoid volume fraction (striped areas fig.3) has little effect on the UTS and YS. Similar results have been attained from Oñoro et al. in [9].

7
 


Figure 3: a) Ultimate Tensile Strength and b) Yield Strength for Al­2014 composite materials [8] 

Experiments on cycle stress and strain pointed out different cycle softening and hardening behavior at different temperatures: from 293 K to 423 K the hardening ability of the composite increase with the temperature, while from 473 K to 573 K a softening behavior, increasing with temperature, has been measured. Srivatsan et al. [10] reported the same results for 2009 alloy reinforced with SiC. Moreover, they observed an increase in elongation with temperature from 27 °C to 150 °C by 7% and an increase in area reduction up to 8%. Therefore, ductility increases with increasing temperature.

References 1. Cao Mao-sheng, Guan Chang-bin, Xu Jia-qiang, Nanomaterials introduction．Haerbin; Haerbin University Press, 2001.97-44. 2. Jiang Huan-lin, Wang Jian-ning, Zhang Jun. Application of nanometer materials. Journal of Qinghai University, 2002, 20(1):34-36． 3. Lai Hua-qing , Fan Hong-xun, Xu Xiang. Metal matrix composites and application on automobile．R and D of Vehicle, 2001, 4:42-44. 4. L.Fischer, Literature Survey Report: Nano-Dispersion Strengthening of Aluminum, Introduction to research, 2004, University of Colorado 5. T.S. Srivatsan, Meslet Al-Hajri, C. Smith, M. Petraroli, The tensile response and fracture behavior of 2009 aluminum alloy metal matrix composite, Materials Science and Engineering, A346 (2003), pp. 91-100. 6. Z.Yang, J. Han, W.Li, W.Wan, S.Kang, Study on fracture behavior of SiCp/A356 composites, RARE METALS, Vol. 25, Spec. Issue (2006), pp .168-174. 8
 


7. M.Samuel, Reinforcement of recycled aluminum-alloy scrap with Saffil ceramic fibers, Journal of Materials Processing Technology 142, 2003, pp. 295–306. 8. M.Vedani, F.D‘Errico, E.Gariboldi, Mechanical and fracture behaviour of aluminium-based discontinuously reinforced composites at hot working temperatures, Composites Science and Technology 66 (2006), pp. 343–349. 9. Oñoro, Salvador, Cambronero, High-temperature mechanical properties of aluminium alloys reinforced with boron carbide particles, Materials Science and Engineering A 499, 2009, pp. 421–426. 10. T.S. Srivatsan, Meslet Al-Hajri, C. Smith, M. Petraroli, The tensile response and fracture behavior of 2009 aluminum alloy metal matrix composite, Materials Science and Engineering, A346 (2003), pp. 91-100.

3. Conventional manufacturing processes 3.1 Introduction The fabrication and study of artificial particulate composites in the form of granular metal can be dated back in the 1960s [1]. In the last decade, the industrial application of nanoscale metal composites has consistently increased, and improvements in composite processing methods are therefore pivotal for enhancing their commercial applicability. So far, many techniques have been employed to synthesize the nano-composite materials. Some of them have been extensively applied, while others are relatively new and require further understanding. The present chapter has been devoted to underline the shortcomings of conventional manufacturing processes, which are the causes of low-quality products or commercial unfeasibility. This analysis has been pursued in order to exclude some of these routes from our investigation, and in the same time to select the processes that are most promising and that will be optimized and modeled in our studies (see Chapter 5).

3.2 Liquid state techniques Liquid metal is generally less expensive and easier to handle than powders, and the shape flexibility constitutes a significant advantage. Liquid state processes are generally fast often easy to be scaled-up. Despite this, they are affected by the lack of wettability of the reinforcement and by interfacial reactivity. Moreover, they are often limited to low-melting point metals. Liquid state routes can be sorted into three major categories [2,3,4]: • •

