This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Mg-Tri Calcium Phosphate Composite is an ideal material for biomedical application, however the poor mechanical properties restrict its application. Friction Stir Processing (FSP) is a potential way for improving the properties of Mg-Tri Calcium Phosphate Composites for biomedical application. In this project, the effect of FSP on Mg-4Zn-2TCP composite and the influence of different processing parameters have been investigated. The effects of FSP on Mg-4Zn alloy was also studied as a comparison. Optical microscopy and SEM were used to study the microstructure evolution after FSP. Microhardness testing were carried out to measure the mechanical property of the materials. Particle size distributions were also measured to quantify the level of particle refinement and the homogeneity of the TCP clusters.
After FSP, the microstructure of Mg-4Zn-2TCP composite and Mg-4Zn alloy was greatly refined. The ultra fine grains were generated in the nugget zone. The grain size reduced about 20 times for Mg-4Zn alloy, and about 10 times for Mg-4Zn-2TCP composite. The microhardness of Mg-4Zn alloy increased from about 43 Hv to about 50 Hv. For Mg-4Zn-2TCP composite, the microhardness also improved a lot and slower travel speed gave rise to a higher hardness. The big TCP clusters were broken up to strip-like morphology. The average cluster size reduced 5 times after FSP, and the homogeneity of the TCP cluster increased dramatically. During the FSP, the MgZn eutectic compounds dissolved into the matrix or broken up to very small size. In addition, the casting porosity disappeared in the nugget zone after FSP.
Table of Contents
Mg-Tri Calcium Phosphate (TCP) composites have very good biocompatibility. The magnesium alloy matrix is non-toxic and can be engineered to dissolve in our body or corrode at a pre-determined rate. TCP compound is similar to one of the mineral in human bone. So it is an ideal material for bone repair. However, the mechanical properties of cast Mg-TCP composites may not satisfy the requirement of biomedical applications for containing coarse TCP particles and their segregation. They need to be refined and better distributed by downstream processing. As the cast Mg-TCP composites are very hard and brittle, so it will be easily fractured if we choose the traditional thermomechanical processes such as hot extrusion rolling or forging etc. In this project, a severe deformation method which involves Friction Stir Processing (FSP) will be studied to refine the TCP particles and improve the microstructure of cast Mg-TCP composites, therefore obtain good mechanical properties
The Mg-TCP composite will be prepared by solidification from the melting Mg alloys. The cast contains coarse particles and segregation of the TCP particles. They need to be refined and better distributed by downstream processing. FSP is a novel approach closely related to Friction Stir Welding that involves translating a rotation tool that is plunged into the material surface. The material is heated and deformed by friction. When the tool is translated, very large strains (>100) are generated in their deformation zones that are known to lead to excellent high levels of particle refinement and revealed significant breakdown of reinforcement.
The Aims of the project are to demonstrate the effects of FSP on the Mg-TCP composites. Microstructure evolution and material flow during FSP and their mechanisms will be studied. By changing the weight fractions of TCP particles and different FSP parameters (rotational rate of the and its traverse speed along the line of joint), the best conditions will be selected to optimize the properties of the Mg-TCP composites in order to achieve the biomedical application requirement. Optical and scanning electron microscopy will be used to analyze the microstructure of the product. Besides, Micro-hardness test will also be involved to measure the mechanical properties of the materials.
2 Literature Review
2.1 Magnesium, Magnesium Alloys
2.1.1 Pure Magnesium
Magnesium is a king of light metal with Atomic Weight 24.4, and its abundance rank the sixth place on earth. Magnesium is ductile and very easy to machine compared with other metals. The Young's modulus of magnesium at 20â„ƒ is 45 GPa; hardness of magnesium is HB 36 and the Poisson's ratio is 0.35. The density of magnesium at 25â„ƒ is 1.738 g/cm3. At the melting point (650â„ƒ), it decreases to 1.65g/cm3. So there is an expansion in volume with the increase of temperature, which is 4.3% at the melting point .
