Investigating Local Deformation In Superplasticity Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The effect of tensile deformation at 500 and a strain rate of on the microstructure, mechanical and texture characteristics is about to be investigated using the grid method on the transverse direction of the test specimen (AA8090). Similar works have been done on the AA8090 but with other techniques. The gridding method is expected to give a thorough incite in to the grain translation and the microstructural evolution within the grain.


Superplasticity is the ability of a polycrystalline material to undergo large uniform elongation with an unusually high strain rate sensitivity of flow stress (m-value). Unlike conventional materials that rely on work hardening to develop neck resistance, superplastic materials attain this because of an exceptional sensitivity of the flow stress to strain-rate. Superplasticity arises under well-defined conditions such that either there is a characteristic microstructure that is stable during the tensile test, or there are special environmental conditions during deformation. Materials in the first group are said to exhibit "structural superplasticity", sometimes referred to as "isothermal" or "micrograin" superplasticity. Those of the second group exhibit "environmental superplasticity". Structural superplasticity is by far the most important commercially and has received much more attention than environmental superplasticity. It is this m-value that gives the material a high resistance to necking. The m-value is defined by:

Where is the stress; is the strain rate and m is the strain rate sensitivity of flow stress.

For superplastic behaviour, m would be greater than or equal to 0.5 and for an aluminium alloy the value would be approximately 0.5. The m-value is a very important parameter in superplastic behaviour.

Experimentally the m-value can be inferred from the interpolation of the strain rates of the stress-strain responses of the tensile test results.

Fig. 1. Test strain result for a typical superplastic material (Al-Cu-Zr) showing the stress-strain response and corresponding m-values [1].

Because of this forming processes similar to those employed with thermoplastics and glass are possible with thermoplastics. The use of superplastic has invited considerable interest from various aspects of engineering field especially the aerospace industry. Superplastic forming of nickel-base alloys has been used to form turbine discs with integral blades while diffusion bonding and superplastic forming of titanium alloy is used to produce fans and compressor blades for aero-engines. Aluminium alloys can be used in the fabrication of airframe control surfaces. For deformation in uniaxial tension/elongation to failure in excess of 200% is usually indicative of super plastic behaviour although several materials can attain excessive extensions of about 1000%.

Deformation is typically performed at elevated temperatures and low strain rates with the presence of a fine microstructure. For superplastic behaviour, a material must be capable of being processed into a fine equiaxed grain microstructure which will remain stable during deformation. The grain size of superplastic materials is usually in the range of 2 to 10.

Fig. 2. A typical tensile sample a) before and b) after uniaxial deformation

Structural Superplasticity

Structural superplasticity arises when a material is deformed above approximately half the absolute melting point provided it has a uniform, fine (< 10p, m diameter) equiaxed grain size that is stable during deformation. In early research this stable microstructure was obtained by heavily hot-working alloys of eutectic or eutectoid composition containing approximately equal proportions of both phases with approximately the same final grain size. In some cases phase transformation also produced the correct microstructure directly. However a low-volume fraction of stable second phase particles is also effective in obtaining a fine stable grain size provided it has a size and distribution capable of inhibiting grain growth. This has led to the development of a range of commercially important superplastic aluminium-base alloys.

Superplastic forming and superplastic materials application

Industrial application of superplasticity was first considered in the 1960's. IBM was at that time investigating both sheet and bulk forming of zinc-aluminium eutectoid for computer parts. Around the late 1970's the automotive industries starting investigating the use of zinc-aluminium alloy. The early 1970's say two interesting breakthrough: the development of the dilute superplastic aluminium alloy named SUPRAL [3] and the development of the SPF titanium technology [17].

Over the past thirty years many alloy systems have been studied ranging from the early duplex eutectic and eutectoid (Zn-Al, Al-Cu, Al-Si, Al-Ca and Al-Pd) to standard or slightly modified alloys. The alloys AA2004 [3] and AA7475 [18] are used primarily in the aerospace industry. AA5083 is used because of its superior corrosion resistance and durability. In general terms, titanium alloys are used for their high strength ratio, excellent resistance to heat and corrosion and density properties in fan blades, tanks and low pressure compressor stages, and also in exhaust nozzles. At high temperatures, titanium alloys is replaced by nickel based alloys, like for example in the high pressure compressor, combustion chamber and high and low pressure turbines. Stainless steels like jethete are used in static parts of the compressor and bearings among others. Aluminium alloys can be used in the compressor casing, inlet ring and cone, while composites may be used for the fan casing, fan blades and cowls.

