Materials Engineering UK

Welcome to MatEngUk!

Marc Graham — Tue, 02 Jun 2009 15:00:00 +0100

Hello and Welcome to the newest thing in materials engineering to hit the web since a photon hit an atom! Hopefully I can shed a bit more light on a few subjects that I’ve written about over the past few years as a Materials Engineering Student at Loughborough University. The Idea of this is to spread a bit of the knowledge that I have gained over my time as a student and also to help others in their work themselves.

Here is a few to get you started, I hope they are up to the standard of what you come to expect from university students, many thanks for looking, and I hope you enjoy the content!

Advanced Testing and Characterisation

Marc Graham — Tue, 30 Jun 2009 11:28:00 +0100

Advanced Testing and Characterisation

Artefacts study

Introduction

The common DVD player works in much the same way as a CD player. Just like its sister product, the information on a DVD is coded as a series of pits on the disks surface. These pits are arranged in a spiral structure in order to be read off by a laser within the DVD player. This laser is the key reason why DVDs have superseded CD’s in the world of data storage. DVD’s use a shorter wavelength for the laser that enables it to place the pits in a much denser arrangement, therefore enabling the same area of disk to hold much more data.

Within a DVD player, its main components consist of a disk drive mechanism, which uses a spindle to hold the disk and a motor to spin it, a printed circuit board, which contains all of the electrical components to convert the data being read into a usable format, and an optical system assembly, which is the laser that reads the information off the DVD. All of these parts can be broken down into their component pieces, which will consist of a range of metals and polymers that have been specifically chosen for their use by their physical and sometimes electrical properties.

There are two types of plastics in production today, the first of which are classified as thermoplastics, they fall under this classification if they possess the properties of being remouldable under heat and pressure. Thermoplastics, such as polyethylene and polypropylene are used in production in low stress applications where they are not put under any load as thermoplastics have a tendency to display creep when under extreme loads.(1) The other category is Thermosetting plastics, these are polymers which form a highly crosslinked molecule upon heating. The reaction is irreversible but the highly crosslinked structure adds properties of heat and chemical resistance, and also a high dimensional stability due to the chains being unable to slip therefore reducing ductility.

In this experiment, 9 parts have been taken from the internals of a DVD player to be analysed with respect to what material they are made from and how they were processed during production. The metallic parts of the DVD player are to be mounted in Bakelite on a metallographers mounting press, whilst the plastic elements can be Identified using the process of Simple Identification, whilst any hard to identify metals can be analysed using the Transmission Electron Microscope. The TEM works on the same basic principle as a light microscope but instead of light, it uses electrons. A TEM uses the much lower wavelength of electrons to get a resolution nearly a thousand times better than a conventional microscope. It uses electromagnetic lenses to focus a beam of overhead electrons into a very thin beam.(3) This beam then travels through the specimen you want to study and then any unscattered electrons hit a fluorescent screen which shows the image as dark and light density patches.Four metallic samples and five polymeric samples were chosen to give a wide range of materials, giving a large spectrum of possible results.

Experimental Procedure

The testing of the polymeric samples was done using a process called Simple Identification. This involves a number of simple and conclusive tests that, when referenced against a table of results, a clear material can easily be seen.

The first test was a density test to measure whether it floated or not. Care was taken to ensure that surface tension was broken and that no air was trapped inside the specimen. The next test for the polymers was a test called the Beilstein test. This involved heating the end of a copper wire in a Bunsen flame until the flame turned colourless, using this hot wire to burn through the polymer on test left an amount of residue on the wire. When put back in the flame, the presence of halogen within the polymer could be seen if the flame turned green/blue. Testing after this was more of a visual test rather than chemical, firstly the specimens were tested to see if they could be torn or cut easily, this gave another property that could be cross referenced to eliminate some polymers. The heating test was also used, this was useful to gauge whether or not the polymer was a thermoplastic i.e. whether it melted or not, and also, using litmus paper, the pH of the reaction in the escaping vapours could be measured. Lastly the sample was left in the Bunsen flame to burn, this allowed us to see whether the sample was self extinguishing, what colour the smoke that was given off was, whether it gave off particles, and also what kind of smell the smoke had as this is an easy way to detect some polymers, such as PE smelling like candle wax.