At constant pressure, the specific heat capacity Cp at 25â„ƒ is 1.028 kJ/kg.K . It goes up with the increase of temperature. The relationship between temperature and Cp can be expressed as:
2.1.2 Magnesium Alloys
Magnesium alloys are mixtures of magnesium with other metals like zinc, aluminum, manganese, copper, silicon, rare earths and zirconium. Magnesium is known as the lightest structural metal. The lattice structure of magnesium alloys is hexagonal, which has a fundamental affect on the properties of Mg alloys. Plastic deformation of the hexagonal latticed metals is more complex than cubic lattice materials such as copper and steel etc. So magnesium alloys are mainly used as cast alloys, however, more attention has been drawn to wrought magnesium alloys after 21century .
Magnesium alloys are classified by short codes denoting the approximate chemical composition by weight. For example, AZ81 contains 8% aluminum and 1% zinc; AS41 contains 4% aluminum and 1% silicon. If Al is added to the Mg alloys, about 0.3 wt. % of Mn is always present to refine its microstructure; if Al and Mn are present, Zr is usually absent at about 0.8 wt.% for the same reason . Zr can only be used in Al Mn free alloys for the formation of intermetallic. The convention to tell the main elements in Mg alloy is shown in table 2.1.
Table 2.1 Convention for alloy elements designation
Letter Alloying elements
Rare earth metals
Different alloying elements give different properties of magnesium alloys. The fig. 2.1 shows that the development of different magnesium alloys .
Fig. 2.1 Development of different magnesium alloys
2.1.3 Strengthen Mechanisms
The mechanisms that restrict the movement of dislocation in Mg alloys can strengthen the Magnesium alloys.
Strain hardening is also called Work Hardening, which is a mechanism of metal strengthening led by plastic deformation. Dislocations that are a kind of line defects can give rise to a lattice strain field within crystal structure. So the bonds close to the dislocations will be under compression and the bonds beyond dislocations are under tension. Therefore, the repulsive or attractive forces associated with the strain fields can impede dislocation motion. On the other hand, dislocations generated by deformation accumulate and interact, and they will form immobile jogs and locks, providing obstacles to further dislocation motion. As the influence made by these two processes, the relation between dislocation density (ρ) and yield strength (-σy) can be given by :
-σy = Gbρ1/2
where b is the Burgers vector, and G is the shear modulus. As a result, when Magnesium and Magnesium alloys are being continuously strained beyond its elastic zone, the increase in dislocation density results in strengthening of metals.
Solid Solution Hardening
Solute atoms also have effects on the strengthening of materials. Solute atoms that is defined as point defects in Magnesium and in Mg alloys alter the lattice structure and generate misfit strain field. There are two kinds of solute atom, which are substitutional, where the solute atoms take the place of the solvent atoms, or interstitial, where the solute atoms are added between the solvent atoms. As a result, the solute atoms cause lattice distortions that impede dislocation motion. A larger substitutional atom will exert a compressive strain on the surrounding matrix atoms, while a smaller substitutional atom will cause a tensile strain. Therefore, interaction between the strain fields imposed by the solute atoms and the strain field associated with dislocations will hinder dislocation motion. Generally, Increasing the concentrate and shear modulus of the solute atoms will increase the yield strength of the material .
Strengthening by Grain Size Refinement
A smaller grain size shows a great effect to the strength of Mg alloys. The yield stress increases with decrease of the grain size. The grain size dependence of the yield stress (and tensile strength) can be expressed by the Hall-Petch relationship [6, 7]:
σy =σo+ K ⁄ d1/2
where d is the average grain diameter, σ0 is a constant and K is the stress intensity factor for plastic yielding.
Second phase precipitates contribute most significantly to improving the strength of Mg alloys. The strength of precipitate strengthened polycrystalline magnesium alloys depends on the particle size, particle distribution and volume fraction of the particles and the nature of the interface between the particles and the matrix (coherent or incoherent). Dislocations may be able to shear small particles but may not cut through large incoherent particles. Bowing is the only way that dislocations can go through large particles [8, 9]. As a result, a relatively fine and close distribution of particles is required for the precipitates to act as effective barrier to dislocation motion, by resisting both cutting and bowing, and the strengthening effect can be estimated using Ashby Orowan strengthening equation given by:
-σ= 0.13Gmbλ-1 ln(r/b)
where -σ is Orowan strengthening caused by the second phase particles, and b is Burger vector of the matrix. Gm is the shear modulus, λ is the interparticle spacing, and r is the radius of the particles.