Titanium alloys such as Ti6-Al4-V and Ti6-Al2-Sn4-Zr2-Mo and other Ti alloys are the most employed material in SPF formed parts. It is used mainly for casings and hot parts around engines, ducts handling hot air, exhaust nozzles and engine components in fan, compressor and auxiliary systems.

There has been a lot of argument about the mechanism responsible for superplasticity in materials and models have been proposed to go with each argument.

The conventional argument: Model types

Dislocation model

This model takes into consideration that when grain boundary slide, there is stress concentration whenever the sliding is obstructed. The relaxation of the stress concentration can be limited by the rate at which dislocations are emitted at the source, the rate at which they cross the grain or the rate at which they are absorbed into the grain boundaries. The principle of glide is assumed because of the lack of obstacles which inhibit dislocation motion. The pile-up of dislocations adjacent to the grain boundaries are thought to develop and provide a back stress against the sliding grains would have to provide more dislocation. And this will enable small increment in the grain boundary slide

Fig. 3. Showing the dislocation model theory

If grain boundary sliding is accommodated by dislocation climb and glide within the grain boundary then the afore-mentioned model cannot hold. Several arguments have been raised against the dislocation based models of superplasticity, namely:

The dislocation pile-up models do not predict a threshold stress for superplastic deformation.

There is no implicit mechanism by which the crystal lattice of either the sliding or accommodating grains could rotate in the lattice pile-up model of Ball and Hutchison [14]. Grain elongation is implicit in any model involving dislocation glide/climb on a limited number of slip systems

Dislocation pile-ups are not observed experimentally. Furthermore, at the high temperatures at which deformation takes place pile-ups would not be expected since the average stress is low.

Diffusion model

Mass transport occurs in grains due to the differences in stress dependant chemical potential on adjacent grain boundaries. Sliding is accommodated by a gradual change in grain shape as mass is transferred by diffusion. The grain boundary migration restores the original equiaxed shape of the grain but in a rotated orientation.

Fig. 4. Showing the diffusion model theory

Several objections have been to the accommodation of superplastic flow by diffusion, namely:

The diffusion paths originally proposed by Ashby and Verrall HYPERLINK ""[9]HYPERLINK "" required that diffusion takes place in different directions on opposite sides of the same grain boundary. As diffusion is driven by the stress acting perpendicular to the grain boundary this is physically impossible.

Deformation in the Ashby-Verrall [9] model is not symmetrical.

If grain boundary sliding is accommodated solely by diffusion then the lattices of the individual grains cannot rotate. The rotations shown by Ashby and Verrall are only apparent and result from grain boundary migration.

Elongated grains should be apparent in the microstructure.

The threshold stress is predicted to decrease with increasing grain size, contrary to the experimental evidence. Moreover, the predicted threshold stress is significantly less than that measured.

The grain switching event can only be invoked once giving a maximum strain of 0.55. The diffusion paths were later modified by Spingarn and Nix so that each grain within the cluster underwent the same change, maintaining symmetry of deformation and a more realistic shape transient

Grain boundaries sliding

Grain boundary occurrence in a rigid system would bring about voids in the microstructure. The voids would contract and expand with respect to the grain translation. Many superplastic materials do not cavitate so grain boundary sliding is therefore accommodated. Even when the cavities are observed their distribution is homogeneous. If the accommodation process is rapid at a high enough temperature then grain boundary sliding could be seen as the controlling mechanism.

Fig. 5. Showing the grain boundary sliding model

Despite all the models available, it has been widely agreed that grain boundary translation is the mechanism responsible. Superplastic materials are classified according to their microstructure: the banded and the equiaxed microstructure.

Fig. 6. (a) Equiaxed microstructure (b) Banded microstructure

AA8090 alloy is a superplastic material which possesses bands of similar orientation (Fig. 6)

The AA8090 has layers of materials near the sheet surfaces with a completely different microstructure and crystallography from the entire material especially that near the mid-plane of the sheet. This is due to the effect of rolling where a partially reversed shear is superimposed on the plane strain as the sheet surface are approached.


The aim of the project is to measure the local deformation and its relationship with changes in its microstructure. It has been kind of a mystery as to what actually goes on inside a superplastic material. A lot work has been done to try to understand this. By carrying out these experiments i hope to fully understand how the grains translate with respect to the deformation. To achieve this, the following objectives have to be completed;

Obtaining the stress-strain responses and m values during the tensile tests. The m value is got from the interpolation of the high and low strain rate segments of the stress-strain curves.

Calculation of texture from the electron back scattered diffraction (EBSD) data.

Tracking the offsets and quantifying the amount of deformation in a more accurate way. This is achieved by analysing the deformed grids on the surface of the specimen.