An alternative and altogether more accurate way of testing a polymer to determine its molecular make up is by means of Infra Red Spectroscopy. The use of infrared rays is based on the principle first given by the Beer-Lambert law.(2) This states that the amount of rays absorbed is wholly dependant upon the chemical composition of a given material. When the material is hit with the rays, the outer electrons get excited. This only occurs at differing energy levels. When testing polymers, infrared is used at a frequency of 0.8-25 µm, which correlates with the frequency required to excite the outer electrons. This leads to the infrared light getting absorbed by the outer electrons, which creates a lower density of that frequency which can be detected and set on a graph.(3) This printout can then be compared to past printouts to determine by which peaks and troughs are evident, which material it is. This process was needed for a sample that was an elastomer band used for driving a pulley system within the DVD drive.

In order for the metallic samples to be properly analysed under the microscope to determine structure and composition, they had to be mounted and polished. The samples were first cut into small pieces, mounted in Bakelite using the metallographers mounting press, and finely polished. The polishing process was time consuming but necessary, starting off with coarse silicon carbide paper and working onto finer and finer grades, the scratches were finally rubbed out of the surface of the metal using the polishing wheels. Care was taken to ensure the correct polishing wheel was used depending on whether or not the metal was ferrous or not. This was first tested using a magnet. Using initial estimates as to what the metal was, the correct etching liquid was used to finely etch the metal to bring out the internal structure more clearly. The samples were then viewed and photographed under a microscope to determine what the metals were by their structure.

One metallic sample that appeared at first glance to be a steel of some kind did not respond to the etching agent of 2% Nital, so the agent was changed for 10% Nital. Again, the metal remained unchanged and the structure did not appear, at this point the metal could’ve been a number of things including a type of Stainless Steel or possibly a Chromium alloy, so to determine exactly what it was it was sent to the Transmission Electron Microscope.

Results

On looking at the first of the metallic samples, which was a screw threaded piece that rotated to move the laser reader, it was noticed that where the sample had snapped off of the original piece, the break revealed a structure not dissimilar to that of a ceramic which led us to believe that the process by which it was made was that of a powdered metal possibly by injection forming. It was not until the molecular structure was revealed after the etching in Carapellas Reagent , which is made up for ferric chloride, hydrochloric acid and alcohol, that this was found not to be the case. The screw was found to be a Brass of composition 60% Copper, 40% Zinc.

(1) (2)

The second piece that was also suspected to be brass was an element of the electrical side of the DVD player, during the process of mounting it in Bakelite and polishing it, a coating was rubbed off leaving a brass coloured metal, the coating can be seen on the side of figure (2), but due to the extremely thin section of metal that was left on the side, the composition of the metal coating could not be deduced. Looking at the molecular structure when comparing it to the other brass section we can see there is only one phase, leading us to the conclusion that it is a brass of composition 70% Copper, 30% Zinc.

Brass in its standard form is mostly a solid solution of zinc, dissolved into copper. It can be seen from the microstructures of the second brass sample that around the 30% zinc mark, the zinc atoms can fit in amongst the copper atoms without disrupting the crystallographic arrangement of the pre-existing copper atoms. As the percentage of zinc increases within the brass, the mass density, colour and average molecular spacing gradually change, saying this, it is worth noting that the shape of the actual structure and how the atoms are positioned stays the same. Looking at the first brass sample, where the amount of copper gets to around the 40% mark a new phase starts within the copper-zinc solid solution. The first phase no longer increases their zinc content as the percentage of zinc increases, it merely makes the molecules in the second phase larger and more numerous

An interesting thing to notice on the 70:30 copper in figure (1) is the twinning that has occurred on some of the grain boundaries. This is when a part of the atomic lattice is deformed so that it forms a mirror image of the undeformed lattice next to it. This process leaves small but well-defined regions of the crystal deformed. This does not distort the properties of the metal by too large an amount but in the process of forming twins within the structure it places some slip systems in a better direction for shear stress to affect the metal, so can indirectly increase the chances for additional slip within the structure. This can create a more ductile metal which is not always wanted.(4)

The first of the steels is the outer casing for the internal motor that was used to drive the wheel that moved the laser along a screw thread. This was found to be magnetic and upon inspection the results pointed to low carbon steel. Looking closer at the shape of the grains, they are not elongated or equiaxed, which leads us to believe that the steel has been annealed. Annealing has the effect of getting rid of any internal stresses in a metal, creating a metal with better stress resistance properties. You can see on the structure photograph the sample contains both pearlitic and ferritic regions, although being a low carbon steel, there is a lot more ferrite. The darker areas of the picture are the pearlite regions, and the lighter areas, the ferrite.

