Page 192 - Handbook of Properties of Textile and Technical Fibres

Silk: fibers, films, and compositesdtypes, processing, structure, and mechanics 169

5.3.6 Silk nanomechanics

The relationship between the chain structure and stressestrain behavior has been deter-
mined by Raman analysis under controlled strain.
It has long been understood that a relationship exists between the molecular chain
arrangements and the stressestrain curves (see Colomban, 2002 and references here-
in). Gosline et al. (1999) discussed the links between the mechanical properties and the
microstructure of spider silk. Experimental confirmation has been provided by nano-
mechanics, i.e., the quantitative study of the bond wave number shift measured by
Raman scattering under controlled tensile/compressive stress/strain applied to a single
fiber (Marcellan et al., 2003; Colomban et al., 2005, 2006; Gouadec and Colomban,
2007). These studies confirm that the mechanical properties are driven by the main
phase, the amorphous one, and in most of the cases, even in polymers with a high
crystalline content, the crystalline phase does not play a significant role in determining
the form of the tensile curve.
Obviously the instant lengthening/untwisting of a helix is consistent with an
obvious plateau in the tensile curve, see Fig. 5.3(a) (Hearle, 2000; Kreplak et al.,
2004; Paquin and Colomban, 2007; Woscieszak et al., 2014). On the other hand, no
phase transition plateau is expected for flat ribbon conformations. However, it is diffi-
cult to predict the effects of twists and distorted helices on the tensile curve. Knot
formation is possible and in this case a flat plateau can be ruled out. This can be consis-
tent with the hardening observed for Type IV curves (Fig. 5.3(a)). Sliding and opening
between b-like ribbons can explain a straight stressestrain curve.
The low wave number modes assigned to chain translation are strain-dependent
(Colomban et al., 2008a; Colomban, 2013). This is similar to Amide III modes and
some other modes of the 1200e1600 cm 1 region obtained by experiment at the
bond level with macroscopic strains applied to the single fiber (Sirichaisit et al.,
2000, 2003). These data reflect rather well the macroscopic mechanical behavior
with a first elastic behavior up to 2% of strain and then a more or less defined plateau.
The analysis of the nNeH modes (Colomban and Dinh, 2012; see Dinh, 2010 for
details) is also very informative on the structural changes due to tension or compres-
sion. The NeH modes are very sensitive because of the elasticity of the NeH bond
when the H.X hydrogen bond is modified (Novak, 1974; Gruger et al., 1995). It
has long been established that a reliable relationship exists between XeH(X ¼ O
or N) stretching wave numbers and the distance with the nearest acceptor (O or H)
to a hydrogen bond. Using the correlation established by Gruger et al. (1995) the cor-
responding distance of hydrogen bonding has been calculated. It is obvious that the
geometrical distortion is weak and can be attributed mainly to intrachain bonding.
However, this demonstrates that a lengthening of the chain bond takes place in the
elastic regime and that the bonds are not so affected by the stress applied at the macro-
scale. Fig. 5.17 compares the wave number shift as a function of applied macroscopic
tensile strain for different types of B. mori fibers (Colomban et al., 2012b; Dinh, 2010).
For each Type, I to IV, the shape of the Raman shift versus applied strain, i.e., the
mechanical behavior at the chemical bond scale (subnanometric scale) is identical to
that observed on single fiber tensite tests, as observed for all studied polymer fibers