Deformable layers

The mechanism of chemical adhesion is probably best studied and demonstrated by the use of silanes as adhesion promoters. However, it must be emphasized that the formation of chemical bonds may not be the sole mechanism leading to adhesion. Details of the chemical bonding theory along with other more complex theories that particularly apply to silanes have been reviewed [48,63]. These are the Deformable Layer Hypothesis where the interfacial region allows stress relaxation to occur, the Restrained Layer Hypothesis in which an interphase of intermediate modulus is required for stress transfer, the Reversible Hydrolytic Bonding mechanism which combines the chemical bonding concept with stress relaxation through reversible hydrolysis and condensation reactions. 

Deformable layers—coupling agents produce a tough, flexible layer. 

The stress curve sharply increases when the steric component appears upon compression. The initial thickness of a deformed layer is equal to be half the distance Dq obtained by extrapolating the sharpest initial increase to stress zero. The value Do is 21 1 nm, which is close the thickness of two molecular layers (19.2 nm) of the a-helix brush, calculated using the CPK model and the orientation angles obtained by FTIR analysis. We have calculated the elastic compressibility modulus Y,... 

For phases made of tilted molecules, measurements show that the inclination of the unit cell can be higher or smaller than the molecular tilt. This is not surprising since the crystallographic unit cell is defined by complex interactions between the deformed layers. 

Fig. 9 Few possible combinations of tilt and polarization for B1 and Blrev phases. In B1 phase molecules are (a) not tilted or (b) tilted in the blocks. The tilt is denoted by one part of the molecule being thicker. This is a part that is tilted out of the paper plane. B lrev structures can be made of the deformed synclinic, antiferroelectric layers with the tilt in the defect region connecting the blocks (c) lower or (d) higher than in the block itself or from (e) the anticlinic antiferroelectric deformed layers or from (f) the synclinic layers where polarization alternates in the neighboring blocks of the same layer...

Metals can also be electropolished to produce a smooth surface that is free of the deformed layer obtained by mechanical polishing I80]. The metal of interest is set up as the anode in a conducting liquid and undergoes a controlled corrosion reaction. 

TWO SEPARATE ALTERED or damaged layers classically have been recognized on metal surfaces formed by cutting- or polishing-type processes namely, an amorphous-like "Beilby" layer and a plastically deformed layer. Modern work indicates that the Beiiby layer is not, in fact, formed by the common important methods of surface preparation but that a deformed layer always is. The detailed structure of this layer is reviewed. Some consideration is also given to residual elastic stresses, surface topography, and embedded abrasive. 

The most satisfactory general description of the plastically deformed layer has been obtained from the examination of metallographic taper sections of the surfaces concerned. Investigations of 70 30 brass surfaces (16,7,10) have been particularly informative because etching techniques axe available to develop in the taper sections indications of prior plastic deformation that can be interpreted in some detail (17-21). Hie present description will be based on the metallographic approach, the more quantitative data obtained by x-ray and electron diffraction techniques being used to supplement it. 

Two distinct zones can be recognized in the plastically deformed layer namely, an outer fragmented layer of severer deformation, and a subcutaneous layer of general and more minor deformation. 

This zone, characterized by the fact that it differs from the substrate only in being subjected to comparatively mild deformation of a simple type, constitutes the bulk of the deformed layer. It can be recognized metallographieally only in those special cases where indications of prior plastic deformation can be developed by etching, or a similar technique. 

Fig. 2. Taper sections of a brass surface abraded on l/0 grade emery paper. Taper ratio 8.2. (a) Etched in ferric chloride reagent and showing the inhomogeneous distribution of the deformation close to the surface. X 100G before reduction for publication, (b) Etched to develop slip-line traces, showing the full extent of the plastically deformed layer. X 25(4 before reduction for publication. The fragmented layer, which is more clearly shown in Fig. 1, is also discern-able in these micrographs.

The phenomena so far described apply only to the effects produced by any one particular operation. It is often necessary in practice to use a sequence of different operations the deforma tion in the final surface will then be that characteristic of the final operation only if the pre-existing deformed layer is removed completely at each stage of the sequence. This can be achieved in metallographic practice if suitable precautions are taken (7, 36, 37), bat often would not be achieved in an industrial sequence. The final surface would then contain the residuals of one or more of the deformed layers produced by the earlier stages of preparation, and these could be much more extensive than the deformed layer produced bv the final stage itself. 

Most of the measurements that have been made of the depth of the deformed layer produced by various preparation procedures are of doubtful significance for one or both of two reasons. Firstly, it. is not certain that the layer investigated is representative of the procedure concerned and free from the residuals of layers produced at early stages of preparation. Secondly, no information is available on the sensitivity of the method used to detect deformation this can have a considerable influence on the depth value determined (see Fig. 5), although the results of any one investigation may still be comparative within themselves. 

The measurements have also so far been confined to soft metals and alloys in the annealed condition and it is difficult at present to predict from these results the likely depths in harder alloys it can only be assumed as a rough approximation that the depth will be inversely proportional to hardness (36,37). It may even be doubted by some whether there can be such a thing as a plastically-deformed layer in very hard metals, but there is good reason to believe that this can be so. An extensive zone of this nature has, in fact, been identified beneathindentationhardness impressions in fully-hardened high-carbon steels (35,55). It is again necessary to remember that the surface deformation occurs under very special conditions. 

The presence of residual elastic stresses is inevitable in a surface which contains a plastically deformed zone whose thickness is limited compared with that of the bulk s pecimen. Alternatively, the relief of these stresses may cause distortion of the specimens in cases where the thickness of the two is comparable. Very little work has been done on this important subject and, so far as can be ascertained, none that can be related to the complexities of die plastically-deformed layer. Considerable complications are introduced because the residual stresses may be of thermal as well as mechanical origin and because those of mechanical origin may be altered by the thermal effects. 

Information obtained from the study of metal surfaces, particularly when it is related to similar studies of ionic crystalline solids, provides some guide to the effects to be expected in semiconductors, which are considered hi the complimentary paper of this volume (70). An extensive plastically-deformed layer should be present, in spite of the extreme brittleness of these materials in bulk, but some surface shattering, particularly the development of cleavage facets, could also occur. 

Normally the craze microstructure is not directly visible in the scanning electron microscope. Thus, an etching procedure using oxygen ions was employed to remove the plastically deformed layer at the sample surface. Control measurements on uncrazed samples showed that this procedure does not lead to artifacts. The surfaces were coated with gold to reduce surface charging. 

Finally, the scenario that arises from the isotopic studies of deformed, layered pyroxenites involves two possible origins ... 

Fig. 1. Convection of material into a stable craze fibril through an active deformation layer.

---