Literature review

The first observation of what seemed to be superplastic behaviour was made in the late 1920s for a Cd-Zn eutectic system at 20°C and a strain rate of about , elongation of 361% was observed and also a maximum elongation of 405% at 120°C and a strain rate of . Jenkins reported a maximum elongation of 410% for a Pb-Sn eutectic at room temperature and at a strain rate of. But the most amazing observation was witnessed by Pearson in 1934; he observed an elongation of 1950% without failure for a Bi-Sn alloy. The same phenomenon was also observed in the old Russia and the name superplasticity was used to describe elongation and ductile behaviour by Bochvar and Sviderskaya in 1945.

The word superplasticity is got from the latin prefix super meaning excess and the greek word plastikos which translates to give form to. The early Russian papers used sverhplastichnost which was later translated to superplasticity [10, 11, 12]. The thorough review of the later papers on superplasticity by Underwood [13] gave rise to the endless scientific and technological interest in superplasticity. Ball and Hutchison [14] proposed the grain sliding theory in 1969. In that year Prensnyakov published the first book on superplasticity in Russian [16, 19].

By the late 1960s superplasticity was receiving considerable research attention in both North America and the UK and the industrial potential of the property was being considered. In 1968 British Leyland described work with an industrial application: the production of panels in an essentially Zn-22 % A1 eutectoid alloy for the automobile industry and refrigerator components. More recently a number of companies have described applications for the Zn-22 % AI eutectoid, and other superplastic alloys based on copper, titanium, stainless steel and nickel have been developed for industrial applications.

Fig. 7. A tensile test specimen showing the normal and the transverse direction.

Several tests and methods have been used to try and fully understand the microstructural evolution in superplasticity. Observations are made of changes in the appearance of specimen surface.

Surface Markers

Surface markers were used to observe the displacement across the grain boundaries in material. This has helped to identify creep as a major characteristic in polycrystalline materials. There has been few investigations involving markers. The work on alumina stringers in Al-Zn-Mg by Matsuki et al [5]. This shows that offset were being developed at the grain boundaries which was translated as grain boundary sliding and grain translation. Before the tensile testing is done, one of the surfaces is polished using alumina suspension. Then the scratch marks are applied either parallel or perpendicular to the tensile axis. After deformation the surfaces are observed using a scanning electron microscopy with secondary electron imaging. And alignment of the grains are analysed and measured.

Grid method

Deformation at the small microstructural scale is best studied using the gridding technique on a polished specimen surface. The gridding is done using the focused ion beam (FIB) milling using a FEI Nova 600. There are two types of grid used: the course grid which has a spacing of 25µm and a fine grid with a spacing of 2µm. The width and depth of both of the gridding types is about 0.3µm. The grids are then examined using a scanning electron microscopy (SEM). The surface topography of the deformed specimens is also measured using a phase-shift optical interferometer. The course grids reveal the development of offsets at grain boundaries in both materials. The grain boundary offsets are measured to give an offset probability distribution. From the grid data's it is possible to get a value for the intragranular strain.

A number of studies have used the banded microstructure to argue that grain translation does not occur during superplastic deformation. One of these studies uses the alignment of similarly oriented subgrains as internal markers. It was argued that the grain translation model proposed by Ashby-Verall [9], would destroy such alignment after undergoing strains. The markers do not also the rotation of grains.

With the AA8090, it was proposed that superplasticity was a result of high strain rate sensitivity intragranular slip. As the rate sensitivity of slip increases, the texture changes due to slip within each grain decreases. The plastic anisotropy of the individual grains does not change.

Key equipment and procedures

The equipment that will be used this work are as follows:

Focused ion beam milling using FEI 600

Scanning electron microscopy

Superplastic tensile testing machine

Electron back scatter diffraction (EBSD) map

Focused ion beam milling

The focused ion beam miller uses gallium () primary ion beam which hits the sample surface and sputters a small amount of material, and leaves the surface as either secondary ions ( or ) or neutral atoms (). The primary beam also produces secondary electrons (). As the primary beam rasters on the sample surface, the signal from the sputtered ions or secondary electrons is collected to form an image.


Fig. 8. A diagram of the focused ion beam (FIB) milling [6]

The controlled grids are done with the FIB milling using FEI 600 prior to tensile testing. This is a significant process in the analyzation of the sample specimen.