Infiltration techniques Agitation techniques 9





•

Spraying

Infiltration techniques Infiltration consists of preparing a porous “perform” of the reinforcement and infiltrating its pores with the molten metal. Liquid-phase infiltration is not straightforward, mainly because of difficulties with wetting the ceramic reinforcement by the molten metal. According to Asthana [3], two forces must be overcome to achieve a good bonding and dispersion of the reinforcement: capillary forces and viscous drag of fluid motion through perform interstices. Evans et al. [4] observed that capillarity never favors the process: from an “energetic” standpoint, metals generally do not bond to non-metals. One therefore cannot simply “place” the metal in contact with the ceramics material. To overcome the capillary forces that lead to non-wetting, the chemistry of the system must be modified, or an external pressure is applied most to the metal to force the contact and enhance the wettability. Chemical modifications include coating the reinforcement (for instance with Nickel), adding special elements to the matrix, or using special atmospheres and very high temperatures. It has been observed [3,4,5] that this can lead to unstable phases and limits the range of matrix alloys that can be used. When a mechanical force is used, this energetic barrier could be overcome without altering the composite chemical composition. Porosity is reduced, the structure refined and the interfacial bond quality improved.


       Figure 4: schematic representation of pressure­driven infiltration [5]. 

However, when the infiltration of the preform occurs readily, reactions between the former and the molten metal may take place and significantly degrade the properties of the composite [5]. For this reason, the preform is often coated to reduce the interfacial reactions, thus increasing the complexity and costs of the process. 10
 


 

There are some disadvantages associated with the use of high pressures to combine the ceramic reinforcement and the matrix metal: perform fragmentation, deformation and unevenly reinforced castings [3]. Furthermore, pressures of the order of ten atmospheres (1 MPa) are needed to drive the metal into 1 µm wide pores [4]. Since in nanocomposites fabrication it is desirable to infiltrate significantly smaller pores, heavy equipment is necessary to withstand the high pressure. Moreover, when the reinforcement interface is wide, as for nanoscale composites, the interfacial energy that must be overcome to ensure wettability is higher, and the pressure that must be applied increases as a consequence. In fact, the threshold pressure, able to guarantee a complete wetting, is the key parameter for infiltration techniques. Kaptay [7] calculated its value in the case of liquid metal infiltrated into a ceramic preform made of closely packed, spherical particles. He assumes that the liquid metal has to reach a certain depth of immersion in the preform at which the interfacial forces becomes zero. This depth is called equilibrium depth:

= R

Where R is the particle radius, W is the adhesion energy and

the interfacial energy between

liquid and vapor. It has been observed that when the partially infiltrated liquid metal reaches the equilibrium depth, further infiltration will be ensured. According to this criterion, the pressure to be applied can be derived according to the following expression:

As we can see, with increasing adhesion energy the pressure to be applied is considerably reduced. However, due to the high surface tension of liquid metals, good adhesion can be ensured only due to chemical reaction between the constituents. Despite this, most of the times chemical reactions give rise to unwanted phases (such as carbures) that can be harmful for the microstructure. Therefore, a satisfactory infiltration can be ensured only through the application of very high pressures. Specifically, the above expression indicates that the lower the particle radius, the higher the threshold pressure. When pressures in the GPa range are applied, the technique allows for a synthesis of multi-phase nanomaterials. Gierlotka et al. [6] used a toroid high-pressure high-temperature cell at pressures up to 7.7 GPa and temperatures up to 2000 °C for the infiltration of a nanoporous matrix prepared by compacting nanopowders of high-hardness materials such as Al₂O₃, SiC, or diamond. The preform grain size was about 10 nm. For what concerns the perform fabrication, the costs of a nano-dimension ceramic preform are industrially unaffordable, as well as the fabrication ease of the preform itself [1]. In addition to pressure-driven infiltration, vacuum-driven infiltration has also been used for some matrix-reinforcement systems. Since the magnitude of the negative pressure that can be achieved is limited, vacuum infiltration is usually coupled with the abovementioned methods of wettability enhancement [3]. Lorentz Forces have also been applied to assist the introduction of the metal in the preform. However, the preform position must be suitably oriented with respect to the force axis, and the 11
 


frequency of the current, which induces the electromagnetic force is effective only within a very narrow range of values [3].

Agitation techniques Stir mixing techniques, widely utilized to mix micron size particles in metallic melts, have been recently modified for dispersing small volume percentages of nanosized reinforcement particles in metallic matrices. Their main advantage is the capability over other processes in rapidly and inexpensively producing large and complex near-net shape components, but there are difficulties in mixing nanosized particles in metallic melts that can’t be overcome with the traditional stirring methods. Agglomerates of particles are usually observed in the solidified microstructure. This is due to the increase in surface area connected with the reduction of particle size, which enhances the difficulty of particle introduction and homogeneous dispersion through the melt. The restraints of mixing techniques when dealing with nanoscale reinforcement will be illustrated in the present paragraph.