2.1.4 Deformation Behavior
According to von Mises  at least 5 independent slip systems are required for polycrystalline magnesium to deform without grain boundaries rupture. Magnesium is close-packed hexagonal structure, therefore, at room temperature there are only 2 independent slip systems available. The number of the basal slip systems (shown in the table 2.2 and fig. 2.2) is much less than that needed. In order to achieve deformation, other non-basal slip systems must be activated or twinning must occur. The brittleness of magnesium at low temperature is a reason of the limited number of independent slip systems.
Table 2.2 Slip systems for Mg and Mg alloys
Relative CRSS (room temperature)
Independent slip system
Fig. 2.2 Slip systems of Mg alloys
Table 2 indicates that first order prismatic slip systems and second order pyramidal slip systems have great effect on the deformation behavior of Magnesium and Mg alloys. In order to improve the deformation ability of Mg, it is necessary to activate of non-basal slip systems. Researchers have achieved the acceptance that the activity of a non-basal slip system, for example, the C critical resolved shear stress of non-basal slip, has dependence on precipitates, solute atoms and temperature. The critical resolved shear stress for slip in the second order pyramidal system is proportional to temperature. The critical resolved shear stress of prismatic slip system falls down very dramatically with increase of the temperature . Generally, above 250â„ƒ sufficient slip systems are active and Mg show good ductility. Therefore, Mg alloys are always rolled or extruded above this critical temperature.
Twinning is also known as a deformation mechanism that occurs intensely even at very low temperatures and low strains. Twinning may occur in the first stage of the deformation and can change the orientation of the basal planes so that they can slip easier. Twinning can lead to asymmetry of the materials for it accommodates deformation easier when compressed than stretched parallel to basal plane.
2.2 Biomedical Application
2.2.1 Biomedical Application of Magnesium Alloys
Magnesium alloys are considered to be a new kind of degradable biomaterials which have attracted great attention of the investigators recently. The main advantages of magnesium alloys are their mechanical properties, biocorrosion properties and biocompatibilities .
Magnesium and magnesium alloys have very low density that range from 1.75 to 2.1 g/cm3, which is much less than that of the biomedical Ti alloy (4.5-4.6 g/cm3) and close to that of the human bone (1.9- 2.2 g/cm3)
The fracture toughness of magnesium and magnesium alloys are better than that of most ceramic materials used in biomedical field.
The elastic modulus, 41-45 GPa, of the magnesium and magnesium alloys is close to that of the human bones, which avoids the stress shielding effect.
Magnesium is an element needed by hunman body. It is the fourth most abundant cation in the human body and also necessary for metabolism. There is approximately 25g magnesium stored in every human body and almost half of the total content stored in bone tissue. Magnesium is a cofactor for many enzymes and stabilized the structures of DNA and RNA.
Magnesium and magnesium alloys have a fast corrosion rate in the physiologic environment with the standard electrode potential of -2.37 V, and pure magnesium metal exhibits even poorer corrosion resistance in Cl- containing physiologic environment. So Mg can be engineered to dissolve in our body or corrode at a predetermined rate without taking it out after bones are repaired.
In conclusion, biomaterials should have high strength, good biocompatibilities and proper corrosion rate that can match tissue healing rate. A lot of element can be used as alloying element for Mg alloys in order to achieve the suitable mechanical properties and corrosion rates.