The second steel that was tested had to be sent to the TEM to get analysed due to the fact that none of the etching techniques managed to bring out a usable structure. The results were as follows:

Element	Wt %	At %
SiK	0.59	1.16
MoL	0.69	0.4
CrK	12.61	13.38
FeK	86.1	85.06

The results of the scan show the metal to be a ferritic stainless steel. Stainless steel is defined to be an iron-carbon alloy with at least 10.5% Chromium content, it becomes highly resistant to corrosion after 12% due to a protective film of chromium oxide that forms on the metal surface.

For the polymeric samples, most were tested using the process of simple identification, but due to the nature of the elastomer band, Infra red spectroscopy was used to gain a better understanding of what the material was. Below are the results for the simple identification tests that were carried out on the other polymer samples.

Sample 1 – Mounting base	Sample 3 – Front Display
Hard to cut, Brittle	Brittle
Bubbling flame.	Floats
Sinks	Black smoke
Flammable	Self extinguishing
Sample 2 - Fan Casing	Sample 4 - Motor housing
Cuts but brittle	Sinks
self extinguishing	Yellow flame upon burning
Green flame on Beilstein test	Cuts easily
Sinks
Turns litmus Red

For the first sample that was tested you could tell from the marks made by the ejector sprues that the piece was injection moulded, so immediately this suggests a thermoplastic. When flexed, the material showed signs of whitening, again indicating a thermoplastic. Another indication of this is that it softened when heated, indicating that is was a semi crystalline thermoplastic. When put into water, the polymer showed signs of having higher density than water so we can rule out it being a polyolefin. From this information and the fact that it was brittle this sample is simply Polystyrene.

The second sample is slightly anomalous, the signs point to a polyolefin, but this clashes with the fact that it is self-extinguishing. On closer inspection it was found to be a three part lamination which incorporated a film of aluminium foil which would explain the polymers ability to self extinguish.

The third sample, the fan casing was similar to the first sample in the fact that it was injection moulded. This therefore indicates a thermoplastic. Also, looking at the fact that the polymer could be cut relatively easily points us to a polymer which is highly amorphous in its composition. The red litmus paper is a sign that the fumes given off are of an acidic nature, this combined with the green flame in the Beilstein test leads us to the conclusion that it must be Polyvinyl Chloride.

The fourth sample was a surround for a small electric motor, again it was injection moulded and showed signs of being a semi crystalline thermoplastic. On the same premise as the first sample I would be confident to say this sample was Polystyrene as well.

The fifth polymer to be tested was tested using Infra Red Spectroscopy. It is a highly cross linked elastomer which proved very difficult to determine what it was. Below is the graph and results of the IR test.

Initially it had a definite resemblance to raw butadiene, but due to the fact this material was not used in these kinds of applications, that was ignored. Nitrile rubber was another possibility that would be very suitable to the application, but there was no peak representing that. There is a large peak at 1000 suggesting some kind of silicon rubber, but this had to be discounted for being too far out of the plausibility range. The only plausible material it could be is a highly crosslinked butadiene polymer.

Discussion

When designing a product to go into production, many different things have to be looked at when trying to pick the correct material for your application. What stresses are going to be applied to it, how the component might fail, how it will be made to a high level of quality but with the lowest level of cost, and many other factors. In this experiment just 9 pieces were taken from the hundreds that were in the DVD player to analyse.