Restraints of agitation techniques Mixing techniques must face three main process difficulties: 1. Introduction of spherical, ceramic particles into liquid metals; 2. Avoiding coagulation of particles in the liquid metallic phase; 3. Ensuring engulfment (avoiding pushing) of ceramic particles by the approaching solidification front. The physics of particle engulfment lays at the basis of a homogeneous particle dispersion and absence of particle agglomerates in the metal. The analysis will underline the influence of particle radius on the phenomenon.

The dynamics of agglomeration The poor wetting between reinforcement material and the metal matrix presents a barrier to the incorporation of the dispersoid phase into the melt. The immersion of solids into liquids requires substitution of a solid-gas interface by an equivalent solid-liquid interface, and can lead to absorption or generation of energy [11,12]. The energetic of solid immersion into liquids are determined by the energy associated with the solid-gas interface, liquid gas interface and solid-liquid interface. By definition, the surface energy ɣ is “the energy required to create a unit area of new surface”, and represent the extra energy possessed by the surface atoms due to the decrease in bond length between the surface atoms and interior atoms [8,11,12]. When a particle is split into two 12
 


smaller particles, the number of broken bonds contributes to define the surface energy of each of the two particles according to the expression [11]:

where N is the number of the broken bonds, ε is half of the bond strength and

the number of

atoms per unit area on the new surface. Particle size (cm)

Surface energy (J/g)

0.1

Total surface area (cm²) 28

0.01

280

5.6 x

0.001

2.8 x

5.6 x

(1 µm)

2.8 x

0.56

(1 nm)

2.8 x

560

5.6 x

            Table 1: Variation of surface energy with particle size (1 g of sodium chloride) [12]. 

Changes in the size range from micron scale to nanometer, lead to great changes in physical and chemical properties of the material. In Table  1it is shown the scatter of seven order of magnitude in the surface energy when the nanometer scale is reached. The massive increase in surface energy makes particle wetting from the melt more difficult, as the surface energy of the system itself is increased [11]. It is known that the tendency of a system towards stability is associated with the minimum of the Gibbs free energy [11, 12, 13, 14]. Therefore, as the dimension of nanostructured material reduces, the reduction of surface energy is the driving force for equilibrium. There are basically three mechanisms throughout this reduction may occur [12,15]: 1. Partilce sintering 2. Ostwald ripening 3. Agglomeration For the first two, the reader is referred to [9,11] for further explanation, since they are related to particular process conditions. For what concerns agglomeration, it is the most common phenomenon occurring when a solid (particles) comes into contact with a non-wetting medium [14,15]. Agglomeration is strictly bonded to the development of a gas-solid-liquid configuration that favors such situation. This is 13
 


clear when the change in the Gibbs energy of formation is analyzed for gas-liquid-solid phases contributing to the formation [14]: ∆G = (

(T, P) -

(T, P)) +

  T is the temperature, P the pressure in the liquid,

∆ and

+

∆

+

∆

 

the chemical potentials of gas and

the liquid; ∆S is the change in interfacial areas. It is clear from the expression above that in the presence of a solid-liquid interface with cavities filled with gas, the Gibbs energy of formation increases thus diminishing the wettability of the system.

            Figure 5: vapor cavities on the liquid­solid interface of a solid inclusion [14].   

When particles are small and the gravity effect becomes negligible, Van der Waals attraction force together with Brownian motion play the main roles. Van der Waals force is weak and becomes significant only at a short particle distance, and it has been attributed to the existence of interactions between gas molecules or atoms. Hamaker [16] in 1937 stated that such interactions exist also between particles and modified the formulation of Van der Waals through the additivity concept (single atoms or molecules make up the particle). When a gaseous phase occupies the cavities, which are located on the particle Van der Waals forces become attractive (negative). The reason why the attraction between particles is caused, is that they contribute to reduce the Gibbs free energy of the following amount:

Where A is the Hamaker constant depending on the polarization properties of the molecules on the particle surface and the medium between them, r is the particle radius and H the interparticle distance. It can be noted from Figure 6a that the electrostatic repulsion is overcome by the Van der Waals attraction force (marked in red) when the interparticle distance reaches a minimum at around 1 nm (blue dot). For smaller values the Born repulsion of adjacent electron clouds begins to play a role. The ionic strength influences the energy of interaction: as it is shown in Figure 6b the overall 14
 


interaction undergoes a sharp transition from repulsive to attractive as a function of ion strength itself.