Fig. 2.4 Relationship between Zn and the tensile strength of magnesium alloysAs Ca Mn and Zn are essential for human body, these three elements should be the best alloying element. The matrix material used in this project is Mg-Zn alloy. Some researches  also revealed that the zinc could improve the corrosion potential of Mg in simulated body fluid (SBF) and reduce the corrosion rate. The matrix material for this project is Mg-4Zn alloy. The phase diagram of Mg-Zn alloy is shown in Fig. 2.3, and it indicates that largest solubility of Zn in solid Mg is 8.4% (340â„ƒ), so if the concentration of Zn is less than 3%, the MgZn phase will not present in the magnesium matrix. Also, with the decrease of the temperature, the solubility goes down. Therefore, quenching and ageing can be used in magnesium alloys. On the other hand, the alloying element Zn also has influence on the tensile strength (shown in the fig. 2.4). The magnesium alloys achieve best mechanical properties when the concentration of the Zn is 6%. But the concentration of Zn should not be too high in the wrought magnesium alloys because of the formation of the loose structure, and this will lead to poor machinability .Macintosh HD:Users:xunanking:Desktop:Screen Shot 2012-07-27 at 15.44.05.pngMacintosh HD:Users:xunanking:Desktop:Screen Shot 2012-09-02 at 00.11.39.png
Fig. 2.3 Phase diagram of Mg-Zn alloy
2.2.2 Biomedical Application of Tri Calcium Phosphate
Tri calcium phosphate [Ca3(PO4)2, TCP] is a kind of hydroxyapatite which usually having a P-whitlockite crystal structure.  It is one of the most important biodegradable ceramic materials in biomedical field, because it is very similar with the inorganic part of body bone and tooth, which the majority is made from hydroxyapatite, salts of calcium (phosphate and carbonate). Besides, another important reason is that TCP (dense or porous state) contains the property of osteosynthesis (stimulate bones to grow). Tri calcium phosphate can be synthesized by a classical ceramic technology with the reaction:
2NH4. H2. PO4 + 3CaCO3 3CO2 + 3H2O+Ca3(PO4)2 + 2NH3
Table 2.3 The mechanical properties of bone, metallic and calcium phosphate implant materials
However, the mechanical properties (shown in table 2.3) of calcium phosphates are too poor. They are brittle and not able to be used as structural materials So, their applications are restrict in coatings of other materials and body bone fillers areas . Recently, some studies have been reported that TCP can be used as reinforcement for magnesium matrix composite in order to achieve the advantages of the both parts. Thermoplastic polymers filled with a ceramic powder (TCP/CPLA composite) are researched in Japan and this composite show very good mechanical properties and biocompatibility which is suitable as a bone repair material .
2.2.3 Biomedical Application of Mg-TCP Composite
Magnesium matrix composites are widely used in biomedical applications because of their improved mechanical properties compared with their monolithic alloy matrix and the good biocompatibility of both magnesium and TCP. Moreover they also have a lower corrosion rate than magnesium alloys . The effort is initially put on continuous fiber reinforced composites, but the route of this process is complex and the cost is very high. Therefore, more researches are done on the particle reinforced Mg composites after that because of their low cost, high modulus, strength and wear resistance and ability for welding . Many results suggested that Mg-TCP composite is a suitable material for bone repairing.
In the research of HE Sheng-ying , the Mg-Zn-Zr/β-TCP composites were cast in a vacuum inductive furnace with the protection of argon gas, and extrusion and ageing were involved to heat-treat the composite. Results show that the grain size of the composite was much finer than that of the monolithic metal matrix. By doing the tensile tests, the increase of the ultimate tensile strength (UTS) and the elongation of composites were observed when β-TCP was added. At the same time, the liquid alloy infiltration technique was also used to fabricate a novel MgCa-hydroxyapatite/ tricalcium phosphate (HA/TCP) composite by Gu. XN . Recently, high shear solidification and equal channel angular extrusion (ECAE) were involved to synthesis nano β-tricalcium phosphate (β-TCP) particles reinforced composite Mg-2Zn-0.5Ca-1β-TCP. The result showed that uniform grain structure with β-TCP particle clusters of 5-25 μm in size evenly distributed in the magnesium alloy matrix while the ECAE processing led to further microstructural refinement and a uniform dispersoid of β-TCP particles in the matrix, giving rise to an increase in both the hardness and the corrosion resistance for the material. The formation of a passive surface film consisting of β-TCP nano particles was considered to be an important reason for the increased corrosion performance .
