Metallography for cast aluminum alloys.

A look at yesterday and today's technology for cast aluminum metallography can help foundrymen optimize their microstructural analysis.

There are two approaches for the metallograpahic examination of cast aluminum alloys - the traditional and the contemporary. The traditional approach is manual and relies upon an operator's hand and experience to prepare the specimen for analysis. The contemporary approach is automated and relies on the equipment to execute the specimen preparation decisions of the operator. The key to choosing the right approach is to determine which preparation procedure provides the quality and number of specimens a foundry requires.

This article discusses the traditional approach to the preparation of cast aluminum specimens and the automated contemporary approach. Although a foundry may embrace one approach over the other, it is important to understand the basics of both to optimize the results of microstructural analysis.

Traditional Specimen Preparation Methods

Thirty years ago, a specimen preparation method for aluminum alloys called the "traditional approach" was perfected. After sectioning and mounting (if necessary) the specimen, it is ground manually ("hand" polished) with a series of finer and finer silicon carbide (SiC) waterproof papers. In most cases, the sequence is 120-, 240-, 320-, 400- and 600-grit papers (ANSI/CAMI grit size scale) using water both as a coolant/lubricant and as a cleanser to wash the abrasive remains and metal swarf (remains) off the paper surface.

If grinding is done by hand, the orientation of the specimen is changed from 45 [degrees] to 90 [degrees] between papers. The goal is to hold the specimen perpendicular to the paper while moving it from edge to center as the wheel rotates at 150 or 300 rpm for 1-2 min/paper.

[TABULAR DATA FOR TABLE 1 OMITTED]

After the SiC paper sequence, polishing begins. A 6 [[micro]meter] diamond paste is used for the first step with either canvas or nylon polishing cloths and a suitable lubricant/extender. The specimen is held perpendicular to the cloth surface as it is rotated in a direction counter to the wheel rotation for 2 min. While some metallographers use a 1-[[micro]meter] diamond paste on a nylon cloth for the next step, most use an aqueous 1-[[micro]meter] magnesium oxide (MgO) slurry on a high-nap wool broadcloth.

After using the cloth, it must be thoroughly washed, treated with a dilute (1:1) aqueous hydrochloric (HCl) solution and stored in this solution, so that magnesium carbonate does not form on the cloth. If magnesium carbonate does form, the contaminated cloth will scratch specimens being prepared.

Unlike most metals and alloys, final polishing using 0.3 [[micro]meter] alpha-alumina and 0.05 [[micro]meter] gamma-alumina aqueous suspensions is not appropriate for aluminum alloys because it does not produce acceptable final polished surfaces. Another option is vibratory polishing with a solution of 2 parts propylene glycol and 1 part water added to the MgO abrasive to prevent carbonation when using a vibratory polisher. These additions provide a complexing action that prevents carbonate formation.

While this traditional approach to specimen preparation procedure is satisfactory for producing acceptable matrix microstructures free of preparation artifacts, it is not good for edge preservation, as edges are often rounded and out of focus when viewed with the light microscope. Many techniques are used as cures for this problem, such as hightech mounting compounds, filler materials and other edge supporters, but good edge retention is still difficult to obtain.

Another problem for aluminum alloys prepared using the traditional approach is the control of relief of the second phase particles - the intermetallic compounds found in aluminum alloys. The wool cloth, with its high nap, promotes height differences between the particles and the matrix that makes higher magnifications difficult to use. In addition, relief around voids (due to gas evolution or micro-shrinkage cavities) develops, making determination of void content difficult as the interface between void and matrix is not sharp. This is due to the edge of the void not being flat.

These hand preparation procedures have been widely replaced by semiautomated devices. Statistical process control and other quality control programs have been introduced requiring image analysis devices to measure microstructural features. Emphasis has been placed on developing better preparation procedures, particularly with respect to relief control at intermetallics and voids.