                    Figure 6: a) interaction energy between two spherical particles; b) energy of interaction for different ionic  strength [15]. 

In addition to Van der Waals potential, Brownian motion ensures the continuous collision between particles. Particles move simply because of their thermal energy, which lays at the basis of Brownian motion. Brownian motion can be defined as the incessant random motion exhibited by microscopic particles immersed in a fluid. Specifically, a suspended particle is randomly bombarded from all sides by molecules coming from the liquid. A. Einstein demonstrated that if the particle is very small, it behaves like a gas molecule and move continuously under these hits which cause it to be displaced in the liquid. The magnitude of the displacement follows a Gaussian statistic distribution according to the relation: d= 
 
 where η is the viscosity of the medium, r the radius of the particle, T the temperature and K the Boltzmann’s constant. As the relation shows, the displacement increases with decreasing radius, thus the probability for a collision to occur is enhanced. It has been demonstrated [15] that for a particle size smaller than 3.5 µm, Brownian motion totally dominates the agglomeration mechanism over other mechanisms such as shear.

Spraying The spray process is generally automated and quite fast, but it is essentially a liquid metallurgy process. In this process, droplets of molten metal are sprayed together with the reinforcing phase 15
 


and collected on a substrate where metal solidification is completed [4] (Figure 7a). Alternatively, the reinforcement may be placed on the substrate, and molten metal may be sprayed onto it. The inert gas is used to atomize the molten metal in to the droplets, as in the Osprey process, developed by Alcan International Ltd. Among the advantages of spray deposition, deleterious reaction products are generally avoided because the time of flight of the composite particles is extremely short [2]. The process itself is relatively inexpensive (less than powder metallurgy), and the method produces a fine grain structure due to the very high cooling rates.
 





    Figure 7: a) Spray­forming process [5]; b) Porosity in an aluminum matrix composite via spray deposition [17]. 

  






 In thermal spray synthesis many variables are involved, therefore it‘s very difficult to control all the process parameters at the same time. A schematic of the process is shown in Figure . For instance, the porosity level depend upon the thermal conditions, impact velocity and spray density or mass flow. Generally, a high degree of residual porosity is observed in the material (Figure 8), so that it has to undergo further processing [2,3]. The porosity is mainly due to the tendency of the particle to stay at the stream boundaries, giving rise to a inhomogeneous dispersion in the final piece [7]. Some authors [2] have found evidences indicating that particles may surround the droplets during flight. Also, if the deposition rate is higher than the solidification rate, liquid metal may be present at the surface of the substrate. In addition to this, the equipment costs are very high compared to other liquid-state techniques and large amounts of waste powder, which must be collected and disposed, are produced [2].


16
 


Figure 8: schematic of the variables involved in spray deposition process [18]. 

3.3 Solid-state processing Solid-state processes are named as “top-down” techniques. Interfacial and surface wetting issues are considerably diminished. This is due to the fact that both phases remain in the solid state, where diffusivity is much lower [2,4]. A fine scale can be attained, although the cost of the powder is significantly higher than the one of the bulk metal [3,4]. Also, when the process involves attrition and high temperatures (high energy ball milling), chemical modification of the initial constituents is likely to happen [8,12]. The final products are generally affected by a high amount of porosity, which strongly decrease the fatigue resistance and requires further metal working.

Powder metallurgy The group of processes commonly defined as powder metallurgy involves different techniques that could be employed to produce the matrix powder itself and to compact it thereafter (cold isostatic pressing, hot pressing, sintering). A schematic of a typical powder metallurgy procedure is shown in Figure 9. Most of the prior work in synthesizing nanocomposites involves the use of powder metallurgy techniques, which are usually affected by high costs. Powder metallurgy has been used [9] to manufacture nano-metric Al₂O₃ particle reinforced aluminum. The starting Al₂O₃ mean particle size in this case was 50 nm. The process involved wet mixing (aluminum powder mixed with varying volume fraction of Al₂O₃ powder in a pure ethanol slurry), following by drying 17
 