Penetration depth


Elementary charge

particle size


velocity at 30 kV

velocity at 2 kV

Momentum at 30 kV

Momentum at 2 kV




In polymer at 30 kV

In polymer at 2 kV

In iron at 30 kV

In iron at 2 kV

Ga+ ion


0.2 nm

1.2 .10-25 kg

2.8.105 m/s

7.3.104 m/s

3.4.10-20 kgm/s

8.8.10-21 kgm/s

nm range

up to 30 kV

pA to nA range

60 nm

12 nm

20 nm

4 nm

Fig. 9. Table showing the specification properties of a focused ion beam milling machine.

Scanning electron microscopy

The scanning electron microscopy of the model FEI Sirion FEG-SEM will be used to analyse the bulk texture (the grains and the grain boundary) with the HKL channel 5 electron back-scattered diffraction (EBSD) acquisition software and with a step size of 0.5. The data's obtained are plotted on VMAP software. The types of signals produced by an SEM include secondary electrons, back-scattered electrons (BSE), characteristic X-rays, light (cathodoluminescence), specimen current and transmitted electrons. SEM magnification  can be controlled over a range of up to 6 orders of magnitude from about 10 to 500,000 times. SEM's may have condenser and objective lenses, but their function is to focus the beam to a spot. In a SEM, as in scanning probe microscopy, magnification results from the ratio of the dimensions of the raster on the specimen and the raster on the display device. Assuming that the display screen has a fixed size, higher magnification results from reducing the size of the raster on the specimen, Magnification is therefore controlled by the current supplied to the x, y scanning coils, or the voltage supplied to the x, y deflector plates, and not by objective lens power. Samples must be of an appropriate size to fit in the specimen chamber and are generally mounted rigidly on a specimen holder called a specimen stub. Several models of SEM can examine any part of a 6-inch (15 cm) semiconductor wafer, and some can tilt an object of that size to 45°. SEM specimens must be electrically conductive, at least at the surface, and electrically grounded to prevent the accumulation of electrostatic charge at the surface. Objects made of metal require little special preparation for SEM except for cleaning and mounting on a specimen stub. Nonconductive specimens tend to charge when scanned by the electron beam, and especially in secondary electron imaging mode, this causes scanning faults. They are therefore usually coated with an ultrathin coating of electrically-conducting material, commonly gold, deposited on the sample either by low vacuum sputter coating or by high vacuum evaporation.

To be able to fully collate the grain orientation correlation the electron back scatter mapping is used together with the scanning electron microscopy (SEM).

Electron back scatter diffraction (EBSD) mapping

Electron Backscatter Diffraction (EBSD) obtains crystallographic information from tilted polished samples in a scanning electron microscope (SEM). The electron beam strikes the sample and the electrons are elastically scatter beneath the surface. It utilises the principle of Bragg's law. The EBSD is a kind of map which tells us more about the bulk texture. A sample area is designated and collection parameters selected (step size, area, sampling method). The system will then automatically raster the electron beam accordingly, and at each point information relating to the EBSP at each raster point can be recorded to create a dataset (e.g. crystal orientation, pattern quality). From this data that is obtained numerous maps, charts and plots are generated. Some of these include grain orientation maps, grain boundary maps, image quality maps, grain size charts, misorientation charts and texture plots.

Superplastic tensile testing machine

The superplastic tensile tester that will be used is made by Alcan International Limited. The difference between this machine and an ordinary tensile testing machine is it has a compartment where the specimen can be heated and controlled.

Tensile tests are usually carded out on strip or machined samples with either circular or rectangular cross section. Test pieces are screwed into or gripped in jaws and stretched by moving the grips apart at a constant rate while measuring the load and the grip separation. This data is plotted as load against extension and then converted to engineering stress (load/original area) against engineering strain (fractional change in length over the test section assuming the deformation is uniform). The actual stress and strain may be calculated if the true cross section is measured during the test.


Electron back scatter diffraction (EBSD) mapping

Electron Backscatter Diffraction (EBSD) obtains crystallographic information from tilted polished samples in a scanning electron microscope (SEM). The electron beam strikes the sample and the electrons are elastically scatter beneath the surface. It utilises the principle of Bragg's law.

Result expected.

The work i am about to undertake which will involve applying of the grids unto the transverse plane of the specimen and deforming it, has never been done in that manner but we expect to get results similar to that which has the grids applied to the normal direction of the specimen.

Firstly to overcome the effect of different textures at different planes, the will be machined using a fly-cut milling this is to remove 1mm from each rolled surface. The result leaves a material which is close to being homogeneous in thickness through-out in thickness in all directions.

Tensile deformation

The tensile specimen are heated to test temperature and held for 20mins prior to deformation. One of the specimens will be tested to failure with an alternating strain rate which switches between strain intervals. The interpolation of the high and low rate segments to give a continuous low and high rate curve will allow the strain rate sensitivity (m-value) to be determined.