Contemporary Specimen Preparation Methods

Along with equipment improvements have come new consumable products that have altered the traditional approach to metallography. Introduction of automation to specimen preparation has prompted development of a number of new surfaces that can be used to remove damage due to sectioning and achieve planarity for all of the specimens in the holder.

For a semi-automatic system, ordinary SiC paper can be used for the first step (planar grinding) as it is very effective for aluminum alloys. For a fully automatic system, SiC is acceptable for the planar grinding step, but it is not an ideal solution because it must be changed after each holder is prepared. Another option with an automatic system is a resin-bonded diamond grinding disc. Resin-bonded discs are available in 8- and 12-in. diameter sizes. The discs consist of a high density of small spots containing diamond abrasive of specific size within a carrier over the disc surface.

The procedure in Table 1, with vibratory polishing as the final step, is used in the automated preparation of a variety of cast aluminums. In this method, the contra direction means that the specimen [TABULAR DATA FOR TABLE 2 OMITTED] holder (which can hold a number of specimens, for example, six 1.25-in. diameter mounts) is rotating in the direction opposite that of the platen. Contra rotation provides more aggressive cutting action than complementary rotation where the holder and platen rotate in the same direction.

Other abrasive surfaces also can be used for the planar grinding step, such as 240- or 320-grit water-cooled SiC abrasive paper. In addition, the second step uses a hard woven, pressure-sensitive adhesive cloth with 9-[[micro]meter] diamond abrasive, however, a napless, hand-woven polyester cloth with diamond paste (natural or synthetic diamond) may be used with an extender/lubricant fluid. Then, while the specimen preparation system is running, diamond abrasive in slurry form is added.

The third step uses a napless, nonwoven synthetic chemotextile pad with 3-[[micro]meter] diamond abrasive. Another approach is to charge a new cloth with diamond paste and then recharge the cloth with the liquid suspension while the machine is running. In working with aluminum alloys, a napless, hard woven, synthetic acetate polishing cloth also yields excellent control of flatness and relief.

The final step uses a colloidal silica suspension on a napless, chemically-resistant, synthetic polishing cloth to promote flatness. Colloidal silica is a successful alternative to MgO. Although this is a very simple four-step specimen preparation procedure, it yields excellent results. To further improve the preparation for reproduction-quality micrographs, a 30-min vibratory polish using colloidal silica can be performed.

Etching

After the specimen has been prepared using either traditional or contemporary procedures, it should be examined in the as-polished condition before etching. Once that is complete, only a few etchants are needed for the bulk of metallographic work on cast aluminum alloys, although many others exist for special purposes, such as phase identification.

Table 2 lists the most commonly used etchants. Prior to the development of wavelength-dispersed spectroscopy on an electron microprobe analyzer and energy-dispersed spectroscopy on a scanning electron microscope, selective etching experiments were conducted as an aid to intermetallic phase identification.

To illustrate the value of vibratory polishing as a final polishing procedure, Fig. 1 shows three aluminum-silicon (Al-Si) alloys containing (1 to r) 7.15, 11.82 and 19.85% Si that were etched with dilute hydroflouric acid (HF) after being prepared as described in Table 1. Figure 2 shows the same alloys after a 30-min vibratory polish using colloidal silica abrasive followed by etching with dilute HF. While both procedures show the proper structure for routine work, the vibratory polishing has removed the few remaining scratches and made the structure crisper to the eye.

Control of the etch time also is critical to success in revealing the microstructure. As an example, Fig. 3 shows the same three specimens after vibratory polishing and with a light dilute HF etch, while Fig. 4 shows the same three specimens after vibratory polishing but with a proper etch depth. Unfortunately, it is impossible to closely define etching times, as the correct etch time and depth must be determined empirically and will vary with examination magnification. For high-magnification work, a more shallow etch depth is desired due to the more limited depth of field of the light microscope, while for low-magnification examination a deeper contrast etch is best.

In each of these micrographs, the surfaces are flat and free from unwanted relief problems and height differences between the silicon particles and the aluminum matrix are very minor. Thus, even at high magnification, good micrographs can be taken.