at 150 ºC then cold isostatic pressing (CIP, as opposed to HIP) to compact the powders. The compacted powders were sintered in a vacuum at 620 ºC (approximately 60 ºC below the melting temperature of aluminum). Clustering has been observed, and its occurrence increases with decreasing particle size. Ma et al. [19] fabricated via powder metallurgy nanometric silicon-nitride particulate-reinforced aluminum composite. They reported the presence of agglomerates, which were visible because of the black color of the reinforcement phase. Peng et al. [20] create a novel, unique, and simplified process for producing aluminum metal matrix nanocomposites reinforced with nanoscale aluminum oxide particles having a controlled volume fraction and uniform distribution in the aluminum matrix. The novelty is the employment of a technique that utilizes the Al₂O₃ surface layers existing on all aluminum particles as ceramic reinforcement. However, the process does not allow a satisfactory control in the phase of layers break-up and spreading. In addition to this, the effectiveness and the scalability of the method have not been proved yet. Moreover, in order to avoid agglomeration, the aluminum matrix alloy powder should have a particle size close to that of the nano-sized reinforcement. Further working of products attained by powder metallurgy may cause the reinforcement phase to break up and deform the surrounding matrix, leading to stress concentration and non uniform particle size [5].


         Figure 9: powder processing, hot pressing, and extrusion process for particulate reinforced composites [5].   

18
 


Mechanical attrition and alloying Mechanical alloying was initially invented in 1980 to form small-particle dispersion-strengthened metal alloys [8]. In the 1990s, the method of high-energy milling gained a lot of attention as a non-equilibrium solid-state process resulting in materials with nanoscale microstructures. A variety of ball mills have been developed for different purposes including tumbler mills, attrition mills, shaker mills, vibratory mills, planetary-mills [12]. The basic process of mechanical attrition is illustrated in Figure 10.


 Figure 10: schematic sketch of the process of mechanical attrition of metal powders [8]. 

  High-energy milling forces can be obtained by using high frequencies and small amplitudes of vibration [12]. Due to the continuous severe plastic deformation, a continuous refinement of the internal structure of the powder particles to nanometer scales occurs during the attrition by the mills, which are made by a dense material such as tungsten or steel. The temperature rise during this process is normally between 100 C° and 200 C°. In the high energy ball milling process, alloying occurs as a result of repeated breaking up and welding of the component particles. During mechanical attrition, the metal powder particles are subjected to severe plastic deformation from collisions with the milling tools. Consequently, plastic deformation at high strain rates occurs within the particles and the average grain size can be reduced to a few nanometers after extended milling [12]. It can provide metastable structure such as nanocomposite structure with high flexibility and scaling to industrial quantities can be easily achieved [8]. The aluminum based nano-composites with the trade-name DISPAL, which are reinforced with Al₄C₃ particles, are manufactured using this method [9]. Despite this, the process has several drawbacks. 19
 


First of all, contamination by the milling tools and the atmosphere usually occurs. Milling of refractory metals (i.e. tungsten) in a shaker for extended periods of time can result in levels of iron contamination of more than 10 at% if high vibrational or rotational frequencies are employed [8,12]. To prevent these phenomena, the process has to be carried out in inert atmospheres and the mills coated. Also, high-energy ball milling can provoke chemical reactions due to the conversion of the mechanical energy into heat [12]. For instance, ductile materials are difficult to ball mill due to particle coarsening, as a result of the advent of chemical reaction with the matrix element. This may occur during milling, or at a later stage during heat treatment [9]. Zhang et al. [21] proved that, when producing metal–ceramic nanocomposite powders by high energy milling of metal powders and ceramic powders, there exists a lower limit of the particle size below which reduction of the particle size cannot be achieved using milling, since the stresses required for further particle refinement cannot be reached during the process. This statement can be explained through the following equation:

where

is the required fracture stress, c the fracture toughness and

the defect size in the

material. As ceramic materials generally have low fracture toughness, the fracture stress is not very high, making the particles easier to be fractured. However, when the ceramic particles are reduced to a nanometer sized level, the likelihood of having internal defects and surface notches are reduced considerably. In this case, will approach the theoretical strength of the ceramic, which is about 1/30 of its Young‘s modulus. As an example, the Young‘s modulus of SiC is approximately 450 GPa. This means that the impact stress has to be over 15 GPa to fracture a “perfect” – meaning with equal to zero - SiC particle. This stress would be very difficult to be achieved with a conventional high energy mechanical mill. Moreover, nanoparticles produced by attrition have a relatively broad size distribution and varied particle shape and geometry. To conclude, although mechanical attrition can produce very fine particles, this process is difficult to design and control so as to produce desired particle size and shape. It is also limited to materials with very poor thermal conductivity [11].