Starting with the first piece, the 60/40 brass was most likely rolled on the production like due to the fact that when you look at the grain structure inside, all the grains are in 1 direction. This creates a strong and standardised structure within the screw that has high tensile properties and as they will be put under constant torsional load as they turn. This piece is a key part to the DVD player and so has to be designed so it doesn’t corrode or break. Brass is a good choice as it is non corrosive and yet still cheap to manufacture.

The second piece of Brass was part of an electrical connection which had a metallic coating to it which was most likely applied by the process of electroplating. Making the brass element the cathode in the reaction allows it to be evenly coated with whichever metal the producer deems necessary.

The steel casing for the motor is a very simple press forming involving probably thousands of identical parts being punched out of a large sheet of thin steel sheet at the same time. Low carbon steel was used for this application due to its low cost and ductile structure being easy to form into different shapes.

The stainless steel element was part of a runner that the laser moved upon inside the DVD reader. This is a vital component that needs to be smooth at all times, even if some moisture were to get into the DVD player, if this part were to corrode, the player would stop working. Stainless was used to stop any corrosion, and although it is slightly more expensive, the cost is worth the peace of mind. The stainless steel would be made through a process of drawing.

Both of the Polystyrene pieces will have been made in the same way. Injection moulding, although the initial cost is high for the mould tool, is a very quick and effective way of producing many identical and intricate parts. Injection moulding works best with thermoplastics of a lower molar mass than average, ensuring a steady flow is maintained.

Conclusion

Upon breaking down a DVD player into its smaller component parts you can tell by very simple tests what the material is and why the manufacturer chose it. The process of Simple ID and IR can tell most polymers apart, but not always conclusively, additives can not always be recognised.

References

William D Callister Jr – Materials science and engineering, an introduction – ISBN 0471320137

Vernon John – Introduction to engineering materials – ISBN 033394917x

Wiliam F Smith – Foundations of materials science and engineering –

ISBN 0070592920

Chapman and Hall – Materials Science – ISBN 0412341506

Dietrich Braun – Simple methods for identification of plastics –

ISBN 1569902801

Precipitation Hardening of a Heat Treatable Alloy

Marc Graham — Tue, 30 Jun 2009 11:25:00 +0100

Precipitation Hardening of a

Heat Treatable Alloy

MPB014

Abstract

When looking at Aluminium alloys, the process of Precipitation Hardening is a relatively fast and cheap way of increasing the strength and hardness. This comes hand in hand with reduced ductility and toughness, and if aged for too long, the alloy can be subject to a process called over-aging whereby a visible precipitation appears in the microstructure and the hardness levels start to decrease.[1]

As the samples were left in the furnace for longer and longer amounts of time, precipitates of Mg2si appear within the structure of the Al. This precipitate is much harder and more brittle than the Aluminium which explains the rise in hardness figures.

Introduction

Precipitation hardening is a widely used process to increase the hardness in a usually non-ferrous metal that have high level of malleability. It was first introduced in the early 20^th century on a material called Duralumin by a professor in America. In essence, the process produces a uniform dispersion of a very fine coherent precipitate in a softer, more ductile base matrix.

The process has three stages, Solution treatment, Quenching, and Ageing. In the Solution treatment stage, the alloy is heated and held above the solvus temperature to create a solid solution. This stage dissolves the precipitate and creates a homogeneous un-segregated solution that is the starting point of the process.[2]

The quenching stage is just cooling the alloy very rapidly. This forces the particles in solution to solidify as they are as they don’t have time to diffuse to potential nucleation sites. After the quench, the structure contains Aluminium particles with Magnesium silicate in solid solution. The Ageing process is where the supersaturated solid solution is heated below the solvus temperature, this allows some magnesium silicate atoms to diffuse to nucleation sites and grow.[2] The best properties can be obtained by stopping this process before it gets to the stage of Over-ageing.

The industrial process of making aluminium drinks cans utilises this process extensively. To obtain the perfect can, the compromise between the amount of hardness needed to keep the can stable and the amount of ductility needed in order that the can form properly is vitally important.

Cans are produced by a method called Impact-Extrusion, whereby a punch is driven into a circular slug of Aluminium creating the base and sides all in one piece.[1] If the Aluminium alloy used is too hard, there will be cracks appearing in the furthest corners of the base, but if the aluminium is too soft, a phenomenon called ‘earing’ occurs, where metal is forces out the top and over the edge, forming a lip that needs to be trimmed off before the can can be used.

