Normal force phenomena

4 [16] shows the negative effect in the format usually seen, as log Nj plotted against log dy/dt, for a 16.4 wt% solution of racemic 350 000 PEG + m-cresol. 

The obviously notable features of with increasing shear rate are the sequence  

After a brief background, we will return to a detailed discussion of the features noted above, as well as other subtle observations. 

The early developments in this field have been reviewed [6]. The earliest mention of negative that we have encountered for any system was by Hutton [35], who measured normal stresses in lubricating greases. He concluded that this strange behavior was a manifestation of a directional effect, involving the gap-setting servomechanism. When this mechanism was switched off, a more regular type of result was obtained.  

Normal stress measurements for some MLC nematics was reported to be consistent with that of a second-order fluid, that the low frequency limit of G /co equaled the low shear limit of N /(dy/dty [36]. Coleman and Markowitz demonstrated that for a second-order fluid in slow Couette flow, the viscoelastic contribution to the normal thrust must have a sign opposite to the inertial contribution on thermodynamic grounds [37]. A textbook by Walters stated that the measurements of first normal stress difference have invariably led to a positive quantity except for one case which was later found to be in error [38]. Adams and Lodge reported the possible observation of a negative value for Nj for solutions of poly isobutylene + decalin [39]. This result was obtained by a combination of obtained from radial 

---

At the same time that we were confronted with the negative normal force phenomenon, we also discovered an odd morphological phenomenon, that of the shear-induced band structure. Many polymer liquid crystal systems have been seen to exhibit striations perpendicular to the shearing direction upon cessation of shear. To our knowledge, the formation of striations or bands was first mentioned in passing by Aharoni [12], who did not specify the shearing direction. This phenomenon has also been confirmed repeatedly and has also now been satisfactorily explained, as discussed in section 11.6. 

We expected to find interesting science, but we were not expecting to wander into a phenomenon which was so unexpected, so counterintuitive that we ourselves assumed it was an artifact. We learned that, indeed, inertia is a source of potential artifact which can produce a positive-to-negative normal force transition in isotropic polymer solutions, as shown in Figure 11.1 [7]. 

Despite the numerous confirmations of the negative phenomenon, it has still been widely stated that the flow of all polymer systems exhibits only positive primary normal forces (i.e. a positive Nj, the first normal stress difference) [8, 9]. Even subsequent reviews and research papers on the specific subject of lyotropic main chain liquid crystal polymers have not mentioned the confirmed negative effect [10], and even equivalent shear measurements on the identical solutions did not report the negative effect [11]. 

Which wear mechanism dominates in a given situation is not always obvious and often two or more wear mechanisms may act jointly. Because all models predict the same dependency on hardness, normal force and sliding velocity, other aspects must be studied to distinguish wear mechanisms. Useful information on wear mechanisms is obtained from microstructural observation of the worn surfaces and of wear debris. Finally, we note that three of the above wear mechanisms involve only mechanical interactions, while oxidative wear is in fact a tribocorrosion phenomenon, combining mechanical wear and high temperature oxidation. 

Tack refers to the adhesion of two surfaces of the same rubbery polymer. When two such surfaces are pressed together and subsequently pulled apart, the maximum force necessary to break the junction depends on the initial time of contact and the normal force applied, as well as the rate of separation and the temperature and other variables. " From the dependence on temperature and polymer molecular weight, it can be inferred that the effectiveness of the bond depends partly on the interdiffusion of molecules across the interface and hence on molecular motions which are reflected in viscoelastic properties in the terminal zone. - However, the effectiveness depends also on the ultimate properties of the polymer itself as discussed in Section E below, and the phenomenon is still not fully understood. 

The interaction of two alkali metal atoms is to be expected to be similar to that of two hydrogen atoms, for the completed shells of the ions will produce forces similar to the van der Waals forces of a rare gas. The two valence electrons, combined symmetrically, will then be shared between the two ions, the resonance phenomenon producing a molecule-forming attractive force. This is, in fact, observed in band spectra. The normal state of the Na2 molecule, for example, has an energy of dissociation of 1 v.e. (44). The first two excited states are similar, as is to be expected they have dissociation energies of 1.25 and 0.6 v.e. respectively. 

However, the assumption of molecule orientation normal to the surface is not convincing enough for this author, and it does not consist well with the results of the molecular d5mamics simulations for the alkane confined between solid walls. An example in Fig. 3 shows that the chain molecules near the wall are found mostly lying parallel, instead of normal, to the wall [6]. This means that the attractions between lubricant molecules and solid wall may readily exceed the inter-molecule forces that are supposed to hold the molecules in the normal direction. Results in Fig. 3 were obtained from simulations for liquid alkane with nonpolar molecules, but similar phenomenon was observed in computer simulations for the functional lubricant PFPE (per-fluoropolyether) adsorbed on a solid substrate [7], confirming that molecules near a solid wall lie parallel to the surface. 

The high dielectric constant of water normally militates against the formation of ion-pairs for simple salts because a high dielectric constant reduces the strength of the electrostatic forces. The phenomenon is more readily observed in solvents of low dielectric constant for a typical mono-monovalent salt, ion-pair formation takes place only when the dielectric constant is less than 41 (Fuoss Kraus, 1933). 

Another well-known phenomenon is the Weissenberg effect, which occurs when a long vertical rod is rotated in a viscoelastic liquid. Again, the shearing generates a tension along the streamlines, which are circles centred on the axis of the rod. The only way in which the liquid can respond is to flow inwards and it therefore climbs up the rod until the hydrostatic head balances the force due to the normal stresses. 

---