The study of magnesium-fluorapatite (FA) nanocomposite material was done in Isfahan University of Technology , which reveals that the addition of magnesium-fluorapatite nanoparticle reinforcements to Mg alloys can better the mechanical properties and lower the corrosion rate. Furthermore, the formation of an apatite layer on the surface of the alloys was faster, which can protect the matrix and provide the osteoconductivity of Mg alloys for biomedical applications.
A. Martin  indicated that the addition of the ceramic reinforcements SiC in Mg-6Zn alloy result in enhancement of the Young modulus up to 60%, and the two-step artificial aging treatment contribute significantly to the mechanical properties of the composite in which ultimate tensile strength became 120MPa higher than the monolithic metal matrix.
In conclusion, a lot of work has been done on searching and fabricating new biomedical Mg matrix composites to achieve the increase of the mechanical properties. However, the limitation is due to the microstructure of the composite material is always undesirable. The particles suffer a large range of sizes and do not distribute evenly. Moreover, the toughness of the materials is low. So in order to further improve the mechanical properties and simplify the complicated fabrication process, more attention should be drawn on how to modify the microstructure of composite. Some deformation methods such as hot extrusion , hot forging , multidirectional forging  and equal channel angular pressing (ECAP)  were investigated. Although the grain size and particle distribution are improved in these methods, the size of the particle did not decrease apparently and the process is relatively complex, moreover the strong texture were developed during the processing. In order to overcome those shortcomings, so was born the Friction Stir Processing that can provide the severe deformation to materials and excellent modification of microstructure. This project is going to study the mechanism and parameter of Friction Stir Processing as well as its effect on microstructure modification so that satisfied properties for biomedical application can be achieved.
2.3 Friction Stir Welding
Friction stir welding (FSW) is a solid-state joining technique that was created by Wayne Thomas in UK in 1991. The initial applications are to weld aluminum alloys . The principle of FSW is very easy to understand. A high-speed rotational tool with a pin and shoulder is plunged into the specimens to be welded and travel along the line of joint (fig. 2.5). The tool has two primary functions. The first one is heating of specimen and the second one is moving of material to achieve the welding. The heating is generated duo to the friction between the tool and the specimen and plastic deformation of specimen. The localized heat makes the material around the pin soft. Then the rotation and translation of the tool result in moving of material around the pin. Because of various geometrical features of the tool, the material movement around the pin can be quite complex. During fraction stir welding process, the material suffers a severe plastic deformation at high temperature, leading to forming of equiaxed and fine recrystallized grains . The nice microstructures in FSW give rise to good mechanical properties.
Fig.2.5 Schematic drawing of Friction Stir WeldingFSW is known to be one of the most important developments in metal welding recently and because of the energy efficiency, environment friendliness, and versatility. Therefore, it is called as a green welding technology. The key advantages are shown in the table 2.4.
Table 2.4 Key benefits of friction stir welding
2.4 Friction Stir Processing
According to the basic principles of FSW, Mishra and his colleagues recently developed friction stir processing (FSP) as an effective tool for microstructural modification method [30, 31]. In such processing, the designed tool is plunged in a monolithic specimen for microstructural modification and this can lead to specific mechanical property improvement. For example, severe deformation and high strain rate were obtained in aluminum alloys by friction stir processing which lead to grain refinement and a high level of second phase particle breakdown. Besides, friction stir processing was also used to make surface metal matrix composites .
2.4.1 Tool Design
Fig. 2.6 Schematic drawing of the FSW tool
The tool design is a very important factor in FSP. By using a carefully designed tool piece, the quality of the process can be improved and productivity can be increased. An FSP tool is made up of a pin and a shoulder (as shown in fig. 2.6). The tool has two main functions that have been mentioned in chapter 2.3. The first one is the heat generated by the friction between pin and specimen. Some additional heat obtains from deformation of alloys. The tool is inserted into the materials until the shoulder touches the specimen. The main part of the heat comes from the friction between the shoulder and specimen. Therefore the relative size of pin and shoulder play a critical role in tool design, while the other features of the tool are not so important. The shoulder also provides confinement for the heated volume of material. The second one is the moving and stirring of the alloys. The distributions of microstructure and post-weld properties as well as process loads are always determined by the tool design. Generally a concave shoulder and threaded cylindrical pins are used.