References 1. M. Di Ventra, S. Evoy, J. R. Heflin, “Introduction to nanoscale science and technology”, Springer, 2004 2. Suresh, Mortensen, Needleman, “Fundamentals of metal matrix composites”, Buttleworth-Heinemann ed., 1993 20
 


3. R. Ashtana, “Solidification Processing of Reinforced Metals”, Trans. Tech Publications, 1997 4. A. Evans, C. San Marchi, A. Mortensen, “Metal Matrix Composites in Industry: An Introduction and a Survey”, Springer, 2003 5. Krishan Kumar Chawla, “Metal matrix composites”, Birkhäuser, 2006 6. Gierlotka, Synthesis of Metal-Ceramic Nanocomposites by high pressure infiltration, Science24.com 7. G.Kaptay, “Interfacial criteria for producing ceramic reinforced metal-matrix composites”, Proc. Int. Conf. High Temperature Capillarity 29 June-2 July 1997, Poland 8. P. M. Ajayan, L. S. Schadler, P. V. Braun, “Nanocomposite science and technology”, Wiley-VCH, 2003 9. L.Fischer,” Literature Survey Report: Nano-Dispersion Strengthening of Aluminum”, Introduction to research, 2004, University of Colorado 10. Vozken Adrian Parsegian, “Van der Waals forces: a handbook for biologists, chemists, engineers, and physicists”, Cambridge University Press, 2006 11. Guozhong Cao, “Nanostructures & nanomaterials: synthesis, properties & applications”, Imperial College Press, 2004 12. C. C. Koch, “Nanostructured materials: processing, properties, and applications”, William Andrew, 2006 13. Fast, “Interaction of Metals and Gases”, Academic Press, 1965. 14. M. Cournil, F.Gruy, P. Cugniet, P. Gardina, H. Saint-Raymond, “Model of aggregation of solid particles in nonwetting liquid medium”, Centre SPIN, URA CNRS 2021, Ecole Nationale Supérieure des Mines de Saint-Etienne. 15. “Particle collision and aggregation”, Oceanography 540--Marine Geological Processes--Autumn Quarter 2002 16. Rhonda Lee-Desautels, “Theory of van der Waals Forces as Applied to Particlate Materials”, Educ. Reso. for Part. Techn. 051Q-Lee 17. Thomas Seefeld, Emil Schubert and Gerd Sepold,” Spray Deposition of MMC Composites by Laser Spraying with Particle Co-injection”, BIAS Bremen Institute of Applied Beam Technology 18. B.Onur, Nanocomposites. 19. Z.Y. Ma, Y.L. Lia, Y. Liang, L F. Zheng , J. BP, S.C. Tjong, “Nanometric Si3N 4 particulate-reinforced aluminum composite”, Materials Science and Engineering A219, 1996, pp. 229-231 21
 


20. Peng et al., “Manufacturing method for aluminum matrix nanocomposites”, United States Patent, 7297310 21. D.L. Zhang, J. Liang, J. Wu, “Processing Ti3Al–SiC nanocomposites using high energy mechanical milling”, Materials Science and Engineering A 375–377 , 2004, pp. 911–916

4. Selected Manufacturing Processes 4.1 Ultrasonic Cavitation Based Solidification High-intensity ultrasonic waves (above 25 W/cm²) can generate strong non-linear effects in the liquid such as transient cavitation and acoustic streaming (a liquid melt flow due to an acoustic pressure gradient) [1]. They are mostly responsible for dispersive effect for homogenizing, degassing for reduced porosity and refining microstructure [2].

Experimental parameters and design of experiment As it is shown in Figure 11, the A356 alloy is melted in a graphite crucible (2 lb. capacity) using an electric resistance heating unit and protected by Argon gas. The ultrasonic probe (Permendur probe) is made of Niobium (Nb) which can withstand high processing temperature with minimum ultrasonic cavitation induced erosion [1]. An alternative arrangement [3] consists in ultrasonic transducers arrayed outside, and coupled with the crucible.

                                                              Figure 11: Schematic of experimental setup [1].  22
 


Nano-sized particles (β-SiC with average size