MY3200 Quantitative Metallography - Laboratory 1
The main purpose of this laboratory exercise is to familiarise you, the student, with the process of preparing a sample for observation using the metallurgical microscope and to quantify the microstructure that is observed. Each student will prepare a sample of steel. The steels supplied range in carbon content from 0.05 to 1.0 wt%, some are plain carbon steels, others are low alloy steels. The steels are available in a number of heat treated conditions, including as fabricated (bar stock), annealed, normalised, quenched and finally quenched and tempered.
A secondary objective of the this and the following laboratories is to observe how (a) alloy composition changes the microstructure for a given heat treatment, (b) how heat treatment changes the microstructure for a given composition and (c) how structure is related to mechanical properties such as hardness, tensile strength and ductility.
Sample Preparation
The micro or macrostructure that is observed often depends on the section that is viewed, so selecting a plane to observe is important. In this exercise the transverse section of bar stock will be examined.
In general, sample preparation of opaque materials such as metals involves obtaining a flat mirror-like finish which can be examined by reflected light.
The sequence of operations is typically as follows:
(a) If the sample is too small or has an awkward shape, it should be mounted in hot or cold setting plastic for easy handling. Note that hot mounting raises the temperature of the sample to 150°C for about 15 minutes and this can effect substantial microstructural changes in some materials
(b) By means of a belt sander or wet abrasive wheel, grind a flat surface on the sample and bevel the corners and edges. Avoid heating or deforming the sample.
(c) Using wet silicon carbide papers of successively finer grades, obtain a finely ground surface covered with even parallel scratches from the smoothest paper. Excessive pressure is not required. It is important to wash the sample and your fingers before moving on to a finer paper in order to avoid carrying over the coarser abrasive particles. The common rule is to grind on each paper for twice as long as it took to remove the scratches from the previous grade, turning the specimen through approx. 90° at each change.
(d) Having thoroughly washed away all traces of grinding material, a final surface is prepared by polishing the surfaces on a polishing wheel which is covered with a fine abrasive paste or slurry. A variety of polishing cloths and polishes are available commercially.
The polishing wheels provided are to be used with a suspension of 6µm diamond paste in kerosene followed by 0.05µm Al2O3 in water. The final polish will give a satisfactory mirror surface, but again it is important to clean the sample and one's fingers to avoid carrying over coarse particles to the final wheel. Again, excessive pressure is not required.
Etching Sources of Contrast
Following the preparation of a near-perfect mirror surface, it is usual to etch the specimen lightly to reveal details of the microstructure. Metals are opaque to light of visible wave lengths and the contrast across the surface is obtained by reflection. There are three principal sources of contrast:
A. Chemical attack may be sensitive to the crystal orientation, leaving exposed small facets of planes of certain {HKL} indices. The result is that the light is reflected from adjacent crystals in different directions and consequently reveals the grain structure as areas of variable darkness.
B. Parts of the structure of high energy are the grain boundary interfaces or two phase interfaces. Depending on the nature of the chemical attack, these regions may be preferentially dissolved so that their intersections with the polished surface are delineated as fine grooves or line traces.
C. Variation in chemical composition within an alloy inevitably means that the etchant attacks some regions more rapidly than others, either removing one phase preferentially, or staining parts of the structure to give color contrast. A mixture of two phases, a and b, might then etch to give a stepped profile.
D. Note that in addition to A, B, and C, surface contrast may also be obtained by optical methods such as the use of polarized light or interference techniques.
Materials which have non-cubic crystal structures rotate the plane of polarization of light in transmission or reflection depending on the orientation of the crystal with respect to the optical axis. In reflection a sample does not need to be etched to produce contrast.
The incidental light is plane polarized by insertion of a "polarizer" slide into the optical path while a second slide is inserted in the optical path behind the reflector and can be rotated axially; this is called the "analyser." If the planes of polarization of the polarizer and analyzer are crossed at right angles, no light will pass through to the eye unless light from the samples has been rotated in transmission or reflection.
Microstructural Observation
Obtain a representative area and produce a micrograph showing the overall structure. A second, higher resolution picture might be required to show detail.
The grain size and volume fraction of ferrite/pearlite in the annealed/normalised steels must be measured. Grain size is measured using a linear intercept technique, the volume fraction of each of the microconstituents is measured using point counting.
1. Point Counting.
In the point counting method a square grid is superimposed over the sample microstructure. The grid spacing should be similar in size, or smaller than the features being measured. The number of points to be counted depends on both the volume fraction of the phase and the desired level of precision in the result. On a rectangular grid consisting of 10x10 or 100 points (the total number of points is PT), the number of grid intersections, points, intersecting the phase of interest is counted, Pa.
The volume fraction of the phase is then
if this is repeated for n fields of view then
clearly for each field of view the value of fa will vary so there will be an error in fa.
enables the standard error to be calculated. However, this appraoch does not allow the calculation of the number of points required to get a given error, for that we use
where s is the error, fa, the volume fraction of phase a and N, the total number of points. You should aim for an error of ±1%.
The intensity ranges from black (0) to white (255) and is known as an 8-bit image since 28= 256 gray levels. It is up to you to determine the threshold limits of gray level between which the phase you are interested in lies. Having found the lower and upper limits of gray scale, the image is converted into a binary (black/white) image wherein all the gray levels between the upper and lower thresholds are converted to black (1) and the remainder revert to white(0). It is now simply a matter of counting the number of black pixels and dividing by the total number of pixels to get the volume fraction for that field of view. Using the image analyser, repeat the point counting exercise to obtain the volume fraction of pearlite. You should note that image analysis is not very forgiving of poor sample preparation or microstructural artifacts
In the above image, containing 75046 pixels, the mean intensity is 161.66 with 0 being no black and 255 being all black. thus the area (volume) fraction of pearlite is 161.66/255 = 0.634 or 63.4%.
EXPERIMENT
For your steel, use point counting to determine the volume fraction of pearlite. You will need to make the measurements using at least 10 fields of view. You should deterime the volume fraction such that the error in the volume fraction is ± 5% or better. After completing the point count from each filed of view determine (a) the volume fraction and (b) the error using the cumulative number of counts for the pearlite. You should plot this measured volume fraction as a function of the total number of points counted to see how the measured volume fraction varies as the number of points counted increases. Include on your graph the limits of ±1 standard deviation. Take one field of view at random and produce a digitized image of the steel, convert the image to a binary and determine the volume fraction of pearlite. BE CAREFUL that you select only the pearlite and that any pixels due to artifacts are eliminated prior to measurement by using the erosion/dilation technique.
3. Grain Size - Linear Intercept
A straight line is drawn across the sample and its length is noted - to do this you will need to use the stage micrometer to determine the real length of the line, l.
The error in the measurement is
where sd is the standard deviation of all of the measurements of d that were made and N is the number of grains measured. It follows from the above equation that for a given error, s, the number of grains to be measured can be estimated. Obviously the process is iterative since sd changes as the number of measurments is increased.