Low-Ferrite Austenitic Stainless Steel
Foundries are often called upon to produce special grades of stainless steel. One of the factors which can be controlled is the volume percent of ferrite. A separate information sheet discusses the question “Why are stainless steel castings magnetic?” and describes some of the advantages and disadvantages of having the structure of the metal partially ferritic. But which factors determine the amount of ferrite and how are they controlled? And how is the amount of ferrite determined?
But first, what is ferrite? ASTM A800 defines ferrite as “the ferromagnetic, body-centered cubic, microstructural constituent of variable chemical composition in iron-chromium-nickel Alloys. This may be formed upon solidification from the molten metal (delta ferrite) or by transformation from austenite or sigma phase on cooling in the solid state (alpha ferrite).” In comparison, austenite is the non-ferromagnetic, face-centered cubic phase which is found in iron-chromium-nickel Alloys. It is important to note that both phases are Alloyed, that is, they contain high levels of chromium, nickel, molybdenum, etc. Ferrite should not be confused with “free iron” or “unAlloyed iron” which may be on the surface of either wrought or cast material.
The principal factors which determine the amount of ferrite are chemical composition, casting section thickness, and heat treatment cycle.
- Composition - ASTM Specification A800 explains how the average amount of ferrite in the castings poured from a heat of molten Stainless steel may be estimated from the chemical composition. Using the formulae given in A800, the amounts of the “ferrite promoting elements” and the “austenite promoting elements” are combined into a Chromium equivalent and a Nickel equivalent, respectively. Then the ratio of the two (Creq/Nieq) is used to determine the expected ferrite from a graph or from a table. The ferrite promoting elements are Cr, Mo, Si, and Cb. The austenite promoting elements are Ni, C, Mn, and N. Ferrite is reduced by lowering the ferrite promoting elements (the Creq) and/or raising the austenite promoting elements (the Nieq). Of course, each of the elements must also be held within the limits established for the specific grade being produced.
- Section thickness - The ferrite content will vary from one location to another in a given casting. In general, a thicker section will have a higher ferrite than a thinner section in the same casting. This effect is due primarily to the lower rate of solidification in the thicker section. The correspondingly lower cooling rate during quenching after heat treatment may also have an effect. By way of further explanation, the material used to develop the relationship given in ASTM A800 was taken from sand cast test bars, one to six inches thick , and solution heat treated at temperatures from 2000 to 2150 F for times of 1 to 4 hours. It should be pointed out that the Foundry has very little control over the section thicknesses of the casting; these are determined by the part designer and the function of the casting. Some sections are cast thicker than originally designed to provide machining stock or to provide improved casting integrity.
- Heat treatment - Heat treatment temperature and the hold time at that temperature are known to affect the final ferrite content of the castings. However, the research has not yet been done to permit the prediction of the amount of change or even the direction of the change given a starting chemistry and section thickness. Fortunately, the effect of heat treatment is usually small relative to those of composition and casting section thickness. Ferrite is normally considered to be beneficial. It provides increased strength and cracking resistance. In fact, the nuclear power industry specifies a minimum of 5 to 7 % ferrite to assure the required level of weldability (avoidance of “microfissures” during welding). This same minimum ferrite level provides much superior castability (resistance to tearing during solidification) as compared with fully austenitic grades.
Now, some comments about determining the amount of ferrite present in a given casting. There are three principal techniques: metallographic, magnetic, and calculation from the composition.
- Metallographic - This method is normally considered the “referee” method and is used to calibrate the other methods. A sample is removed from the casting (thus it is destructive) or from a representative test coupon (which may or may not actually be representative as may be concluded from the above discussion). The sample is polished and then chemically etched to provide the necessary optical contrast between the ferrite and austenite phases. The polished surface is then examined using a metallurgical microscope. The volume percent ferrite is determined using any of a number of well established “manual point counting” techniques. If the solidification rate is rapid (such as in thin walled castings or in weld deposits), the ferrite phase is very fine and even the point counting method is subject to considerable error. This is the primary reason that the welding industry has opted to assign a “ferrite number” based on magnetic response rather than to use the volume percent ferrite based on metallographic measurements.
- Magnetic - This method depends on the fact that ferrite is magnetic while austenite is non-magnetic. Two variants of this technique are commonly used: magnetic attraction and magnetic permeability.
- Magnetic attraction - In these instruments, a magnet is placed in contact with the metal surface and the force required to pull it away is measured. This force is an indication of the amount of ferrite present. Typical instruments in this category are the Magna-Gage and the Elcometer. A third instrument, the Severn Gage, determines whether the sample being tested has a greater or lesser attraction to a magnet than does a reference standard; it thus “brackets” the ferrite content between two available standards.
- Magnetic permeability - These instruments apply a low frequency current to a coil which is placed in contact with the metal surface. The electronic circuitry measures the change in magnetic properties of the coil and converts these changes into ferrite content. A typical instrument in this category is the FeritScope. Note that the magnetic instruments must all be calibrated against standards having known ferrite contents. Note, too, that the magnetic methods all are dependent on the magnetic characteristics of the ferrite which, in turn, are determined by the precise composition, the morphology, and the thermal history of the ferrite.
- Calculation - The formula for calculating the ferrite content from the composition is given in ASTM A800 which was discussed above. This method relies on the accuracy of the chemical analysis; the uncertainty of the chemical analysis is the reason for the scatter band in A800.