Is there any new discovery about grain size and ductility...
Smaller grain sizes result in increasing strength and toughness and loss of tensile ductility. Note that this loss is not a major problem like what happens in strain hardening. However, if you consider a inhomogeneous cast microstructure, homogenizing (and the resultant grain refinement) increases both strength and ductility!
Reducing the GS does not decrease the ductility. Indeed, it is the one strengthening mechanisms that, if at all, it may actually increase it as explained above. Other strengthening mechanisms often decrease the ductility, although not always.
For example: cold deformation, via extrusion or rolling hardens the material at the expense of ductility. This happens because the strain hardening is exhausted during the extrusion/rolling, and that brings the onset of necking forward (see below).
Precipitation hardening may decrease the ductility too, depending on whether it occurs without increasing the strain hardening rate. Keep in mind that the Considere criterion states that necking starts when S= dS/de, where S is the true stress and e the true plastic strain. If S increases trough an additional strengthening necking will start earlier unless dS/de is increased in proportion. The latter is possible (i.e., preserving the ductility while making the alloy stronger) if the strengthening involves, for example, a mixture of coherent and non coherent particles. The former increase S and the latter increase dS/de via the Orowan mechanism.
1) Tensile ductility is generally controlled by parameters, viz. strain hardening (K and n in Hollomon equation) and strain rate sensitivity of flow stress, m. If Hollomon equation (connecting true stress and true plastic strain) is valid, then strain hardening parameters are strength co-efficient, K and strain hardening exponent, n. You may know that the true strain at necking is almost equal to n (εnecking = n). The onset of necking in tension is postponed/delayed thus improving ductility when m (strain rate sensitivity) and n (strain hardening exponent) are high. Please see Hart's instability criterion ("Theory of the tensile test", E. W. Hart, Acta Metall., 15,1967, pp. 351–355) and instability occurs when :- (n/ε) + m = 1. Otherwise, ε instability = n / (1 - m) and m = 0 gives you the Considere criterion.
2) One has to know how grain size will effect strain hardening and strain rate sensitivity, m. In other words, how ductility is influenced by grain size. It is generally known that n increases with increasing grain size (in certain range) and decreases with increasing volume fraction of second phase for a given particle size. It is the "mean free path" (MFP) of dislocations which is an important parameter that controls the strain hardening behavior and the n value.
3) You may please see the reference:- notes by Prof. Yngve Bergström on "The Hollomon n – value, and the strain to necking in steel"- for this, please see the link if you have access:- http://www.plastic-deformation.com/paper8.pdf
4) You may also see further references:- (i) "Strain hardening, strain rate sensitivity, and ductility of nanostructured metals", Y.M. Wang, E. Ma, Materials Science and Engineering A 375–377 (2004) 46–52. (ii) "The Relationship between the Strain-hardening Exponent n and the Microstructure of Metals", ZHANG FAN, HUANG MINGZHI and SHI DEKE, Materials Science and Engineering, A122 (1989) 211-213.
Bharat Heavy Electricals Limited
Indian Institute of Technology Bombay
University of Zagreb
José Gregorio La Barbera-Sosa
Central University of Venezuela
Sai Spurthi Institute of Technology
University of Akron
Indian Institute of Technology Delhi
National Institute of Technology Rourkela