2.4.2 Processing Parameters
Transverse and Rotational Speed
In a successful friction stir processing, the pin travel speed (mm/min) and the rotational speed (rpm) of the tool the need to be carefully selected. The material around the tool will have to be locally raised to a temperature range in which severe plastic deformation can readily occur, and at the same time minimizing the forces acting on the tool. In general, the heat input to the specimen increases with higher rotation speed or decreasing travel speed. Insufficient heat generation can cause voids and defects to develop within the stir zone, and the friction stir tool can also be damaged under high processing forces. On the other hand, too high a temperature can cause the surface layer of the alloys to melt, which affect the microstructure by resolidification.
Plunge Depth and Tilt
The plunge depth means the distance that shoulder is plunged into the surface of specimen. This generates necessary pressure on surface of specimen, and the pressure can forge the material at the rear of the tool. A too low plunge depth can cause an inaccurate welding or even the development of voids within the welds. On the other hand, too high a plunge depth can lead to damage of the tool. Furthermore, large amounts of flash will be generated with too high a plunge depth, and this could then result in a large thickness mismatch between the processed zone and the parent material. Tilt is the angle that the pin tilts from the vertical direction. This will impose a more effective forging of the material at the rear of the pin, but too much tilt reduce the heat generated because of the decrease of the contact surface between shoulder and specimen. As a result, the tilt angle is usually set no greater than 4Ëš, and a plunge depth of 0.2-0.3 mm is applied to increase the area of contact.
2.4.3 Microstructure Evolution
During the FSP, material being processed suffers a server deformation and high temperature treatment, and the microstructure of the specimen is highly modified. Generally three zones with different microstructure characterization of grains and precipitates will form, which are processed zone (PZ), thermo mechanically affected zone (TMAZ), and heat affected zone (HAZ) (shown in fig. 2.7). Microstructure variations in three zones have important influence on mechanical properties of the materials after FSP .
Fig. 2.7 A typical macrograph showing various mircostructure in FSP
The processed zone (nugget zone) is located in the center of weld in which material will move around the pin and suffer a severe plastic deformation. With in nugget zone, the equiaxed recrystallized grains with very small size form because of the recrystallization, so this zone is also called dynamically recrystallized zone (DXZ); Precipitates coarsening or dissolution also may occur because of temperature rise; The coarse particles are also broken to smaller sizes by the stir of pin. Furthermore, after the FSP onion ring structure sometimes form in the nugget zone (figs. 2.7 and 2.8b). Generally there is low dislocation density in recrystallized grains. However, some researchers claimed that high density of sub-boundaries, sub-grains, and dislocations present in the fine recrystallized grains of the stirred zone .
With different tool geometry, processing parameter, temperature, and thermal conductivity of the material, many kinds of nugget zone shapes have been observed by researches. But two shapes are most common, which are basin-shaped nugget zone that widens near the upper surface and elliptical nugget zone (shown in fig. 2.8).
Fig. 2.9 Typical microstructure of Thermo-mechanically Affected Zone Fig. 2.8 Nugget zone of FSP
Thermo mechanically Affected Zone
Thermo mechanically affected zone (TMAZ) - a transition zone is between the nugget zone and heat-affected zone. TMAZ suffer both temperature rise and plastic deformation during FSP. A typical micrograph of thermo mechanically affected zone is shown in fig. 2.9. The thermo mechanically affected zone is a high deformation zone. The parent metal grains were elongated in an upward flowing pattern around the stirred zone. The thermo mechanically affected zone experienced plastic deformation, but there is no recrystallization because the deformation strain is not enough. Dissolution of some precipitates sometimes occurs in TMAZ due to high temperature generated by friction during FSP. The thermal cycle of thermo mechanically affected zone can determine the extent of dissolution. Besides, it was revealed that a high density of sub-boundaries was usually observed in the grains of the thermo mechanically affected zone.
