Page 337 - Fiber Fracture
P. 337

FRACTURE OF NATURAL POLYMERIC FIBRES                                 319























           Fig.  6. Fracture  surfaces produced  in  lepidopteran  silk  by  hand-drawing  under  liquid  nitrogen.  Different
           degrees of  ductility  are  evident in  the  two  materials, suggesting that  there  are  significant  microstructural
           differences  between  them.  Left:  Bombyx  mori  (domesticated  silkworm)  cocoon  silk.  The  white  line
           emphasises  the  fact  that  the  fibre  (bave) consists  of  two  filaments  (brins).  Right:  Rothschildiu  erycinu
           cocoon silk. The sample has deformed to an extent that precludes identification  of the individual brins.

           of the Weibull modulus m:
             F=l-exp(-$)      m



           where F is the cumulative probability that a sample has a fracture strength of n, and no
           is a constant. If  F  is to be independent of  (T,  which is the hallmark of an ideally brittle
          material  (section “Some Thoughts on the Meaning  of  ‘Brittle”’), Eq. 2 requires  m  to
          be equal to zero. Since practical Weibull analysis admits to non-zero values of m, it is
          implicit that ductile contributions to failure can be accommodated too.
             The use of Weibull statistics to characterise failure probabilities in batches of natural
           fibre is therefore appropriate, and can highlight the existence of significant microstruc-
          tural differences between different materials. Comparison of different materials is most
           straightforward if similar sample volumes can be used, but this is difficult if the materi-
          als have different and irregular fibre cross-sections. A further complication is introduced
          by viscoelasticity, which will make strength depend on the detailed stress-strain  profile
           of the material and the rate at which samples are loaded. Comparisons should therefore
          be conducted  at defined and reproducible strain rates. It must be recognised, however,
           that comparison  of  different  polymers  under  exactly equivalent  conditions  can  never
          be achieved in  practice, because it is most  unlikely that different polymers  will have
           identical  viscoelastic  characteristics.  Silks  are  especially  ‘unusual’:  while  increased
          deformation rates are associated with higher strength, higher stiffness and lower elon-
          gation to failure in most viscoelastic materials, silk exhibits an increased elongation to
          failure  (Kaplan et al.,  1997). This observation  supports the  idea that the  propagation
          characteristics of defects, rather than bulk viscoelastic behaviour, governs the fracture of
          natural silk fibres.
   332   333   334   335   336   337   338   339   340   341   342