The other specimen will be deformed at a constant rate of. These specimens are now used to provide samples for examination and those samples are taken from the minimum cross sections region of the deformed gauges. The minimum cross sections are now measured and are used to correct the stress-strain data for effects of inhomogeneouity in deformation. The tensile behaviour of the AA8090 is shown in fig. 10.

Fig. 10. Showing the stress -strain and m value for uniaxial tension of AA8090 tested in the rolling direction at 530°c [7]

The m-values are shown on the same graph and are constant with strain. There will be some difference before and after the correction has been made. After the correction the curve is expected to rise to a maximum stress and a strain of about 0.3 the and there is a decrease in the stress which indicates that there will be a small increase of the strain towards the maximum strain. There will be a strain hardening behaviour as shown in fig. 10. The plastic strain hardening and the strain hardening index will be derived from the stress-strain data. It is discovered that the hardening index is relatively high before reducing to a negative value at strains above 0.4.


The surface of the specimen gauges are ground and polished before the grids are milled unto it. They are then examined using the SEI Sirion field emission gun-scanning electron microscope (FEG-SEM) which is equipped with an HKL electron back scatter diffraction (EBSD) acquisition system. The fig. 11. Shows an example of what the orientation map is to look like.

Fig. 11. Orientation maps from the SEM/EBSD data, showing the microstructural development due to tensile deformation in AA8090 [7].

The banded microstructure we have at the beginning will be transformed to an equiaxed structure.


Measurements will be taken with a coarse step size (2µm) over fields of about in the transverse plane. The data's will be used to derive the orientation distribution functions (ODF) using the harmonic series method with series truncation and an assumption of orthotropic specimen symmetry. There will be a change of texture associated with a reduction in the maximum orientation density.

Surface grids

The grids show development of offsets at grain boundaries in the specimens. One of the most notable observations is the appearance of the extension zones that occur at the grain boundaries. Sufficient information is extracted from interference topography and from the grid results. The offset of the grids will be measured and used to produce offsets probability distributions. From the data inferred from the grids, it is possible to determine a value for the intragranular strain.

With the Al-Cu-Zr alloy, during the early stages of straining, induced crystallization took place. This was attributed to strain-enhanced subgrain growth with the development of increased misorientation at the subgrain level. This process was used to produce a fine crystallized microstructure which possesses fine superplastic behaviour. If superplasticity is defined as the phenomenon where strain localization is resisted by an unusually high train rate sensitivity for crystalline solids then the results here show that the materials is superplastic with the banded microstructure. During superplastic forming, the rate of diffusion and boundary migration changes due to the deformation. The grain translation as the same order as the grain switching model of Ashby and Verrall [9] and its derivatives would be expected with grain sliding. This happen because the grain centres would separate along the tensile axis and this results in gaps. The resulting gaps would fill the grains by movement sideways in the direction of the sheet plane normal.

If grain translation were occurring it would have an effect on the profile of the layered structure near the midplane of the material used. A lack of grain translation shows that the grain boundary sliding cannot be the main mechanism for deformation. Rotation due to grain boundary sliding could occur even without sliding causing grain translation. Often the reduction in texture is seen as evidence of grain boundary sliding. There is some evidence for the occurrence of grain boundary sliding in alloys. This was observed by Chokshi and Mukherjee [8], from measurement of the displacement of the surface markers lines, analysed that grain boundary sliding was responsible for 75% of the total straining under optimum superplastic conditions. This was after a tensile pre-strain of 0.6, by which stage the layered structure would have disappeared.

The misorientation of the boundary is likely to affect the sliding potential. In aluminium boundaries with misorientation of less than 5 degrees will not support grain boundary sliding. Then at orientations greater than 5 degrees, the sliding rates increase with an increase in the orientation angles. In the middle layer of the sheet, the high angle boundaries according to observations would be most likely to slide but lie predominantly low in the sheet plane and therefore would be subjected to a very low resolved shear stress. The increase in the population of the high- angle boundaries in spatial orientation which occurs at large enough strains would be expected to give a significant increase in the rate of sensitivity. This is because grain boundary sliding must occur at a lower stress than other mechanisms.

In general if is needed to be rejected as a significant aspect of superplastic deformation then the texture change must be explained by a different mechanism.

Time schedule

In the course of this work, i shall be carrying out series of experiments such as micromilling, Scanning electron microscopy. All these, reporting the observations and the final write up will have to be carried out properly and efficiently with respect to time. Below is a gannt chart showing how i fully intend maximise my time.