Heat Affected Zone
Between TMAZ and parent material is the heat-affected zone (HAZ). This zone undergoes a temperature rise cycle, but does not experience any plastic deformation (shown in fig. 2.7). The heat generated by FSP result in precipitates coarsening or dissolution and recover of materials. So the grain size in this zone may be larger than that of the parent material.
2.4.5 Material Flow
During friction stir processing, the movement of material (material flow) caused by the rotating tool is very important in determining the microstructure of the PZ, but he whole deformation process is so complicated that the way which material moves around the shoulder and the pin is still not fully understood. fig. 2.5 shows a schematic drawing of the process. While the tool travels across the specimen along the welding direction, material on the left and right side of tool have different deformation behaviors [35, 36], which are known as the 'advancing side' and 'retreating side respectively. The advancing side is the side that the travel direction is the same with the rotational direction of the tool, whereas the retreating side is the side that the travel direction is opposite to the rotational direction of the tool. Most of these investigations have suggested that the generation of the DRZ is governed by an extrusion process, with the shoulder, the pin, the backing plate, and the cold parent material effectively acting as an extrusion chamber .
The material on the advancing side has been observed to travel around the rotating pin . The material travels a long distance around the retreating edge before being finally deposited behind its original position. Hence the material behind the tool on the advancing side is highly deformed. In contrast, the material on the retreating side will be extruded from the front of the tool to the rear without traveling round the tool. So the material on the retreating side moves only a short distance, and experiences a lower level of deformation. Some researchers also noticed that the distance that material move from the tool front to its rear was limited to only one pin diameter behind the initial position of the material. Besides, it has also been observed that mixing and stirring between the material on the advancing and retreating sides rarely take place in the bulk where the pin dominate flow, but occur frequently at the surface of the specimen under the shoulder [37, 38]. This is because the shoulder rather than the pin mainly determines the surface material flow.
In the nugget zone, onion rings that have been motioned before will form. The mechanism for this phenomenon is complex and still under debate. The idea that the FSP is an extrusion process is supported by Krishnan . Onion rings found in the nugget zone is a direct evidence of characteristic material transport phenomena occurring during FSP. It was suggested that the friction stir welding process could be thought to be simply extruding one layer of semicylinder in one rotation of the tool and a cross-sectional slice through such a set of semicylinder results in the familiar onion ring structure. On the other hand, Biallas et al.  suggested that the formation of onion rings was attributed to the reflection of material flow approximately at the imaginary walls of the groove that would be formed in the case of regular milling of the metal. The induced circular movement leads to circles that decrease in radii and form the tube system. In this case, it is believed that there should be thorough mixing of material in the nugget region.
2.4.6 Modification of Magnesium Alloys and Composites by FSP
A lot of research has been down on the modification of magnesium alloys by FSP. In 2005, some researchers in Japan produced the high-strength Mg-Y-Zn alloy plate by FSP after casting. Single, overlapped double and multiple-pass FSP were carried out. Ultra-fine-grained microstructures were observed in nugget zone and the Micro-Vickers hardness tests showed that overlapped FSP could result in twice hardness compared with parent material . Furthermore, in 2006 A.H. Feng suggested that FSP of Mg-Al-Zn casting led to significant breakdown and dissolution of the coarse, network-like eutectic b-Mg17Al12 phase distributed at the grain boundaries and significant grain refinement (15μm), therefore the tensile properties of the casting improve dramatically .
FSP is also a possible way to fabricate magnesium matrix composite. For example, 5-10 vol.% nano-sized SiO2 was incorporated into an AZ61Mg alloy matrix to form bulk composites by this new method. The nanoparticles were uniformly dispersed within nugget zone after four FSP passes, and the grain sizes of the composites varied within 0.5-2 μm, and the hardness of the composite is almost twice as high as that of the parent material and showed high strain rate superplasticity .