Experimental Procedure

Eight identical pieces of Aluminium alloy HS30, with composition within the limits of 0.4-1.5% Magnesium, 0.6-1.3% Silicon, 0.6% maximum Iron, 0.4-1.0% Manganese (the balance being Aluminium) were placed in a holding furnace at 525°C.

These were solution treated in the furnace for 15 minutes, whereupon seven of the samples were removed from the furnace and quenched in cold water. The furnace was then switched off and the eighth specimen allowed to cool inside the furnace very slowly. Care was taken to ensure that the temperature did not rise above the maximum solution temperature for the alloys as it is very close to the solidus of the Aluminium alloy.

One specimen was tested as a freshly quenched sample, the others were put in another furnace at 200°C and individually taken out to be tested at intervals of 2.5, 5, 10, 20, 40, and 80 minutes. These samples were cooled in water before each test.

The samples were tested using a Vickers Hardness machine with a 5kg load. Three hardness impressions were made on the outside of each sample, taking the average as our usable result.

The ductility of each sample was obtained using an Erichson cupping tester. Each sample was tested by forcing a ball bearing to deform the surface, stopping when the first cracks appeared. The height of the deformation was measured, thus giving a ductility result.

Results

	1	2	3	Avg	Ductility (mm)
0 (Water Quenched)	49.70	48.00	52.40	50.03	9.61
2.5	53.90	55.30	56.80	55.33	9.59
5	63.80	65.80	64.10	64.57	9.47
10	84.30	82.40	82.10	82.93	9.08
20	91.30	92.50	90.90	91.57	8.92
40	91.40	94.20	95.80	93.80	8.85
80	89.90	91.30	88.70	89.97	9.50
160 (Furnace Cooled	37.60	37.20	36.90	37.23	10.57

Discussion

The process of precipitation hardening is a solution treatment process. In this process, the alloys is first heated above the solvus temperature and held until a homogeneous solid solution is produced. This step reduces the segregation within the original alloy.

When the aluminium alloy is heated to 525°C in this process, the Magnesium and silicon molecules react and combine to form Magnesium silicate or Mg2Si. At this point, the Mg2Si is in solution with the Aluminium, much like salt in water. When the first sample is quenched in water, the magnesium silicate particles do not have time to diffuse to potential nucleation sites so are trapped as a solid solution. This mean the Aluminium is at its purest form in the freshly quenched sample as the impurities are trapped outside the aluminium grains. This sample is different to the others as putting the others in an oven at 200°C artificially age-hardens them.

As can be seen in the graph, at the 2.5 minute stage, the hardness increases, this is due to very fine particles of Mg2Si coming out of solution and appearing within the Aluminium grains. These extremely fine grains are much harder and more brittle than Aluminium, and as these are now within the grains of Aluminium, the properties of the alloy start to change.

As the time continues, more and more Mg2Si appears creating a harder and harder alloy, with ever decreasing ductility. This continues until a point on both of the graphs at around the 50 minute mark where the hardness starts to decrease and ductility starts to increase. This is the point where the alloy moves into the stage of Over-ageing. This is where the alloy has been in the furnace for too long, and the particles of Mg2Si come together and form course particles.[4] The particles are larger than they were in previous stages, but the process of clumping together means that there is less surface area exposed. This in turn means that there is more Aluminium particles exposed, so the alloy starts to return to the original properties of Aluminium, hence the higher ductility and lower hardness.

Conclusion

As the process continues, the alloy releases Mg2Si
Mg2Si is harder and more brittle, so as more is produced, hardness goes up, ductility goes down.
This happens until over-Ageing occurs at roughly the 50 minute stage.
After this, more Al is exposed aiding ductility but hampering hardness.

