A D V A N C E D M A T E R I A L S & P R O C E S S E S | S E P T E M B E R 2 0 2 2 2 2 When a liquid metal is poured, its surface oxide film can be folded over and the top dry surfaces of the film come into contact (Fig. 1). There is no bonding between these two films, so the double film acts as a crack in suspension in the metal. Similarly, droplets and splashes come together, dry side to dry side of their surface oxide films, so that double films as cracks (bifilms) are always formed[1]. Turbulent pouring can fill the liquid metal with cracks. This process also occurs in socalled vacuum conditions because the vacuums associated with the melting and pouring of metals contain sufficient oxygen to form a surface film. However, the film is much thinner, so the metal certainly contains less oxygen, and is thereby concluded to be “cleaner,” but probably contains the same area of cracks[2]. Vacuummelting and casting is therefore seen to be ineffective in eliminating these serious defects in metals. It is important to realize that the underside of the oxide film on liquid metals is in atomic contact with the melt, having grown from the melt surface, atom by atom. Thus, although the interior interfaces of the bifilm are perfectly dry and unbonded, its exterior interfaces are perfectly wetted, being in intimate atomic contact with the matrix. This unique structure enables the bifilm to interact with metals, controlling properties and structures in a unique way. For instance, the bifilm’s exterior faces appear to be favored substrates for the formation of inclusions and second phases in all alloy systems investigated so far, including alloys of tin, magnesium, aluminium, iron and steel, nickel, and titanium. If the inclusion grows only on one face of the bifilm, then under stress the opening of the bifilm will appear to be the inclusion decohering from the matrix. On the other hand, if the inclusion has grown on both sides of the bifilm, the central bifilm will open under stress, giving the appearance that the inclusion has cracked because of the stress. The regular association of cracks through or at the side of inclusions and second phases is taken by this author to be conclusive evidence for the preferred formation of inclusions and second phases on bifilms. Inclusions appear to form only on bifilms—and the bifilms are often in the grain boundaries, explaining the confusion; inclusions form on bifilms, not grain boundaries. In general, the bifilms in metals are especially thin, with a thickness measured in nanometers. As a consequence, they have tended to escape notice for several thousand years. The purpose of this short article is to draw attention to the well-known features of fracture surfaces for which there are currently no coherent explanation, but which the presence of bifilms explains simply and naturally. Similarly, despite the absence of a theoretical failure initiation mechanism, many failure modes for metals exist, but in every case, this author draws attention to the existence of good evidence[2] that the single reason for the initiation of failure in nearly every case is the bifilm. The bifilm is not intrinsic to metals; it is an externally introduced extrinsic defect. Because bifilms need not be present if casting technology were to be changed, all the failure modes would cease to exist. If metals were not loaded with unnecessary stress raisers such as bifilms, hacksaw cuts, and designed-in sharp corners, metals would, for the first time, enjoy the intrinsic resistance A REVIEW OF ASSUMPTIONS OF FAILURE ANALYSIS A look at how bifilms are created during casting may offer clues to eliminating cracks, tensile fracture, fatigue failure, pitting, and stress corrosion cracking failures. John Campbell* University of Birmingham, U.K. *Member of ASM International PERSPECTIVE Fig. 1 — The oxide film on the liquid metal surface being entrained by folding and splashing during surface turbulence.
RkJQdWJsaXNoZXIy MTYyMzk3NQ==