References

The Metallurgy of Aluminium and Aluminium Alloys - Robert John Anderson,
The Science and Engineering of Materials - Donald Askeland
An Introduction to Materials Engineering and Science for Chemical and Materials Engineers - Brian S. Mitchell
Foundations of Materials Science and Engineering - William Fortune Smith

Surface Flaws in Glass

Marc Graham — Tue, 30 Jun 2009 11:25:00 +0100

The effect Etching with Hydrofluoric Acid has on

Surface Flaws in Glass

Introduction

Glass is used abundantly in everyday life all through the world. It is a versatile and useful material which can be formed into different shapes and sizes very easily. Glass in its purest form is an exceptionally strong material, unfortunately, pure, unflawed glass is extremely hard to produce and is not available on a large scale. What gives glass the brittle properties we know it to have are the surface flaws. These small cracks are caused by chemical or mechanical damage to the surface. These cracks are weak spots and can propagate causing the glass to fail.

These cracks have a very small tip radius and create high stress concentrations at the tip end. This is calculated by the formula:

1 + 2 (√l/R)

Where l = depth of the crack

R = radius.

Experiment Procedure

Involving acid in an experimental procedure meant that safety rules had to be adhered to at all times. These involved the used of rubber protective gloves at all time as Hydrofluoric acid is a highly dangerous and corrosive substance which can injure easily. Spillages need to be made visible by covering the work surfaces in tissue. Any contact with the acid must be washed thoroughly and be treated with magnesium paste and glycerol.

The main aim of the experiment is to see the effect hydrofluoric acid had on the surface flaws on glass. The Glass samples were sectioned into groups of three that would be subjected to the acid for differing intervals. One group was left un-etched, the other groups, etched for five, ten, and fifteen minutes respectively in hydrofluoric acid.

To measure the load the glass can take before breaking, the Hounsfield Tensometer with 1000N and three point loading attachment was set up.

Using a micrometer, the diameters of all the specimens were measured, creating an average for each piece. This was needed in order to work out the tensile stress for each sample.

The experimental procedure was as follows;

Place group 1 samples in a beaker of 6% Hydrofluoric acid solution for five minutes.
Remove samples from solution and wash thoroughly in alcohol.
Dry samples with hot air dryer.
Repeat process for both the ten minute group, and fifteen minute group.
Position samples in Hounsfield tensometer three point loading attachment.
Cover with a sheet of polythene and test the sample to breaking point.

To work out the tensile stress, the following equation is used;

F = (8WL)/( πD^3)

W = Applied maximum load

L = Distance between pair of supports = 30

D = Diameter of sample

Results

Sample	Dia (mm)	Dia (mm)	Av Dia (mm)	Max Load (N)	Tensile Strength (Nmm^-2)
Unetched
1	4.96	4.95	4.96	299.0	187.192
2	4.95	4.96	4.96	307.0	192.201
3	4.96	4.96	4.96	285.0	178.427
5 min etch
1	4.96	4.96	4.96	460.0	287.988
2	4.95	4.95	4.95	175.0	110.226
3	4.93	4.93	4.93	401.5	255.980
10 min etch
1	4.96	4.96	4.96	318.5	199.400
2	4.95	4.96	4.96	382.0	239.155
3	4.96	4.96	4.96	779.5	488.014
15 min etch
1	4.93	4.93	4.93	551.5	351.614
2	4.95	4.96	4.96	398.0	249.172
3	4.96	4.96	4.96	671.5	420.400

Conclusion

The main conclusion that was found through doing the experiment was that the longer the glass was exposed to the acid, the higher the tensile stress became, This is due to the acid eroding away the tip of the crack to create a larger radius at the end. This process of widening the radius of the tip lowers the stress concentration at the point where the crack would normally propagate, so it takes more to make the crack grow.

When compared, glass and metals have roughly the same tensile properties, but due to the fact that glass has a very small amount of plastic deformation, when a force is applied, it acts brittle, whereas in metals, the plastic property allows the cracks to propagate much slower and this will only happen when the elastic limit is reached.

From the graph we can see that there are a lot of scattered results. This is due to the glass samples having many different crack arrangements within them. To have a fair test the amount of samples would have to be vastly increased and then averaged to create a better understanding of how much the Hydrofluoric acid affects the tensile Properties of Glass.

References

Brian S Mitchell : An Introduction to Materials Engineering and Science for Chemical and Materials Engineers

Milton Ohring : Engineering Materials Science - Page 160